Oral Presentations

OP01: MIS Periop Optimization

ID: AGS26/131

A001: A 2 year retrospective analysis of clinical outcomes for patients treated for single level lumbar disc herniations with radiculopathy between uniportal interlaminar endoscopic spine surgery and biportal endoscopic spine surgery

Uday Raman 1 , Nikhil Kadanthode 1 , Sreejith Chandran Menon 1 , Midhun P. C 1

1 Meitra Hospital, Kozhikode, India

Introduction: Minimally invasive spine surgery consists of both uniportal interlaminar endosocopy and biportal endoscopy. Both the methods are widely used in treatment of lumbar disc herniations. The study was conducted for comparing the clinical outcomes between the two methods namely biportal endoscopic spine surgery and uniportal interlaminar endoscopic spine surgery for the treatment of single level lumbar disc herniations with radiculopathy. Material and Methods: A total of 79 patients were followed up with single level lumbar disc herniations that underwent endosopic spine surgeries out of which 58 underwent uniportal interlaminar endoscopy and 21 underwent biportal endoscopy in the past 2 years and had a minimum follow up of 6 months post surgery. All Patients that were selected had single level disc herniations with radiculopathy. The patients preoperatively were assessed for pelvic parameters and Phirrman grading. The patient’s outcomes were measured in the form of surgery duration, duration of hospital stay, post operative complications, visual analogue scale (VAS) for back and leg pain and the Oswestry disability index (ODI). Results: The VAS scores and ODI decreased significantly in two groups after surgery. The results demonstrated that the mean operating time for biportal endoscopy was significantly higher but the radicular pain reoccurrence was higher in the uniportal group. The length of hospital stay, complication rates, Oswestry Disability Index score, VAS score for back and VAS score for leg were comparable for both groups at 1 day, 1 month and 6 months follow up. Conclusion: According to our study and data, both group of patients had comparable early postoperative back pain relief. With similar complication rates and comparable outcomes in terms of clinical outcomes. Nevertheless, both surgical techniques are safe and effective and are a viable option for lumbar disc herniations.

ID: AGS26/135

A002: Improvement in low back pain following endoscopic decompression for spinal stenosis

Maria Auron 1,2 , Yihyun Roh 2 , Maria Pedraza 2 , Win Boonsirikamchai 3 , Yi-Hao Liang 4 , Jung-Hoon Kim 2 , Jin-Sung Kim 2

1 Spine Surgery Unit, Department of Orthopedics, Hadassah Medical Center, Hebrew University of Jerusalem, Jerusalem, Israel,

2 Spine Center, Department of Neurosurgery, Seoul St. Mary’s Hospital, College of Medicine, The Catholic University of Korea, Seoul, South Korea,

3 Department of Orthopedics, Bhumibol Adulyadej Hospital, Bangkok, Thailand,

4 Orthopedic Department, Guangdong Provincial Hospital of Chinese Medicine, Guangzhou University of Chinese Medicine, Guangzhou, China

Introduction: Decompression surgery for lumbar spinal stenosis (LSS) effectively relieves leg symptoms, but its impact on low back pain (LBP) is considered unreliable. In this study, we evaluated the improvement in LBP following full-endoscopic decompression for LSS and analyzed the clinical and radiological factors that may influence these changes. Material and Methods: Consecutive patients who underwent full-endoscopic decompression for LSS between 2017 and 2024 in our hospital were retrospectively analyzed. Pre- and postoperative clinical and radiographic status were assessed for patients with a follow-up of ≥ 1 year. The primary outcome was reaching a minimally important clinical difference (MCID) in Visual Analogue Scale (VAS) for LBP. Secondary outcomes included the change in VAS leg pain, Oswestry disability index and modified MacNab criteria. Radiological assessment included spinal canal cross-sectional area, Modic type 1 change, sagittal vertebral axis, Cobb angle, percentage of vertebral slippage, lateral listhesis, pelvic tilt, sacral slope, pelvic incidence, lumbar lordosis, segmental lordosis and PI minus LL. Logistic regression analyses were performed to identify factors associated with achieving MCID, and subgroup analysis assessed the impact of low-grade degenerative spondylolisthesis (SPL) on outcomes. Results: A total of 64 patients (mean age: 63.8 years, 53.1% female) were analyzed with a mean follow-up of 24.8 months. 23 patients (35.9%) had low-grade SPL. MCID was achieved in 84.5% of patients at 12 months and 79.7% at last follow-up. Radiographic analysis showed improved postoperative sagittal lumbopelvic alignment after surgery, although no specific radiographic parameters were identified as predictors of clinically meaningful back pain improvement. Greater baseline back pain severity was associated with achieving MCID (OR = 2.31, 95% CI 1.31 - 4.08, p = 0.004), and low-grade SPL was not associated with inferior clinical outcomes. Conclusion: Full-endoscopic decompression for LSS is associated with clinically meaningful improvement in LBP in the majority of patients. Higher preoperative back pain was identified as a significant predictor of achieving MCID, and the presence of low-grade SPL did not negatively affect outcomes. Our findings support the efficacy and long-term stability of endoscopic decompression for lumbar spinal stenosis, including degenerative low-grade spondylolisthesis.

ID: AGS26/3084

A003: Minimally invasive surgery for repair of spontaneous ventral spinal cerebrospinal fluid leaks

David Rowland 1 , Dwij Mehta 2 , Indran Davagnanam 3 , Ayman Qureshi 3 , Peter Cowley 3 , Salwa Kamourieh 2 , Sanjay Cheema 2 , Manjit Matharu 2 , Parag Sayal 1

1 Victor Horsley Department of Neurosurgery, National Hospital for Neurology and Neurosurgery, London, United Kingdom,

2 Headache and Facial Pain Group, University College London (UCL) Queen Square Institute of Neurology, London, United Kingdom,

3 Lysholm Department of Neuroradiology, National Hospital for Neurology and Neurosurgery, London, United Kingdom

Introduction: Spontaneous intracranial hypotension (SIH) secondary to ventral dural defects in the spine is an increasingly recognised pathology. This is a debilitating condition with symptoms of orthostatic headache with associated audiovestibular and cognitive disturbance. Sequelae of untreated leaks include siderosis and frontotemporal brain sagging syndrome. Material and Methods: We described the largest UK series of surgical management of ventral CSF leaks and our MDT approach for diagnosis and subsequent management. Results: Forty-five patients, diagnosed through a specialist SIH MDT, underwent surgical management of a ventral CSF leak over a 4-year period. The most common symptom was an orthostatic headache. Twelve patients had a diagnosis of superficial siderosis pre-operatively. All patients underwent myelography to identify the leak and location. Surgery to repair the CSF leak was predominantly by a minimally invasive, dorsal, transdural approach with 360 degree sealing of the dural defect. There was a median duration of 8 months follow up. 80% patients had improvement in their symptoms at follow up. 39% had complete resolution, 41% had partial improvement and 20% had no change in their symptoms. Those with improvement in their symptoms tended to have a shorter duration of symptoms prior to surgery (median 1.8 years for those with improvement vs median 4.4 years without improvement, p = 0.19). There was a significantly higher proportion of patients who did not improve post operatively in the subset with a pre-operative diagnosis of superficial siderosis. 50% of patients with siderosis pre-operatively had no improvement compared to 10% of patients without siderosis showing no improvement (p = 0.02). All patients had radiological resolution of the associated spinal epidural collection on follow up MRI scan. There were four post operative complications in separate patients; arachnoiditis resulting in lower limb paraesthesia, development of post operative raised intracranial pressure requiring a VP shunt, revision surgery for repair of their ventral leak and a pulmonary embolism. Conclusion: We have described the development of an effective MDT approach for surgical management of ventral CSF leaks. The majority of patients had symptomatic improvement and resolution of the ventral leak radiologically. Those with a longer duration of symptoms or a pre-operative diagnosis of superficial siderosis were less likely to improve. We would recommend offering early surgery and counselling patients on the risks of developing siderosis without repair of the leak. Long term radiological follow up assessing for siderosis should be considered in patients who decline surgical management.

ID: AGS26/1794

A004: Objective gait assessment after unilateral biportal endoscopic discectomy: a prospective smartphone-based analysis of spatiotemporal recovery in lumbar disc herniation

Ali Kaplan 1 , Begüm Aslantaş Kaplan 1 , İlker Solmaz 2

1 Tokat Erbaa State Hospital, Tokat, Turkey,

2 Private Practice, Ankara, Turkey

Introduction: Lumbar disc herniation (LDH) significantly impairs gait function through pain, neuromuscular deficits, and compensatory movement patterns. While unilateral biportal endoscopic (UBE) discectomy has demonstrated favorable clinical outcomes through patient-reported outcome measures (PROMs), objective functional assessment remains limited. Traditional laboratory-based gait analysis, though accurate, is impractical for routine clinical use. Smartphone-based gait analysis offers an accessible alternative for objective functional assessment. This study aimed to prospectively evaluate the restorative effect of UBE discectomy on spatiotemporal gait parameters using validated smartphone technology, quantifying baseline deficits compared to healthy controls and measuring postoperative recovery. Material and Methods: This single-center prospective observational study enrolled 42 consecutive patients with single-level LDH (L4-L5 or L5-S1) scheduled for UBE discectomy and 45 age-and gender-matched healthy controls. All procedures were performed by a single board-certified neurosurgeon using standardized UBE technique. Objective gait assessment utilized the validated Gait Analyzer smartphone application with participants wearing Samsung Galaxy S24 positioned at L3 vertebra. Three 10-meter walking trials at self-selected pace were recorded preoperatively and at 6-week follow-up for patients, once for controls. Measured parameters included gait velocity, cadence, step time, step length, step-time and step-length symmetry, and vertical center-of-mass excursion (Vert-CoM). Subjective outcomes included Visual Analog Scale (VAS) for leg pain and Oswestry Disability Index (ODI). Statistical analysis employed independent t-tests, paired t-tests, mixed-model ANOVA for level-stratified analysis (L4-L5 vs L5-S1), and correlation analysis between subjective and objective measures. Results: Eighty-seven participants were analyzed Groups were comparable for age (45.1 ± 10.5 vs 43.3 ± 9.9 years, p = 0.399) and gender distribution (p = 0.964). Preoperatively, patients demonstrated significant gait impairment compared to controls: reduced velocity (0.83 ± 0.18 vs 1.14 ± 0.15 m/s, p < 0.001), lower cadence (88.2 ± 12.3 vs 106.8 ± 7.1 steps/min, p < 0.001), prolonged step time (756 ± 109 vs 549 ± 138 msec, p < 0.001), shorter step length (0.55 ± 0.13 vs 0.64 ± 0.07 m, p < 0.001), increased asymmetry (step-time: 38 ± 28% vs 9 ± 4%, p < 0.001; step-length: 34 ± 21% vs 9 ± 3%, p < 0.001), and reduced Vert-CoM (1.57 ± 0.63 vs 2.57 ± 0.51 cm, p < 0.001). At 6 weeks postoperatively, patients showed significant improvement in all parameters (all p < 0.001). Gait velocity, cadence, step time, step length, and step-time symmetry normalized to control levels (all p > 0.05), while step-length symmetry (13 ± 7% vs 9 ± 3%, p = 0.001) and Vert-CoM (2.28 ± 0.65 vs 2.57 ± 0.51 cm, p = 0.007) improved but remained different from controls. Level-stratified analysis revealed superior recovery at L4-L5 versus L5-S1 for velocity (+0.48 ± 0.40 vs +0.19 ± 0.27 m/s, p = 0.015) and step length (+0.24 ± 0.19 vs +0.03 ± 0.16 m, p = 0.001). VAS decreased from 8.12 ± 1.13 to 0.88 ± 0.96 (p < 0.001) and ODI from 70.1 ± 13.0 to 21.9 ± 10.0 (p < 0.001). Pain reduction correlated with gait improvements (r = -0.375 to 0.567, p < 0.05), while ODI changes showed no significant correlations with objective measures. Conclusion: UBE discectomy produced improvements in objective gait function by 6 weeks postoperatively. While most spatiotemporal parameters normalized, residual deficits in step-length symmetry and vert-CoM suggest incomplete neuromuscular recovery requiring targeted rehabilitation. Level-specific recovery patterns indicate L4-L5 decompression yields superior spatial gains while L5-S1 cases benefit more from cadence increases, reflecting root-specific biomechanical differences. The divergence between pain reduction correlating with objective improvements while disability scores did not emphasizes the complementary nature of subjective and objective outcome measures.

ID: AGS26/2201

A005: Clinical outcomes of microscopic vs. endoscopic decompression for stable lumbar degenerative spondylolisthesis: a subgroup analysis of a randomized controlled trial

Luca Ambrosio 1,2 , Yi-hyun Roh 2 , Jung-Hoon Kim 2 , Maria Pedraza 2 , Maria Auron 2 , Jin-Sung Kim 2

1 Fondazione Policlinico Campus Bio-Medico, Rome, Italy,

2 The Catholic University of Korea, Seoul, South Korea

Introduction: Degenerative lumbar spondylolisthesis (DLS) affects approximately 5-8% of adults and is frequently associated with lumbar spinal stenosis (LSS). In patients without frank instability, decompression alone can provide favorable outcomes while avoiding fusion. Unilateral laminotomy for bilateral decompression (ULBD) is an established technique that minimizes bony removal and preserves the facet joints, thereby reducing the risk of iatrogenic instability. Traditionally performed using a microscopic approach, endoscopic ULBD is being increasingly performed with promising outcomes. The aim of this study was to compare clinical and radiographic outcomes of patients with DLS treated with microscopic versus endoscopic ULBD. Material and Methods: We conducted a multicenter, prospective, assessor-blinded, non-inferiority randomized controlled trial including adults (20-80 years) with one- or two-level central LSS (Schizas ≥ B and DLS grade ≤ 2) randomized to microscopic or full-endoscopic ULBD. From this cohort, patients with DLS were analyzed. Baseline demographics, perioperative parameters (incision length, drain volume, operative time, length of stay, postoperative creatine phosphokinase [CPK], postoperative dural sac area), clinical outcomes (visual analog scale [VAS] for back and leg pain, Oswestry Disability Index [ODI], European Quality of Life 5-Dimensions-5-Level [EQ-5D-5L], walking capacity, Patient and Observer Scar Assessment Scale [POSAS], and satisfaction rate), and radiographic outcomes (slippage, segmental lordosis, lumbar lordosis) were assessed at 2 weeks, 3, 6, and 12 months. Results: Of 110 randomized patients, 21 who underwent endoscopic ULBD (68.9 ± 6.2 years, 8 males, 13 females) and 15 who underwent microscopic ULBD (68.7 ± 9.2 years, 12 males, 3 females) had DLS and were included in this subgroup analysis. Apart from a slightly higher body mass index in the microscopic group (26.3 ± 1.9 vs. 24.6 ± 2.6 kg/m 2 , p = 0.035), baseline characteristics were comparable. Endoscopic ULBD was associated with significantly lower postoperative CPK levels (135.3 ± 90.5 vs. 205.4 ± 84.8 mg/mL, p = 0.002), drain volume (84.0 ± 80.7 vs. 206.7 ± 100.9 mL, p < 0.0001), and incision length (13.6 ± 9.9 vs. 45.2 ± 11.2 mm, p < 0.0001), whereas microscopic ULBD was characterized by a shorter operation time (118.0 ± 36.4 vs. 159.2 ± 28.3 min, p = 0.0005). No difference in postoperative dural sac area was found. Both groups experienced significant improvements in VAS back and leg pain (p < 0.001), ODI (p < 0.05), EQ-5D-5L (p < 0.01), and walking capacity (p < 0.05) from baseline. Microscopic ULBD resulted in significantly better 12-month leg pain relief (0.6 ± 0.9 vs. 2.1 ± 2.3, p < 0.05), quality of life (6.3 ± 1.3 vs. 8.2 ± 2.9, p < 0.0001), and satisfaction rates (80.8 ± 28.9 vs. 96.4 ± 4.4%, p < 0.05), and 6-month walking duration (71.7 ± 37.9 vs. 46.5 ± 18.7 min, p < 0.05). Radiographic outcomes showed no significant intergroup differences. At one year, two cases of DLS progression were documented (one per group) and were not associated with unfavorable clinical outcomes. Conclusion: In patients with DLS, microscopic ULBD provided superior improvements in pain, disability, quality of life, walking capacity, and satisfaction vs. endoscopic ULBD. However, the clinical relevance of these differences remains uncertain. Careful patient selection is essential, and larger randomized trials are needed to validate these subgroup findings.

ID: AGS26/841

A006: Ultra-early intervention with percutaneous kyphoplasty for acute osteoporotic vertebral compression fractures: efficacy and safety outcomes from a high-volume county hospital center

Wenjie Wu 1 , Li Huo 1 , Chuanqi Zou 1 , Junxiong Cao 1 , Boxing Wang 1 , Qian Cheng 1

1 The People's Hospital of Banan District, Chongqing, China

Introduction: The optimal timing for performing percutaneous kyphoplasty (PKP) in acute osteoporotic vertebral compression fractures (OVCFs) remains controversial, particularly regarding the risk of cement leakage in hyperacute cases. While delayed intervention is conventional, ultra-early surgery (< 72 hours) may offer superior pain relief and functional recovery. This study presents a large-scale prospective analysis from a high-volume county hospital, evaluating the feasibility, safety, and efficacy of ultra-early PKP. Material and Methods: From January 2023 to January 2025, 526 patients with single-level acute OVCFs were prospectively enrolled. Patients were divided into two groups based on symptom-to-surgery time: Group A (ultra-early, ≤ 72 hours; n = 302) and Group B (conventional, > 72 hours to 2 weeks; n = 224). All procedures were performed by the same surgical team using a uniform technique with high-viscosity cement. Primary outcomes included visual analog scale (VAS) score, Oswestry Disability Index (ODI), vertebral height restoration, and cement leakage rate. Secondary outcomes included operative time, hospital stay, and patient satisfaction. Results: The ultra-early group (A) demonstrated significantly better outcomes in pain reduction at 24 hours postoperatively (mean VAS: 2.3 ± 0.8 vs. 3.6 ± 1.1, p < 0.01) and shorter average hospital stay (4.5 ± 1.2 days vs. 7.8 ± 2.4 days, p < 0.01). Notably, the cement leakage rate was lower in Group A (6.1% vs. 11.4%, p < 0.01). Vertebral height restoration was significantly greater in the ultra-early group (88.5% ± 6.7% vs. 76.3% ± 8.9%, p < 0.01). There were no significant differences in operative time or overall complication rates. Patient satisfaction scores were higher in Group A (9.2/10 ± 0.7 vs. 8.1/10 ± 1.4, p < 0.01). Conclusion: Ultra-early PKP performed within 72 hours of symptom onset is not only feasible but also associated with superior clinical outcomes, reduced cement leakage, and greater cost-effectiveness due to shortened hospitalization. This study provides compelling evidence that challenges traditional timing paradigms and highlights the advantages of high-volume regional centers in implementing early intervention protocols. These findings support the adoption of ultra-early PKP as a standard of care for eligible patients.

ID: AGS26/247

A007: Time frame classification of unilateral biportal endoscopy for learning curve and relation with complications

Ahmed Alqatub 1 , Ghazwan Hassan 2 , Menna Alzaafarany 3 , Yasameen Resen 3

1 Alhasan Spine Center, Alhasan Almujtaba Teaching Hospital, Karbala, Iraq,

2 Middle East Private Hospital, Baghdad, Iraq,

3 Alhasan Spine Center, Karbala, Iraq

Introduction: The surgery of Unilateral Biportal Endoscopy Discectomy pass through different stages. AO spine task force wrote on 10 steps for doing the surgery. In this study we classified the surgery into 6 steps each has its beginning and ends and we studied the relation of the time to the complications and the learning curve and the improvement of the time scale over the running surgeries. Material and Methods: A prospective case series study. We studied our first 79 cases in doing UBE the study started from October 2023 to June 2025. We excluded 16 case due to the complexity of the surgery, double crush and associated unilateral Laminotomy for bilateral decompression. And we divided the time into 6 zones according to the following: A: Zero to record: The time of preperations, identification of the entry point, incision & muscle detachment to entry the scope. B: Start to Bear bone: The time from the record (scope entry) to end of rooming (control of bleeding, creating workspace, clearing the spino-laminar junction). C: Bear bone to Flavetomy: Laminotomy (by kerrison, osteotome or speed burr), and flavetomy (entering the epidural space). D: Foraminotomy & Lateral recess decompression: Identification of the traversing root, lateral recess decompression and foraminotomy. E: Discectomy: Performing the discectomy & annuloplasty. F: Lag time: The lag time, after annuloplasty to endoscope withdrawal or drain insertion. Results: Linear regression results with side and level as predictors: Moderate influence: right-side adds +1.24, L5-S1 adds +6.77 time units for Step B; level has notable effect. Statistical significance of side and level effects on procedure times: level significantly affects Step B times; side not significant. Association of procedure times (A-F) with complications alone: Significant association; complications tend to increase Step A times. Conclusion:

• We concluded that Significant downward trend in Step C times with case progression (R 2 ≈ 31%). This Suggests operator learning, improved efficiency, or team familiarity for this surgical step.

• Step B times significantly longer in L5-S1 surgeries (p ≈ 0.01).

• Step A times significantly longer in cases with complications (p ≈ 0.046). • Right-sided surgeries show mild trends toward longer times in some steps (especially Step F), but not statistically significant.

ID: AGS26/2950

A008: Implementation of a multi-institutional enhanced recovery after surgery (ERAS) for elective spine surgery: two-year Michigan Spine Surgery Improvement Collaborative (MSSIC) experience

Anisse Chaker 1 , Kylie Springer 1 , Kari Jarabek 1 , Michael Melhem 1 , Heegook Yeo 1 , Jian-hui Hu 2 , Jad Khalil 3 , Kenneth Easton 4 , Richard Easton 5 , Doris Tong 6 , Miguelangelo Perez-Cruet 7 , Noojan Kazemi 8 , Ilyas Aleem 8 , Muwaffak Abulhak 1 , Victor Chang 1

1 Department of Neurosurgery, Henry Ford Hospital, Detroit, United States,

2 Center for Health Policy & Health Services Research, Henry Ford Health Detroit, United States,

3 Corewell Health William Beaumont University Hospital, Royal Oak, United States,

4 University of Michigan Health - West, Wyoming, United States,

5 Corewell Health Beaumont Troy Hospital, Troy, United States,

6 Michigan State University - Ascension Providence Hospital, Southfield, United States,

7 Department of Neurosurgery, Corewell Health William Beaumont University Hospital, Royal Oak, United States,

8 University of Michigan, Ann Arbor, United States

Introduction: Enhanced Recovery After Surgery (ERAS) protocols have been introduced to several surgical specialties with the intention to reduce peri-operative adverse events and improve various health-system quality metrics including length of stay and patient satisfaction. Though this has been studied in the setting of spine surgery, the vast majority are single institution protocols biased toward academic teaching hospitals. There is a paucity of literature studying the utilization of a multi-institutional ERAS protocols that is applicable to community hospitals without associated training programs. We present the Michigan Spine Surgery Improvement Collaborative (MSSIC) two-year experience implementing a multi-institutional ERAS protocol for elective spine surgery. Material and Methods: The MSSIC database was queried for patients across 29 hospitals and three ASCs who underwent elective cervical or lumbar spine surgery from January 2022 to the end of December 2024. Patients were grouped based on whether they underwent the ERAS protocol. The protocol consisted of pre-op measures, including pre-surgical patient education classes and carbohydrate rich drink prior to surgery; intra-op measures including multimodal opioid sparing analgesia; post-operative measures including ambulation within 8 hours of surgery. Outcomes analyzed include patient demographics, complications, and patient-reported outcomes. A multivariate analysis was conducted to control for confounders. Results: 15,633 patients were included (10,816 lumbar patients and 4817 cervical). Lumbar patients who underwent the ERAS protocol were more likely to be discharge home (RR 1.06; CI 1.03-1.09, p < 0.001), return to work within 90 days (RR 1.24; CI 1.01-1.52, p = 0.04), and achieve MICD in PROMIS functional score at 90 days (RR 1.11; CI > 1.00-1,24, p = 0.04). Notably, lumbar patients undergoing the ERAS protocol were less likely to have a prolonged hospital stay (RR 0.82; CI 0.72-0.93, p = 0.003). Cervical patients who underwent the ERAS protocol were more likely to be discharged home (RR 1.11; CI 1.07-1.15, p < 0.001) and were less likely to have a prolonged length of stay (RR 0.82; CI 0.69-0.98, p = 0.03) or have a readmission within 30 days (RR 0.65; CI 0.43-0.98, p = 0.04). Conclusion: In our 2-year experience implementing an ERAS protocol in elective spine patients, we have demonstrated its feasibility and potential in decreasing length of stay and 30-day readmissions while increasing home discharge, return to work, and functional status. The use of an elective spine ERAS protocol should be considered by health systems performing elective spine surgery as a quality-improvement driven initiative regardless of multi-institutional status.

ID: AGS26/3448

A009: Clinical outcomes of amniotic membrane use following endoscopic lumbar discectomy: a propensity-matched analysis

Dia Halalmeh 1 , Yusuf Ansari 2 , Arwa Alabide 3 , Rahul Kumar 4 , Saqib Hasan 5

1 University of Iowa Hospitals and Clinics, Iowa City, United States,

2 Temple University, Philadelphia, United States,

3 University of Kufa, Kufa, Iraq,

4 University of Miami Miller School of Medicine, Miami, United States,

5 Golden State Orthopedics and Spine, Oakland, United States

Introduction: Endoscopic lumbar discectomy has been demonstrated to successfully relieve or reduce pain due to disc herniation. However, residual pain and recurrent disc herniation are significant unresolved problems. A possible solution is the use of amniotic membranes, which have been shown to reduce inflammation and promote healing. The aim of this study is to compare pain scores, ODI, and recurrent disc herniation for patients following endoscopic discectomy with and without the use of amniotic membranes. Methods: Retrospective review of 333 patients undergoing endoscopic lumbar discectomy between 2019-2024 was performed. Patients were dichotomized into the amniotic group if amniotic membrane tissue was placed into the annular defect following discectomy and control group if the patient did not receive amniotic membranes following discectomy. Patient demographic variables and outcomes of visual analog scale (VAS) for leg pain, back pain, Oswestry disability index (ODI), paresthesia, and recurrent disc herniation were collected retrospectively via chart review. Repeated measures ANOVA was used to compare pain scores and ODI between amniotic and control group over time. Multivariable logistic regression was used to identify risk factors of disc re-herniation. Results: Of the 333 patients, 100 (30.0%) were in the amniotic group and 233 (70.0%) were in the control group. Mean operative times were not significantly different between the group (86.8 vs 85.3 mins, p = .69). Mean ODI was significantly different between amniotic and control groups at 12 months (9.1 vs 19.8, p = .003, respectively). VAS leg or back pain scores were not significantly different between groups. On multivariable logistic regression analysis, use of amniotic membranes had a decreased risk of disc re-herniation (OR: 0.23 [95% CI: 0.04-0.84], p = .02) and postop dysesthesia (OR: 0.14 [0.01-0.75], p = .01) after controlling for revision surgery, preop ODI, and VAS leg pain score. There was no significant difference in complications (i.e. CSF leak, infection, transient neurological deficits, conversion to open, or epidural hematoma) between the two groups. Conclusion: The use of amniotic membranes following endoscopic discectomy was associated with significantly lower ODI, decreased risk of re-herniation, and dysesthesias as compared with the control group.

OP2: Function and Recovery in ASD

ID: AGS26/171

A010: Vacuum discs are associated with greater sagittal imbalance and disability in adult spine deformity patients

Kira Skaggs 1 , Javier Castro 1 , Jayme Koltsov 1 , Alexa Pius 1 , Taylor Johnson 2 , Matthew Follett 3 , Tanmaya Sambare 4 , John Kleimeyer 1 , Kirkham Wood 1 , Todd Alamin 1 , Lex Mitchell 1 , Geoffrey Riley 1 , Eugene Carragee 1 , Serena Hu 1

1 Stanford University, Palo Alto, United States,

2 Scottish Rite for Children, Dallas, United States,

3 Synergy Orthopedics, San Diego, United States,

4 Los Angeles County Harbor-UCLA Medical Center, Los Angeles, United States

Introduction: Intervertebral Vacuum Discs (VDs) signify advanced degenerative changes in the spine and are frequently observed in older patients. Despite their prevalence, VDs' clinical implications in adult spinal deformity (ASD) remain underexplored. This study assesses the prevalent characteristics of VDs and their impact on spinal alignment, pain, and surgical outcomes within an ASD cohort. Material and Methods: This retrospective analysis included consecutive patients (2013-2019) undergoing surgery for spinal deformity with lumbar spine CT scans from a single institution. Preoperative CTs were evaluated for the presence and size of VDs. Radiographic parameters were measured pre- and postoperatively. Patient-reported outcome measures were systematically collected at the initial preoperative visit and postoperatively at 6 weeks, 6 months, 1 year, and 2 years. Mixed effects models were used to assess outcomes (α = 0.05). Results: The final cohort included 135 patients [60% female, aged 70 (68, 72) years]. VDs were present in 86% of patients. Preoperatively, patients with VDs experienced greater leg pain [+0.9 (0.3, 1.6) vs. no VD; p = 0.004]. Patients with large VDs had higher preoperative back pain vs. those with small VD [+1.0 (0.5, 1.6); p < 0.001] or no VD [+0.6 (0.1, 1.1); p < 0.027]. Patients with VDs exhibited significant sagittal imbalance, lower preoperative lumbar lordosis [-6.3 (95% CI: 8.8, 3.8) vs. no VD; p < 0.001], greater LL-PI mismatch [+4.9 (95% CI: 2.4, 7.4) vs. no VD; p < 0.001], and lower SL [-2.0 (95% CI: -3.6, -0.5) vs. no VD; p < 0.001]. Surgical interventions, particularly Anterior Lumbar Interbody Fusions (ALIFs), were more frequently utilized at levels with VDs (ALIFs: 17.5% VD vs 6.5% no VD, p < 0.001). ALIFs were associated with a greater change in segmental lordosis than AntLat [+7.0 (2.5, 11.5); p < 0.001] and no interbody fusion [+8.8 (4.4, 13.1); p < 0.001] when a VD was present. Despite VDs being associated with greater pain and deformity preoperatively, surgical outcomes were independent of VD presence, size, or number. Larger improvements in NPRS leg pain were observed when a VD was present [-0.91 (95% CI: -1.63, -0.18) vs. no VD; p = 0.014]. Large VDs were also associated with larger decreases in back pain at 6 weeks post-op versus small VD [-1.3 (-2.0, -0.6); p < 0.001] and no VD [-1.2 (-1.8, -0.6); p < 0.001]. VDs at any level were associated with larger pre- to postoperative increases in LL [+3.4 (95% CI: 0.7, 6.0) vs. no VD, p = 0.015] and larger decreases in PI-LL mismatch [-4.0 (95% CI: -7.1, -1.0) vs. no VD; p = 0.008] in univariate analysis. Conclusion: This study establishes the near ubiquity of VDs in ASD surgical patients (86%), higher than previously reported in other degenerative diagnoses. Preoperatively, patients with VDs experienced greater back and leg pain, and exhibited greater sagittal imbalance. Despite their worse preoperative condition, ASD patients with VDs achieved similar outcomes to those without VDs. The larger improvements in LL and PI-LL mismatch were largely attributable to increased utilization of interbody fusions, particularly ALIFs, at VD levels.

ID: AGS26/1535

A011: Upper thoracic spine alignment: the missing piece of the alignment puzzle to improve GAP and T4L1PA targets

Nuri Demirci 1 , Yasemin Yavuz 2 , Munish Gupta 1

1 Department of Orthopaedic Surgery, Washington University in St. Louis, Saint Louis, United States,

2 Department of Biostatistics, Faculty of Medicine, Ankara University, Ankara, Turkey

Introduction: Proximal junctional kyphosis (PJK) remains one of the most frequent and clinically significant complications following adult spinal deformity (ASD) surgery. Despite advances in corrective strategies, the durability of alignment remains a challenge. Metrics such as the Global Alignment and Proportion (GAP) score and T4-L1 pelvic angle mismatch (T4L1PA) have been proposed as alignment targets to mitigate PJK risk. However, their predictive performance diminishes once patients fall within target ranges, suggesting that current metrics may inadequately account for upper thoracic sagittal balance. To address this limitation, we evaluated a novel parameter, the T1-T9 pelvic angle mismatch (T1T9PA) hypothesizing that it provides additional discriminatory power and complements existing metrics for PJK prediction. Materials and Methods: A prospective cohort of 207 ASD patients who underwent thoracic-to-pelvis fusion with ≥7 instrumented levels and pelvic fixation, all with ≥ 2 years of follow-up was analyzed. PJK was defined as a ≥ 20° increase in proximal junctional angle from baseline or revision surgery for PJK. Demographic and surgical characteristics, including age, body mass index (BMI), sex, level of upper instrumented vertebra (UIV), and osteotomies, were recorded. Six-week postoperative radiographs were assessed for GAP score, T4L1PA, and T1T9PA. GAP scores were categorized as proportioned (≤ 2), moderately disproportionate (3-6), or severely disproportionate (≥ 7). T4L1PA was classified as within target (-1° to +3°) or off target. Receiver operating characteristic (ROC) analysis determined the optimal cutoff for T1T9PA. Associations with PJK were analyzed using chi-square testing and multivariate logistic regression. Results: The mean patient age was 62.1 ± 11.1 years, mean BMI 29.7 ± 5.7, and 77.8% were female. Upper thoracic UIVs (≤ T6) were present in 72.9% of cases, pedicle subtraction osteotomies in 20.8%, and vertebral column resections in 6.3%. Over a mean follow-up of 46 months (range 24-113), 32 patients (15.5%) developed PJK. PJK incidence increased across GAP categories: 9.6% in proportioned, 17.9% in moderately disproportionate, and 21.7% in severely disproportionate patients, though this trend was not statistically significant. ROC analysis identified a T1T9PA cutoff of 10° as optimal. Stratification showed that mismatch ≥ 10° was strongly associated with higher PJK rates in moderately (27.0% vs. 9.8%, p < 0.05) and severely disproportionate groups (37.5% vs. 4.5%, p < 0.05), but not in proportioned patients (18.2% vs. 8.3%, p > 0.05). T4L1PA within target was associated with lower PJK incidence (9.2%) compared to off-target alignment (20%). Among off-target patients, PJK risk increased from 9.4% with mismatch < 10° to 32.1% with mismatch ≥ 10° (p < 0.05). In contrast, among those within target T4L1PA, the difference between mismatch < 10° (7%) and ≥ 10° (18.8%) did not reach statistical significance. Conclusion: In this long fusion to the pelvis ASD cohort, PJK developed in 15.5% of patients during long-term follow-up. While GAP score and T4L1PA provided limited discriminatory power, a T1T9PA mismatch ≥10° was independently associated with increased PJK risk, particularly in patients with disproportionate GAP or off-target T4L1PA. These findings highlight the importance of upper thoracic alignment in sagittal balance and suggest that T1T9PA may serve as a robust alignment target for surgical planning aimed at reducing proximal junctional complications.

ID: AGS26/1563

A012: Functional gait analysis in adult spinal deformity: insights from butterfly diagram and three-foot-zone evaluation

Jung-Hee Lee 1 , Ki Young Lee 1 , Woo Jae Jang 1 , Hong-sik Park 1

1 Kyung Hee University Hospital, Seoul, South Korea

Introduction: Adult spinal deformity (ASD) is characterized by progressive sagittal malalignment, resulting in impaired posture, mobility limitations, and reduced quality of life. While radiographic parameters such as the SRS-Schwab classification and the Global Alignment and Proportion (GAP) score remain the cornerstone of diagnosis and surgical planning, they cannot adequately capture functional deficits, particularly gait disturbances. Lumbar degenerative kyphosis (LDK), redefined as “drop body syndrome,” represents a homogeneous ASD subtype associated with severe sagittal imbalance and stooped posture. This study aimed to quantify gait characteristics in LDK patients using a treadmill-based pressure sensor system, focusing on center-of-pressure (CoP) trajectories, butterfly diagram visualization, and three-foot-zone plantar pressure analysis. Material and Methods: This retrospective study included 371 patients with LDK (mean age 71.4 ± 6.1 years) who underwent corrective surgery, compared with 72 healthy controls (mean age 57.5 ± 13.1 years). Inclusion criteria were sagittal malalignment (sagittal vertical axis > 50 mm, pelvic incidence-lumbar lordosis mismatch > 10°, pelvic tilt > 25°), paraspinal muscle atrophy on imaging, and clinical features consistent with LDK. Gait assessments were conducted preoperatively using a treadmill-based pressure sensor system, recording step length, stride length, step width, stance phase, cadence, and walking velocity. CoP trajectories were visualized as butterfly diagrams, and three-foot-zone analysis quantified weight transfer from heel to forefoot. Results: Compared with healthy controls, LDK patients showed significantly reduced step and stride length, cadence, and walking velocity, alongside increased step width, stance phase, and stride time. Swing phase and single-limb support were markedly decreased. CoP analysis revealed shortened gait line and single-limb support line, diminished anterior-posterior excursion, and increased variability, producing a characteristic “flattened linear” butterfly diagram with preserved lateral symmetry but impaired centralization. Three-foot-zone analysis demonstrated shortened heel-to-forefoot transfer time, increased midfoot loading, and significantly reduced forefoot pressure, indicating impaired weight progression and push-off mechanics. Conclusion: Butterfly diagram and three-foot-zone analysis provide distinctive quantitative and visual markers of gait dysfunction in LDK patients, beyond conventional radiographic assessment. The findings highlight midfoot-dominant gait strategies, reduced forefoot push-off, and flattened linear CoP trajectories as disease-specific features of LDK. These tools may serve as valuable adjuncts for functional evaluation, diagnosis, and rehabilitation planning in ASD.

ID: AGS26/1648

A013: Effect of perioperative sarcopenia treatment on the development of proximal junctional disease in adult spine deformity

Yücel Bilgin 1 , Saltuk Buğra Güler 2 , Merve Çetin 3 , Selcan Akesen 4 , Burak Akesen 1

1 Department of Orthopedics and Traumatology, Bursa Uludag University, School of Medicine, Bursa, Turkey,

2 Department of Orthopedics and Traumatology, Mamak State Hospital, Ankara, Turkey,

3 Department of Nutrition and Dietetics, Bursa Uludag University, School of Medicine, Bursa, Turkey,

4 Department of Anesthesiology and Reanimation, Bursa Uludag University, School of Medicine Bursa, Turkey

Introduction: This study aimed to determine the effect of perioperative sarcopenia treatment on the development of postoperative proximal junction disease in elderly adult patients undergoing spinal deformity surgery. Material and Methods: Patients who underwent surgery with the diagnosis of adult lumbar degenerative deformity were included in the study. Inclusion criteria were determined as being 60 years of age or older, having a diagnosis of sarcopenia, having had 3 or more levels of long-level instrumentation, not having received any previous treatment for sarcopenia, and having a minimum of 1-year postoperative follow-up. Patients were evaluated for sarcopenia according to the Asian Criteria. DEXA was used to determine muscle mass. 6-meter walking speed and 5-times sit-to-stand test were used to determine muscle functions. Lumbar paraspinal muscle mass and psoas muscle mass were measured using lumbar MRI (magnetic resonance imaging). Patients were divided into two groups as those who received perioperative sarcopenia treatment and those who did not. Nutritional regulation, protein-supplemented diet and resistance exercises were used in sarcopenia treatment (preoperative 3 months and postoperative 9 months). Patients were compared in two groups in terms of proximal junction disease development and clinical scores (VAS and ODI). Results: A total of 84 patients were included in the study. Perioperative sarcopenia treatment was applied to 35 patients (3M, 32F). Perioperative sarcopenia treatment was not applied to 49 patients (9M, 40F). In the group of treated patients, 2 patients developed proximal junction disease. In the untreated group, 9 patients developed postoperative proximal junction disease. In the statistical comparison, the difference was determined to be significant (p = 0.012). Postoperative 1st year clinical scores (VAS and ODI) were significantly better in the sarcopenia treatment group (p = 0.005 and p = 0.032). Conclusion: According to our study, perioperative sarcopenia treatment (protein-rich diet and exercise) increases muscle mass based on DEXA and radiological imaging, significantly reduces the development of proximal junctional disease after advanced age spinal deformity surgery, and positively affects postoperative 1st year clinical results (VAS and ODI scores).

ID: AGS26/1057

A014: Mobilization matters: early versus late ambulation after three column osteotomy in adult spinal deformity - Predictors and outcomes

Christopher Lai 1 , Bradley Hammoor 1 , Joshua Coan 1 , Benjamin Varieur 1 , Jordan Gasho 1 , Theresa Chua 1 , Marcos Gonzalez 1 , Stuart Hershman 1

1 Massachusetts General Hospital Department of Orthopaedic Surgery, Boston, United States

Introduction: Early ambulation is associated with improved outcomes and shorter length of stay (LOS) after spine surgery. Three column osteotomies (3CO) are useful in addressing rigid spinal deformity in adult spinal deformity (ASD) but increase complications and risk. Predictors of delayed ambulation and the impact of late ambulation after 3CO on the postoperative course are not known. Therefore, the purpose of this study is twofold: 1) to investigate predictors of delayed ambulation and 2) to assess the impact of early vs. late ambulation on patient outcomes after 3CO. Material and Methods: Patients undergoing 3CO for ASD from 2016 to 2023 in a large academic hospital system were included and stratified by early ambulation (EA) or late ambulation (LA). LA was defined as ambulating after postoperative day 2. Perioperative outcomes, complications, and patient-reported outcome measures (PROMs) were analyzed. Multivariable regressions were performed with length of stay, discharge to facility, and inpatient complications as outcome variables. A mixed-effects model assessed within- and between-group improvements in PROMs over time. Significance level was set at p < 0.05. Results: 198 patients met criteria for analysis. BMI [29.6 (7.6) vs. 28.2 (6.7), p = 0.017], levels fused [10 (6.5) vs. 9 (5), p = 0.002], EBL [2625 (2000) vs. 1800 (1725), p < 0.001], and operative time [464 (182) vs. 416 (118), p < 0.001] were higher in LA. Postoperatively, LA had significantly higher rates of ICU admission [45.9% vs. 13.2%, p < 0.001], inpatient complications [76.8% vs. 39.7%, p < 0.001], RBC transfusion [64.8% vs. 46.1%, p = 0.012], LOS [9 (5) vs. 5 (3), p < 0.001] and discharge to facility [79.5% vs. 33.8%, p < 0.001] than EA. On multivariable regression, ICU admission [3.71 (1.70-8.72), p = 0.002] and operative time ≥ 448 minutes [2.28 (1.12-4.66), p = 0.023] were significant predictors of LA. LA was a significant independent predictor of increased LOS [OR 2.4 (1.2-4.1), p < 0.001], discharge to facility [OR 7.60 (3.21-19.3), p < 0.001], and inpatient complication [OR 3.84 (1.81-8.30), p < 0.001] when adjusted for age, sex, comorbidities, and surgical characteristics. On mixed effects modeling, EA had significant improvements in Global Physical scores from baseline to 1 year [3.23 (0.57-5.89), p = 0.014] and baseline to 2-years [3.75 (1.03-6.47), p = 0.004], Global Mental scores from baseline to 2-years [2.41 (0.33-5.15), p = 0.02], and Physical Function scores from baseline to 1-year [3.80 (1.38 - 6.22), p = 0.001] and baseline to 2 years [0.34 (1.93 - 2.61), p = 0.002]; LA only experienced significant improvement in Global Physical function from baseline to 1 year [2.84 (0.55-5.12), p = 0.009]. Finally, LA experienced significantly less improvement in PROMIS Physical Function scores from baseline to 1-year (p = 0.017) and baseline to 2-year (p = 0.018) compared to EA patients. Conclusion: ICU placement and increased operative time are associated with late ambulation after 3CO surgery. Late ambulation is predictive of increased length of stay, inpatient complications, and discharge to a facility. Late ambulation is associated with inferior PROM improvements up to 2 years postoperative. Routine ICU admission after 3CO should be minimized, and early mobilization should be emphasized to optimize hospital course and patient recovery.

ID: AGS26/1774

A015: Biomechanical advantages of antero-posterior accessory rods configuration in 4-rod construct after pedicle subtraction osteotomy

Daniela Bruno 1 , Chiara Cini 2 , Giulia Cavazzoni 1 , Gaia Youssef 1 , Martina Bellavita 1 , Bruna Maccaferri 2 , Giovanni Ciani 2 , Alessandro Gasbarrini 2 , Luca Cristofolini 1 , Francesco Vommaro 2 , Marco Palanca 1

1 Alma Mater Studiorum - University of Bologna, Bologna, Italy,

2 IRCCS Rizzoli Orthopaedic Institute, Bologna, Italy

Introduction: Pedicle Subtraction Osteotomy (PSO) is an invasive surgical technique used to restore sagittal alignment in severe adult spinal deformities [1,2]. However, the risks of post-operative complications, such as pseudoarthrosis [3,4] and rod breakage [4,5], are still high. Multi-rod constructs, particularly 4-rod configurations, have been introduced to improve stability [3,4] and prevent rod failure [6,7]. To date, however, no study has investigated the biomechanical impact of positioning of the accessory rods relative to the primary rods. The aim of this study was to compare a conventional 4-rod construct with accessory rods in medio-lateral configuration (ML) versus an alternative 4-rod construct with accessory rods in antero-posterior (AP) configuration, focusing on construct stiffness, and rod deformation. Material and Methods: Five thoracic porcine spine segments (T5-T11) were used to grant similarity among the specimens and reduce confounding factors. The extreme vertebrae were embedded in poly-methyl methacrylate (PMMA), to be mounted in a testing machine. A PSO was performed at T8 of each segment, followed by posterior fixation (two levels above and two levels below the osteotomy) using a 4-rod construct (two primary rods and two accessory rods). Accessory rods were first placed in ML configuration (i.e. positioned laterally and parallel to the primary rods) through side-to-side connectors. Biomechanical tests were performed with an electrodynamic spine simulator (MIB4.0-MAS2-S, MIB, Italy) applying pure moments in axial rotation (AR) ( ± 7.5 Nm) and flexion (F) (+7.5 Nm). After testing in the ML configuration, the accessory rods were positioned in antero-posterior (AP) configuration (i.e. placed posteriorly to the primary rods within the coronal plane) and testing was repeated. Stiffness was evaluated in the two configurations. Mechanical deformations of the primary rods were measured at 5Nm using triaxial strain gauge rosettes bonded posteriorly at the osteotomy level. Comparisons between ML and AP were analyzed with paired t-tests (p < 0.05). Results: Stiffness between did not change significantly between the two configurations (ML vs. AP) in both axial rotation and flexion (AR: 0.81 ± 0.17 Nm/° vs 0.81 ± 0.18 Nm/°, F: 1.49 ± 0.26 Nm/° vs 1.46 ± 0.30 Nm/°). In axial rotation, the maximum and minimum principal deformations of the primary rods showed a slight increase in AP compared to ML, without statistically significant differences (maximum: +9.9%; minimum: +14.8%). Conversely, in flexion, deformations of the primary rods decreased significantly in AP, with nearly half the magnitude of ML (maximum: -47%, p = 0.027; minimum: -45%, p = 0.029). Conclusion: In terms of stiffness, no relevant differences were found between the two configurations, suggesting that spinal mobility is not substantially affected and that the potential for bone healing and osteointegration is preserved. Conversely, the AP configuration led to a marked reduction in deformation in the primary rods during flexion, which may lower the mechanical demand on the instrumentation and, consequently, reduce the risk of rod fatigue failure after PSO.

References

1. Dorward et al. Neurosurg Focus 2010

2. Bridwell, et al. Spine 2006

3. Gupta et al. Oper Neurosurg 2018

4. Hyun et al. Spine 2014

5. Smith, et al. J Neurosurg Spine 2017

6. La Barbera et al. Eur Spine J 2018

7. Hallager, et al. Spine 2016

Acknowledgments: HEU-Funded METASTRA project (grant 101080135)

ID: AGS26/1851

A016: Low normalized total psoas area predicts greater change in early stage sagittal alignment following adult spinal deformity surgery

Michael Jian Wen Chen 1 , Han Jo Kim 2 , Robert Uzzo 2 , Stephane Owusu-Sarpong 2 , Andrea Pezzi 2 , Gabrielle Dykhouse 2 , Atahan Durbas 2 , Matthew Cunningham 2 , Francis Lovecchio 2

1 China Medical University Hospital, Taichung, Taiwan,

2 Hospital for Special Surgery, New York, United States

Introduction: To evaluate whether preoperative sarcopenia, defined by low normalized total psoas area (NTPA), predicts postoperative deterioration of sagittal alignment in unfused spinal segments after adult spinal deformity (ASD) surgery. Material and Methods: This study retrospectively reviewed 103 patients who underwent lower thoracic fusion to the pelvis for ASD between 2014 and 2022. NTPA was measured at L3 and L4. Patients were stratified into “Low NTPA” and “Normal NTPA” groups based on validated sex-specific thresholds. Radiographic sagittal parameters were assessed at immediate post-OP, discharge, 6 weeks, 6 months, 1 year, and 2 years postoperatively. Primary outcomes included changes in pelvic tilt (PT), T1 pelvic angle (T1PA), T4 pelvic angle (T4PA), L1 pelvic angle (L1PA), L4 pelvic angle (L4PA), lumbar lordosis (L1-S1), distal lumbar lordosis (L4-S1), thoracic kyphosis (T4-T12), and proximal junctional angle (PJA). Temporal changes within-group were analyzed using paired t-tests with Bonferroni correction. Results: A total of 103 patients (mean age 65.2 years; 33 males, 70 females) were analyzed. Fifty (48.5%) were classified as Low NTPA and 53 (51.5%) as Normal NTPA. Baseline demographics, ASA class, DEXA T-scores, and sagittal parameters were similar, with no difference in UIV distribution. At 6 weeks, Low NTPA patients showed marked thoracic kyphosis progression with increases in T1PA, while Normal NTPA patients maintained alignment. PT increased significantly in the Low NTPA group at 6 months post-OP. T4-L1 PA mismatch was evident for both groups as early as 6 weeks post-OP. By 1 year, gradual sagittal changes were also evident in the Normal cohort. Revision rates for junctional failure were comparable (Low NTPA: 8%; Normal NTPA: 9.4%). Conclusion: Low NTPA predicts early loss of sagittal correction and kyphotic change after ASD surgery, most evident within the first 6 weeks, underscoring the role of paraspinal musculature before fusion. Pelvic retroversion was subsequently recruited to maintain sagittal balance in patients with low NTPA. Muscle status should be incorporated into preoperative risk stratification and optimization.

ID: AGS26/2533

A017: Are we achieving what the computer tells us to? Analysis of lumbar pedicle subtraction osteotomies with pre-bent rods

Rafael Garcia de Oliveira 1 , Kento Yamanouchi 1,2 , Kenneth Nguyen 1 , Renzo Laynes 1 , Takeshi Fujii 1,2 , Iyan Younus 1,3 , Aiyush Bansal 1 , Jean Christophe Leveque 1 , Philip Louie 1 , Rajiv Sethi 4 , Venu Nemani 1

1 Virginia Mason Medical Center, Seattle, United States,

2 Keio University, Tokyo, Japan,

3 Vanderbilt University, Nashville, United States,

4 University of Washington, Seattle, United States

Introduction: Precise restoration of sagittal balance is a critical goal in adult spinal deformity surgery. Computer-assisted planning allows for patient-specific alignment targets and rod pre-bending, theoretically improving the accuracy of surgical correction. However, the correlation between planned and achieved alignment goals using pre-bent rods remains unclear, representing a significant gap in understanding how well current technological tools translate into actual surgical outcomes. This study aims to compare preoperative planning to postoperative spinopelvic parameters in patients undergoing lumbar PSOs using pre-bent rods to identify potential discrepancies and inform future improvements in surgical planning techniques. Materials and Methods: This study is a single-center retrospective analysis that included all patients who underwent planned PSO surgery with ≥ 6 levels of thoracolumbar fusion at our institution between 2018 - 2022. All cases utilized UNID-based preoperative planning for PSO at the same level and used pre-bent rods. Preoperative predicted parameters were compared with actual postoperative spinopelvic alignment, including vertebropelvic angles and PSO segmental angles. The primary outcomes were differences between planned and achieved sagittal vertical axis (SVA), pelvic tilt (PT), sacral slope (SS), thoracic kyohosis (TK), and L1 pelvic angle (L1PA). Results: Of the 51 eligible patients, 20 met the study criteria and were included in the analysis. The mean age was 66.8 years, with 60% female. The PSO level was L3 in 15%, L4 in 45% and L5 in 40% of the patients. The mean fused levels were 12 ± 2.8, number of rods was 5 ± 1.2, the mean difference of the PSO Angle was 0.1° ± 8. The combination of software planning and pre-bent rods inaccurately simulated SVA by an average of 46.5 mm, Pelvic Tilt by 9°, Sacral Slope by 9.4°, Thoracic Kyphosis by 5.2° and L1PA by 2° (p < 0.05). 45% of the patients were out of the ideal L1PA range of ± 2°. Conclusion: While PSOs using pre-bent rods generally achieved the planned PSO angle and lumbar lordosis goal, this study demonstrated significant discrepancies between important parameters of the sagittal alignment. Specifically, PT/SS changes were underestimated, thoracic kyphosis increased more than planned, and SVA correction was underachieved. Ultimately, those discrepancies led to missing some of the most important alignment goals. These findings suggest that current planning tools may require refinement to improve the accuracy of predicted postoperative alignment using pre-bent rods and minimize the risk of suboptimal outcomes.

ID: AGS26/2699

A018: Determining the expected standing pelvic tilt and sacral slope for a given pelvic incidence in healthy asymptomatic individuals

Fatemeh Alavi 1 , Christopher J. Nielsen 1 , Michael Kelly 2 , Jeffrey Hills 3 , Stephen Lewis 1

1 University of Toronto, University Health Network (UHN), Toronto, Canada,

2 Rady Children’s Hospital, University of California San Diego, San Diego, United States,

3 UT Health San Antonio, San Antonio, United States

Introduction: Pelvic parameters such as pelvic tilt (PT) and sacral slope (SS) play a crucial role in sagittal spinal alignment and biomechanical balance for a patient with a fixed pelvic incident (PI). Although PT and SS are dynamic and posture-dependent, their relation in the standing posture was not fully recognized. Understanding the relationship between these variables is essential to evaluate spinal alignment and to plan corrective interventions. In this study, we propose a model that describes PT as a function of SS, adjusted by a PI-dependent factor to account for individual pelvic morphology. This approach offers a more personalized framework for assessing sagittal balance across varying pelvic types. Material and Methods: We reviewed standing full-spine low dose stereoradiographs of an international cross-sectional study on 320 healthy volunteers over 18 years of age from four continents, without evidence of spinal disease or spine surgery. The exclusion criteria were spondylolisthesis, coronal Cobb angle >10°, or an Oswestry disability index >20%. Institutional review board approval was obtained by participating centers. Radiographic parameters including PI, PT, and SS were measured. To facilitate subgroup analysis, individuals were categorized based on their PI values into discrete ranges: <30°, 30–40°, 40–50°, 50–60°, 60–70°, 70–80°, and >80°. Linear regression models were fit and r 2 computed to assess associations between pelvic incidence, pelvic tilt and sacral slope. Results: We realized that both sacral slope and pelvic tilt are correlated strongly and moderately with PI (r 2 =0.58, r 2 =0.45). While there was no association between sacral slope and pelvic tilt when considering all PI values, by categorizing PI values into different subgroups, we found strong correlation between SS and PT within each PI subgroup. A normal PT relative to SS can be described using the equation: PT = –0.8 × SS + k, where k accounts for individual variability in pelvic morphology: k = 0.9 × PI – 3.5. This model demonstrated a coefficient of determination (R 2 ) ranging from 0.60 to 0.99 across different PI ranges, indicating variable but generally strong explanatory power depending on pelvic anatomy. Only 10% of healthy asymptomatic individuals showed a sacral slope exceeding the expected value by over 10 degrees. Conclusion: The proposed PI-dependent model of PT relative to SS provides a quantitative framework to assess normal pelvic orientation across varying sagittal alignments. By establishing expected PT and SS values for a specific PI, clinicians can better distinguish between physiological alignment and compensatory mechanisms in patients with spinal deformity. Understanding the PT-to-SS ratio in the context of an individual's PI can aid in identifying deviations from normative alignment and support

OP3: Epidemiology in Spine

ID: AGS26/1428

A019: Anterior cervical discectomy and fusion versus cervical disc arthroplasty for cervical degenerative disc disease: alarming rates of bias

Joshua Davood 1 , Henry Avetisian 1 , Daniel Rusu 1 , Aidan Lindgren 1 , Marco Napolitano 1 , Joshua Khorsandi 2 , Apurva Prasad 1 , Joseph Borna 1 , Joseph Noorvash 1 , Nicole Kahen 1 , William Karakash 1 , Marc A. Abdou 1 , Jeffrey C. Wang 1 , Raymond Hah 1 , Kiran Alluri 1

1 Department of Orthopaedic Surgery, Keck School of Medicine of the University of Southern California, Los Angeles, United States,

2 Kirk Kerkorian School of Medicine at UNLV, Las Vegas, United States

Introduction: Cervical degenerative disc disease (CDDD) is commonly treated with anterior cervical discectomy and fusion (ACDF) or cervical disc arthroplasty (CDA), but their comparative effectiveness remains debated. Systematic reviews and meta-analyses are crucial for guiding clinical decision-making, yet spin bias-overstating benefits or downplaying limitations-remains prevalent in orthopaedics. We aimed to quantify the prevalence and types of spin and to appraise methodological quality among systematic reviews and meta-analyses comparing ACDF and CDA. Materials and Methods: A systematic search of PubMed, Web of Science, and Embase was conducted in September 2024 to identify systematic reviews and meta-analyses. Inclusion required: meta-analysis or systematic review; CDDD; direct single- or multi-level ACDF vs CDA comparison of treatment; English language. Two independent reviewers screened abstracts and full texts, then assessed the included articles for the nine most spin categories defined by Yavchitz et al. and AMSTAR-2. Journal impact factor, publication year, and yearly/total citations were extracted for each article. Associations between AMSTAR-2 rating, yearly citations, and journal impact factor and the number of types of spin present or presence/absence of spin were assessed using Poisson regression modeling and binary logistic regression modeling, respectively. Results: Of 351 articles, 77 articles met inclusion criteria. Spin was present in 49 (64%) articles; number of spin types per article ranged 0-4. The most frequently observed spin type was Type 5 (the conclusion claims the beneficial effect of the experimental treatment despite a high risk of bias in primary studies), found in 30 (39%) of the articles. All spin-positive articles favored CDA. AMSTAR-2 assessment rated 1 (1%) articles as high, 0 as moderate, 12 (16%) as low, and 64 (83%) as critically low. Journal impact factors ranged from 0.3-10.1 (mean 2.53); yearly citations ranged from 0-18.71 (mean 4.65). Binary logistic regression models found AMSTAR-2 rating significantly predicted presence of spin (OR: 0.182 [0.048-0.689], p = 0.012) while yearly citations and impact factor did not (both p > 0.05). Lastly, Poisson regression models found yearly citations and impact factor did not significantly predict amount of spin (both p > 0.05) while AMSTAR-2 rating fell just short of significant correlation (IRR: 0.517 [0.261-1.023], p = 0.058). Conclusion: Spin bias was strongly apparent in CDA vs ACDF systematic reviews and meta-analyses. Type 5 was the most prevalent spin type found, indicating conclusions often overlooked uncertain evidence. AMSTAR-2 ratings emphasize the need for improving and standardizing methodology in orthopaedic research. Lower AMSTAR-2 ratings were associated with presence of spin, whereas journal impact factor and yearly citations were not-suggesting editorial prestige and popularity do not safeguard against biased framing. Spin-positive articles uniformly favored CDA, suggesting there may be external influences such as industry funding driving bias. Limitations include spin classification subjectivity and restriction to English-language reviews. Clinicians should interpret abstract claims cautiously and appraise full texts. Greater transparency will reduce misinterpretation and better support decisions between CDA and ACDF.

ID: AGS26/1738

A020: SIRIS Spine, the Swiss National Spinal Implant Registry: design and early findings

Thorsten Jentzsch 1 , David Bellut 2 , Ralph T Schaer 3 , Daniel Haschtmann 4 , Emin Aghayev 5

1 Balgrist University Hospital, University of Zurich, Zurich, Switzerland,

2 University Hospital Zurich, Zurich, Switzerland,

3 Inselspital, Bern University Hospital, Bern, Switzerland,

4 Schulthess Klinik, Zurich, Switzerland,

5 Eurospine, Uster, Switzerland

Introduction: Practice patterns in spine surgery vary substantially, underscoring the need for high-quality, real-world data. National implant registries can track outcomes and trends to guide clinical and policy decisions. SIRIS Spine is one of the first mandatory, nationwide spinal implant registries, established in Switzerland to enhance transparency and quality in spine care. The objective of this study is to describe the design and early findings of the registry. Material and Methods: SIRIS Spine is operated by EUROSPINE under the SIRIS Foundation with data services from NEC Software Solutions. It is endorsed by Swiss professional societies and has been embedded in the national quality framework (ANQ) since 2021. All hospitals must report eligible procedures and cover registration costs. Governance is provided by a steering committee and a scientific board. Collected variables include patient demographics, surgical details, and implant characteristics; electronic patient-reported outcome measures are optional. Participating centers receive regular benchmarking reports. Results: Initial inclusion (2021) targeted adults undergoing one- to two-level posterior lumbar fusion for degenerative or spondylolisthetic disease and revision procedures; in 2022, the scope expanded to thoracolumbar osteoporotic vertebral augmentations. From 2021-2024, 18,386 surgeries were recorded from 91 hospitals, involving 16,770 patients and 117,276 implants from 67 manufacturers. The leading indication was degenerative disease (60%), followed by osteoporotic fractures (14%) and non-degenerative spondylolisthesis (5%); mean age was 67 (standard deviation 14) years. Transforaminal lumbar interbody fusion was the most common technique (58%). The overall reoperation/revision rate was 7%, with higher rates among older, obese, and higher-risk (American Society of Anesthesiologists class) patients. Conclusion: A mandatory nationwide spinal implant registry is feasible and yields actionable insights. Early SIRIS Spine data demonstrate the value of structured, longitudinal capture for monitoring care quality. Priorities include streamlining inclusion criteria and strengthening data completeness and reliability to support wider clinical application and international benchmarking.

ID: AGS26/177

A021: A critical appraisal of reporting bias in systematic reviews and meta-analyses addressing AI and machine learning in spine surgery

Kenneth Nguyen 1 , Sydney Pattison 1,2 , Isabel Iral 1,2 , Madison Alessi 1,2 , Kento Yamanouchi 1,3 , Philip Louie 1

1 Virginia Mason Medical Center, Seattle, United States,

2 University of Washington, Seattle, United States,

3 Keio University, Tokyo, Japan

Introduction: Artificial intelligence (AI) and machine learning (ML) are rapidly transforming spine surgery by enabling advancements in diagnosis, surgical planning, intraoperative guidance, and postoperative outcome prediction. Despite their increasing integration, the quality of evidence supporting these technologies and the accuracy with which results are communicated remain uncertain. In particular, systematic reviews and meta-analyses synthesizing AI/ML applications in spine surgery may be prone to reporting bias in the form of spin, which can exaggerate benefits, understate limitations, and ultimately mislead clinical decision-making. To date, no study has comprehensively evaluated the prevalence of spin and the methodological quality of these reviews. The purpose of this study was to critically appraise systematic reviews and meta-analyses on AI and ML in spine surgery, assessing for spin and overall reporting quality. Methods: This study was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) guidelines. Peer-reviewed systematic reviews were collected from PubMed, Scopus, and Embase databases using the search (“AI” OR “artificial intelligence” OR “machine learning”) AND (spine surgery) AND (“systematic review” OR “meta-analysis”). The abstracts were assessed for the 15 most common types of spin. A Critical Appraisal Tool For Systematic Reviews (AMSTAR 2) was used to evaluate the quality of the studies. Several study characteristics were analyzed including PRISMA adherence, year of publication, and Level of Evidence. The association of these characteristics with the presence of spin, as well as the number of spin types present, was determined using t-tests, analysis of variance (ANOVA), Fischer Tests, and Spearman’s Rank Coefficients. Results: The initial database search identified 273 studies, of which 89 duplicates were removed. An additional 121 studies were removed after title and abstract screening because they did not meet the inclusion criteria. The remaining 60 studies were assessed for eligibility after which a total of 38 studies were included. Spin was present in the abstract of 32 out of 38 studies (84.2%). Each type of spin was observed in at least one study’s abstract with the exception of spin type 4. The three most common types of spin were type 3 (15/38, 39.47%), followed by type 5 (13/38, 34.21%), and type 9 (15/38, 39.47%). Lower journal impact factors and lower AMSTAR 2 confidence ratings were both significantly associated with the presence of spin (p < 0.05). Conclusion: This study addresses the lack of evaluation of reporting bias in systematic reviews and meta-analyses of AI and ML in spine surgery. Our findings reveal a high prevalence (84.2%) of spin in abstract reporting, with common types including overstatement of benefits and underreporting limitations. The presence of spin was significantly associated with lower journal impact factors and reduced methodological quality. These results underscore the need for improved transparency and rigor in research synthesis to ensure accurate communication of AI/ML applications. Addressing these biases is crucial for reliable clinical decision-making and the advancement of evidence-based spine surgery

ID: AGS26/2177

A022: Utilization patterns and clinical outcomes of spinal cord stimulators for Parkinson's disease patients: a national claims database analysis

Sujai Arakali 1 , Kashif Qureshi 1 , Jose Anzueto 1 , Rahul Jayaram 1 , Fabrizio Darby 1 , Khaled Alok 1 , Carlos Goulart 1 , Zion Zibly 1 , Mihir Gupta 1

1 Yale University School of Medicine, New Haven, United States

Introduction: Chronic pain affects 60-80% of Parkinson's disease (PD) patients, significantly impacting quality of life. Spinal cord stimulation (SCS) has emerged as a potential therapeutic option for managing chronic pain in PD patients. However, limited large-scale data exists regarding utilization patterns, demographic characteristics, and clinical outcomes of SCS for PD patients. This study aims to evaluate national trends in SCS utilization among PD patients and assess associated clinical outcomes compared to non-PD SCS recipients. Methods: Using the PearlDiver Mariner claims database (2010-2022), we identified PD patients (ICD-9: 332.0, ICD-10: G20) above the age of 50 who received permanent SCS implantations (CPT: 63655). Three cohorts were analyzed: PD patients with SCS (PD-SCS, n = 1,008), PD patients without SCS (PDnoSCS, n = 1,581,155), and SCS patients without PD (SCSnoPD, n = 52,721). Demographic characteristics, comorbidities, and clinical outcomes were compared using chi-square tests and t-tests. Propensity score matching was performed using age, gender, Charlson Comorbidity Index (CCI), coronary artery disease, diabetes, obesity, and tobacco use as covariates. Logistic regression was used to identify predictors of major complications within 60 days and revision surgery within 2 years. Results: SCS utilization among PD patients increased significantly from 2010-2022, with the PD-SCS/total SCS ratio rising from 0.56% to 1.88% with notable increases in utilization observed from 2010-2011 (58% increase) and 2014-2015 (57% increase). PD-SCS patients were older (mean age 71.1 vs 65.5 years, p < 0.001) and had higher comorbidity burden (CCI 3.26 vs 2.40, p < 0.001) compared to SCSnoPD patients. After propensity score matching, no significant differences were found between PD-SCS and SCSnoPD patients in major complications within 60 days (p = 0.619) or 2-year revision rates (p = 0.104). Logistic regression identified that 30-day and 60-day readmissions were strong predictors of 60 day major complications (p < 0.001). For 2-year revision surgery, female gender (p = 0.046), healthcare-related social determinants (p = 0.041), and 30-day and 60-day readmissions were significant predictors (p < 0.001). Conclusion: This claims database analysis reveals increasing utilization of SCS among PD patients over the past decade, with PD-SCS patients demonstrating higher age and comorbidity burden compared to non-PD SCS recipients. Propensity score matching showed equal complication rates between the two groups, indicating no significant additional risk for performing the procedure in PD patients compared to non-PD patients. While age and readmission rates predict short-term complications, gender, healthcare-related socioeconomic factors, and readmission rates significantly influence long-term revision rates. These findings provide valuable insights for clinicians considering SCS as a pain management strategy in PD patients and highlight the importance of comprehensive preoperative assessment including healthcare socioeconomic factors and readmission risk.

ID: AGS26/2832

A023: Trends in robot-assisted spinal fusion surgeries: an analysis by age group, race, ZIP Code income quartile, and urban-rural classification

Connor Burke 1 , Matthew Blackwell 2 , Andrew Brumett 2 , Anoop Chinthala 2 , Arshjeet Singh 2 , Raj Lavadi 2 , Gordon Mao 2

1 Shanghai Medical College of Fudan University, Shanghai, China,

2 Indiana University School of Medicine, Indianapolis, United States

Introduction: Recent literature has suggested that robot assistance in spinal fusion surgery provides enhanced perioperative outcomes. With the potential benefits new surgical techniques can have, it is important that trends in adoption be identified to provide guidance for future efforts in surgical resource allocation and promoting equity. This study analyzes the trends by age group, race, urban-rural location, primary payer, and median household income of patient’s ZIP code in robot-assisted spinal fusion surgeries. Material and Methods: A retrospective serial cross-sectional analysis was performed on the 2010-2021 National Inpatient Sample (NIS). Age group, race, urban-rural classification, primary payer, and median household income of patient’s ZIP code were analyzed for 34,911 cases of robot-assisted spinal fusion surgery in comparison to 5.19 million cases of free-hand spinal fusion surgery. Survey-weighted Poisson regression with a difference in differences comparison was used to evaluate whether demographic changes occurred across sequential three-year epochs. Results: Compared with free-hand, robotic representation showed a decline in central-metropolitan cases (DiD RR 0.55, 95% CI 0.46-0.66, p < 0.001) and uptake in small-metropolitan counties (RR 2.78, 1.05-7.37, p = 0.040) as well as among Medicaid-insured patients (RR 6.53, 1.02-42.01, p = 0.048). Pediatric cases showed a steep decline in the robotic cohort (RR 0.20, 0.11-0.37, p < 0.001). Robotic cases showed no significant association between race and robotic utilization after accounting for age group, urban-rural classification, primary payer, and median household income of patient’s ZIP code. Median household income of the patient’s ZIP code also displayed no significant association with robotic utilization after adjusting for covariates. Conclusion: The present study indicates a change in the population utilizing robot-assisted spinal fusion surgeries. These trends are reflective of diffusion of this new technology. Geographically, the decline in central metropolitan utilization alongside the greater uptake in smaller metropolitans suggests shifts from urban-based teaching hospitals to community hospitals. Financially, increased Medicaid representation demonstrates broader acceptance of robotics in patient care, rather than confinement to privately insured patients. Furthermore, robotic expansion occurred without evidence of worsening disparities by race or income. Among the low volume of pediatric cases that occurred, the disproportionate decline of robotic usage may reflect concerns of safety and the appropriateness of this technology for younger patients.

ID: AGS26/343

A024: Investigation of the normative values of spinal sagittal parameters in the pediatric population

Osman Ceylan 1 , Muhammet Kocaoğlu 1 , Cem Pekgül 2 , Cem Karabulut 1 , Deniz Kargin 1 , Ahmet Keskin 1 , Ali Öner 1

1 Baltalimanı Kemik Hastalıkları Hastanesi, Istanbul, Turkey,

2 Gülhane Eğitim Araştırma Hastanesi, Ankara, Turkey

Introduction: There are relatively few studies in the literature addressing normative values of sagittal spinal parameters in childhood and adolescence. The aim of this study was to investigate sagittal spinal parameters and establish normative values in a normal pediatric population aged 1-18 years. Materials and Methods: We retrospectively reviewed radiographic data of individuals who underwent full-spine orthoradiographic evaluation for various indications at our institution between January 2010 and July 2023. A total of 394 individuals (189 males, 205 females) were analyzed, and sagittal plane parameters were comprehensively evaluated. Patients aged 1 and 2 years were excluded due to inadequate imaging quality. Considering the growth velocity in the pediatric population as well as statistical data distribution, subjects were categorized into four age groups: 3-5, 6-10, 11-15, and 16-18 years. Results: In the overall pediatric population, mean values were as follows: Cervical lordosis (CL, C2-C7) 14.30° ± 10.17, Thoracic kyphosis (TK, T4-T12) 37.12° ± 9.00, Lumbar lordosis (LL, L1-S1) 53.38° ± 10.40, Pelvic incidence (PI) 42.08 9.22° ±, Pelvic tilt (PT) 5.73° ± 7.42, and Sacral slope (SS) 36.43° ± 7.41. Additionally, Spino-Sacral Angle was 126.25° ± 8.46, Spino-Cranial Angle 79.41° ± 11.60, T1 slope 27.08° ± 7.90, Oc-C2 angle 20.41° ± 9.97, and C1-C2 angle 30.92 ± °9.73. Beyond these overall averages, the age- and group-specific variations of each parameter are also presented in our study. Conclusion: Sagittal alignment has been increasingly investigated over the past two decades, and its clinical significance continues to expand. Establishing normative values of sagittal parameters is crucial for maintaining spinal health and optimizing deformity management. Defining reference values plays a fundamental role in treatment planning for spinal deformities such as scoliosis, in the development of surgical strategies, and in the optimization of conservative methods, particularly brace therapy.

ID: AGS26/217

A025: The epidemiology of elective sacroiliac joint fusion: a study of clinical pain pathways

William Adams 1 , Jefferson Waters 1 , Kaitlyn Maniscalco 1 , Ara Khoylyan 1 , Michael Pheasant 2 , Tan Chen 3

1 Geisinger Commonwealth School of Medicine, Scranton, United States,

2 Geisinger Northeast Orthopaedic Residency, Wilkes-Barre/Scranton, United States,

3 Geisinger Musculoskeletal institute, Division of Orthopaedic Spine Surgery, Wilkes-Barre/Scranton, United States

Introduction: The frequency of sacroiliac joint fusion (SIJF) is rapidly rising, with the yearly volume increasing by an average of 33.5% between 2010 and 2021. This rate is expected to grow further at 53.3% over the next four years. While several studies have investigated the etiologies of sacroiliac joint (SIJ) pathology, the epidemiology, clinical course, and demographics of patients with refractory SIJ pain leading to elective SIJF have not yet been elucidated. This study aimed to (1) describe the distribution of common pain pathways and clinical courses leading to elective SIJF and (2) describe the demographics of patients who have undergone elective SIJF. Material and Methods: A retrospective review was conducted identifying 323 consecutive SIJFs performed between 2011 and 2024 at a large integrated healthcare system. Demographic data and clinical history were collected. Patients were assigned to common pain pathways. Each pain pathway represented a distinct clinical course leading to the development of refractory SIJ pain and subsequent SIJF. Descriptive statistics were performed. Results: Two hundred seventy patients were included in the study. One hundred eighty-three (67.8%) were female and 87 (32.2%) were male. The average patient age was 53.7 ± 13.0 years old. The average BMI was 30 ± 5.8. One hundred forty-five (55%) patients had a history of tobacco use. The most common pain pathway included 128 (47%) patients with new onset refractory and injection-confirmed unilateral SIJ pain leading to SIJF. Forty-three (16%) SIJF patients had previous lumbar fusion not involving the lumbosacral junction. Thirty-four (13%) SIJF patients had prior single-level L5/S1 fusion, and 31 (12%) patients had prior multi-level lumbar fusion including the L5/S1 junction. Thirty-four (13%) patients had prior contralateral SIJF. Two hundred and seventeen (81%) patients ended the study with a unilateral SIJF. Fifty-three (19%) patients ended the study with bilateral SIJF, of which 19 (6%) had lumbar fusion prior to their initial SIJF. One hundred forty-two (53%) patients had prior single-level lumbar fusion, multi-level lumbar fusion, or contralateral SIJF. Conclusion: Diverse pain pathways and clinical courses exist for patients undergoing elective sacroiliac joint fusion for refractory and injection-confirmed SIJ pain. A majority of SIJF patients had a previous history of lumbar fusion or contralateral SIJF. Female patients predominated in the study population at a ratio of 2:1, and a majority of patients undergoing elective SIJF had a history of smoking. This study provides further insight into the epidemiology of SIJ pathology and the demographics of patients who undergo elective SIJF.

ID: AGS26/760

A026: The relationship between smartphone screen time and cervical spondylosis: a cross-sectional study

Siravich Suvithayasiri 1 , Warayos Trathitephun 1 , Jackapol Kamolpak 1 , Rathapoom Suwanaratana 1 , Monchai Ruangchainikom 2

1 Chulabhorn Hospital, Chulabhorn Royal Academy, Bangkok, Thailand,

2 Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok, Thailand

Introduction: The concept of “text neck” posits that prolonged forward head posture from smartphone use accelerates cervical spine degeneration due to increased biomechanical loads. While associations between smartphone use and neck pain symptoms are frequently reported, particularly in younger populations, there is a distinct lack of objective evidence investigating whether the duration of smartphone exposure directly corresponds to structural, radiographic changes characteristic of cervical spondylosis. This study aimed to bridge this gap by examining the association between objectively measured smartphone screen time and the presence and severity of radiographically defined cervical spondylosis in a working-age adult population. Material and Methods: A cross-sectional study was conducted on 178 adult participants. Data collection included demographics, clinical history, and Neck Disability Index (NDI) scores. The primary exposure variable is smartphone usage, which was objectively measured using the built-in device software, recording the average daily screen time (in minutes) and the average daily pickups over one week. All participants underwent cervical spine radiography. The primary outcome, cervical spondylosis, was quantified using the Cervical Degenerative Index (CDI). Participants were dichotomized into Spondylosis-Absent (maximum CDI = 0) and Spondylosis-Present (maximum CDI ≥ 1) groups. Statistical analysis involved comparing group characteristics, using univariate and multivariate logistic regression to adjust for confounders (age and BMI), and Spearman correlation to assess relationships between screen time, NDI scores, and ordinal CDI severity. Results: Of the 178 participants (mean age 41.1 years; 79.2% female), 155 (87.1%) exhibited radiographic evidence of spondylosis. Participants with spondylosis were significantly older (mean age 42.1 vs. 34.7 years, p < 0.001) and had higher baseline NDI scores (28.1% vs. 20.7%, p = 0.005). However, smartphone usage metrics were nearly identical between groups. Average daily screen time was 388.9 minutes for the spondylosis-present group versus 390.1 minutes for the spondylosis-absent group (p = 0.973). Daily pickups and duration of smartphone ownership also showed no significant differences. Univariate analysis identified age (OR 1.08 per year, p < 0.001) and NDI score (OR 1.05 per point, p = 0.007) as predictors of spondylosis, but screen time demonstrated no association (OR 1.00, p = 0.973). In the multivariate analysis, age remained the only significant independent predictor (adjusted OR 1.07, p = 0.005), while screen time remained non-significant (adjusted OR 1.00, p = 0.895). Spearman correlation analysis confirmed no significant relationship between screen time duration and either CDI severity (ρ = -0.015) or NDI score (ρ = -0.004). Conclusion: In this cohort of working adults, objectively logged smartphone screen time duration and usage frequency were not associated with the presence or radiographic severity of cervical spondylosis. Age emerged as the sole independent predictor of degenerative changes. While a correlation between degeneration severity and neck disability was observed, screen time did not correlate with either outcome. These findings challenge the assumption that screen time duration is a primary driver of structural cervical degeneration, suggesting that preventive strategies should prioritize established risk factors and ergonomics rather than focusing solely on time spent on devices.

ID: AGS26/785

A027: Impact of GLP-1 agonist therapy on postsurgical outcomes following multilevel correction for adult spinal deformity: a propensity score-matched analysis

Cassandra Dr'Amico 1 , Benjamin Jacques 1 , Robert Ferdon 2 , Jason Silvestre 1 , Stephen Lewis 3 , Christopher Nielsen 3 , Charles Reitman 1 , James Lawrence 1 , Robert Ravinsky 1

1 Medical University of South Carolina, Charleston, United States ,

2 University of South Carolina - School of Medicine, Columbia, United States ,

3 University of Toronto, Toronto, Canada

Introduction: GLP-1 receptor agonists (GLP-1As), widely used in diabetes management, have recently been linked to reduced postoperative complications. Their role in spinal surgery, however, remains underexplored. This study examines outcomes in adult spinal deformity (ASD) patients undergoing deformity correction with and without GLP-1A therapy. Materials and Methods: This multi-center, retrospective cohort study was performed using the TriNetX Global Collaborative Database (2005-2025) utilizing CPT and ICD-10 codes for patients undergoing spinal deformity correction due to ASD. Patients prescribed GLP-1As within 1 year of surgery were 1:1 propensity-score matched with patients who were not using GLP-1As. Cohort balancing was achieved according to patient demographics and comorbidities. Surgical outcomes between groups were analyzed at 1- and 2-year intervals. Statistical significance was defined as p < 0.05. Results: At 1- and 2-years following surgery, patients taking GLP-1As had significantly lower odds of pseudoarthrosis, hardware failure, wound dehiscence, infection, thromboembolic events, readmission, and mortality. Conclusion: Our findings reveal a statistically significant reduction particularly in pseudoarthrosis, hardware failure, readmission and mortality associated with patients treated with GLP-1As. These results align with the recent literature, pointing to a potential complementary therapy in management of ASD. Further research characterizing the mechanism by which GLP-1As exert their effects on postoperative spinal physiology is warranted to assess their utility in the optimization of patient outcomes.

OP4: EOS and Beyond

ID: AGS26/159

A028: Complications and curve progression in EOS patients with extended distraction surgery intervals in growing rod surgery - A retrospective cohort study

Jianqiang Wang 1 , Yong Hai 1

1 Beijing Chaoyang Hospital, Beijing, China

Introduction: Early-onset scoliosis (EOS), defined as spinal deformity presenting before age 10, is typically managed with growing rod surgery to halt curvature progression. Periodic distractions of the rods allow simultaneous control of deformity progression and preservation of spinal growth potential. Current protocols mandate distractions every 6 months, though some regions extend intervals to 1 year based on local resources. However, clinical experience reveals that many patients exceed 1-year intervals (up to 2 years) for scheduled distractions due to poor compliance and financial burden. Material and Methods: We conducted a retrospective analysis of 59 EOS patients who underwent growing rod surgery at our institution between August 2012 and June 2022. Radiographic parameters were collected to compare the differences in outcomes and complications between patients who underwent lengthening surgeries at intervals of less than 12 months versus those with intervals greater than 12 months. Results: In this study, 29 complications were found in 59 patients. The complication rate was significantly lower in the group with lengthening intervals of less than 12 months compared to the group with intervals longer than 12 months (35.3% vs. 68%, p < 0.05). Additionally, patients in the less than 12-month interval group showed less progression of the main curve, a lower maximum kyphosis angle, and greater height increases in the fixed segments (d = 7.53 ± 3.69 cm, p < 0.05). In contrast, patients with lengthening intervals longer than 12 months were more likely to experience sagittal plane imbalance (p < 0.05). Conclusion: Lengthening intervals greater than 12 months have a higher rate of complications and increase the risk of curve progression and spinal imbalance, and it is unwise to increase the interval time just to minimize the number of surgeries before the final fusion.

ID: AGS26/1670

A029: Hybrid fixation vs bilateral VBT in AIS: striking the balance between durability and motion

Altug Yucekul 1 , Feyzi Kilic 2 , Fedde Weenink 3 , Gokhan Ergene 4 , Sahin Senay 5 , Elif Gizem Carus 2 , Tais Zulemyan 2 , Caglar Yilgor 6 , Ahmet Alanay 1

1 Acibadem Mehmet Ali Aydinlar University School of Medicine, istanbul, Turkey ,

2 Comprehensive Spine Center, Acibadem Maslak Hospital, Istanbul, Turkey ,

3 Vrije Universiteit Amsterdam Faculty of Medicine, Amsterdam, The Netherlands ,

4 Department of Operating Room Services, Acibadem Mehmet Ali Aydinlar University Vocational School of Health Sciences, Istanbul, Turkey ,

5 Department of Cardiovascular Surgery, Acibadem Mehmet Ali Aydinlar University School of Medicine, Istanbul, Turkey ,

6 Koc University Hospital Spine Center, Istanbul, Turkey

Introduction: Hybrid fixation-thoracic posterior instrumentation and fusion combined with lumbar vertebral body tethering (VBT)-and bilateral VBT (thoracic and lumbar) are increasingly used motion-preserving strategies for adolescent idiopathic scoliosis (AIS). Hybrid constructs offer durable correction in the thoracic spine while aiming to preserve lumbar mobility, whereas bilateral VBT provides flexibility across both regions but may carry higher tether-related complication risks. To date, no direct comparison between these two approaches has been reported. This study presents the first matched, follow-up-equalized analysis comparing hybrid fixation and bilateral VBT, evaluating radiographic, patient-reported, pulmonary, complication, and implant-survival outcomes. Materials and Methods: Prospectively collected data were retrospectively analyzed. A 1:1 matched cohort (15 pairs) was created using preoperative thoracolumbar/lumbar Cobb (± 5°) and Sanders stage (± 1), incorporating CVM (± 2). For each pair, analyses were capped at the shorter follow-up (pairwise cap: median 25.3 months, IQR 23.8-28.1). Upper thoracic (UT), main thoracic (MT), and thoracolumbar/lumbar (TLL) Cobb angles were compared at preoperative, first-erect, and latest follow-up ≤ cap. SRS-22 domain changes (pre to latest ≤ cap) and pulmonary function (percent-predicted FVC, FEV1) were compared as Hybrid change - Bilateral change (ΔΔ). Complications were assessed with McNemar’s test; implant survival (time to first tether break) was analyzed by Kaplan-Meier with log-rank. Results: Fifty AIS patients (25 Hybrid, 25 Bilateral) were included. Mean age was 13.3 ± 1.3 years (Hybrid) and 12.9 ± 2.0 years (Bilateral). Lenke types in the Hybrid group were Type 1 (6), Type 2 (3), Type 4 (4), and Type 5 (2) with lumbar modifiers B (35.7%) and C (64.3%). The Bilateral group included Type 1 (5), Type 2 (4), Type 3 (3), and Type 6 (3) with lumbar modifiers B (13.3%) and C (86.7%). Preoperative UT, MT, and TLL Cobb angles were 36.8° ± 10.5°, 62.1° ± 17.5°, and 50.4° ± 8.5° (Hybrid) vs 29.9° ± 9.6°, 59.7° ± 10.2°, and 49.9° ± 8.6° (Bilateral). Group comparability (Hybrid - Bilateral) was UT +7.2° (p = 0.152), MT +2.2° (p = 0.678), and TLL +1.1° (p = 0.080). At first-erect, differences favored Hybrid: UT -12.5° (p = 0.0017), MT -20.0° (p < 0.001), and TLL -5.5° (p = 0.049). At latest ≤ cap, differences persisted: UT -10.1° (p = 0.020), MT -17.1° (p < 0.001), and TLL -9.7° (p = 0.014). SRS-22 ΔΔ were: Function -0.47 (p = 0.014), Pain -0.17 (p = 0.539), Self-image +0.39 (p = 0.298), and Mental Health -0.51 (p = 0.113). Pulmonary function changes were not significant between groups: ΔΔ FVC -4.0 (p = 0.561), ΔΔ FEV1 -1.5 (p = 0.802). Complications were fewer in Hybrid (tether breaks 2) vs Bilateral (9) (p = 0.023). Kaplan-Meier survival free of tether break was higher in Hybrid: 12 months 1.00 vs 0.857; 24 months 0.825 vs 0.455 (log-rank p = 0.015). Conclusion: In this first matched, follow-up-equalized comparison of motion-preserving strategies in AIS, Hybrid fixation achieved lower UT/MT/TLL Cobb angles both immediately and at mid-term, and demonstrated higher implant survival with fewer tether failures. Bilateral VBT showed greater improvement in SRS-Function, while other SRS domains and pulmonary outcomes were comparable. These findings suggest Hybrid fixation may offer greater durability, whereas Bilateral VBT may confer functional advantages. Larger, longer-term studies are needed to validate these results.

ID: AGS26/2185

A030: Comparison of magec rods and APC technique in the management of early onset scoliosis

Uzair Rashid 1

1 Ghurki Trust Teaching Hospital, Lahore, Pakistan

Background: Early Onset Scoliosis (EOS) provides special challenges because continuous spinal deformity correction and preservation of growth must be in balance. The two least invasive approaches that are currently applied to EOS are magnetically controlled growing rods (Magec) and the Active Apex Correction (APC) technique. This study compares their efficacy, safety, and clinical outcomes. Materials and Methods: 60 pediatric patients (4-10 years old) with EOS, treated in 2021-2024, were put under analysis. Magec rod implantation was done on 30 patients, and the APC technique was also done on 30 patients. Measures of outcomes were curve correction (Cobb angle), complication rates, frequency of reoperation, and quality of life (QoL). Preoperative radiographic measurements were recorded at 6, 12, and 24 months. T-tests and Chi-squares were used to do a statistical analysis. Results: Both groups achieved significant postoperative curve correction (Magec: 45% ± 10%; APC: 40% ± 12%, p < .05). Magec rods were associated with fewer reoperations (10% vs. 20%, p < .05) but higher implant-related complications (15% vs. 8%, p < .05), including rod breakage and magnetic motor failure. Patients treated with APC demonstrated superior QoL scores at 12 months (p < .05), likely due to less invasive adjustments. Growth preservation was comparable across groups (p > 0.05). Conclusion: Magec rods and the APC technique are effective in the management of EOS with comparable results in terms of deformity correction and growth preservation. Magec rods minimize reoperational rates but have more risks of implant complications, whereas APC demonstrates superior results in QoL. Decisions on treatment must be customized based on the nature of the curves and on patient/family preferences. The Magec rod is expensive, while APC is an economic technique. We need long-term data and results to determine which technique is more effective.

ID: AGS26/2405

A031: Preoperative pelvic incidence-thoracic kyphosis mismatch predicts correction of pelvic incidence-lumbar lordosis mismatch in Scheuermann’s kyphosis

Kylie Callan 1 , Joshua Klatt 1 , Joshua Speirs 1 , Laura Bellaire 1 , Joel Turtle 1

1 University of Utah Hospital, Salt Lake City, United States

Introduction: Scheuermann’s kyphosis is the most common structural kyphosis in adolescents. Treatment ranges from observation to surgical intervention, with recommendations largely based on curve magnitude, and surgery typically indicated for kyphosis greater than 75 degrees. This approach does not account for overall spinal and pelvic geometry. The objective of this study was to determine whether pelvic incidence-thoracic kyphosis (PI-TK) mismatch is a better predictor of surgical outcomes than thoracic kyphosis alone, with the hypothesis that PI-TK mismatch more accurately reflects functional alignment. Material and Methods: Patients with Scheuermann’s kyphosis who underwent surgery were identified using CPT and ICD codes. Preoperative, one-month postoperative, and six-month postoperative radiographs were reviewed when available. Measurements included maximum kyphosis, maximum lordosis, T1-T12 kyphosis, pelvic incidence, and sagittal balance. Demographic information was collected from chart review. Correlations between preoperative PI-TK mismatch and preoperative PI-lumbar lordosis (LL) mismatch were assessed. Paired t-tests compared preoperative and postoperative PI-LL mismatch. Additional analyses examined the relationship of preoperative PI-TK mismatch and preoperative maximum kyphosis with PI-LL correction. Results: Fifty-three patients had complete radiographs at six months postoperatively and were included in the analysis. Twenty-four (45.3%) were female. The mean age at surgery was 15.9 ± 1.7 years. The mean BMI was 24.6 ± 7.1. Preoperative PI-TK and PI-LL mismatch demonstrated a strong correlation (r = 0.67, p < 0.00001). PI-LL mismatch significantly improved from preoperative (-36.4°) to postoperative (-16.4°) at six months (p < 0.0001). Preoperative PI-TK mismatch correlated modestly with PI-LL correction (r = -0.28, p = 0.04), while preoperative maximum kyphosis was not significantly associated with PI-LL correction (r = 0.23, p = 0.11). Conclusion: This study demonstrates that PI-TK mismatch is more closely related to postoperative sagittal alignment correction than maximum kyphosis, supporting the hypothesis that mismatch is a more clinically relevant marker of pathology than curve magnitude alone. The strong association between PI-TK and PI-LL mismatch further suggests that PI-TK may serve as a surrogate marker of disability, similar to the established role of PI-LL in the adult spine. By highlighting the limitations of using kyphotic angle alone, these findings underscore the importance of incorporating spinopelvic parameters into surgical decision-making for Scheuermann’s kyphosis. PI-TK mismatch may improve patient selection, enhance preoperative counseling, and provide a more accurate framework for predicting outcomes. Future prospective studies including patient-reported measures are warranted to validate PI-TK mismatch as a marker of functional disability and surgical indication in this population.

ID: AGS26/2780

A032: Acute abdominal compartment syndrome following scoliosis surgery in a patient with Rett syndrome

Mesut Kılıç 1

1 Department of Orthopaedics and Traumatology, Private Kolmed Hospital, Amasya, Turkey

Introduction: Rett syndrome is a rare neurodevelopmental disorder often associated with scoliosis. Surgical correction carries high complication risks.This paper aims to present a case of acute abdominal compartment syndrome (AACS) as a rare, life-threatening complication of scoliosis surgery in Rett syndrome. Material and Methods: A 17-year-old female with Rett syndrome underwent T3-iliac posterior instrumentation and pelvic fixation. Intraoperatively, AACS developed, requiring urgent decompressive laparotomy. Results: After spinal correction, the patient developed hypotension and bradycardia, requiring inotropic support. Echocardiography showed compression of the right ventricle and inferior vena cava. Intraoperative ultrasonography revealed collapse of the vena cava and renal veins, gastric distension, bowel wall edema, and free pelvic fluid. Fluid drainage did not improve findings. Bladder pressure measured 25 mmHg; arterial blood pressure fell to 30/10 mmHg. Emergency laparotomy with Bogota bag was performed. Hemodynamics improved, the abdomen was closed on day 3, the patient was extubated on day 5, transferred to the ward on day 10, and discharged. Conclusion: Scoliosis surgery in Rett syndrome has high complication rates. To our knowledge, this is the first report of AACS in such a patient. If untreated, mortality approaches 100%.

ID: AGS26/3046

A033: The predictive value of proximal femur maturity index (PFMI)-based nomogram for distal adding-on following posterior hemivertebra resection and fusion

Youwei Ai 1 , Yongdi Wang 1 , Hong Ding 1 , Ce Zhu 1 , Zhuojie Xiao 1 , Limin Liu 1

1 West China Hospital, Sichuan University, Chengdu, China

Introduction: Distal adding-on is a common complication in patients with congenital scoliosis due to hemivertebra (HV) after posterior hemivertebra resection and fusion, associated with skeletal maturity status. However, the Sanders staging (SS) and the distal radius and ulna (DRU) classification require a hand-wrist radiograph, and the accuracy of the Risser sign in patients with scoliosis is a matter of concern. Recently, a novel proximal femur maturity index (PFMI) has been proposed as a reliable tool to determine skeletal growth. Material and Methods: Patients with congenital scoliosis due to HV who underwent posterior hemivertebra resection and fusion at our department between 2012 and 2021 were selected, and their demographic data and radiographic assessments were collected. We carried out a ROC curve analysis to assess the efficacy of the PFMI in independently predicting distal adding-on. The least absolute shrinkage and selection operator (LASSO) method was performed to select the risk factors and then establish a nomogram to predict the rate of distal adding-on. The concordance index (C-index), the receiver operating characteristic (ROC) curve and the calibration curve were used to evaluate the performance of the nomogram. Results: 108 patients met the inclusion criteria. 22 patients (20.3%) developed distal adding-on. The AUC of the PFMI is 0.756 (95% confidence interval [CI]: 0.658-0.853) and the optimal threshold is 2.5 (sensitivity: 72.7%, specificity: 74.4%). After LASSO regression analysis, seven potential risk factors were selected, including BMI, Risser sign, location of HV, preoperative Cobb angle, preoperative TK, preoperative SVA, and the PFMI. The area under the receiver operating characteristic curve (AUC) of the nomogram was 0.919, the C-index was 0.919, and the calibration curve of the nomogram demonstrated good results in predicting adding-on. Conclusion: A patient with a PFMI grade of 2 or less has a high risk of developing distal adding-on after posterior hemivertebra resection and fusion. This PFMI-based nomogram provides a valuable tool for assessing the incidence of distal adding-on.

ID: AGS26/286

A034: Hip reduction surgery versus scoliosis correction in non-ambulant neuromuscular patients. Which surgery as a first step?

Mostafa Meshneb 1 , Mohammed Patel 1 , Nasir Quraishi 1

1 Queens Medical Centre Nottingham University Hospitals, Nottingham, United Kingdom

Introduction: Children and young adults with neuromuscular disorders have a high incidence of both spine and hip deformities, the severity of which is associated with severity of the neurological compromise. There are not widely accepted clinical or radiological guidelines to facilitate decision making on which of these conditions should be primarily addressed. The aim of this study was to evaluate the outcome of either primary scoliosis or hip surgery in children and young adults with neuromuscular disorders and its effect on progression of the other deformity, as well as the need for of a second operation. Material and Methods: A retrospective study was conducted on all children and young adults with neuromuscular related synchronous hip subluxation / dislocation and scoliosis undergoing hip or scoliosis surgeries in our institution between 2012 and 2021, with a minimum follow up of 24 months. Demographic and operative data were collected; hip migration percentage (MP), Cobb angle of the main curve and degree of pelvic obliquity were measured pre- and post-operatively at final follow up. The need for correction of the “other” remaining pathology after the primary operation was analyzed. Statistical analysis was performed using SPSS version 23 with p < 0.05 for significance. Results: 40 neuromuscular patients (25cerebral palsy, 11syndromic, 2SMS and 2DMD) with synchronous hip displacement and scoliosis were included. Twenty patients with an average age of 10.2 years (6-17) had hip correction surgery performed primarily (hip first), with a mean follow up of 54.9 (14-96) months compared to a second group consisting of 20 patients of average age 12.4 years (10-24) who had scoliosis correction first), with a mean follow-up of 40 (13-60) months. In the “hip first” group, pelvic obliquity, hip MI and Cobb angle were 16.8°, 70%, and 49°, respectively. At final follow up, the mean pelvic obliquity and Cobb angles significantly progressed post-operatively to 27.2° (p = 0.003) and 82.2° (p = 0.0016) respectively. 18 patients (90%) required scoliosis correction after the hip surgery. In the “scoliosis first” group, the mean pelvic obliquity, hip MP and Cobb angle were 21.2°, 49 % and 65.5°, respectively. At final follow-up, both pelvic obliquity and the Cobb angle significantly improved to 8.44° (p = 0.003) and 23.4° (p = 0.0013) respectively. In 11/20 (55%) patients, the hip MP had significantly increased following the spinal surgery to 65% (p = 0.0017), but only 5/20 (25%) patients underwent hip surgery after scoliosis correction. Conclusion: In neuromuscular patients presenting with synchronous hip displacement and scoliosis deformity, corrective scoliosis surgery is associated with significant correction of pelvic obliquity and lower rates of secondary hip surgery. On the other hand, hip surgery does not reduce the risk of pelvic obliquity and scoliosis deformity progression.

ID: AGS26/529

A035: Impact of preoperative halo gravity traction on posterior spinal fusion outcomes in patients with osteogenesis imperfecta: the first comparative analysis

Paolo Brigato 1 , Salim Al Rawahi 2 , Yousef Aljamaan 3 , Oded Rabau 4 , Kai Sheng 2 , Neil Saran 2 , Jean Ouellet 2

1 University Hospital Campus Bio-Medico of Rome, Rome, Italy ,

2 McGill University, Montreal, Canada ,

3 Imam Abdulrahman bin Faisal University, Dammam, Saudi Arabia ,

4 Yitzhak Shamir Medical Center, Tel Aviv, Israel

Introduction: Osteogenesis Imperfecta (OI) is a rare connective tissue disorder frequently complicated by severe scoliosis, which is resistant to bracing due to bone fragility and ligamentous laxity. Posterior spinal fusion (PSF) with pedicle screws has been shown to achieve up to 60% of coronal curve correction. Preoperative halo-gravity traction (HGT) may further improve deformity, respiratory function, and nutritional status, though it carries inherent risks. This study assesses whether HGT provides meaningful perioperative and radiographic benefits over PSF alone in this rare and vulnerable patient population. Methods: The study included 36 pediatric OI patients with spinal deformity, treated between January 2002 and Decembre 2020, with a minimum 2-year follow-up. The primary outcome was the coronal correction rate (CR) of scoliosis curves. Secondary outcomes included major postoperative complications (classified according to the Modified Clavien-Dindo-Sink Classification (MCDS), operative time, blood loss, length of hospital stay (LOS) and coronal plane loss of correction at the last FUP. Statistical analysis was performed using Mann-Whitney U test, and Chi-Squared tests, with significance set at 0.05. An a priori power analysis (80% power, medium effect size 0.3, α = 0.05) indicated that a minimum of 10 patients per parameter was required to detect the intended effect. A p-value < 0.05 was considered statistically significant for all comparisons. Results: The patients were divided into a HGT+PSF group (N = 19) and a PSF-only group (N=17), which were comparable preoperatively regarding sex, age, OI type, curve characteristics, and bone mineral density. Postoperative CR for major and minor curves was 60.2 ± 20.6% and 66.5 ± 19.6% in the HGT+PSF group, and 55.1 ± 19.8% and 37.7 ± 16.7% in the PSF group (p = 0.337 and p = 0.003). At last follow-up, CR was 51.1 ± 14.2% and 38.8 ± 25.9% for HGT+PSF, and 44.9 ± 21.6% and 25.2 ± 13.3% for PSF (p = 0.298 and p = 0.238). Mean blood loss was 1235.2 ± 736.8 mL versus 1368.3 ± 976.1 mL (p = 0.972), and total surgical time was 443.1 ± 92.7 min versus 410.1 ± 81.1 min (p = 0.490) for the HGT+PSF and PSF groups, respectively. Mean LOS was 12.6 ± 6 days versus 9.5 ± 2.9 days (p = 0.186). Perioperative complications occurred in 11 patients (57.9%) in the HGT+PSF group and 5 patients (29.4%) in the PSF group (p = 0.367), with major complications (MCDS Grade ≥ III) more frequent in HGT+PSF (4 patients, 36%) than PSF (1 patient, 20%). Conclusions: Preoperative HGT provided modest additional coronal correction but did not reduce perioperative complications, blood loss, operative time, or LOS compared to PSF alone. While it may offer radiographic benefits, its routine use should be carefully considered against potential risks and resource demands in pediatric OI patients. Future prospective, multicenter studies are warranted to better define the role of HGT in this population and optimize surgical planning and outcomes.

ID: AGS26/913

A036: Too young or just in time? Outcomes of posterior hemivertebra resection under age three

Fedde Weenink 1 , Bing Wui Ng 2 , Feyzi Kilic 3 , Bayron Valenzuela C 4 , Akinbo Olaolu 5 , Elif Gizem Carus 3 , Tais Zulemyan 3 , Nuri Demirci 6 , Caglar Yilgor 7 , Altug Yucekul 8 , Ahmet Alanay 8

1 Vrije Universiteit Amsterdam Faculty of Medicine, Amsterdam, The Netherlands ,

2 Spine Unit, Hospital Pakar Kanak-kanak UKM, Kuala Lumpur, Malaysia ,

3 Comprehensive Spine Center, Acibadem Maslak Hospital, Istanbul, Turkey ,

4 Department of Neurosurgery, Hospitla Clínico Universidsd de Chile, Santiago, Chile ,

5 Department of Neurological Surgery, University College Hospital (UCH), Ibadan, Nigeria ,

6 Department of Orthopaedic Surgery, Washington University in St. Louis, Saint Louis, United States ,

7 Koc University Hospital Spine Center, Istanbul, Turkey ,

8 Acibadem Mehmet Ali Aydinlar University School of Medicine, istanbul, Turkey

Introduction: The ideal timing of hemivertebra resection remains debated, as earlier intervention may optimize correction but surgery under the age of three poses unique anatomical and technical challenges. This study aimed to evaluate the safety profile, radiological outcomes, and specific technical aspects of posterior hemivertebra resection in this distinct patient population. Materials and Methods: A retrospective review was conducted on patients who underwent hemivertebra resection before the age of three. Inclusion criteria were the presence of congenital spinal deformities caused by single or multi-level hemivertebrae and a minimum follow-up of 2 years. Patients with additional segmentation failures or prior spinal surgery were excluded. Patients who underwent a 2-rod construct were compared with those who underwent 3- or 4-multirod constructs. All resections were carried out using modern surgical techniques, including multimodal neuromonitoring, intraoperative CT navigation, microsurgical dissection with meticulous hemostasis, and multirod/hook-assisted stabilization strategies. In semi-segmented HVs, enucleation rather than decancellation was performed to minimize bleeding and preserve correction. Results: A total of 19 patients (23 hemivertebrae resections) met the criteria. The mean surgical age was 23.5 ± 7.5 months, and the mean follow-up was 74.3 ± 41.1 months. The mean coronal segmental Cobb angle was reduced from 33.0° ± 9.3° preoperatively to 5.2° ± 4.7° postoperatively and was measured at 6.0° ± 6.8° at the latest follow-up. The initial correction rate of 85.4% ± 11.0% was well-maintained at 82.8% ± 18.4% at the final follow-up. For the cranial compensatory curve, the mean initial correction was 64.4% ± 31.0%, which improved to 80.0% ± 26.5% at the latest follow-up. For the caudal curve, the initial correction was 61.1% ± 19.3%, and this was maintained at 63.7% ± 33.2% at the final follow-up. The mean operative time was 359.6 ± 93.5 minutes, with a mean intraoperative blood loss of 183.3 ± 124.5 ml and a mean hospital stay of 5 ± 2 days. Six complications were recorded in five patients: one early implant failure, two cases requiring growing rod implantation due to new curve formation independent of the resected segment, one instance of hemivertebra regrowth, and two superficial wound infections. None of the patients experienced any neurological deterioration. While radiographic correction rates were comparable between the 2-rod and multi-rod groups (p > 0.05), postoperative casting was required for all patients with 2-rod constructs, whereas none was needed for patients in the multi-rod group. Conclusion: Posterior hemivertebra resection in children under 3 years of age is a neurologically safe and highly effective method for deformity correction when performed with modern surgical techniques and by experienced teams. Early intervention offers permanent results with an acceptable complication profile and should be considered a definitive treatment strategy in this patient group. Furthermore, the application of multi-rod constructs and hook-assisted compression may provide superior stability, protecting patients from the burden of postoperative casting.

OP5: Non-Operative Treatment Options

ID: AGS26/263

A037: Comparative safety and health service utilization for gabapentin versus non-steroidal anti-inflammatory drugs for cervical radiculopathy: a retrospective cohort study of academic centers

Anthony Baumann 1 , Robjert Trager 2 , Omkar Anaspure 3 , Dalton McGlamery 4 , Shahab Yazdanpanah 4 , Keegan Conry 1 , Gordon Preston 1 , Jacob Hoffmann 1

1 Cleveland Clinic Akron General, Akron, United States ,

2 Department of Rehabilitation Services, University Hospitals, Cleveland, United States ,

3 Perelman School of Medicine, University of Pennsylvania, Philadelphia, United States ,

4 Northeast Ohio Medical College, Rootstown, United States

Introduction: Off-label use of gabapentin for neuropathic pain conditions such as cervical radiculopathy (CR) remains controversial due to limited evidence of clinical benefit and safety concerns (e.g., opioid co-prescription, adverse drug events [ADEs], dizziness, and hospitalization risk), necessitating further research. This study explores safety risks and health service utilization in adults with newly diagnosed CR, comparing gabapentin to nonsteroidal anti-inflammatory drugs (NSAIDs) to inform clinical practice. Material and Methods: This retrospective propensity-matched cohort study using the TriNetX database included opioid-naïve adults (≥ 18 years) with newly-diagnosed CR without history of spinal surgery, spinal neoplasms, myelopathy, opioid-use disorder or cervical injection. Patients were divided into gabapentin or NSAID cohorts based on prescription at CR diagnosis, and matched for demographics, prescriptions, and comorbidities. Outcomes included incidence and risk ratios (RRs) of safety and health service utilization markers assessed through one year. Significance was evaluated at P<.0036 (Bonferroni-corrected). Results: After matching, there were 23,379 patients (mean age: 53 years; 57% female) per cohort. About 14% of patients per cohort were co-prescribed opioids at CR diagnosis. Compared to the NSAID cohort, the gabapentin cohort had a statistically significant (p < .0036) increased risk of posterior cervical fusion (RR: 3.23), opioid-related disorder (RR: 2.87), cervical corticosteroid injection (RR: 2.52), opioid-related ADEs (RR: 2.32), anterior cervical discectomy and fusion (RR: 2.10), acute respiratory failure (RR: 1.80), cervical spine magnetic resonance imaging (RR: 1.43), inpatient visits (RR: 1.38), oral opioid prescription (RR: 1.34), and cervical spine radiographic imaging (RR: 1.21). Risk of dizziness, falls, somnolence, and emergency department visits was not significantly increased (p > .0036). Cohorts were similar in terms of data quality and healthcare engagement per multiple negative control outcomes. Conclusion: Gabapentin prescription for newly-diagnosed CR in adults was associated with increases in opioid prescription, adverse safety events, and health service utilization compared to NSAIDs through one year, corroborating its lack of guideline recommendation for CR. No outcome favored gabapentin over NSAIDs. Gabapentin was occasionally co-prescribed with opioids at CR diagnosis, raising concerns.

ID: AGS26/787

A038: GLP1 receptor agonists in patients with spinal cord injury: a propensity-matched analysis of clinical and functional outcomes

Hanna von Riegen 1 , Alireza Mirahmadi 1 , Mario Keko 1 , Jason Pittman 1 , Sapan Gandhi 1

1 Department of Orthopedics, Beth Israel Deaconess Medical Center, Harvard Medical School in Boston, MA, Boston, United States

Introduction: Spinal cord injury (SCI) is associated with high morbidity and complex rehabilitation trajectories. Glucagon-like peptide-1 receptor agonists (GLP1-RAs) have demonstrated neuroprotective and anti-inflammatory properties in preclinical models, but their real-world safety and impact in SCI populations remain unclear. The objective of this study is to evaluate and compare the functional and clinical outcomes associated with GLP1-RA use in individuals with SCI, using real-world data from a multi-institutional database. Material and Methods: A retrospective cohort study was conducted using the 2017-2024 data from TriNetX Global Collaborative Network. Adults with SCI were stratified by GLP1-RA use and matched 1:1 (N = 268 per group) via propensity scores based on relevant demographic, clinical, and laboratory variables (including age, sex, race, BMI, diabetes, and socioeconomic indicators). The index event was spinal surgery (CPT-coded decompression, instrumentation, or arthrodesis) in patients with a documented SCI diagnosis. Routinely documented clinical data from electronic health records of participating institutions were available for up to one year after the index surgery and extracted through the TriNetX platform for analysis. Primary outcomes included functional status (mobility, spasticity, bladder/bowel dysfunction), opioid use, fatigue, readmission, ICU stay, mortality and complications (UTI, sepsis, DVT, pressure ulcers). Risk differences and Kaplan-Meier survival estimates were used to compare outcomes between the groups. Results: GLP1-RA use was associated with lower rates of opioid use at 3 months (81.2% vs. 95.5%, p < 0.001) and 1 year (85.8% vs. 97.4%, p < 0.001). Rates of fatigue (18.3% vs. 25.4%, p = 0.047) and readmission (1-year: 63.4% vs. 73.5%, p = 0.012) were also reduced in the GLP1RA group. While ICU admissions were lower in the GLP group, the difference was not significant (28.7% vs. 36.2%, p = 0.065). There were no significant increases in complications such as infection, wound issues, or mortality. Functional outcomes (mobility, neurogenic dysfunction, sensitivity) remained comparable across groups. Conclusion: In this propensity-matched analysis, GLP1-RA use in SCI patients was associated with reduced opioid use, less fatigue, and fewer readmissions without increasing the risk of complications or mortality. These findings support potential benefits of GLP1-RA use in SCI populations and suggest further investigation of potential mechanisms of benefit.

ID: AGS26/2444

A039: Impact of a transitional care clinic on 30-day readmissions after elective spine fusion surgery: a retrospective cohort analysis

Denyse Pitak 1 , Puya Alikhani 2 , Mohsen Rostami 2 , Diego Soto Rubio 2 , Dana Saleh 2 , Luba Ayzenshtat 1

1 Tampa General Hospital, Neuroscience Service Line, Tampa, United States ,

2 University of South Florida Department of Neurosurgery, Brain and Spine, Tampa, United States

Introduction: Hospital readmissions following elective spine fusion surgery remain a persistent challenge, impacting patient recovery, healthcare resource utilization, and institutional quality metrics. While predictors of readmission have been widely studied, few evidence-based strategies have demonstrated success in reducing these events in spine surgery. Transitional Care Clinics (TCCs), designed to bridge the critical gap between hospital discharge and outpatient follow-up, have shown benefits in other surgical populations but remain unexplored in the spine population. The role of TCC programs in optimizing post-discharge outcomes across all spinal fusion types are not well understood. This study evaluated the impact of a structured, multidisciplinary TCC program on 30-day readmissions following elective spine fusions surgery at a large academic medical center. Material and Methods: We performed a retrospective cohort analysis of adult patients (≥18 years) undergoing elective cervical, thoracic, or lumbar spine fusion between May 2024 and May 2025. Exclusion criteria were trauma, infection, malignancy, and emergent cases. The intervention group participated in a structured TCC program implemented from November 2024 to May 2025. Key program elements included a follow-up visit (in-person or telehealth) within 7 days of discharge, comprehensive wound evaluation, medication review, adjustment of pain management, physical therapy coordination, and escalation of care if required. The pre-intervention group (May-October 2024) received standard discharge instructions only. The primary outcome was 30-day hospital readmission. Secondary outcomes included emergency department (ED) visits and skilled nursing facility (SNF) use. Fischer’s exact test was used to examine the association between TCC participation and readmission. Results: There was a total of 345 patients who underwent an elective cervical, thoracic, or lumbar spinal fusion between May 2024 and May 2025. The pre-intervention cohort had 202 patients, while the post-intervention cohort had 143 patients. Pre-intervention cohort patients had a readmission rate of 21.3%, while post-intervention cohort patients had a rate of 5.6%. Fisher's exact test revealed a significant association between TCC intervention and readmission, showing that patients with TCC intervention had a 78% decrease in odds of readmission compared to those without TCC intervention (odds ratio [OR] = 0.22, 95% CI (0.08, 0.49), p = 0.00039). Conclusion: A structured TCC program significantly reduced 30-day readmissions after elective spine fusion, demonstrating its potential as a high-impact intervention in spine surgery. By providing timely follow-up, addressing complications early, and coordinating ongoing care, TCCs enhanced recovery and improved patient outcomes. The findings suggest that transitional care can be an essential component of modern spine surgery programs and may serve as a model adaptable to diverse health systems. Implementing this approach has the potential to improve quality outcomes, advance value-based care, and reduce the burden of preventable readmissions. Future prospective studies in broader settings are needed to validate these findings, evaluate cost effectiveness, and ensure that this care model can be applied across institutions and healthcare systems worldwide.

ID: AGS26/2524

A040: Bracing and physiotherapeutic scoliosis specific exercises (PSSE) can be an alternative treatment option for severe scoliotic patients who refused surgery

Nikos Karavidas 1

1 Schroth Scoliosis & Spine Clinic, Athens, Greece

Introduction: According to Scoliosis Research Society (SRS) surgical indication for Adolescent Idiopathic Scoliosis (AIS) is above 40 ο . However, many patients refuse to be operated, trying brace treatment. To date, no strong evidence exists for non-operative treatment in large scoliotic curves. Our purpose was to investigate the efficacy of a combined therapy with brace and Physiotherapeutic Scoliosis Specific Exercises (PSSE) in severe scoliosis during growth spurt. Material and Methods: Prospective cohort study. 78 patients (73 females and 5 males) received treatment by 3D corrective brace and PSSE. Our inclusion criteria were at least one structural curve with Cobb angle > 40 ο , Risser 0-2, age > 10 years and less than 1 year after first menstruation. Average Cobb angle was 53.4 ο for thoracic curves (41 ο - 99 ο ) and 52.4 ο (40 ο - 85 ο ) for lumbar curves, mean Risser 0.6 and age 12.6 years. 18 curves (23.1%) were single (15 thoracic and 3 lumbar) and 60 double (76.9%). Outcome parameters were Cobb angle post-treatment, Ange Trunk Rotation (ATR), Trunk Appearance Clinical Evaluation (TRACE) scale, SRS-22 questionnaire score, and number of patients underwent operation. Mean follow-up was 35.5 months. Statistical analysis performed by Analysis of Variation (ANOVA) and Chi-square tests. Results: Totally, 31 (39.7%) subjects remained stable, 24 (30.8%) improved > 5 ο and 23 (29.5%) progressed > 5 ο . For the improved curves, 12 were thoracic, 7 lumbar and 5 both thoracic and lumbar. For the progressed curves, 15 were thoracic, 3 lumbar and 5 both thoracic and lumbar). Cobb angle post-treatment improved to 52.2 ο (p = 0.09) for thoracic and 45.9 ο , (p = 0.002) for lumbar curves. A statistically significant reduction was reported for ATR, thoracic reduced from 13.2 ο to 10.4 ο (p = 0.004) and lumbar from 11.3 ο to 8.6 ο (p = 0.01). TRACE scale also significantly decreased from 8.3 to 6.4 (p = 0.01) and SRS-22 total score improved from 73.1 to 80.2 (p = 0.002). Mean in-brace correction (IBC) was 36.8% for thoracic and 34.8% for lumbar curves. In double curves progression rate was 33.3% and in single curves 16.7% (p = 0.004). Higher IBC and good compliance with the treatment protocol led to less progression rate. 9 patients (11.5%) underwent spinal fusion. Conclusion: Conservative treatment achieved a success rate of 70.5% in a group of scoliotic curves above 40 ο with a high risk of progression at the peak of growth. A significant improvement was detected for trunk rotation (ATR), body symmetry measured by TRACE scale and quality of life measured by SRS-22 questionnaire. Non-operative treatment by bracing and PSSE can effectively treat severe scoliosis and reduce the need for surgery. The results were significantly better in single than double curves and in subjects with good compliance and IBC.

AGS26/676

A041: Evaluation of the acute corrective effect of customized scoliosis braces produced using 3D imaging technology on Cobb angle - Single center study

Ahmet Hamdi Olçar 1 , Oğuzhan Gök 1 , Hayri sucu 1 , Tuhan Kurtulmuş 1 , özgür Karakoyun 2

1 Sancaktepe Prof.Dr.Ilhan Varank Training and Research Hospital, Istanbul, Turkey, 2 Doç.Dr.Özgür Karakoyun Private Practice, Tekirdağ, Turkey

Introduction: This study aims to evaluate the acute (short-term) corrective effect of customized scoliosis braces, produced using three-dimensional (3D) imaging and computer-aided design/computer-aided manufacturing (CAD/CAM) technology, on the Cobb angle in individuals diagnosed with Adolescent Idiopathic Scoliosis (AIS). Material and Methods: In this retrospective study conducted between 2022 and 2023, 55 AIS patients (45 females, 10 males) with Cobb angles ranging between 20° and 45° were included. A 3D body surface scan was taken for each patient, and personalized thoracolumbosacral orthosis (TLSO) braces were designed using CAD. Thoracic, thoracolumbar, and lumbar Cobb angles were measured from pre- and post-brace radiographs and compared statistically using the paired t-test. Results: Significant reductions in Cobb angles were observed post-bracing in all regions (p < 0.001), with average correction rates of 72% in the thoracic, 85% in the thoracolumbar, and 71% in the lumbar regions. The highest correction was achieved in the thoracolumbar segment. Conclusion: Short-term findings indicate that personalized braces designed via 3D scanning and CAD/CAM offer strong acute correction of the Cobb angle in AIS treatment. Further multicenter, prospective, and controlled studies are needed to assess long-term effectiveness.

ID: AGS26/576

A042: Conservative management of the adult spinal deformity patient: a biopsychosocial pain self-management intervention versus care-as-usual - A comparative treatment-based cohort study with one-year follow-up

Evelien Hoevenaars 1 , Michiel Beekhuizen 1 , Ahmet Alanay 2 , Yann Philippe Charles 3 , Frank Kleinstueck 4 , Ibrahim Obeid 5 , John O'Dowd 6 , Ferran Pellisé 7 , Javier Pizones 8 , Maarten Spruit 1 , Marinus De Kleuver 9 , Miranda van Hooff 1 9

1 Sint Maartenskliniek, Nijmegen, The Netherlands ,

2 Acibadem University Maslak Hospital, Istanbul, Turkey ,

3 Hôpitaux universitaires de Strasbourg, Strasbourg, France ,

4 Schulthess Klinik, Zurich, Switzerland ,

5 Pellegrin Bordeaux University Hospital, Bordeaux, France ,

6 Hampshire Hospitals NHS Trust, Basingstoke, United Kingdom ,

7 Hospital Universitari Vall d’Hebron, Barcelona, Spain ,

8 La Paz University Hospital, Madrid, Spain ,

9 Radboudumc, Nijmegen, The Netherlands

Introduction: The prevalence and societal burden of disease of adult spinal deformity (ASD) is increasing. Currently, no evidence based non-operative treatment exists for ASD patients with chronic low back pain (CLBP) and negative impact on quality of life (QoL). The aim of this study was to compare a biopsychosocial intervention to care-as-usual in symptomatic, conservatively managed ASD patients. Material and Methods: A comparative cohort study of a biopsychosocial intervention cohort (n=95) and a care-as-usual cohort (n=192) was conducted. Inclusion criteria of both cohorts were age ≥ 18, no surgical indication, Oswestry Disability Index (ODI)>20, coronal Cobb angle ≥ 20, idiopathic (AdIS) or degenerative (de novo) scoliosis (DDS), baseline and one year follow-up (1yFU) ODI data. The biopsychosocial intervention was a 100-hour interdisciplinary group program, based on cognitive behavioural principles and consisted of acceptance and commitment therapy, physical exercise and pain education. The care-as-usual consisted of pain medication, brace, physical therapy, osteopathy/chiropractic and/or injections or no treatment. Primary outcome was functional status (ODIv2.1a). Secondary outcomes were pain intensity (NPRS 0-10), QoL (SF36) and pain medication use. Clinical relevance was a minimal important clinical change (MCIC; improvement in ODI ≥ 10 points) and a patient acceptable symptom state (PASS; ODI ≤ 22). Results: Demographics of the biopsychosocial intervention cohort: median age 57 (IQR 14) years, 87% female, median Cobb angle 28° (IQR 13) and 42% AdIS - 58% DDS. Demographics of the care-as-usual cohort differ mainly in Cobb angle with a median 45° (IQR 28) and ASD diagnosis 71% AdIS - 29% DDS. Interestingly, the correlation between coronal Cobb angle size and functional status (ODI) at baseline was small (rs = -.02; p = .74; n = 287). At 1yFU the patients in the intervention cohort showed significantly greater improvement in ODI compared to the care-as-usual cohort (z = -4.26, p = <.001). Clinical relevance: 49% of the patients in the intervention cohort reached MCIC and 34% reached a PASS, compared to 22% and 18% respectively in the care-as-usual cohort. All secondary outcomes followed a similar pattern of greater improvement in the intervention cohort. In the intervention cohort the number of patients not using pain medication increased from 20 patients at baseline to 49 patients at 1yFU, and the number of patients using opiates halved. Conclusion: This study shows clinically relevant and significantly larger improvement in function, pain, and quality of life in the biopsychosocial intervention cohort, compared to the care-as-usual cohort at 1yFU in symptomatic, conservatively managed ASD patients. In addition, it shows a decrease in pain medication use in the intervention cohort. This non-operative, biopsychosocial intervention for symptomatic ASD patients seems a promising treatment option that could help diminish the increasing burden of ASD for patients and healthcare costs.

ID: AGS26/730

A043: Multidisciplinary tumor board evaluations reduce complications in nonoperative patients with spinal tumors

Thomas Buquet 1 , Fidaa Al-Shakfa 2 , Danielle Boulé 2 , Véronique Freire 1,2 , Giuseppina Laura Masucci 1,2 , Zhi Wang 1,2 , Jesse Shen 1,2

1 Université de Montréal, Montréal, Canada ,

2 Centre hospitalier de l'Université de Montréal, Montréal, Canada

Introduction: Patients with spinal tumors (ST) often present complex cases that require multidisciplinary management. Tumor boards (TB) facilitate patient-centered clinical decision-making, but few centers offer TB and, to our knowledge, no prior studies have evaluated their clinical effectiveness compared to decisions made by a single clinician. This study aims to assess whether clinical outcomes for patients with ST are improved when decisions are made through a TB rather than by an individual physician. Material and Methods: This is a single-center retrospective study including all patients with primary or metastatic ST who were managed non-surgically and evaluated by a TB composed of at least a musculoskeletal radiologist, a radiation oncologist, and a spine surgeon compared to a group of patients with no presentation to a TB, both during the period between January 2019 and June 2023. Demographic data, tumor type, choice of treatment and complications were documented. Results: Non-surgical treatment decisions and complications rates were compared between 122 patients presented to a TB and 106 patients not presented to a TB during the same period. Patients discussed at a TB had less stereotactic body radiation therapy (SBRT) (39.3% vs 64.2%, p < 0.001) and less palliative radiotherapy (17.2% vs 31.1%, p < 0.05) but underwent vertebroplasty more frequently (13.1% vs 0.9%, p < 0.001). Conservative approaches such as observation (13.9% vs 0.9%, p < 0.001) or no intervention recommended (13.9% vs 2.8%, p < 0.05) were more frequently chosen in the TB group. Overall, patients in the TB group were less likely to receive any intervention (72.1% vs. 96.2%, p < 0.001). At more than 90 days, complication rates were significantly lower in the TB group (2.5% vs. 12.3%, p < 0.05), with radiation-induced fractures being the most common complication. These fractures occurred less frequently in the TB group (1.6% vs. 11.3%, p < 0.05). When comparing only patients who did receive an intervention, TB discussion was still associated with fewer complications (3.4% vs 12.7%, p < 0.05). Conclusion: Multidisciplinary tumor board discussion significantly influenced non-surgical treatment decisions in patients with ST, favoring more conservative approaches and reducing overall intervention rates. TB presentation was also associated with significantly fewer 90-day complications. These findings support the integration of TB discussions in the clinical management of patients with ST to optimize their treatments. This work supports further multicenter prospective studies evaluating the effectiveness of TB evaluations.

ID: AGS26/1412

A044: Prescription pattern of rigid cervical collar use after cervical fusion: a national multicenter survey of spine surgeons in Mexico

Cinthya Alejandra Varas-Aguilar 1 , Alexis Manuel Mazariegos-Perez 1 , Hector Ismael De Haro-Segura 1 , Daniel Torres-Fragoso 2 , Alfredo Javier Moheno-Gallardo 1

1 Highly Specialized Medical Unit of Traumatology, Orthopedics, and Rehabilitation “Dr. Victorio De La Fuente Narváez”, Mexico City, Mexico ,

2 Hospital Angeles Lindavista, Mexico City, Mexico

Introduction: The use of rigid cervical collars in the postoperative management of cervical fusion has been debated for decades. Originally described by Smith and Robinson in 1958, collars were initially reserved for multilevel procedures. Over time, their use expanded based on the belief that limiting cervical motion enhances spinal stability, reduces pain, and provides patients with a sense of safety. However, no consensus currently exists on specific indications or the optimal duration of postoperative collar use. Material and Methods: A nationwide survey was designed and administered to spine surgeons in Mexico, including both orthopedic surgeons and neurosurgeons. The Delphi methodology was employed through the SurveyMonkey platform, with open-ended and multiple-choice questions. The questionnaire was developed by two spine surgeons, two orthopedic residents, and one spine fellow. Participation was anonymous, and results were analyzed using IBM SPSS 26.0. Descriptive statistics were applied, including measures of central tendency and dispersion for quantitative variables, and absolute frequencies and percentages for qualitative variables. Results: A total of 109 spine surgeons responded to the survey. The majority were male (93.6%). More than half (57.8%) worked in both public and private sectors, and 29.4% had 10-20 years of professional experience. Regarding surgical volume, 33% reported performing 16-30 cervical fusions per year. Rigid cervical collars were prescribed by 72.5% of surgeons, used in an average of 68% of their postoperative patients. Pathology type was considered by 63.3% of respondents, with longer immobilization prescribed for traumatic compared to degenerative cases (6.54 vs. 5.94 weeks, p = 0.964). Osteosynthesis material influenced decisions in 55% of cases, with self-locking interbody cage associated with longer immobilization (mean 5.15 weeks). Patient comorbidities were taken into account by 59.6% of surgeons, particularly osteoporosis, which was associated with the longest immobilization times (mean 7.57 weeks). Similarly, 55% of respondents considered the number of fused levels, prescribing longer collar use when more than three levels were involved (mean 7.58 weeks). The primary reason cited for prescribing a rigid cervical collar was patient comfort and improved perception of safety, reported by 65.1% of surgeons. Correlation analysis showed no significant association between years of professional experience and collar use (Spearman’s Rho = -0.121, p = 0.209). Comparisons between practice settings revealed no significant differences, although surgeons working exclusively in private practice demonstrated a greater tendency to prescribe collars (32.4% vs. 18.4%, p = 0.39). Conclusion: This national survey demonstrates that rigid cervical collar use after cervical fusion remains a frequent practice among spine surgeons in Mexico, despite the absence of standardized prescription guidelines. The most influential factors in decision-making were pathology type (traumatic vs. degenerative), osteosynthesis material, number of levels fused, and the presence of comorbidities, particularly osteoporosis. Professional experience and practice setting did not significantly influence prescription patterns. Importantly, patient comfort and perceived safety remain key reasons for prescribing collars, highlighting the subjective elements involved in postoperative care. These findings underscore the need to develop evidence-based clinical guidelines and establish national consensus to unify practice standards regarding postoperative immobilization in cervical fusion patients.

ID: AGS26/944

A045: ERAS - An opportunity for spinal surgery - Findings from a systematic review

Caroline Büchel 1 , Caroline Treanor 2 , Benjamin Davies 3 , David Anderson 4 , Michael Fehlings 5 , Carl Zipser 1

1 Balgrist University Hospital, Zurich, Switzerland ,

2 The school of Physiotherapy, Royal College of Surgeons in Ireland, 123 St Stephens Greens, Dublin, Dublin, Ireland ,

3 Division of Neurosurgery, Department of Clinical Neurosciences, University of Cambridge, Cambridge, United Kingdom, Cambridge, United Kingdom ,

4 School of Health Sciences, Faculty of Medicine and Health, University of Sydney, Sydney, Australia., Sydney, Australia ,

5 Division of Neurosurgery and Spinal Program, University of Toronto and Krembil Brain Institute, University Health Network, Toronto, ON, Canada., Toronto, Canada

Introduction: Enhanced Recovery After Surgery (ERAS®) is a multimodal perioperative care pathway providing evidence-based recommendations to improve functional outcomes after surgery. ERAS pathways have been implemented in various surgical disciplines, including spine surgery. A consensus statement for lumbar spinal fusion has been published by the ERAS® society in 2021, yet recommendations for cervical spine surgery spinal fusion are lacking. Since guideline development depends on precise information on the ERAS landscape, we conducted a systematic review to identify areas of spinal surgery where ERAS was investigated, as well as the applied interventions and outcomes of these studies. Knowledge about the spinal surgery sub-specialities using ERAS will help to inform the development of future ERAS guidelines in spine surgery. Material and Methods: We conducted a systematic review (PROSPERO no. CRD42025638293) evaluating primary research articles which were available in English. Study inclusion was based upon screening of title and abstract, with no limits set regarding publication date. A SWiM analysis of the efficacy of ERAS in comparison to traditional care was done for studies with a control group. Additionally, a meta-analysis for RCTs evaluating ERAS in the field of lumbar spine surgery was performed. Results: Eighty-one studies were included, of which a majority were cohort studies (n = 62/81, 76.5%), followed by RCTs (n = 7/81, 8.6%). Most articles addressed the use of ERAS in the field of degenerative spine pathologies (n = 35/81, 43.2%), followed by deformity (n = 29/81, 35.8%). The median number of interventions implemented was twelve. Interventions most commonly used were patient education (n = 60/81, 74.1%), recommendations on urinary catheter usage (n = 51/81, 63%), postoperative analgesia and early mobilisation (both n = 61/81, 75.3%) and early oral intake (n = 49/81, 60.5%). The most frequently assessed outcome measures were length of stay (LOS) (n = 65/81, 92.9%) and complication rates (n = 40 /81, 49.4%). A comparison of LOS between ERAS and traditional care (TDC-) patients demonstrated a shorter LOS for the ERAS group in 56 studies (93.3%, n = 60). Similarly, meta-analysis showed a statistically non-significant tendency for ERAS shortening LOS for 1 day in lumbar spine patients [95%CI -2.77, 0.71; p = 0.25]. Decreased complication rates, with a median complication rate of 8.8% for ERAS patients vs. a median of 15.6% for TDC patients, were noted. Total hospital costs were reduced in 15 articles (93.8%, n = 16), with a median total cost reduction of 1161.95$. Conclusion: Most evidence exists for the use of multimodal interventions in degenerative spine and deformity surgery, focusing on reducing LOS and complication rates, and optimizing pain management. ERAS interventions appear to effectively improve functional outcomes and reduce costs. There is need for further investigation of personalised ERAS recommendations based upon a patient diagnosis. Furthermore, despite being frequently studied, optimising pain management is a challenge, partially owed to unspecific pain assessments.

OP6: Biologics and Bone Regeneration and Beyond

ID: AGS26/1503

A046: Proteome changes of structural compromised annulus fibrosus collected from degenerative disc patients

Taylor Bader 1 , Mohammed Salaam 1 , David Hart 1 , Holly Sparks 1 , Paul Salo 1 , Antoine Dufour 1 , Ganesh Swamy 1

1 University of Calgary, Calgary, Canada

Introduction: Intervertebral discs (IVDs) are comprised of a central gelatinous nucleus pulposus (NP) that is sheathed in layers of fibrocartilaginous annulus fibrosus (AF) and endplates that together act as the principal connection between vertebrae. The IVD is primary composed of core matrisome proteins, including collagens and proteoglycans. Due to limited perfusion and sparse cellularity, IVDs lose their structural integrity through progressive breakdown as individuals age. Disc degeneration (DD) is an escalation of this breakdown through proteolytic extracellular matrix (ECM) remodeling that potentially leads to chronic pain, spinal deformities, and highly invasive surgery. Despite its prevalence, mechanisms of DD remain poorly defined, especially within the AF. The aim of this study is to apply a multifaceted approach combining mechanics, structure, and proteome analysis of human AF explants to provide a robust characterization of degenerative changes. We compared age-matched controls to DD individuals who had structurally compromised AF tissue measured by reductions in shear modulus (SM) and histological degeneration grading. We hypothesize that accompanying these changes are changes in the core matrisome of the AF. Material and Methods: AF tissues were obtained from minimally degenerative (nonDD) donors through the Southern Alberta Tissue Donation Program with lumbar radiographs to ensure normal morphology. DD AF tissue was obtained from patients undergoing spinal surgery at Foothills Medical Center (Ethics ID: REB 18-1308) with standing X-rays to confirm no cofounding deformity diagnosis and T2-weighted MRI images to grade degeneration using a modified Pfirrmann grade. From a larger cohort of biomechanically assessed tissues, an age-matched cohort was generated using a propensity matching algorithm (matchit) in R. The cohort included anterior AF tissue from 5 nonDD (37 ± 8; 2 females, 3 males) and 5 DD (37 ± 7; 3 females, 2 males) individuals. Each AF tissue was divided into three adjacent 5mm 3 sections, two for biomechanical assessment and histological processing, and one frozen for proteome analysis. Mechanical shear testing of anterior 5mm 3 AF tissue cubes was performed in circumferential (left-right) and radial (anterior-posterior) shear loading. Four strain intervals of 2.5% with 20-minutes of relaxation were applied, totaling 10% strain. Shear modulus (SM) was calculated using a line of best fit between each of the most relaxed points. These same cubes were then embedded and sectioned for hemoxylin and eosin (H&E) staining and histology scoring. Degeneration was scored followed ORS Spine standardized histological scoring, using only the AF section. AF cellularity, lesions, and extracellular matrix were scored between 0-3, with higher scores noting increased degenerative changes. Two-sample two-tailed t-tests were used to compare differences in SM and histology degradation scores between groups and sexes. Proteomic analysis used quantitative detergent solubility profiling (QDSP). Soluble proteins were separated using five total buffers, each increasing in concentrations of detergents. Buffers 0 & 1, non-ionic detergents that gently isolate soluble proteins, were combined to form fraction 1 (soluble proteins). Buffers 2 & 3 included ionic detergents to denature proteins and were combined to form fraction 2 (partially soluble proteins). Lastly, the remaining pellet was isolated as fraction 3 (insoluble proteins). Fractions 1, 2, and 3 of each AF tissue included a bioinformatics pipeline of LC/MS-MS, data-independent analysis (DIA), DIA neural network (DIA-NN) spectra matching, STRING-db pathway analysis, and MSstats statistics (adjusted p < 0.05, log 2 fold-change > 2). Normalized intensities of nonDD and DD compared changes in proteins and protein solubility between fractions. Results: SM was significantly reduced in the DD individuals (radial SM: 7 ± 4; circumferential SM: 32 ± 21 kPa) when compared to age-matched nonDD individuals (radial SM: 59 ± 35; circumferential SM: 195 ± 65) (p < 0.05). Radial SM was significantly increased in nonDD females compared to nonDD males (p < 0.05). Histological degradation scores were significantly increased in DD individuals (4 ± 1) when compared to nonDD individuals (2 ± 1) (p < 0.05). 1083 proteins were identified across all the samples. Notable elevated collagens in soluble (fraction 1) nonDD AF tissue included a 2-fold increase in COL1A1, COL1A2, and 2.5-fold increase in COL4A2 when compared to DD AF tissue. Insoluble matrisome proteins (fraction 3) upregulated in nonDD individuals included COL5A1 (3-fold), COL5A2 (2-fold). Additional changes were seen in proteoglycans, glycoproteins, and ECM regulators.

Conclusion: Collection of IVD tissues from patients undergoing surgery for these painful conditions provides the opportunity to explore mechanical, biological, and structural properties of ex vivo human IVD. Although not accounting for the potential inflection point (when the degeneration and pain initially develop) in DD, understanding the cross-sectional differences of age-matched individuals provides a unique understanding into potential pathological remodeling.

ID: AGS26/1689

A047: Does the fetus intervertebral disc acquire microbiome during the development? A metagenome analysis

S. Rajasekaran 1 , Chitraa Tangavel 2 , Divya Arunachalam 2 , Nandakumar M 2 , Gnanaprakash Gurusamy 1

1 Ganga Hospital and Medical Centre, Coimbatore, India , 2 Ganga Orthopaedic Research and Education Foundation, Coimbatore, India

Introduction: The microbiome of the fetal Intervertebral disc (IVD) is unexplored till date and this study is attempted to document their microbiome. The IVD in healthy adults is avascular; the route of microbial entry inside the disc needs to be addressed. The objective of this study is to understand whether the microbial inoculation happens during the development of the fetus and the type of microbial species it acquires from the mother. Also to understand whether the fetuses acquire similar microbiomes. The outcome of this study is to identify and list the microbial population acquired and explore the bacterial diversity in the IVD components: Nucleus pulposus (NP); Annulus fibrosus (AF); Endplate (EP). This study would answer when the IVD acquires its microbiome. Material and Methods: Nine fetal IVD encompassing NP (n = 8), AF (n = 9), and EP (n = 9) components. 16S rRNA sequencing was used to study the bacterial diversity in the nine fetal IVDs. The tissue sample from IVD components NP (n = 8), AF (n = 9), and EP (n = 9) were used for microbial DNA extraction. Universal bacterial primers were used for V3-V4 region amplification and were sequenced in three batches. The sequenced reads were analysed using QIIME-2 pipeline. ConQuR tool was used to mitigate the batch effect. The alpha and beta diversity were analysed for the inter and intra species diversity among the IVD components. Results: A total of 33 bacterial genus were identified with sample positivity above 25%. In AF, EP and NP 14, 20 and 12 bacterial genus were found respectively. Among them ten bacterial genus Aeromonas, Albidovulum, Arthrobacter, Bacillus, Corynebacterium, Propionibacterium, Pseudomonas, Staphylococcus and Streptococcus were the common among all the fetal IVD-NP, AF and EP components. Diversity analysis shows higher alpha diversity in AF when compared to EP and NP. Beta diversity analysis shows significant difference (p = 0.001) in the microbial composition of NP, AF and EP meaning the microbiome is specific to each component according to the microenvironment even during the developmental stage. Conclusion: The human IVD is not sterile, it acquires the microbiome during the developmental stage as evidenced in this study. The bacterial genus observed among the AF, EP and NP correlates with the previous studies reported in normal adult IVD components. The species richness is more in AF compared to NP and EP. The fetus during the developmental stage is vascular with rich blood supply and would be the route for this microbial inoculation during the IVD development. This study is the first to document the presence of bacterial microbiome in the IVD components.

ID: AGS26/1715

A048: A semi-synthetic oxysterol (Oxy133) as a novel alternative to rhBMP-2 for bone regeneration

Marc Abdou 1 , Matthew Gallo 1 , Nicole Hang 1 , Mirbahador Athari 1 , Hua Shen 1 , Fergui Hernandez 1 , Farhad Parhami 2 , Jeffrey C. Wang 1 , Raymond Hah 1 , Kiran Alluri 1

1 Keck School of Medicine of the University of Southern California, Los Angeles, United States ,

2 MAX BioPharma, Inc., Santa Monica, United States

Introduction: Large bone defects remain a major challenge in orthopaedic surgery, especially when biological capacity for healing is limited. Recombinant human bone morphogenetic protein-2 (rhBMP-2) is widely used but requires supraphysiologic dosing and carries high costs and adverse effects, including heterotopic ossification. Oxysterols are cholesterol-derived signaling molecules with osteoinductive potential. Oxy133, a semi-synthetic oxysterol, has demonstrated strong preclinical performance in spinal fusion models. This study compared Oxy133 to rhBMP-2 in a rat femoral critical-sized defect model. Material and Methods: Twelve-week-old male Lewis rats underwent creation of a 6-mm mid-diaphyseal femoral defect stabilized with plate and cerclage fixation. Animals were randomized to one of three groups: Oxy133 (40 mg/200 µL, n = 14), rhBMP-2 (10 µg/200 µL, n = 14), or DMSO control (n = 5). A collagen sponge scaffold was loaded with the assigned substrate. Healing was evaluated radiographically at 4, 8, and 12 weeks. Microcomputed tomography (µCT) quantified bone volume fraction (BVF). Histology confirmed cortical and trabecular bridging. Biomechanical torsion testing assessed stiffness, torque, and energy to failure. Results: By 8 weeks, 100% of Oxy133-treated defects achieved complete radiographic union versus 86% in the rhBMP-2 group; no control animals healed. At 12 weeks, all but one rhBMP-2 specimen had healed. Radiographic scores averaged 4.7 ± 0.4 (Oxy133) vs. 4.6 ± 0.8 (rhBMP-2). µCT confirmed similar BVF (6.04 ± 2.03% vs. 5.82 ± 1.41%, p = 0.28). Histology showed robust cortical bridging and trabecular bone in both Oxy133 and rhBMP-2 groups, with no bone formation in controls. Biomechanical testing revealed comparable mechanical integrity; rhBMP-2-treated femurs demonstrated greater torsional stiffness (p = 0.04), while other parameters were equivalent. One case of heterotopic ossification occurred in the rhBMP-2 group; none were observed with Oxy133. Conclusion: Oxy133 achieved bone regeneration comparable to rhBMP-2 while promoting earlier union and avoiding complications. Its distinct mechanism of action and lack of adverse events suggest potential safety advantages over rhBMP-2. These findings support Oxy133 as a promising, scalable, and cost-effective alternative for bone regeneration, warranting further evaluation in large-animal models and translational studies.

ID: AGS26/1756

A049: Unveiling the microbiome of intervertebral degeneration with shotgun metagenomics

S. Rajasekaran 1 , Gnanaprakash Gurusamy 2 , Chitraa Tangavel 3

1 Ganga Orthopaedic Research and Education Foundation, Coimbatore, India ,

2 Ganga Hospital and Medical Centre, Coimbatore, India ,

3 Ganga Research Centre, Coimbatore, India

Background: The global human population faces various health challenges including non-communicable diseases such as low back pain (LBP) which significantly impact quality of life. LBP is lifestyle oriented, having high impact on the individuals health comparing other life threatening diseases. It arises from disruptions within the intervertebral discs (IVDs) located between the vertebrae of the spinal cord. The IVD disease leading to degeneration is multifactorial and one such factor is sub-clinical infection hypothesized in our earlier reports (Rajasekaran et al., 2017, 2019, 2020, 2021,2023). As evidenced earlier the disc harbors unique microbiome as this study is proposed to explore the abundance of the entire microbial species through shotgun metagenomics. Methods: Shotgun metagenomics was used to identify the diverse microbial species in IVD. The IVD control tissue samples were collected from brain dead organ donors (n = 20), diseased tissue from patients undergoing surgery diskectomy with phenotypes modic changes (MC) (n = 18) and non-modic (NMC) (n = 22). DNA was extracted (Green & Sambrook, 2014) and shotgun sequencing was performed with Illumina Novoseq platform and the data obtained was analyzed with Kraken 2(v-2.2.3) tool. The relative abundance of the bacterial species were analyzed with Bracken (Bayesian Re-estimation of Abundance after Classification with Kraken). Alpha and beta diversity analysis were performed with statistics using Mann-Whitney U test and PERMANOVA (permutational multivariate analysis of variance) test respectively. The differential abundance were analysed using MaAsLin2 tool. Results: A total of 1816 bacterial species were identified in control, in MC 2141 and in the NMC samples it was 2141 species respectively. With a median relative abundance of > 1% a total of 202 species was identified in the control IVD tissue, whereas 116 and 146 species were found in the diseased MC and NMC tissue samples respectively. The alpha diversity revealed a rich species diversity with a significant difference in the control and NMC samples (p = 0.03) and the beta diversity analysis using PCoA resulted a significant difference (p = 0.002) between the control and diseased (MC & NMC) population. Across all the samples (Control., MC & NMC) the top abundant bacterial species was found to be Klebsiella pneumoniae, Pseudomonas aeruginosa and Corynebacterium amycolatum, Acidovorax sp. BLS4 . However the order of abundance varies according to the phenotypes. The bacterial diversity in diseased samples (MC & NMC) was less.Presence of Pseudomonas aeruginos a phages substantiates the P.aeruginosa as a natural inhabitant of IVD. Conclusion: Bacteria such as Pseudomonas aeruginosa, Stenotrophomonas maltophilia, Burkholderia gladioli, Stutzerimonas stutzeri Ralstonia insidiosa, Leifsonia shinshuensis, and Leifsonia.sp. WHRI.6310E showed a positive correlation in diseased samples there by providing clue leading to dysbiosis. The study evidenced to strengthen our hypothesis that IVDD caused by infection due to the P.aeruginosa abundance. As it facilitates co-bacterial species to polymicrobial infection which may influence the disease progression by modulating the host immunity and the IVD microbiome. Further richness of Psedomonas phages in disease group substantially prove P.aeruginosa presence in IVD and its possible role in IVDD.

ID: AGS26/1826

A050: Cranial and caudal shifts: a genetic perspective on transitional vertebrae and a unified enumeration algorithm

Kadir Abul 1

1 Department of Orthopaedics and Traumatology, Başakşehir Çam and Sakura City Hospital, Istanbul, Turkey

Introduction: Transitional vertebrae exemplify the interface between developmental genetics and clinical anatomy. Variations such as cervical ribs, hypoplastic T12 ribs, and sacralized L5 are not random anomalies but predictable outcomes of homeotic transformations driven by Hox gene expression boundaries. Understanding these cranial and caudal shifts offers a basic science framework for redefining spinal enumeration and addressing the persistent challenge of wrong-level surgery. Materials and Methods: We conducted a conceptual review synthesizing developmental biology, anthropological evidence, and radiological studies of transitional vertebrae. Emphasis was placed on the principle of Hox gene colinearity, which organizes vertebral identity along the cranio-caudal axis. We analyzed reported correlations between rib morphology, thoracolumbar transitional vertebrae (TLTV), lumbosacral transitional vertebrae (LSTV), and their cranial counterparts such as C7 transverse apophysiomegaly and cervical ribs. These observations were used to propose a unified spinal enumeration algorithm. Results: A reproducible pattern emerged across all three transitional zones. Caudal shifts, such as elongated T12 ribs forming accessory ribs at L1, frequently co-occurred with distinct S1-S2 disc spaces. Conversely, cranial shifts, such as hypoplastic or absent T12 ribs, correlated with sacralization of L5, including Castellvi-type variants. These lumbosacral changes were often mirrored in the cervical spine, where hypertrophic C7 transverse processes or cervical ribs were present. Together, these findings confirm that transitional vertebrae represent systematic, genetically patterned transformations rather than isolated anomalies. Conclusion: Cranial and caudal shifts can be understood through developmental genetics and consistently identified on imaging. We propose a unified spinal enumeration algorithm that defines the 20th vertebra as L1 and the 25th as S1, independent of rib morphology. Whole-spine imaging remains the gold standard for accurate numbering, yet awareness of these genetic and morphological patterns allows surgeons to anticipate anomalies even on limited radiographs. Integrating these insights into clinical practice may reduce wrong-level surgery and harmonize vertebral nomenclature.

ID: AGS26/214

A051: Serum deprivation induces autophagic flux and apoptosis in human nucleus pulposus cells: implications for intervertebral disc degeneration and potential therapeutic strategy

Jong-Beom Park 1 , Young-Yul Kim 1

1 The Catholic University of Korea College of Medicine, Seoul, South Korea

Introduction: Nutritional deprivation is a critical pathological factor in intervertebral disc degeneration (IVDD), given the avascular nature of the disc and its dependence on diffusion through the endplate for nutrient supply. This study investigated the effects of serum deprivation on autophagic flux and apoptosis in human nucleus pulposus (NP) cells, with implications for regenerative strategies in IVDD. Material and Methods: Hima NP cells were isolated from degenerated human lumbar disc tissues which surgically obtained from surgeries of lumbar disc herniations. NP cells were cultured under varying fetal bovine serum (FBS) concentrations (10%, 1%, and 0%) to simulate progressive nutrient deprivation. Morphological changes, cell viability, DNA content, and metabolic activity were evaluated. Western blotting quantified LC3, P62, HMGB1, and cleaved caspase-3. Apoptotic cell populations were measured by fluorescence-activated cell sorting (FACS). Immunofluorescence staining for LC3, P62, and cleaved caspase-3 assessed autophagic and apoptotic localization. RT-qPCR was performed to analyze expression levels of autophagy-related genes (LC3, P62) and apoptosis-related genes (Caspase-3). Results: Serum deprivation induced marked reductions in NP cell viability, altered morphology, and decreased metabolic activity in a dose- and time-dependent manner, with maximal effects at 48 hours in 0% FBS. Lower serum concentrations triggered increased autophagic activity, evidenced by LC3-II accumulation and P62 degradation, as well as enhanced expression of LC3 mRNA. Chloroquine treatment confirmed increased autophagic flux by preventing lysosomal degradation of LC3-II. HMGB1 translocated from nucleus to cytoplasm under nutrient stress, further supporting autophagy activation. Concurrently, caspase-dependent apoptosis was observed, with elevated cleaved caspase-3 protein and increased Caspase-3 mRNA levels. FACS analysis revealed significant increases in apoptotic cell populations under 0% FBS conditions. Immunofluorescence confirmed co-localization of LC3 and cleaved caspase-3, indicating simultaneous autophagy and apoptosis under nutrient deprivation. Conclusion: Serum deprivation robustly induces autophagic flux and caspase-dependent apoptosis in human NP cells, highlighting the critical role of nutritional supply in disc cell survival. The interplay between autophagy and apoptosis under nutrient stress may represent a key mechanism in IVDD progression. Clinically, these findings underscore the potential of therapeutic strategies aimed at restoring nutrient microenvironments or modulating autophagy-apoptosis balance-such as enhancing endplate permeability, supplying metabolic substrates, or pharmacologically regulating autophagy-to preserve NP cell viability and promote intervertebral disc regeneration.

ID: AGS26/2978

A052: Risk of peritoneal organs violation from two different entry points for lumbar endoscopic surgery: an imaging study

Rafael Barreto Silva 1 , Gabriel Pokorny 2 , José Carlos Barbi Gonçalves 1 , Luís Eduardo De Lábio Parra 1 , Carlos Almelunge 1 , Thiago Coutinho 1

1 Instituto de Coluna de Campinas, Campinas, Brazil ,

2 Independent Researcher, São Paulo, Brazil

Introduction: In an attempt to find an ideal entry point for each patient, some authors have proposed the use of preoperative CT or MRI scans to identify the location of the retroperitoneal structures and then assess how horizontal a lateral access could be. However, one of the main difficulties with this analysis is that the retroperitoneal structures move to a more anterior position, between the dorsal position (CT or MRI position) and the prone position (surgical position). The aim of this study is analyze the position of the retroperitoneal structures and their movement using CT scans in prone and dorsal decubitus, as well as to verify the theoretical risk of retroperitoneal injury at each of the following entry points. Methods: The inclusion criteria was as follows, Patients over the age of 18, who have consented to take part in the study by filling in the free consent form, and who have no medical contraindications to undergoing CT scans. And the exclusion criteria was patients with any pathology that makes it impossible to take the measurements proposed in the study. Patients who, for whatever reason, have a low-quality image exams. The risk of violation of peritoneal content was assessed both in dorsal and ventral decubitus from two different starting points (red line and green line). Data was compiled in a spreadsheet and analyzed using R software. Categorical variables were described as count and percentage. Population-averaged effects were estimated while accounting for within-patient correlation and sparse cells by fitting a penalized GEE for small-sample with bias-adjusted standard Effects were summarized as estimated marginal means and pairwise contrasts (via emmeans) on the probability scale, with the GL versus RL contrast estimated within each level × decubitus. For robustness, cluster-bootstrap (1,000 resamples) percentile confidence intervals were obtained. Results were expressed as odds ratios and marginal probabilities with 95% confidence intervals. Results: 28 patients were included in the work, of those 16 were female (57%), with the mean age of 39 years old. In a penalized GEE with MB small-sample variance, we found significant main effects of reference line (F(1,27) = 105.8, p < 0.0001), spinal level (F(3,27) = 5.2, p = 0.006), and decubitus (F(1,27) = 80.0, p < 0.0001), but not side of approach (F(1,27) = 1.0, p = 0.336). Also, there was a significant interaction between spine level and decubitus (F(3,27) = 6.6, p = 0.002). As for the pairwise estimated marginal means comparison regarding the risk of peritoneal violation in ventral decubitus. At L1-L2 , the risk with the GL was 4% (95% CI 0.7-20%) versus 61% (46-73%) with the red line ( p < 0.0001 ). At L2-L3 , the GL yielded 0.6% (0.4-9%) compared with 36% (21-53%) for the RL ( p < 0.0001 ). The same pattern held at lower levels: 0.6 % vs 14% at L3-L4 and 7 % vs 45% at L4-L5 . Conclusion: Across all lumbar levels and in both positions, the GL reference line was associated with substantially lower risk of peritoneal violation than RL . Interactions involving the access point were not significant, indicating a stable GL benefit across levels and positions.

ID: AGS26/458

A053: Nasal chondrocyte spheroids survive and integrate in degenerated intervertebral disc

Raphael Schmid 1,2 , Janhavi Apte 2 , Elias Schulze 3 , Igor Červenka 2 , Salim Darwiche 4 , Karina Klein 4 , Andrea Barbero 2 , Stefan Schären 1 , Olga Krupkova 1,2 , Karoliina Pelttari 2 , Ivan Martin 2 , Arne Mehrkens 1,2

1 Spine Surgery, University Hospital Basel, Switzerland, Basel, Switzerland ,

2 Department of Biomedicine, University of Basel & University Hospital Basel, Basel, Switzerland ,

3 Institute of Pathology, University Hospital Basel, Basel, Switzerland ,

4 Musculoskeletal Research Unit (MSRU), University of Zurich, Zurich, Switzerland

Introduction: Painful intervertebral disc disease (IDD) is a prevalent condition associated with catabolic degenerative changes in the intervertebral disc (IVD). Current treatments provide limited relief, underscoring the need for innovative therapies capable of functioning in the harsh IVD microenvironment. Nasal chondrocytes (NC) have shown promise in adapting to degenerated heterotopic transplantation sites, where they form cartilaginous matrix. This study explores the use of NC-derived 3D spheroids (NCS) for cell-based repair of degenerated IVDs. Material and Methods: NCS were generated from human or sheep NC and metabolically preconditioned (at reduced glucose and oxygen levels) in medium containing interleukin-1 receptor antagonist (IL-1Ra), to enhance their resilience in target microenvironment low in oxygen and nutrients. Proteomic analysis of preconditioned NCS (pNCS) responses to IDD-mimicking conditions was conducted. pNCS were then injected into: (i) native human or bovine degenerated IVDs ex vivo and (ii) a sheep model of Chondroitinase ABC (ChABC)-induced IDD in vivo. Post-injection pNCS integration and chondroprotective effects in native IVD were assessed. Results: Compared to non-pNCS, pNCS underwent chromatin changes, increased glycolytic activity, and reduced proliferation, while maintaining the expression of chondroprotective proteins (Aggrecan, Collagen type II, SOX-9). Sustained release of IL-1Ra from pNCS prevented the production of inflammatory mediators in IDD microenvironment. In ex vivo degenerated IVD cultures, pNCS survived, and NC spread into the surrounding NP tissue within 7 days of injection. In a pilot sheep study (n = 2), ChABC caused moderate IDD by 8 weeks, at which point pNCS were injected. Following injection, pNCS/NC localized in the IVD, spread, and survived for at least 2 months. Conclusion: Following pre-conditioning, NCS were able to adapt to the IDD microenvironment and remained protected from pro-inflammatory insults. NCS survived and integrated into native degenerated IVD tissue in both ex vivo and in vivo models. A larger in vivo study (n = 9 sheep/18 IVDs) is currently ongoing to further evaluate the safety and efficacy of this approach.

ID: AGS26/3138

A054: Molecular expression of leptin, leptin receptor variant-2, and FGF23 in hypertrophic bone from lumbar spinal stenosis patients: a novel bone-based qPCR profiling study

Ozan Aydogdu 1 , Resat Unal 2 , Derya Hazer Rosberg 3

1 Muğla Eğitim ve Araştırma Hastanesi, Mugla, Turkey

2 Department of Molecular Biology and Genetics, Molecular Cell Biology Division, Faculty of Science, Mugla Sitki Kocman University, Mugla, Turkey

3 Lund University Skåne, University Hospital, Lund, Sweden

Introduction: Obesity is an established risk factor for lumbar spinal stenosis (LSS). While prior studies have evaluated molecular mechanisms in ligamentum flavum hypertrophy, bone-derived molecular data in LSS remain lacking. This study investigated bone expression of Leptin (LEP), Leptin Receptor Variant-2 (LEPR-V2), and fibroblast growth factor-23 (FGF23) in patients undergoing surgery for LSS. To our knowledge, this is the first bone-based molecular profiling of these markers in LSS surgical specimens. Materials and Methods: This prospective study included patients who underwent decompression for LSS at Muğla Sıtkı Koçman University Education and Research Hospital between 2018 and 2019. Patients were stratified by BMI into Control (≤ 24 kg/m 2 ) and Experimental (> 24 kg/m 2 ) groups. Hypertrophic bone samples were harvested intraoperatively, RNA was extracted, converted to cDNA, and quantitative PCR (qPCR) performed for LEP, LEPR-V2, and FGF23. Relative expression levels were calculated by the 2^ΔΔCt method. Spinopelvic parameters including lumbar lordosis, pelvic tilt, sacral slope, pelvic incidence, and Cobb angle were measured pre- and post-operatively from standing lateral radiographs. Results: Thirty-seven patients were analyzed (Control: n = 11; Experimental: n = 26). Molecular markers were non-normally distributed. Across all patients, BMI was not significantly correlated with LEP (ρ = -0.079, p = 0.657), LEPR-V2 (ρ = 0.254, p = 0.148), or FGF23 (ρ = -0.089, p = 0.615). Between BMI groups, no significant differences were detected (LEP p = 0.174; LEPR-V2 p = 0.326; FGF23 p = 0.850). However, a non-significant upward tendency in LEP and LEPR-V2 expression was observed in higher BMI patients, suggesting a potential link between obesity-related signaling and hypertrophic bone changes in LSS. Conclusion: This study provides the first evidence of bone-derived molecular expression of LEP, LEPR-V2, and FGF23 in LSS. Although no statistically significant associations with BMI were demonstrated, the observed tendency for higher LEP and LEPR-V2 expression in patients with elevated BMI highlights a possible mechanistic role of adipokine pathways in hypertrophic bone remodeling. These results extend prior ligamentum-based findings and underline the importance of further multicenter studies integrating larger cohorts to validate the contribution of obesity-related molecules in LSS pathogenesis.

OP7: Preventing Junctional Problems

ID: AGS26/191

A055: Too much of a curve? Rethinking spinal alignment strategies for high pelvic incidence patients

Devender Singh 1 , Matthew Geck 1 , John Stokes 1 , Eeric Truumees 1

1 Ascension Texas Spine and Scoliosis, Austin, United States

Introduction: Sagittal alignment is essential for spinal stability, particularly in patients with high pelvic incidence (PI > 60°). Achieving ideal sagittal balance in these patients remains challenging due to the complex interaction between pelvic morphology and spinal curvature. Existing alignment strategies may fail to provide optimal balance, leading to postural compensation and increased mechanical stress on the spine. This study assesses the effectiveness of three alignment strategies in high PI patients, evaluating compliance with key sagittal alignment parameters. We hypothesize that higher PI patients will experience greater difficulty in achieving optimal sagittal alignment due to increased pelvic morphology demands. Among the three alignment strategies (PI-LL = 10, PI-LL = 0, PI-LL = -10), certain strategies will fail more frequently, particularly in aligning C2PL and T1 slope. Material and Methods: A cohort of 20 patients with PI > 60° underwent computational modeling of three sagittal balance strategies: PI-LL = 10° (less_curve), PI-LL = 0° (ideal_curve), and PI-LL = -10° (more_curve). Each strategy was tested using three alignment goals: C2PL = 0 mm, T1PA = 10°, and T1PA = 15°. Plans were analyzed based on three criteria: T1 slope between 15°-35°, T1 pelvic angle (T1PA < 20°), and C2PL between +9 mm and -33 mm. Results: The results revealed significant variability in compliance rates across strategies. The highest compliance was observed in the C2PL = 0 mm strategy, reaching 35%, whereas the lowest compliance occurred in the T1PA = 10° strategy, with only 0%. The most frequently failed parameter was T1 Slope 15°-35, which showed failure rates exceeding 100% in some cases. Patients with high PI exhibited significant difficulty in maintaining T1 slope within the ideal range, suggesting that standard alignment strategies may not fully compensate for the increased lumbar lordosis demands of these individuals. Failure to align sagittal parameters correctly could lead to increased thoracolumbar stress, heightening the risk of postoperative instability and adjacent segment disease. Conclusion: These findings suggest that conventional spinal alignment strategies may not fully accommodate the postural needs of high PI patients. The inability to achieve consistent sagittal alignment underscores the need for refined planning techniques that incorporate global spinal balance adjustments, particularly in the thoracolumbar region. Future studies should explore dynamic models that integrate thoracic and pelvic compensatory mechanisms for improved surgical outcomes.

ID: AGS26/965

A056: The impact of post-operative sagittal alignment on the development of proximal junctional kyphosis following adult spinal deformity surgery: a retrospective cohort analysis

Puya Alikhani 1 , jade vanderpool 1 , Dana Saleh 1 , Ahmad Alkadri 1 , Diego Soto Rubio 1 , Mohsen Rostami 1

1 University of South Florida, Tampa, United States

Introduction: Proximal Junctional Kyphosis (PJK) is a common and debilitating complication following long-segment fusion for adult spinal deformity (ASD). While numerous risk factors exist, the role of achieving optimal post-operative sagittal alignment remains a critical area of investigation. Material and Methods: A retrospective cohort study was conducted on 87 patients who underwent thoracolumbar fusion for ASD. Patients were divided into two cohorts: the PJK Cohort (cases, n = 38), who developed radiographically confirmed PJK, and the Control Cohort (n = 49), who underwent surgery without developing PJK. Pre- and post-operative radiographic parameters-Pelvic Incidence (PI), Lumbar Lordosis (LL), Pelvic Tilt (PT), Sagittal Vertical Axis (SVA), and PI-LL mismatch-were measured. The delta (Δ) for each parameter (post-operative minus pre-operative value) was calculated to quantify the magnitude of surgical correction. Independent samples t-tests were used to compare the mean changes between cohorts. Results: The Control Cohort demonstrated a significantly greater magnitude of improvement in key sagittal alignment parameters compared to the PJK Cohort. The mean improvement in lumbar lordosis (ΔLL: -23.5° vs. -16.2°, p = 0.037) and the mean reduction in pelvic tilt (ΔPT: -7.5° vs. -3.9°, p = 0.042) were both statistically greater in the Control Cohort. A strong trend towards a greater reduction in PI-LL mismatch was also observed in the Control Cohort (ΔPI-LL: -21.1° vs. -14.6°, p = 0.069). No significant difference was found in the change in SVA between groups (p = 0.204). A pattern of residual post-operative malalignment, particularly a PI-LL mismatch > 10°, was frequently observed in the PJK Cohort. Conclusion: Successful restoration of sagittal balance is paramount in preventing PJK. This study demonstrates that patients who developed PJK received a less robust correction of their sagittal deformity, leaving significant residual malalignment. The statistically significant differences in LL and PT correction, and the strong trend in PI-LL mismatch, indicate that inadequate correction shifts excessive biomechanical stress to the proximal junction, leading to mechanical failure. These findings underscore that achieving optimal sagittal alignment is a critical, modifiable surgical objective for mitigating the risk of PJK.

ID: AGS26/1527

A057: Four milligrams per level: defining the BMP threshold to prevent rod fracture and proximal junctional kyphosis in adult spinal deformity surgery

Nuri Demirci 1 , Yasemin Yavuz 2 , Munish Gupta 1

1 Department of Orthopaedic Surgery, Washington University in St. Louis, Saint Louis, United States

2 Department of Biostatistics, Faculty of Medicine, Ankara University, Ankara, Turkey

Introduction: Bone morphogenetic protein (BMP) is widely used to enhance fusion in adult spinal deformity (ASD) surgery, but its optimal dosing strategy remains undefined. The ideal dose per level that decreases the risk of mechanical complications has not been established. Methods: A prospective cohort of 207 ASD patients who underwent thoracic-to-pelvis fusion with ≥ 7 levels and ≥ 2 years of follow-up was analyzed. BMP dose per level was calculated and categorized as < 3, 3-3.99 (< 4), and ≥ 4 mg/level. Outcomes were rod fracture (RF) and proximal junctional kyphosis (PJK), assessed by Kaplan-Meier analysis and multivariable Cox models stratified by osteoporosis. Inverse probability of treatment weighting (IPTW) was applied using a propensity score including age, sex, BMI, osteoporosis status, number of fused levels, PSO, rod material, rod size, accessory/satellite rods, and surgical approach to balance baseline characteristics between high- and low-dose BMP groups. Results: Mean age was 62.1 ± 11.1 years, BMI 29.7 ± 5.7, and 77.8% were female; 20.8% underwent PSO and 6.3% VCR. Over a mean 46-month follow-up, 20 patients (9.7%) developed RF and 32 (15.5%) developed PJK. The mean BMP dose per fused level was 3.2 ± 1.0 mg, with a median of 3.0 (IQR 2.7-3.6, range 0.8-6.0). For RF, crude event rates declined stepwise with higher BMP dose per level: 15.6% with < 3 mg/level, 8.1% with 3-3.99 mg/level, and 3.1% with ≥ 4 mg/level. For PJK, corresponding crude rates were 25.0%, 10.8%, and 12.5%. Kaplan-Meier estimates showed 36-month RF risks of 13.4%, 6.0%, and 0.0% across these dose groups. After IPTW adjustment, 36-month PJK risk decreased from 13.8% with < 4 mg/level to 2.5% with ≥ 4 mg/level (HR 0.29, 95% CI 0.07-1.14, p = 0.076). In Cox regression, higher BMP dose per level was significantly associated with a lower hazard of PJK (HR 0.39 per mg/level, 95% CI 0.23-0.66, p < 0.001) and showed a nonsignificant protective trend for RF (HR 0.75, 95% CI 0.40-1.43, p = 0.15). Conclusion: At 36 months, rod fracture risk decreased from ∼13% with < 3 mg/level to 0% with ≥ 4 mg/level, while proximal junctional kyphosis risk decreased from ∼14% with < 4 mg/level to ∼2-3% with ≥ 4 mg/level. These findings suggest that the ideal BMP dose should be at least 4 mg per level fused meaningfully improves construct durability and lowers late mechanical complications.

ID: AGS26/1784

A058: Effect of fusion construct length on standing balance in adults with spinal deformity

Bethany Wilson 1 , Sara McMahan 1 , Kyle Robinson 1 , Isador Lieberman 2

1 Texas Back Institute, Plano, TX, United States , 2 Texas Back Institute, Frisco, TX, United States

Introduction: Adult Spinal Deformity (ASD) often requires surgical intervention involving fusion with constructs of varying lengths. While longer constructs are needed in some patients to address the extend of the deformity, this results in reduced spinal mobility compared to shorter constructs. While spinal alignment is commonly discussed among physicians dealing with spinal deformity, standing balance is often overlooked in these discussions. Compromised standing balance may negatively impact performance of daily activities and is important to consider when performing spinal fusions. The purpose of this study was to determine if construct length in ASD surgery effected standing balance. Material and Methods: This study was a post hoc analysis of prospectively collected data with 80 patients who underwent surgery for ASD and 35 asymptomatic control subjects (C). Patient reported outcome measures included visual analog scale for midback (VASmb) low back (VASlb), and leg (VASleg) pain, and the Oswestry Disability Index (ODI). Balance measures based on center of pressure (COP) included range of sway in the sagittal and coronal planes (ROSs/ROSc), total sway distance (TSD), sway ellipse area (SEA), and sagittal/coronal sway velocity (SVs/SVc). ASD patients were divided into two cohorts based on construct length: Short Construct (SC) involved 3-6 instrumented levels (n = 40; 12 males/28 females, mean age 63.6 ± 13.3 yrs, mean BMI 28.5 ± 5.6 kg/m 2 ) and Long Construct (LC) involved 7-14 instrumented levels (n = 40; 5 males/35 females, mean age 58.1 ± 17.5 yrs, BMI 27.3 ± 5.2 kg/m 2 ). Balance in these patient groups was compared with that in asymptomatic control subjects (C; n = 35; 15 males/20 females, age 39 ± 13.3yo, BMI 24.7 ± 3.6 kg/m 2 ). Evaluations were conducted preoperatively (P0), 3 months post-surgery (P3), and 1-year post-surgery (P12). At each visit, patients were assessed with VAS and ODI and performed three eyes-open quiet standing tests on force plates to assess COP motion and related measure. Results: ASD patient groups were demographically similar, while C were significantly younger, taller, and had lower BMI. Preoperatively, LC patients reported higher VASmb scores compared to SC patients (SC: 2.7 ± 2.9 | LC: 4.3 ± 3.1), which improved significantly at P3 while SC remained diminished (SC: 1.7 ± 2.5 | LC: 2.4 ± 2.8) and P12 (SC: 1.4 ± 2.3 | LC: 1.8 ± 2.7). Preoperative VASlb (SC: 2 ± 1.9| LC: 6.5 ± 2.9) and ODI (SC: 39.4 ± 14.7 | LC: 38.9 ± 15.7) scores also improved by P3, where VASlb (SC: 6.2 ± 3.2| LC: 2.3 ± 2.4) and ODI: (SC: 28.8 ± 18.4 | LC: 28 ± 16.6). These values were sustained at P12. Preoperatively, SC and LC groups showed significantly worse COP balance measures than C. These include ROSs, ROSc, TSD, SEA, SVs, and SVc. At P3, SC ROSc, ROSs, SEA and SVc were similar to control levels, while LC measures remained impaired. By P12, most SC improvements persisted, remaining comparable to controls, while LC showed no changes. All p < 0.05. Conclusion: Patients with shorter constructs achieved postoperative standing balance comparable to healthy controls, with improvements sustained at one year postoperatively. In contrast, patients with longer constructs did not experience similar balance improvements at any follow-up. These findings offer insights to guide expectations concerning balance after ASD surgery.

ID: AGS26/287

A059: Incidence and risk factors of postoperative motor deficits after spine deformity surgery: results of the SDIM (spinal deformity intraoperative monitoring) study

Juan Sardi 1 , Justin Smith 1 , Brett Rocos 2 , Lawrence Lenke 3 , Ferran Pellisé 4 , Yong Qiu 5 , Ahmet Alanay 6 , Nasir Quraishi 7 , Randolph Gray 8 , Saumyajit Basu 9 , Go Yoshida 10 , Amer Aziz 11 , Stephen Lewis 12

1 Department of Neurosurgery University of Virginia, Charlottesville, United States ,

2 Departments of Neurosurgery and Orthopedic Surgery, Duke University, Durham, NC, United States ,

3 Columbia University Department of Orthopedic Surgery, New York, NY, United States ,

4 Hospital Universitari de la Vall d’Hebron, Barcelona, Spain ,

5 Spine Surgery, Drum Tower Hospital of Nanjing University Medical School, Nanjing, Jiangsu, China ,

6 Department of Orthopedics and Traumatology, Acibadem Mehmet Ali Aydinlar University School of Medicine, Istanbul, Turkey ,

7 Queen's Medical Centre Campus, Nottingham University Hospitals NHS Trust, Nottingham, United Kingdom ,

8 The Children`s Hospital Westmead, St. Leonards, Australia ,

9 Kothari Medical Centre, Kolkata,, India ,

10 Hamamatsu University School of Medicine, Hamamatsu, Japan ,

11 Ghurki Trust Teaching Hospital, Lahore, Pakistan ,

12 Toronto Western Hospital, Toronto, ON, Canada

Introduction: Spinal deformity surgery has advanced significantly over the past two decades, allowing for more ambitious correction of complex pathologies. Despite these advances, procedures remain associated with substantial risk, particularly new-onset postoperative motor deficits (NOPMD), which have profound consequences for patient function, quality of life, and healthcare costs. Reported incidences vary widely, reflecting heterogeneous patient populations, inconsistent outcome measures, and reliance on retrospective data. This analysis aimed to determine the overall incidence of NOPMD and to identify clinical, radiographic, and intraoperative factors associated with their occurrence. We also sought to distinguish cord- from root-level injury patterns and assess the diagnostic accuracy of neuromonitoring. Material and Methods: This was a post-hoc analysis of the SDIM registry, which prospectively enrolled patients aged 10-80 years undergoing spinal deformity correction at 20 international centers. Eligible patients had major scoliosis (Cobb ≥ 80°), high-grade spondylolisthesis, or required posterior column or three-column osteotomies. Patients were stratified by surgical level into cord-level (T12 and above) and non-cord-level (L1 and below) groups. Neuromonitoring included MEP, SSEP, and EMG across standardized muscle groups. NOPMD was defined as ≥ 1-point decrease in ASIA-LEMS immediately after surgery. Alerts were based on ≥ 50% reductions in MEP/SSEP amplitude or sustained EMG activity. Baseline demographic, radiographic, and surgical data were compared between groups, and logistic regression identified independent predictors of NOPMD. Results: A total of 542 patients met inclusion criteria, with 195 undergoing non-cord-level and 347 cord-level procedures. The overall incidence of NOPMD was 10.1% (n = 55), significantly higher in the non-cord-level cohort (15.9%) compared with the cord-level group (6.9%). Deficits most commonly involved proximal lower extremity muscles, particularly hip flexors and knee extensors. At last follow-up, 58.2% of affected patients demonstrated complete recovery, with comparable rates between cord- and non-cord-level groups. IONM alerts were observed in 10.8% of non-cord-level and 16.1% of cord-level cases. Alerts were significantly more frequent in patients who developed NOPMD (29% vs. 7.3% in non-cord-level; 50% vs. 13.6% in cord-level; both p < 0.01). However, sensitivity of IONM for detecting root-level deficits remained poor, ranging from 5.9% to 30.8%, while specificity exceeded 93% across all muscle groups. Univariate analysis identified older age, higher BMI, posterior decompression, interbody device use, and IONM alerts as significant risk factors. On multivariate analysis, only age (OR 1.02, 95% CI 1.00-1.04, p = 0.019) and greater number of monitored muscle groups (OR 1.22, 95% CI 1.08-1.37, p = 0.002) independently predicted NOPMD. Conclusion: In this multicenter, prospectively collected cohort, the incidence of NOPMD was 10.1%, with deficits occurring more than twice as frequently in non-cord-level compared with cord-level spine surgeries. These deficits were predominantly consistent with nerve root rather than spinal cord injuries, most often affecting proximal muscle groups. Despite the use of standardized intraoperative neuromonitoring (IONM), sensitivity for detecting root-level injuries remained poor, underscoring a critical limitation of current practice. Multivariate analysis identified older age and greater surgical complexity as independent predictors. These findings highlight the urgent need for improved neuromonitoring strategies and emphasize meticulous surgical technique and vigilance.

ID: AGS26/3331

A060: Does ideal PI-LL alignment reduce compressive forces at the thoracolumbar junction in normal individuals?

Fatemeh Alavi 1 , Christopher J. Nielsen 1 , Jeffrey Hills 2 , Michael Kelly 3 , Stephen Lewis 1

1 University of Toronto, University Health Network (UHN), Toronto, Canada

2 UT Health San Antonio, San Antonio, United States , 3 Rady Children’s Hospital, University of California San Diego, San Diego, United States

Introduction: Restoring spine alignment with the target of Pelvic Incidence–Lumbar Lordosis (PI-LL) mismatch less than 10° is recommended to achieve good clinical outcomes. However, it exhibits significant variability in normal population, does not account for variability in segmental lumbar lordosis and it is unclear whether this alignment target results in lower spinal forces at the junctions. With compression force at the junction felt to be a leading factor in proximal junctional failure, this study aims to investigate the forces at the thoracolumbar junction in asymptomatic individuals that have alignments with and without ideal PI-LL mismatch. Material and Methods: We used radiographic data of an international cross-sectional study on 319 healthy volunteers over 18 years of age from four continents, without evidence of spinal disease or spine surgery. Institutional review board approval was obtained by participating centers. Radiographic parameters including PI and L1-S1 lordosis (LL) were measured based on EOS images. Individualized musculoskeletal (MSK) models based on sex, weight, height, and spine curvature were built and spinal forces were calculated using MSK analysis. Individuals were divided into two groups based on PI-LL values: Matched (PI-LL<10°) and mismatched (PI-LL>10°). In order to exclude the effect of PI on spinal force, individuals were also clustered in two groups based on PI values: PI<55° and PI>55°; and ideal mismatch was studied within subgroups. Spinal compressive and shear forces at thoracolumbar levels (T12/L1 and L1/L2) were compared between different subgroups. Results: 119 individuals (37.3%) were mismatched with a mean PI-LL of 16 ° ± 4.4°. The PI value of the group with mismatch was 49.9 ± 10.3° vs. 53.7 ± 10.1° in the matched group (p=0.001). Shear forces at L1 and L2 levels were significantly lower in individuals with matched PI-LL(p<0.001), however, compressive forces were not different between the groups (p=0.08, p=0.5). For 121 individuals (37.9%) with PI>55, no significant difference in both shear and compression was found between alignments with and without matched PI-LL values (p=0.36, p=0.78). For the 198 individuals with PI<55° and matched PI-LL, significantly lower shear force at the upper lumbar region was noted compared to the mismatched group (p<0.001), however, compression was not different between the groups. Conclusion: While targeting ideal PI-LL match for individuals with low PI (PI<55°) can be effective in decreasing shear force at the upper lumbar region, it did not result in lower spinal compression forces in all range of normal population. This finding demonstrated the limitation of the ideal PI-LL value as a sagittal alignment target. Further investigation is needed to identify more reliable alignment targets that better reflect the spinal forces at the junctional levels.

ID: AGS26/3385

A061: Adult spinal deformity treated with ≤ 4-level fusion: multicenter retrospective analysis of radiographic, clinical, and HRQoL data

Akinbo Olaolu 1 , Feyzi Kilic 2 , Elif Gizem Carus 2 , Tais Zulemyan 2 , Caglar Yilgor 3 , Ibrahim Obeid 4 , Yann Philippe Charles 5 , Ferran Pellisé 6 , Javier Pizones 7 , Frank Kleinstuck 8 , Altug Yucekul 9 , Ahmet Alanay 9 , European Spine Study Group ESSG 10

1 Department of Neurological Surgery, University College Hospital (UCH), Ibadan, Nigeria ,

2 Comprehensive Spine Center, Acibadem Maslak Hospital, Istanbul, Turkey,

3 Koc University Hospital Spine Center, Istanbul, Turkey ,

4 Spine Surgery Unit, Polyclinique Jean Villar, Brugge, France, Brugge, France ,

5 Service de Chirurgie du Rachis, Hôpitaux Universitaires de Strasbourg, strasbourg, France ,

6 Spine Surgery Unit, Hospital Universitari Vall d'Hebron, barcelona, Spain ,

7 Spine Surgery Unit, Hospital Universitario La Paz, Madrid, Spain, Madrid, Spain ,

8 Spine Surgery Unit, Schulthess Klinik, Zurich, Switzerland., Zurich, Switzerland ,

9 Acibadem Mehmet Ali Aydinlar University School of Medicine, Istanbul, Turkey ,

10 Spine Surgery Unit, Hospital Universitari Vall d'Hebron, Barcelona, Spain

Introduction: Short-segment (≤4 levels) fusion is frequently used for adult spinal deformity (ASD) to restore alignment and function while limiting morbidity. Mechanical complications are well recognized after long-segment fusion; however, their incidence, predictors, and impact in short-segment constructs remain poorly characterized, and the spectrum of these complications, their radiographic/clinical predictors, and their relationship to health-related quality of life (HRQoL) are insufficiently defined. We aimed to describe outcomes after short-segment fusion, compare patients with and without complications, evaluate HRQoL change and MCID attainment, and identify independent radiographic and clinical risk factors for complications. Materials and Methods: We retrospectively queried a multicenter ASD database for consecutive adults undergoing posterior instrumented fusion of ≤ 4 levels. Patients with at least 2-year follow-up were included. Radiographs were analyzed; key parameters included major Cobb, apical translations, RPV/RLL/RSA, segmental/lumbar lordosis and thoracic kyphosis, and GAP/Roussouly categories. HRQoL instruments (ODI, COMI Back, SF-36 PCS/MCS, SRS-22 subtotal/total) were collected from baseline to last available follow-up (“Last”). Group comparisons used Mann-Whitney U and Fisher/χ 2 tests. Candidate predictors (p < 0.10) entered multivariable logistic regression to model “any complication”; distinct clinical and radiographic models were built. Discrimination and calibration were summarized by AUC and Brier score. Results: A total of 138 (F/M) 102/36 patients were analyzed. Age at surgery was 64.5 ± 12.8 years and follow-up was 59 months (median; 24-159). (73.9%/26.1%). Primary diagnoses were degenerative in 72.3%, idiopathic in 17.5% and other in 10.2%. 22.5% of the patients had prior spine surgery. Total surgical time was 235.1 ± 120.0 min; operative blood loss was 538.1 ± 471.2 mL; total transfused blood was 169.2 ± 327.6 mL with a mean 9.96 ± 8.47 days hospitalization. 38 (27.5%) patients experienced ≥ 1 complication. Complications were 26.3% Implant-related problems, 23.7% adjacent segment disease, 18.4% fractures (pedicle /sacrum), 15.8% junctional kyphosis (PJK/DJK), 7.9% pseudarthrosis and 7,9 others, respectively. In univariate radiographic comparisons, early postoperative (6W) residual-deformity measures, between non-complicated vs complicated groups were RPV (-9.45 ± 6.91, -4.99 ± 5.35; p = 0.002), PT apical translation to C7 (1.54 ± 2.45, 6.52 ± 5.87; p = 0.013), GAP score (6.52 ± 4.02, 4.03 ± 3.55; p = 0.003), and Major Cobb (28.20 ± 15.49, 21.84 ± 13.36; p = 0.032). Baseline GAP and Roussouly category distributions did not differ between groups. From baseline to Last, both groups statistically improved across HRQoL instruments. Between-group testing COMI Back change (non-complicated: Δ = 3.42 ± 2.46, p < 0.001 and complicated: Δ = 5.07 ± 1.64, p < 0.001) ( p = 0.046) and SF-36 PCS (Δ = 3.73 ± 11.82, p = 0.465 and Δ = 13.38 ± 9.71; p = 0.002, respectively) (p = 0.029) were significant, while differences in ODI and SRS-22 changes were not significant. MCID attainment for ODI, SF-36 PCS, and SRS-22 was observed in both groups. The radiographic multivariable model retained two 6W predictors-RPV (OR = 1.11 per unit; 95%CI 1.03-1.20; p = 0.004) and proximal thoracic apical translation to C7 (OR = 1.34 per unit; 95%CI 1.06-1.71; p = 0.016)-with acceptable discrimination (AUC = 0.73, Brier = 0.17). A clinical model identified prior spine surgery and current tobacco use as independent risk factors (AUC = 0.67; Brier = 0.17). Conclusion: In ≤ 4-level ASD fusions, approximately one quarter of patients experienced a complication, yet clinically meaningful improvements in pain, disability, and disease-specific HRQoL were achieved in both groups. Even in this deformity cohort not primarily addressed with extensive corrective surgery, patients achieved substantial, complication-independent improvements in HRQoL. Prior spine surgery and active smoking emerged as the strongest clinical risk factors for complications. Future comparative studies should clarify selection drivers and outcomes in similarly deformed patients treated with short-segment versus long-segment fusion.

ID: AGS26/537

A062: The effect of preoperative rehabilitation on adult spinal deformity patient outcomes, patient recovery kinetics, and cost effectiveness

Peter Passias 1 , Kyriakos Chatzis 1 , Mohammad Daher 1 , Francesca Totis 1 , Peter Tretiakov 1 , Anthony Baumann 2 , Tobi Onafowokan 1 , Max Fisher 1 , Ethan Cottrill 1

1 Duke University, Durham, United States , 2 Cleveland Clinic, Ohio, United States

Introduction: Adult spinal deformity (ASD) is a debilitating disorder of the lumbar and thoracic spine that significantly impacts patient quality of life. Surgical correction is a proven treatment option that provides functional restoration and pain relief. Preoperative rehabilitation programs have recently been implemented to prepare patients for surgery and to promote patient health. The purpose of this study was to identify the effect of preoperative rehabilitation on surgical ASD patient outcomes. Material and methods: This is a retrospective analysis of a consecutive series of 899 ASD Patients. ASD patients with baseline (BL) and two-year (2Y) follow-up were included if they had preoperative rehabilitation data. Patients were divided into 2 groups: those who had preoperative rehabilitation (Prehab+) and those who did not (Prehab-). Prehab consisted of physical and mental components. Patients were excluded if they presented with any of the following at BL: severe neurological deficit ( < 3/5), minimal ambulation, or current depression/anxiety. Normalized HRQL scores at BL and follow-up intervals (6W, 1Y, 2Y) were generated. Normalized HRQLs were plotted and area under the curve was calculated, generating one number describing overall recovery (Integrated Health State [IHS]). Cost was calculated using the PearlDiver database. Cost per Quality-Adjusted Life Year (QALY) at 2Y were calculated. Means comparison tests and Multivariable regression analysis assessed patient reported outcomes and cost adjusting for baseline and surgical characteristics. Results: The study included 464 patients, with 186 in the preoperative rehabilitation group (Prehab+) and 278 in the non-rehabilitation group (Prehab-). Prehab+ patients were older (65.5 ± 11.8 vs. 57.2 ± 14.9 years; p < 0.001), had a higher Charlson Comorbidity Index (CCI) (2.2 ± 1.7 vs. 1.5 ± 1.6; p < 0.001), and a higher BMI (28.4 ± 6.0 vs. 26.2 ± 5.0; p < 0.001). They also exhibited higher rates of comorbidities such as hypertension, arthritis, and diabetes compared to the Prehab- group. Despite these differences, Prehab+ patients achieved significantly better outcomes. Prehab+ patients had shorter lengths of stay (LOS) (7.0 ± 4.3 vs. 7.8 ± 4.1 days; p = 0.045) and lower estimated blood loss (1177.0 ± 1325.3 ml vs. 1532.8 ± 1410.7 ml; p = 0.006). Transfusion rates were also lower in the Prehab+ group (54% vs. 67%; p = 0.003). Early discharge was more common in the Prehab+ group, with significantly more patients discharged within three days (13.0% vs. 3.2%; p < 0.001). SICU admission rates were lower in Prehab+ patients (51.6% vs. 65.4%; p = 0.003). Prehab+ patients had a lower rate of overall complications (69.9% vs. 78.4%; p = 0.038) and medical complications (16.7% vs. 24.8%; p = 0.036). The reoperation rate was significantly reduced in the Prehab+ group (18.3% vs. 33.5%; p < 0.001). Multivariate analysis confirmed that Prehab+ patients had a lower risk of complications (OR: 0.6; 95% CI: 0.42-0.87; p = 0.035) and reoperation (OR: 0.38; 95% CI: 0.24-0.61; p < 0.001). Prehab+ patients also demonstrated superior health-related quality of life (HRQL) outcomes at two years, including higher rates of achieving the minimal clinically important difference (MCID) in SF-36 PCS at 6 weeks (42.9% vs. 15.1%; p = 0.003) and SRS-22r total score at one year (OR: 3.11; 95% CI: 1.16-8.4; p = 0.024). Cost-effectiveness analysis revealed that Prehab+ patients had a lower cost per QALY at two years ($14,463 vs. $45,515; p < 0.05). Conclusion: Preoperative rehabilitation appears to be significantly associated with a shorter length of stay, better reported clinical outcomes, greater utility gained, costs by procedure and cost effectiveness at two-year follow-up following adult spinal deformity surgery

ID: AGS26/84

A063: Posterior vertebral column resection with preservation of posterior ligamentous complex for severe angular spinal kyphotic deformity: a comparative study

Lijin Zhou 1 , Honghao Yang 1 , Yong Hai 1

1 Beijing Chao-Yang Hospital, Beijing, China

Introduction: Angular spinal kyphotic deformity could result in significant cardiopulmonary dysfunction, poor self-image, and neurological deficits. Posterior vertebral column resection (PVCR) is a highly effective technique for correcting spinal malignment. However, the limitations such as instability from radical removal of posterior elements associated with a high rate of neurological complications. The purpose of this study was to compare the effectiveness and safety of modified PVCR and traditional PVCR in the treatment of patients with angular spinal kyphotic deformity. Material and Methods: A retrospective cohort study was conducted. Patients who underwent PVCR for angular spinal kyphotic deformity with a minimum of 24 months follow-up from March 2018 to November 2022 were included. These patients had a focal kyphosis (FK) with deformity angular ratio greater than 20. Patients were divided into the modified PVCR group and the traditional PVCR group based on the procedures. The modified PVCR, which preserved the posterior ligamentous complex (PLC) throughout the process, was performed for patients who were admitted to the hospital after October 2020. Demographic information, radiological parameters, surgical information, health-related quality of life, and complications were collected and compared between the groups. Results: A total of 62 patients with angular spinal kyphotic deformity were recruited in this study. The mean age at surgery was 42.29 ± 12.42 years. Among the patients, 31 were in the modified PVCR group and 31 were in the traditional PVCR group. The FK was corrected from 125.54° ± 15.83° to 48.19° ± 16.07° in the modified PVCR group and 119.77° ± 17.10° to 57.20° ± 15.43° in the traditional PVCR group. The correction rate of FK in the modified PVCR group was significantly greater than the traditional PVCR group (62.26% ± 10.24% vs. 51.46% ± 15.49%, p = 0.002). There were no significant differences in the operation time, estimated blood loss, and health-related quality of life between the two groups. However, the incidence of neurological complications was 16.1% in the modified PVCR group, significantly lower than the 38.7% observed in the traditional PVCR group (p = 0.046). Conclusion: By preserving the PLC throughout the osteotomy, the modified PVCR outperformed the traditional technique in terms of both neurological complications and deformity correction.

OP8: Tumor Mapping and Staging

ID: AGS26/1065

A064: Management of painful spinal metastases: a systematic review of GRADE-based evidence across pain types

W. Christopher Newman 1 , Konstantina Matsoukas 1 , Marie-Laure Vial 2 , Ori Barzilai 1 , Ilya Laufer 3 , on behalf of AO Spine Knowledge Forum Tumor

1 Memorial Sloan Kettering Cancer Center, New York, United States ,

2 AO Foundation, AO Network Clinical Research, Davos, Switzerland ,

3 NYU Langone Health, Department of Neurosurgery, New York, United States

Background: Spinal metastases affect up to 40% of patients with cancer and are a major cause of morbidity, primarily due to pain. The AO Spine Knowledge Forum Tumor recently formalized six pain types in spinal metastatic disease: mechanical-axial, mechanical-radicular, inflammatory-axial, inflammatory-radicular, neuropathic-central, and neuropathic-peripheral pain. Clear classification of these types allows a more rigorous alignment of interventions with pain mechanisms. We conducted a systematic review to assess the quality of evidence supporting different therapeutic strategies for each pain type using the GRADE framework. Methods: Following the Preferred Reporting Items of Systematic Reviews and Meta-Analysis (PRISMA) methodology, we conducted a systematic review of PubMed (NLM) database on March 17, 2025 for English-language studies published in the last five years (2020-2025) reporting pain outcomes for patients with spinal metastases treated with radiation therapy (RT), surgery, vertebral augmentation, ablation techniques, interventional injections, neuroablative procedures, and pharmacologic interventions. Studies were screened for explicit pain endpoints. Evidence quality was appraised using the Grading of Recommendations, Assessment, Development, and Evaluation (GRADE) system, categorizing interventions as high, moderate, low, or very low quality of evidence. Results: A total of 148 articles were included. High-quality evidence supports RT and opioids for inflammatory-axial pain, with strong and consistent pain reduction across randomized trials and meta-analyses. Mechanical-axial pain demonstrates high-quality evidence for kyphoplasty and vertebroplasty, and moderate-quality evidence for surgical stabilization. Inflammatory-radicular pain has moderate-quality evidence for both RT and surgical decompression, while mechanical-radicular pain is supported by moderate-quality evidence for surgical intervention. For vertebral body ablation (radiofrequency or cryoablation), evidence for inflammatory-axial pain is moderate but limited by a lack of randomized data. Intrathecal drug delivery demonstrates moderate-quality evidence for neuropathic-central and neuropathic-peripheral pain, particularly in refractory cases. Neuroablative approaches such as spinal cord stimulation, cordotomy, and dorsal root entry zone lesioning are supported by low-quality evidence, largely from retrospective series. Evidence for epidural steroid injections and neurolytic blocks remains low, with small heterogeneous studies and short follow-up periods. Conclusion: Inconsistent and vague definitions of pain types limit the overall quality of pain outcomes across various studies. Despite this, high-quality evidence exists for RT, opioids, and vertebral augmentation techniques, while other modalities remain supported by moderate or low-quality evidence. Adoption of standardized pain type definitions in future trials will improve the ability to refine and personalize evidence-based care.

ID: AGS26/1306

A065: A novel rat caudal spine metastasis resection model for therapeutic implant screening

Nitai Klein 1,2 , Ateeque Siddique 1 , Michael H 1,3 , Lisbet Haglund 1,2 , Derek Rosenzweig 1

1 Division of Orthopedics - McGill University Health Center, Shriner's Hospital for Children - Canada, Montreal, Montreal, Canada

2 Shriner's Hospital for Children - Canada, Montreal, Canada

3 University of Conneticut Health Center, Farmignton, United States

Introduction: The spine is the most frequent site of osseous metastases. Surgical resection often leaves critical-sized bone defects and residual tumor cells that drive recurrence. Polymethylmethacrylate (PMMA) cement is commonly used for reconstruction due to its strength and accessibility, but it lacks therapeutic function. We developed a doxorubicin (DOX)-loaded PMMA composite using mesoporous silica nanoparticles to enable sustained local drug release. While effective in vitro against breast and prostate cancer models, its performance in a relevant in vivo spinal setting has not been established. The aim of the study was to develop an in vivo spinal metastasis resection model and evaluate the efficacy of DOX-PMMA in reducing tumor recurrence. Material and Methods: Luciferase-tagged MDA-MB-231 breast cancer cells were encapsulated in alginate-gelatin hydrogel and implanted into caudal vertebrae of immunodeficient rats. Six weeks later, tumors were resected and reconstructed with either control PMMA or DOX-PMMA. Tumor progression was monitored by bioluminescence and CT imaging. At 10 weeks, blood tests, micro-CT, histology, and immunohistochemistry (Caspase-3, Ki-67) were performed. Results: Tumors remained viable and stable until resection. Ten weeks after cement implantation, bioluminescence was significantly reduced in the DOX-PMMA group compared with controls (p < 0.05). Control animals demonstrated elevated lactate dehydrogenase and creatine kinase, consistent with disease progression. Micro-CT and histologic analyses are ongoing. Conclusion: This novel rat caudal vertebra model provides a reproducible platform for testing spinal biomaterials. Local delivery of DOX via PMMA cement reduced tumor recurrence without systemic chemotherapy. This strategy holds promise for improving local disease control and quality of life in patients with spinal metastases.

ID: AGS26/2251

A066: International Delphi Consensus on defining the optimal surgical composite outcome in primary malignant bone tumors of the spine

Mark Maclean 1 , Rafael De la Garza Ramos 2 , Raphaële Charest-Morin 3 , Michael Weber 4 , Markian Pahuta 5 , Shalin Patel 6 , Arjun Sahgal 7 , Laurence Rhines 8 , Cordula Netzer 9 , Nicolas Dea 3 , Jorrit-Jan Verlaan 10 , Alessandro Gasbarrini 11 , Jeremy Reynolds 12 , Ori Barzilai 13 , Chetan Bettegowda 14 , Stefano Boriani 15 , Charles Fisher 3 , Ziya Gokaslan 16 , Áron Lazáry 17 , Ilya Laufer 18 , Daniel Sciubba 19 , John Shin 20 , C. Rory Goodwin 21

1 Dalhousie University, Department of Surgery, Halifax, Canada ,

2 Department of Neurological Surgery, Montefiore Medical Center Albert Einstein College of Medicine, New York, United States ,

3 Vancouver General Hospital, Vancouver, Canada ,

4 University of Connecticut, Farmington, United States ,

5 McMaster University, Hamilton, Canada ,

6 MD Anderson, Houston, United States ,

7 Department of Radiation Oncology, Sunnybrook Health Sciences Centre, University of Toronto, Toronto, Canada ,

8 MD Anderson Cancer Center, Houston, United States ,

9 University Hospital of Basel, Basel, Switzerland ,

10 University Medical Center Utrecht, Utrecht, Netherlands ,

11 IRCCS Istituto Ortopedico Rizzoli, Bologna, Italy ,

12 Oxford University Hospitals NHS Foundation Trust, Oxford, United Kingdom ,

13 Memorial Sloan Kettering Cancer Center, New York, United States ,

14 Johns Hopkins University School of Medicine, Baltimore, United States ,

15 University of Bologna, Bologna, Italy ,

16 Brown University, Providence, United States ,

17 Buda Health Center, Budapest, Hungary ,

18 New York University Langone Health, New York, United States ,

19 Northwell Health, Manhasset, United States ,

20 University of Pennsylvania, Philadelphia, United States ,

21 Duke University, Durham, United States

Introduction: No single indicator captures the essence of a desirable surgical outcome for primary malignant bone tumors of the spine. Composite outcome measures have been used to comprehensively define the success of achieving an optimal surgical outcome in general surgical oncology but not spinal surgery. To achieve consensus in defining a composite surgical outcome measure for use in spinal primary malignant bone tumor research. Material and Methods: A modified Delphi approach was employed by the AO Spine Knowledge Forum Tumor. Cross-sectional surveys were followed by in-person consensus voting, in accordance with Core Outcome Measurement in Effectiveness Trials recommendations. Included outcome indicators were assigned objective measurement parameters. Consensus was defined as ≥ 70% agreement among respondents. Results were analyzed descriptively. Results: The initial response rate was 61% (42 of 69). Respondents represented 12 countries and 32 institutions. Most (93%) respondents felt a composite measure should be used to assess surgical outcomes in patients treated for a primary spinal bone tumor. Nineteen clinically relevant outcomes were identified and rated. Four outcome indicators attained consensus and were assigned measurable parameters. The final composite measure was comprised of: (1) achieving the desired surgical plan and intended resection margins; (2) absence of Spinal Adverse Events Severity system version 2 grade 3 (or higher) surgical complications at 90 days; (3) absence of a major unplanned neurological deficit (permanent loss of function in extremity or bowel/bladder control) at 12 months; and (4) maintaining or improving Eastern Cooperative Oncology Group performance status at 12 months. Conclusion: The AO Spine Knowledge Forum Tumor developed a composite outcome measure (OSCO-P) for research use in primary malignant bone tumors of the spine. This consensus-derived measure aims to capture the most complete oncologic resection with minimal morbidity and optimal functional recovery.

ID: AGS26/2531

A067: Influence of high BMI on prognosis in patients undergoing surgery for spinal metastases

Maxwell Sahhar 1 , Ahmed Sulieman 2 , Malcolm Hamilton-Hall III 2 , Yesha Parekh 2 , Myung-Jin Cha 2 , Andrew Kim 2 , Goutham Yalla 2 , Daniel Lubelski 3 , Sang Hun Lee 2

1 Brown University Warren Alpert Medical School, Providence, United States ,

2 Johns Hopkins University Department of Orthopedic Surgery, Baltimore, United States ,

3 Johns Hopkins University Department of Neurological Surgery, Baltimore, United States

Introduction: Body mass index (BMI) is a widely used clinical metric with complex implications in surgical oncology. In line with population-wide increases to BMI, more patients undergoing surgery for spinal metastasis have a high BMI. While obesity is traditionally considered a risk factor in spine surgery, the impact of high BMI on patients undergoing decompression and instrumented fusion (DIF) for spinal metastases remains undefined. Material and Methods: The American College of Surgeons National Surgical Quality Improvement Program (NSQIP) database was queried from 2016 to 2022 for patients who underwent DIF with ICD-10 codes for metastatic spinal disease. Generalized additive models (GAM) were used to assess the relationship between BMI and 30-day postoperative mortality and major complications, adjusting for demographic, clinical, and procedural covariates. BMI was modeled as a smooth term to account for non-linear associations. Results: A total of 2,256 patients were included in the study. The mean age was 62.8 years, and the mean BMI was 27.4 kg/m 2 . The overall 30-day mortality rate was 6.7%, with patients undergoing thoracic and cervical DF having higher mortality than patients undergoing lumbar DF (p = 0.003). BMI was a significant non-linear predictor of mortality (EDF = 3.253, p = 0.0011) and major complications (EDF = 1.000, p = 0.0464). Mortality risk increased substantially below a BMI of 22.5 kg/m 2 , with a 4.45 times higher risk and 2.26 times higher risk for patients at a BMI >18.5 and 18.5-22.5, respectively (p <0.001 and p = 0.002). No significant increase in risk was seen in patients of higher BMIs, even in patients exceeding a BMI of 40 kg/m 2 . Functional dependence and pulmonary comorbidities were independently associated with mortality. Major complications occurred more frequently in patients with low BMI and functional impairment but these trends were not modified by spinal region in adjusted analyses. Conclusion: High BMI, even when exceeding standardized obesity cutoffs, is not a predictor of mortality and major complications following decompression and instrumented fusion for spine metastasis. Low BMI of less than 22.5 is a significant non-linear predictor of mortality and major complications. These findings suggest that BMI can serve as a practical proxy for physiological reserve in preoperative risk stratification for spine metastasis surgery.

ID: AGS26/3202

A068: Augmented reality-assisted resection of intradural extramedullary spinal tumor: a single-center experience

Giacomo Cammarata 1 , Francesco Certo 1 , Angelo Basile 1 , Sara Pappalardo 1 , Erica Grasso 1 , Giuseppe Sotera 1 , Giuseppe Barbagallo 1

1 Department of Neurosurgery, Universitary Hospital G. Rodolico-San Marco“, Catania, Italy

Introduction: The integration of augmented reality (AR) into neurosurgical practice has recently gained momentum, with the goal of enhancing intraoperative visualization and surgical precision. In spinal surgery, AR-assisted techniques offer the potential to improve the accuracy of tumor localization and resection while minimizing the need for repeated fluoroscopy or extensive bony exposure. Intradural extramedullary (IDEM) spinal tumors, which are often benign but may be infiltrative or malignant, represent an ideal setting to evaluate AR accuracy due to the need for precise localization before dural opening. This study presents a single-center experience with AR-assisted resection of IDEM spinal tumors, focusing on accuracy, histological variability, and the correlation between morphometric parameters and AR performance. Material and Methods: We retrospectively reviewed 30 patients (mean age 56.4 years, range 12-77) who underwent AR-assisted IDEM tumor resection between January 2024 and August 2025. Demographic and clinical data, tumor characteristics, surgical approach, and histopathological findings were collected. The primary endpoint was intraoperative AR shift (mm), defined as the distance between AR-predicted and true tumor localization. Secondary analyses investigated correlations between AR shift and tumor morphometrics (antero-posterior [AP], cranio-caudal [CC] diameters, tumor volume), histological subtype, and spinal region. Statistical evaluation was performed using Pearson’s correlation and analysis of variance. Results: The mean AR shift was 2.3 ± 1.0 mm, consistently within an acceptable intraoperative margin. Correlation analyses showed that larger tumors were associated with greater AR shift, particularly for cranio-caudal diameter (r = 0.64, p < 0.01), while volume (r = 0.47, p = 0.04) and AP diameter (r = 0.29, n.s.) had weaker associations. These findings confirm that elongation along the cranio-caudal axis is the most relevant predictor of reduced AR accuracy. Histology significantly influenced performance. Meningiomas and schwannomas showed the lowest shift (1.2-2.3 mm), reflecting their circumscribed nature, while more infiltrative or malignant lesions demonstrated higher values: plasmocytoma (3.0 mm), lymphomas (3.2 mm), paraganglioma (2.9 mm), and germ cell tumor (3.8 mm). Between-group comparison showed a significant difference in AR shift between circumscribed and infiltrative histologies (p = 0.03). Regional analysis revealed comparable accuracy across levels: thoracic 2.45 mm, cervical 2.35 mm, and lumbar 2.03 mm, without significant differences (p = 0.42). Variability was slightly higher in the thoracic spine, possibly due to respiratory and anatomical factors. Conclusion: AR-assisted spinal tumor resection proved highly accurate, with mean shifts consistently below 3 mm. Accuracy was greatest for circumscribed lesions and decreased modestly with larger or infiltrative tumors. Statistical analyses highlighted a strong correlation with cranio-caudal extension and significant differences across histological categories, supporting AR reliability in well-defined tumors. These results align with emerging literature reporting sub-3 mm accuracy in AR-guided spine procedures, confirming reproducibility in oncological contexts. Beyond precision, AR reduced reliance on intraoperative fluoroscopy and improved surgical orientation, particularly in challenging thoracic cases. Our findings support the feasibility and safety of AR integration in spinal oncology, suggesting it may complement navigation and intraoperative imaging. Larger, prospective studies are warranted to validate long-term outcomes and explore its role in complex, multi-level, or recurrent tumors.

ID: AGS26/3288

A069: Development of the Spinal Metastasis Staging (SMS) System: an international multidisciplinary expert panel and survey study

Harmen Kuijten 1 , Roxanne Gal 1 , Nicolien Kasperts 1 , Joanne van der Velden 1 , Jorrit-Jan Verlaan 1

1 University Medical Center Utrecht, Utrecht, The Netherlands

Introduction: If left untreated, spinal metastases typically progress from an asymptomatic stage to advanced stages with debilitating pain, spinal instability, and neurological deficits. Timely recognition is critical, yet diagnostic and referral delays remain common as many patients first present to non-spine specialists. While several guidelines exist for managing spinal metastases, these are designed mainly for spine surgeons and radiation oncologists, treating patients who are already diagnosed with spinal metastases. This leaves a gap for simple classification and referral tools aimed at non-spine clinicians. To address this problem, we developed a preliminary Spinal Metastasis Staging (SMS) system and a corresponding referral tool. Material and Methods: The SMS system defines four stages reflecting the natural progression of spinal metastases: SMS I, asymptomatic; SMS II, inflammatory (non-mechanical) pain; SMS III, mechanical pain and/or instability; and SMS IV, neurological deficits and/or cord compression. Based on these stages, we developed a referral tool in the form of a decision tree, designed to guide clinicians on referral priorities and urgency. This referral tool can be presented in different formats (e.g., interactive online versions), but for evaluation purposes it was distributed as a concise, color-coded, two-sided pocket map. The preliminary SMS system and pocket map were revised through two regional and one international expert panel rounds. The refined pocket map was then evaluated in an international online survey distributed via professional societies and networks including AO Spine. Quantitative data were analyzed descriptively, and qualitative feedback underwent thematic analysis for incorporation into a next version. Results: Expert panels endorsed the SMS system and suggested refinements, including clearer definitions of pain types, flowchart simplification, and radiologist involvement for consistent SINS and Bilsky reporting. In the subsequent international survey, with 120 respondents, 89% considered the pocket map easy to understand, 82% reported no missing or unclear steps or decision points, and 97% judged its format suitable. Among the 33 non-spine specialists (28% of respondents), 91% indicated the pocket map would support more timely or appropriate referrals, and 61% reported they would use the tool frequently or always, 33% occasionally, and only 6% never, mainly because they already rely on personal experience or collegial discussion. Conclusion: The SMS system and corresponding referral tool, evaluated in the form of a pocket map, were positively received in both expert panels and an international survey, with feedback supporting further refinement. The findings highlight the tool’s strong potential to facilitate earlier recognition and referral of patients with spinal metastases. The feedback is currently being incorporated into the next version for dissemination among non-spine clinicians. Further work will comprise clinical validation, using the SMS system as a classification framework to examine whether the proposed stages represent clinically distinct patient groups and how they relate to outcomes such as quality of life, survival, and healthcare costs.

ID: AGS26/3325

A070: Separation surgery with postoperative SBRT to treat for spinal oligometastatic disease - Evaluation of carbon fibre reinforced PEEK instrumentation in comparison to standard titanium fixation

Priyanshu Saha 1 , Sebastian Siller 1 , Kofi Cox 1 , Zion Hwang 1 , Febisayo Sidiku 1 , David Gower 1 , Francis Johnston 1 , Marios Papadopoulos 1 , Salvatore Russo 2 , Timothy Bishop 1 , Jason Bernard 1 , Darren Lui 1

1 St George's University Hospitals NHS Foundation Trust, Department of Complex Spine Surgery, London, United Kingdom ,

2 Imperial College Healthcare NHS Trust, London, United Kingdom

Introduction: Oligometastatic disease (≤ 5 extracranial metastases) is hypothesised as the intermediary state between localised and systemic metastatic disease. However, in terms of spinal oligometastatic disease, data on oncological outcome after radical surgical treatment are still lacking in literature. The same applies to the theoretical advantages CFR-PEEK instrumentation in this specific patient cohort compared to the use of titanium fixation. Material and Methods: We retrospectively analysed our prospectively collected consecutive cohort of patients undergoing Separation Surgery for spinal oligometastatic disease between 2018- 2023 at out quaternary level 1 spinal centre; whenever possible postoperative SBRT was applied post adjuvant alongside with systemic oncological treatment. Patient demographics, surgical and oncological characteristics as well as outcomes with regards to hardware malfunction, local recurrence rates and mortality were investigated and compared between the group with CFR-PEEK (CF) instrumentation and a control group with standard titanium (Ti) fixation. Follow-up was at least 12 months in every patient which ranged up to 60 months. Results: A total of 25 oligometastatic MSCC patients (mean age 59.4, male: 52%) all with Bilsky Score 1c or greater) underwent Separation Surgery using CFR-PEEK fixation. Varying tumour histologies were included and compared to 20 patients that underwent separation surgery with standard Ti fixation. Mean operation time, intraoperative blood and length of inpatient stay for Separation Surgery was 234.8 minutes, 726.7 ml and 21.0 days without any statistical differences between both CF and Ti groups. Postoperative complication rate was 6% (no significant differences between the CFR-PEEK and Ti group). At the last interval scan (mean: 17 months post-operative), the rate of construct failure or screw breakage was 0% in both CFR-PEEK and Ti groups. 2 patients in the CFR-PEEK group and 3 patients of the Ti group failed to receive postoperative adjuvant SBRT due to inpatient death, autonomy, and non-local recurrent disease. While overall 1-year survival and local tumour control rates 100% in those that received post operative SBRT, it was markedly worse in those patients failing adjuvant SBRT (50% vs 0%); there were no significant differences in oncological survival between the CFR-PEEK and Ti groups (p = 0.4299). Conclusion: Separation Surgery plus adjuvant SBRT are feasible in spinal oligometastatic disease and enable a favourable prognosis. The use of CF instrumentation in this context is equally safe compared to standard titanium fixation and may be beneficial due to its material-inherent advantages allowing superior planning and delivery of radiotherapy and the reduced artefacts on postoperative imaging.

ID: AGS26/396

A071: Utilization of carbon fiber spinal instrumentation in primary spinal tumors: outcome analysis from the AO Spine Primary Tumor Research and Outcomes Network

Gisberto Evangelisti 1 , Marie-Laure Vial 2 , Feng Wei 3 , Alessandro Gasbarrini 1 , Laurence Rhines 4 , Ziya Gokaslan 5 , Jeremy Reynolds 6 , Alessandro Luzzati 7 , Disch Alexander C. 8 , Praveen Mummaneni 9 , Daniel Tobert 10 , Michelle Clarke 11 , Áron Lazáry 12 , Chetan Bettegowda 13 , Riccardo Cecchinato 14 , Stefano Boriani 15 , Ori Barzilai 16 , Cordula Netzer 17 , Arjun Sahgal 18 , Jorrit-Jan Verlaan 19 , Charles Fisher 20 , Ilya Laufer 21 , Nicolas Dea 20

1 Istituto Ortopedico Rizzoli, Bologna, Italy ,

2 AO Spine, Davos, Switzerland ,

3 Department of Orthopaedic, Peking University Third Hospital, Beijing, China, beijing, China ,

4 Department of Neurosurgery, The University of Texas MD Anderson Cancer Center, Houston, Texas, USA, Houston, United States ,

5 Department of Neurosurgery, The Warren Alpert Medical School of Brown University, Providence, Rhode Island, USA, Providence, United States ,

6 Oxford University Hospitals NHS Trust, Oxford, UK, Oxford, United Kingdom ,

7 Department of Orthopaedics, Istituto Ortopedico Galeazzi, Milan, Italy, Milan, Italy ,

8 University Comprehensive Spine Center, University Hospital Carl Gustav Carus at the TU Dresden, Dresden, Germany, Dresden, Germany ,

9 Department of Neurosurgery, University of California, San Francisco, California, USA, San Francisco, United States ,

10 Department of Othopedic Surgery, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts, USA., Boston, United States ,

11 mayo clinic, Rochester, United States ,

12 National Center for Spinal Disorders, Budapest, Hungary, budapest, Hungary ,

13 Johns Hopkins Medical Center, Baltimore, United States ,

14 Istituto Ortopedico Galeazzi, Milano, Italy ,

15 Post graduate program of orthopedics, University of Bologna, Bologna, Italy ,

16 Department of Neurosurgery, Memorial Sloan Kettering Cancer Center,, New York, United States ,

17 Department of Spine Surgery, University Hospital of Basel, Basel, Switzerland ,

18 Departments of Radiation Oncology and Surgery, University of Toronto, Toronto, Canada ,

19 Department of Orthopaedics, University Medical Center Utrecht, Utrecht, Netherlands ,

20 Combined Neurosurgical and Orthopaedic Spine Program, University of British Columbia, Vancouver, Canada ,

21 Department of Neurosurgery, NYU Langone Medical Center, New York, United States

Introduction: The use of carbon fiber-reinforced polyetheretherketone (CFR-PEEK) instrumentation in spinal surgery has emerged as an alternative to traditional titanium-based implants in recent years. This study aims to evaluate the use and performance of carbon fiber instrumentation in patients with primary spinal tumors focusing on adverse event risk, overall survival and oncologic outcomes. Material and Methods: Data were collected from the AO Spine Primary Tumor Research and Outcomes Network (PTRON), a multicenter international registry. The primary endpoint was the incidence of adverse events, while secondary endpoints included the risk of developing at least one adverse event, overall survival, local tumor control, and disease progression. Results: A total of 359 patients enrolled in the PTRON registry met the inclusion criteria and were included in the study. Among them, 84 receiving carbon fiber implants (23%), 243 titanium implants (68%), and 32 a combination of both (9%). Median follow-up times were 1.9 years (IQR: 0.4-3.0), 1.3 years (IQR: 0.5-2.4) and 1.1 years (IQR: 0.5-1.7), in the carbon fiber, titanium and combination groups, respectively. The estimated risk of having at least one postoperative AE, including construct failures with and without loss of correction, wound infection (deep or superficial), non-union and wound dehiscence, was statistically comparable between patients receiving carbon fiber, titanium or a combination of both carbon fiber and titanium implants (p > 0.05). Kaplan-Meier survival analyses and multivariable Cox proportional hazards models showed no significant differences in overall survival, recurrence-free survival, or progression-free survival between the implant groups. Conclusion: Carbon fiber spinal implants demonstrate a comparable safety and oncologic profile to traditional titanium implants while allowing for improved tumor surveillance and easier radiation delivery.

ID: AGS26/415

A072: Correlation between Adamkiewicz artery level and tumor localization in metastatic spine disease: implications for preoperative embolization

Wei-Lin Pan 1 , Yen-Heng Lin 1 , Yu Cheng Huang 1 , Fon-Yih Tsuang 1

1 National Taiwan University Hospital, Taipei, Taiwan

Introduction: The Adamkiewicz artery (AKA), also known as the great anterior radiculomedullary artery, is the largest radiculomedullary artery of the spinal cord. Typically arising between T8 and L2, it provides the dominant contribution to the anterior spinal artery (ASA) and perfuses the lower two-thirds of the spinal cord. For major spinal surgeries such as total en bloc spondylectomy, preoperative transarterial embolization (TAE) of radiculomedullary arteries is recommended to reduce blood loss, operative time, and improve visualization. However, when the AKA connects to the ASA, embolization carries a high risk of spinal cord infarction and is often spared, potentially compromising surgical hemostasis. We reviewed metastatic spine tumor cases undergoing preoperative TAE, focusing on patients in whom the AKA was spared, to examine whether a correlation exists between tumor level and AKA level. Material and Methods: Patients with metastatic thoracic or lumbar spinal tumors who underwent preoperative TAE between 2022 and 2024 were retrospectively reviewed. Cases with skipped or diffuse patterns were excluded. Tumor levels were numerically encoded (C7 = 0, T1 = 1 … L5 = 17); in multilevel involvement, the mean of the highest and lowest levels was assigned. Angiographic records were reviewed to identify the AKA; in patients with two AKAs (n = 4), the artery closest to the tumor was used. Spearman’s rank correlation was employed to assess the relationship between tumor and AKA levels, and a Binomial test evaluated whether tumors occurred within ±1 vertebral level of the AKA more frequently than expected by chance. Results: Forty-eight patients with identified AKAs were included. Spearman’s correlation showed a strong positive association between tumor level and AKA level (ρ = 0.884, p < 0.001). Binomial analysis revealed that 54% of tumors were located within ±1 vertebral level of the AKA, significantly higher than the expected random probability of 18% (p < 0.001). Conclusion: In patients with metastatic spinal tumors undergoing TAE, tumor levels were strongly correlated with the level of the AKA, and tumors were significantly more likely to occur near the artery than expected by chance. These findings suggest that ASA supply through the AKA may influence the localization of spinal metastases. Recognition of this association may improve embolization planning and surgical strategy, balancing hemostatic control with spinal cord safety.

OP9: Complex Cervical Cases

ID: AGS26/1012

A073: The correlation between changes in radiological cervical parameters and facial asymmetry

Joana Hankollari 1 , Nikollaq Leka 2

1 Albanian University Tirana, Tirana, Albania

2 University of Medicine Tirana, Tirana, Albania

Introduction: In my daily practice as a rheumatologist, one of the most frequent complaints I deal with is related to the cervical spine. When posture changes of the cervical spine occur, it is often obvious that facial asymmetry exists. The aim of this study is to identify changes in the cervical spine that correlate with facial (a)symmetry. Identifying a potential correlation would be of great clinical significance, as doctors who treat cervical spine pathologies would be more cautious about problems related to facial asymmetry. Material and Methods: 32 patients aged 18-64 years were included in this prospective study who had complaints about their cervical spine. Cervical radiographs were used to assess parameters such as C2-C7 Lordosis (CL), T1 Slope (T1S), Thoracic Inlet Angle (TIA), Neck Tilt (NT), C2-C7 Sagittal Vertical Axis (C2-C7 SVA), cervical and cranial tilting for each patient. The assessment of facial asymmetry was done through inspection and P-A cephalometry. Results: Changes in cervical radiographic parameters occurred in 53.13 % of patients. The parameters that were most altered were: C2-C7 Lordosis (CL), T1 Slope (T1S), C2-C7 Sagittal Vertical Axis (C2-C7 SVA), cervical and cranial tilting. Facial asymmetry was found in 64.71% of patients who experienced changes in these parameters, whereas only 20% of cases without cervical radiological parameters had facial asymmetry. Conclusion: Facial asymmetry is more prevalent in patients with changes in cervical radiological parameters than in those without cervical radiological parameters changes.

ID: AGS26/2957

A074: Cervical pedicle inlet screws: clinical results of a novel trajectory for navigated screw placement in the sub-axial cervical spine

Christopher Martin 1 , Bishop Anyiwe 1 , Regan Carlson 1 , Kari Odland 1

1 University of Minnesota, Minneapolis, United States

Introduction: Lateral mass screws are widely used in the sub-axial cervical spine, but have poor pullout strength and result in a screw tulip that is medial. This medial position makes passing a rod to a nearby pedicle screw more difficult, which has particular relevance in long constructs crossing the cervico-thoracic junction. Sub-axial pedicle screws have been described, but carry a risk of neuro-vascular injury. Furthermore, many sub-axial pedicles are simply too small to safely pass a screw. The cervical pedicle inlet (CPI) trajectory involves a lateral starting point similar to a true pedicle screw which allows for easier rod passage ( Figure 1 ), but the tip of the screw stops in the dense bone at the pedicle inlet ( Figure 2 ), thus limiting neurovascular risk. Our group recently published a cadaveric biomechanical study showing that CPI screws have a higher pullout strength than lateral mass screws. However, limited clinical data exist. This study aims to report the early clinical experience with this new screw trajectory, with an emphasis on screw related complications and clinical outcomes. Methods: We retrospectively reviewed 227 CPI screws from 128 consecutive cases from a single surgeon at a single center. All patients were ≥ 18 years and underwent posterior cervical fusion with navigated CPI screw placement in the sub-axial cervical spine between 2020 and 2025. All screws were placed with intra-operative O-Arm navigation. CT scans were available on all patients and were used to classify breaches with a modified Gertzbein-Robbins scale. Results: 146 of 227 screws (64%) were grade 0 (completely within bone). 72 screws (32%) had a grade 1 (< 1 mm) cortical breach. 9 screws (4%) had a grade 2 breach (> 1 mm). All of these were recognized on the intra-operative check spin with the O-arm and 4 were repositioned while the other 5 were removed. There were no neurovascular injuries or mechanical failures, and no patient had radicular pains or required a return to the OR as a result of screw related complications. There was no patient where a CPI could not be placed. Conclusion: Use of image-guidance allows for safe placement along the CPI trajectory, with no recorded screw related complications. For patients with cervical pedicles that are too small to accept a true pedicle screw, the CPI trajectory is a reasonable alternative. The lateral starting point of the CPI trajectory facilitates rod passage to other pedicle screws in the construct, and may have particular applications in minimally invasive approaches that utilize a paramedian incision.

ID: AGS26/1472

A075: Extension of fusion to the cervical spine versus upper thoracic spine for the management of proximal junctional kyphosis of thoracolumbar fusion

Ahmed Sulieman 1 , Yesha Parekh 1 , Virginie Lafage 2 , Renaud Lafage 2 , Breton Line 3 , Christopher Ames 4 , Thomas Buell 5 , Shay Bess 3 , Robert Eastlack 6 , Jeffrey Gum 7 , Munish Gupta 8 , Richard Hostin 9 , Han Jo Kim 10 , Gregory Mundis 11 , Peter Passias 12 , Christopher Shaffrey 12 , Justin Smith 13 , Khaled Kebaish 1 , Sang Hun Lee 1 , International Spine Study Group (ISSG) 14

1 Department of Orthopaedic Surgery, The Johns Hopkins University, Baltimore, United States,

2 Lenox Hill Hospital Department of Orthopaedic Surgery, New York, United States,

3 Denver International Spine Center, Denver, United States,

4 University of San Francisco, San Francisco, California, United States,

5 Department of Neurological Surgery, University of Pittsburgh Medical Center, Pittsburgh, PA, United States,

6 Scripps Clinic, La Jolla , California, United States,

7 Leatherman Spine Center, Louisville , Kentucky, United States,

8 Washington University in St Louis, St. Louis, Missouri, United States,

9 Southwest Scoliosis Institute, Plano, TX, United States,

10 Hospital for Special Surgery, New York, United States,

11 San Diego Spine, California, United States, United States,

12 Departments of Neurosurgery and Orthopaedic Surgery, Duke University School of Medicine, Durham, NC, United States,

13 Department of Neurosurgery, University of Virginia Health System, Charlottesville, VA, United States,

14 International Spine Study Group Foundation, Denver, Colorado, United States

Introduction: Proximal junctional kyphosis (PJK) management after thoracolumbar fusion requires extension of fusion to the proximal spinal segments. Unlike extensions to the less mobile thoracic segments, crossing the cervicothoracic junction involves more mobile cervical segments and creates a different biomechanical influence and clinical outcomes. No study has compared the outcomes of extending fusion to the cervical versus Upper Thoracic (UT) spine. Material and Methods: This is a retrospective review of multicenter, prospective cervical deformity database. Patients with thoracic PJK who underwent revision with extension of fusion to either the cervical or UT (T1 or T2) spine were identified. Patients with cervical upper instrumented vertebra (UIV) were subdivided into lower cervical (LC; C4-7) and upper cervical (UC; and occiput-C3) groups. Baseline demographics, surgical variables, radiographic outcomes, 2-year health-related quality of life scores, complications, and revision rates were analyzed. Results: A total of 51 patients were included, with a mean age of 60.4 ± 12.9 years, and 91% were female. Of those, 12 patients were in the UT group (Preoperative UIV n: 8 at T1-4, 2 at T5-8, 2 at T9-12), 20 in the LC group (Preoperative UIV n: 18 at T1-4, 2 at T9-12), and 19 in the UC group (Preoperative UIV n: 14 at T1-4, 1 at T5-8, 4 at T9-12). Demographics and Charlson Comorbidity Index were similar across groups. EBL, operative time were similar while levels fused were more in the UC, followed by LC and the UT group. All groups had similar radiographic measurements, including change in proximal junctional angle. Instrumentation failure was higher in LC group (25%, vs. 0% in UC and 8% in UT p = 0.03). Reoperation rate was 0% in UT, 30% in LC and 10.5% in UC (p = 0.071) and recurrent PJK was 0% in UT, 15% in LC and 10.5% in UC (p = 0.427). Postoperative radiographic and HRQols were similar across the groups at 2 years. Conclusion: Stopping fusion at the upper thoracic spine (T1-2) did not increase complications or reoperation rates compared to extending fusion to the cervical spine. However, LC UIV (C4-7) had higher instrumentation failure, recurrent PJK, and revision rates than UC (C2-3) and UT groups. Crossing cervicothoracic junction may not be beneficial to avoid junctional level complications.

ID: AGS26/1934

A076: A decisional algorithm for the surgical treatment of kyphotic degenerative cervical myelopathy

Marco Ajello 1 , Stefano Colonna 1 , Filippo Lacatena 1 , Ludovico Maria Comite 1 , Francesco Maria Dell'Accio 1 , Luca Ceroni 2 , Nicola Marengo 1 , Diego Garbossa 1 , fabio cofano 1 , Alexander R. Vaccaro 3

1 Neurosurgery Unit, Department of Neuroscience “Rita Levi-Montalcini”, University of Turin, Turin, Italy

2 Department of Psychology, University of Turin, Turin, Italy

3 Rothman Orthopaedic Institute at Thomas Jefferson University, 925 Chestnut St, Philadelphia, PA, 19107, USA, Philadelphia, United States

Introduction: The optimal surgical treatment of kyphotic degenerative cervical myelopathy (DCM) is still object of debate. Cervical kyphosis directly correlates with spinal cord compression, making restoration of cervical lordosis a major objective of the surgical treatment. We present a decisional algorithm thought to guide the surgeon in the choice of the optimal surgical treatment in DCM patients. Material and Methods: A multicenter retrospective analysis was performed on adult patients who underwent surgery for kyphotic DCM. Surgical planning followed an algorithm based on: (1) disease extension (< 3 vs. ≥ 3 levels), (2) deformity flexibility on extension radiographs, and (3) localization of ankylosis. According to these criteria, procedures included anterior-only, posterior-only, combined anterior-posterior, or posterior-anterior-posterior approaches. Clinical outcomes were evaluated using patient-reported outcome measures (PROMs), including the mJOA score, arm and neck pain on the Numeric Rating Scale (NRS), the Neck Disability Index (NDI), and the EQ-5D questionnaire. Radiological outcomes were assessed on postoperative static radiographs obtained at ≥ 6 months, measuring segmental and global cervical lordosis (C2-C7), T1 slope, and C2-C7 SVA. Patient satisfaction was determined using the North American Spine Society satisfaction index (NASS-SI). Results: A total of 117 patients treated with anterior-only (n = 42; 35.8%), posterior-only (n = 36; 30.7%) or combined (n = 39; 33.5%) cervical arthrodesis were included in the analysis. Application of the algorithm achieved effective kyphosis correction across groups, with postoperative segmental angles at least neutral compared to preoperative values (p < 0.05). Clinical scores (mJOA, NRS-arm/neck, NDI, EQ-5D) improved significantly in all groups (p < 0.05). Combined approaches were associated with higher complication rates compared with anterior- and posterior-only groups. Conclusion: The proposed algorithm enabled effective correction of kyphotic deformity, with postoperative complications comparable to those reported in the literature. Favorable clinical and radiological outcomes support its applicability in guiding the optimal surgical approach for kyphotic DCM. Future larger studies are essential to validate these results and the applicability of the algorithm in this population.

ID: AGS26/2922

A077: K-line tilt as a predictor of postoperative kyphotic deformity after laminectomy for cervical myelopathy and its correlation with clinical outcomes

Laksh Agrawal 1 , Mihir Bapat 1 , Bharat Patel 1 , Amandeep Gujral 1 , Ashutosh Tripathi 1 , Rushikesh Jadhav 1 , Akshaj Sharma 1

1 Nanavati Max Super Speciality Hospital, Mumbai, India

Introduction: Laminectomy is widely performed for multilevel degenerative cervical myelopathy (DCM) and K-line-positive (K-line (+)) ossification of the posterior longitudinal ligament (OPLL). However, postoperative kyphotic deformity develops in up to one-fifth of patients and is associated with inferior neurological recovery, often necessitating additional stabilization. The risk is particularly high in patients with preoperative straight alignment. While K-line tilt-a radiographic measure of the angle between the K-line and a vertical reference-has been validated as a predictor of deformity after laminoplasty and anterior procedures, its role in laminectomy remains undefined. This study aimed to determine whether preoperative K-line tilt predicts postoperative kyphosis following laminectomy, establish an optimal cutoff value, and evaluate its relationship with clinical outcomes. Material and Methods: Seventy patients undergoing laminectomy for multilevel DCM or K-line (+) OPLL between 2017 and 2022 were retrospectively reviewed, with a minimum follow-up of two years. Only patients with non-kyphotic preoperative alignment (C2-C7 lordotic angle ≥0°) were included. Radiographic parameters assessed included K-line tilt, C2-C7 lordotic angle, C2-C7 sagittal vertical axis (SVA), and C7 slope. Neurological function was assessed with the Japanese Orthopaedic Association (JOA) score, and recovery rate was calculated. Multivariate analysis was used to identify predictors of postoperative kyphosis, and receiver operating characteristic (ROC) analysis was performed to determine the cutoff K-line tilt angle. Results: Postoperative kyphotic deformity (C2-C7 lordotic angle <0°) was observed in 8 patients (11%). These patients demonstrated significantly lower mean JOA recovery rates compared with those without deformity. Preoperative K-line tilt emerged as an independent predictor of postoperative kyphosis. A cutoff value of 21° was identified; patients with preoperative tilt >21° had a higher incidence of postoperative deformity and poorer neurological recovery. Conclusion: Preoperative K-line tilt is a reliable predictor of postoperative kyphotic deformity following laminectomy for cervical myelopathy in K-line (+), non-kyphotic spines. A tilt angle >21° confers a high risk of postoperative deformity and suboptimal recovery. In such cases, laminectomy alone may be insufficient, and supplemental stabilization with anterior or posterior fusion should be considered to prevent deformity and optimize outcomes.

ID: AGS26/332

A078: Surgical safety strategies for cervical kyphosis in ankylosing spondylitis based on modified C7 PSO

Canhua Ye 1 , Huiren Tao 1

1 The University of Hong Kong - Shenzhen Hospital, Shenzhen, China

Introduction: Ankylosing Spondylitis (AS) is a chronic inflammatory disease primarily affecting the axial skeleton, with a global prevalence of 0.1%-1.4% and higher incidence in males (3.4:1). Cervical kyphosis is a severe late complication, causing restricted vision, dysphagia, and poor quality of life. While cervical osteotomy is effective, its complexity raises risks. This study explores modified C7 PSO osteotomy strategies, aiming to provide safety and efficacy insights for clinical practice. Material and Methods: A retrospective analysis was conducted on the clinical data of 4 patients with ankylosing spondylitis and cervical kyphosis. All patients underwent C7 modified PSO surgery. Surgical details such as positioning, operative time, surgical strategies, intraoperative blood loss, and transfusion volume were recorded. Radiographic parameters, including the cervical sagittal vertical axis (CSVA), Cobb angle of cervical kyphosis, and Chin-brow vertical angle (CBVA), were assessed pre- and post-operatively to evaluate the surgical outcomes. Results: The modified PSO procedure was performed at the C7 level for all 4 patients. The upper instrumented vertebrae (UIV) were primarily C3 or C4, while the lower instrumented vertebrae (LIV) were mainly T3 or T4. The mean surgical time was 470 minutes (range: 394-531 minutes), with a mean intraoperative blood loss of 875 ml (range: 600-1100 ml). No abnormal motor-evoked potentials (MEP) signals were recorded during the surgery. The average postoperative follow-up period was 37 months (range: 4-55 months). Radiographic results showed that the mean cervical kyphosis Cobb angle corrected from 23.4° preoperatively to 2.4° postoperatively, the CBVA improved from 98.4° to 59.6°, and the mean CSVA reduced from 80.2 mm to 59.6 mm. During the follow-up period, one patient experienced duodenal bulb bleeding and bilateral femoral condyle fractures after a fall, while another patient had a dural tear during surgery, resulting in postoperative intracranial air and headache. There were no complications such as loosening, infection, or sensory/motor deficits. Conclusion: With effective perioperative management, modified C7 PSO surgery is a safe and effective treatment for cervical kyphosis in ankylosing spondylitis patients. It significantly corrects the cervical deformity and improves the horizontal gaze function, providing a safe and effective treatment option for these patients, with important clinical implications.

ID: AGS26/895

A079: Chiari I malformation and scoliosis: does posterior-fossa decompression influence the curve progression? A systematic review of the literature and meta-analysis

Davide Palombi 1 , Sergio De Salvatore 2 , Paolo Brigato 2 , Leonardo Oggiano 3 , Gianpiero Tamburrini 1 , Luca Massimi 1 , Timothée De Saint-Denis 4

1 Department of Pediatric Neurosurgery, Università Cattolica del Sacro Cuore di Roma, Rome, Italy

2 Department of Orthopaedics, Campus Bio Medico di Roma, Rome, Italy

3 Department of Orthopaedics, Ospedale Pediatrico Bambino Gesù di Roma, Rome, Italy

4 Service de Orthopedie et Chirurgie Reparatrice de l'Enfant, Hopital Armand Trousseau, Paris, France

Introduction: Chiari I malformation (CIM) is frequently associated with scoliosis, and syringomyelia (SyM). Posterior fossa decompression (PFD) is often performed before spinal fusion, but its impact on scoliosis progression remains debated. This systematic review and meta-analysis aimed to assess scoliosis outcomes after PFD in pediatric patients and to identify predictive factors of curve behavior. Material and Methods: A systematic literature search was performed in PubMed, Cochrane, Embase, Medline, Scopus, and Web of Science up to January 2025, following PRISMA guidelines. Eleven retrospective studies comprising 380 pediatric patients with CIM and scoliosis treated with PFD were included. Primary outcome was scoliosis stabilization or improvement after PFD. Pooled proportions were calculated using random-effects models, and meta-regression was used to identify predictors of outcome. Results: The mean age at surgery was 10.6 ± 1.8 years, and the mean preoperative major curve angle was 34.0° ± 7.4°. Syringomyelia was present in the majority of cases, with postoperative improvement documented in 81.6% of patients. The pooled success rate (improved or stable scoliosis) after PFD was 57% (95% CI: 49-65%), while curve progression occurred in 45% (95% CI: 37-54%). Subgroup analysis of patients undergoing PFD with duroplasty showed a similar success rate (54%, 95% CI: 45-63%). Meta-regression demonstrated that older age at surgery (OR per +1 year = 1.24, 95% CI: 1.10-1.39, p < 0.001) and larger preoperative Cobb angle (OR per +5° = 0.79, 95% CI: 0.65-0.96, p = 0.012) were associated with higher risk of curve progression. In the duroplasty subgroup, only age remained significant. Conclusion: PFD leads to stabilization or improvement of scoliosis in more than half of pediatric patients with CIM, but a substantial proportion still experience progression, often requiring spinal fusion. Younger age and smaller preoperative Cobb angles are associated with better outcomes. Early identification of risk factors and a multidisciplinary approach are essential to optimize surgical planning.

ID: AGS26/921

A080: The hemodynamic and complication impact of anterior column release in adult spinal deformity

Diego Soto Rubio 1 , Schahin Salmanian 1 , Ahmad Alkadri 2 , Eric Cohen 2 , Daven Mejica 1 , John Lingam 3 , Shrinit Babel 1 , Dana Saleh 1 , Mohsen Rostami 1 , Puya Alikhani 1

1 Department of Neurosurgery, Brain and Spine, University of South Florida, Morsani College of Medicine, Tampa, United States ,

2 USF Health Morsani College of Medicine, Tampa, United States ,

3 University of South Florida, Tampa, United States

Introduction: Complex adult spinal deformity (ASD) correction often requires extensive resection of posterior or middle column elements, which can result in substantial intraoperative blood loss. Traditional open surgical techniques are frequently performed in a single stage, increasing the risk of significant blood loss during one operative event. Anterior column release (ACR), performed via a retroperitoneal approach, has been proposed to mitigate these risks; however, ACR often requires staged procedures, which could potentially offset this advantage. This study aimed to evaluate whether ACR reduces estimated blood loss (EBL), transfusion requirements, and perioperative medical complications compared with open osteotomy-based approaches. Material and Methods: We retrospectively reviewed 146 patients who underwent ASD correction between 2016 and 2022 with ≥ 1 year of follow-up. Demographic and surgical data were collected, including the number of instrumented levels, type of osteotomy, staging, total EBL, and transfusion requirements. Total EBL was calculated as the cumulative blood loss from all operative stages. The number of packed red blood cell (PRBC) units transfused intraoperatively and postoperatively was also recorded. Results: Thirty-two patients underwent ACR and were compared with 114 patients treated using open approaches, including pedicle subtraction osteotomy (PSO), intradiscal osteotomies, and Smith-Peterson osteotomies (SPO). Baseline characteristics, including age, BMI, and preoperative Oswestry Disability Index (ODI), were similar between groups. ACR patients had significantly shorter instrumented constructs compared with non-ACR cases (6.97 ± 2.41 vs. 9.04 ± 2.86 levels, p < 0.001). Percutaneous screw placement was more common in ACR cases (31.3%). ACR demonstrated substantially lower EBL compared to open approaches (634 ± 446 mL vs. 954 ± 778 mL, p = 0.001) and reduced transfusion requirements (1.34 ± 1.78 vs. 2.23 ± 2.35 PRBC units, p = 0.030). The rates of perioperative medical complications-including pulmonary embolism, myocardial infarction, pneumonia, urinary tract infection, pneumothorax, and sepsis-were low and did not differ significantly between groups (all p > 0.17). Conclusion: ACR demonstrated a significant hemodynamic advantage, with lower blood loss and decreased transfusion requirements compared to open osteotomy-based techniques. However, this reduction in EBL and transfusions did not translate into differences in overall medical complication rates. These findings suggest that the choice between ACR and open approaches should be primarily guided by the desired surgical goals and alignment correction, rather than expectations of improved perioperative recovery.

ID: AGS26/936

A081: Innovative coronal plane parameters for cervical spine deformity: significance of coronal plane horizontal gaze

Ahmed Sulieman 1 , Yesha Parekh 1 , Virginie Lafage 2 , Renaud Lafage 2 , Breton Line 3 , Christopher Ames 4 , Thomas Buell 5 , Shay Bess 6 , Jeffrey Gum 7 , Robert Eastlack 8 , Munish Gupta 9 , Richard Hostin 10 , Han Jo Kim 11 , Peter Passias 12 , Gregory Mundis 13 , Justin Smith 14 , Christopher Shaffrey 12 , Khaled Kebaish 1 , Sang Hun Lee 1 , International Spine Study Group (ISSG) 15

1 Department of Orthopaedic Surgery, The Johns Hopkins University, Baltimore, United States,

2 Lenox Hill Hospital Department of Orthopaedic Surgery, New York, United States,

3 Denver International Spine Center, Denver, United States,

4 University of San Francisco, San Francisco , California, United States,

5 Department of Neurological Surgery, University of Pittsburgh Medical Center, Pittsburgh, PA, United States,

6 Denver International Spine Center, Denver , Colorado, United States,

7 Leatherman Spine Center, Louisville , Kentucky, United States,

8 Scripps Clinic, La Jolla, California, United States,

9 Washington University in St Louis, St. Louis , Missouri, United States,

10 Southwest Scoliosis Institute, Plano, TX, United States,

11 Hospital for Special Surgery, New York, United States,

12 Departments of Neurosurgery and Orthopaedic Surgery, Duke University School of Medicine, Durham, NC, United States,

13 San Diego Spine, California, United States, United States,

14 Department of Neurosurgery, University of Virginia Health System, Charlottesville, VA, United States,

15 International Spine Study Group Foundation, Denver, Colorado, United States

Introduction: Cervical spine deformity often involves malalignment in both the coronal and sagittal planes. However, in contrast to the numerous sagittal measures of cervical deformity, fewer coronal plane metrics exist, and they may not fully capture coronal malalignment in severe cervical kyphosis. We introduce innovative coronal parameters to enhance the evaluation of cervical coronal alignment. Material and Methods: Retrospective analysis of data from 15 US sites collected from 2013 to 2024. Conventional metrics were C2-C7 sagittal and coronal Cobb angle (CCA), C2-C7 sagittal vertical axis (SVA), T1 slope, chin-brow vertical angle, slope of line of sight (SLS), and C7-S1 coronal vertical axis (CVA). We introduce 6 novel coronal parameters: bilateral orbits-coronal angle (BOCA), bilateral clavicle-coronal angle, bilateral iliac crest-coronal angle, bilateral hip-coronal angle, nasion-T1 CVA, and nasion-S1 CVA. HRQoL was assessed using the Neck Disability Index, modified Japanese Orthopaedic Association score, Patient-Reported Outcomes Measurement Information System Pain Interference and Physical Function subscales, and Veterans RAND 12-Item Health Survey Physical and Mental Component Summary scores. We examined correlations between novel and conventional parameters, as well as HRQoL scores. Subgroup analysis was performed in patients with cervical kyphosis (C2-C7 Cobb angle > 0°). Multivariate linear regression was used to assess associations between changes in novel coronal parameters and postoperative sagittal alignment, adjusting for age, baseline modified Japanese Orthopaedic Association score, and number of levels fused. Alpha = .05. Results: Among the novel coronal parameters, BOCA correlated with CCA (p = .01); BOCA, bilateral clavicle-coronal angle, and bilateral hip-coronal angle correlated with C7-S1 CVA (p = .04, .04, and < .01, respectively). Nasion-T1 and nasion-S1 CVAs correlated with C7-S1 CVA (p = .04 and <.01, respectively). BOCA and nasion-S1 correlated with C2-C7 SVA (p < .01); nasion-S1 correlated with T1 slope (p = .02). Nasion-T1 correlated with C2-C7 sagittal Cobb angle and T1 slope (both, p = .02). In the cervical kyphosis subgroup, BOCA correlated more strongly with C2-C7 SVA (p < .01) and chin-brow vertical angle (r = 0.52, p = .02) than with CCA (p = .09 and r = 0.43, p = .05, respectively). No correlations were observed between novel parameters and baseline HRQoL measures. On multivariate regression, the change in BOCA independently predicted 6-week postoperative C2-C7 SVA (p = .02), and the change in nasion-S1 was associated with postoperative T1 slope (p = .03). No other coronal parameters were associated with postoperative sagittal alignment. Conclusion: BOCA, nasion-T1 and nasion-S1 CVA are valuable adjuncts to conventional parameters for evaluating coronal cervical spine deformity. BOCA demonstrated stronger correlations with sagittal alignment metrics and may be a more reliable parameter than CCA in patients with combined sagittal and coronal deformity.

OP10: Contruct Failures and Others

ID: AGS26/1436

A082: Perioperative risk stratification in anterior cervical discectomy and fusion: the importance of metabolic syndrome vs obesity

Lane McCoy 1 , Travis Kotzur 1 , Brian Neuman 1,2

1 Washington University School of Medicine in St. Louis, St. Louis, United States

2 Barnes Jewish Hospital, St. Louis, United States

Introduction: While obesity is frequently used as a surrogate for perioperative risk in anterior cervical discectomy and fusion (ACDF) surgery, body mass index (BMI) alone does not reflect the underlying metabolic derangements that may contribute to adverse outcomes. Metabolic syndrome (MetS) may offer improved stratification of surgical risk. This study investigates the association between MetS and perioperative outcomes in ACDF surgery and evaluates its predictive utility relative to obesity alone. Material and Methods: A retrospective analysis was performed on patients from TriNetX databases. MetS was defined by the presence of at least three of the following: type 2 diabetes mellitus, hypertension, hypertriglyceridemia, high central adiposity, or low high-density lipoprotein. Obesity was defined as a BMI >30 kg/m 2 . Patients were stratified into four cohorts: Metabolically healthy (MetH), MetS, Obese MetH, and Obese with another MetS component (without full MetS diagnosis). Primary outcomes included postoperative infection, mechanical complication, postoperative spinal implant infection (PSII), pseudoarthrosis, revision, deep venous thrombosis (DVT) and pulmonary embolism, sepsis, wound disruption, and readmission. Propensity score matching was used to control for baseline demographics and Charlson Comorbidity Index. Comparative analyses were conducted between MetS and each of the other cohorts, as well as between obesity alone and MetH. Statistical significance was defined as p < 0.05. Results: A total of 92,724 patients, aged 40-90 years, who underwent primary ACDF were matched and included among all cohorts. MetS patients had significantly increased risk in all outcomes compared to MetH patients, where greatest increases were observed in wound disruption (RR 2.45; CI 1.96-3.06; p < 0.001), postoperative infection (RR 2.41; CI 2.02-2.87; p < 0.001), and PSII (RR 1.96; CI 1.63-2.37; p < 0.001). MetS patients also had significantly increased risk in all outcomes compared to Obese MetH patients, with exception to pseudoarthrosis. The most significant increases were observed in readmission (RR 2.06; CI 1.91-2.21; p < 0.001), wound disruption (RR 1.92; CI 1.43-2.58, p < 0.001), and PSII (RR 1.89; CI 1.43-2.49; p < 0.001). Obese MetH patients were shown to have no significant increased risk for mechanical complication, PSII, or revision when compared to MetH patients. Likewise, MetS patients who were not obese were shown to have similar increased risks compared to MetH in many outcomes, specifically wound disruption (RR 2.07; CI 1.19-3.59; p = 0.008), postoperative infection (RR 2.11; CI 1.29-3.44; p = 0.002), mechanical complication (RR 1.61; CI 1.09-2.38; p = 0.015), and PSII (RR 3.26; CI 1.97-5.41; p < 0.001). Conclusion: MetS emerged as a more robust and comprehensive predictor of perioperative complications than obesity alone. Across nearly all measured outcomes, MetS was associated with substantially higher risks compared to both metabolically healthy patients and obese patients without metabolic derangements. Notably, these elevated risks persisted even among non-obese individuals with MetS, underscoring that adverse metabolic health drives much of the observed perioperative morbidity. In contrast, obesity without accompanying metabolic dysfunction did not significantly increase the risk for certain complications. These findings highlight the need to adopt a more nuanced approach that incorporates metabolic health status into preoperative counseling and optimization protocols to improve surgical outcomes.

ID: AGS26/2389

A083: Systematic review of surgical and complications outcomes after multilevel ALIFs

Igor Barreira 1 , Rafael Moriguchi 1 , Rodrigo Amaral 1 , Luiz Pimenta 1 , Fernando Marcelino 1 , Amabile Galeti 1 , Leandro Resende 1 , Igor Arthur Costa 1 , Gabriel Pokorny 1 , Jullyene Pokorny 1 , Bianca Soares 1

1 Instituto de Patologia da Coluna, São Paulo, Brazil

Introduction: Multilevel (≥ 2-level) ALIF, might carry higher complication burden than single-level ALIF, owing to greater anatomical complexity, and challenges that are usually avoided in single-level (L5-S1) approaches. As indications for multilevel ALIF expand to spinopelvic realignment, robust safety data are needed; existing studies are small and use inconsistent complication definitions. This systematic review and meta-analysis therefore seeks to quantify surgical and postoperative risks, guiding multidisciplinary planning and optimizing patient outcomes. Material and Methods: A rapid systematic review was performed (PRISMA-2020) in MEDLINE, Embase, Scopus, and CENTRAL from inception to 08 August 2025. We included English-language cohort or comparative studies that reported ≥ 2-level anterior lumbar interbody fusion (ALIF) and provided extractable data on peri-operative complications and/or at least one of the following continuous outcomes: operative time, estimated blood loss (EBL), and length of stay (LOS). Two reviewers independently screened titles/abstracts, assessed full texts, and extracted events, means, standard deviations, and sample sizes; discrepancies were resolved by consensus. (PROSPERO: CRD420251119413). Complication proportions were pooled with a logistic-normal random-effects generalized linear mixed model (GLMM) using the Paule-Mandel REML estimator and a 0.5-cell continuity correction. Continuous outcomes were meta-analysed with DerSimonian-Laird random effects on raw means. Statistical heterogeneity was quantified with τ 2 , and 95 % prediction intervals (PI) were calculated with a t-distribution (df = k - 2) to estimate the range for a future comparable study. Leave-one-out analyses explored single-study influence. Results: Twelve multilevel-ALIF cohorts were quantitatively analysed. Complications were reported in six studies (n = 30 948). A logistic-normal random-effects model yielded a pooled complication rate of 17 %, with a 95 % PI from 0.4 % to 92 %, indicating extreme between-study heterogeneity (τ 2 = 1.77). Across nine multilevel-ALIF cohorts (total = 24 797 patients, 108 vascular events), the pooled risk of vascular injury was 2 %, with minimal residual heterogeneity, and with a narrow 95 % prediction interval (1.9 %-3.0 %). Five studies contributed operative metrics. Mean surgical time pooled to 144 min (95 % PI 13 - 275 min; τ 2 = 1 400), and mean blood loss to 208 mL (95 % PI ≈ 0 - 504 mL; τ 2 = 7 126). Length-of-stay (LOS), available in the same five cohorts, averaged 5.3 days; its 95 % PI ranged from 1.4 days (overnight discharge) to 9.2 days, with heterogeneity largely driven by a single Korean series (τ 2 = 1.20). Leave-one-out analyses confirmed robustness of pooled complication and LOS estimates except when the highest-risk or longest-stay study was omitted, which reduced the pooled complication rate to 7 % and LOS to 4.4 days, respectively. Conclusion: Overall, contemporary multilevel ALIF demonstrates wide performance ranges across institutions: most centres report operative times under 2 h, blood loss under 250 mL, and hospitalisation under 1 week, but marked outliers can inflate pooled estimates and prediction bands. Evidence on ≥ 2-level ALIF remains sparse and highly inconsistent. Across 12 cohorts, the average complication rate (17 %) and wide 95% prediction interval (0.4 - 92%), compared to a very small vascular complication rate (2%), prediction interval (1.9% to 3%), suggests that different concepts of complications evaluation might generate a significant heterogeneity across studies possibly inflating the rate of overall complications, when compared to a more standardized criteria such as vascular injury.

ID: AGS26/2619

A084: A study on the pattern of vertebral marrow edema during normal healing process after lateral lumbar interbody fusion and its relationship with early postoperative lumbar symptoms

Xiaopeng Li 1,2,3 , Feng Li 3 , Qiang Yang 1,2,3

1 Capital Medical University Affiliated Beijing Friendship Hospital,Beijing, China,

2 Tianjin Hospital, Tianjin, China,

3 Weifang People’s Hospital, Weifang, China

Introduction: Lateral lumbar interbody fusion (LLIF) demonstrates significant clinical efficacy for lumbar degenerative diseases. Postoperative vertebral bone marrow edema, manifesting as distinct patterns on lumbar MRI, was observed in patients with low back pain and, notably, also in asymptomatic individuals. To investigate vertebral bone marrow edema patterns during normal LLIF healing and their association with early postoperative lumbar symptoms to guide clinical rehabilitation and complication prevention. Material and Methods: A single-center retrospective study with a minimum follow-up of 6 months. We conducted a single-centre, retrospective analysis of consecutive patients who underwent trans-psoas LLIF for degenerative lumbar disorders between September 2019 and January 2023. Baseline demographics, peri-operative details, and imaging data were extracted from the institutional database. Post-operative Visual Analogue Scale (VAS) back pain and Oswestry Disability Index (ODI) scores were obtained by structured telephone interviews at 3 days, 3 months, and 6 months. Associations between clinical outcomes, bone mineral density (DXA) and body-mass index (BMI) were examined with SPSS software. Results: Post-operative vertebral bone-marrow oedema was classified on day-3 MRI as absent edema type, cage-surrounding type, or diffuse type. The incidence of BME was higher after three-level than single- or double-level fusions (70.6 % vs 44.9 % and 35.3 %, respectively; P < 0.05), with cage-surrounding oedema predominating in the latter groups (≈29 % each). VAS back pain scores did not differ between BME patterns at any time-point (P = 0.136, r < 0.2). On post-operative day 3, patients without oedema had lower ODI scores than those with cage-surrounding or diffuse oedema (P = 0.001), yielding a moderate positive correlation (r = 0.466, P < 0.01); this association disappeared at 3 and 6 months (P > 0.05, r < 0.2). Higher bone mineral density and BMI were observed in the no-oedema group (both P < 0.05). Weak correlations linked BME with increasing age (r = 0.393, P < 0.01) and with lower BMD (r = –0.301) and BMI (r = –0.287). Conclusions: This study is the first to delineate the vertebral bone marrow edema patterns that accompany normal healing after LLIF and to examine their relationship with early post-operative symptoms. All observed edema was contiguous with the cage. No significant association emerged between BME pattern and pain perception, whereas early lumbar function was modestly influenced by the presence of edema; this influence waned over time. Advanced age, lower bone mineral density and lower BMI were weakly but significantly associated with BME development. These findings facilitate discrimination of physiologic edema from pathologic events, refine outcome assessment and inform tailored post-operative rehabilitation after LLIF.

ID: AGS26/2693

A085: Preoperative prediction of ICU need following high-risk cervical spine fusion

Saleh Mostafa 1 , Omar Saleh 2 , Abduljabbar Alhammoud 1

1 University of Arizona, Tucson, United States

2 Eastern Virginia Medical School, Norfolk, United States

Introduction: Non-elective cervical spine fusion is associated with substantial risk of post-operative patient complications, including respiratory failure, hemodynamic instability, and the need for intensive care unit (ICU) admission or prolonged hospitalization. Novel patient risk assessment tools may improve perioperative planning and decrease the likelihood of acute complications following non-elective cervical spine fusion. We evaluated machine learning modeling techniques to risk stratify patients for prolonged hospital stay or ICU admission based exclusively on pre-operative factors. Materials and Methods: Adults (≥ 18) undergoing cervical fusion were identified in a large, real-world cohort (MIMIC-IV, Beth Israel Deaconess Medical Center) using ICD-9/10-PCS and CPT procedure codes, yielding 2,229 cervical spinal fusion procedures. The primary outcome was a composite of major postoperative complications requiring ICU-level care, defined as either an unplanned ICU transfer within 48 hours or the need for specific life-sustaining therapies including invasive mechanical ventilation, continuous vasoactive infusions, invasive hemodynamic monitoring, continuous renal replacement therapy, or massive transfusion [≥ 4 units PRBCs/24h]. A gradient boosted tree (GBT) was trained to predict ICU admission using only patient preoperative data. All predictors were restricted to factors available at admission or during surgical planning, including planned surgical approach, patient comorbidities, admission type, labs, vitals, and demographics. Model performance was evaluated using the Area Under the Receiver Operating Characteristic (AUROC) curve, accuracy, recall, and precision. Results: The GBT model generalized and discriminated exceptionally well regarding the need for ICU admission, achieving an AUROC of 0.879 and an overall accuracy of 0.83. The most significant factors associated with ICU admission included the indication for surgery such as tumor, infection, trauma, or degenerative causes, patient BMI, Charlson Comorbidity Index, approach (eg. anterior versus posterior), and patient pre-operative hemoglobin. Conclusion: Machine learning models, using only readily available preoperative data, demonstrate significant potential in accurately risk-stratifying patients for ICU-level care prior to non-elective cervical spine fusion. This predictive tool offers a critical advancement in perioperative planning. Proactive identification of high-risk patients can optimize resource allocation, enhance patient safety, and ultimately improve clinical outcomes.

ID: AGS26/3249

A086: Surgical treatment of Stage III-IV sacral and ischial pressure ulcers: patient-reported quality of life and functional outcomes from a 9-year institutional cohort

Amir Human Hoveidaei 1 , Daniel Over 2 , Sanyam Kulshrestha 1 , Janet Conway 1

1 International Center for Limb Lengthening, Rubin Institute for Advanced Orthopedics, Sinai Hospital of Baltimore, Baltimore, United States

2 Albany Medical College, Albany, United States

Introduction: Sacral and ischial pressure wounds are a devastating complication of prolonged immobility. Surgical intervention has been shown to play a critical role in the management of advanced stage III or IV ulcers, however there is limited research evaluating the efficacy of different surgical techniques and perioperative protocols. Although some articles have evaluated patient specific outcomes using general survey tools, no research to date has evaluated long term quality of life standards using a validated, pressure-ulcer specific patient reported outcome measure tool. This study aims to address this gap in knowledge by assessing outcomes following surgical intervention and strict perioperative protocol adherence in patients with high grade pressure ulcers. Domains of interest include Pressure Ulcer Quality of Life (PU-QoL) survey scores and functional outcome scores using the Barthel Index tool. Material and Methods: A cross-sectional study was conducted among patients who underwent surgical management of stage III or IV sacral and/or ischial pressure ulcers at our institution between 2015 and 2024. Surgical management followed a standardized two-stage protocol consisting of radical debridement with intraoperative cultures, followed by negative-pressure wound therapy and definitive flap or non-flap reconstruction. Postoperative care included 3-4 weeks of strict offloading, progressive sitting with Roho cushions, and pressure-relief education. Diverting colostomy was performed based on proximity of wound. Postoperative quality of life was assessed using the PU-QOL questionnaire. Functional independence was measured using the Barthel Index, ranging from 0 to 100, with higher scores reflecting greater independence. Descriptive statistics were used to summarize demographics and survey scores. Results: Of the 25 patients who underwent surgical management for sacral and/or ischial pressure ulcers, 4 were deceased at the time of analysis and excluded, resulting in a final cohort of 21 patients. The majority were male (90%), with a mean age of 55.24 ± 13.83 years. The mean preoperative ulcer size was 9.4 cm. One-third (33.3%) had primary ulcers, while 66.7% presented with recurrent wounds. Flap-based reconstruction was performed in 40% of patients. 57.1% of ulcers were ischial, 28.6% sacral, and 14.3% involved both sites. Postoperative quality of life, evaluated after an average follow-up of 47.5 months using the PU-QOL tool, revealed a mean total score of 11.39%. Domain-specific analysis showed the highest burden in Pain and Discomfort (16.67%), followed by Everyday Activities (13%), and Emotional Distress (12.06%). Functional independence, measured by the Barthel Index, showed a mean score of 71.19 ± 8.28, reflecting moderate levels of independence. Conclusion: Surgical management of advanced pressure ulcers is a necessary yet complicated procedure that confers a great impact on functional outcomes and quality of life post-procedure. This novel study elucidated promising long-term outcomes in a cohort of patients undergoing this procedure along with a strict perioperative treatment protocol. Additionally, this study provides preliminary insight into the social and functional outcomes challenging this patient population and should be used to more specifically address the needs of this community.

ID: AGS26/3356

A087: Expandable cages as independent risk factor for cage subsidence in transforaminal lumbar interbody fusion: a retrospective analysis of 530 patients

Thibault Grundschober 1,2,3 , Gregor Fischer 1,2 , Linda Bättig 1,2 , Yesim Yildiz 1,2 , Laurin Feuerstein 1,4 , Lorenzo Bertulli 1,2 , Francis Kissling 1,2 , Thomas Schöfl 1,4 , Nader Hejrati 1,2 , Daniele Gianoli 1,4 , Stefan Motov 1,2 , Benjamin Martens 1,4 , Martin Stienen 1,2 , Felix Stengel 1,2

1 Interdisciplinary Spine Center, Cantonal Hospital of St. Gallen & Medical School of St. Gallen, H-OCH Health Ostschweiz, St. Gallen, Switzerland, St. Gallen, Switzerland ,

2 Department of Neurosurgery, Cantonal Hospital of St. Gallen & Medical School of St. Gallen, H-OCH Health Ostschweiz, St. Gallen, Switzerland, St. Gallen, Switzerland ,

3 University of Basel, Medical School, Basel, Switzerland ,

4 Department of Orthopedic Surgery, Cantonal Hospital of St. Gallen & Medical School of St. Gallen, H-OCH Health Ostschweiz, St. Gallen, Switzerland, St. Gallen, Switzerland

Introduction: Cage subsidence (CS) remains a significant complication following transforaminal lumbar interbody fusion (TLIF), with potential impact on fusion success and clinical outcomes. While various risk factors have been identified, the type of interbody cage in development of CS requires further clarification. This study aimed to evaluate independent risk factors for CS at 3 months after TLIF, with particular focus on expandable versus static cage design. Material and Methods: We conducted a single-institution retrospective analysis of 530 patients who received 648 cages during single- or multi-level TLIF procedures between November 2018 and May 2025. Patients included in the cohort received 481 expandable (74.2%) and 167 static interbody TLIF cages (25.8%). CS was assessed radiographically at 3-month follow-up. Multivariable logistic regression analysis was performed to identify independent risk factors, including patient demographics, radiological parameters, cage characteristics, and surgical variables. Results: CS at 3 months occurred in 25.2% of cages (163/648). In multivariable analysis, expandable cage type emerged as the strongest independent risk factor for CS (OR 4.62, 95% CI 2.05-10.42, p < 0.001). Additional independent risk factors included cage position non-anterior (middle or posterior disc space: OR 2.21, 95% CI 1.31-3.71, p = 0.003), 15° vs. 8° lordotic cage shape (OR 1.89, 95% CI 1.12-3.18, p = 0.017), and small cage size (OR 1.76, 95% CI 1.05-2.95, p = 0.033). Univariable analysis confirmed the cage type effect, showing significantly higher CS rates for expandable versus static cages (31.6% vs 6.6%, χ 2 = 41.20, p < 0.001). Patient factors including age, BMI, frailty score, bone density (mean HU), ASA grade, and smoking status were no independent risk factors for CS. Conclusion: In this analysis, expandable cage design represented the most significant independent risk factor for CS following TLIF, with a nearly 5-fold increased odds compared to static cages. Surgical technique factors, particularly cage positioning and sizing, also contribute to CS risk. These findings suggest that static cages may offer superior mechanical stability in the early postoperative period, challenging the presumed advantages of expandable cage technology. Careful consideration of cage selection and optimal surgical technique is essential for minimizing subsidence risk in TLIF procedures.

ID: AGS26/3422

A088: Does preoperative nutritional status influence complications following scoliosis surgery in patients with GMFCS IV-V cerebral palsy?

Nevzat Gönder 1 , Musa Bilgin 2 , Çağrı Karabulut 3 , Beytullah Unat 4 , Emre Kırkgöz 1 , Ramazan Erol 1 , İlkan Kişi 1

1 Gaziantep University School of Medicine, Gaziantep, Turkey

2 Aksaray State Hospital, Aksaray, Turkey

3 Pazarcık State Hospital, Kahramanmaraş, Kahramanmaraş, Turkey

4 Gaziantep Islam Science and Technology University, School of Medicine, Gaziantep, Turkey

Introduction and Objective: Scoliosis is a considerable health issue in individuals with cerebral palsy (CP). Scoliosis associated with CP arises from imbalanced muscle tone, and the likelihood of its development escalates with age, correlating with the severity of CP as indicated by the Gross Motor Function Classification System (GMFCS). By the age of 18, 50% of patients with GMFCS levels IV or V have clinically moderate to severe scoliosis. Owing to the associated muscular and nutritional deficiencies, these patients are predisposed to post-surgical difficulties. While preoperative nutritional status is crucial for postoperative complications in most procedures, it necessitates more meticulous evaluation in this patient cohort due to preexisting nutritional issues. The objective of this study was to investigate the correlation between preoperative nutritional status and complications following surgery. Materials and Methods: A retrospective study was conducted on the medical records of patients diagnosed with GMFCS IV-V CP who were followed up and treated for scoliosis between March 2019 and August 2023 with minimum 18 month follow-up. Children with immunodeficiency and syndromic diseases were excluded from the study. Patients were evaluated in two groups: Group A, consisting of 27 patients who developed postoperative complications, and Group B, consisting of 22 patients who did not develop complications. Demographic features, Cobb angle, amount of bleeding, white blood cell (WBC), preoperative and postoperative hemoglobin (hgb) and platelet (Plt) values were compared between the groups. The patients' preoperative nutritional status was assessed using the Prognostic Nutritional Index (PNI). Complications were graded according to the Modified Clavien-Dindo-Sink Complication Classification System. Results: Twenty-seven patients in group A and 22 patients in group B were included in the study. The mean age of patients in group A was 14.22 ± 2.3, while in group B was 13.5 ± 2.5 (p = 0.304). No significant difference was found between the groups in terms of WBC (p = 0.596), preoperative hgb (p = 0.17), postoperative hgb (p = 0.447), plt (p = 0.771), amount of bleeding (p = 0.793), Cobb angle (p = 0.104) and sex (p = 0.964), respectively. Per the Clavien-Dindo-Sink Complication Classification System, 3 patients experienced grade 1 complications, 5 patients had grade 2, 10 patients presented with grade 3, 6 patients encountered grade 4a, 2 patients faced grade 4b, and 1 patient suffered grade 5 issues. PNI in Group A (31.19 ± 4.24) was significantly lower than in Group B (43.1± 2.44) (p < 0.001). Furthermore, patients in Group A (13.37 ± 12.92) had to stay in the hospital longer than those in Group B (5.63 ± 1.7) (p < 0.001). Conclusions: Preoperative nutritional status, quantified by the PNI, emerged as the significant predictor of complications in GMFCS IV-V cerebral palsy patients undergoing scoliosis surgery. Lower PNI was strongly associated with higher complication rates and prolonged hospitalization. These findings highlight the critical role of nutritional optimization in improving outcomes for this vulnerable population.

ID: AGS26/3479

A089: Does crossing the cervicothoracic junction in multilevel posterior cervical arthrodesis impact reoperation rates for adjacent segment disease? A retrospective analysis from the University of Michigan Hospital

Jawad Saad 1 , Peyton Goethe 2 , Noojan Kazemi 2

1 Central Michigan University, Saginaw, United States , 2 University of Michigan, Ann Arbor, United States

Introduction: Degenerative cervical spine disease as a cause of pain and disability, often requiring posterior cervical decompression and fusion, particularly for multilevel disease in aging patients. Determining the lower instrumented vertebra, which is the lowest point of the fusion, poses both anatomical and biomechanical challenges. This particularly occurs at the cervicothoracic junction (CTJ), where the flexible cervical spine transitions into the rigid thoracic spine. Ending fusion at C7 may place increased stress on the C7-T1 disc, potentially accelerating adjacent segment disease (ASD), while extending fusion into the thoracic spine may reduce this risk but limit postoperative mobility. Current literature presents conflicting data regarding whether crossing the CTJ reduces the need for reoperation due to ASD and other complications. Compare the rates of reoperation due to degeneration and complications such as pain, wound issues, and hardware failure between patients undergoing multilevel posterior cervical fusion terminating at C7 versus those extending across the cervicothoracic junction fusing at T1-T4. Material and Methods: This study is a retrospective cohort analysis using data from the University of Michigan Hospital Department of Neurosurgery. This included 393 adult patients who underwent elective multilevel posterior cervical fusion for degenerative cervical spine disease. Patients were divided into two cohorts: those whose fusions ended at or before C7 (C1-C7, n = 132) and those with constructs extending into the thoracic spine (T1-T4, n = 261). Reoperations were categorized as either for fusion extension due to degeneration or other complications like pain, hardware failure, or wound issues. Postoperative complications were also analyzed, and statistical significance was assessed using Fisher’s Exact Test (p < 0.05). Results: Among 393 patients, reoperation rates were lower in the CTJ cohort compared with those terminating at C7 (7.3% vs. 9.8%). Neurologic complications were frequent overall, with weakness affecting roughly one-quarter of patients in both groups, while radicular symptoms were higher in the CTJ group (21.8% vs. 18.9%). Wound-related complications, including surgical site infection and wound dehiscence, were also more common with CTJ extension. Additionally, dysphagia and the need for postoperative feeding tube placement occurred more often in the CTJ group, highlighting the increased morbidity associated with extending constructs across the cervicothoracic junction. Conclusion: Crossing the CTJ did not significantly reduce reoperations or surgical site complications. While pain was more frequent in the T1-T4 group, our findings differ from prior literature suggesting a protective effect of thoracic extension. Further investigation is warranted to clarify the clinical value of crossing the CTJ.

ID: AGS26/542

A090: High psychosocial impact of complications and maladaptive coping strategies among spine surgeons

Rishi Kanna 1 , S. Rajasekaran 1

1 Ganga hospital, Coimbatore, India

Introduction: Surgical complications are an inevitable part of spine practice, yet their psychosocial and professional impact on surgeons remains underexplored. This study surveyed practising spine surgeons to assess the burden of complications, coping mechanisms, and burnout. Material and Methods: An anonymous invited survey was completed by 50 practicing spine surgeons. The data collected included demographics, complication frequency, psychosocial impact, coping strategies, financial considerations, and burnout domains. Descriptive analysis was performed. Results: Respondents were predominantly male (100%) orthopedic surgeons (96%) with fellowship training (86%). Most practiced in tier-1 cities (54%) under private institutions (74%). The majority had 5-20 years of experience and performed 100-400 spine surgeries annually. Nearly all surgeons (96%) admitted being psychologically affected, reporting guilt (42%), depression (38%), and anxiety (36%). Multiple psychological domains were affected in 86% of respondents, particularly sleep (80%) and general well-being (78%). Emotional response was stronger when complications involved children (56%) or known patients (56%). Coping strategies included support from colleagues (74%), mentors (52%), and family (54%), alongside personal habits such as music (34%), sports (34%), and alcohol use (34%). Financial concessions were made by two-thirds of surgeons through fee or institutional cost reduction. Burnout was prevalent, with moderate-to-severe exhaustion in 44%, depersonalization in 40%, and poor professional accomplishment in 48%. While 74% perceived spine surgery as having a higher risk of complications compared to other specialties, only 12% contemplated career changes. Institutional support was rated “good” by 56%, “mild” by 34%, and absent by 10%. Conclusion: Complications in spine surgery exert profound psychosocial and professional consequences on surgeons, with high prevalence of burnout and maladaptive coping. Structured institutional support and targeted interventions are urgently needed.

OP11: Thoracolumbar Burst Care

ID: AGS26/319

A091: Bracing vs no bracing for the nonoperative management of A3/A4 thoracolumbar burst fractures: subanalysis from an international prospective multicenter study

Brian Karamian 1 , Charlotte Dandurand 2 , James Connelly 1 , Marcel Dvorak 2 , Klaus Schnake 3 , Mohammad El-Sharkawi 4 , S. Rajasekaran 5 , Miguel Hirschfeld 6 , Lorin Benneker 7 , Sebastian Bigdon 7 , Andrei Joaquim 8 , Harvinder Singh Chhabra 9 , Eugen Cezar Popescu 10 , Jin Tee 11 , Cumhur Oner 12 , Alex Vaccaro 13 , Gregory Schroeder 13 , Richard Bransford 14

1 University of Utah, Salt Lake City, United States ,

2 The University of British Columbia, Vancouver, Canada ,

3 Paracelsus Private Medical University Nuremberg, Nuremberg, Germany ,

4 Assiut University, Asyut, Egypt ,

5 Ganga Hospital, Coimbatore, India ,

6 Hospital Costa del Sol, Malaga, Spain ,

7 University of Bern, Bern, Switzerland ,

8 State University of Campinas, Sao Paulo, Brazil ,

9 Sri Balaji Action Medical Institute, New Delhi, India ,

10 Emergency Hospital, Iasi, Romania ,

11 The Alfred Hospital, Melbourne, Australia ,

12 University Medical Centers, Utrecht, Netherlands ,

13 Thomas Jefferson University Hospital, Philadelphia, United States ,

14 University of Washington, Seattle, United States

Introduction: The role of bracing in the non-operative management of thoracolumbar burst fractures remains controversial. The objective of this study is to compare the clinical outcomes of patients treated with or without a stabilizing brace for neurologically intact thoracolumbar burst fractures. Material and Methods: This post-hoc subanalysis of a prospective, multicenter study examined patients with type A3 or A4 thoracolumbar burst fractures without neurologic deficit who were treated nonoperatively with either a stabilizing brace or no brace. The primary outcomes were achieving the minimal clinically important difference (MCID) and patient acceptable symptom state (PASS) in the Oswestry Disability Index (ODI). Secondary outcomes included return to work, mental health, pain, and overall patient satisfaction with treatment. Results: In total, 76 patients met inclusion criteria for the study including 48 patients (63%) treated with bracing and 28 (37%) patients treated without bracing. Time-to-event analyses indicated no differences in achieving MCID or PASS in ODI between patients with and without a stabilizing brace (all p > 0.05). Significantly better scores were seen in the bracing group when compared to the non-bracing group at 3 months in terms of pain and overall satisfaction with care (both p < 0.05). Anxiety/depression scores were better in the bracing group at all follow-up visits except the 3-month follow-up visit when compared to the non-bracing group (all p < 0.05). Conclusion: We found no statistically significant differences in PROMs, return to work, or satisfaction at 1 year between patients treated with or without bracing. However, bracing was associated with improved pain, anxiety/depression scores, and satisfaction in the early period after injury. Bracing may offer advantages in terms of patient comfort and mental health in the short term but likely has no bearing on longer term outcome in neurologically intact patients with thoracolumbar burst fractures.

ID: AGS26/1204

A092: A novel surgical approach to thoracolumbar fractures: our experience

Andrea Cattalani 1 , Dora Riva 2 , Matteo Vitali 1 , Stefano Marasco 3 , Andrea Barbanera 1

1 Neurosurgery Unit, Department of Surgery, SS. Antonio e Biagio e Cesare Arrigo Hospital, Alessandria, Italy

2 Neurosurgery Unit, Department of Clinical-Surgical, Diagnostic and Pediatric Sciences, University of Pavia, Pavia, Italy

3 SSD di Neurochirurgia, Ospedale Umberto Parini di Aosta, Aosta, Italy

Introduction: The global burden of spinal fractures on health, and consequently on quality of life, has experienced an upward trend in recent years. The majority of these fractures (70%) occurs at the thoraco-lumbar junction. Notwithstanding the effort to standardize surgical indications concerning fractures, the discourse surrounding the optimal treatment options continues to be a topic of ongoing debate. The objective of this study was to present our personal experience concerning thoraco-lumbar fractures through the implementation of various techniques, including a novel approach. This approach involves the positioning of pedicle screws one level above and one level below the fractured vertebra, complemented by a stenting device and short pedicle screws at the fracture site. Each procedure was evaluated based on its efficacy in correcting deformities and sustaining said corrections. Material and Methods: This retrospective study included post-traumatic thoracolumbar fractures treated between January 2020 and March 2025 at a single institution (AOU S.S. Antonio e Biagio e Cesare Arrigo, Alessandria, Italy). A total of 31 patients were included in our analysis. Inclusion criteria were the following: (1) type A (subtype A2, A3, and A4) and type B fractures according to the AOSpine Thoracolumbar Spine Injury Classification System; (2) age > 18; (3) last follow-up at least 6 months after surgery. Our cohort of patients was divided into three groups based on the type of surgical procedure they underwent: group A (short-segment fixation plus stenting AND short pedicle screws at the fracture site), group B (short-segment fixation plus stenting OR short pedicle screws at the fracture site), and group C (long-segment fixation). CT scans and/or AP and lateral radiographs were performed pre-operatively, post-operatively, at 3 months, and at last follow-up. For each radiological image, Cobb’s kyphosis angle was taken into consideration. The degree of kyphosis correction was subsequently calculated by comparing the pre- and immediately post-operative Cobb’s angles. The following radiological controls (at 3 months and at last follow-up) were compared in order to assess the potential loss of correction.Quantitative data were reported as median and interquartile range (IQR) or mean and standard deviation (SD). Categorical variables were expressed as counts and percentages. Associations between patients baseline characteristics and type of surgery were assessed through two-tailed Kruskall-Wallis test or two-tailed Exact Fisher’s test. p < 0.05 was considered statistically significant. Results: In Group A, considered as the reference one, Cobb’s angle significantly decreased in the post-operative follow-up compared to baseline (β = -8.71, p < 0.001). Said correction was then significantly maintained in all the following controls (at 3-months and at last follow-up). Group C exhibited a statistically significant increase in Cobb’s angle over time compared to Group A at all time points. Changes in Cobb’s angle over time in Group B were similar to those observed in Group A. Conclusion: Short-segment fixation combined with stenting and the placement of short pedicle screws at the fractured vertebra seems to be a valid and effective technique to reduce post-traumatic kyphosis angle in thoraco-lumbar fractures and to maintain such correction over time.

ID: AGS26/221

A093: The AO Spine TL Classification System in acute thoracolumbar fractures: choice of treatment and implementation in clinical setting

Vahid Buyukayten 1 , Milan Ridderikhof 1 , Jos Twisk 1 , Olivier Groot 2 , Job Doornberg 3,4 , Frank Bloemers 1

1 Amsterdam UMC, Amsterdam, The Netherlands ,

2 UMC Utrecht, Utrecht, The Netherlands ,

3 UMC Groningen, Groningen, Netherlands ,

4 Flinders University, Adelaide, Australia

Introduction: The AO Spine TL Classification System does not provide explicit treatment advice for all thoracolumbar fractures. Further evaluation of its clinical applicability and consideration of incorporating additional relevant parameters associated with poor outcome is especially warranted. Therefore, this study investigated the association between this classification system and fracture level in deciding to treat patients with acute traumatic thoracolumbar fractures surgically or conservatively. The secondary aim was to evaluate the implementation of the classification system in clinical practice. Material and Methods: All adult (≥ 18 years) patients who presented in the Emergency Department of a level-one trauma centre, with one or more acute thoracolumbar (T1-L5) fractures were retrospectively included between 01/10/2016 and 30/06/2021. Exclusion criteria were: (1) pre-existing thoracolumbar fractures, (2) >20% missing data, (3) patients who suffered non-survivable injuries, and (4) pathological fractures. The primary outcome was definitive fracture management dichotomized into conservative or surgical treatment. Potential effect modifiers included: age, sex, comorbidities, trauma mechanism, Glasgow Coma Scale, Injury Severity Score, number of fractures and neurological injuries. The potential effect modifiers were adjusted for potential confounding. For the analyses binary logistic regression was used. The secondary outcome was adherence to the AO Spine TL Classification System in treatment decisions, calculated as the proportion of cases treated (conservatively or surgically) in line with its recommendations. Adherence was classified as poor (< 30%), intermediate (30-80%), or excellent (> 80%). Results: In total, 553 patients were included of which 387 (70%) patients received conservative treatment, and 166 (30%) patients underwent surgery. Of all fractures, 376 (68%) were type A0-A3, 46 (8.3%) type A4, 34 (6.2%) type B1, 81 (15%) type B2, and 16 (2.9%) type C. Patients with type A4 fractures were treated surgically more often than patients with type A0, A1, A2, or A3 fractures (OR = 31.85; 95% CI = 13.21-76.80; p ≤ 0.01). Fracture level was not associated with choice of treatment (OR = 0.93; 95% CI = 0.61-1.42; p = 0.75). For both associations, none of the investigated eight parameters were effect modifiers (all p > 0.10). Adherence of physicians to the morphology components of the classification system was 95% as to types A0-A3, 86% as to B2-B3, and 100% as to C. As to A4 and B1, adherence could be determined, as the choice of treatment is left to the discretion of the physician. Adherence to the neurological and modifier components of the classification system could not be determined, due to a lack of documentations by physicians. Conclusion: As expected, patients with type A4 fractures are treated surgically more often than patients with A0, A1, A2, or A3 fractures. The analyses of patient and vertebral fracture characteristics showed no effect modifiers. The implementation of the AO Spine TL Classification System regarding the morphological components appears to be excellent. However, the implementation of the overall algorithm of the classification system remains suboptimal.

ID: AGS26/2722

A094: Removal versus retention of posterior spinal implants in patients with healed thoracolumbar junction fractures: analysis of clinical and radiographic outcomes

Raboudi Hamza 1 , Jlidi Mohamed 1 , mouldi lamouchi 1 , Karim Mallek 1 , Ismail Karray 1 , Jihed Bechikh 1 , Sami Mbarki 1 , Hedi Atig 1 , Walid Bouaicha 1 , Selim Daas 1

1 Orthopedics and Traumatology Department, Hospital of Mohamed Taher Maamouri, Nabeul, Tunisia

Introduction: Implant removal following successful bone union is a common practice in orthopedic surgeries, including spinal procedures. General indications for spinal implant removal include infection, nonunion, symptomatic misplacement, implant breakage or loosening, local irritation, or growth disturbance. However, there is controversy regarding the benefits and indications for routine spinal implant removal in asymptomatic patients with successful unions. The objective of our study is to evaluate the impact of implant removal on functional and radiological outcomes. Material and Methods: We conducted a retrospective study of 120 patients operated on for thoracolumbar junction fractures between January 2010 and January 2020. Patients were divided into two groups: Group A including 38 patients who underwent implant removal, and Group B including 82 patients without implant removal. Functional and clinical outcomes were evaluated by the Visual Analogue Scale (VAS), Oswestry Disability Index (ODI), and Dallas Pain Questionnaire. Radiological outcomes were evaluated by measuring regional kyphosis, vertebral kyphosis, traumatic regional angulation, and Beck index at postoperative and the last follow-up. The evaluation of the results was done by calculating the loss of correction of radiological parameters between the postoperative radiographs and the last follow-up radiographs as well as at final follow-up. Results: The mean age was 34.84 years in Group A and 50.40 years in Group B. Group A included 17 women and 21 men, whereas Group B included 29 women and 53 men. Diabetes was present in 8 patients in Group A and 12 patients in Group B. The mean BMI was 27 in Group A and 28.12 in Group B. The fractured vertebrae were distributed as follows: in Group A, 15 cases involved T12, 22 involved L1, and 1 involved both T12 and L1; in Group B, 28 cases involved T12, 52 involved L1, and 1 involved both T12 and L1. Functional assessment showed no significant differences between the two groups: the mean Visual Analog Scale (VAS) score was 2.53 in Group A versus 2.33 in Group B (p = 0.43); the mean Oswestry Disability Index was 23.11 in Group A versus 25.08 in Group B (p = 0.51); and the Dallas Pain Questionnaire score was 15.40 versus 16.38, respectively (p = 0.49). Regarding the radiological outcomes, we didn’t noted a significant difference between the two groups in terms of the loss of reduction of vertebral kyphosis, regional kyphosis, traumatic regional angulation, and Beck index did not differ significantly between the two groups. Furthermore, UCLA grading at the last follow-up showed no statistically significant differences between the two groups (p = 0.31). Conclusion: Our findings suggest that implant removal does not significantly affect either functional outcomes or radiological outcomes after thoracolumbar junction fractures. Further studies are required to identify specific subgroups of patients who may benefit from implant removal.

ID: AGS26/283

A095: The effect and mechanism of methylprednisolone combined with ADSCs which over-expression BDNF in the treatment of acute spinal cord injury in rats

Wenchen Ji 1

1 Department of Orthopaedics, the First Affiliated Hospital of Xi’an Jiaotong University, Xi'an, China

Introduction: To evaluate the effect and mechanism of Methylprednisolone combined with ADSCs which over-expression BDNF in the treatment of acute spinal cord injury in rats. Material and Methods: 48 spinal cord injury rats were divided into four groups: Control group (just cut off the spinal cord and no treatment), Methylprednisolone group, ADSCs/BDNF-GFP group and the combined group. BBB scores, BDNF expression and cell apoptosis rate were compared between four groups at 1w,2w and 4w after operation. Expression of Bax and Bcl-2 protein which were related with apoptosis were detected 4w after operation. Results: Compare to other groups, the combined group could obviously improve rats’ lower limbs function after SCI, could significantly reduce the rate of cell apoptosis, promote the expression of Bcl-2 and inhibit the expression of Bax. Conclusion: The neurological function of rats with spinal cord injury can be obviously improved by ADSCs with over-expressed BDNF combined with Methylprednisolone. The main mechanism may be improving expression of BDNF and decrease apoptosis of spinal cord cells.

ID: AGS26/338

A096: Patient Reported Outcomes following Operative and Non-Operative Management of A3 vs A4 Thoracolumbar Burst Fractures

Barry Kweh 1 , Charlotte Dandurand 2 , Alex Vaccaro 3 , Klaus Schnake 4 , Mohammad El-Sharkawi 5 , Richard Bransford 6 , Harvinder Singh Chhabra 7 , Ulrich Spiegl 8 , Dimitri Hauri 9 , Cumhur Oner 10 , Gregory Schroeder 3 , Jin Tee 1,11,12

1 National Trauma Research Institute, Melbourne, Australia, 2 Combined Neurosurgical and Orthopedic Spine Program, Vancouver General Hospital, University of British Columbia, Vancouver, Canada, 3 Rothman Institute, Philadelphia, United States, 4 Center for Spinal and Scoliosis Surgery, Malteser Waldkrankenhaus St. Marien Erlangen, Germany, Erlangen, Germany, 5 Department of Orthopaedic and Trauma Surgery, Faculty of Medicine, Assiut University, Egypt, Egypt, 6 Department of Orthopaedics and Sports Medicine, Harborview Medical Center, University of Washington, Seattle, United States, 7 Department of Orthopaedics and Spine Surgery, Ganga Hospital, Coimbatore, Tamil Nadu, India, 8 Clinics of Trauma Surgery, Orthopaedics, and Reconstructive Surgery, München Klinik Harlaching, Sanatoriumplatz 2, 81545, Munich, Germany, 9 AO Foundation, Stettbachstrasse 6, 8600, Dübendorf, Switzerland, 10 University Medical Centers,, Utrecht, Netherlands, 11 Alfred Hospital, Melbourne, Australia, 12 Monash University, Melbourne, Australia

Introduction: The quality of life that patients experience after sustaining thoracolumbar burst fractures is often overlooked in preference of radiographic features or neurological function. The objective of this study was to evaluate functional and quality of life outcomes following operative and non-operative management of thoracolumbar burst fractures. Material and Methods: A prospective multi-centre observational study of neurologically intact patients who sustained traumatic thoracolumbar burst fractures managed surgically or non-surgically was performed. Outcome measures included the EQ-5D index, EQ-5D visual analogue scale (VAS), pain numerical rating scale (NRS), the pioneering AO Spine Patient Reported Outcome Spine Trauma (PROST) tool and patient satisfaction. Follow-up time points included at discharge, 6 weeks, 3 months, 6 months, 1 year and 2 years. The Wilcoxon rank sum test was used to evaluate differences between A3 and A4 fracture morphologies for continuous variables whilst the Cochran-Armitage trend test was utilized to evaluate differences in the ordinal outcome anxiety and depression. Mixed models for repeated measures (MMRM) were used for continuous outcomes using unstructured covariance. A generalized estimating equations (GEE) model was employed for the ordinal outcomes of anxiety and depression.. The models were adjusted for key variables including sex, age, BMI, smoking status. TLICS score and treatment modalities. Results: 198 patients were identified. The EQ-5D index determined there was no statistically significant difference for A3 and A4 fractures at 2 year follow-up for operative (p=0.48) and non-operative management (p=0.51). A similar finding was noted with the EQ-5D VAS measure for surgical (p=0.33) and non-surgical treatment (p=0.21). Adjusted mixed effects model for the EQ-5D index and EQ-5D VAS confirmed patients with operatively and non-operatively managed A3 (p<0.001) and A4 fractures (p<0.001) improved at every follow-up checkpoint up to 2 years. A GEE model demonstrated there was a higher proportion of patients who were less anxious or depressed among operatively managed patients with A4 fractures compared to A3 fractures at 2 year follow-up (OR 7.7, 95% CI 1.5 − 40.5, p=0.02) but this was not true for those non-operatively treated (OR 0.8, 95% CI 0.1 − 4.4, p=0.78). MMRM for pain NRS and AO Spine PROST suggested there was a significant improvement in outcome compared to discharge for both surgically and non-surgically treated patients (p<0.01) but no significant difference between the surgical subgroups. Patient satisfaction was similar between A3 and A4 fractures for non-operatively (p=0.62) and operatively treated patients (p=0.50) at final follow-up). Conclusion: Contemporary management of thoracolumbar burst fractures in neurologically intact patients has evolved to the extent that there is improvement in function, pain and disability for both A3 and A4 fractures regardless of whether surgical or non-surgical management is pursued. We demonstrated that the cohort of patients who sustain the more severe A4 morphology experienced improved mental health outcomes upon final 2 year follow-up if operatively managed. Ultimately patient satisfaction was similar on long-term follow-up. These findings suggest that larger randomized studies using the AO Spine Thoracolumbar Injury Classification System are still required to adequately comprehend the biopsychosocial effect of these fractures.

ID: AGS26/3052

A097: Against all odds: six-year outcomes of thoracolumbar burst fracture surgery in a district hospital of Bangladesh

Mahamudul Hasan 1,2

1 Bangladesh Spine and orthopaedic Hospital, Dhaka, Bangladesh

2 Satkhira Medical College Hospital, Satkhira, Bangladesh

Introduction: Thoracolumbar burst fractures represent one of the most frequent and challenging spine injuries. While outcomes from tertiary centers are well documented, there is a paucity of evidence from district-level hospitals in low- and middle-income countries where resources are limited. This study evaluates surgical outcomes over six years from a single surgeon operating in a district hospital in Bangladesh, highlighting the factors that influence recovery despite infrastructural constraints. Material and Methods: A retrospective review was conducted of consecutive patients with thoracolumbar burst fractures (T10-L4) who underwent surgery between 2019 and 2025. Demographics, mechanism of injury, ASIA grade, time to surgery, fracture characteristics, operative details, and complications were recorded. Outcomes included neurological recovery, functional status (ODI/ambulation), radiographic parameters (kyphotic angle, canal compromise), and complications. Univariate and multivariable analyses were used to identify independent predictors of poor outcome. Results: A total of 163 patients mean age 32 years; 62% male were included with a median follow-up of 9 months. Neurological improvement was observed in 92% of patients; mean kyphosis correction was excellent, at final follow-up. Independent predictors of poor outcome were delayed surgery > 72 hours, severe preoperative deficit (ASIA A-B), canal compromise > 50%, and age > 60 years . Despite limited imaging, implants, and critical care support, complication rates remained within reported global ranges. Conclusion: Thoracolumbar burst fracture surgery can achieve satisfactory neurological and radiological outcomes even in resource-constrained district hospitals. Timely surgery, early diagnosis, and pragmatic operative strategies are key determinants of success. These results underscore the importance of strengthening district-level spine care capacity in low- and middle-income countries.

ID: AGS26/1900

A098: Reliability and validity of the AO Spine Patient Reported Outcome Spine Trauma (PROST) version 1.1 in individuals with longstanding spinal cord injury

Marcel Post 1 , Aline Hakbijl 1 , Sylvia Egberts 1 , Janneke Stolwijk 1 , Rutger Osterthun 2 , Tijn van Diermen 3 , Marcel F. Dvorak 4 , Richard Bransford 5 , Andrei Joaquim 6 , Harvinder Singh Chhabra 7 , Cumhur Oner 8 , Said Sadiqi 8,9

1 Rehabilitation Center De Hoogstraat, Utrecht, The Netherlands ,

2 Rijndam Rehabilitation Center, Rotterdam, The Netherlands ,

3 Sint Maartenskliniek, Nijmegen, The Netherlands ,

4 University of British Columbia, Vancouver, Canada ,

5 University of Washington School of Medicine, Seattle, United States ,

6 State University of Campinas, Campinas-Sao Paulo, Brazil ,

7 Sri Balaji Action Medical Institute, New Delhi, India ,

8 University Medical Center Utrecht, Utrecht, The Netherlands ,

9 St. Antonius Hospital, Utrecht, The Netherlands

Introduction: The AO Spine PROST (PROST) is a validated measure of health-related quality of life (HRQoL) designed for use in individuals with spine trauma with no neurological deficit or incomplete spinal cord injury (SCI). For use in individuals with motor complete traumatic or non-traumatic spinal cord disease (SCI/D), a slightly adapted PROST 1.1 was developed. This study investigates the reliability and validity of the PROST 1.1 in individuals with longstanding motor-complete SCI/D. Materials and Methods: Among three specialized rehabilitation centers in the Netherlands, a total of 72 Individuals with motor complete SCI/D were included of whom 62 also completed the second administration after a median of 7 (IQR 5-11) days. Internal consistency (Cronbach’s alpha), test-retest reliability (Intra-Class Correlations (ICC with 95% Confidence interval (CI)) and concurrent validity (Spearman correlations ≥ .50 with reference measures) were investigated. Results: Median age was 54 (Inter-Quartile Range 40-64) years, 37% had non-traumatic SCI/D, 63% ASIA Injury Severity grade A, 79% paraplegia, and median time since onset was 20 (IQR 8-34) months. An approximately normal distribution without floor or ceiling effects was observed in the total PROST 1.1 score. Alpha was good (.90) and the ICC was good (.84; 95% CI .75-.90). Most correlations between the PROST 1.1 and the reference measures were strong (≥ .50) as expected, ranging from .27 with the VAS-score of the EuroQoL 5D-5L up to .70 with the sum-score of the EuroQoL 5D-5L. Conclusion: The slightly adapted PROST 1.1 appears to be applicable in individuals with motor-complete SCI/D to evaluate their HRQoL

ID: AGS26/605

A099: Return to work after treatment for thoracolumbar burst fractures: results from a prospective multicenter international cohort study

Mohamed Ali 1 , Charlotte Dandurand 2

1 Prince Mohamed Ben AbdelAziz Hospital, Riyadh second Cluster, Riyadh, Saudi Arabia

2 Vancouver General Hospital University of British Columbia, Vancouver, Canada

Introduction: The appropriate treatment option for neurologically intact patients with thoracolumbar (TL) burst fractures, surgery or nonoperative therapy, is still contested. This study aims to compare the time to return to work (RTW) between surgical treatment versus nonsurgical treatment of TL burst fractures in neurologically intact patients. Material and Methods: Clinical and outcome data were obtained from a prospective, observational multicenter cohort study that compared the surgical and nonsurgical treatment of TL burst fractures in neurologically intact patients. We evaluated the RTW information via the patient diary utilized to score the Indirect Cost Questionnaire. Baseline demographics, clinical and injury details were compared between patients who returned to work (RTW) and those who did not (NRTW) by 3 months and 1 year. ODI scores at baseline and 3 months were compared between patients who returned to work and those who did not by 3 months. Baseline and 1-year ODI scores were compared between patients who returned to work and those who did not by 1 year. Kaplan-Meier survival curves were plotted with median event times and a 95% CI calculated separately for patients with non-operative and operative management. Kaplan-Meier curves were compared using a log-rank test. A multivariable Cox regression model was fitted to compare the hazard ratio (HR) of RTW between treatment groups. Results: 198 eligible and enrolled patients without major protocol deviations were recruited from fourteen sites across North America, Europe, the Middle East, and Asia. 122 patients (61.6%) were treated surgically, and 76 (38.4%) were treated nonsurgically. Patients who had not returned to work by 3 months were more frequently smokers (19.2% vs 8.2%, p = 0.04) and experienced more frequently a low-energy trauma (25.6% vs 12.9%, p = 0.04) and had a higher ODI at 3 months (mean (SD) 21.2 (14.6) versus 11.1 (11.0), p < 0.01) than RTW patients. Patients who had not returned to work by one year were significantly more frequently of Asian ethnicity (22.7% vs 15.0%) or of other origin (9.1% vs 0.8%), were more often smokers (31.8% vs 10.2%, p = 0.01), had a shorter time from injury to treatment decision (mean (SD) 1.4 (1.5) vs 2.4 (2.2), p = 0.02) and had higher ODI values at baseline (mean (SD) 70.3 (16.4) vs 55.4 (19.7) vs p < 0.01) and at 1 year (mean (SD) 16.0 (14.4) vs 6.4 (8.4) vs p < 0.01). No significant differences were found in primary injury type or other patient and injury characteristics, and treatment modality. Median time to RTW did not significantly differ between surgical and nonsurgical groups (84.0 days [95% CI: 67.0-96.0] vs. 120.0 days [95% CI: 71.0- 136.0], p = 0.236). Multivariable Cox regression showed no significant difference between the surgical and the non-operative treatment groups (HR: 1.23, 95%-CI: 0.79, 1.92). Conclusion: When considering RTW by 3 months and 1 year, smoking and higher ODI at the respective timepoints correlated with no return to work. Otherwise, all patient demographics and injury characteristics did not correlate with RTW. There was no statistically significant difference in time to RTW between surgical and nonsurgical treatment of TL burst fractures in neurologically intact patients within one year.

OP12: Harnessing the Immune System

ID: AGS26/734

A100: Sterile wound healing or deep infection? Serum protein signatures identify mechanistic pathways after spinal surgery: a prospective matched-cohort study

Siegmund Lang 1 , Pia Fehrenbach 2,3 , Daniel Haschtmann 4 , Tamas Fekete 4 , Maja Gocević 4 , Puk Kwant 2 , Esther de Jong 3 , T. Fintan Moriarty 2 , Sibylle Grad 2 , Markus Loibl 4

1 University Hospital Regensburg, Regensburg, Germany,

2 AO Research Institute Davos, Davos, Switzerland,

3 Amsterdam UMC, Amsterdam Institute for Infection and Immunity, University of Amsterdam, Amsterdem, Netherlands,

4 Schulthess Clinic Zurich, Zurich, Switzerland

Background: Surgical site infection (SSI) after spinal fusion remains a severe complication with high morbidity. Conventional markers such as CRP or leukocyte count often lack specificity to distinguish SSI from sterile post-operative inflammation. Little is known about the mechanistic pathways that drive SSI in contrast to normal healing. This study therefore aimed to identify plasma protein signatures that not only differentiate SSI from uncomplicated recovery but also provide insights into the underlying immunopathology using targeted proteomics. Materials and Methods: Between October 2020 and May 2024, 43 spine-surgery patients were prospectively enrolled, 22 with SSI and 21 without SSI. After exclusion for matching and technical criteria, 17 infected patients (I) and 20 non-infected controls (NI) were analyzed. Controls were matched to infected patients for age, sex, and surgical invasiveness. Plasma samples were collected pre-operatively and 6 weeks post-operatively in non-infected patients, and at SSI diagnosis in infected patients. Ninety-two proteins were quantified using the OLINK inflammation panel (NPX log2 scale). Group comparisons were performed using Wilcoxon rank-based tests. The study was approved by the institutional ethics committee, and all patients gave written informed consent. Results: The mean age was 63.4 ± 15.7 years in infected patients and 63.0 ± 10.3 years in controls (p = 0.91), with 47.1% vs. 45.0% females (p = 0.87). The surgical spectrum ranged from single-level decompressions to multilevel fusions without differences in operated levels. Median time to SSI diagnosis was 18.8 days, and Staphylococcus aureus was the most frequent isolate (n = 10). In NI patients, post-operative changes reflected a controlled sterile response: Compared to pre-operative levels, EN-RAGE (p = 0.0012) and IL-33 (p = 0.0433) increased, reflecting the systemic response to the surgical intervention, while 4E-BP1 (p = 0.0335), SIRT2 (p = 0.0216), STAMBP (p = 0.0195), and Flt3L (p = 0.0084) decreased, suggesting attenuation of translational and metabolic stress. Comparing NI and I patients postoperatively displayed a pronounced inflammatory signature in I patients with higher levels of IL-17A, IL-6, IL-8, CXCL1, CCL3, and MCP-3, driving neutrophil recruitment. Angiogenic mediators VEGFA, HGF, and TGF-α were elevated compared to NI patients, indicating infection-associated tissue remodeling. Importantly, an elevated osteolytic protein cluster was evident in I patients: OSM, MMP-1, and TNFSF14 were upregulated, consistent with infection-driven matrix degradation, while TRANCE (RANKL)was paradoxically reduced (all markers p < 0.001). Conclusion: Proteomic profiling distinguished sterile wound healing from SSI. While uncomplicated healing was marked by transient alarmin release with resolution of stress pathways, SSI was characterized by an IL-17/chemokine surge, angiogenic activation, and induction of osteolytic mediators, supporting the development of biomarker panels for early and specific SSI detection.

ID: AGS26/1669

A101: Immune profiling identifies specific cell alterations in postoperative spinal infections: a prospective matched-control study

Siegmund Lang 1 , Pia Fehrenbach 2,3 , Daniel Haschtmann 4 , Tamas Fekete 4 , Maja Gocević 4 , Puk Kwant 2 , Esther de Jong 3 , T. Fintan Moriarty 2 , Sibylle Grad 2 , Markus Loibl 4

1 University Hospital Regensburg, Regensburg, Germany ,

2 AO Research Institute Davos, Davos, Switzerland ,

3 Amsterdam UMC, Amsterdam Institute for Infection and Immunity, University of Amsterdam, Amsterdem, The Netherlands ,

4 Schulthess Clinic Zurich, Zurich, Switzerland

Introduction: Surgical site infections (SSIs) are serious complications following spine surgery. They are associated with increased patient morbidity, longer hospitalization, and the need for complex revision procedures. The aim was to identify immune cell alterations associated with SSIs and to understand how these differ from those in patients without SSIs. Materials and Methods: A total of 43 patients undergoing lumbar spine surgery were prospectively enrolled. After quality control and matching, 17 patients who developed SSIs (group I) and 20 matched non-infected patients (group NI) were included for immune profiling. Matching parameters included age, sex, BMI, and surgical invasiveness. Peripheral blood mononuclear cells (PBMCs) were isolated from blood samples drawn postoperatively and analyzed using high-dimensional mass cytometry (43 markers). Cell types and activation/exhaustion markers were identified using uniform manifold approximation and projection (UMAP) dimensionality reduction, PhenoGraph clustering, and statistical comparisons. Results: The mean age of patients was 63.4 ± 15.7 years in group I and 63.0 ± 10.3 years in group NI. The gender distribution was similar (47% vs. 45% female). SSIs occurred after a median of 18.8 days and were predominantly caused by Staphylococcus aureus (n = 10), coagulase-negative staphylococci (n = 2), and Escherichia coli (n = 2). Surgical procedures included both minor (e.g., decompression) and major interventions (e.g., multilevel fusions), equally distributed between groups. Mass cytometry identified 46 immune metaclusters, of which 30 differed significantly between infected and non-infected patients. Six clusters were upregulated in the infection group and 24 downregulated. Innate immune cells such as NK cells (p = 0.0043), monocytes (p = 0.0357), myeloid DCs (p = 0.0011), plasmacytoid DCs (p = 0.0458), and CD1 + DCs (p = 0.0057) were reduced in infected patients. Furthermore, γδ T cells (p = 0.0070) and double-positive T cells (p = 0.0152) were also decreased. In contrast, B cells (p = 0.0305) and Th17 cells (p = 0.0152) were elevated in infected patients. While total CD4 + and CD8 + T cell counts did not differ significantly, activation markers showed differences on CD4+ T cells- CD69 + CD4 + T cell (p = 0.0023) and HLA-DR + CD4 + T cell (p = 0.0032)) were increased in infected patients. Moreover, PD-1 + CD4 + T cell (p = 0.0141) were increased in infected patients. HLA-DR + CD8 + T cells (p = 0.0424) and OX-40 + CD8 + T cells (p = 0.0470) were reduced in infected patients. Discussion: This study suggests that patients developing SSIs following spinal surgery exhibit a distinct immune cell signature. Infected patients showed a suppressed innate immune profile. These findings support the relevance of high-dimensional immune profiling for better understanding the host response to SSI. Further research is warranted to validate immune-based markers for clinical risk stratification and early diagnosis.

ID: AGS26/117

A102: In vivo segmental vertebral kinematics in patients with degenerative lumbar scoliosis

Fei Xu 1 , Weishi Li 1

1 Orthopaedic Department, Peking University Third Hospital, Beijing, China

Introduction: This study aimed to find a standard of the vertebra kinematics during functional weight-bearing activities in degenerative lumbar scoliosis (DLS) patients. Material and Methods: Fifty-four patients were involved into this study with forty-two in DLS group and twelve in the control group. The three-dimensional (3D) vertebral models from L1 to S1 of each participant were reconstructed by computed tomography (CT). Dual-orthogonal fluoroscopic imaging, along with FluoMotion and Rhinoceros software, was used to record segmental vertebral kinematics during functional weight-bearing activities. The primary and coupled motions of each vertebra were analyzed in patients with DLS. Results: During flexion-extension of the trunk, anteroposterior (AP) translation and craniocaudal (CC) translation at L5-S1 were higher than those at L2-3 (9.3 ± 5.1 mm vs. 6.4 ± 3.5 mm; p < 0.05). The coupled mediolateral (ML) translation at L5-S1 in patients with DLS was approximately three times greater than that in the control group. During left-right bending of the trunk, the coupled ML rotation at L5-S1 was higher in patients with DLS than that in the control group (17.7 ± 10.3° vs. 8.4 ± 4.4°; p < 0.05). The AP and CC translations at L5-S1 were higher than those at L1-2, L2-3, and L3-4. During left-right torsion of the trunk, the AP translation at L5-S1 was higher as compared to other levels. Conclusion: The greatest coupled translation was observed at L5-S1 in patients with DLS. Coupled AP and ML translations at L5-S1 were higher than those in healthy participants. These data improved the understanding of DLS motion characteristics.

ID: AGS26/2270

A103: Free calcium and the extracellular calcium-sensing receptor regulate intervertebral disc degeneration and calcification

Michael Grant 1 , Laura Epure 1 , John Antoniou 1 , Fackson Mwale 1

1 McGill University, Montreal, Canada

Introduction: Degenerative disc disease (DDD) is a common cause of lower back pain. Calcification of the intervertebral disc (IVD) has been correlated with DDD, and is especially prevalent in scoliotic discs. The appearance of calcium deposits has been shown to increase with age, and its occurrence has been associated with several other disorders such as hyperparathyroidism, chondrocalcinosis, and arthritis. Trauma, vertebral fusion and infection have also been shown to increase the incidence of IVD calcification. Our data indicate that Ca 2+ and expression of the extracellular calcium-sensing receptor (CaSR) are significantly increased in mild to severely degenerative human IVDs. In this study, we evaluated the effects of Ca 2+ and CaSR on the degeneration and calcification of IVDs. Material and Methods: IVD tissue and cells were isolated from donor tissue. Immunohistochemistry (IHC) was performed to determine expression of CaSR. Free calcium levels were measured and compared between Thompson graded discs. Western blotting and RT-qPCR were performed on cultured human NP and AF cells to demonstrate expression of CaSR, matrix proteins aggrecan and collagen, and catabolic enzymes following incubation in either increasing concentrations of Ca 2+ [1.0-5.0 mM], CaSR allosteric agonist (cincalcet, 1 uM), or IL-1b [5 ng/mL] for 7 days. Ex vivo bovine IVD organ cultures were prepared using PrimeGrowth Disc Isolation System (Wisent Bioproducts). IVDs were cultured in 1.0, 2.5 mM Ca 2+ or with cinacalcet for 21 days. Histology and Western blotting were performed to measure changes matrix composition. Structural stiffness of disc tissues due to calcification was determined using the MACH-1 mechanical testing system (Biomomentum, Laval, Quebec). Results: Human disc tissue demonstrated significant increases in free calcium from early (G2) to severe (G4) degeneration (p < 0.001; n = 6-15 discs per grade). Treatment with either Ca 2+ the CaSR agonist or Il-1b significantly decreased matrix protein synthesis (aggrecan and Col II; p < 0.001; n = 4). However, only IL-1b upregulated catabolic enzyme expression (p < 0.0001; n = 4). Similar effects were observed in whole organ cultures, as Ca 2+ and cinacalcet decreased proteoglycan and collagen content. Ca 2+ -treated IVDs demonstrated with calcium deposits in NP and AF tissues yet both Ca 2+ and CaSR agonist increased alkaline phosphate expression (p < 0.01; n = 4). Biomechanical studies on Ca 2+ -treated IVDs demonstrated significant increases in stiffness of NP and AF tissues (p < 0.001 and p < 0.05, respectively; n = 3). Interestingly, IL-1b dose-dependently increased CaSR expression in IVD cells and in IVD organ cultures. Conclusion: Our results suggest that changes in the local concentrations of calcium and activation of CaSR occur in disc degeneration and effect matrix protein synthesis, calcification and IVD biomechanics. Inflammation may be a driving force for CaSR expression in disc cells enhancing Ca 2+ -induced degeneration. The effects of Ca 2+ on IVD degeneration should be considered when investigating repair strategies of the disc.

ID: AGS26/302

A104: Molecular characterization and classification of chordoma dictate individualized therapeutic strategies

Tong Meng 1

1 Shanghai General Hospital, Shanghai, China

Introduction: Chordoma, a rare and heterogeneous mesenchymal malignancy, presents with distinct clinical and biological behaviors. To guide individualized therapeutic strategy, we described the comprehensive clinical features of chordoma along with its underlying biological characteristics. Additionally, the accurate molecular classification and specific therapeutic options were also determined. Material and Methods: We enrolled a large cohort of chordoma including 102 patients and described their clinical, imaging and histopathological features. Tandem mass tag-based proteomic analysis was performed to assess overall protein levels and molecular classification was defined by the nonnegative matrix factorization clustering. Associations between clinical-proteomic features and overall survival were assessed by Lasso regression and Cox proportional hazards regression models. Potential treatment strategies were further evaluated by chordoma cell, cell- and patient-derived xenografts according to molecular subtypes. Results: In the final cohort of chordoma, a total of 63 patients (61.76%) experienced recurrence and 37 patients (36.27%) died of all causes with the median survival time of 43 (range, 1-192) months. We classified chordoma into three molecular subtypes of bone microenvironment dominant, mesenchymal derived, and mesenchymal to epithelial transition mediated, which were also identified as prognostic factors, along with resection mode and distant metastasis, in the multivariable Cox proportional hazards model. The three subtypes exhibited distinct biological traits of osteoclastogenesis and immunogenicity, oxidative phosphorylation, and receptor tyrosine kinase activation, suggesting targeted therapeutic strategies of Denosumab, S-Gboxin and Anlotinib, respectively. Notably, these approaches demonstrated positive treatment outcomes for each subtype in vivo and in vitro , as well as in patient-derived xenograft models. Conclusion: Our work sheds light on clinical-proteomic characteristics of chordoma that deepens our understanding of this rare disease and provides candidate precision treatment strategies for chordoma according to molecular classification, underscoring their potential for clinical application.

ID: AGS26/3136

A105: Dual mechanisms of susceptibility: distinct immune and inflammatory landscapes of type 1 and type 2 diabetes in murine spinal implant-associated infection and the potential for metabolic intervention

Thomas Olson 1 , Trevor LLoyd 1 , Rene Chun 1 , Autreen Golzar 1 , Joshua Wiener 1 , Chris Hamad 1 , John Adams 1 , Nicholas Bernthal 1 , William Sheppard 1

1 UCLA, Los Angeles, United States

Background: Diabetes mellitus is a well-recognized risk factor for infection and wound complications following spinal and orthopedic surgery. However, clinical studies often group type 1 diabetes mellitus (T1DM) and type 2 diabetes mellitus (T2DM) together, overlooking their distinct pathophysiology. This study directly compared infection burden, wound healing, and immune response between T1DM and T2DM in a murine spinal implant-associated infection model. Methods: T1DM was induced by streptozotocin injection and T2DM by high-fat, high-sucrose diet in C57BL/6J mice. Stainless steel spinal implants were placed, with select groups inoculated with bioluminescent Staphylococcus aureus (Xen36). Infectious burden was quantified longitudinally by noninvasive bioluminescent imaging and post-mortem colony forming units. Wound integrity was assessed by Kaplan-Meier survival of dehiscence. Systemic immunity was profiled by complete blood counts and multiplex cytokine analysis. Rescue was evaluated in both cohorts using semaglutide to evaluate the impact of metabolic modulation. Results: Both T1DM and T2DM mice exhibited significantly greater infectious burden compared with controls. T2DM mice demonstrated the highest and most sustained bacterial load, with bioluminescence peaking early and remaining elevated. T1DM mice showed a delayed but pronounced infectious peak, suggesting differential bacterial persistence. Wound complications were most severe in T1DM, with rapid and frequent dehiscence occurring within the first postoperative week. T2DM animals exhibited intermediate rates and timing of wound breakdown. Control wounds in both cohorts demonstrated zero cases of dehiscence. Cytokine profiling revealed distinct inflammatory landscapes. T2DM mice had elevated baseline cytokine levels, including C-reactive protein, G-CSF, GM-CSF, and IL-6, but mounted only minimal increases after infection, consistent with a state of chronic inflammation and blunted acute responsiveness. In contrast, T1DM mice displayed lower baseline cytokines yet demonstrated more robust infection-induced increases, indicating an acutely reactive but less durable immune profile. Complete blood counts supported these differences: both T1DM and T2DM groups had lower total white blood cells, lymphocytes, and red blood cell counts compared with controls, though neutrophil increases after infection were proportionally greater in diabetics. Semaglutide treatment in T1DM and T2DM mice improved glucose tolerance and decreased both peak and cumulative infectious burden. Wound dehiscence was delayed and less frequent compared with untreated T2DM, supporting the potential for metabolic intervention to enhance host defense. Conclusions: T1DM and T2DM impair defense against spinal implant infection through distinct mechanisms. T2DM is characterized by obesity, chronic systemic inflammation, and an attenuated acute immune response, while T1DM exhibits lower baseline immune tone but rapid, fragile responses accompanied by severe wound failure. Both groups demonstrated impaired hematopoiesis and reduced adaptive cell counts, contributing to heightened infection susceptibility. Metabolic modulation with semaglutide mitigated infectious and wound complications, underscoring the potential for tailored perioperative strategies in both diabetic subtypes.

ID: AGS26/3346

A106: The impact of histone deacetylase (HDAC) inhibition on neurobehavioural outcomes in preclinical models of traumatic and non-traumatic spinal cord injury: a systematic review

Natalia Jogodzinska 1 , Caleb Cole 1 , Jamie Brannigan 1 , Renuka Chintapalli 1 , Benjamin Davies 1 , Mark Kotter 1 , Oliver Mowforth 1

1 Division of Neurosurgery, Department of Clinical Neurosciences, University of Cambridge, Cambridge, United Kingdom

Introduction: Spinal cord injury (SCI) is a traumatic injury resulting in significant life-changing disability. Elucidating the molecular processes associated with SCI may help to design novel therapeutics targeted at improving patient outcomes. Current pharmacological candidates include histone deacetylase (HDAC) inhibitors, which have been studied in the context of cancer, neurodegenerative and rheumatological disease. The anti-inflammatory properties of HDAC inhibitors have been postulated to be of value in SCI. The objective of this review was to synthesise the impact of HDAC inhibitors on neurobehavioural outcomes in preclinical studies of traumatic and non-traumatic SCI and to evaluate the suitability of HDAC inhibitors for clinical trials in patients with SCI. Materials and Methods: The review was prospectively registered with PROSPERO (CRD42023477882) and conducted following PRISMA 2020 guidelines. MEDLINE and Embase were searched. Studies of animal models of traumatic or non-traumatic spinal cord injury evaluating the effect of HDAC inhibitors on neurobehavioural outcomes were eligible for inclusion. The SYRCLE checklist was used to assess for risk of bias. Screening, data-extraction and risk of bias assessments were completed in duplicate. Results: Of 10,549 studies identified by database searching, 42 studies met inclusion criteria. Animals used were rats (n = 28), mice (n = 13) and rabbits (n = 1). SCI models included spinal cord contusion (n = 24), epidural compression (n = 2), compression with a vascular clip (n = 6), hemisection (n = 5), ischaemia/reperfusion injury (n = 4) and dorsolateral funiculus crush (n = 1). Valproate was the most frequently used HDAC inhibitor (n = 20). Other HDAC inhibitors included 4-phenylbutyric acid (4-PBA; n = 7), RGFP966 (n = 3). trichostatin A, tubastatin A, entinostat, PCI-34051, scriptaid, CI-994, TMP269, vorinostat, 3-TYP, SW-100, ACY1215 were each evaluated in a single study. Three studies used Sirt1 inhibitor EX527 administered simultaneously with an activator molecule: melatonin (n = 1), oxymatrine (n = 1) and MLN4924 (n = 1). Pain and anxiety were evaluated in one study, in which significant improvement was demonstrated. Locomotor function was assessed in 98% (41/42) of studies; The direction of effect demonstrated improvement in locomotor outcome in 73% (30/41) of studies, deterioration in 5% (2/41) of studies and no effect or conflicting findings in 22% (9/41) of studies. Conclusion: HDAC inhibitors are associated with functional motor recovery and improved anxiety and pain scores in preclinical models of SCI. However, the results should be interpreted with caution as risk of bias of included studies was unclear. These results support investigation of HDAC inhibitors as targets for translational research and clinical trials in patients with spinal cord injury.

ID: AGS26/1586

A107: Reprogramming the immune microenvironment with an antimicrobial peptide-loaded nano-hydroxyapatite/collagen hydrogel for enhanced bone regeneration in osteomyelitis

Jinzheng Liang 1 , Liang Wang 1 , Wensheng Zhang 1

1 Department of Orthopedics, The third Affiliated Hospital, Southern Medical University, Academy of Orthopaedics, Guangdong Province, Guangdong Provincial Key Laboratory of Bone and Joint Degeneration Diseases, Guangzhou, China

Introduction: Osteomyelitis is a refractory bone infection characterized by progressive bone destruction, sequestrum formation, and high recurrence rates. Current treatments are often limited by bacterial biofilm formation, antibiotic resistance, and insufficient bone regeneration. This study aimed to develop a collagen-based hydrogel co-loaded with an antimicrobial peptide (AMP) and nano-hydroxyapatite (nHA) (Col-AMP/nHA) to achieve integrated antibacterial, immunomodulatory, and osteogenic effects, and to investigate its underlying mechanisms. Material and Methods: The pro-inflammatory effects of Staphylococcus aureus and its component peptidoglycan (PGN) on primary bone marrow-derived macrophages (BMDMs) and rat tibiae were first evaluated. A cationic hydrogel system was constructed by conjugating a positively charged AMP into a collagen hydrogel and incorporating nHA, enabling the capture of anionic pathogen-associated molecular patterns (PAMPs, e.g., PGN) and damage-associated molecular patterns (DAMPs, e.g., free DNA). The antibacterial and anti-biofilm efficacy of the hydrogel against S. aureus was assessed. Using RAW264.7 cells, HEK-Blue™ hTLR2 cells, and primary BMDMs, the regulatory effects of the hydrogel on macrophage polarization and osteoclast differentiation under PGN stimulation were investigated via ELISA, immunofluorescence, and single-cell RNA sequencing. Its osteoinductive capacity was evaluated in a co-culture system with rat bone marrow mesenchymal stem cells (BMSCs). Finally, a rat tibial osteomyelitis model infected with S. aureus was established, and the hydrogel was implanted after debridement. Therapeutic outcomes were comprehensively assessed through histological, radiological, and molecular analyses. Results: PGN derived from S. aureus significantly induced M1 macrophage polarization and pro-inflammatory responses. The Col-AMP/nHA hydrogel exhibited broad-spectrum antibacterial activity and potent anti-biofilm effects. It effectively neutralized PGN, inhibited TLR2/NF-κB pathway activation, reversed M1 polarization in RAW264.7 cells, and suppressed excessive osteoclast differentiation. Single-cell transcriptomics further confirmed that the hydrogel cleared PGN and blocked NF-κB signaling. Moreover, it markedly enhanced the expression of osteogenesis-related genes and mineralized nodule formation in BMSCs. In vivo, treatment with Col-AMP/nHA effectively cleared the infection, attenuated local inflammation, and significantly promoted new bone formation in the defect areas. Conclusion: This study successfully developed a multifunctional collagen-based hydrogel system that integrates the antibacterial and immunomodulatory functions of AMP with the osteogenic properties of collagen and nHA, enabling multi-target regulation of the osteomyelitic microenvironment. By clearing PAMPs/DAMPs to mitigate excessive inflammation and promoting osteogenic differentiation to accelerate bone regeneration, this strategy presents a transformative potential for the clinical treatment of drug-resistant osteomyelitis.

ID: AGS26/946

A108: A clinico-molecular classification scheme for ossification of ligamentum flavum: based on 1034 cases

Longjie Wang 1 , Shuo Tian 1 , Jialiang Lin 1 , Chuiguo Sun 1 , Weishi Li 1

1 Peking University Third Hospital, Beijing, China

Introduction: Thoracic ossification of the ligamentum flavum (TOLF) is characterized by ectopic bone formation within the ligamentum flavum and often leads to severe myelopathy. Its etiology remains unclear, hindering targeted management. This study aimed to establish a novel classification system integrating clinical, radiological, and molecular characteristics to elucidate its underlying pathogenesis. Material and Methods: We conducted a retrospective analysis of 1,034 surgically treated TOLF patients at our institution from January 2010 to December 2023. The study incorporated four clinical dimensions: consensus clustering (n = 76), epidemiological distribution (n = 1034), imaging characteristics (n = 399), and clinical feature assessment (n = 586). Molecular analysis was performed on surgical specimens from 11 representative cases. Results: Our analysis revealed two distinct TOLF subtypes with characteristic profiles: (1) The degenerative type (75% of cases) primarily affected elderly patients (mean age 60 years) with normal body mass index (BMI), with localized involvement (≤ 3 levels) in the upper/lower thoracic spine. This subtype demonstrated significant biomechanical alterations, including changes in C2-7 SVA and LL, as supported by finite element analysis showing stress changes in the ligamentum flavum. Molecular analysis revealed the activation of wound-healing, ossification, and inflammatory pathways. (2) The metabotropic type (25% of cases) predominantly affected younger patients with abdominal obesity and metabolic syndrome, who exhibited multilevel involvement with preserved spinal curvature but elevated blood lipid levels and distinct dysregulation of lipid/energy metabolism pathways. Conclusion: A novel TOLF classification system consisting of two subtypes, namely, degenerative and metabotropic, each with distinct clinical, biomechanical, and molecular profiles, was developed. The degenerative subtype is primarily associated with mechanical stress and the ageing process, whereas the metabotropic subtype is linked to systemic metabolic dysfunction. This framework provides valuable insights into the diverse pathogenesis of TOLF and could inform future research and personalized treatment strategies.

OP13: Imaging That Predicts

ID: AGS26/1270

A109: Comparative analysis of MRI-based VBQ Score and CT-based HU with DEXA-Based T-Score in assessing BMD in degenerative spine

Ajoy P Shetty 1 , Karthik Ramachandran 1 , Arul kumaran N 1 , S. Rajasekaran 1

1 Ganga Medical Centre and Hospital, Coimbatore, India

Introduction: In spine surgery, accurate assessment of bone mineral density (BMD) is critical for surgical planning and postoperative outcomes. Although DEXA remains the standard method for assessing osteoporosis, its accuracy is reduced in degenerative spinal conditions due to artefactual interference. This study examines the diagnostic performance of CT-derived Hounsfield Units (HU) and MRI-based Vertebral Bone Quality (VBQ) scores by comparing them with dual-energy X-ray absorptiometry (DEXA) T-scores in patients with degenerative spine. Material and Methods: This observational study included patients who underwent lumbar spine MRI, CT, and DEXA scans within a three-month interval. VBQ scores were calculated from T1-weighted sagittal MRI images using vertebral and cerebrospinal fluid signal intensities. CT HU values were measured from L1-L4 vertebrae in axial, sagittal, and coronal planes. DEXA T-scores of the spine and femur were used as reference standards. Results: In a cohort of 104 patients (mean age 59.9 ± 11.9 years; M: F = 46:58), CT HU values showed strong positive correlations (r = 0.745 spine, 0.680 femur), while VBQ scores showed a negative correlation with DEXA T-scores (r = -0.727 spine, -0.783 femur). Optimal thresholds for osteopenia were a HU value of ≤ 163 and a VBQ score of ≥ 2.2, and for osteoporosis, a HU value of ≤ 105 and a VBQ score of ≥ 2.6. HU values and VBQ were inversely correlated (r = -0.869). Among 25 patients with OVCF, DEXA misclassified 8 cases, whereas HU and VBQ correctly identified 21 and 23 patients, respectively, as osteoporotic. Conclusion: VBQ and CT HU values are reliable surrogate markers for BMD assessment in the degenerative spine. These modalities offer practical alternatives to DEXA, particularly when DEXA is unavailable or confounded, and can enhance opportunistic screening and preoperative planning.

ID: AGS26/1426

A110: Utility of intraoperative ultrasound in the surgical treatment of spine trauma injuries

Ashraf El Naga 1 , Anthony DiGiorgio 1 , Sama Zazai 1 , Nesa Milan 1 , Dana Hazem 1 , Michael Huang 1 , Phiroz Tarapore 1 , Geoffrey Manley 1 , Mark Xu 1 , David Gendelberg 1

1 University of California, San Francisco, San Francisco, United States

Introduction: Intraoperative ultrasound (IOUS) has been increasingly described in spine trauma with much of the focus being on prognostication of spinal cord injury. IOUS can be a valuable adjunct in spine surgery, offering real-time visualization of the neural elements withing the dura. Despite its potential, its adoption remains limited, and evidence supporting its routine use is still evolving. This study aims to evaluate the clinical utility of IOUS in spine surgery, focusing on its role in intraoperative decompression assessment and decision-making. Material and Methods: A retrospective review of the operative notes of all patients who underwent acute surgery for traumatic spine etiologies was performed over a 5-year period. Patient and injury demographics were collected. The operative notes were reviewed to identify cases in which intraoperative ultrasound was used and whether any change to the operative course was undertaken based on the ultrasound findings. Descriptive statistics and Chi-squared tests were employed. Results: 525 patients who underwent surgery for acute spine trauma were identified between January 2020 and December 2024. Based on the operative report, intraoperative ultrasound was documented in 143 patients (113 cervical injuries, 15 thoracic injuries, and 15 lumbar injuries). Laminectomies were performed in all these cases (136 for symptomatic spinal cord or conus level compression, 4 for acute cauda equina syndrome, and 3 cases with nerve root compression present). Ultrasound confirmed decompression of the neural elements along the length of decompression in 120 (83.9%) cases. Intraoperative ultrasound findings altered treatment in 23 (16.1%) cases: The laminectomy was extended proximally or distally in 16 (11.2%) of cases. Expansile duraplasty was undertaken in 4 (2.8%) patients. A transpedicular decompression was performed based on ultrasound in one patient with a burst fracture and the neck was repositioned to generate more lordosis to address ventral compression in one cervical case. There was no noted difference in the proportion ultrasound guided intervention based on underlying spine region or the presence of bony injury (p = 0.784, 0.243 respectively). Conclusion: Intraoperative ultrasound is a useful intraoperative modality particularly in confirming decompression or providing actionable information regarding the adequacy of posterior based decompression. Given its low cost and relative ease of use, surgeons may find this modality helpful in intraoperative assessment of posterior based decompressions, particularly in spine trauma.

ID: AGS26/2499

A111: Comparative analysis of DXA, vBMD, CT-HU, and VBQ score as independent risk factors and correlates of cage subsidence following LLIF

Hanming Bian 1 , Qiang Yang 1,2 , Chao Chen 1 , Dong Zhao 1 , Gang Liu 1

1 Capital Medical University Affiliated Beijing Friendship Hospital, Tianjin Hospital, Tianjin, China

Introduction: While it is established that cage subsidence following LLIF is significantly associated with vertebral bone mineral density, there remains a lack of predictive models based on preoperative CT data to accurately forecast individualized subsidence risk. Therefore, this study aims to investigate whether four bone mineral density (BMD) measurement methods-DXA T-score, vertebral BMD (vBMD), CT Hounsfield unit (HU) value, and Vertebral Bone Quality (VBQ) score-can serve as independent risk factors for cage subsidence after Lateral Lumbar Interbody Fusion (LLIF), and to compare their correlations with the extent of cage subsidence. Material and Methods: A retrospective case analysis was conducted to explore the relationship between the four BMD measurement methods and cage subsidence after LLIF. Patients who underwent single-level LLIF for lumbar degenerative diseases between January 2019 and December 2023 were included. DXA T-scores were obtained from medical records, while lumbar CT-HU values, vBMD, and VBQ scores were measured based on preoperative CT and MRI scans. Cage subsidence was assessed using X-rays at 12 months postoperatively, and the cage subsidence distance (CSD) was measured. Logistic regression analysis was used to identify risk factors for cage subsidence. Pearson correlation analysis was employed to examine the correlation between identified risk factors and CSD, and Williams' test was used to compare the statistical differences in correlation coefficients. Results: According to the inclusion/exclusion criteria, 94 patients who underwent single-level LLIF were included. Among them, 61 were female (64.9%), with a mean age of 63.2 ± 8.1 years. Based on the presence of cage subsidence at 12 months postoperatively, patients were divided into two groups: the subsidence group (34 cases, mean CSD 3.22 ± 0.61 mm) and the non-subsidence group (60 cases, mean CSD 1.07 ± 0.73 mm). There were no significant differences in demographic data or preoperative disc height (DH), lumbar lordosis (LL), or segmental lordosis (SL) between the two groups (p > 0.05). The subsidence group had significantly lower DXA T-scores (-1.87 ± 0.43 vs. -0.66 ± 0.31, p < 0.05), CT HU values (88.37 ± 22.85 vs. 168.44 ± 20.71, p < 0.05), and vBMD values (74.48 ± 10.75 vs. 127.61 ± 13.12, p < 0.05), and a significantly higher VBQ score (4.61 ± 1.08 vs. 3.77 ± 0.96, p < 0.05) compared to the non-subsidence group. Logistic regression analysis identified CT HU value (OR = 0.902, p = 0.003), vBMD (OR = 0.961, p < 0.001), and VBQ score (OR = 1.328, p = 0.037) as independent risk factors for cage subsidence after LLIF. Pearson correlation analysis showed that CT HU value (r = -0.579, p < 0.001) and vBMD (r = -0.703, p < 0.001) were negatively correlated with CSD, while VBQ score was not significantly correlated with CSD (r = 0.293, p = 0.086). Williams' test indicated that the correlation coefficient between vBMD and CSD was significantly greater than that between CT-HU and CSD (p = 0.039). Conclusion: vBMD, CT HU value, and VBQ score can serve as independent risk factors for cage subsidence, but vBMD demonstrates a stronger correlation with the extent of subsidence. Therefore, vBMD is selected as the preferred B parameter for predicting cage subsidence and will be used as the BMD evaluation parameter in subsequent methods for calculating vertebral mechanical properties based on BMD.

ID: AGS26/2547

A112: Body roundness matters? Two-year lumbar fusion outcomes stratified by BRI (Body Roundness Index)

Arvind G Kulkarni 1

1 Mumbai Spine Scoliosis & Disc Replacement Center, Bombay Hospital, Breach Candy Hospital, Mumbai, India

Introduction: Body Roundness Index (BRI) is a novel anthropometric measure reflecting central obesity and visceral adiposity, which may influence surgical outcomes. This study evaluated the impact of BRI on functional recovery in patients undergoing single level MIS-TLIF (Minimally Invasive - TLIF). Material and Methods: Consecutively operated single level MIS-TLIF patients between January 2020-January 2023 were stratified into two groups based on sex-specific BRI cut-offs: Group 1 (high BRI, n = 99; men > 3.965, women > 3.757) and Group 2 (low BRI, n = 166; men ≤ 3.965, women ≤ 3.757). Groups were matched for age and sex. Functional outcomes were assessed preoperatively and at 2-year follow-up using the Oswestry Disability Index (ODI) and Visual Analog Scale (VAS) for leg pain. Results: A total of 290 patients underwent single level MIS-TLIF in the said period in the age group of 35 to 93 years. Fifteen patients had incomplete follow-up data. At 2 years, both groups demonstrated significant improvements in disability and pain, with no intergroup statistical significance: Group 1 (high BRI): mean ODI improvement 15.2 points, mean VAS improvement by 3.8 points; BRI 5.2 ± 0.9. Group 2 (low BRI): mean ODI improvement 15.6 points, mean VAS improvement 4.0 points; BRI 2.8 ± 0.6 (p > 0.05). Patients with degenerative spondylolisthesis showed greater ODI improvement than those with lytic spondylolisthesis (p = 0.02), while VAS outcomes were similar (p = 0.251). Conclusion: Preoperative BRI, based on sex-specific cut-offs, did not significantly affect long-term functional outcomes after single level MIS-TLIF. Both high and low BRI groups achieved meaningful disability improvement, suggesting that central obesity, as measured by BRI, may not compromise recovery.

ID: AGS26/2708

A113: How low can we go? Can we further reduce radiation exposure during scoliosis radiographic examination?

Rachelle Imbeault 1 , Marjolaine Roy-Beaudry 1 , Sylvain Deschenes 1 , Justin Dufresne 1 , Stefan Parent 1

1 CHU Sainte-Justine, Montreal, Canada

Introduction: From diagnosis, pediatric patients with idiopathic scoliosis require frequent radiological evaluations. Due to their longer life expectancy and greater radiosensitivity, technologies must minimize radiation dose while optimizing image quality. This study aims to compare two biplane radiography systems, one using photon counting and dose modulation, the other using energy integration and fixed dose throughout acquisition. Material and Methods: Fifty patients were recruited between June 2023 and July 2024. All patients had a diagnosis of idiopathic scoliosis, with an average Cobb angle of 29.7° ± 15.0° and a mean age of 13.7 years. Each patient underwent two successive biplanar examinations (PA and lateral views), one using the dose modulating system and the other using a fixed dose system. Entrance skin dose was measured for 30 patients, using optically stimulated luminescence dosimeters located near seven radiosensitive anatomic areas. Acquired X-rays were randomized and evaluated by one pediatric orthopedic surgeon and one pediatric radiologist. Statistical analysis was performed using Wilcoxon signed-rank tests. Results: Dose analysis on 30 patients allowed to observe an average dose reduction of 48% (range: -26% to 71%) when dose modulation was used. Dose distribution shows a 63% reduction in the cervicothoracic region, while the reduction was 29% in the lumbosacral region. X-ray analysis revealed that while decreasing the dose, dose modulation also improves global image quality (p-value ≤ 0.001). Furthermore, analysis of specific anatomical structures showed that visibility is improved for more than 84% of the anatomical landmarks being scored (p-value ≤ 0.05). Conclusion: Results indicate that dose modulation can reduce patient exposure to radiation by 48%, while improving global image quality. The greater reduction in the cervicothoracic region suggests better protection of the radiosensitive organs located in this area.

ID: AGS26/2742

A114: Vascular window at the surgical corridor for L5-S1 ALIF according to pelvic incidence

Alexandry Carvalho 1 , Cristiano Menezes 2 , Juan Cárdenas 2 , Alfredo Guiroy 3 , Alberto Gotfryd 4 , Jerônimo Milano 5 , Cristian Correa Valencia 7 , Gabriel Lacerda 6 , Facundo Van Isseldyk 8 , Matias Pereira Duarte 9 , Martin Gagliardi 10 , Jefferson Soares Leal 11

1 Clínica Ortomed, Teresina - Piauí, Brazil,

2 Federal University of Minas Gerais (UFMG), Columna Institute, Belo Horizonte, Brazil,

3 Clínica de Cuyo, Mendoza, Argentina,

4 Santa Casa de São Paulo, Medical School and Hospitals, São Paulo, Brazil,

5 Neurological Institute of Curitiba, Curitiba, Brazil,

6 Hospital Felicio Rocho, Belo Horizonte, Brazil,

7 Clinica Las Condes, Santiago, Chile,

8 Hospital Privado de Rosario, Rosario, Argentina,

9 Clínica La Pequeña Familia, Junin, Argentina,

10 Northern Ontario School of Medicine University, Thunder Bay, Ontario, Canada,

11 Federal University of Minas Gerais, Belo Horizonte, Brazil

Introduction: Anterior lumbar interbody fusion (ALIF) at the L5-S1 level is a widely accepted technique for addressing degenerative, deformity, and instability conditions of the lumbosacral junction. Its advantages include direct disc space access, restoration of disc height, and favorable sagittal alignment correction. However, in a subset of the population where the confluence of the common iliac veins is located more distally, mobilization of the left iliac vein may be technically demanding. Spinopelvic parameters, particularly pelvic incidence (PI), are key determinants of sagittal spinal alignment and pelvic morphology. PI, a fixed morphological parameter in adults, influences the orientation of the sacrum and, consequently, the spatial relationship between the sacral plateau and surrounding vascular structures, potentially altering the position of the iliac vessels relative to the L5-S1 disc. The present study aimed to investigate whether pelvic incidence values are associated with differences in vascular window dimensions. Material and Methods: A multicenter, retrospective, observational cohort study was conducted from January 2020 to August 2025. Adult patients (> 18 years of age) who underwent standing whole-spine radiographs and lumbosacral magnetic resonance imaging were included. Patients with prior lumbosacral/abdominal surgery, spinal trauma, tumor, or infection were excluded. Demographic data such as age, sex and ethnicity were assessed. In addition, the main parameters measured were pelvic incidence and vein-free vascular window, defined as the ratio of the interiliac distance (between the right and left iliac veins) to the total L5-S1 disc space, expressed as a percentage. The patients were divided into 3 groups: group 1: PI < 45°, group 2: PI 45°-60° and group 3: PI > 60°. Results: Seventy-nine patients were included so far, being 60,7% female, with a mean age of 57 years/old. The main indications for imaging examinations in the patients included in the study were: low back pain (48.1%), sciatica (15.1%) and low back pain associated with radicular pain (12.6%). Among the main associated diseases were: degenerative disc disease (39.7%), spondylolisthesis (14.2%), degenerative scoliosis (10.2%), lumbar stenosis (8.9%). In relation to the parameters evaluated, the mean PI was 53.8° +- 12.4° (51.1° in men and 55.5° in women). The most frequent group according to PI was group 2, corresponding to 41.7% (n33), followed by group 3 31,6% (n25) and group 1 26,5% (n21). The mean value of the vein-free vascular window according to each group was: 54.74% (group 1), 55.92% (group 2) and 64.65% (group 3), however the relationship between high/low PI and the vascular window was not statistically significant (p = 0.247). Conclusion: Preliminary results suggest that patients with a high pelvic incidence (> 60°) tend to have a larger average vein-free vascular window. However, further study with a larger cohort and a more even distribution among the three groups is needed to validate these observations and elucidate the possible association between pelvic incidence and vascular anatomy.

ID: AGS26/2791

A115: The predictive value of multifidus degeneration in osteoporotic vertebral compression fracture (OVCF) patients with kyphosis deformity

Junyu Li 1 , Zimo Wang 1 , Gengyu Han 1 , Hanwen Zhang 2 , Zhuo Ran Sun 1 , Yongqiang Wang 1 , Miao Yu 1 , Weishi Li 1 , Lin Zeng 3 , Yan Zeng 1

1 Department of Orthopaedics, Peking University Third Hospital, Beijing Key Laboratory of Spinal Disease Research, Engineering Research Center of Bone and Joint Precision Medicine, Beijing, China ,

2 Peking University Health Science Center, Beijing, China ,

3 Clinical Epidemiology Research Center, Peking University Third Hospital, Beijing, China

Introduction: Osteoporotic vertebral compression fracture (OVCF) causes pain, kyphosis and neurological damage, which significantly affect patients’ quality of life. Patients with OVCF are often elderly and have severe osteoporosis, which makes preoperative symptom more serious, postoperative recovery worse and the incidence of postoperative complications high. The paraspinal muscles have been well studied in adult spinal deformities, but there is no conclusive evidence that their findings can be applied to OVCF. The purpose of this study was to evaluate the associations between multifidus (MF) parameters including fat infiltration (FI), relative functional sectional area (rFCSA), relative gross cross-sectional area (rGCSA) and the sagittal parameters, symptom score, and postoperative complications. Material and Methods: The study included 108 OVCF patients with kyphosis deformity who underwent corrective surgery and were followed for 2 years. MRI were performed preoperatively to evaluate the paraspinal muscle morphology, including MF fat infiltration (MFFI), relative functional cross-sectional area (MFrFCSA), and MF relative gross cross-sectional area (MFrGCSA). VAS, ODI, JOA, and SRS-22 were conducted preoperatively. Preoperative, postoperative and last-follow up spine sagittal parameters were recorded, as well as sagittal balance, loss of correction results and improvement and deterioration of sagittal parameters. The occurrence of postoperative mechanical complications, including adjacent segment disease, screw loosening, proximal junctional kyphosis, and distal junctional problem were recorded. We analyzed the relationship between MF degeneration and the above parameters. Results: Strong correlation was observed in VAS and MFFI (rr = 0.597, p = .000),MF rFCSA (rr = -0.520, p = .001) and MF rGCSA (rr = -0.461, p = .005), as well as ODI and MF rFCSA (rr = 0.336, p = .042). Preoperatively, strong correlations were observed between MF rFCSA and LL (rr = -0.320, p = .010), TLK (rr = -0.271, p = .026), TK (rr = -0.251, p = .048). MF rGCSA and LL (rr = 0.259, p = .039), TLK (rr = -0.247, p = .043), TK (rr = -0.273, p = .030), GK (rr = -0.381, p = .002) were also strongly correlated. Our study showed strong correlations between MF FI and TLK loss (rr = 0.406, p = .003), TK loss (rr = 0.332, p = .045);MF rGCSA and SVA loss (rr = -0.367, p = .050), TPA loss (rr = -0.404, p = .030); MF rGCSA and TPA loss (rr = -0.401, p = .031), MF FI and GK loss (rr = 0.397, p = .027). MF FI was significantly higher in the complication-presence group (p = .045). Conclusion: Multifidus degeneration is significantly associated with QoL, sagittal parameters and mechanical complications in OVCF patients with kyphosis deformity. The pathological changes of paravertebral muscles should be included in the surgical strategy and postoperative paravertebral muscle rehabilitation should be adopted to improve the clinical outcomes of OVCF patients.

ID: AGS26/3477

A116: Preoperative paraspinal muscle quality is an independent predictor of prolonged hospitalization after surgical intervention for lumbar spine trauma

Nathan McLaughlin 1 , Zachary Lowe 1 , Advika Srinivas 1 , Amir Fanaei 1 , Mauricio Avila 1

1 Saint Louis University, St. Louis, United States

Introduction: Paraspinal muscle atrophy is a known predictor of poor outcomes in degenerative spine disease, but its prognostic role in acute spinal trauma is not well characterized. We hypothesized that poor preoperative paraspinal muscle quality, characterized by high fatty infiltration, is associated with worse short-term outcomes in patients undergoing surgery for acute traumatic lumbar spine injuries. Methods: A retrospective cohort study was conducted on adult patients who underwent surgical intervention for acute lumbar trauma in a regional level 1 trauma. Preoperative T1-weighted MRI was used to measure paraspinal cross-sectional area (CSA) of erector spinae (ES) and multifidus (MF) at L3-L4 and L4-L5 levels. Fatty infiltration (FI) of paraspinal muscles was graded on a 0-5 scale and analyzed across four groups (Goutallier classification, 0 being normal muscle) (0, 1, 2, and 3+). Primary outcomes were total hospital length of stay (LOS) and 90-day complications. Associations were assessed using Kruskal-Wallis, Mann-Whitney U, and Chi-square tests. Multivariable linear regression was used to determine if FI was an independent predictor of log-transformed LOS after controlling for patient age. Results: A total of 83 eligible patients were included. A significant association was found between FI grade and hospital LOS (Kruskal-Wallis, p = 0.005). Median LOS was similar for patients with FI grades 0, 1, and 2 (7.2, 7.0, and 7.0 days, respectively), but more than doubled for patients with grade 3+ FI (17.4 days). In multivariable regression, FI remained independently associated with longer LOS after adjusting for age (β = 0.186, p = 0.047), corresponding to an approximate 20% increase in LOS per grade increase. In contrast, muscle CSA was not significantly associated with LOS (p > 0.05). Neither FI grade nor CSA was significantly associated with 90-day complication rates or a post-operative loss of independent ambulation (p > 0.05). Conclusion: Preoperative paraspinal muscle quality, as measured by fatty infiltration, is a significant predictor of prolonged hospitalization following surgery for acute lumbar spinal trauma. This association remains independent of patient age. Muscle quality appears more prognostic for short-term recovery than overall muscle size, offering a valuable MRI-based metric for preoperative risk stratification and resource allocation after surgery.

ID: AGS26/42

A117: Evaluating the prognostic role of computed tomography hounsfield units in anticipating spinal outcomes post-instrumentation: a systematic review and meta-analysis

Chinedu Egu 1 , Elie Najjar 1 , Spyridon Komaitis 1 , Edidiong Essiet 1 , Samuel Akintunde 1 , Vusumuzi Sibanda 1 , Khalid Salem 1 , Opinder Sahota 1 , Nasir Quraishi 1

1 Centre for Spinal Studies and Surgery, Queens Medical Centre, Nottingham, United Kingdom

Introduction: The use of Hounsfield unit (HU) measurements derived from preoperative computed tomography (CT) scans has emerged as a promising surrogate for assessing bone mineral density (BMD). This systematic review aims to elucidate the role of HU in predicting spinal outcomes in patients undergoing spinal instrumentation. Material and Methods: A comprehensive systematic review of the English-language literature was performed across multiple databases, focusing specifically on adult patients who underwent spinal instrumentation. Adhering to PRISMA guidelines, studies published between 2017 and 2024 that investigated the correlation between HU measurements and spinal outcomes were included. Data extraction and subsequent meta-analysis were conducted utilizing a random effects model. The methodologi cal quality of the included studies was appraised using the Newcastle-Ottawa Scale, and statistical analyses were executed with R software. Results: A total of thirty-five retrospective studies met the inclusion criteria, encompassing a cohort of 3,927 patients. The reported rates of complications included cage subsidence at 35.5%, pedicle screw loosening at 27.9%, proximal junctional failure at 28.6%, and pseudoarthrosis at 66.7%. Notably, lower HU values were significantly associated with an increased incidence of complications across all studies. The delineated HU cutoff thresholds for predicting specific complications were as follows: 198 HU for lumbar interbody fusion cage subsidence; 116 HU and 126 HU for pedicle screw loosening in lumbar and thoracic spine fusions, respectively; 151 HU for proximal junctional failure in thoracolumbar spinal fusion; and 240-260 HU for pseudoarthrosis following anterior odontoid screw fixation. HU measurements were typically calculated as the mean HU of the vertebrae in the operational segments: cervical (C2-C7), thoracic (T1-T12), lumbar (L1-L4), or sacral (S1). Conclusion: This meta-analysis substantiates a significant correlation between reduced HU values and spinal complications following instrumentation. Specifically, thresholds of below 130 HU for thoracic and lumbar pedicle screw loosening, below 200 HU for lumbar cage subsidence, approximately 150 HU for proximal junctional failure in the thoracolumbar region, and below 260 HU for pseudoarthrosis subsequent to anterior odontoid screw fixation were identified. The mean vertebral HU of the surgical segments serves as a reliable metric for this assessment.

OP14: Personalizing Degenerative Care

ID: AGS26/1023

A118: Sustainability in lumbar decompression surgery - Small tweeks leading to greater wins

Tahir Khaleeq 1 , Siddhant Kapoor 2 , Shahnawaz Haleem 2

1 Queen Elizabeth Hospital, Birmingham, United Kingdom , 2 Royal Orthopaedic Hospital, Birmingham, United Kingdom

Background: Healthcare contributes ∼5% of global CO 2 emissions, with operating theatres among the most energy- and waste-intensive hospital areas. Spinal surgery is particularly resource-heavy, due to high implant use, complex technology, long procedures, reliance on single-use items, and transport of heavy trays from central sterile services. A recent study estimated that a single lumbar microdiscectomy generates ∼478 kg CO 2 e. This evaluation sought to quantify inefficiencies in routine spinal surgery and identify opportunities to reduce cost, waste, and environmental impact. Aim: To assess the utilisation of instruments and single-use items in lumbar decompression surgery, and to quantify associated financial and carbon costs. Methods: A prospective service evaluation was undertaken at a tertiary spinal centre, analysing 20 consecutive lumbar decompressions. For each case, the standard spinal tray list was cross-referenced with instruments actually used, as recorded by the scrub team. Instruments consistently unused across all cases were identified. Data were also collected on single-use consumables, implants, and waste. Procurement and sterile services provided sterilisation costs, and carbon emissions were estimated using published values (66-77 g CO 2 e per instrument per cycle). Findings were corroborated against theatre logs and operative notes. Results: Of 46 instruments in the Spinal Extras tray, approximately 40% were never used across any of the 20 procedures. These could be safely removed from the tray, reclassifying it from Category G (41-50 items; £59.30 per cycle) to Category F (31-40 items; £48.83 per cycle). This reduction equates to a saving of £20.47 per case. Based on 553 lumbar decompressions performed in 2024-25, annual savings would be at least £10,789, with potential for greater reductions if all consistently unused instruments were removed. From an environmental perspective, removing 6-10 unused instruments would reduce 770-1400 g CO 2 e per case, equivalent to 823 kg CO 2 e annually. Additional unquantified waste was noted from routinely opened but unused single-use drapes and consumables, further increasing the environmental burden. Conclusion: This study highlights the hidden financial and carbon cost of unnecessary instruments and consumables in spinal surgery. Tray rationalisation alone could save >£10,700 and >400 kg CO 2 e annually at a single centre, with greater benefits possible if consistently unused instruments are removed. In addition, substantial waste was observed from single-use items, underlining the need for broader strategies targeting consumables, implants, and sterilisation processes. Tray optimisation represents an achievable “quick win” that complements wider sustainability initiatives and supports alignment with NHS Net Zero targets.

ID: AGS26/1101

A119: Improved outcomes using a novel low-modulus 3D-printed porous titanium alloy cage (Osteo Match) versus PEEK cage in osteoporotic patients undergoing lumbar interbody fusion: an early-stage follow-up with CT scans

Weishi Li 1

1 Peking University Third Hospital, Beijing, China

Introduction: The cage subsidence and failure of fusion occur frequently, due to diminished bone density and compromised mechanical strength of osteoporotic spine. The elastic modulus of current cage structure remains higher than the vertebral bone modulus in osteoporotic patients. The modulus mismatch between the cage and host bone predisposes to localized stress concentration. We firstly investigated the clinical use of a novel low-modulus 3D-printed porous Ti-alloy cage named Osteo Match, which has elastic modulus of < 1GPa matching the bony elastic modulus of osteoporotic patients. This study demonstrates that Osteo Match cage is a promising alternative for enhancing surgical outcomes in osteoporotic patients undergoing lumbar fusion. Material and Methods: We selected the patients undergoing short-segment (≤ 2 levels) lumbar fusion from the database of a prospective cohort study registered in ClinicalTrials.gov ( NCT05981222 ). The general information including age, gender, body mass index (BMI, kg/m2), bone mineral density (BMD), operation time, blood loss, cage type (PEEK or Osteo Match), cage height. Cage subsidence of over 2mm in depth and the fusion status were assessed at 6 months follow-up with CT scans. VAS, ODI, and JOA were recorded for analyzing clinical outcomes. Results: A total of 103 patients were followed-up at 6 months after the surgery. All of the postoperative clinical scores were significantly improved compared with the preoperative scores (p < 0.001). No revision surgery was required and no fatigue cracking of Osteo Match cage was observed. The ratio of female was higher in patients using PEEK cages, but other baseline factors including BMD measured in T-score and CT-HU were comparable between the two types of patients. There were 139 fusion segments in total, including 42 segments in the Ti-alloy group, and 97 segments in the PEEK-group. At 6 months follow-up, the cage subsidence rate was significantly lower in Ti-alloy group than in PEEK group (21.4% vs. 63.9%, p < 0.001). The use of Osteo Match Ti-alloy cage (OR 0.321; 95%CI 0.126-0.817, p = 0.017) was the independent protective factor of occurrence of cage subsidence. The Ti-alloy group achieved higher fusion rate than the PEEK-group (64.3% vs. 44.3%, p = 0.031). Conclusion: Compared with PEEK cages, the novel low-modulus 3D-printed porous Ti-alloy cage has advantages in reducing subsidence risk and promoting interbody fusion in osteoporotic patients. Multicenter study with longer follow-up duration is needed for further investigation.

ID: AGS26/1384

A120: Effectiveness of tranexamic acid in elective lumbar spine surgery cases: a single-center ambispective investigation

Constantinos Chaniotakis 1 , Nikolaos Gkantsinikoudis 1 , Mikail Chatzivasiliadis 1 , Stylianos Kapetanakis 1,2

1 Spine Department and Deformities, Interbalkan European Medical Center, Thessaloniki, Greece

2 Academic Orthopaedic Department, Aristotle University of Thessaloniki, Thessaloniki, Greece

Introduction: Intravenous Tranexamic Acid (ITA) represents a pharmacologic measure to reduce intraoperative hemorrhage. Effectiveness of TA in lumbar spine surgery has been reported by several scholars in recent literature. However, high-quality data regarding precise safety-efficacy profile remain scarce. Aim of this study is to present our experience with utilization of TA in elective standard lumbar spine surgery cases, involving Microdiscectomy (MD) and Posterior Instrumented Fusion (PIF) cases for Lumbar Disc Herniation (LDH) and Lumbar Spinal Stenosis (LSS), respectively. Material and Methods: Two hundred and eighty (n = 280) patients were enrolled in this ambispective investigation. Ninety (n 1 = 90) patients represented obese individuals (BMI > 30 kg/m 2 ) treated with MD for single-level LDH, whilst rest one hundred and ninety (n 2 = 190) were managed with decompression-PIS for LSS. MD patients and PIF patients were evaluated retrospectively and prospectively, respectively, resulting in this ambispective evaluation. Patients were separated in two groups according to ITA administration, the ITA group (administration of single dose 10 mg/kg ITA preoperatively, 47 patients in LDH group and 98 patients in PIF group) and control group (no ITA administration, 43 patients in LDH group and 92 patients in PIS group). Studied outcome measures were intraoperative blood volume loss, postoperative Hemoglobin (Hb) reduction, operation duration and hospitalization time. Results: No significant differences in terms of demographic parameters and comorbidities were recorded between two groups. Intraoperative blood loss, postoperative Hb reduction, operation duration and overall hospitalization times were statistically significantly reduced in ITA group in comparison with control group. This difference was evident in overall comparison as well in LDH and PIF groups, distinctly. Conclusion: Our results showed that ITA is significantly effective in reducing intraoperative hemorrhage, operation as well hospitalization duration in obese patients with LDH -who demonstrate inherently greater operation times and blood loss in contrast to their non-obese counterparts- as well as in patients with LSS subjected to PIF. No ITA-associated complications were observed, advocating for its safety, even in elderly individuals with underlying comorbidities.

ID: AGS26/176

A121: Outcomes of health metrics after lumbar decompression and fusion

Jahan Aslami 1 , Praveer Vyas 2 , Daniel Altman 1,2

1 Drexel University College of Medicine, Pittsburgh, United States , 2 Allegheny Health Network, Pittsburgh, United States

Introduction: Lumbar degenerative disease is a prevalent condition that may require surgical intervention. While functional outcomes following lumbar fusion are well-documented, its impact on physical health metrics remains underexplored. The Patient-Reported Outcomes Measurement Information System (PROMIS) provides concise and standardized population-level assessments of physical, mental, and social health. We hypothesized that increased function following lumbar fusion would lead to long-term improvements in health metrics. Material and Methods: We retrospectively reviewed patients who underwent lumbar decompression and fusion at a single institution, excluding those with chronic debilitating conditions (e.g., COPD). Patient demographics and physical health metrics-including BMI, heart rate, blood pressure, A1C, lipid profiles, and PROMIS scores-were collected at one year preoperatively and at 1, 2, and 3 years postoperatively. Paired t-tests were used to compare postoperative values to preoperative baselines, with Benjamini-Hochberg correction applied for multiple comparisons. p < 0.05 denotes significance. Results: A total of 1,336 patients were included over six years. Triglycerides decreased from 149.3 ± 103.8 mg/dL preoperatively to 125.3 ± 55.9 mg/dL at 3 years postoperatively (mean reduction 24.0 mg/dL, p = 0.0016). Total cholesterol decreased from 175.5 ± 37.6 mg/dL to 165.4 ± 42.7 mg/dL at 3 years (mean reduction 10.1 mg/dL, p = 0.0017). A1C declined from 6.61 ± 1.19% to 6.45 ± 0.82% at 3 years (mean reduction 0.16%, p = 0.31), and LDL dropped from 98.6 ± 31.3 mg/dL to 93.5 ± 37.4 mg/dL (mean reduction 5.1 mg/dL, p = 0.20), though the change were not statistically significant. Diastolic blood pressure showed significant improvement, decreasing from 80.0 ± 8.7 mmHg to 77.5 ± 8.7 mmHg at 3 years (mean reduction 2.5 mmHg, p < 0.001). Systolic blood pressure decreased from 133.2 ± 13.9 mmHg to 131.6 ± 15.1 mmHg (mean reduction 1.6 mmHg, p = 0.056), narrowly missing statistical significance. No significant differences were observed in BMI (31.8 ± 5.9 to 31.8 ± 6.2, p = 0.64), HDL (49.8 ± 14.5 to 49.8 ± 16.3, p = 0.98), or heart rate (74.6 ± 13.1 to 72.8 ± 12.1, p = 0.56). At one year, all PROMIS domains showed significant improvement. Anxiety decreased from 51.3 ± 8.9 to 47.1 ± 8.7 (p = 0.009), depression from 50.6 ± 8.0 to 46.0 ± 8.7 (p < 0.001), fatigue from 58.2 ± 8.3 to 53.0 ± 8.4 (p = 0.014), pain interference from 67.8 ± 5.0 to 59.1 ± 9.3 (p < 0.001), and sleep disturbance from 57.8 ± 8.1 to 54.2 ± 8.9 (p = 0.008). Physical function increased from 34.0 ± 4.9 to 39.9 ± 8.9 (p < 0.001), social participation from 38.6 ± 6.1 to 46.5 ± 9.8 (p < 0.001), and upper extremity function from 38.2 ± 6.0 to 45.6 ± 9.1 (p = 0.002). PROMIS data at two and three years were insufficient for further longitudinal analysis. Conclusion: Our findings suggest that lumbar decompression and fusion may contribute to improved physical health metrics. All metrics except for BMI and HDL decreased from baseline. These results suggest improved patients’ physical health. This may be due to increased mobility post-fusion, as well as more regular outpatient follow-up. All PROMIS scores improved significantly at one year, aligning with existing literature. Further, the mean difference in PROMIS scores exceeded the minimal clinically-important differences reported in the literature (usually 2-5). These findings underscore the procedure’s value across multiple domains, including reductions in anxiety, depression, fatigue, and sleep disturbance, as well as improvements in physical function and social participation. Limitations include the inability to control for medications, which may have influenced blood pressure, A1C and lipid values.

ID: AGS26/2814

A122: Sex specific analysis of the prognostic factors associated with reaching minimally clinically important difference one-year post surgery across common degenerative lumbar conditions: a CSORN study

Aazad Abbas 1 , Karlo Pedro 2 , Chris Bailey 3 , Philippe Phan 4 , Jennifer Urquhart 3 , Raja Rampersaud 1,5 , CSORN Investigators

1 Division of Orthopaedic Surgery, University of Toronto, Toronto, Canada

2 Division of Neurosurgery, Toronto Western Hospital, Toronto, Canada

3 Combined Orthopaedic and Neurosurgical Spine Program, London Health Sciences Centre, London, Canada

4 Division of Orthopaedic Surgery, The Ottawa Hospital, Ottawa, Canada

5 Schroeder Arthritis Institute, Division of Orthopaedic Surgery, Toronto Western Hospital, Toronto, Canada

Introduction: Up to one-third of patients experience limited benefit following surgical intervention for common degenerative lumbar conditions. There is significant variability in reported prognostic factors with studies being limited by the number of variables and/or sample size as well as a specific diagnosis. Furthermore, although there is growing recognition that outcomes vary by sex, there has been relatively limited work taking a sex-specific approach to investigating predictors of outcome in this clinical population. As such, the objective of this study is to determine unique prognostic factors associated with patient reported pain related disability outcome one-year following spine surgery in males and females. Material and Methods: Data from the Canadian Spine Outcomes and Research Network (CSORN) database (2015-2023) was analyzed for patients with common degenerative lumbar diagnoses. Multivariate logistic regression models were created to predict minimally clinically important difference (MCID) achievement based on 30% improvement in the Oswestry Disability Index (ODI) at one-year follow-up. Sex stratified analysis was conducted to analysis differences between men and women. Patient diagnosis and a comprehensive list of patient factors (n = 110 variables) were used as inputs. Stepwise selection minimized the Akaike Information Criterion (AIC), with significance set at p < 0.05. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated for each variable. Results: The study included 6,214 patients, with diagnoses of lumbar spinal stenosis-LSS (2,175), degenerative disc-DD (456), disc herniation-LDH (1,279), degenerative spondylolisthesis-DLS (1,591), isthmic spondylolisthesis-IS (361), and degenerative scoliosis-DS (352). The average (SD) age was 60.1 (13.4) with men and women evenly distributed (3,142 (50.6%) and 3,072 (49.4%), respectively). A summary of factors across both cohorts is found in Table 1. Key common predictive factors across both male and female cohorts were age, baseline ODI score, comorbidities, DLS, IS, smoking, unemployed/other work status, those on worker’s compensation, and daily narcotics use. Factors that were only significant in the male cohort were postgraduate education, employed but not working, legal consultation obtained, medication for back pain, medication duration use > 1 year, intermittent narcotics use, and expected improvement in sporting activities. Factors that were only significant in the female cohort were BMI, baseline back and leg pain, and over the counter medication use.When combining both cohorts and adjusting for sex, the factors that were significant but not in each cohort separately with ORs (95% CI) were living with a partner 1.23 (1.06, 1.43) and less than high school education 0.8 (0.66, 0.97). Conclusion: This study outlines prognostic factors related to achieving MCID for pain related disability across common degenerative lumbar conditions one-year following surgery. A number of factors uniquely influenced outcomes for males and females, suggesting that simply adjusting for sex effects in regression analyses may not be sufficient to identify sex differences. This has implications for the interpretation and use of current prognostic evidence; and future development of predictive algorithms for ODI outcome following spine surgery. It also has implications for patient education and decision-making, and suggests that interventions to optimize outcomes may require a sex-specific approach.

ID: AGS26/2958

A123: Impact of preoperative glucagon-like peptide analog therapy on 90-day and 1-year complications following laminectomy in overweight adults: a large multicenter propensity-matched study

Muaz Wahid 1 , Zuhair Zaidi 1 , Asim Mohamed 1 , Salah Aoun 1

1 UT Southwestern Medical Center, Dallas, United States

Introduction: Glucagon-like peptide (GLP) analogs are widely prescribed for obesity and diabetes, yet perioperative effects in spine surgery are incompletely defined. This study represents one of the first large-scale, decompression-only analyses linking preoperative GLP exposure to postoperative outcomes after laminectomy/laminotomy among overweight adults, with effects assessed at 90 days and 1 year. Material and Methods: Adults with overweight or obesity undergoing cervical, thoracic, or lumbar laminectomy/laminotomy were identified in the TriNetX US Collaborative Network (72 health systems; 2010-2024). Prior fusion or bariatric surgery was excluded. Exposure was ≥ 6 months of GLP prescriptions; comparators were non-users. Outcomes (sepsis, acute kidney injury [AKI], deep-vein thrombosis [DVT], myocardial infarction [MI], pneumonia, post-laminectomy syndrome, wound dehiscence, surgical site infection, stroke, and death) were assessed from postoperative day 1 through day 365. Propensity score matching across 33 covariates yielded balanced cohorts (n = 5,680 each). Odds ratios (ORs) with 95% confidence intervals (CIs) were estimated. Results: After matching, 11,360 patients were analyzed (median follow-up, 365 days per group). At 90 days, GLP exposure was associated with lower odds of sepsis (OR 0.38, 95% CI 0.25-0.56), DVT (0.26, 0.16-0.43), post-laminectomy syndrome (0.45, 0.32-0.62), wound dehiscence (0.38, 0.27-0.51), surgical site infection (0.37, 0.28-0.50), and death (0.30, 0.17-0.54; all p < 0.001); associations for AKI, MI, pneumonia, and stroke were not significant. At 1 year, benefits persisted for sepsis (0.52, 0.40-0.68), DVT (0.41, 0.29-0.57), post-laminectomy syndrome (0.45, 0.37-0.55), wound dehiscence (0.47, 0.36-0.62), surgical site infection (0.43, 0.33-0.56), and death (0.48, 0.35-0.65). AKI showed borderline reduction (0.80, 0.65-1.00; p = 0.046); MI, pneumonia, and stroke remained non-significant. Absolute effects at 1 year underscore clinical impact: mortality was 1.04% in GLP users vs 2.15% in non-users (absolute risk reduction [ARR] 1.11%; approximate NNT ≈ 90); sepsis was 1.7% vs 3.3% (ARR 1.6%; NNT ≈ 63). At 90 days, representative event rates likewise favored GLP exposure (e.g., DVT ∼0.5% vs 1.8%; surgical site infection ∼1.3% vs 3.4%), aligning with 60-70% relative reductions. These patterns concentrate in infection, wound-healing, and thromboembolic domains, with neutral signals for macrovascular ischemia (MI, stroke) and pneumonia, suggesting pathway specificity rather than a global reduction in all postoperative risks. Conclusion: Preoperative GLP exposure was associated with consistent, domain-specific risk reductions-approximately 40-70% lower odds and meaningful absolute benefits-in sepsis, surgical-site morbidity, thromboembolism, and mortality after laminectomy, with effects sustained through 1 year. To our knowledge, no prior decompression-only spine cohort has demonstrated such consistent, domain-specific perioperative benefits of GLP therapy. The combination of large effect sizes, persistence across time horizons, and selective pathway signal (infection/wound/thromboembolism, not MI/stroke) provides actionable insight: GLP therapy functions as a potentially modifiable perioperative risk factor rather than solely a metabolic treatment. These findings support evaluating GLP continuation/optimization within perioperative pathways for overweight patients undergoing decompression, with prospective studies needed to test causality, timing, and class-specific effects.

ID: AGS26/631

A124: The differential impact of tobacco, non-tobacco nicotine, and cannabis on postoperative complications after lumbar decompression

Abdel Diab 1 , Reme Arhewoh 1 , Omar Diab 2 , Rakesh Patel 1 , Ilyas Aleem 1 , Joshua Piche 1

1 University of Michigan, Ann Arbor, United States , 2 University of Texas Medical Branch, Galveston, United States

Introduction: Tobacco smoking is widely recognized for its negative impact on surgical outcomes. However, the differential effects of non-tobacco nicotine products and marijuana use on postoperative outcomes remain poorly characterized. With increasing adoption of e-cigarettes, smokeless tobacco, and legalized marijuana, understanding their implications in lumbar decompression surgery is crucial for perioperative risk stratification. This study examines the distinct associations of tobacco, non-tobacco nicotine dependence (NTND), and marijuana use to better inform perioperative risk stratification. Material and Methods: A retrospective cohort analysis was conducted using the TriNetX database, encompassing 104 healthcare organizations. Propensity score matching was employed to balance cohorts for age, gender, BMI, and key comorbidities including diabetes, hypertension, and CKD. Primary outcomes included 90-day complications, readmissions, and reoperations. Secondary outcomes encompassed sepsis, surgical site infection, myocardial infarction, acute renal failure, and mortality. Long-term outcomes were assessed at 2 years. Statistical analyses included risk differences, odds ratios (OR) with 95% confidence intervals, and Kaplan-Meier survival analysis for reoperation rates. Results: The study analyzed three cohorts: tobacco users (N = 3,012), cannabis users (N = 1,367), NTND users (N = 1,019), and respectively matched controls. After matching, baseline characteristics were well-balanced across groups. Tobacco use was significantly associated with increased risks of surgical site infection (2.4% vs 1.7%, p < 0.001; OR 1.42), sepsis (1.0% vs 0.6%, p = 0.001; OR 1.58), myocardial infarction (0.7% vs 0.4%, p = 0.001; OR 1.76), acute renal failure (1.7% vs 1.2%, p < 0.001; OR 1.44), 90-day readmissions (7.8% vs 5.6%, p < 0.001; OR 1.44), and mortality (0.4% vs 0.2%, p = 0.037; OR 1.60). However, tobacco users exhibited lower reoperation rates at two years (3.6% vs 4.3%, p = 0.004; OR 0.83), an observation requiring cautious interpretation due to potential survivorship bias and variations in follow-up patterns. Cannabis use was linked to higher 90-day readmission rates (8.4% vs 5.4%, p = 0.002; OR 1.61) but did not significantly impact other complications. NTND showed no increased risk across any measured outcomes. Conclusion: Tobacco use remains a critical modifiable risk factor for short-term complications and mortality following lumbar decompression surgery. Cannabis use was primarily associated with elevated 90-day readmission rates, underscoring the need for targeted perioperative monitoring. In contrast, NTND demonstrated no measurable impact on postoperative outcomes. These findings highlight the importance of tailored preoperative counseling for tobacco and marijuana users to mitigate risks and improve surgical outcomes. The observed association between tobacco use and reduced reoperation rates warrants further investigation, as it may reflect underlying data artifacts or survivorship biases. Future research should explore the long-term effects of NTND and cannabis use, particularly as their societal adoption continues to grow, to enhance perioperative risk assessment and patient care strategies.

ID: AGS26/3155

A125: 3D morphometric evaluation of surgical strategies for lumbar sagittal alignment restoration

Mete Özer 1,2 , Ersin Taşkın 1 , Tamer Atan 1 , Bedri Karaismailoğlu 2,3,4 , Mehmet Ali Talmac 1 , Önder Aydıngöz 3

1 Başakşehir Çam And Sakura City Hospital, Istanbul, Turkey

2 Istanbul University - Cerrahpaşa, Institute of Graduate Studies, Istanbul, Turkey

3 Istanbul University - Cerrahpaşa, Cerrahpaşa Faculty of Medicine, Istanbul, Turkey

4 Istanbul University-Cerrahpasa, CAST (Cerrahpasa Research, Simulation and Design Laboratory), Istanbul, Turkey

Introduction: This study compared the sagittal correction achieved by different surgical techniques for lumbar sagittal balance using three-dimensional model-based morphometric analysis. Material and Methods: Six lumbar CT datasets were reconstructed into three-dimensional models with Mimics and 3-Matic software. Four surgical scenarios were simulated at standardized levels: (1) single-level Smith-Petersen osteotomy (SPO) at L4-5, (2) single-level anterior lumbar interbody fusion (ALIF) at L4-5, (3) two-level Ponte osteotomy (L3-4 and L4-5), and (4) two-level oblique/lateral lumbar interbody fusion (OLIF/LLIF) with posterior pedicle screw fixation (L3-4 and L4-5). MATLAB-based morphometric analysis was performed to measure global and segmental lordosis, pelvic parameters, foraminal height, and disc height. Results: All techniques provided measurable improvements in sagittal alignment. Single-level ALIF and SPO achieved comparable angular correction, with ALIF yielding greater disc and foraminal height restoration. Two-level Ponte and OLIF/LLIF approaches resulted in cumulative angular gains, with Ponte showing more effective posterior column shortening and OLIF/LLIF demonstrating significant disc and foraminal height restoration. Conclusion: Different surgical techniques for lumbar sagittal balance provide distinct morphometric advantages. While single-level ALIF offers the strongest segmental correction with substantial height restoration, SPO provides posterior-based angular gain. Two-level Ponte osteotomy and OLIF/LLIF achieve comparable cumulative lordotic correction through different mechanisms-posterior column resection versus anterior column distraction. These findings highlight the importance of 3D model-based evaluation in guiding the selection of surgical strategy tailored to patient-specific deformity patterns.

ID: AGS26/607

A126: Single-position lateral-decubitus ALIF for lumbar spinal pathologies: a retrospective cohort study

Joseph Boktor 1 , Youssef Sadek 1 , Ahmed Nenaa 1 , Alok Tiwari 1

1 Royal Orthopaedic Hospital, Birmingham, United Kingdom

Introduction: Single position lateral decubitus Anterior Lumbar Interbody Fusion (ALIF) was introduced in the early 2000s and gained wider acceptance in last ten years with first introduction in Europe in 2019. Early reports suggest that this approach reduces operative duration, anaesthetics’ time, diminishes intraoperative blood loss, shortens hospital stay, and may lower complication rates. Despite these promising clinical benefits, lateral-decubitus ALIF remains relatively novel, with a paucity of published series and limited follow-up. The aim of this study is to review a consecutive series of lateral-decubitus ALIF procedures, detailing our surgical technique as well as clinical and radiographic outcomes, aiming to characterise its efficacy and safety profile and to inform strategies for enhancing patient outcomes following this approach. Material and Methods: Retrospective review of prospectively collected data from consecutive single surgeon series of patients who underwent a single-position lateral ALIF procedure with navigated posterior pedicle-screw insertion at the Royal Orthopaedic Hospital between June 2022 and March 2025.Patient demographics, relevant comorbidities, and details on pathology were extracted from electronic medical records. Operative variables recorded comprised: time for patient positioning, Interval from induction of anaesthesia to recovery, skin-to-skin operative duration, estimated blood loss, intraoperative and postoperative complications, and length of hospital stay. Sagittal-plane correction was quantified by measuring sacral slope, slip Boxall angle and disc height on lateral radiographs. Results: A total of 18 patients were included on this study. Mean age was 47.9 ± 12.1 years. The average body mass index (BMI) was 32 ± 6.26. 8 were males. Four out of the 18 patients (22.2%) were performed as half day list. The mean skin-to-skin surgical time averaged 191 ± 39.8 minutes. Median blood loss was low, at 100 mL, although 2 cases (11.1%) reported blood loss of over as sustained intra-operative complications (left common iliac vein bleeding). Both patients were satisfied at 6 months & 1 years follow up. No postoperative complications. Mean hospital stay was 4.2 ± 1.9 days. Mean Disc height increased from 0.88 ± 0.42 mm to 1.74 ± 0.47 mm (p < 0.0001). Mean Slip angle improved from +2.3° ± 5.8° (segmental kyphosis) to -8.7° ± 5.4° (lordosis), (p < 0.0001). Mean sacral slope increased from 39.2° ± 7.5° pre-operatively to 40.7° ± 6.8° post-operatively (p = 0.263). Conclusion: Single position ALIF with navigated posterior screws has good outcome and patient satisfaction. Our series showed that although main risk is higher possibility of vascular injury, this can be adequately managed with no long-term effects. With improved learning curve and staff awareness, two cases can be done in normal day List.

OP15: TB of the Spine

ID: AGS26/1675

A127: Machine learning-based prediction of in-hospital mortality in spondylodiscitis: a multicenter retrospective analysis

Felix Corr 1 , Felix Stengel 1 , Janna Hastings 2 , Dustin Grimm 1 , Carolin Albrecht 3 , Ann-Kathrin Jörger 3 , Sonja Häckel 4 , Aline Albrecht 4 , Valérie Lupi 4 , Philipp Störrle 4 , Marisa Gandía 5 , Carmen Novo 5 , Joshua Hägler 6 , Sara Lener 7 , Siegmund Lang 8 , Faizan Kareem 1 , Krauss Philipp 9 , Lorenzo Bertulli 1 , Martin Stienen 1 , Stefan Motov 1

1 Department of Neurosurgery & Interdisciplinary Spine Center, HOCH Health Ostschweiz, Cantonal Hospital of St. Gallen, St. Gallen, Switzerland ,

2 Medical Knowledge and Decision Support, University of St. Gallen, St. Gallen, Switzerland ,

3 Neurosurgery, Technical University Munich, Munich, Germany ,

4 Orthopedics, Inselspital, Bern University Hospital, Bern, Switzerland ,

5 Neurosurgery, Hospital Universitario La Paz, Madrid, Spain ,

6 Neurosurgery, Cantonal Hospital Aarau, Aarau, Switzerland ,

7 Neurosurgery, Medical University Innsbruck, Innsbruck, Austria ,

8 Universitätsklinikum Regensburg, Regensburg, Germany ,

9 Neurosurgery, University Hospital Augsburg, Augsburg, Germany

Introduction: Spondylodiscitis is a severe spinal infection characterized by a heterogeneous clinical presentation and significant in-hospital mortality. Current risk stratification tools offer limited predictive power. This study investigates the potential of machine learning to identify predictors of in-hospital mortality using routinely collected clinical data. Material and Methods: In this retrospective multicenter study conducted across five European centers between 2018 and 2023, clinical data from 191 patients diagnosed with spondylodiscitis were analyzed. After excluding cases with missing survival outcomes, 170 patient datasets were included for model development. A random forest classifier was trained to predict binary in-hospital mortality based on 15 clinical variables, including age, BMI, Systemic Inflammatory Response Syndrome (SIRS) und Quick Sequential Organ Failure Assessment (QSOFA), active malignancy, diabetes, drug and steroid use, and treatment modality. The dataset was randomly split into training (75%) and test (25%) sets. Model performance was assessed using accuracy on the held-out test set, and feature importance was evaluated via mean decrease impurity and permutation-based analysis. Results: The model achieved an accuracy of 76.7% on the test set, indicating moderate predictive performance. Key predictors of in-hospital mortality included age, BMI, QSOFA score, active cancer, alcohol use, diabetes, and treatment type. While treatment modality was not among the top-ranked features, it contributed to model accuracy. The confusion matrix revealed 32 true positives, 1 true negative, 10 false negatives, and 0 false positives. Attempts to predict hospital length of stay resulted in poor performance (accuracy: 20%), highlighting the need for additional data or alternative modeling approaches for this outcome. Conclusion: This study demonstrates that a machine learning approach using routinely available clinical variables can moderately predict in-hospital mortality in patients with spondylodiscitis. These findings support further refinement and prospective validation of the model to enhance clinical decision-making and individualized risk assessment.

ID: AGS26/2721

A128: Surgical outcomes of posterior-only surgery for thoracolumbar spinal tuberculosis: a five-year case series of 83 patients from a high-burden region

Phuc Diep 1 , Hanh Nguyen 1 , Truc Vu 1 , Tuan Ha 1

1 Hospital for Traumatology and Orthopaedics, Ho Chi Minh city, Viet Nam

Introduction: Thoracolumbar spinal tuberculosis (TLSTB) is a common form of musculoskeletal tuberculosis and remains a significant health burden in developing countries. Although anterior approaches have long been the standard, posterior-only surgery has emerged as a promising alternative, offering advantages in deformity correction, reduced invasiveness, and shorter hospitalization. However, robust evidence on its long-term outcomes and complication profile remains limited. This study provides context-specific evidence by evaluating clinical and radiological outcomes, as well as perioperative complications, of posterior-only surgery in TLSTB. Materials and Methods: This retrospective case series included 83 TLSTB patients who underwent posterior-only long-segment pedicle screw fixation with debridement and reconstruction using either structural autograft, an anterior distraction device, or a titanium mesh cage over a 5-year period. Clinical outcomes included Visual Analog Scale (VAS), ASIA scale, and Oswestry Disability Index (ODI). Radiological parameters were kyphotic Cobb angle, fusion by Bridwell criteria, and hardware failure. All variables were measured preoperatively, postoperatively, and at final follow-up. Multivariable regression analysis was performed to identify predictors of poor outcomes. Intraoperative and postoperative complications, including dural tear, nerve root sacrifice, wound-related complications, implant-related issues and recurrence were systematically documented. Results: The mean age was 41.6 ± 5.3 years, with a mean follow-up of 28 ± 6.1 months. VAS improved from 7.1 ± 1.84 to 1.8 ± 0.52 (p < 0.001). Neurological recovery was achieved in 92% of patients, with 73% improving by ≥ 2 ASIA grades. ODI decreased from 64.2 ± 10.74 to 24.6 ± 6.23 (p < 0.001). Kyphotic angle was corrected from 47.4° ± 23.94° to 16.7° ± 4.78° (p < 0.001). Solid fusion (Bridwell grade I-II) was achieved in 87%, while hardware failure occurred in 7.2%. Recurrence occurred in 2.4%, and the overall complication rate was 13.2%. Severe neurological deficit, delayed-onset paralysis, and baseline kyphosis were identified as predictors of poorer outcomes. Conclusion: Posterior-only surgery provides significant improvements in pain, neurological function, and deformity correction in TLSTB, supporting its role as a valuable alternative to conventional approaches. Nonetheless, limitations such as hardware failure, nonunion, and residual risks highlight the need for careful patient selection and long-term follow-up.

ID: AGS26/1873

A129: Reassessing neurological risk in spinal tuberculosis: the dominant role of mechanical instability over others

S. Rajasekaran 1 , Harikrishna R.R 2 , Sri Vijay Anand KS 2 , Ajoy P Shetty 2 , Rishi Kanna 2

1 Ganga Hospital and Medical Centre, Coimbatore, India

2 Ganga Medical Centre and Hospital, Coimbatore, India

Background: Spinal tuberculosis (TB) remains a major cause of morbidity worldwide, with neurological deficits occurring in approximately 10-30% of cases. While traditional predictors such as canal encroachment, kyphosis, and cord signal changes have received attention, the independent role of mechanical instability in neurologic impairment is unclear. This study reassesses the contribution of instability, as quantified by the Tuberculosis Spine Instability Score (TSIS), and compares its predictive value to established radiological factors in thoracic spinal TB. Methods: A retrospective review was conducted at a single tertiary center, including 122 patients diagnosed with thoracic spinal TB. Clinical, and radiological data were collected, with neurological status classified using the Frankel grading system and spinal stability using TSIS by Prof.S.Rajasekaran. Statistical analysis included univariate and multivariate logistic regression to determine independent risk factors for neurological deficit, with additional analysis of the interaction between instability with epidural compression and Canal encroachment area (CEA). Results: Univariate analysis showed that age, CEA, cord signal changes, epidural compression, and instability were significantly associated with neurological deficit. On Multivariate analysis, although CEA demonstrated a notable association, definite instability, as defined by TSIS, carried higher odds for neurological deficit (adjusted odds ratio = 8, p = 0.001) compared to it, while cord signal change and epidural compression losing statistical significance. The predicted probability of neurological deficit ranged from 1% (stable spine with no epidural compression) to 63% (unstable spine with epidural compression), confirming instability as the most prominent risk factor across subgroup. In stable spine, elevated CEA correlated with neurological deficit, but in unstable spines, the difference in CEA was less prominent, underscoring the predominant impact of instability over canal encroachment. Conclusion: Mechanical instability, as defined by TSIS, is the most robust and independent predictor of neurological deficit in thoracic spinal TB, superseding conventional radiological measures. Incorporating TSIS into routine evaluation enables early identification and prioritization of high-risk patients for surgical stabilization, thereby providing a practical, evidence-based shift from compression-based to instability-guided management and helping to prevent irreversible neurologic injury.

ID: AGS26/543

A130: Fluoroscopic guidance gives more useful results than CT guidance in transpedicular spine biopsy

Rishi Kanna 1 , Sai Keerthan 1 , Ajoy P Shetty 1

1 Ganga Hospital, Coimbatore, India

Introduction: Even though spinal biopsy is an important investigatory tool to diagnose spinal lesions (infections or tumors), the optimal image guidance to perform a biopsy is still unclear. While both Computerized tomography (CT) and Fluoroscopy are commonly used, the selection of one over the other is largely subjective based on availability of resources, clinician’s preference, and cost of care. Material and Methods: Case records of consecutive patients with spinal lesions (n = 206), inconclusively diagnosed as infection/primary tumor/metastasis in MRI and whom underwent a percutaneous biopsy were studied. The biopsy was performed through either CT (CTB, n = 101) guidance in the radiology suite or fluoroscopy (FLB, n = 105) guidance in the operating room. All CTB were done under local anesthesia with Jamshidi needle. FLB was performed under LA (n = 60) or GA (n = 45), with both Jamshidi needle and thin tissue forceps. The tissue was subjected to pyogenic culture, GenXpert test, histopathological examination and tubercular culture. The results were grouped into three categories: Highly Useful, Useful and Not useful. Results: FLB had a significantly higher number of useful results than CTB. In FLB, 58 (55.2%) were highly useful, 31 (29.5%) were useful, and 16 (15.2%) were not useful, while in CTB, 51 (50.5%) were highly useful, 6 (5.9%) were useful and 44 (43.6%) were not useful. FLB identified 89/105 disease-positive cases (sensitivity - 84.8%), while CTB identified 57/101 disease-positive cases, yielding a sensitivity of 56.4% (p = 0.0001, 4.29 higher odds and 1.5 times higher chances of predicting the disease). A high number of “not useful” results in CT guided biopsy were noted in the lumbar region (50.8%) and in patients with suspected infection (49.1%). Conclusion: Fluoroscopy guided biopsies performed in operating suite with combination of Jamshidi needle and tissue forceps are better than CT guided biopsies in planning further treatment, especially in patients with spinal infections.

ID: AGS26/2995

A131: Surgical site infection from posterior cervical fusion: a 10 year institutional experience

Ben Carnovale 1 , Samuel Wakelin 1 , Nicole Don 1 , Ayesha Waheed 1 , Hansen Deng 1,2 , Kojo Hamilton 1,2,3 , Nitin Agarwal 1,2,3

1 Department of Neurological Surgery, University of Pittsburgh School of Medicine, Pittsburgh, PA, United States

2 Department of Neurological Surgery, University of Pittsburgh Medical Center, Pittsburgh, PA, United States

3 Neurological Surgery, Veterans Affairs Pittsburgh Healthcare System, Pittsburgh, PA, United States

Introduction: Surgical Site Infection (SSI) is a significant concern postoperatively to spine surgeries, contributing to extended hospital stays, increased healthcare costs, morbidity, and mortality. In many cases, SSI requires re-opening of the surgical wound and a wash-out, as well as pharmaceutical treatment. Furthermore, the presence and severity of these infections may vary based upon surgery type and location, with posterior cervical fusion contributing unique risk factors for susceptibility to and spread of infection. The regimen surrounding these SSIs typically is well defined, involving empirical antibiotics preceding characterization of pathogens that inform a full treatment course. However, further investigation of these infections may better inform empirical regimens for patients presenting with potential SSI following posterior cervical decompression and fusion (PCDF). In this study, we aimed to characterize SSI following PCDF at a single, quaternary care institution and characterize the species and resistance profiles of the isolated organisms. Materials and Methods: We performed a retrospective review of posterior cervical fusions at a single institution over the past ten years. The electronic medical record was used to extract demographic, clinical, surgical, and outcome variables including age, sex, comorbidities, surgical indication, operative details, and presence of an infection. For each SSI, the time to wash-out, pathogenic organism, antibiotic type and duration, and resistance matrices (where performed) were recorded. Associations between SSI incidence and demographic, clinical, and surgical factors including body mass index (BMI), age, diabetes status, smoking status, surgical indication (degenerative, deformity, trauma, revision), and number of levels fused were assessed with univariate statistics and multiple logistic regression. Additionally, Gram-negative and Gram-positive pathogens were separated for a separate subanalysis. Results: Records for 590 posterior cervical fusions were recorded, with 47 noted infections (8.0%). The average time to wash-out was 31.4 days. BMI was positively associated with risk of infection in both univariate (p = 0.018, Odds Ratio 1.0464) and multiple logistic regression (p = 0.0272, Odds Ratio 1.0457). Coagulase negative Staphylococcus (20), Staphylococcus aureus (16), and Propionibacterium species (9) were the most common organisms detected. Seventeen SSIs included multiple pathogens. Patients with SSI had an average antibiotic course of 30.7 days. Sensitivity matrices were recorded for 35 of the positive cultures, with 20 (57%) exhibiting resistance to at least one antibiotic. The most commonly-resisted drugs were Oxacillin (11) and Erythromycin (6). Operative time was positively associated with odds of gram-negative bacterial infection (p = 0.038, Odds Ratio: 1.005). Conclusion: Surgical site infection following posterior cervical fusion remains a clinically significant complication, with an incidence of 8% in this cohort. Elevated BMI was identified as an independent risk factor, underscoring the importance of patient optimization prior to surgery. Additionally, increased operative time was associated with increased risk of gram-negative infection, potentially indicating a more severe infectious phenotype in longer operations. Coagulase-negative Staphylococcus and Staphylococcus aureus were the most isolated organisms, with over half of cultures demonstrating resistance to at least one antibiotic, most frequently oxacillin and erythromycin. These findings highlight the need for heightened awareness of resistant pathogens in tailoring empirical antibiotic regimens in patients with surgical site infection.

ID: AGS26/710

A132: Prevalence of spinal tuberculosis among extrapulmonary TB including MDR spinal TB: a systematic review

Julian Scherer 1,2,3 , Cornelia Gnägi 4 , Harvinder Singh Chhabra 5 , S. Rajasekaran 6

1 University of Cape Town, Cape Town, South Africa

2 University Hospital of Zurich, Zurich, Switzerland

3 Center for Musculoskeletal Surgery, Charite - University Medicine, Berlin, Germany

4 Tygerberg Hospital, Cape Town, South Africa

5 Sri Balaji Action Medical Institute, Delhi, India

6 Ganga Hospital, Coimbatore, India

Introduction: Spinal Tuberculosis (STB) is thought to represent most cases amongst Musculoskeletal TB (MSKTB) and can result in spinal instability, neurological fallout, severe pain and general TB symptoms. The WHO tuberculosis report does not report MSKTB cases and therefore also does not allow to estimate STB cases. When digging into international literature, data on prevalence and incidence of STB are derived from small case series, which leads to the question whether the epidemiological numbers on Extrapulmonary TB (EPTB) as well as MSKTB including STB are reliable. Similarly, the numbers on drug-resistant STB are questionable. The aim of this systematic review was to assess published cohorts examining different forms of tuberculosis to unravel a more precise prevalence of STB and MDR-STB. Material and Methods: We conducted a systematic review of the literature according to the PRISMA checklist and algorithm, aiming to assess all articles reporting on EPTB and STB. We performed an initial search of Web of Science TM for eligible investigations. The search terms were: (“extrapulmonary tuberculosis”) AND (“spinal tuberculosis”) (All Fields). We limited the search to cohort studies, randomized controlled trials and large case series, while systematic reviews, meta-analyses, small case series and case reports were excluded. We included papers published in English until November 2024. Results: The initial search revealed 180 publications of which 16 were included into the final synthesis. Five papers (Group A) were focusing on EPTB, Pulmonary TB (PTB) and STB comprising a total of 12,017 patients. The majority of patients had PTB (80.6%), and 19.4% had EPTB. Only 292 patients had STB which reflects 2.4% amongst all assessed patients and 12.5% amongst all EPTB patients. Three studies out of Group A were focusing specifically on STB and MSTK and revealed that less than a quarter (17.7%) of all EPTB cases were MSKTB and STB was the most frequent entity amongst MSKTB cases (67.5%). Eleven studies focused on EPTB and STB cases (Group B). However, most of the EPTB cases in these publications were MSKTB cases which may skew the results. Synthesis of these papers revealed a total of 7,736 patients of which 65.7% were STB cases. Nine studies specifically analyzed STB amongst MSKTB cases with a frequency of 67.5%. Taking all publications together which strictly analyzed STB and MSKTB cases (n=12; from Group A and B), the majority (67.3%) of all MSTKB cases accounted for STB. Only three amongst all assessed studies in this review analyzed drug resistance and showed that 11.83% (range 8.5% to 14.5%) of all STB cases were MDR-STB cases. Conclusion: Our results indicate that the incidence of spinal TB amongst EPTB as well as the incidence of MDR-STB in general might be underestimated. However, these data should be handled with caution since the assessed studies did not strictly aim to determine incidence or prevalence rates. Further, the assessed studies were mainly of retrospective nature and the study sites were not equally internationally distributed. Thus, prospective international registries should be implemented to further investigate the true incidence of STB and MDR-STB.

ID: AGS26/312

A133: Intraoperative 0.05% chlorhexidine gluconate (CHG) irrigation reduces surgical site infections in thoracolumbar spine trauma

Dallas Kramer 1 , Pauline Hoffman 1,2 , Caitlin Barrett 1,2 , Praveer Vyas 1 , Seung Jeong 1 , Daniel Altman 1,2 , Alex Yu 1,2

1 Allegheny Health Network, Pittsburgh, United States,

2 Drexel University College of Medicine, Pittsburgh, United States

Introduction: Surgery for traumatic thoracolumbar injuries carries nearly twice the risk of surgical site infection (SSI) compared to elective fusion for lumbar degenerative disc disease. Efforts to mitigate infection risks include dilute antibiotic and/or antiseptic (povidone-iodine) irrigation and vancomycin powder. Despite being available as a topical antiseptic in spine surgery for over 50 years, to date there is no literature on the use of intraoperative chlorhexidine gluconate (CHG) irrigation. Material and Methods: Retrospective review of patients undergoing surgery for traumatic thoracolumbar injuries between 2017-2023. Rates of SSI were compared among cases using diluted antibiotic irrigation with or without the addition of 0.05% CHG irrigation (450mL/bottle). Clinical and surgical data were collected from the electronic health records, including: age, sex, BMI, obesity, diabetes mellitus, tobacco use, spine region, number of intervertebral levels, surgery time, estimated blood loss, and subfascial vancomycin powder. Statistical analysis was performed using student’s t-test and chi-squared or Fisher’s exact tests (p < 0.05). Potential confounding variables for SSI were evaluated using bivariate analysis and multiple logistic regression. Results: 210 patients (82 control, 128 CHG) met inclusion criteria. There were no significant differences in patient demographics or surgical data besides greater utilization of vancomycin powder in the CHG group (43.0% vs. 20.7%, p < 0.005). There was a trend towards significantly fewer SSI with CHG irrigation (4.7% vs. 12.2%, p = 0.083). After controlling for potentially confounding variables including vancomycin powder, there was a significantly reduced risk of SSI with CHG irrigation (aOR 0.26 [95% CI; 0.06-0.83]; p = 0.028). The number needed to treat (NNT) to prevent one SSI was 11 patients. Conclusion: This is the first investigation of intraoperative 0.05% CHG irrigation during spine surgery, showing a significant reduction in the incidence of SSI in the setting of thoracolumbar spine trauma.

ID: AGS26/484

A134: Oral iron supplementation after lumbar decompression and fusion is associated with increased infection risk

Melissa Romoff 1 , Michael Kim 1 , Emily Tse 1 , Courtney Spitzer 1 , Andy Ton 1 , Vivian Chen 1 , Ryan Le 1 , Abhinav Sharma 1 , Justin Chan 1 , Emily Mills 1 , Hao-Hua Wu 1 , Don Park 1 , Yu-po Lee 1 , Nitin Bhatia 1 , Sohaib Hashmi 1

1 Univeristy of California, Irvine, Orange, United States

Introduction: Oral iron is commonly used to treat anemia, particularly in surgical patients who may be at risk of perioperative blood loss. While iron is known to positively drive red blood cell production, recent evidence suggests elevated free serum iron may promote bacterial overgrowth and impair immunity. In orthopedic and spinal procedures, postoperative infections are a major cause of surgical dissatisfaction, as they can lead to hardware failure and increased morbidity. However, the impact of postoperative iron supplementation on infection risk following lumbar spine surgery remains unexplored. Given the rising prevalence of spine surgeries and routine oral iron administration, greater insight into its effects on postoperative infection rate is critical. Material and Methods: A retrospective cohort study was conducted using data from the TriNetX United States Collaborative Network, a database including over 117 million patients. Patients who underwent lumbar spine fusion and decompression surgery between 2015 and 2024 were identified using Current Procedural Terminology (CPT) codes. Two cohorts were assembled: patients who received oral iron supplementation within 60 days of surgery and those who did not receive iron of any supplementation before or after surgery. Propensity score matching was performed to minimize confounders between groups based on key variables, age, sex, comorbidities, preoperative anemia, and postoperative blood transfusion status. Statistical analyses included the calculation of odds ratios (ORs) to evaluate relative risk, Kaplan-Meier survival analysis to measure infection-free survival probability, and hazard ratios (HRs) between groups. The primary outcome was the incidence of spinal fixation device-related infections within one year postoperatively. Results: A total of 48,528 patients met the study criteria (24,264 in each cohort). The incidence of spinal fixation device-related infections was significantly higher in the iron-supplemented cohort (0.874%) compared to the non-iron cohort (0.495%), with a statistically significant difference (p < 0.0001). Patients receiving oral iron had a 76.4% higher odds of developing infections compared to the non-iron cohort (OR: 1.764, 95% CI: [1.415-2.215], p < 0.0001) Kaplan-Meier survival analysis revealed a significantly increased hazard in the iron cohort, with 76.5% higher risk of developing an infection within one year postoperatively (HR: 1.765, 95% CI [1.412-2.206], p < 0.0001). Conclusion: Postoperative oral iron supplementation is associated with a significantly increased risk of spinal fixation device-related infections following lumbar fusion and decompression surgery. The increased infection rate may be attributed to iron’s role in fostering bacterial growth and modulating immune response; however, future prospective studies are needed to further elucidate the underlying mechanisms and identify strategies to improve patient outcomes. These findings underscore the need to critically evaluate routine iron supplementation practices in surgical patients.

ID: AGS26/2874

A135: Age- and localization-dependent pathogen distribution and mortality in pyogenic spondylodiscitis: insights from 10,801 nationwide cases

Siegmund Lang 1 , Volker Alt 1 , Nicolas H. von der Höh 2

1 University Hospital Regensburg, Regensburg, Germany , 2 University Hospital Leipzig, Leipzig, Germany

Introduction: Pyogenic spondylodiscitis is a rare but life-threatening spinal infection with rising incidence, particularly among elderly patients. Although Staphylococcus aureus is traditionally considered the leading pathogen, clinical experience suggests that Gram-negative bacilli and Enterococci become more relevant with age, potentially impacting empiric therapy. Large-scale, population-based confirmation of these shifts has been lacking. Methods: We performed a retrospective analysis of 10,801 patients with pyogenic spondylodiscitis from German nationwide hospital discharge data (InEK, 2024). Patients were stratified by age (< 65, 65-79, ≥ 80 years) and by infection site (cervical, thoracic, lumbar, multisegmental). Pathogen groups and resistant organisms (MRSA, VRE, MRGN) were identified from ICD-10 codes. Outcomes included in-hospital mortality. Statistical analyses comprised χ 2 tests and logistic regression with odds ratios (OR) and 95% confidence intervals (CI). Results: Lumbar infections were most common (50.7%), followed by thoracic (25.3%), multisegmental (12.5%), and cervical (11.5%). In-hospital mortality increased stepwise with age (2.6% < 65 y; 5.1% 65-79 y; 10.9% ≥ 80 y) and differed by localization, being lowest in lumbar infections (6.2%) and highest in multisegmental disease (12.9%). Pathogen distribution showed clear age-related shifts. Staphylococci overall declined significantly with age (35.7%, 28.5%, 21.3%; p < 0.001). Within this group, Staphylococcus aureus accounted for the majority of isolates but decreased more steeply (24.8%, 18.1%, 12.9%), whereas coagulase-negative and other staphylococci remained relatively stable (10.9%, 10.4%, 8.4%). In contrast, Enterobacterales (17.3%, 21.4%, 21.4%) and Enterococci (3.9%, 8.7%, 11.8%) increased with advancing age (both p < 0.001). Pseudomonas spp. were more frequent in elderly patients (up to 4.6 %), while anaerobes decreased (4.8% to 2.0 %). Resistant organisms were rare overall (1.5-3.2%) but rose with age (p < 0.001). By localization, S. aureus and other staphylococci predominated in cervical and thoracic disease (34.5% and 30.7%), while Enterobacterales and Enterococci were most common in lumbar and multisegmental infections (up to 23.5% and 13.4%, respectively; p < 0.001). Mortality analyses within age strata revealed distinct patterns. In patients < 65 years, Enterococci (7.4%, OR 3.27, 95% CI 1.53-7.00) and resistant organisms (12.2%, OR 5.55, 95% CI 2.11-14.59) carried the highest risk, while Staphylococcal infections were not associated with increased mortality. Among those aged 65-79 years, mortality was significantly elevated for Staphylococci (8.4%, OR 2.34, 95 % CI 1.79-3.07), Enterobacterales (7.7%, OR 1.79, 95% CI 1.34-2.39), Enterococci (8.7%, OR 1.89, 95% CI 1.29-2.78), Pseudomonas spp. (10.3%, OR 2.25, 95% CI 1.40-3.60), and resistant organisms (11.6%, OR 2.54, 95% CI 1.46-4.44). In patients ≥ 80 years, S. aureus and other staphylococci (15.5%, OR 1.72, 95% CI 1.36-2.17) and Enterobacterales (13.6%, OR 1.39, 95% CI 1.10-1.77) remained significantly associated with mortality, while other groups did not. Conclusion: This nationwide study demonstrates that both pathogen distribution and mortality in pyogenic spondylodiscitis are strongly dependent on age and localization. With increasing age, Gram-negative bacilli and Enterococci gain importance while S. aureus declines. Mortality increases stepwise with age and is linked to Enterobacterales, Enterococci, and resistant organisms in younger and middle-aged adults, and to staphylococci in the elderly. These findings support the need for age- and localization-adapted empiric treatment strategies to improve early coverage in pyogenic spondylodiscitis.

OP16: Cervical Trauma

ID: AGS26/393

A136: Surgical management of traumatic C2-C3 subluxation with bilateral carotid artery dissection

Vinicius Ricieri Ferraz 1 , Marcelo Ochoa Coelho de Souza Furlan 2 , Heber Martim 3 , Cesar Massami Yukita 3

1 Dr. José de Carvalho Florence Hospital, The Regional Hospital of São José dos Campos - Dr. Rubens Savastano, São José dos Campos, São Paulo, Brazil

2 Pontifical Catholic University of Paraná, Curitiba, Brazil

3 Department of Neurosurgery, The Regional Hospital of São José dos Campos - Dr. Rubens Savastano, São José dos Campos, São Paulo, Brazil

Introduction: Traumatic cervical spine injuries associated with vascular lesions are infrequent but present significant surgical challenges. An infrequent condition involves C2-C3 subluxation combined with bilateral carotid artery dissection, necessitating rapid diagnosis and a multidisciplinary treatment approach. This article outlines the surgical and endovascular techniques employed to manage such complex cases, highlighting critical technical aspects that facilitated effective stabilization and vascular preservation. Material and Methods: A 48-year-old male patient experienced a high-energy motorcycle collision resulting in severe cervical trauma. Imaging studies identified traumatic subluxation of the C2-C3 vertebral levels, bilateral dissection of the internal carotid arteries, and complete occlusion of the left vertebral artery. Due to the elevated risk of secondary ischemic injury, a multidisciplinary approach was implemented. Surgical stabilization was carried out via a posterior approach utilizing lateral mass and pedicle screw fixation extending from C1 to C4. During the procedure, meticulous care was taken to preserve vascular integrity and minimize cervical manipulation, thereby reducing the likelihood of further arterial compromise. Following spinal stabilization, the patient underwent an adjunctive endovascular intervention. Digital subtraction angiography confirmed bilateral carotid dissections, and subsequent stent placement coupled with antithrombotic therapy was performed to re-establish vascular flow and mitigate the risk of thromboembolic events. Results: The posterior fixation provided immediate spinal stabilization and restored cervical alignment without intraoperative complications. Postoperative imaging confirmed proper hardware placement and effective decompression. The endovascular procedure successfully re-established carotid lumen patency, and subsequent perfusion studies indicated improved cerebral blood flow. The patient’s neurological condition remained stable throughout hospitalization, with gradual motor recovery during rehabilitation. At 12-month follow-up, spinal fusion was solidly achieved, vascular patency was maintained, and no ischemic events occurred. Conclusion: The coexistence of traumatic cervical instability and bilateral carotid dissection necessitates a meticulously coordinated therapeutic approach. Posterior cervical fixation offers dependable stabilization with minimal disturbance to injured vessels, whereas endovascular therapy stands as a safe and efficacious modality for restoring carotid flow. This case underscores the significance of integrating advanced spinal instrumentation with endovascular techniques to enhance patient outcomes in cases involving concurrent spinal and vascular trauma. Early multidisciplinary intervention is paramount to avert catastrophic neurological sequelae and to secure long-term functional recovery.

ID: AGS26/1457

A137: Prognostic significance of C7-T1 involvement in unstable subaxial cervical fractures treated surgically: a multicenter retrospective study

Guillermo Ricciardi 1,2,3 , Guisela Quinteros Rivas 4 , Ignacio Cirillo 5,6,7 , Manuel de Jesus Ambrosio Vicente 8,9 , Edgar Márquez García 10 , Sinval Dorneles 11 , Jhofre Estuardo Portilla Fárez 12 , Eduardo Laos 13 , Carla García-Ramos 14 , Alberto Gotfryd 15 , Diego Zambonin 16 , Mauricio Cardoso 17 , Charles Carazzo 18 , Ratko Yurac 19 , Alfredo Guiroy 20

1 Centro Médico Integral Fitz Roy, Sanatorio Guemes, Hospital General de Agudos Dr. Teodoro Alvarez, Buenos Aires, Argentina

2 Sanatorio Guemes, Buenos Aires, Argentina

3 Hospital General de Agudos Dr. Teodoro Álvarez, Buenos Aires, Argentina

4 Clínica Alemana, Santiago, Chile

5 Hospital del Trabajador, Clínica Las Condes, Universidad del Desarrollo, Santiago, Chile

6 Clínica Las Condes, Santiago, Chile

7 Universidad Andrés Bello, Santiago, Chile

8 Clínica de Columna Dr. Manuel Dufoo Olvera - Secretaría de Salud de la Ciudad de México, Mexico, Mexico

9 Hospital General de Morelia, Mexico, Mexico

10 Hospital de Traumatología Dr. Victorio de la Fuente Narvaez - Instituto Mexicano del Seguro Social, Mexico, Mexico

11 Hospital Estadual de Urgências Governador Otávio Lage de Oliveira, Goiana, Brazil

12 Hospital General Isidro Ayora, Loja, Ecuador

13 Hospital Nacional Guillermo Almenara Irigoyen, Lima, Peru

14 Instituto Nacional de Rehabilitación, Mexico, Mexico

15 Santa Casa de Misericórdia de São Paulo, Sao Paulo, Brazil

16 Hospital Cristo Redentor, Porto Alegre, Brazil

17 Hospital João XXIII, Belo Horizonte, Brazil

18 Neurosurgery, University of Passo Fundo, São Vicente de Paulo Hospital - Passo Fundo - RS, Brazil, Passo Fundo, Brazil

19 Spine Unit, Department of Traumatology, Clínica Alemana, Santiago, Chile, Santiago, Chile

20 Neurosurgery. Elite Spine Health and Wellness Center (Florida, USA, Florida, United States

Introduction: Subaxial cervical fractures occur between C3 and C7, most frequently involving C5-C7, while the C7-T1 junction accounts for nearly 10% of cases. This cervicothoracic transition zone presents unique biomechanical and anatomical challenges, with prior studies suggesting higher neurological compromise, fracture-dislocation rates, and surgical complications. The objective was to determine whether C7-T1 involvement is a prognostic factor for early complications in surgically treated unstable subaxial cervical fractures. Methods: We conducted a multicenter retrospective cohort study across 13 Latin American spine centers (2014-2023). Adults aged 18-65 with traumatic unstable fractures (A3, A4, or C) between C3 and C7-T1 were included. Exclusion criteria were pathological, fragility, or ankylosing spondylitis-related fractures, and spinal infections. The primary outcome was early postoperative complications (within 90 days). The main independent variable was injury level (C7-T1 vs. other subaxial). Additional variables included demographics, comorbidities, neurological status, injury type, surgical features, and hospital course. Logistic regression identified independent predictors. Results: Of 541 patients, 474 met criteria (77.6% male; mean age 39.3 years). C7-T1 injuries represented 11.4%. Overall, 25.9% developed complications, most commonly pneumonia (12.4%), urinary tract infection (7.8%), and surgical site infection (2.7%); mortality was 4%. Multivariate analysis identified AIS grade A (OR 3.75), AIS grade B (OR 2.81), polytrauma (OR 3.38), and F4 facet fractures (OR 1.67) as independent predictors. Injury level was not significantly associated. Conclusion: Early complication rates were comparable between C7-T1 and other subaxial levels. Neurological status, polytrauma, and facet injury patterns (not vertebral level) were the key predictors of morbidity.

ID: AGS26/1837

A138: Anterior spinal artery visualization on CT angiography as a prognostic marker in acute cervical spinal cord injury

Baptiste Boukebous 1 , Amal Abayed 2 , Pierre Antoine Allain 2 , Myriam Lamamri 2 , Mohamed Zoghlami 2 , Bilal El Hachache 2 , Marc-Antoine Rousseau 1

1 APHP, Université Paris Cité, Paris, France

2 APHP, Paris, France

Introduction: There are very few data regarding the involvement of the anterior spinal artery (ASA) in acute cervical spinal cord injuries (SCI), including decompensation of cervical spondylotic myelopathy, traumatic fractures or dislocations, or spinal cord injury without radiological abnormalities (SCIWORA). Some authors have reported the consistent visualization of this artery on CT angiography in severe spinal cord trauma, whereas others have noted that it is poorly visualized. We hypothesize that, when visible, the presence of the ASA in the cervical region on CT angiography represents an indicator of initial severity and a predictor of poor neurological recovery. Materials and Methods: A total of 37 patients admitted to the intensive care unit for acute cervical trauma were analyzed (mean age 40 years, mean follow-up 145 days): 16 SCIWORA, 5 cases of cervical spondylotic myelopathy decompensation, 7 fracture-dislocations treated surgically, and 9 minor fractures treated with a cervical collar (e.g., transverse process). None of these patients had associated traumatic brain injury or thoracolumbar spinal cord injury. The ASA was assessed on CT angiography performed as part of the initial trauma evaluation. Two independent readings were performed, one of them blinded to the neurological status. Initial and last follow-up neurological status were assessed using the American Spinal Injury Association (ASIA) Impairment Scale. The Simplified Acute Physiology Score (SAPS) was collected to assess overall trauma severity. Results: At admission, the ASIA distribution was: 12/37 A, 6/37 B, 2/37 C, 6/37 D, and 11/37 E. The ASA was detected in 13/37 patients (35%) by both readers, with inter-observer agreement in 35/37 cases. Ten out of twenty patients graded ASIA A, B, or C (50%) had a visible ASA, compared with 3/17 (18%) in ASIA D or E (p = 0.04). No significant difference in SAPS was observed between patients with or without a visible ASA. Considering the ASIA score as a progressive numerical scale (A = 1, E = 5) and accounting for each patient’s potential for recovery (e.g., an ASIA E patient has no recovery potential, whereas an ASIA A patient may gain up to 4 points), patients without a visible ASA had a recovery potential of +72% relative to their baseline, compared with +33% for patients with a visible ASA, corresponding to an average 40% reduction in neurological recovery potential for patients with a visible ASA (CI95%[-69%, -10%], p = 0.01). For example, among the 5 patients initially graded ASIA A without a visible ASA, all improved to at least ASIA D; among the 7 patients initially ASIA A with a visible ASA, only 2 improved to at least D. All SCIWORA patients without a visible ASA reached at least ASIA D at the last follow-up, regardless of their initial neurological status. Conclusion: Whether due to arterial spasm, increased blood flow, occlusion, or raised cerebrospinal fluid pressure, the reasons why the ASA became visible in patients with poorer prognosis SCI remain unclear. However, this observation paves the way for further investigations. The visibility of the ASA could already be considered an early prognostic factor for neurological recovery.

ID: AGS26/228

A139: Mortality risk in type ii dens fractures: the role of pneumonia and neurocognitive disorders after anterior screw fixation in a German level 1 trauma center

Kenny Hagar 1 , Zaccaria Richard 1 , Jan Wulf 1 , Wolfgang Böcker 1 , Carl Neuerburg 1 , Suero Eduardo 1

1 LMU München, MUM - Muskuloskelettale Universitätsmedizin München, Trauma and Orthopedic Surgery, Munich, Germany

Introduction: Due to demographic changes and the globally progressing aging of the population, the number of osteoporotic vertebral fractures is steadily increasing. Dens fractures account for approximately 15-20% of all cervical spine fractures. Depending on associated comorbidities, this type of fracture may carry an increased risk of mortality following surgical treatment using ventral screw osteosynthesis. The aim of this study is to evaluate the clinical outcomes of anterior screw fixation in patients with type II dens fractures and to identify the influence of various comorbidities on mortality in this patient population. Material and Methods: This retrospective study included 82 patients with a mean age of 75.7 years who were treated between January 2011 and December 2019 at two hospital locations and by two specialized departments. All patients underwent anterior screw fixation for type II dens fractures according to the Anderson and D”Alonzo classification. The study aimed to assess the influence of demographic and clinical factors on mortality in this context. We first conducted a Cox regression analysis to evaluate potential risk factors. Next, we identified predictors of overall mortality using logistic regression models, with postoperative death as the dependent variable. Results: The occurrence of postoperative delirium significantly increased both 30-day and overall mortality (p < 0.05). Similarly, patients who developed postoperative pneumonia had significantly higher mortality rates (p < 0.05). A pre-existing diagnosis of dementia at the time of surgery was also associated with a significantly increased 30-day mortality (p < 0.05). Other comorbidities, however, did not show a significant impact on mortality (p > 0.05). Conclusion: Our results indicate that patients developing complications such as delirium or pneumonia after ventral screw osteosynthesis for type II dens fractures are at a significantly increased risk of mortality. Additionally, pre-existing dementia correlates with a higher postoperative mortality rate. The risk of death increases substantially when these factors occur in combination. Therefore, effective prevention and management of delirium and pneumonia are critical in the postoperative care of these patients. Although dysphagia was not associated with increased mortality, early involvement of speech-language therapy following surgery may be beneficial. Surgical complications themselves did not influence mortality. The optimal management of type II dens fractures remains a subject of ongoing scientific discussion.

ID: AGS26/2297

A140: The role of Hounsfield Units in predicting cage subsidence after lateral lumbar interbody fusion: a systematic review and meta-analysis

Chen Zhang 1 , Zachary Chu 2 , Jonathan Boey 3 , Reuben Soh 3

1 Faculty of Medicine and Health, University of New South Wales, Sydney, Australia

2 Yong Loo Lin School of Medicine, National University of Singapore, National University Health System, Singapore, Singapore

3 Department of Orthopaedic Surgery, Singapore General Hospital, Singapore, Singapore

Introduction: This systematic review and meta-analysis examined cohort studies investigating the relationship between vertebral Hounsfield unit (HU) measurements on preoperative computed tomography (CT) scans and postoperative cage subsidence (CS) in patients undergoing lateral lumbar interbody fusion (LLIF). As DEXA scans may inaccurately reflect local bone quality, vertebral HU on CT scans have emerged as a viable alternative to predict CS. This study aimed to identify a potential HU threshold predictive of CS in LLIF for future studies to explore and verify. Material and Methods: Four databases were searched for studies reporting CS rates and preoperative segmental HU in patients undergoing LLIF, including oblique lateral interbody fusion, extreme lateral interbody fusion, anterior-to-psoas, and transpsoas techniques for degenerative spinal conditions. Random-effects meta-analyses were conducted. Outcomes were pooled segmental HU values and the predictive performance of different HU thresholds for CS. Intra-study risk of bias was rated using the Newcastle-Ottawa scale, and evidence certainty was assessed with the GRADE approach. Results: Five retrospective cohort studies totaling 396 patients (271 without CS, 125 with CS) were included. The pooled segmental HU was 112.1 (95%-CI: 96.3-128.0) for patients with CS and 142.6 (95%-CI: 127.5-157.6) for patients without CS. The pooled mean difference was 29.9 (95%-CI: 14.4-45.4, p < 0.01, p egger = 0.51). A threshold of ≤ 120 HU had the best predictive performance for CS (Sensitivity: 75.9%, 95%CI: 31.9% to 95.5%, Specificity: 88.0%, 95%CI: 75.1% to 94.7%, AUC: 0.903). Conclusion: This study affirms that HU is a potentially useful predictive tool for CS in patients undergoing LLIF for degenerative spinal diseases. Future studies should explore the cut-off value of 120 HU to validate its predictive capability.

ID: AGS26/400

A141: Management and outcomes of coincident traumatic atlantoaxial fractures: a level 1 trauma center case series

David Caldwell 1 , Vardhaan Ambati 1 , Jun Oh 1 , Ryan Wilson 1 , Hunter Yamada 1 , John Yue 1 , Ashraf El Naga 1 , David Gendelberg 1 , Rajiv Saigal 1 , Anthony DiGiorgio 1 , Praveen Mummaneni 1 , Nima Alan 1

1 University of California San Francisco, San Francisco, United States

Introduction: Coincident traumatic atlantoaxial fractures are common among older patients. These fractures pose challenges for management and outcomes data for patients with fractures are limited. The current study provides an institutional experience for clinical management. Material and Methods: We retrospectively reviewed 68 cases at Zuckerberg San Francisco General Hospital (San Francisco, California) with traumatic atlantoaxial fractures between 2003 and 2024. Univariate analyses were performed to identify factors associated with inpatient mortality, operative versus non-operative treatment, and fracture union. Multivariable logistic regression, using best subset logistic regression with a maximum of two predictors due to the number of patients, as well as multivariable logistic regression using expert selected variables, was used to identify factors associated with mortality and operative treatment. Univariate odds ratios from logistic regression were used to analyze fracture patterns. Results: Coincident traumatic atlantoaxial fractures predominantly happen in elderly individuals (74.2 years, range 23.0-100.0), and 48.5% of patients had neurological deficits. Univariate analyses demonstrated patients who expired during admission had high rates of neurologic deficits (82.4% vs. 37.3%, p = 0.001). These fractures have high rates of bony non-union in conservatively managed cases (44.4% of 9 patients with follow-up), while no non-union was observed amongst surgical cases. A smaller proportion of females underwent surgery (23.5% of surgical cases, vs 54.9% of managed nonoperatively cases, p = 0.025). Of 51 patients who survived hospitalization, 4 (7.8%) died within 1 year, and 9 (17.6%) in total (longest follow-up 126 months). On univariate logistic regression analysis, lateral structures were often injured together (occipital condyle with C1/C2 transverse foramen (odds ratio 8.33, p < 0.05), C1/C2 transverse foramen with the C2 lateral mass (odds ratio 5, p < 0.05), and C1 Landells Type III fractures with the C2 lateral mass (odds ratio 3.75, p < 0.05). On multivariable analysis, the presence of neurological deficits was associated with inpatient mortality (p = 0.004), while there were no significant associations between operative treatment and predictor variables. Conclusion: Traumatic atlantoaxial fractures occurred in a predominantly elderly and have a high rate of bony non-union with non-operative management, though fibrous nonunion can occur. Patients with this fracture pattern have high inpatient and post-discharge mortality rates. Surgery was not statistically associated with either expiration during admission or long-term survival. For management, these fractures require individualized decision making, and we propose that in select patients, surgery can help maximize the chances of bony union.

ID: AGS26/575

A142: A rare case of delayed intra cranial hypotension with unilateral multiple cranial nerve palsies from cerebro spinal fluid leak in a whiplash injury

Debasish Hajra 1 , Ankita Hajra 2 , Akshima Ghosh 3 , Radu Beltechi 4

1 University Hospital of Coventry and Warwickshire NHS Trust, Coventry, United Kingdom

2 University College London, London, United Kingdom

3 University of Oxford, Oxford, United Kingdom

4 University Hospital of Coventry and Warwickshire NHS trust, Coventry, United Kingdom

Introduction: Intracranial hypotension resulting from cerebrospinal fluid (CSF) leak in the neuraxis is divided into either primary or spontaneous intracranial hypotension (SIH) and secondary, like in trauma or iatrogenic cases. We present a rare secondary intracranial hypotension, with multiple unilateral cranial nerve palsies and to our knowledge not reported so far in literature. Material and Methods: We present a 33-year-old male who was diagnosed with a whiplash injury after a road traffic accident. He had persistent headaches and developed unilateral VI, VII and VIII cranial nerve palsies. This is rare and literature search has shown that there is either single unilateral cranial nerve palsy with dural leak or multiple unilateral cranial nerve palsies in epidural analgesia without a dural leak. MRI and lumbar puncture with opening pressure of 5.8 cms of water, strongly suggested intra-cranial hypotension. CT myelogram confirmed overflow of dye at C1-C2 level. Nearly 30-70% of SIH, improves with epidural blood patches but definitive cure in type 1, type 2 and type 3 (classification by Schievink WI et all), mostly requires surgical closure. Results: Our patient opted for surgery. He had a posterior hemilaminectomy of C1 and C2 and repair of leak. He recovered well and was discharged. Conclusion: Delayed CSF leak and the resultant intracranial hypotension has always been an entity which has been the cause of persistent headache and is commonly misdiagnosed, resulting in significant delay in the initiation of managements and the risks associated with treatment for disorders that mimic intracranial hypotension. The postural headache, which is the clinical hallmark of intracranial hypotension, is most likely due to the direct result of reduced CSF volume with loss of buoyancy resulting in caudal displacement of the brain, causing traction and irritation on pain-sensitive structures. over the time this becomes less prominent as the untreated CSF leak becomes chronic. This case highlights cervical whiplash in RTAs as an important and under-considered cause of traumatic dural tear and CSF leakage with a delayed presentation. Associated multiple unilateral cranial nerve palsies make our case unique, and reports of such presentation have not been seen in literature so far. Increasing the awareness of this type of headaches and associated cranial nerve palsies whenever they are present is required to decrease the high rate of misdiagnosis.

ID: AGS26/581

A143: Micro-structural MRI is a biomarker which predicts baseline heterogeneity & outcomes after acute cervical spinal cord injury-insights from the RISCIS imaging sub-study

Mohammed Ali Alvi 1 , Mouad Elganga 2 , Karlo Pedro 1 , Raman Abbaspour 2 , Michael Fehlings 1

1 University Health Network, University of Toronto, Toronto Western Hospital, Toronto, Canada ,

2 University of Toronto, Toronto, Canada

Introduction: Early prognostic biomarkers are essential to optimize management strategies and improve the design of clinical trials in spinal cord injury (SCI). The RISCIS Trial (Safety and Efficacy of Riluzole in Acute Spinal Cord Injury Study), found that patients treated with riluzole demonstrated improvements in functional recovery in patients with an acute cervical SCI (ASIA A-C). In the current analysis, we analyzed a subset of patients enrolled in the RISCIS trial with acute (within 72 hrs after SCI) MRIs to assess the association of SCI lesion morphometrics with baseline and follow up neurological and functional outcomes. Materials and Methods: A total of 24 patients (12 Riluzole and 12 Placebo) were analyzed. For each patient, T2W sequence MRIs (obtained within 72-hrs of injury) were processed using the SpinalCordToolbox (SCT) v7.1 sct_deepseg lesion_sci_t2 command. Quantitative metrics including ventral, dorsal and total tissue bridge width and lesion volume were extracted. We examined the association between lesion morphometrics and baseline motor and functional scores as well as the change in motor and functional scores. We also assessed interactions effect between treatment (Riluzole) and imaging biometrics. Results: The mean age of the cohort was 52.5 years and 71% (n = 17) were males. The distribution of ASIA grade at baseline was: ASIA A 25% (n = 6), ASIA B 16.7% (n = 4), ASIA C 54.2% (n = 13), and ASIA D 4.2% (n = 1), and was also found to be comparable between the two groups (p = 0.433). At baseline, we observed strong correlations for ventral tissue bridge width (VTBW) and total tissue bridge width (TTBW) with total motor score (R2 = 0.303, p = 0.016; and R2 = 0.288, p = 0.020), and with upper motor score (R2 = 0.333, p = 0.010; and R2 = 0.266, p = 0.030) after adjusting for age and ASIA grade. At follow-up, we observed a strong correlation between TTBW and spinal cord independence measure (SCIM). The interaction between treatment type (Riluzole or Placebo) and TTBW (p = 0.042) or VTBW (p = 0.049) was found to be significant for total motor score change. The extent of preserved tissue bridge was predictive of recovery in motor score at follow-up for patients receiving Riluzole. We also analyzed a global outcome score, a Global rank-sum score calculated by summing pooled ranks across ANCOVA-adjusted endpoints: Total motor score, SF-36 physical component, and SCIM. We found strong correlation between TTBW and the global outcome (B = 6.794, p = 0.044) as well as VTBW (B = 9.98, p = 0.046) and the global outcome. Conclusion: Our results indicate that quantitative MRI metrics obtained from a semi-automated AI-facilitated pipeline may help prognosticate outcomes after acute cervical SCI and may thus be used for potential risk stratification, as well as for targeted personalized interventions.

ID: AGS26/876

A144: MRI signal intensity and extension as predictors of clinical outcomes in cervical spinal cord injuries

Enrico Giordan 1 , Andrea Valenti 1 , Luca Chinellato 1 , Giuseppe Canova 1

1 Aulss 2 Marca Trevigiana, Treviso, Italy

Introduction: Traumatic cervical spinal cord injury (cSCI) is a complex condition with unpredictable neurological outcomes. While magnetic resonance imaging (MRI) is the gold standard for initial assessment, its ability to predict functional recovery is unclear. This study explores whether the volumetric extent and signal intensity of cSCI on preoperative MRI correlate with neurological and functional outcomes after early surgical decompression and rehabilitation. Material and Methods: This retrospective study involved 45 adult patients with acute cervical spinal cord injury (cSCI) who underwent urgent spinal decompression within 12 hours of injury. Preoperative cervical MRI T2-weighted sequences assessed signal intensity and cSCI volume (3D volumetric analysis, %VSCI) and peak hyperintensity (SCIir). Functional and neurological outcomes were measured preoperatively, postoperatively, and after rehabilitation using the ASIA Impairment Scale, modified Japanese Orthopaedic Association (mJOA) score, and Barthel Index (BI). Multiple regression analyses, adjusted for age and sex, examined the relationships between imaging parameters and outcomes. Results: Regression analysis revealed that increased %VSCI was independently associated with lower preoperative mJOA scores (p = 0.014) and lower post-rehabilitation mJOA scores (p = 0.001). No significant associations were found between %VSCI and ASIA score variation (p = 0.234), nor between SCIir and pre- or postoperative ASIA or mJOA scores (p > 0.5 in all cases). The mean mJOA score improved from 8.9 ± 3.2 to 12.6 ± 5.1 (p < 0.001), and 40.9% of patients reached ASIA E at final follow-up. Conclusion: This retrospective study demonstrates that neither the volumetric extent nor the signal intensity of cSCI on preoperative MRI-T2weighted images is a reliable predictor of postoperative neurological recovery. Although associations were found with baseline disability scores, no significant correlations with clinical improvement emerged after decompression and rehabilitation. These findings suggest that MRI characteristics alone cannot capture the complexity of SCI prognosis. In conclusion, while MRI remains a cornerstone in the diagnostic workup of cSCI, it should not be solely relied upon to guide prognostication.

OP17: Adult Deformity Decisions

ID: AGS26/72

A145: The effectiveness and safety of staged halo-pelvic traction combined with posterior spinal fusion in the treatment of severe rigid spine deformity: a prospective cohort study

Lijin Zhou 1 , Honghao Yang 1 , Yong Hai 1

1 Beijing Chao-Yang Hospital, Beijing, China

Introduction: Severe rigid spinal deformities present significant challenges in correction surgery due to complexity and associated comorbidities. To mitigate the surgical risks, preoperative halo-pelvic traction (HPT) have been employed. This study aims to evaluate the effectiveness and safety of staged HPT combined with posterior spinal fusion (PSF) in the treatment of severe rigid spine deformity. Material and Methods: This is a prospective cohort study. From 2020 to 2022, 61 consecutive patients (mean age 26.2 years) with severe rigid spine deformity who underwent staged HPT combined with PSF with a minimum 24-month follow-up were recruited. Radiographic parameters, clinical information, pulmonary functions tests, and perioperative complications were recorded. Results: The mean preoperative coronal Cobb angle was 114.2° ± 38°, and the mean MK was 105.8° ± 34.7°. Following the HPT (mean duration 19.2 weeks), the mean coronal Cobb angle were corrected to 55.3° post-traction (50.6%) and 47.4° after PSF (58.3%); the mean MK angle were corrected to 52.6° post-traction (49.5%) and 38.1° after PSF (63.4%). The overall complication rate during HPT was 16.4%, while surgery-related complications were 18.0%, with no permanent neurological deficits observed. Conclusion: Staged HPT combine with PSF is effective and safe for patients with severe rigid spine deformities. HPT could mitigate the severity of spine deformity, minimize the need for invasive three-column osteotomies, and reduce the risk of complications for correction surgery.

ID: AGS26/1203

A146: Underestimated subtype in coronal type C degenerative scoliosis: absence of lumbar fractional curve increase postoperative coronal imbalance risk

Yong Qiu 1 , Zezhang Zhu 1 , Zhen Liu 1 , Ming Wang 1 , Xiaodong Qin 1 , Saihu Mao 1 , Benlong Shi 1

1 Division of Spine Surgery, Department of Orthopedic Surgery, Nanjing Drum Tower Hospital, The Affiliated Hospital of Medical School, Nanjing University, Nanjing, China, Nanjing, China

Introduction: To investigate the influence of distal lumbar fractional curve (FC) on postoperative coronal imbalance (CIB) in patients with coronal type C degenerative scoliosis (DS) undergoing sequential correction techniques. Material and Methods: Patients with coronal type C DS undergoing sequential correction were included. Based on the presence of FC (L3-L5 Cobb angle > 10° with the convex side opposite to the major curve), patients were divided into FC and Non-FC groups. At final follow-up, patients were further categorized into CIB+ (coronal balance distance > 3 cm) and CIB- groups. Demographics, radiographic parameters, surgical data, patient-reported outcomes, and complications were compared using chi-square and Student t-tests. The cutoff values for L4 and L5 tilt were determined based on maximal chi-square values. Results: A total of 43 patients (61.5 ± 7.4 years) were enrolled, with 21 in the FC group and 22 in the Non-FC group. Compared to the FC group, the Non-FC group had a significantly higher incidence of postoperative CIB (54.5% vs 14.3%, p < 0.001), which persisted at the final follow-up (36.4% vs 9.5%, p < 0.001). The Non-FC group exhibited greater L5 tilt both postoperatively and at final follow-up. The correction rates for L4 tilt, and L5 tilt were significantly higher in the CIB- group (p < 0.05). Additionally, patients in the CIB- group demonstrated significantly lower L4 tilt, L5 tilt, and apical vertebral translation at final follow-up compared to the CIB+ group. The optimal correction targets were defined as L4 tilt ≤ 13° and L5 tilt ≤ 7°. Conclusion: This study identified two subtypes of coronal type C DS based on distal lumbar FC. Non-FC patients had a higher risk of postoperative CIB. Achieving L4 tilt ≤ 13° and L5 tilt ≤ 7° in Non-FC patients may reduce the incidence of CIB. For FC patients, conventional sequential correction techniques can achieve satisfactory coronal balance.

ID: AGS26/2019

A147: Optimizing spinal realignment: a comparative analysis of correction and complications of osteotomy techniques in adult spinal deformity

Diego Soto Rubio 1 , Cesar Carballo 1 , Kiana Yeganeh 2 , Schahin Salmanian 1 , Bryan Clampitt 3 , Samantha Schimmel 3 , Mohammadmahdi Sabahi 4 , Dana Saleh 1 , Mohsen Rostami 1 , Jay I. Kumar 1 , Puya Alikhani 1

1 Department of Neurosurgery, Brain and Spine, University of South Florida, Morsani College of Medicine, Tampa, United States

2 Ponce Health Sciences University School of Medicine, Ponce, Puerto Rico, USA, Ponce, United States

3 USF Health Morsani College of Medicine, Tampa, United States

4 Department of Neurological Surgery, Pauline Braathen Neurological Center, Cleveland Clinic Florida, Weston, FL, USA, Weston, United States

Introduction: Osteotomies are fundamental for correcting adult spinal deformity (ASD). We compared the effectiveness of anterior column realignment (ACR), pedicle subtraction osteotomy (PSO), intradiscal osteotomy (IDO), and Ponte osteotomies in achieving spinopelvic correction, their clinical outcomes, and complications. Material and Methods: We retrospectively analyzed 146 patients who underwent posterior fusions for ASD correction between 2016 and 2022, considering only patients with ≥ 1 year of follow-up. Patients were grouped according to the osteotomies with the most significant impact on sagittal alignment change. IDO, PSO, ACR or Ponte. Spinopelvic parameters, including pelvic tilt (PT), pelvic incidence (PI), sacral slope (SS), lumbar lordosis (LL), PI-LL mismatch, and sagittal vertical axis (SVA), and their changes from preoperative to postoperative images were compared. Surgical and clinical variables were collected, including mechanical complications (proximal junctional kyphosis [PJK], and failure [PJF], different types of hardware failure [HF]), estimated blood loss [EBL], packed red blood cell [PRBC] transfusions, and length of stay [LOS]. Clinical status was measured with the Oswestry disability index [ODI]. Revision-free survival time was analyzed using Kaplan-Meier curves, with patients followed from index surgery until revision or last follow-up, and differences between osteotomy types were assessed. Results: A total of 146 patients underwent ASD correction with IDO (n = 23), PSO (n = 21), ACR (n = 32), or Ponte (n = 70) osteotomies. Groups were comparable in age, BMI, preoperative disability, and most spinopelvic parameters. PSO achieved the greatest sagittal correction (ΔLL = 29.7 ± 19.1°, ΔPI-LL mismatch = -82.6 ± 16.1, ΔSVA = -74.6 ± 51.6), while IDO and ACR produced intermediate corrections, and Ponte the least. EBL and PRBC units transfused were lower in ACR and Ponte, corresponding to shorter instrumented constructs. PJK occurred most frequently in ACR (31.3%) and Ponte (21.7%), while IDO had the lowest rate (8.7%). Hardware complications were common but similar across groups, with screw pullout more frequent in ACR. Kaplan-Meier analysis of revision-free survival up to 50 months showed no significant differences among groups (p = 0.478), with the earliest reoperations occurring in Ponte, followed by ACR and PSO. Conclusion: PSO achieved the greatest sagittal correction, while IDO and ACR provided intermediate correction. IDO demonstrated the highest revision-free survival, with the earliest reoperations occurring in Ponte, followed by ACR and PSO. These findings highlight the importance of osteotomy selection in optimizing long-term durability and maintaining alignment.

ID: AGS26/1259

A148: Is crossing the thoracolumbar junction necessary? Reevaluating upper instrumented vertebra selection to minimize proximal junctional complications

Atahan Durbas 1 , Francis Lovecchio 1 , Tejas Subramanian 1 , Renaud Lafage 2 , Christopher Ames 3 , Shay Bess 4 , Robert Eastlack 5 , Munish Gupta 6 , Richard Hostin 7 , Khaled Kebaish 8 , Eric Klineberg 9 , Gregory Mundis 10 , David Okonkwo 11 , Frank Schwab 2 , Christopher Shaffrey 12 , Justin Smith 13 , Virginie Lafage 2 , Han Jo Kim 1 , Andrea Pezzi 1

1 Hospital for Special Surgery, New York, United States

2 Lenox Hill Hospital, New York, United States

3 University of San Francisco, San Francisco, United States

4 Denver International Spine Center, Denver, United States

5 Scripps Clinic, La Jolla, United States , 6 Washington University in St Louis, St. Louis, United States

7 Southwest Scoliosis Institute, Plano, United States

8 Johns Hopkins University School of Medicine, Baltimore, United States

9 UTHealth McGovern Medical School, Houston, United States

10 San Diego Spine, La Jolla, United States

11 University of Pittsburgh Medical Center, Pittsburgh, United States

12 Duke University, Durham, United States

13 University of Virginia Health System, Charlottesville, United States

Introduction: The choice of upper instrumented vertebra (UIV) in adult spinal deformity (ASD) surgery remains controversial, particularly regarding whether constructs should extend above the thoracolumbar junction (TLJ). Proponents of crossing and reaching above the TLJ suggest that this reduces proximal junctional kyphosis (PJK) and associated mechanical complications, though supportive evidence is limited. Furthermore, traditional PJK definitions may not clearly distinguish between physiological adaptation and true pathological changes at the TLJ. This study aimed to evaluate whether extending fusion across the TLJ reduces proximal junctional pathology compared with stopping below it. Material and Methods: We conducted a retrospective analysis of ASD patients undergoing instrumented fusion from either T11/T12 (above the TLJ, ATLJ) or L1/L2 (below the TLJ, BTLJ) to the sacrum or pelvis, with a follow-up period of at least 2 years. Exclusion criteria included prior fusions with UIV above L3 or using prophylactic tethers in their index surgery. Primary outcomes were proximal junctional complications: PJK (proximal junctional angle [PJA] ≥ 10°/ΔPJA ≥ 10°), proximal junctional failure (PJF; PJA ≥ 28°/ΔPJA ≥ 22° or proximal extension of fusion), and abnormal PJA (≥ 2 SD from normative UIV-specific means). Propensity score matching (PSM) was performed based on demographics, baseline alignment, and L1 pelvic angle correction. Results: A total of 131 patients (88 females, 43 males; 84 ATLJ, 47 BTLJ) with a mean age of 65.3 ( ± 9.7) years, BMI of 28.8 ( ± 5.0) kg/m 2 , and mean PI of 57.8° ( ± 11.4) were included, with an average follow-up of 2.3 years. In the ATLJ group (n = 84), UIVs were T11 in 86% and T12 in 14%. In the BTLJ group (n = 47), UIVs were L1 in 46% and L2 in 54%. The rate of PJK was higher in the ATLJ group compared to BTLJ (67% vs. 32%, p < 0.001), while the rates of PJF (20% vs. 15%, p = 0.4) and abnormal PJA (45% vs. 51%, p = 0.5) did not differ. After PSM, each group consisted of 34 patients, with a well-balanced distribution (ATLJ: T11/T12, 88%/12%; BTLJ: L1/L2, 56%/44%). Postoperative alignment parameters were largely comparable, although BTLJ patients demonstrated slightly greater PI-LL mismatch (2° vs. 8°, p = 0.001) and less kyphotic T10-L2 segment (13° vs. 6°, p = 0.04). ATLJ patients continued to have higher PJK rates (68% vs. 32%, p = 0.004), whereas PJF (21% vs. 12%, p = 0.3) and abnormal PJA (31% vs. 47%, p = 0.6) remained similar. Stratified analysis by specific UIV confirmed no differences in PJF (p = 0.55) or abnormal PJA (p = 0.34) between groups. Conclusion: Crossing the TLJ does not reduce proximal junctional complications in ASD surgery. Fusing to the pelvis with a UIV of L1 or L2 is an acceptable strategy in ASD surgery, as the actual rates of proximal junctional pathology are similar to those observed when constructs cross the TLJ. The higher rate of PJK observed in ATLJ patients may reflect limitations of current PJK definitions in differentiating between pathological and mechanical changes at the TLJ, rather than indicating a true clinical disadvantage.

ID: AGS26/1318

A149: How imaging speaks: the role of supine versus standing radiology in adult degenerative deformity surgery

Buse Sarigul 1 , Sedat Dalbayrak 2 , Mesut Yilmaz 2

1 Umraniye Research and Training Hospital, Department of Neurosurgery, Istanbul, Turkey

2 Medicana Atasehir Hospital, Department of Neurosurgery, Norospinal Akademi, Istanbul, Turkey

Introduction: With an aging population, the incidence of degenerative spinal deformities is increasing. Surgical treatment in these patients remains challenging due to high complication and revision rates, advanced age, and comorbidities. Accurate radiological evaluation, in correlation with clinical findings, is crucial for determining surgical levels, the necessity of osteotomy, and the type of osteotomy to be performed. This study aimed to investigate the impact of standing versus supine imaging on surgical planning in degenerative deformity. Material and Methods: A retrospective review was conducted on 86 consecutive patients who were diagnosed with degenerative spinal deformity. Preoperative imaging included standing full-spine scoliosis radiographs (AP/lateral), dynamic radiographs, MRI, and CT with AP/lateral scanograms. All images were analyzed using RadiAnt DICOM Viewer. Parameters measured included spinopelvic parameters (PI, PT, SS), coronal Cobb angle, lumbar lordosis (LL), thoracic kyphosis (TK), sagittal vertical axis (SVA) and PI-LL mismatch. Vertebral rotation was graded by Nash-Moe. Measurements obtained from standing radiographs were compared with those from supine MRI/CT. Their effect on fusion length and osteotomy type (segmental vs. major) was assessed. Logistic regression analysis was performed to identify predictors of major osteotomy requirement. Results: 86 patients were included in the study. 82 were female and 4 were male. The mean age was 68.71 ± 7.47 years. The mean coronal Cobb angle was 32.57 ± 9.83° on standing radiographs and 20.41 ± 8.72° on supine imaging (p < 0.001). Lumbar lordosis measured 23.34 ± 11.62° in standing versus 32.21 ± 10.87° in supine position (p < 0.001). PI-LL mismatch improved significantly from 16.83 ± 10.24° standing to 9.37 ± 8.61° supine (p < 0.001). SVA reduced from 58.23 ± 34.71 mm to 34.46 ± 29.13 mm (p = 0.004). Vertebral rotation regressed ≥ 1 grade in 28 patients (32.56%). Thoracic kyphosis did not differ significantly. Conclusion: Combined assessment of standing and supine imaging provides a more accurate understanding of the true deformity load and its clinical correlation in degenerative spinal deformity surgery. This approach enables more precise surgical planning, avoids unnecessary long fusions and major osteotomies, and may help reduce complication and revision rates.

ID: AGS26/199

A150: Evaluating clinically meaningful outcomes in adult spinal deformity surgery: MCID analysis of functional and social improvements

Devender Singh 1 , Matthew Geck 1 , Eeric Truumees 1 , John Stokes 1

1 Ascension Texas Spine and Scoliosis, Austin, United States

Introduction: Adult spinal deformity (ASD) surgery aims to alleviate pain and improve physical function and quality of life. Given the complexity of the surgical intervention, it is essential to determine whether postoperative changes are clinically meaningful. The Minimum Clinically Important Difference (MCID) helps to assess the smallest change in a patient's outcome that is perceived as beneficial. We aim to calculate and evaluate the clinical significance of MCID in PROMIS Physical Function, Pain Interference, and Social Role outcomes among adult deformity patients undergoing spinal correction surgery. Measurements were taken preoperatively and at the 1-year postoperative mark. Our hypothesize was that a significant proportion of patients undergoing adult deformity surgery will achieve meaningful improvements in physical function and social role, but fewer will achieve clinically important improvements in pain interference. Material and Methods: We conducted an analysis of 84 adult deformity patients who underwent spinal correction surgery via the traditional posterior open approach. Patient-reported outcomes were collected preoperatively and at 1 year postoperatively, including PROMIS Physical Function (PF), Pain Interference (PI), Social Role (SR), and Scoliosis Research Society (SRS) scores. MCID was calculated using the 0.5 Standard Deviation (SD) method, which is a widely accepted distribution-based method. Results: The mean age of the cohort was 56.2 years, with an equal distribution of male and female patients. The MCID values for each outcome were: 3.21 for PF, 3.32 for PI, and 3.92 for SR. A total of 68.2% of patients achieved the MCID for PF, while only 6.8% for PI. Similarly, 68.2% of patients reached the MCID for SR, suggesting that improvements in physical function and social interaction were more pronounced than improvements in pain. Older patients (≥ 60 yrs.) were more likely to achieve clinically meaningful improvements in PF, with 75% of patients in this age group achieving MCID compared to 60% in patients younger than 60. This may be explained by the fact that older patients with advanced deformity or functional limitations might experience a more noticeable improvement in mobility following surgery, whereas younger patients may already have a higher baseline level of function and experience less perceived benefit from the intervention. Postoperative improvements in SR were strongly correlated with improvements in PF (r = 0.65, p < 0.05), indicating that improved mobility was linked to greater social participation. However, there was no significant correlation between changes in PI and PF, suggesting that pain management remains a challenge in this patient population. Conclusion: The majority of patients undergoing adult spinal deformity surgery experienced clinically meaningful improvements in PF and SR, with over two-thirds of patients achieving MCID for these outcomes. PI, however, showed limited improvement, as only 6.8% of patients achieved MCID. This suggests that while the surgery successfully restores mobility and improves social participation, postoperative pain management requires further attention. The results of this study align with existing literature that reports greater improvements in function and quality of life compared to pain outcomes in this patient population.

ID: AGS26/200

A151: Radiographic predictors of functional and pain outcomes in Scheuermann’s kyphosis: a ROC-based minimal clinically important difference (MCID) analysis

Devender Singh 1 , Matthew Geck 1 , John Stokes 1 , Eeric Truumees 1

1 Ascension Texas Spine and Scoliosis, Austin, United States

Introduction: Scheuermann's Kyphosis (SK) is a structural spinal deformity that often necessitates surgical correction to improve sagittal balance and alleviate pain and functional impairments. The primary goal of such interventions is to achieve meaningful improvements in patient-reported outcomes such as pain relief and functional ability. Radiographic parameters, including Pelvic Incidence-Lumbar Lordosis (PI-LL) mismatch, Thoracic Kyphosis (TK), Lumbar Lordosis (LL), and T1 Pelvic Angle (TPA), are critical in guiding surgical planning and evaluating the success of corrective surgery. However, the relationship between changes in these parameters and patient-reported outcomes remains variable, warranting further investigation into their clinical relevance. This study aimed to evaluate the clinical significance of key radiographic parameters in SK correction by calculating the Minimal Clinically Important Difference (MCID) using the Receiver Operating Characteristic (ROC) curve method. The analysis sought to determine the radiographic thresholds that best predict patient-reported improvements in function and pain, based on Oswestry Disability Index (ODI) and Visual Analog Scale (VAS) scores at 2 years postoperative follow up. Material and Methods: A retrospective analysis of 40 patients who underwent surgical correction for SK was conducted. Key radiographic parameters, including PI-LL mismatch, TK, LL, and TPA, were analyzed at the final follow-up. The MCID was calculated using the ROC curve method, with thresholds defined by the patient-reported outcomes of functional improvement (ODI) and pain relief (VAS). The Area Under the Curve (AUC) was used to assess the predictive power of each radiographic parameter. Comparisons were made with established literature standards to evaluate whether the cohort required more stringent corrections to achieve clinically meaningful outcomes. Results: The cohort consisted of 52.2% male with a mean age of 19.7 years, a mean BMI of 24.8, and Sanders Score of 7.23. The analysis revealed that PI-LL mismatch and TK were the strongest predictors of patient-reported improvements. For PI-LL mismatch, the ROC-based MCID threshold was 6.0°, more stringent than the commonly accepted threshold of ≤10° in the literature. The final PI-LL mismatch values ranged from -7.0° to 8.0° in the cohort. The AUC for PI-LL mismatch was 0.68 for ODI, indicating that it was a fair predictor of functional improvement. In contrast, its AUC for VAS was 0.54, suggesting it had poor predictive value for pain relief. TK demonstrated a significant association with pain relief, with an ROC-based MCID threshold of 0.40° and an AUC of 0.69 for VAS, making it a fair predictor of pain improvement. However, its predictive value for functional improvement was limited, as evidenced by an AUC of 0.51 for ODI. LL had ROC thresholds of 47.0° for ODI and 48.0° for VAS, with AUC values of 0.54 and 0.60, respectively, indicating poor predictive ability for both functional and pain improvements. Similarly, TPA showed limited predictive power, with ROC thresholds of -3.0° for ODI and 9.0° for VAS, and corresponding AUC values of 0.61 and 0.54. These results suggest that while LL and TPA are important for overall sagittal balance, their direct impact on patient-reported outcomes in this cohort was minimal. Conclusion: The ROC-based MCID analysis identified PI-LL mismatch and TK as the most clinically significant predictors of improvement in patients undergoing surgical correction for SK. Stricter corrections in PI-LL mismatch and TK were associated with better patient-reported outcomes, particularly in terms of functional improvement and pain relief, respectively. In contrast, LL and TPA, while essential for sagittal alignment, were poor predictors of meaningful clinical improvements. These findings suggest that surgical planning for SK should prioritize achieving tighter corrections in PI-LL mismatch and TK to maximize patient outcomes. Surgeons should focus on these parameters during preoperative planning and intraoperative correction to enhance patient outcomes, as LL and TPA showed limited predictive value for clinical improvement in this cohort.

ID: AGS26/2005

A152: Impact of pre-operative staged skull-femoral traction followed by spinal osteotomy and posterior instrumented fusion in severe rigid spine deformity: two-year prospective outcomes

Sharif Jonayed 1

1 National Institute of Traumatology & Orthopaedic Rehabilitation (NITOR), Dhaka, Bangladesh

Introduction: Severe rigid spinal deformities (Cobb angle > 90°, flexibility < 25%) remain among the most challenging conditions in spine surgery, with high risks of neurological injury, cardiopulmonary compromise, and suboptimal correction. Pre-operative staged skull-femoral traction (SFT) may facilitate gradual deformity mobilization and pulmonary optimization, potentially reducing the morbidity of subsequent spinal osteotomy and posterior instrumented spinal fusion (PSF). This study evaluates the radiographic, pulmonary, and functional outcomes of this staged approach at three-year follow-up. Material and Methods: Seventeen consecutive patients (11 females, 6 males; mean age 22.4 ± 6.1 years) with severe rigid scoliosis or kyphoscoliosis underwent staged SFT (mean duration 19.1 ± 4.6 days; mean traction weight 36.2% ± 6.1% body weight) followed by spinal osteotomy and PSF between 2021-2023. Minimum follow-up was 24 months (mean 26.8 ± 2.4). Outcomes included coronal and sagittal correction, pulmonary function (FVC % predicted), complications, and SRS-22 functional scores. Results: The mean pre-traction coronal Cobb angle of 116.9° ± 13.4° improved to 73.5° ± 11.2° after traction and further to 50.7° ± 9.6° post-fusion, representing an overall correction of 57%. Sagittal kyphosis decreased from 95.1° ± 15.1° to 55.6° ± 10.2° postoperatively (42% correction). Pulmonary function improved significantly, with mean FVC increasing from 57.6% ± 6.8% predicted preoperatively to 72.3% ± 6.1% at final follow-up (p < 0.05). Functional outcomes also improved, with SRS-22 scores rising from 2.9 ± 0.5 to 4.3 ± 0.4. Complications included two transient neurological deficits that resolved completely and three superficial pin-site infections, with no permanent morbidity or mortality. All patients reported improved balance, posture, and pain relief at follow-up. Conclusion: Pre-operative staged SFT followed by spinal osteotomy and PSF is a safe and effective strategy for severe rigid spinal deformities. This protocol achieved substantial coronal and sagittal correction, significant pulmonary improvement, and durable functional gains with low complication rates. The staged approach represents a valuable alternative to high-risk single-stage osteotomies in carefully selected patients.

Keywords: Rigid scoliosis, skull-femoral traction, spinal osteotomy, posterior spinal fusion, pulmonary outcomes

ID: AGS26/977

A153: Geriatric nutritional risk index as a predictor of postoperative outcomes in elderly patients undergoing spinal deformity surgery

Ataollah Shahbandi 1 , Kevin Wojcik 1 , Saman Shabani 1

1 Department of Neurosurgery, Medical College of Wisconsin, Milwaukee, United States

Introduction: The Geriatric Nutritional Risk Index (GNRI) is widely applied to evaluate malnutrition in older adults, yet its relevance in the context of spine surgery for deformity correction remains unclear. This study aimed to examine the relationship between nutritional status before surgery, as assessed by GNRI scores, and 30-day postoperative outcomes in geriatric patients undergoing spinal deformity surgery. Material and Methods: We conducted a retrospective cohort study using the 2008-2023 American College of Surgeons National Surgical Quality Improvement Program database. Patients aged ≥ 65 years who underwent corrective surgery for spinal deformity were retrieved using Current Procedural Terminology codes or International Classification of Diseases, 9th and 10th Revision. Participants were stratified based on GNRI score: normal (> 98), malnourished (92-98), and severely malnourished (< 92). Receiver operating characteristic (ROC) analyses and multivariable regression models were used to assess the independent association and predictive value of GNRI compared with the 5-item modified Frailty Index (mFI-5) and chronological age for perioperative outcomes, including length of hospital stay, non-home discharge, and 30-day postoperative complications. Results: A total of 2,186 patients were included: 1,705 (78%) had normal nutrition, 332 (15.2%) were malnourished, and 149 (6.8%) were severely malnourished. Malnourished (40.1%) and severely malnourished (68.5%) groups exhibited significantly higher rates of preoperative anemia (27.2% in the normal group, p < 0.001), lower body mass index (28.7 ± 6.3 and 28 ± 6.5 kg/m 2 , respectively, versus 29.7 ± 5.9 in normal group, p = 0.001), and lower albumin levels (3.6 ± 0.2 and 3 ± 0.4 g/dL, respectively, vs 4.2 ± 0.3 in normal cohort, p < 0.001) compared with the normal group. After adjusting for potential confounders, malnourished status was independently associated with significantly greater odds of wound disruption (0.6% versus 2.1%, odds ratio [OR] 3.3, p = 0.014), prolonged mechanical ventilation (4.8% vs 1.8%, OR 2.2, p = 0.019), postoperative transfusion (59% vs 46.7%, OR 1.5, p = 0.001), increased transfusion volume (coefficient 0.452, p = 0.044), and non-home discharge (59.9% vs 47.3%, OR 1.6, p < 0.001) compared to normal patients. Severe malnutrition independently correlated with significantly longer hospital stay (9.7 ± 8.4 versus 6.4 ± 4.5 days; coefficient 2.4, p < 0.001), greater likelihood of prolonged hospitalization (32.9% vs 14.9%, OR 2.1, p < 0.001), and higher rates of non-home discharge (60.4% vs 47.3%, OR 1.6, p = 0.017) compared to normal patients. In ROC analyses, GNRI demonstrated superior discriminatory ability compared with mFI-5 (p = 0.394, p = 0.645, p = 0.179, p = 0.113, p = 0.014, and p = 0.233 respectively) and age (p = 0.277, p = 0.074, p = 0.039, and p = 0.167, p < 0.001, and p < 0.001 respectively) for predicting wound disruption, superficial surgical site infection, prolonged mechanical ventilation, unplanned intubation, the need for transfusion, and prolonged hospitalization. Conclusion: Preoperative malnutrition, as defined by GNRI, is correlated with a greater risk of adverse 30-day outcomes following spinal deformity surgery in older adults. This finding highlights the importance of incorporating nutritional assessment into preoperative evaluation for spinal deformity surgery in the geriatric population.

OP18: Advanced MISS

ID: AGS26/1326

A154: Complications in cervical minimally invasive spine surgery: a proportional meta-analysis of tubular, uniportal, and biportal approaches

Chibuikem Ikwuegbuenyi 1 , Sean Inzerillo 1 , Eesha Gurav 1 , Noah Willett 1 , Mousa Hamad 1 , Galal Elsayed 1 , Osama Kashlan 1 , Roger Härtl 1

1 Department of Neurological Surgery, Och Spine at New York-Presbyterian/Weill Cornell Medical Center, New York, United States

Introduction: Minimally invasive spine surgery (MISS) for cervical pathologies has demonstrated superior clinical outcomes compared to the open approaches, but differences in complication rates among the tubular, uniportal, and biportal techniques remain unclear. This proportional meta-analysis compares complication rates across these techniques utilizing studies published in the last decade. Materials and Methods: Following PRISMA guidelines, we conducted a systematic review and meta-analysis (PROSPERO CRD42024594335). PubMed, Medline, Embase, and Cochrane Library were queried for cervical MISS studies (Jan 2013-Mar 2024) utilizing tubular, uniportal, or biportal approaches. Included studies reported complication rates in adult cohorts. Exclusions were abstracts, reviews, and non-English articles. A random-effects model was applied, and the I 2 statistic assessed heterogeneity. All analyses were conducted using R Studio. Results: Twenty-one studies (1299 patients) were included, with average patient ages ranging from 47 to 74.5 years and 64% male. All studies had low bias risk. Follow-up periods ranged from 3 to 33 months. The pooled complication rate for cervical MISS was 5% (95% CI [3%-7%]), with heterogeneity (I 2 = 59%). Subgroup analysis showed complication rates of 4% (95% CI [1%-10%], I 2 = 70%) for tubular, 6% (95% CI [2% -12%], I 2 = 46%) for uniportal, and 5% (95% CI [2%-8%], I 2 = 39%) for biportal. No statistically significant differences were found (p = 0.85). Nerve injury rates were higher with the uniportal approach (6%, 95% CI [2%-16%], p = 0.02). Dural tears (1%, 95% CI [0%-2%], I 2 = 0%) and postoperative hematomas (0%, 95% CI [0%-3%], I 2 = 0%) had low incidence, with no significant differences between approaches (p = 0.61 and p = 0.78, respectively). Conclusions: Cervical MISS demonstrates a low overall complication rate, with tubular approaches showing a numerically lower risk, though differences were not statistically significant. Larger comparative studies are needed to provide more definitive results for better clinical application.

ID: AGS26/1740

A155: Bipotal endoscopic tranforaminal lumbar interbody fusion (BE-TLIF): evaluation clinical and CT- based radiological outcome

Songwut Sirivitmaitree 1

1 Bandung Crown Prince Hospital, Udon, Thailand

Introduction: Biportal endoscopic spine surgery is a type of minimally invasive spine surgery that uses two small incisions to insert an endoscope and surgical instruments. Biportal endoscope provides various procedures, such as decompression, discectomy, and fusion. BE- TLIF is a surgical technique that combines the advantages of minimally invasive and endoscopic spine surgery. It is a type of spinal fusion surgery through a transforaminal approach that aims to stabilize the spine and relieve pain caused by various spinal disorders, such as degenerative disc disease, spondylolisthesis, and recurrent disc herniation. The advantage of BE- TLIF does not require a tubular retractor which may reduce muscle damage ,independent endoscopic and instrument ports were easily to contralateral decompression and reduce postoperative pain. Biportal endoscopic TLIF is a promising alternative to conventional open or minimally invasive TLIF, but it requires a learning curve and more studies to evaluate its long-term outcomes. Material and Methods: Retrospective cohort study. This study included 72 patients with 81 fusion segments who underwent BE-TLIF between January 2020 and March 2022. All of the patients were observed at least a 12-month follow-up. We collected demographic data, total operation time and length of hospital stay. Clinical outcomes measures included visual analogue scale (VAS) scores for lower back pain and leg pain, Oswestry Disability Index (ODI), Plain film x-ray 1, 3, 6,12 months postoperatively and then every year thereafter. Computed tomography (CT) of the lumbar spine 6 and12 months postoperatively. The fusion rate was assessed using the Bridwell interbody fusion grading system. Cage subsidence was classified into no subsidence, no more than 2 mm, and more than 2 mm, Postoperative complications were also noted. Results: There were significant improvement in VAS for lower back pain from 6.3 ± 0.8 to 1.2 ± 0.5, VAS for leg pain from 5.8 ± 1.3 to 1.2 ± 0.4, ODI from 58.1 ± 11.9 to 13.4 ± 4.6. The p-values were all < 0.001. The average hospital stay was 4.7±1.1 days. The CT studies available for 52 fusion segments. 6 months after surgery showed successful fusion (Bridwell grade I or grade II) in 33 segments (63.5%). 1 year after surgery showed successful fusion (Bridwell grade I or grade II) in 49 segments (94.2%). Significant cage subsidence of more than 2 mm was only noted in 46 segments (56.1%). Complications included 2 dural tear, 1 pedicle screws malposition, and 3 transient paresthesia, in which 1 patient required reoperations. Conclusions: BETLIF may have favorable clinical outcomes, high fusion rate and lower complication. BETLIF may be a good alternative treatment for lumbar spine disorder (stenosis, instability, and spondylolisthesis). Further studies with large samples and long-term follow-up are needed.

ID: AGS26/2583

A156: Quantitative radiological assessment of interlaminar and transforaminal endoscopic lumbar decompression for cauda equina syndrome: a 12-month prospective study

Joel Stoy 1 , Samira Akmal 2 , Santhosh G. Thavarajasingam 3 , Darren Lui 4 , Jin-Sung Kim 5 , Salvatore Russo 2

1 Imperial College School of Medicine, London, United Kingdom

2 Imperial College Healthcare NHS Trust, London, United Kingdom

3 Ludwig-Maximilians-Universität München, Munich, Germany

4 St. George’s University Hospitals NHS Foundation Trust, London, United Kingdom

5 The Catholic University of Korea, Seoul, South Korea

Introduction: Radiological outcomes in cauda equina syndrome (CES) are mostly reported after open laminectomy, with limited data on minimally invasive approaches. Endoscopic lumbar decompression minimizes collateral tissue damage, and we hypothesize that it achieves adequate neural decompression, potentially preserving spinal stability and alignment. No prospective study has quantitatively assessed radiographic outcomes after endoscopic decompression in CES; this study evaluates them over a 12-month period. Material and Methods: We conducted a prospective, single-center cohort study at Imperial College Healthcare NHS Trust between November 2023 and September 2025. We enrolled 21 adults with clinically and radiologically confirmed CES, who underwent endoscopic lumbar decompression. Serial MRIs were obtained at baseline, immediately postoperatively, and at 3, 6, and 12 months. All imaging was independently reviewed by two fellowship-trained spinal surgeons. Assessed parameters included surgical approach, volume of canal decompression, Pfirrmann disc grade, Modic changes, sacral and pelvic angles, extent of facet resection and radiographic complications. Continuous variables were assessed by paired t-tests or Wilcoxon signed-rank tests, and categorical outcomes by chi-square or Fisher’s exact tests. Results: The cohort had a mean age of 47.7 years (9 males, 12 females). Index levels were L4/5 in 10 (47.6%), L5/S1 in 9 (42.9%), and other levels in 2 (9.5%), with L5/S1 predominantly via the interlaminar approach (8/9) and L4/5 via the transforaminal approach (6/10). Complications occurred in 2 patients (9.5%): one postoperative hematoma and one recurrent disc herniation. In the interlaminar group (n = 11), DSCA increased from 31.5 ± 34.8 mm 2 preoperatively to 132.7 ± 49.8 mm 2 at latest follow-up (Wilcoxon Z = -2.936, p = 0.003; Cohen’s d = 2.47), while in the transforaminal group (n = 10), DSCA rose from 13.4 ± 22.6 mm 2 preoperatively to 138.5 ± 41.4 mm 2 at latest follow-up (Wilcoxon Z = -2.803, p = 0.005; Cohen’s d = 4.12). Radiological success (DSCA > 100 mm 2 ) was achieved in 8/11 (72.7%) interlaminar and 9/10 (90.0%) transforaminal patients; > 50% DSCA increase occurred in 9/11 (81.8%) and 10/10 (100%). Median disc degeneration was Pfirrmann IV, with Modic type 2 changes in interlaminar and type 3 in transforaminal groups. Mean disc height was 6.8 ± 2.3 mm anteriorly and 4.3 ± 2.0 mm posteriorly (interlaminar) versus 6.2 ± 2.4 mm and 4.7 ± 1.4 mm (transforaminal). Spinal alignment was preserved (interlaminar, transforaminal): segmental kyphosis (27.8 ± 9.8°, 12.6 ± 6.8°), lumbar lordosis (35.0 ± 9.5°, 29.6 ± 12.1°), sacral slope (27.8 ± 7.8°, 28.6 ± 7.6°), pelvic incidence (56.3 ± 10.2°, 71.5 ± 5.4°), and pelvic tilt (23.0 ± 6.4°, 35.0 ± 6.8°). In the transforaminal group, 8/10 required facet resection (70% complete, 60% partial pediculectomy), and mean laminectomy width was 11.1 ± 7.6 mm. Conclusion: Both interlaminar and transforaminal endoscopic approaches were effective over 12 months, with outcomes comparable or superior to those reported for standard laminectomy in the Preliminary findings from this pilot study suggest the transforaminal approach may achieve greater canal expansion, though interpretation is limited by the small, single-center cohort.

ID: AGS26/2613

A157: To evaluate the efficacy of spinal anesthesia in patients undergoing minimal invasive single level transforaminal lumbar interbody fusion (MIS TLIF) surgery and to compare the results with that of general anesthesia

Vishal Kundnani 1 , Maitreya Patil 1,2

1 Bombay Hospital and Medical Research Center, Lilavati Hospital and Medical Research Center, Guru Nanak Hospital, Mumbai, India

2 Bombay Hospital and Medical Research Center, Lilavati Hospital and Medical Research Center, Guru Nanak Hospital and Medical Research Center, Mumbai, India

Introduction: To evaluate the efficacy of spinal anesthesia in patients undergoing minimal invasive single level transforaminal lumbar interbody fusion (MIS TLIF) surgery and to compare the results with that of general anesthesia. Material and Methods: 178 patients were included in the study, 86 were in general anesthesia and 92 were in spinal anesthesia. Patients aged between 20 to 70 years who had undergone MIS TLIF not responding to 6 weeks of conservative treatment were included. The routine steps of anesthesia for both general and spinal anesthesia were adhered. The visual analogue scale, blood loss, duration of surgery, time from entering operation theatre to time of incision, time of bandaging to exit from operation theatre, time of stay in Post Anesthesia Care Unit (PACU), nausea/vomiting, urinary retention, duration of stay in hospital, peri-operative complications were compiled and assessed. Appropriate statistical analysis was applied. Results: The mean time for entering the operation theatre to the incision; mean time from bandaging to the exit; mean PACU time and the mean hospital stay were significantly lower in the spinal anesthesia group (p < 0.05). The other parameters are comparable except, urinary retention which was significantly higher in spinal anesthesia group (p < 0.05). Conclusion: Spinal anesthesia offers efficient operating room functioning with decreasing overall operation theatre time. It is very efficient alternative technique to general anesthesia which can be considered for elective lumbar surgeries with a lower late of adverse events especially at lower lumbar levels.

ID: AGS26/3064

A158: Endoscopic resection of lumbar intradural spinal tumors: a case series on feasibility, safety, and preliminary outcomes

Facundo Van Isseldyk 1,2 , Vincent Hagel 3

1 Hospital Privado de Rosario, Rosario, Argentina, 2 Endospine Academy, Zurich, Switzerland, 3 Asklepios Hospital Lindau, Spine Center, Lindau, Germany

Introduction: Intradural spinal tumors represent 5-10% of spinal neoplasms and are typically managed with open microsurgical resection, which achieves high rates of gross total resection (GTR) but is associated with substantial morbidity due to extensive muscle dissection and bony removal. Endoscopic spine surgery has proven effective for degenerative pathologies by minimizing tissue disruption, yet its application to intradural lesions remains rare. The technical challenges of dural opening, irrigation management, tumor manipulation within a constrained corridor, and dural closure have limited widespread adoption. This study reports the feasibility, safety, and preliminary outcomes of a fully endoscopic approach for lumbar intradural tumor resection. Material and Methods: We retrospectively reviewed 10 consecutive adult patients who underwent endoscopic resection of lumbar intradural-extramedullary tumors between 2020 and 2024. Inclusion criteria were lumbar intradural-extramedullary tumors confirmed on MRI; exclusion criteria included intramedullary lesions, multisegmental disease requiring multilevel laminectomy, or non-lumbar tumors. Demographic, radiologic, surgical, and clinical data were analyzed. Surgical technique involved a uniportal interlaminar approach with continuous irrigation, endoscopic dural opening, tumor dissection, and dural closure by either suture or patch. Outcomes included extent of resection on postoperative MRI, neurological status via modified McCormick scale, complications, hospital stay, and follow-up imaging. Results: Ten patients (6 women, 4 men; mean age 58.3 years, range 46-72) were included. Tumor types were schwannoma (n = 7), meningioma (n = 2), and neurofibroma (n = 1), all located between L2-L5 (mean size 24.5 mm, range 10-38). Presenting symptoms included radicular pain (n = 6), neurogenic claudication (n = 2), and motor deficits (n = 2). Mean operative time was 96.5 ± 19.7 minutes, with no intraoperative complications. Dural closure was performed with endoscopic suturing in 3 patients and patch placement in 7. GTR was confirmed in all cases. No new neurological deficits or cerebrospinal fluid-related complications occurred. Modified McCormick grades improved in 6 patients (60%) and remained stable in 4 (40%); none worsened. Both patients with preoperative motor deficits improved postoperatively. Hospital stay averaged 1.5 ± 0.5 days. Statistical analysis confirmed a significant improvement in McCormick grades (p = 0.014). Subgroup analysis revealed that patients with sutured closure achieved grade 1 function at follow-up, though operative times were slightly longer. No recurrences were observed during a mean follow-up of 15 months (range 12-18). Conclusion: Endoscopic resection of lumbar intradural-extramedullary tumors is feasible and safe in selected patients, achieving 100% GTR and favorable neurological outcomes without added morbidity. The minimally invasive nature of this technique likely contributed to the short hospital stays and rapid recovery observed. Key technical considerations include irrigation pressure modulation, controlled dural opening, and tailored closure strategies, both of which were effective in preventing CSF-related complications. These findings suggest that endoscopy can extend beyond degenerative pathology into intradural tumor surgery without compromising oncological or functional goals. However, the approach is technically demanding and best performed by surgeons with extensive endoscopic experience. Larger, prospective studies with long-term follow-up are warranted to validate these results and further define the role of endoscopy in intradural tumor management.

ID: AGS26/1579

A159: Unilateral biportal endoscopic with modified arcocristectomy for pinching cervical spondylotic myelopathy: surgical technique and early experiences

Qiang Yang 1,2 Xiaopeng Li 1,2,3

1 Capital Medical University Affiliated Beijing Friendship Hospital, Beijing, China

2 Tianjin Hospital, Tianjin, China,

3 Weifang People’s Hospital, Weifang, China

Introduction: Introduce the surgical technique and early clinical efficacy of the modified arcocristectomy for treating pinching cervical spondylotic myelopathy using unilateral approach for bilateral decompression under cervical unilateral biportal endoscopy (UBE)surgery. This technique is being introduced for the first time and is applied for the first time in the treatment of pinching cervical spondylotic myelopathy. Material and Methods: This study enrolled nine patients with pinching cervical spondylotic myelopathy who underwent modified Arcocristectomy with UBE for unilateral approach and bilateral decompression at our medical center between December 2023 and November 2024. Patients’ demographic data, imaging findings, and perioperative data were collected. Neurological recovery was assessed using the Japanese Orthopaedic Association (JOA) score and the Neck Disability Index (NDI). Imaging parameters, including cervical lordosis, segmental cervical angle at the surgical level, and maximum sagittal diameter of the spinal canal, were measured preoperatively, on the first postoperative day, and at the final followup. The safety and efficacy of the surgery were evaluated by comparing and analyzing these indicators. Results: The study included six men and three women, with an age range of 60-84 years (mean 70.75 ± 7.34 years). The mean preoperative symptom duration was 14.13 ± 16.57 months, and the mean follow-up duration was 14.00 ± 5.13 months. The mean preoperative Japanese Orthopaedic Association (JOA) score was 9.25 ± 3.28, which improved to 12.00 ± 3.21 on the first postoperative day, representing a mean improvement rate of 39.93 ± 15.07%. At the final follow-up, the mean JOA score was 14.38 ± 2.13, with a mean improvement rate of 69.85 ± 12.74%. The mean Neck Disability Index (NDI) score was 42.25 ± 15.00% preoperatively and decreased to 15.50 ± 5.90% at the final follow-up. Imaging results showed that the mean preoperative cervical lordosis was 19.04 ± 8.18°, which decreased to 4.91 ± 5.59° postoperatively and returned to 19.31 ± 7.82° at the final follow-up. The mean preoperative segmental cervical angle at the surgical level was 6.90 ± 3.82°, which decreased to 1.85 ± 2.08° postoperatively and returned to 7.05 ± 3.36° at the final follow-up. The mean preoperative maximum sagittal diameter of the spinal canal at the surgical level was 5.71 ± 2.03 mm, which increased to 11.98 ± 1.91 mm postoperatively. Intraoperatively, the mean anesthesia duration was 161.50 ± 47.11 min, and the mean surgical time was 108.13 ± 47.88 min. Blood loss was minimal. One patient experienced dizziness on the first postoperative day. Conclusion: The innovation of the modified Arcocristectomy surgery under cervical UBE for treating pinching cervical spondylotic myelopathy lies in the extended resection of the posterior-superior margin of the lamina. This modification effectively alleviates spinal cord compression caused by the “pinching effect” during cervical extension. Short-term follow-up results demonstrate significant clinical improvements, offering a new minimally invasive treatment option for this condition.

ID: AGS26/352

A160: Unilateral biportal endoscopic decompression combined with percutaneous pedicle screw fixation for burst fractures

Chun Tseng 1,2,3 , SiHuai Wang 1 , Chien-Min Chen 4 , Hsien-Te Chen 1,2,3

1 Department of Orthopaedic Surgery, China Medical University Hospital, Taichung, Taiwan

2 Spine Center, China Medical University Hospital, Taichung, Taiwan

3 Department of Sports Medicine, College of Health Care, China Medical University, Taichung, Taiwan

4 Division of Neurosurgery, Department of Surgery, Changhua Christian Hospital,, Changhua, Taiwan

Introduction: Thoracolumbar burst fractures often lead to neural compression and spinal instability, traditionally treated with open decompression and fixation. Although effective, open surgery is associated with significant soft tissue trauma, blood loss, and prolonged recovery. Minimally invasive approaches, particularly Unilateral Biportal Endoscopic (UBE) decompression combined with Percutaneous Pedicle Screw (PPS) fixation, have emerged as alternatives that aim to achieve effective decompression and stabilization while minimizing morbidity. This study evaluated the clinical and radiological outcomes of UBE with PPS fixation in patients with burst fractures. Material and Methods: Patients with thoracolumbar burst fractures (T11-L2) presenting with neurological deficits (ASIA B-D), canal compromise > 30%, Cobb angle >15°, and vertebral body height (VBH) loss > 40% were included. Exclusion criteria were pathological fractures, multilevel trauma, or follow-up < 12 months. Surgical treatment consisted of UBE decompression for removal of retropulsed fragments combined with short-segment PPS fixation. Clinical outcomes were assessed with ASIA grade, Visual Analog Scale (VAS), and Oswestry Disability Index (ODI). Radiographic evaluation included Cobb angle, VBH ratio, and canal compromise. Parameters were recorded preoperatively, immediately postoperatively, and at follow-up. Results: Twelve patients were analyzed. • Neurological Recovery: ASIA grade improved in 91.7% of cases, including C→E (n = 5), D→E (n = 5), and C→D (n = 2). • Pain and Function: Postoperative pain and functional status improved significantly. VAS decreased markedly, and ODI scores improved, with most patients shifting from severe to moderate or mild disability. • Radiographic Improvements: The mean Cobb angle improved from 19.5° pre-op to 7.2° post-op, corresponding to ∼63% kyphosis correction. The mean VBH ratio increased from 42.1% to 80.8%, with an average restoration gain of 38.7%; most patients achieved > 90% height recovery compared to adjacent vertebrae. • Complications: The complication rate was low (< 5%), limited to minor implant-related issues, without permanent neurological deterioration or infection. Conclusion: UBE decompression combined with PPS fixation is a safe and effective minimally invasive strategy for thoracolumbar burst fractures. This approach achieves significant neurological recovery, pain relief, kyphosis correction, and vertebral height restoration while minimizing surgical morbidity. These results support UBE+PPS as a viable alternative to conventional open surgery in selected patients with thoracolumbar burst fractures.

ID: AGS26/545

A161: Comparison of uniportal endoscopy (UE) versus unilateral biportal endoscopy (UBE) in the treatment of severe upward-migrated lumbar disc herniation: a multicenter study

Chongwen Wang 1 , Zhi Zhang 1 , Yuanbin Liu 2 , Gang Wu 3 , Yongan Ren 4 , Yi Yuan 5 , Congdi Liu 1 , Nai Mu 1 , Shifu Bao 1

1 Department of Spine Surgery, Chengdu Fifth People’s Hospital, Chengdu, China

2 Spine Center, Suining First People's Hospital, Suining, China

3 Department of Spine Surgery, Dazhou Orthopedic Hospital, Dazhou, China

4 Department of Spine Surgery, Xichang People's Hospital, Xichang, China , 5 Department of Spine Surgery, Dazhou Second People's Hospital, Dazhou, China

Introduction: The incidence of migration nucleus pulposus in patients with lumbar disc herniation ranges from 35% to 72% of which 34% are severe upward-migrated. Currently, percutaneous endoscopic lumbar discectomy (PELD) for severe upward-migrated lumbar disc herniation (SUM-LDH) has the advantages of being minimally invasive, safe, and effective. This study compared the clinical efficacy of uniportal endoscopy(UE) and unilateral biportal endoscopy (UBE) for the treatment of SUM-LDH. Material and Methods: A retrospective analysis was performed on 109 patients with SUM-LDH admitted to five spine centers between May 2019 and June 2022. The UBE group (53 patients, 28 males and 25 females, aged 53.2 ± 7.6 years). The UE group (56 patients, 32 males and 24 females, aged 51.9 ± 9.3 years). Both groups via the contralateral sublaminar approach. The operative time, length of stay(LOS), times of intraoperative fluoroscopy, and postoperative nucleus pulposus residue were compared between the two groups. The visual analogue scale (VAS) scores for low back and leg pain, Oswestry disability index (ODI) were compared before surgery and 1, 3, 6, 12, and 24 months after surgery. The modified Macnab classification were used to evaluate the efficacy at the last follow-up. Results: The follow-up time was 28.6±4.3 months. All operations were successfully completed. The operative time and LOS in the UBE group were (69.3 ± 7.4) minutes and (4.8 ± 1.3) days, respectively; in the UE group, these were (67.9 ± 9.5) minutes and (5.1 ± 1.5) days, respectively, with no statistically significant difference (t = 1.64, -0.75, p > 0.05). The times of intraoperative fluoroscopy in the UBE group was (7.5 ± 3.1) fewer than in the UE group (16.1 ± 5.3), with a statistically significant difference (t = -9.14, p < 0.01). In the UBE group, no nucleus pulposus residue, and one patient experienced transient radiating pain in the affected lower limb after surgery, which was resolved after oral gabapentin. In the UE group, one patient experienced cerebrospinal fluid leakage, and three patients had nucleus pulposus residue. Compared with preoperatively, the VAS scores of low back and leg pain, ODI in both groups were significantly improved after surgery (p < 0.01). However, there was no significant difference in the VAS scores of low back and leg pain, ODI between the two groups (p > 0.05). According to the modified Macnab classification evaluation criteria: in the UBE and UE groups were 93.9% and 91.6%, respectively, with no statistically significant difference (χ 2 = 0.31, p = 0.59). Conclusion: For patients with SUM-LDH, both UE and UBE are effective in relieving pain symptoms. Compared with UE, UBE can removal of the nucleus pulposus under direct vision. Furthermore, it can reduce the probability of residual nucleus pulposus and reduce the times of intraoperative fluoroscopy.

ID: AGS26/553

A162: Outcomes of posterior endoscopic cervical discectomy: a retrospective study of 30 patients

Atif Ali 1 , Muhammad Farrukh Bashir 1

1 Ghurki Trust Teaching Hospital Lahore, Lahore, Pakistan

Introduction: Cervical disc herniation is one of the major causes of cervical radiculopathy and disability. Traditional surgical options such as anterior cervical discectomy and fusion (ACDF) or open discectomy are effective but may lead to complications including fusion-related morbidity, dysphagia, and adjacent segment disease. Posterior Endoscopic Cervical Discectomy (PECD) has recently emerged as a minimally invasive, motion-preserving alternative that aims to reduce these risks while maintaining clinical efficacy. Materials and Methods: This retrospective study included 30 patients (18 males, 12 females; mean age 42.3 ± 8.7 years) diagnosed with single-level cervical disc herniation confirmed by MRI. All patients had persistent radicular pain or neurological symptoms for at least six weeks despite conservative treatment, with clinical findings correlating to imaging. Patients with multilevel disc disease, prior cervical surgery, instability, deformity, infection, trauma, tumor, or significant medical comorbidities were excluded. All underwent PECD, and outcomes were assessed using the Visual Analogue Scale (VAS) for pain, Neck Disability Index (NDI), operative details, complications, and overall satisfaction. Statistical analysis was performed using SPSS v27. Results: At 6-12 months follow-up, patients showed marked clinical improvement. The mean VAS score improved from 7.8 ± 1.1 preoperatively to 2.1 ± 0.9 postoperatively (p < 0.001), while the mean NDI decreased from 61.5 ± 7.8% to 18.4 ± 5.6% (p < 0.001). The average operative time was 82 ± 15 minutes, with minimal blood loss (<50 ml). No major intraoperative or postoperative complications were observed. Two patients (6.7%) experienced transient postoperative dysesthesia, which resolved with conservative management. Patient satisfaction was high at 93%, and no reoperations were required during the follow-up period. Conclusion: Posterior endoscopic cervical discectomy is a safe, effective, and motion-preserving technique for managing single-level cervical disc herniation. It provides significant pain relief, functional improvement, and high patient satisfaction while avoiding the risks associated with fusion procedures. Given its excellent short-term outcomes and low complication rates, PECD represents a valuable alternative to conventional ACDF in appropriately selected patients. Larger multicenter studies with longer follow-up are recommended to confirm the long-term durability of these results.

OP19: State of the Art of Arthroplasty

ID: AGS26/1399

A163: TDR versus fusion - Live systematic review and metanalysis

Igor Barreira 1 , Rafael Moriguchi 1 , Rodrigo Amaral 1 , Luiz Pimenta 1 , Fernando Marcelino 1 , Igor Arthur Costa 1 , Amabile Galeti 1 , Leandro Resende 1 , Jullyene Pokorny 1 , Gabriel Pokorny 1 , Bianca Soares 1

1 Instituto de Patologia da Coluna, São Paulo, Brazil

Introduction: Degenerative lumbar spine disease is a leading cause of chronic low-back pain and socioeconomic burden. Lumbar fusion is the long-standing surgical standard, but loss of motion and adjacent-segment degeneration have fuelled interest in total disc replacement (TDR). We performed a living systematic review and meta-analysis to compare the efficacy, safety and functional outcomes of TDR versus fusion in adults with degenerative lumbar disease. Material and Methods: Monthly searches of PubMed, Embase, CENTRAL, ClinicalTrials.gov and WHO ICTRP (inception to March 2024) identified comparative studies of TDR and fusion reporting clinical or radiological outcomes. Two reviewers independently screened records extracted data and assessed risk of bias (Cochrane RoB-2 for RCTs; Newcastle -Ottawa Scale for cohorts). Pooled odds ratios (OR) for binary outcomes and mean differences (MD) for continuous outcomes were calculated with random-effects models; heterogeneity was explored with I 2 statistics. The review protocol was registered in PROSPERO and set up as a “living” review with quarterly updates. Results: From 1,184 records, 18 studies (six RCTs, 12 cohorts) encompassing 58,447 patients (6 274 TDR, 52 173 fusion) met the criteria. TDR significantly reduced overall complications versus fusion (OR 0.46, 95 % CI 0.29 -0.72; I 2 = 80 %). Pain relief favoured TDR (VAS-back MD -0.93, 95 % CI -1.52 to -0.35; I 2 = 58%). Disability scores (Oswestry Disability Index) did not differ (MD -3.3, 95 % CI -8.83 to 2.11; I 2 = 74 %). Sensitivity and leave-one-out analyses confirmed robustness of the complication and pain findings; evidence certainty was graded “very low” to “low” owing to heterogeneity and study design limitations. Conclusion: Current evidence suggests that lumbar TDR offers comparable functional recovery to fusion while providing modestly better pain relief and a lower incidence of peri-operative and device-related complications. Nevertheless, heterogeneity and predominantly observational data temper confidence. High-quality randomised trials and continuous evidence surveillance via this living review are needed to refine patient selection and inform guideline recommendations.

ID: AGS26/1438

A164: Cervical disc arthroplasty versus posterior cervical foraminotomy as motion preserving surgeries for cervical radiculopathy: a systematic review and meta-analysis

Jia Yi Loh 1 , Xian Jun Ngoh 1 , Zhihong Chew 2 , Yee Gen Lim 1 , Lei Jiang 1

1 Department of Orthopaedic Surgery, Singapore General Hospital, Singapore, Singapore

2 Department of Orthopaedic Surgery, Changi General Hospital, Singapore, Singapore

Introduction: Motion preserving surgery for cervical radiculopathy treats lateral and foraminal disc pathology while causing minimal disruption to the stability of the cervical spine and include cervical disc arthroplasty (CDA) and minimally invasive posterior cervical foraminotomy (MI-PCF) including endoscopic PCF (PECF). This study aims to compare the functional and surgical outcomes between CDA and MI-PCF. Material and Methods: This study includes a systematic review of present literature and meta-analysis comparing motion preserving surgeries of CDA and MI-PCF for treatment of cervical radiculopathy. Results: A total of 575 abstracts were reviewed and assessed and 7 comparative studies between MI-PCF and CDA were included. The mean operative time for patients undergoing CDA was between 38.7 to 110.1 min, while the mean operative time for patients undergoing MI-PCF was between 35.9 to 108.3 min, with MI-PCF being performed 3.4 min faster than CDA. The overall complications rate was 5.4% for CDA and 3.4% for MI-PCF and the overall reoperation rate was 0.9% for CDA and 3.8% for PCF. A random effects model meta-analysis was carried out which showed no significant differences in complication rate but MI-PCF had significantly higher reoperation rates compared to CDA. Patients who underwent PCF had non-statistically significant improvement in functional outcomes of Neck Disability Index (NDI) and Visual Analogue Scale (VAS) neck and arm scores compared to patients who underwent CDA at 6 months postoperatively. Subsequently, patients who underwent MI-PCF had further improvement in NDI and VAS neck and arm scores 1 year post-operatively, with statistically significant improvements in scores compared to patients who underwent CDA. Subgroup analysis of functional outcomes comparing CDA and PECF showed non-statistically significant improvement in VAS scores but poorer NDI scores for patients undergoing PECF at 6 months post operation. Conclusion: MI-PCF has lower complication rates and better short term functional outcomes compared to CDA despite also having higher reoperation rates. Further studies with longer follow up correlating functional outcomes with radiological parameters measuring cervical motion is required to establish its superiority over CDA.

ID: AGS26/1479

A165: Cervical disc arthroplasty in congenital cervical spinal stenosis: outcomes compared to non-stenotic patients

Hannah Shelby 1 , Kiran Alluri 1 , Joseph Shelby 1 , Tara Shelby 2

1 Department of Orthopaedic Surgery, Keck School of Medicine of University of Southern California, Los Angeles, CA, Los Angeles, United States

2 Stanford University School of Medicine, Stanford, United States

Introduction: Cervical spinal stenosis is the narrowing of the spinal canal that can result in cervical nerve root compression and cause significant radicular pain, numbness, and weakness and in more severe cases can progress to cervical spondylitis myelopathy (CSM). Congenital cervical spinal stenosis (CCSS), caused by abnormal development of the pedicle lamina or vertebral column, predisposes patients to symptomatic compression due to the reduced baseline canal diameter. While anterior cervical discectomy and fusion (ACDF) is the most common surgical treatment, it eliminates motion and increases the risk of adjacent-segment degeneration. Cervical disc arthroplasty (CDA) is a motion-preserving alternative with proven safety and effectiveness, but its outcomes in patients with CCSS remain poorly defined. This study examines whether CDA provides comparable clinical results in patients with and without CCSS. Material and Methods: IRB approval was obtained for this retrospective review of prospectively collected data. Patients with radiculopathy or myelopathy undergoing one- or two-level cervical disc arthroplasty at a single institution (2017-2020) were included and stratified by the Torg-Pavlov Ratio (TPR), a clinical index of CCSS (< 0.82 = CCSS; ≥ 0.82 = non-CCSS). Patient reported outcome measures (PROMs) included the Neck Disability Index (NDI), Visual Analog Scale for neck and arm pain (VAS-Neck, VAS-Arm), Short Form-12 Physical and Mental Component Scores (SF-12 PCS, SF-12 MCS), and PROMIS Physical Function (PROMIS PF), collected preoperatively, < 6 months, and ≥6 months postoperatively. Correlations between radiographic parameters and PROMs were assessed, and between-group and within-group comparisons were performed using standard parametric and non-parametric tests. MCID (Minimal Clinically Important Difference) thresholds were applied, with significance set at p < .05. Results: Sixty-two patients were included where the average age was 43 years, and the sex split was 54.8% male. 59 patients were available for TPR analysis, (12 CCSS, 20.3%; 47 non-CCSS, 79.7%). No significant correlations were found between TPR or sagittal canal diameter and PROMs, except for a moderate correlation between TPR and SF-12 PCS improvement at ≥ 6 months. Significant postoperative improvement across most PROMs was seen in both cohorts. CCSS patients did show less consistent gains, with no significant improvement in Arm VAS, SF-12 PCS at < 6 months, or SF-12 MCS at ≥ 6 months. At ≥ 6 months, non-CCSS patients had greater improvement in Neck VAS (p = 0.04) and SF-12 MCS (p = 0.02). MCID achievement was similar between groups for most PROMs, except SF-12 MCS at ≥ 6 months, wherein 18 patients (58.1%) and only 1 patient (10%) in the non-CCSS and CCSS respectively achieved the MCID (p = 0.011). Conclusion: While CCSS is an important risk factor for cervical pathology, its impact on outcomes after CDA remains poorly understood. This study provides evidence that while CDA is safe and effective in both CCSS and non-CCSS patients, certain PROMs and functional gains may be less robust in CCSS patients, highlighting the need for careful preoperative counseling. Our results contrast with prior ACDF literature, where CCSS has been linked to higher adjacent segment degeneration, suggesting CDA may offer a motion-preserving alternative in this population.

ID: AGS26/1493

A166: Long-term outcomes of cervical disc replacement and anterior cervical discectomy and fusion: a review of FDA IDE trials with real-world comparison

Ali Issani 1 , William Karakash 1 , Henry Avetisian 1 , Daniel Rusu 1 , Aidan Lindgren 1 , Zeeshan Ahsan 2 , Dil Patel 1 , David Riopelle 1 , Kevin Kashanchi 1 , Matthew Gallo 1 Marc A. Abdou 1 , , Amyn Malik 3 , Raymond Hah 1 , Kiran Alluri 1 , John Liu 4 , Jeffrey C. Wang 1

1 Department of Orthopaedic Surgery, Keck School of Medicine of the University of Southern California, Los Angeles, United States

2 Department of Emergency Medicine, Aga Khan University, Karachi, Pakistan

3 O'Donnell School of Public Health, UT Southwestern Medical Center, Dallas, United States

4 Department of Neurological Surgery, Keck School of Medicine of the University of Southern California, Los Angeles, United States

Introduction: Degenerative cervical spine disease, presenting with neck pain, radiculopathy, and myelopathy, has historically been treated with anterior cervical discectomy and fusion (ACDF) as the gold standard of treatment. However, concerns about reduced spinal mobility and adjacent segment disease led to the development of cervical disc replacement (CDR), a motion-preserving alternative approved for clinical use in 2007. Nine devices are now FDA-approved, with evidence showing CDR to be non-inferior to ACDF and offering better long-term outcomes in pain relief, function, ASD, and revision rates. These approvals were based on FDA investigational device exemption (IDE) trials, which have faced criticism for limited post-approval oversight. Post-approval studies (PAS) remain underpowered and poorly generalizable. To address these gaps, this study compares long-term outcomes of CDR and ACDF using real-world data from a national database, focusing on subsequent surgery rates relative to FDA IDE trial findings. Materials and Methods: This study used the PearlDiver Mariner Database (2010-2022) to identify ACDF and CDR cases, excluding patients < 18, those with < 1-year follow-up, prior cervical surgery, hybrid constructs, or procedures for trauma, malignancy, or infection. Cohorts were stratified by subsequent cervical surgeries, with complete follow-up required for inclusion at 2-, 3-, 4-, 5-, 7-, and 10-year intervals. FDA-approved IDE trials were identified through PubMed, Scopus, and Cochrane (April 2025), excluding non-FDA devices and follow-up > 10 years. The primary endpoint was subsequent cervical surgery, comparing FDA trial results with real-world PearlDiver outcomes. Event rates were calculated as surgeries per patients at risk, and Kaplan-Meier-style curves were reconstructed from published data using the Guyot algorithm with monotonicity constraints. Survival probabilities were pooled by weighted means, hazard ratios were estimated at discrete intervals, and differences between cohorts were tested using chi-square or Fisher’s exact test (p < 0.05) in R (v3.4.4). Results: A total of 277,146 ACDF and 25,957 CDR patients were identified in PearlDiver, including 174,135 1-level and 103,011 2-level ACDF cases, and 21,995 1-level and 3,962 2-level CDR cases. From literature review, 44 FDA IDE trials and two additional FDA-approved device studies were included from google scholar. Compared with FDA cohorts, PearlDiver patients undergoing 1-level ACDF had significantly higher survival at 2, 5, 7, and 10 years (p < 0.05), while 1-level CDR showed modest but significant advantages at 2, 7, and 10 years. For 2-level ACDF, survival was similar at 2 years but significantly higher in PearlDiver at 5 and 7 years. In contrast, 2-level CDR demonstrated high survival in both cohorts with no significant differences across all time points. Conclusion: In conclusion, this study demonstrates that CDR performs comparably in real-world settings and controlled trials for 1-level procedures, whereas ACDF shows more favorable real-world outcomes with lower subsequent surgery rates than reported in FDA trials. These differences may reflect trial selection bias or stricter adjudication of ACDF outcomes. Our findings validate the safety and efficacy of both procedures, support CDR as a motion-preserving alternative for select patients and underscore the value of real-world data in complementing trial evidence.

ID: AGS26/1939

A167: Long-term clinical outcomes beyond ten years following lumbar total disc replacement: a systematic review and meta-analysis

Arnaud Delafontaine 1,2 , Jean-Baptiste Odent 1 , Jean-Charles Aurégan 1,2,3

1 Department of Orthopedic, Traumatological and Reconstructive Surgery, Prosthetic Replacement Unit of the Spine, Hip and Knee, Antoine Béclère Hospital, APHP, University Paris Saclay, Clamart, France, CIAMS, University Paris-Saclay, Orsay, France, CIAMS, University of Orléans, Orléans, France, Clamart, France

2 CIAMS, University Paris-Saclay, Orsay, France, Orsay, France

3 B3OA, University Paris Cité, Paris, France

Introduction: Lumbar total disc replacement (TDR) aims to preserve segmental motion and potentially reduce adjacent segment degeneration compared with fusion. While mid-term outcomes are well documented, evidence beyond 10 years remains limited. This systematic review and meta-analysis synthesizes the long-term clinical, radiographic, and implant-related outcomes of lumbar TDR, emphasizing differences between early-generation and modern implants. Material and Methods: A systematic review were conducted using PubMed, Scopus, and Web of Science through September 2025. Studies with ≥ 10-year follow-up reporting clinical outcomes after lumbar TDR were included. Outcomes assessed included Oswestry Disability Index (ODI), Visual Analog Scale (VAS) for back pain, patient satisfaction, revision and reoperation rates, implant survivorship, and radiographic parameters (segmental mobility, heterotopic ossification, and spontaneous fusion). Five key cohorts were included: Putzier et al. (2006; Charité I,II and III, 17-year follow-up), Park et al. (2016; ProDisc-II, 10-year follow-up), Lu et al. (2018; Charité III, 15.4-year follow-up), Kitzen et al. (2020; Charité III, 10 -15-year follow-up), and Marnay et al. (2025; 1 -2 level ProDisc L, 7 -21-year follow-up). Random-effects meta-analysis was applied to continuous outcomes, with subgroup analyses comparing early-generation versus modern implants and mono- versus bisegmental replacements. Results: A total of 1847 patients were included. Modern implants, such as ProDisc-II and ProDisc L, demonstrated sustained clinical improvements (mean ODI reduction -22.1 points, 95% CI -24.7 to -19.5), low index-level revision rates (0.7-3% at 10 -12 years), preserved segmental mobility, and high patient satisfaction (> 70%). Early-generation Charité implants achieved significant initial ODI and VAS improvements; however, long-term follow-up revealed high rates of spontaneous fusion or ankylosis (up to 60%), loss of segmental motion, and reoperation rates up to 38%. Radiographic analysis indicated that heterotopic ossification and spontaneous ankylosis were the primary contributors to functional decline in older implants. Functional implants preserved motion and were associated with low incidence of adjacent segment degeneration, whereas ankylosed or fused segments showed degeneration rates comparable to fusion surgery. Subgroup analyses suggested that patient satisfaction correlated more closely with preserved segmental function or spontaneous ankylosis than with radiographic alignment alone. Conclusion: Lumbar TDR can provide durable clinical improvement beyond 10 years. Modern devices such as ProDisc-II and ProDisc L achieve the most favorable long-term outcomes, preserving motion and maintaining low revision rates. Early-generation Charité implants, while initially effective, are associated with higher rates of spontaneous fusion, segmental motion loss, and reoperation, highlighting the critical role of implant design, surgical technique, and patient selection. Radiographic outcomes reinforce that functional preservation, rather than mere anatomical alignment, is key to long-term success. Continued long-term evaluation of newer TDR designs is essential to optimize indications, improve biomechanics, and maximize patient-centered outcomes.

ID: AGS26/2575

A168: Less superior adjacent syndrome and lower reoperation rate. Medium- and long-term results of cervical arthroplasty versus anterior cervical arthrodesis: systematic review and meta-analysis of randomized clinical trials

Jorge Nuñez 1,2 , Anna Taberner 1 , Juan Diego Montenegro 1 , Victoria San Gil 1 , Berta Escudero 1

1 Hospital Universitari MútuaTerrassa, COT & Care Institute SL, Terrassa, Spain

2 Clinica Teknon, Barcelona, Spain

Introduction: To compare medium- and long-term postoperative surgical results, especially the adjacent syndrome rate, adverse event rate, and reoperation rate, of patients operated on with cervical arthroplasty or anterior cervical arthrodesis in published randomized clinical trials (RCTs), at one cervical level. Material and Methods: Systematic Review and meta-analysis. The inclusion criteria for study selection were as follows: (1) randomized controlled trials (RCTs) comparing the two surgical procedures (cervical arthroplasty [CA] and anterior cervical arthrodesis [ACA]), (2) studies evaluating single-level cervical surgery, (3) follow-up time longer than 60 months, and (4) studies reporting at least one outcome related to the variables under investigation. Thirteen RCTs were included. Clinical, radiological, and surgical outcomes were analyzed, with adjacent segment disease rate and reoperation rate defined as the primary endpoints. Secondary variables included the Neck Disability Index (NDI), cervical visual analog scale (VAS), radicular VAS, the 36-item Short Form Health Survey (SF-36) physical and mental component scores, adverse event rate, high-grade heterotopic ossification rate, and range of motion (ROM) at the operated cervical segment. Results: A total of 2,963 patients were analyzed. The cervical arthroplasty group demonstrated a lower rate of superior adjacent segment disease (OR = 0.49, 95% CI: 0.37-0.65, p < 0.001), a lower reoperation rate (OR = 0.47, 95% CI: 0.32-0.70, p = 0.0001), less radicular pain (SMD = -0.17, 95% CI: -0.28 to -0.06, p = 0.002), and better scores in both the Neck Disability Index (SMD = -0.16, 95% CI: -0.29 to -0.02, p = 0.02) and the SF-36 physical component (SMD = 0.13, 95% CI: 0.03-0.24, p = 0.01) compared with anterior cervical arthrodesis. No significant differences were observed in the lower adjacent segment disease rate (OR = 0.70, 95% CI: 0.34-1.44, p = 0.33), adverse event rate (OR = 0.70, 95% CI: 0.34-1.44, p = 0.33), neck pain VAS (SMD = -0.16, 95% CI: -0.35 to 0.02, p = 0.08), or SF-36 mental component (SMD = 0.09, 95% CI: -0.15 to 0.34, p = 0.47). At final follow-up, a mean cervical range of motion of 7.91 degrees was preserved, with a high-grade heterotopic ossification rate of 9.67% in patients who underwent cervical arthroplasty. Conclusion: In the medium and long-term follow-up, cervical arthroplasty showed a lower rate of superior adjacent syndrome and a lower rate of reoperation. No statistically significant differences were found in the rate of inferior adjacent syndrome or in the rate of adverse events.

ID: AGS26/2628

A169: Functional radiographic assessment of 772 patients treated with 1 level lumbar TDR: an analysis of pelvic ROM and L1 race as motion parameters

Thierry Marnay 1 , Guilhaume Geneste 1 , Grégory Edgard-Rosa 1 , Martin Grau-Ortiz 1 , Caroline Hirsch 1 , Georges Negre 1

1 CCV Montpellier, Clinique du Parc, Montpellier, France

Introduction: Multiple studies have shown that lumbar Total Disc Arthroplasty (TDA) is not only safe and effective in treating chronic lumbar degenerative disc disease (DDD) but also provides a better surgical alternative to fusion with respect to patient-reported results, restoring motion, and reducing reoperation rates and severity of adjacent level degeneration. Published evidence is limited regarding analysis of motion restoration and index level motion demonstrating the Global influence of TDR on lumbar spine motion. Material and Methods: In this retrospective 1-level TDA study, we report results up to 21 years follow-up, of 772 patients treated with semi-constrained ball-and-socket lumbar TDA from 1999 to 2013. We compare clinical outcomes and motion analysis of 1-level L5-S1 versus L4-L5 versus L3-L4 patients. Two hundred of these patients had no disc herniation (Group H0), and 350 patients had a disc herniation (Group H1). 222 patients had already been operated on with a discectomy at the index level; of those, 124 had a recurrent disc herniation (Group H3), and 98 had no recurrent disc herniation (post-discectomy syndrome). Indications for TDA were determined on clinical status, long-term low back pain with more recent leg pain, resistant to conservative care (minimum 6 months), radiographic explorations and MRI. Patient level distribution was: 1 at L2-L3, 14 at L3-L4, 234 at L4-L5 and 523 at L5-S1. Clinically, patients were evaluated preoperatively, then postoperatively at 3, 6, 12, and 24 months, then annually. Collected data included Oswestry Disability Index (ODI), Visual Analog back and leg pain Scores (VAS), neurological/physical assessment. Complications, reoperation/revision rates, and perioperative data points were also assessed. We also measured, pre- and postoperatively, pelvic parameters, index level flexion/extension motion, pelvic Range of Motion (ROM) (sacral slope, flexion/extension angle), and L1 RACE (L1 in extension and flexion angle). Results: All Groups showed dramatic and statistically significant ODI reduction by 3 months post-surgery (p < 0.001) and maintained scores over time. ODI improvement ranged from 26-30 points. Although patients with prior surgery took longer to reach final VAS pain levels (back and leg) (6 versus 3 months), there was no significative statistical difference between groups in H classification or according to the level operated on, concerning ODI, VAS back and leg pains reduction at 24 months. Index Revision rate was 0.77% and adjacent level reoperation 2%. The motion analysis showed the index level ROM improved 8.3° at L5-S1 TDA group, 9.3° at L4-L5 TDA, 11.5° at L3-L4. The lumbar extension-flexion gain was 19.7° for L5-S1 TDA group, 12,6° for L4-L5, and 26.8° for L3-L4. The pelvic range of motion gain was 13.83° at L5-S1 TDA group, 18.7° at L4-L5, and 23.8° at L3-L4 TDA. The global motion gain measured by L1 Race was 31.34° in L5-S1 TDA group, 27.2° in L4-L5, and 48.4° in L3-L4. Conclusion: One of the largest groups of TDA patients evaluated, study results demonstrate the robust long-term clinical success of one-level lumbar TDA, as assessed at 7-21 years postoperatively. Patients had significant- and maintained-reduction in disability and pain scores over time, and low rates of index revision/reoperation or ALS, when compared to long-term published fusion data. Flexion extension measures showed substantial improvement at index level, pelvic participation in flexion extension (Pelvic ROM) and global motion from the lumbar-pelvic complex (L1 Race). These motion results explain the quality of clinical results as functional and adjacent level preservation points of view.

ID: AGS26/2962

A170: Comparison of outcomes between single-level cervical disc arthroplasty and anterior cervical discectomy and fusion in nicotine-dependent patients

Ankit Hirpara 1 , Jalen Warren 1 , Nakul Narendran 1 , Christina Cheng 1 , Christopher Furey 1 , Prashant Rajan 1

1 Case Western Reserve University/University Hospitals, Cleveland, United States

Introduction: Nicotine use is frequently associated with increased complications after cervical fusion, including dysphagia, infection, and revision surgery. Moreover, the conventional cage-plate system used in anterior cervical discectomy and fusion (ACDF) may predispose patients to adjacent segment disease, soft tissue injury, and pain. Studies have shown that cervical disc arthroplasty (CDA) may offer advantages compared to ACDF in the general population, but it is unclear whether these benefits extend to nicotine users as well, who are known to be suboptimal surgical candidates. As such, the purpose of this study was to compare rates of post-operative medical complications, opioid use, and revision surgery in nicotine users undergoing CDA versus ACDF. Material and Methods: The TriNetX database, commonly used in the orthopedic literature, was queried to identify patients aged 18 to 60 years with nicotine dependence who underwent isolated, primary single-level ACDF or single-level CDA. Patients in the ACDF cohort were excluded if they had relative or hard contraindications for CDA, including but not limited to cervical fracture, instability, deformity, or disorders of bone density or structure. This was done to ensure that both cohorts consisted of patients who would have been theoretically eligible for either intervention. Cohorts underwent propensity score matching in a 1:1 ratio based on 23 covariates, including demographic variables, medical comorbidities, substance use, and indication for surgery. The following post-operative outcomes were collected: 1) complications and healthcare utilization within 3 months, 2) opioid use within 2 years, and 3) revision surgery within 2 years. Results: Following successful propensity score matching, each cohort contained 1,049 patients. Within 3 months, patients in the ACDF cohort had higher rates of healthcare utilization, including readmission (4.5% vs. 1.8%, p < 0.001) and emergency department visits (13.6% vs. 9.7%, p = 0.005), as well as dysphagia (5.2% vs. 3.2%, p = 0.023). There were no significant differences in medical or respiratory complications. More patients in the ACDF cohort were prescribed opioids at all time points within 2 years (all p < 0.001), and they were given more prescriptions at 1 week (1.4 vs. 1.3, p = 0.015) and 1 month after surgery (2.2 vs. 1.9, p = 0.038). Lastly, patients who underwent ACDF had higher rates of revision cervical fusion at 1 year (2.7% vs. 1.0%, p = 0.004) and 2 years (4.3% vs. 1.1%, p < 0.001). Patients who underwent ACDF also had higher rates of revision cervical surgery at 2 years (5.0% vs. 2.9%, p = 0.013). Conclusion: ACDF, compared to CDA, is associated with higher rates of healthcare utilization, dysphagia, opioid use, and revision surgery in patients who are dependent on nicotine. These results may be used to inform and optimize surgical decision making in this high-risk population.

ID: AGS26/658

A171: Complications following cervical and lumbar artificial disc replacement - A retrospective analysis between 2014 and 2022

Karsten Ritter-Lang 1

1 Orthopedic Surgery, Spine Surgery, Stenum Ortho GmbH, Ganderkesee, Germany

Introduction: Although artificial disc replacement surgery has been an established method for 40 years, it is still viewed critically. This is likely due to the fact that the complications that occur have not been systematically examined to date. A distinction between implant-associated and non-implant-associated complications has rarely been published in the literature. This retrospective analysis distinguishes between early and late complications and differentiates between implant-associated and non-implant-associated complications. It aims to provide information about typical complications associated with different implants and help to avoid potential errors when determining the indication. Material and Methods: The retrospective analysis is based on 2,614 cases following artificial disc replacement surgery between 2014 and 2022 at the Clinic for Orthopaedics Stenum. During the same period, 172 patients underwent revision surgery. Of these, 100 patients were referred from outside the clinic. The evaluation considered the reasons for the necessary revision and then classified them into early and late complications as well as implant-associated and non-implant-associated complications. Implant-associated complications included: subsidence, migration, prosthesis defect, low-grade infection, acquired spondylolisthesis, osteolysis of unclear origin, stick-slip friction, segmental hypermobility, rotational slippage and aseptic loosening. Non-implant-associated complications included: early/postoperative haematoma, early nerve root irritation, early spondylolisthesis, early paresis, early infection, late infection and persistent stenosis. Results: In 158 cases, we observed implant-related causes. In contrast, 14 cases involved complications not related to the implant. Among the implant-related complications, 67 cases involved core defects (Cadisc) alone. These were followed by 31 cases with low-grade infections, 16 cases with acquired spondylolisthesis, 10 cases with unclear osteolysis, 10 cases with stick-slip friction, 8 cases with subsidence, 8 cases with instability/hypermobility, 4 cases with migration, 2 cases with aseptic loosening and 2 cases with rotational slippage. Looking at the total number of treatments and excluding the Cadisc cases, the revision rate is 3.86%. The data also shows that certain specific problems occur more frequently with certain implants and not at all with others. There are 24 cases of early complications and 81 cases of late complications. Conclusion: The available data show that systematic analysis of apparent causes of failure is necessary. It is also clear that the complications that occurred were predominantly not due to the treatment method. At 3.86%, the revision rate is low. This rate can certainly be reduced further through strict indication and appropriate implant selection. In addition, it is evident that late complications clearly predominate, which is why systematic follow-up observation in the first 24 months after surgery is recommended.

OP20: Preventing Infection in Spine

ID: AGS26/125

A172: Expandable cage versus mesh cage in the treatment of vertebral osteomyelitis

Mitsuhiro Nishizawa 1 , Mladen Djurasovic 1 , Steven Glassman 1 , John Dimar 1 , Charles Crawford, III 1 , Benjamin Kostic 1 , Leah Carreon 1

1 Norton Leatherman Spine Center, Louisville, Kentucky, United States

Introduction: Surgical treatment of vertebral osteomyelitis often includes anterior column reconstruction due to large bone defects following debridement. However, consensus regarding the optimal method for anterior column reconstruction in vertebral osteomyelitis has not been established. The purpose of this study is to evaluate differences in clinical outcomes and costs between mesh cages and expandable cages in patients with vertebral osteomyelitis. Material and Methods: We retrospectively reviewed the clinical records of consecutive patients who underwent surgical intervention for vertebral osteomyelitis at a single institution between 2012 and April 2024. Patients who underwent anterior column debridement with either a mesh cage or an expandable cage and posterior fusion involving four or more levels were included. Clinical outcomes, including fusion, nonunion, revision surgery, and recurrence, as well as costs, were compared between the expandable cage and mesh cage groups. Results: A total of 92 patients were included: 20 (22%) in the expandable cage group and 72 (78%) in the mesh cage group. The expandable cage group had a significantly higher nonunion rate (35% vs 9.7%, p = 0.014) and revision rate (30% vs 9.7%, p = 0.032). Costs were also higher in the expandable cage group, with a significant difference in in-hospital costs (Expandable: 77,737 ± 34,586; Mesh: 63,250 ± 27,899; p = 0.028). The number of fused levels (OR = 1.32, 95% CI: 1.05-1.71; p = 0.024) and the use of an expandable cage (OR = 3.27, 95% CI: 1.04-11.02; p = 0.046) were independent risk factors for treatment failure, defined as revision surgery, recurrence, or mortality. Conclusion: The use of an expandable cage for reconstruction after anterior column debridement was associated with higher costs and with increased risks of nonunion, revision surgery, and treatment failure compared to the use of mesh cages.

ID: AGS26/1333

A173: Effect of allograft bone culture findings on surgical site infection

Henar M a Arqueros Martinez 1 , Sara Martos 1 , Pablo Guinea 1

1 Hospital Universitario Fundacion Alcorcon, Madrid, Spain

Introduction: The objective of this study was to analyze the relationship between the results of graft cultures and surgical wound infection after lumbar spinal fusion surgery in our center. Additionally, to study other factors related to postoperative infection in our patients. Material and Methods: A retrospective cohort study was conducted on patients who met the following inclusion criteria: lumbar spinal fusion surgery involving 1, 2, or 3 levels, performed between 2015 and 2022, without previous surgeries, requiring allograft implantation, and with a follow-up period of at least 1 year. The variables collected were: age, Charlson comorbidity index, hospital stay, obesity, smoking, immunosuppressive treatment, levels treated, postoperative transfusion, pre-implant allograft culture results, intraoperative complications, and postoperative infection. These variables were analyzed using SPSS version 29.0. Results: We included 521 patients and 613 allografts used. Twenty patients (3.8%) developed postoperative infection within the first year, most infections diagnosed between 1st-3rd month. Pre-implant culture was positive in 37 cases (7.1%). Of these, 3 (8.1%) developed postoperative infection. In the group with negative cultures, infection occurred in 17 cases (3.5%), not being statistically significant (p > 0.05). In patients with positive cultures and postoperative infection, the same microorganism was isolated in 3 cases (15%). The most frequent microorganisms isolated were coagulase-negative Staphylococcus in allograft cultures, and Staphylococcus epidermidis in post-surgical wound cultures. Other studied variables were not statistically significant for the development of postoperative infection. Conclusion: A positive allograft culture didn’t predict postoperative infection after lumbar spinal fusion surgery in our sample. Additionally, risk factors described in the literature didn’t influence our sample.

ID: AGS26/1444

A174: Impact of the duration of antibiotic prophylaxis on surgical site infection rate in patients undergoing complex spinal surgery for spinal metastases

Omer Ahmed 1,2 , Amina Ait Mouhoub 1,2 , Mahreen Tanveer 2,3 , Jesse Shen 1,2 , Fidaa Al-Shakfa 1,2 , Zhi Wang 1,2

1 Université de Montréal, Montreal, Canada

2 Centre Hospitalier de l’Université de Montréal, Montreal, Canada

3 McGill University, Montreal, Canada

Introduction: Traditional guidelines are increasingly discouraging the use of postoperative antibiotic prophylaxis. However, there is limited evidence in support of these recommendations in high-risk groups such as cancer patients undergoing complex spinal surgery. Considerable variation exists in post-operative antibiotic prescribing practices, often guided by individual surgeon judgement and preference. Hence, this study aims to evaluate whether varying durations of postoperative antibiotics influence surgical site infection rates in this population. Ultimately, the goal is to establish whether prolonged prophylaxis offers an additional protective effect or whether current guidelines are sufficient for this high-risk population. Material and Methods: This single-center retrospective observational pilot study identified all patients surgically treated for spinal metastasis between 2010 and 2024, with instrumentation or fusion spanning ≥ 6 spinal levels. Medical data was extracted from the institutional electronic medical record system ( Oacis) . Patients with primary spinal tumors and who had undergone prior radiotherapy at the site of surgery were excluded. Analysis was conducted in two patient groups based on the antibiotic duration (≤ 24h vs. > 24h). The primary outcome was SSI occurrence within a 90-day postoperative window from the date of surgery. Statistical analyses were conducted using the SAS software (SAS Institute, Cary, NC, USA). A two-sided p-value < 0.05 was considered statistically significant. Results: A total of 139 patients who underwent surgery spanning ≥ 6 spinal levels for spinal metastasis were included. A total of 14 surgical site infections were identified, with an overall rate of 10.07%. Statistical techniques such as the Chi-square test and Ficher’s exact test were used to identify the relationship between ATB duration and SSI rate. The Chi-square test showed no significant difference in infection rates among the two groups with antibiotic durations of > 24 h and ≤ 24 h, with Chi-square p = 0.54, showing no significant difference. The Ficher’s exact test also supported this finding with p = 0.770. Further logistic regression analysis showed that the unadjusted odd ratio was 0.687, which suggests that an antibiotic duration of > 24 h reduces infection odds by ∼31%, compared to a duration of ≤ 24 h. Additionally, the multivariable logistic regression model identified several confounding variables such as diabetes, smoking status, and length of surgery. After adjusting for these confounders, the OR dropped to 0.417, suggesting a stronger protective effect of prolonged antibiotics with a reduction in infection odds by ∼58%. None of the confounder variables reached statistical significance, although smoking status showed a trend towards increased infection odds with OR = 6.166 and borderline significance (p = 0.0678). Conclusion: This study observed a protective effect of prolonged antibiotic prophylaxis in surgically treated spinal metastasis patients, although it was not statistically significant. A trend was observed towards reduced infection odds in the group with extended antibiotics (> 24h), suggesting potential clinical significance despite the absence of statistically significant findings. Prolonged prophylaxis may have a protective effect in patients with spinal metastasis undergoing complex spinal surgery. However, larger, multi-center studies are needed to verify this finding and to establish guidelines for prescribing optimal antibiotic prophylaxis in this population.

ID: AGS26/1454

A175: What is worse: a comparison of solitary versus multifocal pyogenic spondylodiscitis using a nationwide analysis of readmission rates and risk factors

Julius Gerstmeyer 1,2,3 , Anna Gorbacheva 4 , Clifford Pierre 1,4 , Mark Kraemer 1,4 , Colin Gold 1 , Cameron Hogsett 1 , Nicholas Minissale 1 , Alexander von Glinski 3 , Tobias Schulte 3 , Thomas A. Schildhauer 2 , Amir Abdul-Jabbar 1,4 , Rod Oskouian 1,4 , Jens Chapman 1,4

1 Swedish Neuroscience Institute, Seattle, United States

2 Department of General and Trauma Surgery, BG University Hospital Bergmannsheil, Ruhr University Bochum, Bochum, Germany

3 Department of Orthopedics and Trauma Surgery, St. Josef Hospital Essen-Werden, Essen, Germany

4 Seattle Science Foundation, Seattle, United States

Introduction: Spondylodiscitis is a growing infectious condition with significant morbidity. The impact of multifocal involvement remains understudied. This study compared 90-day all-cause readmission rates between patients with solitary versus multifocal spondylodiscitis and identified the associated risk factors. Material and Methods: A retrospective analysis of the 2020 Nationwide Readmissions Database was conducted. Adult patients with primary spondylodiscitis were identified using ICD-10 codes and categorized into solitary or multifocal involvement groups. Demographic, clinical, and surgical data were extracted. Descriptive statistics and multivariate logistic regression were performed. Results: Of 6132 patients, 585 (9.6%) had multifocal disease. Multifocal patients were slightly younger (58.9 vs. 60.3 years; p = 0.049); had longer hospital stays (14.7 vs. 11.4 days; p < 0.001), time to readmission (p < 0.001); and surgery was more common (p = 0.003). Ninety-day readmission rates were similar (35.6% vs. 34.9%; p = 0.766). Type 2 diabetes was the only comorbidity significantly associated with multifocal disease (p = 0.020) and independently predicted readmission (aOR 1.236). Surgery and longer length of stay were protective (aOR 0.743; 0.0990). Conclusion: Multifocal spondylodiscitis is relatively common but not an independent risk factor for readmission. Readmission rates of both cohorts were similar. Surgery and prolonged hospitalization may reduce readmission risk.

ID: AGS26/715

A176: The biopsy dilemma in spinal infections: timing of antibiotics and diagnostic yield

Elie Najjar 1

1 Queen's Medical Centre, Nottingham University Hospitals, Nottingham, United Kingdom

Introduction: Spinal infections such as spondylodiscitis carry a risk of severe morbidity if diagnosis and treatment are delayed. Microbiological confirmation is critical for guiding therapy, yet culture yields vary depending on biopsy technique and timing of antibiotics. The effect of prior antimicrobial exposure on diagnostic yield remains uncertain, particularly between CT-guided and open biopsy. Materials and Methods: We conducted a systematic review in accordance with PRISMA guidelines, searching MEDLINE, EMBASE, Web of Science, Cochrane Library, and Scopus through June 2025. Eligible studies included adults and children with suspected spinal infection undergoing CT-, fluoroscopy-, or open biopsy, with or without prior antibiotics. The primary outcome was culture positivity. Risk of bias was assessed with ROBINS-I, and pooled risk ratios (RR) with 95% confidence intervals (CI) were calculated using fixed- or random-effects models depending on heterogeneity. Results: Thirty-eight studies encompassing > 2,000 patients were included. Nine comparative studies (n = 796) demonstrated that prior antibiotics significantly reduced culture yield (59.6% vs 39.2%; RR = 0.74, 95% CI: 0.63-0.86, p = 0.0001; I 2 = 0%). The reduction was most marked within 72 hours of antibiotic initiation. Open biopsy achieved the highest yield (76.1%) compared with CT-guided (50.0%) and fluoroscopy-guided biopsy (28.6%). Pooled analysis confirmed superiority of open biopsy over CT-guided approaches (RR = 0.66, 95% CI: 0.57-0.76, p < 0.00001), while CT and fluoroscopy showed no significant difference. Conclusion: Prior antibiotic therapy reduces microbiological yield in spinal infections, with the strongest effect observed in the first 72 hours. Open surgical biopsy consistently outperforms percutaneous techniques, even after antibiotic exposure. Whenever clinically feasible, a biopsy-first approach should be prioritised in stable patients, whereas early surgical biopsy should be considered when antibiotics cannot be delayed. Standardisation of microbiological protocols and prospective studies are needed to optimise diagnostic pathways.

ID: AGS26/1587

A177: Ideal classification system including instability score which could guide management

Harvinder Singh Chhabra 1 , Mayukh Mukherjee 1 , Julian Scherer 2 , S. Rajasekaran 3

1 Sri Balaji Action Medical Institute, Delhi, India

2 University of Cape Town, Cape Town, South Africa , 3 Ganga Medical Centre & Hospitals, Coimbatore, India

Background: Spinal tuberculosis (Pott’s disease) accounts for a significant proportion of extrapulmonary TB, leading to vertebral destruction and potential neurological compromise. Multiple scoring systems have been proposed to quantify instability and guide management. Objective: To identify and critically appraise spinal instability scoring systems specific to spinal TB, as well as relevant instruments extrapolated from neoplastic and infectious contexts, summarizing their components, validation outcomes, and potential applicability. Material and Methods: Systematic searches of PubMed, Cochrane Library, Scopus, Web of Science, and Google Scholar were conducted. Two reviewers independently screened records, extracted data, and assessed quality using Newcastle-Ottawa Scale and ROBINS-I. Due to heterogeneity, narrative synthesis was performed. Results: A total of seven scoring systems were identified from studies published between 2013 and 2025, encompassing over 570 patients across development and validation cohorts. Four instruments were TB-specific (Rajasekaran TB Instability Score, TSIS, CSTS, Pediatric Spine Instability Score) and demonstrated excellent inter-rater reliability (ICC 0.88-0.99) and high diagnostic performance (sensitivity 76-94%; specificity 82-100%). Three extrapolated scores (SINS, SISS, SITE) were evaluated in TB cohorts but showed moderate reliability (ICC 0.89-0.96) and variable thresholds for instability. Risk-of-bias assessment indicated moderate methodological quality overall, with common limitations including retrospective design and inconsistent external validation. Narrative synthesis highlighted heterogeneity in score components and cut-offs, underscoring the need for standardized, prospective multicenter studies. Conclusion: Among seven evaluated scoring systems, TB-specific instruments (Rajasekaran, TSIS, CSTS, Pediatric) exhibit superior reliability (ICC ≥ 0.88) and diagnostic accuracy (sensitivity 76-94%; specificity 82-100%). Extrapolated tools offer supplementary perspectives but lack TB-specific nuances. Heterogeneity in parameter selection and validation underscores the need for harmonized core domains-incorporating deformity, vertebral loss, neurological status-and rigorous multicenter prospective validation. Adoption of a standardized, evidence-based instability score will enhance surgical decision-making and improve patient outcomes in spinal tuberculosis.

ID: AGS26/1591

A178: Changing tides in the treatment of spondylodiscitis? Retrospective comparison of mortality and quality of life after surgical and conservative treatment

Victoria Buschmann 1 , Erik Wegner 1 , Daniel Wagner 1,2 , Alexander Wartensleben 1 , Philipp Drees 1 , Stefan Mattyasovszky 1 , Tobias Nowak 1

1 University Medical Center Mainz, Mainz, Germany

2 Centre Hospitalier Universitaire Vaudois, Lausanne, Switzerland

Introduction: The increasing incidence of spondylodiscitis and its potentially severe consequences when not promptly diagnosed highlight the need for further research to improve treatment guidelines, reduce mortality and morbidity and improve the quality of life in patients who suffer from persistent physical limitations. Material and Methods: We collected data from 103 patients, with 8 patients lost to follow-up, who were diagnosed with vertebral osteomyelitis, disc infection or discitis between 2009 and 2018. The primary outcome was the 1-year mortality rate in patients treated with either conservative or surgical intervention, with both groups receiving antibiotic treatment. A standardized questionnaire was used to assess health-related quality of life after treatment by evaluating the European Quality of Life 5 Dimension 5 Level version (EQ-5D-5 L) and the European Quality of Life Visual Analog Scale (EQ-VAS). In addition, we used the Oswestry Disability Index (ODI) and the Parker Mobility Score to identify backpain-related limitations after treatment. Results: The group receiving surgical treatment had a significantly lower 1-year mortality rate (22%) than did those who were treated conservatively, who had a 4-fold greater risk for death in the first year after treatment. With respect to quality of life, 39 patients answered the standardized questionnaires during follow-up, and the questionnaire results revealed no significant difference in limitations in daily life or in health-related quality of life, with a median Parker Mobility Score of 9 for the conservatively treated patients compared with 7.5 for the surgically treated patients. This difference between the groups was not statistically significant, with a p value of 0.216 > α. A similar result was obtained in the evaluation of the ODI, with a medial index of 30% in the conservatively treated group compared with 24% in the surgical group, which was not statistically significant as indicated by a p value of 0.360 > α. Conclusion: The early surgical approach when treating spondylodiscitis is advantageous for identifying the underlying infection and initiating appropriate antibiotic therapy, therefore reducing mortality and resulting in a greater likelihood of full recovery than the conservative treatment does.

ID: AGS26/43

A179: Is povidone-iodine irrigation in spine surgery a game-changer? A meta-analysis of RCTs on surgical site infection prevention

Chinedu Egu 1 , Rodrigo Muscogliati 1 , Weronika Nocun 1 , Hussein Akil 1 , Wasse Saleh 1 , Nasir Quraishi 1 , Khalid Salem 1 , Elie Najjar 1

1 Centre for Spinal Studies and Surgery, Queens Medical Centre, Nottingham, United Kingdom

Introduction: Surgical site infections (SSIs) remain a significant complication in spinal surgery, particularly in instrumented procedures, contributing to increased morbidity, prolonged hospitalization, and healthcare costs. Intraoperative wound irrigation (IWI) using antiseptic solutions such as povidone-iodine (PI) has been proposed as a preventive strategy. This systematic review and meta-analysis of only randomized controlled trials (RCTs) evaluate the efficacy of PI irrigation against control interventions, in reducing postoperative SSIs and examines its effect on spinal fusion outcomes. Material and Methods: Following PRISMA guidelines, a comprehensive search of Ovid MEDLINE, EMBASE, Web of Science, and the Cochrane Library was conducted through March 2025. Only randomized controlled trials (RCTs) comparing PI irrigation to control interventions (saline or no irrigation) in adult spinal surgery were included. The primary outcome was the overall SSI rate. Secondary outcomes included superficial and deep SSI rates, fusion rates, and adverse events. Data were pooled using a random-effects model, and heterogeneity was assessed using the I 2 statistic. Results: Six RCTs involving 1,804 patients (901 PI group, 903 control group) were included. Pooled analysis demonstrated a significant reduction in overall SSI with PI irrigation (0.77% vs. 6.0%; RR = 0.16, 95% CI: 0.08-0.34; p < 0.00001; I 2 = 0%). Subgroup analyses showed significant reductions in both superficial (RR = 0.19; 95% CI: 0.07-0.52; p < 0.001) and deep SSIs (RR = 0.21; 95% CI: 0.08-0.52; p = 0.0009), with absolute infection rates reduced from 2.9% to 0.33% and from 3.2% to 0.4%, respectively. Fusion rates were comparable between groups (RD = 0.03; 95% CI: 0.00-0.06; p = 0.10). Conclusion: PI irrigation significantly lowers the risk of both superficial and deep SSIs in spinal surgery without negatively impacting fusion outcomes. These findings are based on Level I evidence and support the integration of PI irrigation into routine perioperative protocols as a low-cost, effective adjunct for infection prevention.

ID: AGS26/804

A180: Establishment and validation of an infection prevention and control system following vertebral augmentation: a hybrid design study with historical controls and prospective validation

Wenjie Wu 1 , Chuanqi Zou 1 , Li Huo 1 , Xiaojing Si 1

1 The People's Hospital of Banan District, Chongqing, China

Introduction: Surgical site infection (SSI) remains a serious complication of vertebral augmentation procedures (VP/KP), especially in resource-limited settings. This study aimed to develop and validate a feasible, cost-effective SSI prevention strategy tailored for county-level hospitals. Material and Methods: A hybrid step-wedge design was employed, integrating a historical cohort and a prospectively enrolled interventional cohort. The historical cohort included 1000 VP/KP patients (January 2020-June 2024), with SSI cases re-adjudicated per CDC criteria. A refined prevention strategy was developed based on risk factors identified from the historical analysis and multidisciplinary consensus. Key elements included: preoperative optimization of high-risk patients (diabetes with HbA1c > 7%, renal insufficiency with eGFR < 45 mL/min/1.73m 2 ), standardized skin preparation, intraoperative pulsed lavage of bone cement, strict ase technique, and dynamic CRP monitoring postoperatively (alert threshold: CRP > 80 mg/L on postoperative day 3). The prospective cohort (July-December 2024, n = 200) received the intervention. Each patient was matched 1:1 with a historical control using propensity score matching (caliper = 0.2 SD) based on age, sex, diabetes, renal function, surgery type, and number of levels. Primary outcome was a composite of confirmed or highly suspected SSI within 90 days. Results: The historical SSI rate was 6.0%. After matching, 180 pairs were formed. The prospective cohort showed a significant reduction in the composite SSI rate to 2.2% (4/180) compared to 6.7% (12/180) in matched controls (adjusted OR = 0.32, 95% CI: 0.12-0.82, p = 0.017). Subgroup analysis confirmed consistent benefits in high-risk populations: diabetes (aOR = 0.29), renal impairment (aOR = 0.31), and multi-level surgery (aOR = 0.35). Median hospitalization cost increased slightly (¥28,500 vs. ¥27,800, p = 0.09), yielding an incremental cost-effectiveness ratio of ¥14,000 per SSI avoided. The probability of cost-effectiveness exceeded 80% at a willingness-to-pay threshold of 1× GDP per capita. Conclusion: A tailored infection prevention strategy significantly reduces SSI following vertebral augmentation in county-level hospitals, with favorable cost-effectiveness. Key components include patient optimization, cement lavage, and CRP-guided monitoring. This protocol is feasible for implementation in resource-limited settings.

Keywords: Vertebral augmentation, surgical site infection, infection prevention, pulsed lavage, cost-effectiveness, county hospital, propensity score matching

OP21: Effect of Osteoporosis on Spine Care Management

ID: AGS26/1515

A181: Osteoporosis and perioperative management trends in posterior cervical fusion patients between 2012-2022

Apurva Prasad 1 , William Karakash 1 , Justin Zheng 1 , Daniel Rusu 1 , Aidan Lindgren 1 , Henry Avetisian 1 , Marc Abdou 1 , Jeffrey C. Wang 1 , Raymond Hah 1 , Kiran Alluri 1

1 Department of Orthopaedic Surgery, Keck School of Medicine of the University of Southern California, Los Angeles, United States

Introduction: Posterior cervical fusion (PCF) is an increasingly performed surgical procedure for a range of degenerative and traumatic cervical spine conditions. Osteoporosis has been consistently associated with higher complication rates, pseudarthrosis, hardware failure, and revision surgery in spine procedures. In this study, we sought to observe trends in osteoporosis perioperative management in a subset of PCF patients. Material and Methods: Using PearlDiver, patients who underwent multi-level PCF between 2012-2022 were identified using ICD-9/10 codes. Patients under 18 years old, those with less than two years of preoperative follow-up, or indications related to trauma, malignancy, or infection were excluded. We subsequently identified criteria in accordance with the International Society for Clinical Densiometry (ISCD) guidelines for screening: females ≥ 65, males ≥ 70, or younger individuals with one or more established risk factors for osteoporosis. In addition, we identified patients who received DEXA scans using ICD-9/10 codes. Temporal trends were assessed by comparing data from two periods (2012-2016 and 2017-2022), in addition to building multivariate analyses to analyze predictors of DEXA screening. Results : From 2012-2022, 57,384 total patients underwent posterior cervical spinal fusion (PCF) operation. 21,872 patients underwent PCF operation in 2012-2016, compared with 35,512 between 2017-2022. 6045 of these patients had a diagnosis of osteoporosis. 43,086 met ICSD screening criteria in the whole cohort. When stratifying by time period, 70.33% of patients met ICSD screening criteria between 2012-16, compared with 78.01% between 2017-22 (p < 0.001). Fragility fractures significantly increased from 2012-16 (1300, 9.1%), to 2017-22 (2188, 9.96%) (p = 0.007). In trends screening, patients met screening criteria at an increased rate from 2012-22. From multivariate analysis, age and Elixhauser Comorbidity Index (ECI) were significant predictors of DEXA scan prior to PCF operation (both p < 0.0001). Additionally, gender (p < 0.001), Medicare (OR = 0.89, p = 0.00259), Medicaid (OR = 0.70, p < 0.001), and commercial insurances (OR = 1.27, p < 0.001) proved to be significant predictors of preoperative DEXA scan. Osteoarthritis significantly increased chances of a DEXA scan (OR = 1.51, p < 0.001). Additionally, diabetes significantly decreased odds of receiving a DEXA scan in the whole cohort (OR = 0.82, p < 0.001). Conclusion: Between 2012-2022, significantly more patients were osteoporotic. However, patients with Medicaid, as well as diabetics were less likely to receive DEXA scans despite meeting screening criteria. As osteoporosis continues to increase in patients undergoing spine surgery, it is imperative for spine surgeons to recognize its underdiagnosis and optimize bone health preoperatively. Increased utilization of DEXA scans in patient populations at increased risk of osteoporosis should be a key aspect of preoperative planning.

ID: AGS26/1666

A182: Validation of vertebral Hounsfield Unit measurement on CT against DEXA for the diagnosis of osteoporosis: a cross-sectional study

Javier Matta 1 , Maria Suarez 1 , David Florez 1 , Victor Arrieta 1 , Sebastian Forero 1

1 Hospital Militar Central, Bogota, Colombia

Introduction: Osteoporosis is a highly prevalent condition in aging populations, associated with fragility fractures, disability, and high healthcare costs. Dual-energy X-ray absorptiometry (DEXA) is the current gold standard for diagnosis, but its availability is often limited in acute or emergency settings. Computed tomography (CT) is widely used for other clinical indications, and vertebral Hounsfield Units (HU) have been proposed as a surrogate marker for bone mineral density (BMD). Establishing a validated correlation between HU and DEXA values could support opportunistic osteoporosis screening and earlier treatment. Material and Methods: We conducted a cross-sectional diagnostic accuracy study at the Central Military Hospital in Bogotá, Colombia. Adults ≥ 50 years who had undergone non-contrast CT of the thorax, abdomen, pelvis, or lumbar spine within three months were recruited and subsequently underwent DEXA. HU values were measured in vertebrae L1-L4 using standardized regions of interest, excluding fractured or artifact-affected vertebrae. The mean HU per patient was compared with DEXA T-scores. Descriptive statistics characterized the cohort, and Pearson’s correlation coefficient assessed the relationship between HU and T-scores. Receiver operating characteristic (ROC) curves determined HU thresholds for osteoporosis and osteopenia. Multivariate linear regression adjusted for age, sex, BMI, waist circumference, and comorbidities. Results: A total of 186 patients were included (mean age 67.2 ± 8.5 years; 62% women). DEXA identified osteoporosis in 32.3% and osteopenia in 41.4%. HU values showed a strong positive correlation with lumbar spine T-scores (r = 0.78, p < 0.001) and a moderate correlation with femoral neck T-scores (r = 0.65, p < 0.001). ROC analysis demonstrated excellent diagnostic accuracy of HU for osteoporosis (AUC = 0.89; 95% CI 0.84-0.94), with an optimal threshold of 110 HU (sensitivity 91.2%, specificity 87.5%). For osteopenia, the threshold was 135 HU (sensitivity 82.4%, specificity 80.1%). Multivariate regression confirmed HU as an independent predictor of DEXA T-scores after adjusting for confounders (β = 0.71, p < 0.001). Conclusion: This study demonstrates a strong correlation between vertebral HU values on CT and DEXA-derived bone mineral density. HU thresholds of < 110 and 110-135 were reliable for identifying osteoporosis and osteopenia, respectively. These findings validate HU measurement as a feasible opportunistic screening tool for osteoporosis, particularly in acute and emergency care settings where DEXA is not available. Implementing HU-based assessment in routine CT interpretation may enable earlier diagnosis, targeted prevention strategies, and improved management of fragility fracture risk in older adults.

ID: AGS26/222

A183: From imaging to failure load - Hounsfield Units as predictors of bone quality in lumbar spine surgery

Anto Abramovic 1 , Sara Lener 1 , Raphael Gmeiner 1 , Christoph Orban 1 , Leonie Feige 2 , Werner Schmölz 2 , Sebastian Hartmann 1 , Claudius Thomé 1

1 Medical University of Innsbruck, Department of Neurosurgery, Innsbruck, Austria

2 Medical University of Innsbruck, Department of Orthopaedics and Traumatology, Innsbruck, Austria

Introduction: Preoperative assessment of bone quality is becoming increasingly important due to rising surgical volumes in elderly patients and their associated comorbidities. The value of the “gold standard” DXA has been questioned and may not sufficiently reflect the surgically targeted vertebral bodies. Therefore, the aim of this study was to evaluate the clinical utility of Hounsfield Units (HU) from routine CT scans in estimating the trabecular bone structure and mechanical stability of lumbar vertebrae. Material and Methods: In an ex-vivo study, 65 lumbar vertebral bodies (L1-L5) from 53 human donors (mean age 73.0 ± 10.5 years; 22 females) were analyzed; twelve specimens were excluded based on predefined criteria. HU values were measured using routine CT (according to Schreiber et al.), followed by qCT to determine volumetric bone mineral density (vBMD). Cylindrical cores (20 mm × 15 mm) consisting of trabecular bone were extracted and embedded in PMMA. The cores were then subjected to biomechanical testing using axial loading to determine their load to failure (LOF) and linear stiffness (LS). Results: Osteoporosis was defined using qCT (vBMD < 80 mg/cm 3 ); 27 vertebrae (51%) met this criterion. With the used CT scan setting this corresponded to a HU value of 130. Non-osteoporotic vertebrae exhibited higher mean LOF (527 N vs. 319 N; ΔLOF ≈ 208 N; p < 0.001; d > 1.5). HU values correlated significantly with LOF (R 2 = 0.513; p < 0.001). Stiffness of the trabecular bone was also significantly reduced in osteoporotic samples (3.06 N/mm vs. 1.86 N/mm; p < 0.001; d = 1.54). HU values showed a strong positive correlation with stiffness in regression analysis (R 2 = 0.507; p < 0.001). Split-sample validation showed moderate generalizability of the HU-based model (R 2 training: 0.517; test: 0.366), with expected performance decline due to sample size and the simplicity of a single-parameter model. Conclusion: HU values from standard CT scans correlate well with both mechanical strength and stiffness of lumbar vertebral bodies and represent a practical, imaging-based tool for preoperative assessment of bone quality in spine surgery.

ID: AGS26/2480

A184: Mechanical properties estimation of lumbar vertebral endplates based on volumetric bone mineral density

Hanming Bian 1 , Qiang Yang 1 , chao chen 1 , Dong Zhao 1 , Gang Liu 1

1 Tianjin University Tianjin Hospital, Tianjin, China

Introduction: Cage subsidence following lumbar interbody fusion is significantly associated with the biomechanical properties of the vertebral endplate. However, there remains a lack of bone mineral density-based methods for quantitatively assessing these mechanical characteristics. To address this gap, this study aims to measure the volumetric bone mineral density (vBMD), mechanical strength, and elastic modulus of cadaveric lumbar vertebral endplates, investigate the fitting relationship between vBMD and the elastic modulus, establish a regression model, and ultimately develop an imaging-mechanical characteristic calculation method based on the bone density of the endplate region. Material and Methods: Fresh-frozen human lumbar vertebral specimens were obtained, and standardized endplate samples measuring 30 mm × 13 mm × 1.25 mm were cut from the superior and inferior vertebral endplates. Micro-CT was used to measure parameters such as vBMD, bone volume fraction (BV/TV), and bony endplate thickness. A four-point bending test was employed to measure the elastic modulus and yield strength of the endplate samples. Linear regression analysis was used to determine the correlation between the mechanical properties of the endplates and their bone density and structural parameters, and to establish regression equations between endplate bone density (vBMD), bone volume fraction (BV/TV), and elastic modulus. Results: A total of 40 lumbar vertebral endplate samples were obtained, including 20 superior endplate samples with a bony endplate thickness of 0.464 ± 0.144 mm, vBMD of 294.27 ± 58.10 mg/cm 3 , BV/TV of 0.270 ± 0.053, elastic modulus of 355.54 ± 125.83 MPa, and yield strength of 34.01 ± 0.86 MPa; and 20 inferior endplate samples with a bony endplate thickness of 0.571 ± 0.138 mm, vBMD of 367.88 ± 61.64 mg/cm 3 , BV/TV of 0.311 ± 0.061, elastic modulus of 404.59 ± 125.83 MPa, and yield strength of 34.41 ± 1.03 MPa. Univariate linear regression showed a strong correlation between vBMD (superior endplate: R 2 = 0.722, p < 0.05; inferior endplate: R 2 = 0.769, p < 0.05) and BV/TV (superior endplate: R 2 = 0.528, p < 0.05; inferior endplate: R 2 = 0.577, p < 0.05) with the elastic modulus of the vertebral endplate samples. Regression equations between BMD, BV/TV, and the elastic modulus of the endplates were established: ① Superior endplate: E = 2.57 × BMD + 403.10 × (BV/TV) - 523.05; ② Inferior endplate: E = 2.04 × BMD + 621.23 × (BV/TV) - 540.55. Conclusion: The elastic modulus of vertebral endplates is highly correlated with their vBMD and BV/TV. Linear fitting relationships between vBMD, BV/TV, and the elastic modulus of the endplates were established.

ID: AGS26/2509

A185: Lumbosacral transitional vertebrae: ffrequency and implications in radiological analysis

Micaela Besse 1 , Guillermo Ricciardi 2 , Claudio Bulos 3 , Jose Rosado Pardo 3 , Gabriel Genaro Carrioli 2 , Anibal Sarotto 1

1 Hospital General de Agudos Carlos G. Durand, Sanatorio Vitorrio Francin, Buenos Aires, Argentina

2 Hospital Alvarez., Sanatorio Guemes, Buenos Aires, Argentina

3 Sanatorio Franchin, Buenos Aires, Argentina

Introduction: Numerical variations and the lumbosacral transitional vertebrae (LSTV) have been described, being the sacralization the most frequent. Measurement of spinopelvic parameters, and even global balance parameters in patients with LSTV can be affected depending on the endplate used. The purpose of this study was to evaluate the impact of transitionel vertebrae on the measurement of spinopelvic parameters; additionally we examined the relationship between of LSTV and L5 slope. Materials and Methods: Retrospective, descriptive study of a sample of patients, radiographically evaluated with full-spine X-rays between January 2023 and January 2025 at a single center. Results: A total of 110 patients were evaluated, with a predominance of females and a mean age of 36.8 years. A transitional vertebra was identified in 49 patients (44.5%), who showed significant differences in pelvic incidence (PI) and spinopelvic parameters; notably a higher median lumbar lordosis (LL) and a higher mean PI compared to the group without LSTV. Additionally, significant differences were observed in the measurements depending on the endplate used. Conclusion: Spinopelvic parameters (PI, SS, LL, and PT) were significantly increased in patients with LSTV. The L5 slope was significantly higher in patients with sacralization, with a significantly greater L4-L5 segmental lordosis. The presence of LSTV should be identified, as it alters spinopelvic measurement values, for which full-spine imaging remains the gold standard. We suggest using the last mobile segment for these measurements.

Level of evidence: III

Key words: lumbosacral transitional vertebrae, radiologic analysis, spinopelvic parameters, radiologic meassurements

ID: AGS26/2824

A186: Forearm bone mineral density predicts adjacent segment disease following transforaminal lumbar interbody fusion similar to lumbar Hounsfield unit value in patients with lumbar degenerative disease

Qian Chen 1 , Youwei Ai 2 , Hong Ding 2

1 Affiliated Hospital of North Sichuan Medical College, Nanchong, China , 2 West China Hospital, Sichuan University, Chengdu, China

Introduction: Adjacent segment disease (ASD) develops in ≈9 % of patients after lumbar fusion and is a leading reason for revision; low bone mineral density (BMD) is a recognized risk factor. While dual-energy X-ray absorptiometry (DXA) of the lumbar spine can over-estimate BMD in degenerated spines, CT-based Hounsfield units (HU) at the index levels correlate better with ASD, yet lack validated thresholds and require supplementary imaging. Forearm DXA, unaffected by spinal degeneration, accurately reflects axial BMD and predicts osteoporotic complications, but its association with ASD has not been explored. We hypothesize that forearm BMD is an equivalent, radiation-free predictor of ASD after lumbar fusion. Material and Methods: A retrospective analysis was conducted on 336 patients who underwent TLIF for lumbar degenerative pathologies. Participants were stratified into two cohorts based on evidence of ASD: The ASD group and the non-ASD group. Demographic parameters including gender distribution, mean age, body mass index (BMI), smoking status, diabetes mellitus prevalence, and operated spinal levels were systematically recorded. BMD was assessed using dual-energy X-ray absorptiometry (DXA) at the distal one-third of the non-dominant forearm radius and ulna. Additionally, BMD was quantified as Hounsfield units (HU) via computed tomography (CT) imaging of the L1-L4 vertebral bodies. Results: At a minimum follow-up of 24 months, the prevalence of adjacent segment disease (ASD) was 11.6%. Compared to the non-ASD group, patients with ASD exhibited significantly lower BMD in the non-dominant forearm (0.325 ± 0.2 vs. 0.536 ± 0.1, p < 0.001) and lower HU value of the L1-L4 vertebral bodies (109.3 ± 29.1 vs. 140.3 ± 20.0, p < 0.001). Following propensity score matching, statistically significant differences were observed in the forearm T score, BMD and CT HU value between the two groups. In multivariate logistic regression analysis, both forearm BMD (OR 0.882; 95% CI 0.829-0.939; p < 0.001) and HU values of L1-L4 (OR 0.936; 95% CI = 0.903-0.971; p < 0.001) were identified as important risk factors for ASD. The area under the receiver operating characteristic curve (AUC) for forearm BMD and HU value in predicting ASD was 0.823 and 0.804, respectively. A forearm BMD cutoff value of 0.392 demonstrated optimal diagnostic performance for predicting ASD, with a sensitivity of 74.4% and specificity of 84.6%. Conclusion: The forearm BMD was an important risk factor for ASD. The forearm BMD was a valid predictor of ASD in patients undergoing lumbar fusion, as was the CT HU value.

ID: AGS26/410

A187: The impact of preoperative MRI-defined mild instability on outcomes after posterior open-door laminoplasty: a retrospective cohort study

Haijun Tian 1 , Shichang Zhao 1

1 Department of Orthopaedics, Shanghai Sixth People’s Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, Shanghai, China

Introduction: The standard treatment for traumatic cervical spinal cord injury (tCSCI) is dichotomous: fusion for unstable spines and laminoplasty for stable ones. This paradigm fails to address a “gray zone” of patients with mild instability, where the optimal surgical strategy is unknown. The novelty of this study lies in proposing and evaluating a specific set of preoperative MRI criteria to define this cohort, questioning whether motion-preserving laminoplasty is a safe and effective alternative to fusion. We hypothesized that for patients selected by these novel criteria, laminoplasty would provide excellent neurological recovery without compromising long-term spinal stability. Material and Methods: A retrospective cohort study was conducted on 68 patients with tCSCI treated with posterior open-door laminoplasty. The key inclusion criterion was the presence of preoperative MRI-defined mild instability, defined as having one or more of the following without signs of gross instability: 1) Indeterminate posterior ligamentous complex (PLC) injury (high signal on T2/STIR without complete disruption; SLIC DLC score = 1); 2) Minor traumatic disc injury without significant cord compression; or 3) Subtle static malalignment (vertebral translation < 2.0 mm or kyphosis < 7°). Primary outcomes were changes in ASIA Impairment Scale (AIS) grade and ASIA Motor Score (AMS). Secondary outcomes included radiographic stability, assessed by the C2-7 Cobb angle and the development of delayed instability. Mean follow-up was 12.5 months. Results: Significant neurological recovery was observed. At final follow-up, 52 patients (76.5%) improved by at least one AIS grade (p < 0.001), and the mean AMS improved significantly from 64.8 to 85.9 (p < 0.001). Crucially, radiographic stability was maintained across the cohort. There was no significant change in the mean C2-7 Cobb angle (12.8° preoperatively vs. 12.1° postoperatively, p = 0.28). No patient developed progressive kyphosis (> 10°) or delayed spondylolisthesis requiring revision surgery. The overall complication rate was low (5.2%). Conclusion: In a carefully selected cohort defined by specific preoperative MRI criteria for mild instability, posterior open-door laminoplasty is a safe and effective treatment that yields excellent neurological recovery while preserving long-term radiological stability. The use of these nuanced MRI findings to identify suitable candidates is a critical innovation, suggesting that this motion-preserving technique is a viable and advantageous alternative to instrumented fusion for this specific “gray zone” population.

ID: AGS26/594

A188: Impact of screw diameter on the mechanical performance of cement-augmented lumbar spinal fusion constructs

Akar Jani 1 , Nicole DeVries Watson 1 , Douglas Fredericks 1 , Catherine Olinger 1

1 University of Iowa Hospitals and Clinics, Iowa City, United States

Introduction: Pedicle screw instrumentation (PSI) is a well-established surgical approach to treat degenerative spinal conditions. However, osteoporosis severely compromises bone quality, reducing screw fixation strength—measured as pullout strength—by up to 80%. Cement augmentation with polymethylmethacrylate (PMMA), delivered via fenestrated pedicle screws, reinforces the screw-bone interface and can improve pullout strength up to twofold. Increasing screw diameter enhances fixation strength by increasing surface area of the screw-bone interface, but in osteoporotic vertebrae, the benefit of larger screws with cement augmentation may be limited. Augmentation increases construct rigidity, potentially concentrating stress at the screw-bone interface and reducing the advantage of larger screws. Thus, the benefit of larger screws in cement-augmented fixation remains unclear. This study utilizes anatomically accurate osteoporotic-grade synthetic L3 lumbar vertebrae (Sawbones®, Pacific Research Labs) to evaluate how increased screw diameters in cement-augmented constructs affect performance following cyclic loading. We hypothesize that larger-diameter fenestrated screws will have greater cement distribution within cancellous bone, enhancing construct stability and pullout strength. Materials and Methods: Fifty-millimeter length titanium polyaxial fenestrated pedicle screws of 5.5 mm (n = 8) or 6.5 mm (n = 8) diameter were bilaterally implanted into osteoporotic L3 lumbar Sawbones. 3D-printed drill guides ensured consistent screw trajectories parallel to the vertebral endplates. Each fenestrated screw was augmented with 3 cc of PMMA. Non-fenestrated 6.5 mm screws (n = 8) were used for non-augmented constructs. All constructs were stabilized with 100 mm pre-bent rods and mounted to a Material Testing System (MTS) following modified ASTM F1717-21. Implants underwent 90,000 cycles of compressive loading (100-200 N at 3 Hz) with axial rotation (± 1.5° at 1.8 Hz) to simulate three months of walking at 110 steps/min. Additional cement-augmented fenestrated screws (5.5 mm and 6.5 mm, n = 2 each) not subjected to loading served as passive controls. CT scans before and after loading assessed damage, screw-hole volume changes, and cement distribution. Pullout strength was measured according to modified ASTM F543-23. Results: Initial CT confirmed intact cortical shells after screw implantation. Cement injection caused superior endplate extravasation in one of four vertebrae with 5.5 mm fenestrated screws and three of four vertebrae with 6.5 mm fenestrated screws. All screws withstood cyclic loading. Comparison between left and right screws showed no significant differences in pullout strength for 5.5 mm screws (left: 1726.23 ± 333.12 N, right: 1672.72 ± 285.36 N, p = 0.64) or 6.5 mm screws (left: 1806.20 ± 314.90 N, right: 2033.68 ± 106.70 N, p = 0.12). 6.5 mm screws demonstrated higher mean pullout strength than 5.5 mm screws (1919.94 ± 249.37 N vs 1699.47 ± 288.73 N), but this difference was not significant (p = 0.124). Conclusion: Preliminary findings suggest increasing screw diameter does not significantly improve biomechanical performance in cement-augmented fixation. Larger screws were associated with more frequent extravasation, indicating elevated structural risk without increased fixation strength. This study uses osteoporotic Sawbone, standardized screw placement, and combined compression with axial rotation to better replicate physiological conditions. Compared with prior cadaveric models, this novel approach provides a controlled, clinically relevant evaluation of cement-augmented pedicle screw constructs.

ID: AGS26/725

A189: Predicting cage subsidence after TLIF using HU: a combined model with PROMs and machine learning validation

Elie Najjar 1 , Khalid Salem 1

1 Queen's Medical Centre, Nottingham University Hospitals, Nottingham, United Kingdom

Background: Cage subsidence is a common radiographic finding following transforaminal lumbar interbody fusion (TLIF), but its clinical significance remains uncertain. While Hounsfield Unit (HU) values from preoperative CT scans have emerged as surrogates for local bone quality, few studies have validated HU-based models for predicting subsidence risk or explored their impact on functional outcomes. We aimed to develop and validate a simple, clinically applicable model for predicting cage subsidence using HU measured at the fused vertebral level, and to assess the relationship between subsidence and patient-reported outcomes. Methods: This retrospective cohort study included 76 patients who underwent single-level TLIF between 2019 and 2023 with complete preoperative CT imaging and 12-month clinical follow-up. HU values were measured at the fused vertebral bodies, and cage subsidence was defined as ≥ 2 mm vertical migration or disc height loss on CT at 6-12 months. Logistic regression and ROC analysis were used to derive a threshold-based predictive model. Outcomes included the Oswestry Disability Index (ODI), EQ-5D VAS, and patient satisfaction. Results: The overall subsidence rate was 31.6%. HU at the fused level was significantly lower in patients with subsidence (108.3 vs. 135.7, p = 0.048). Logistic regression confirmed HU as an independent predictor (β = −0.0121, p = 0.023); each 10-unit decrease in HU increased the odds of subsidence by 12.1%. A threshold of HU < 130 identified a high-risk group with a 50.0% subsidence rate, compared to 22.0% in the low-risk group (AUC = 0.68). No significant differences in ODI, EQ-5D, or satisfaction were observed between groups at 12 months. Conclusion: HU measured at the fused vertebra is a reliable predictor of cage subsidence following TLIF. A simplified threshold model offers a reproducible tool for preoperative risk stratification. Importantly, radiographic subsidence did not correlate with poorer functional outcomes, reinforcing the need to focus on patient-centered metrics in surgical success.

OP22: Complex Cervical Myelopathy

ID: AGS26/1434

A190: Deep learning fusion of operative notes with structured preoperative data improves 2-year outcome prediction for cervical spinal myelopathy surgery: a QOD study

Eunice Yang 1 , Elan Schonfeld 1 , Praveen Mummaneni 2 , Dean Chou 1 , Mohamad Bydon 3 , Erica Bisson 4 , Christopher Shaffrey 5 , Oren Gottfried 5 , Anthony Asher 6 , Domagoj Coric 6 , Eric Potts 7 , Kevin Foley 8 , Michael Wang 9 , Kai-Ming Fu 10 , Michael Virk 10 , John Knightly 11 , Scott Meyer 11 , Paul Park 8 , Cheerag Upadhyaya 12 , Chun-Po Yen 13 , Juan Uribe 14 , Luis Tumialan 14 , Jay Turner 14 , Regis Haid 15 , Andrew Chan 1

1 Columbia University, New York, United States

2 UCSF, San Francisco, United States

3 University of Chicago, Chicago, United States

4 University of Utah, Salt Lake City, United States

5 Duke University, Durham, United States

6 Carolina Neurosurgery & Spine Associates, Charlotte, United States

7 Goodman Campbell, Indianapolis, United States

8 Semmes-Murphey, Memphis, United States

9 University of Miami, Miami, United States

10 Cornell University, New York, United States

11 Maxim Spine, Morristown, United States

12 University of North Carolina, Chapel Hill, United States

13 University of Virginia, Charlottesville, United States

14 Barrow Neurological Institute, Phoenix, United States

15 Atlanta Brain and Spine, Atlanta, United States

Introduction: Predicting which patients are likely to reach the minimum clinically important difference (MCID) following surgery for cervical spinal myelopathy (CSM) remains challenging due to the limitations of structured data. Multimodal deep learning enables the integration of structured data with surgical notes, offering models that more comprehensively analyze factors contributing to the achievement of MCID. This study aims to 1) determine the optimal encoding and dimensionality-reduction techniques for training fusion models and 2) evaluate model performance in predicting 24-month MCID achievement across patient-reported outcomes (PROs). Material and Methods: We included 1,141 patients from the Quality Outcomes Database (QOD) who had undergone surgery for CSM. MCID thresholds were applied to 2-year patient-reported outcomes (PROs), including the modified Japanese Orthopaedic Association (mJOA), Neck Disability Index (NDI), EuroQol visual analogue scale (EQ-VAS), EuroQol health survey (EQ-5D), and the visual analogue scales for arm pain (VAS-AP) and neck pain (VAS-NP). Operative notes were encoded using both TF-IDF vectorization and BlueBERT. For each PRO, feed-forward neural networks were trained using structured data alone along with three fused datasets for each reduction technique: full, reduced (95% explained variance), and minimal (75% explained variance). Results: Fusion models combining structured data with operative note embeddings could outperform models trained on structured data alone, though the magnitude of improvement varied across outcomes. For EQ-5D, the strongest baseline model using structured data achieved an AUC of 0.82 ± 0.04. Adding operative note embeddings produced a modest gain, with the full TF-IDF fusion model reaching an AUC of 0.85 ± 0.03. The most pronounced improvement was observed for EQ-VAS, where the baseline AUC of 0.73 ± 0.05 increased to 0.82 ± 0.04 following use of the BERT encoder. Model performance on metrics of functional performance was weakest at baseline. Structured-data only models yielded AUCs of 0.74 ± 0.04 for NDI and 0.69 ± 0.05 for mJOA. The TF-IDF fusion model improved NDI MCID prediction to an AUC of 0.76 ± 0.04, and the BERT fusion model demonstrated a substantial improvement in mJOA MCID prediction to an AUC of 0.76 ± 0.04. By contrast, pain-related outcomes showed the least benefit from the operative note fusion method. For VAS-AP and VAS-NP, baseline AUCs were 0.82 ± 0.04 and 0.82 ± 0.03, respectively. The BERT fusion models yielded modest gains for VAS-AP with an AUC of 0.84 ± 0.03, though VAS-NP demonstrated an AUC of 0.82 ± 0.03. Conclusion: Operative notes represent rich repositories of information that capture dimensions of patient health not captured by structured clinical variables. In this study, fusion models integrating operative note embeddings substantially improved long-term MCID prediction for EQ-VAS and mJOA. However, models with already strong baseline performance using structured data showed limited additional benefit from the fusion technique. These findings highlight the value of surgical documentation as a complementary data source for AI tools, ultimately enhancing our ability to identify patients most likely to benefit from surgery.

ID: AGS26/1452

A191: Socioeconomic and clinical factors predict 5-year loss of quality-of-life benefits following surgery for cervical spinal myelopathy: a QOD machine learning analysis

Eunice Yang 1 , Elan Schonfeld 1 , Praveen Mummaneni 2 , Dean Chou 1 , Mohamad Bydon 3 , Erica Bisson 4 , Christopher Shaffrey 5 , Oren Gottfried 5 , Anthony Asher 6 , Domagoj Coric 6 , Eric Potts 7 , Kevin Foley 8 , Michael Wang 9 , Kai-Ming Fu 10 , Michael Virk 10 , John Knightly 11 , Scott Meyer 11 , Paul Park 8 , Cheerag Upadhyaya 12 , Chun-Po Yen 13 , Juan Uribe 14 , Luis Tumialan 14 , Jay Turner 14 , Regis Haid 15 , Andrew Chan 1

1 Columbia University, New York, United States

2 UCSF, San Francisco, United States

3 University of Chicago, Chicago, United States

4 University of Utah, Salt Lake City, United States

5 Duke University, Durham, United States

6 Carolina Neurosurgery & Spine Associates, Charlotte, United States

7 Goodman Campbell, Indianapolis, United States

8 Semmes-Murphey, Memphis, United States

9 University of Miami, Miami, United States

10 Cornell University, New York, United States

11 Maxim Spine, Morristown, United States

12 University of North Carolina, Chapel Hill, United States

13 University of Virginia, Charlottesville, United States

14 Barrow Neurological Institute, Phoenix, United States

15 Atlanta Brain and Spine, Atlanta, United States

Introduction: Quality of life measures are particularly valuable in spine surgery, uniquely capturing patients’ overall physical, mental, and social wellbeing while allowing for objective comparisons across conditions. Initial achievement of the minimum clinically important difference (MCID) in quality-of-life scores may reflect postoperative improvement, but sustained benefit is crucial. Identifying risk factors for long-term decline is essential for expectation setting and appropriate tailoring patient care. Our study thus evaluates preoperative and early postoperative factors associated with EuroQol-5D (EQ-5D) decline between 24 and 60 months. Material and Methods: Our retrospective analysis utilizes the prospective, multicenter Quality Outcomes Database (QOD) cervical spinal myelopathy (CSM) dataset. We included patients who both (1) achieved the EQ-5D MCID at 24 months and (2) had 60-month EQ-5D data collected. The change in EQ-5D between 24 and 60 months was calculated for each patient. Patients at risk of deterioration were defined as those with a decline greater than or equal to a half standard deviation below the mean change, a cutoff that is consistent with the threshold commonly used to define the minimum clinically important difference. To compare the stable and deterioration cohorts, we performed univariate and multivariable analysis (including variables reaching p < 0.20), with the results subsequently validated via linear regression and XGBoost models on continuous 24-to-60-month EQ-5D change. Results: Of the 1085 patients in the initial dataset, 248 patients met our inclusion criteria of EQ-5D MCID achievement at 24 months and EQ-5D collected at 60 months. From this study cohort, 69 (27.8%) patients were considered at risk of deterioration and 179 (72.2%) were considered stable at 60 months. Multivariate analysis identified a significant association between 60-month EQ-5D falloff and arthritis (OR = 2.40, p = 0.01) and higher baseline NDI (Beta = 0.003, p = 0.04), while Medicare insurance was associated with a higher likelihood of 60-month MCID fall-off (OR = 0.51, p = 0.04). Notably, age itself was not a significant predictor of fall-off (p > 0.05). Linear regression analysis and Boruta feature selection on 24-to-60-month change in EQ-5D both showed arthritis as a predictive factor of deterioration risk as well. A SHAP explanation analysis of the XGBoost model trained on features identified as confirmed or tentative using Boruta confirmed the directionality of arthritis as negative for long-term preservation of post-surgical EQ-5D improvements. Conclusion: A proportion of patients who achieve MCID at 24 months may not maintain this improvement long term. Among preoperative characteristics, arthritis emerged as the most significant driver of long-term deterioration, possibly reflecting systemic arthritic disease that worsens QOL beyond the spine alone. Notably, this suggests arthritis outweighs the impact of other common comorbidities in CSM. Future work should evaluate whether this effect differs by type of arthritis, such as osteoarthritis versus rheumatoid arthritis, and the extent to which medical management of these conditions may modify long-term outcomes. Awareness of such socioeconomic and clinical risk factors may help surgeons anticipate which patients are most vulnerable to long-term decline and identify candidates for targeted interventions to sustain quality-of-life improvements.

ID: AGS26/2503

A192: Tandem degenerative spinal stenosis: predicting outcomes using machine learning from a 12-year experience at a tertiary care centre

Subrat Mohapatra 1 , Manoj Phalak 2 , Het Shah 2 , Sachin Borkar 2 , Shashank S Kale 2

1 IMS and SUM Hospital, Bhubaneshwar, India , 2 All India Institute of Medical Sciences (AIIMS), Delhi, India

Introduction: Concurrent cervical and lumbar stenosis, a.k.a tandem spinal stenosis (TSS), is a rare degenerative disease, affecting multiple spinal levels with paucity of literature describing the incidence, natural history, and operative management strategy. The aim was to study the frequency, natural history, and outcome of TSS in surgically treated patients and to design a surgical outcome clinical scoring tool to be used as an adjunct to clinical decision-making in patients with TSS. Materials and Methods: The study was conducted in the Department of Neurosurgery, All India Institute of Medical Sciences (AIIMS), New Delhi. A retrospective review of clinical and radiological hospital records was done for patients treated surgically between January 2007 and December 2020. Functional outcome assessed by the Nurick score, mJOA, and ODI. Cervical surgery included either anterior (discectomy or corpectomy and fixation) or posterior (laminoplasty or laminectomy and fixation), and discectomy ± laminectomy was done for lumbar afflictions. Machine learning using the GPT-4o language learning model was used to create a predictive model for determining surgical success using a combination of pre-operative parameters based on the 12-year dataset. Results: Out of 1673 cases of cervical and/or lumbar degenerative disease surgically treated during this period, concurrent stenosis was identified in 122 cases (96 male, 26 female), out of which three cases were excluded due to mortality/ loss to follow up in the postoperative period (n = 119). The mean age was 59 ± 11.8 years (27-83 years) with a mean postoperative follow-up of 14.25 months (5-60 months). Only cervical decompression was done in 73 patients (61%), and only lumbar surgery was done in 23 patients (19%). Both surgeries were done in 23 patients (19%). Of the patients requiring surgery at both levels, single-stage cervical and lumbar surgery was done in 12 patients (10%), and staged surgery (cervical followed by lumbar) was done in the remaining cases. A total of 84 patients underwent cervical surgery (70.58%), out of which 11 patients needed lumbar surgery in follow-up. The mean interval between the staged procedures was 17 months. The difference in pre- and post-operative functional outcomes on ODI, Nurick score, and mJOA scores independently in all groups of surgery was significant (p < 0.05). The functional outcome difference between the single-stage and two-stage surgery was insignificant (p > 0.05). The clinical scoring tool helps predict surgical benefit in patients with TSS, with a negative predictive value of 69.23%. It can be used as an adjunct for deciding the surgical strategy in TSS patients. Conclusion: Although the incidence of coexisting stenosis is low, but should not be overlooked. Deciding whether to operate on either cervical or lumbar or both levels is a vital dilemma in patients with TSS, and the TSS surgical outcome predictive tool can help in identifying patients who don't require two-level surgery and thus avoid the surgical morbidity associated with both-level surgery.

ID: AGS26/3344

A193: Obstructive sleep apnea and 30-day outcomes after posterior cervical decompression and fusion: insights from a national propensity-matched analysis

Omar Diab 1 , Pouya Alijanipour 2

1 University of Texas Medical Branch, Galveston, United States

2 Department of Orthopedics, University of Texas Medical Branch, Galveston, United States

Introduction: Obstructive sleep apnea (OSA) is common and has been linked to increased perioperative cardiopulmonary complications. However, its relationship to postoperative outcomes following elective posterior cervical decompression and fusion (PCDF) remains uncertain. Leveraging a large national database, this study aimed to evaluate whether a preoperative diagnosis of OSA is associated with 30-day postoperative complications after elective PCDF and to assess the extent to which comorbid conditions linked to OSA may mediate this association. Materials and Methods: We conducted a retrospective, propensity-matched cohort analysis using the TriNetX Research Network. Adult patients (18-80 years) who underwent PCDF between January 2010 and September 2025 were included. The exposure group comprised patients with a documented diagnosis of OSA (within two years prior to surgery), while the control group consisted of patients without OSA. Patients with cervical trauma, malignancy, spinal cord tumors, active infection within 30 days, or those undergoing anterior cervical discectomy and fusion were excluded. All patients were required to have at least one postoperative follow-up within 30 days. Primary outcomes were acute respiratory failure and prolonged mechanical ventilation within 30 days. Propensity score matching (1:1, greedy nearest-neighbor, caliper = 0.1) was performed in two stages. In the first stage, patients were matched based on demographics and procedural variables (number of additional fusion levels and concurrent laminectomy). In the second stage, matching was repeated with the addition of major comorbidities (hypertension, diabetes, chronic kidney disease, obesity, and COPD). Risk ratios (RRs), risk differences (RDs), and 95% confidence intervals (CIs) were calculated, with significance determined using two-sided z-tests and chi-square tests (p < 0.05). Results: A total of 3,586 patients with OSA and 17,268 patients without OSA undergoing PCDF were included. Patients with OSA had significantly higher rates of hypertension (83.4% vs 60.3%), obesity (56.4% vs 21.4%), diabetes (47.5% vs 25.1%), chronic kidney disease (17.4% vs 8.8%), and COPD (20.2% vs 10.8%) compared with those without OSA (p < 0.001 for all). After propensity score matching on demographics and procedural variables, 3,582 patients remained in each group. In this matched cohort, OSA was associated with higher risks of surgical site infection (3.4% vs 2.4%), acute kidney failure (4.7% vs 3.3%), acute respiratory failure (3.8% vs 2.6%), and hospital readmission (14.7% vs 12.8%). However, after additional matching for comorbidities all outcomes were comparable within the two groups. Conclusion: OSA is associated with a substantially higher burden of metabolic and cardiopulmonary comorbidities among patients undergoing PCDF. Apparent associations between OSA and postoperative complications, including infection, renal failure, and respiratory failure, are significantly attenuated after accounting for these comorbidities. These findings highlight the need for OSA to be considered alongside its associated comorbid conditions rather than an isolated risk factor, and its presence alone should not preclude patients from undergoing PCDF.

ID: AGS26/3460

A194: Angiotensin receptor blockers are associated with lower pseudarthrosis and revision rates after multilevel posterior cervical fusion: a propensity-matched analysis

Seungjun Lee 1 , Matthew Kim 2 , Graham Goh 2 , Tony Tannoury 2 , Chadi Tannoury 2

1 Boston University Chobanian and Avedisian School of Medicine, Boston, United States

2 Department of Orthopaedic Surgery, Boston University Medical Center, Boston, United States

Introduction: The renin-angiotensin system (RAS) influences bone remodeling. Angiotensin receptor blockers (ARBs) and angiotensin-converting enzyme inhibitors (ACEIs) act via distinct pathways that could differentially affect arthrodesis and perioperative outcomes after posterior cervical fusion (PCF). However, their impact on PCF outcomes remains unclear. Therefore, the purpose of this study was to compare perioperative complications, revision rates, and resource utilization in hypertensive adults undergoing multilevel PCF treated with ARBs versus ACEIs. We hypothesized that ARB therapy would be associated with lower pseudarthrosis and infection rates without a concomitant increase in adverse events. Material and Methods: The TriNetX Global Collaborative Network was queried to identify hypertensive adults undergoing multilevel PCF using CPT and ICD-10 codes. Cohorts were defined as ARB users or ACEI users based on prescriptions within 6 months before and after surgery. Propensity score matching (1:1) controlled for age, sex, BMI, smoking, osteoporosis, HbA1c, and Charlson Comorbidity Index comorbidities. Outcomes assessed included medical complications at 90 days, and surgical complications, revision rates, and resource utilization at 90 days, 6 months, 1 year, and 2 years. Results: After matching, 4,315 patients were included in both ARB and ACEI cohorts (n = 8,630). ARB use was associated with lower pseudarthrosis rates at 90 days (OR 0.879, p = 0.046), 6 months (OR 0.869, p = 0.023), 1 year (OR 0.843, p = 0.004), and 2 years (OR 0.858, p = 0.008). ARB patients also had lower revision rates at 1 year (OR 0.814, p = 0.049) and 2 years (OR 0.792, p = 0.010). In addition, ARB use was associated with reduced 90-day acute kidney injury (6.0% vs 7.1%; OR 0.837, p = 0.041). No significant differences were observed for other medical or surgical complications, readmissions, or ED utilization. Conclusion: ARB therapy was associated with significantly lower pseudarthrosis rates across all timepoints and fewer revisions at 1-2 years after multilevel PCF. A modest reduction in acute kidney injury was also observed. These findings suggest that RAS inhibitor class influences spinal fusion outcomes and perioperative risk. Prospective studies are warranted to confirm these associations and to guide perioperative optimization in hypertensive spine surgery patients.

ID: AGS26/524

A195: The submandibular retropharyngeal approach: a safe corridor to the upper anterior cervical spine - Anatomical stepwise dissection and clinical case illustration

Alexander F. Küffer 1 , Manuel A. Molina-Vega 2 , Mohammad H. Nour El Dine 3 , Harry Van Loveren 3 , Siviero Agazzi 3 , and Davide Croci 1,3

1 Department of Neurosurgery,

2 Department of Surgical Oncology, Lakeland Regional Medical Center, Lakeland,United States,

3 Department of Brain and Spine, University of South Florida, Tampa, United States

Introduction: Surgical access to the anterior upper cervical spine and craniovertebral junction is technically demanding due to limited space and the presence of critical neurovascular and visceral structures. Traditional anterior routes, such as the transoral, transnasal or anterolateral approaches, are associated with limitations in exposure and increased morbidity. The submandibular retropharyngeal approach (SRA) has been proposed as an alternative to access the upper anterior cervical spine and the cranio-cervical junction, but detailed anatomical documentation and clinical validation remain limited. Materials and Methods: A cadaveric head specimen was prepared with colored vascular injections to enhance anatomical delineation. Stepwise dissection was performed to document the surgical corridor of the SRA, complemented by high-resolution photography and illustrations. Endoscopic assistance was employed to extend cranial visualization to the anterior arch of C1. The approach was compared with transoral, transnasal, and anterolateral techniques in terms of exposure, safety, and surgical accessibility. A clinical case of a patient undergoing anterior C2 corpectomy for metastatic destruction of the C2 vertebral body was presented to illustrate practical application. Results: The SRA provided a safe, direct, and oblique surgical trajectory to the anterior C2 vertebral body and craniovertebral junction, with reduced need for pharyngeal retraction and preservation of the pharyngeal plexus. Endoscopic assistance further enhanced cranial exposure. The illustrative case demonstrated successful anterior reconstruction with a bone strut graft and plate fixation, achieving spinal stability without significant postoperative dysphagia or speech deficits. Conclusion: The submandibular retropharyngeal approach represents a viable alternative to traditional anterior routes for upper cervical spine surgery, combining effective exposure with reduced risk to critical structures. Its integration into neurosurgical practice may broaden surgical options for complex anterior cervical pathologies.

ID: AGS26/661

A196: Postoperative neurologic and functional outcomes in myasthenia gravis patients undergoing cervical spine surgery for myelopathy

Hanna von Riegen 1 , Erica Greiner 1 , Mario Keko 1 , Aimee Boegle 2 , Jason Pittman 1 , Sapan Gandhi 1

1 Department of Orthopedics, Beth Israel Deaconess Medical Center, Harvard Medical School in Boston, MA, Boston, United States

2 Department of Neurology, Beth Israel Deaconess Medical Center, Boston, United States

Introduction: Myasthenia Gravis (MG) is an autoimmune neuromuscular disorder associated with potential postoperative complications, yet data on surgical outcomes in this population remains limited. This study aimed to evaluate whether MG patients experience worse neurologic and functional outcomes following cervical spine surgery compared to non-MG patients. Material and Methods: Using the TriNetX research platform, we conducted a retrospective, multicenter, cohort study. Adult patients with MG (ICD-10: G70.0) who underwent cervical spine surgery for degenerative myelopathy (ICD: M47.12, M48.02) were matched 1:1 to non-MG controls based on age, sex, race, ethnicity and BMI. Patients with malignant bone diseases were excluded. In exploratory analyses, we further evaluated MG subgroups based on preoperative immunosuppressive therapy and included a lumbar spine surgery cohort to assess systemic versus procedure-specific risks. Results: A total of 290 matched patient pairs were analyzed. The average age of the patients in both groups was 65.9 ± 10.5 years, with the majority being white (76.3%), and male (54.0%). MG patients had a significantly higher risk of postoperative dysphagia within 3 months (21.5% vs. 10.9%, p = 0.001) and 1 year (31.4% vs. 15.7% , p < 0.001). Additionally, MG patients showed significantly increased rates of motor and nerve deficits (18.6% vs. 8.4%, p = 0.007), plexus disorders (9.1% vs. 5.2%, p = 0.001), and fatigue at 1 year (15.0% vs. 8.8%, p = 0.025). Post-operative paresis and muscle weakness were higher among MG patients (7.3% vs. 3.0%, p = 0.070, and 13.9% vs. 8.8%, p = 0.059, respectively), although not statistically significant. In the immunosuppression subgroup, most outcomes were comparable, though fatigue was significantly more common among immunosuppressed MG patients. In the lumbar surgery comparison group, MG patients also had higher rates of postoperative dysphagia and neurologic complications, suggesting a broader perioperative vulnerability beyond the cervical procedure itself. Conclusion: MG patients undergoing cervical spine surgery for cervical myelopathy are at increased risk for postoperative neurologic and functional complications. These findings suggest a higher rate of residual or ongoing neurologic conditions in MG patients following surgery and highlight the importance of perioperative planning, MG-specific risk assessment, and multidisciplinary care strategies in this high-risk patient group. Exploratory findings indicate that these risks may not be limited to the cervical spine and could reflect systemic effects of MG, underscoring the need for tailored perioperative strategies.

ID: AGS26/732

A197: Dual-rod modification in posterior cervical instrumentation: minimum 5-year follow-up results

Mehmet Aydoğan 1 , Abdülhalim Akar 1 , Muhammed Fatih Serttaş 2

1 Bosphorus Spine Center, İstanbul, Turkey , 2 Kocaeli City Hospital, Kocaeli, Turkey

Introduction: Posterior cervical instrumentation extending from the upper cervical to the cervicothoracic region often requires excessive rod bending and wide soft tissue dissection, particularly at the C6 level. These technical challenges may complicate the procedure, increase operative time, and necessitate the use of lateral connectors to bridge pedicle and lateral mass screws. To overcome these limitations, we developed a modified double-rod technique for posterior cervical instrumentation and evaluated its clinical and radiological outcomes in patients with multilevel cervical spinal stenosis. Material and Methods: Between 2012 and 2020, 63 patients (mean age: 68.1 years, range: 45-78) with at least three-level cervical spinal stenosis underwent posterior cervical decompression and stabilization using the modified double-rod technique. Of these, 57 patients underwent posterior-only decompression, while 6 patients underwent combined anterior-posterior decompression (3 discectomy, 3 corpectomy). Standard posterior approaches were used to insert pedicle and lateral mass screws, which were then connected by two rods on each side, forming a four-rod construct. Radiological evaluation included measurement of cervical lordosis (C2-C7) on anteroposterior and lateral radiographs, and assessment of fusion by CT and radiographs. Clinical outcomes were assessed using the Japanese Orthopaedic Association (JOA) score, Neck Disability Index (NDI), and Visual Analog Scale (VAS) for arm and neck pain. Mean follow-up was 118.7 months (range: 60-156). Results: All patients had multilevel cervical stenosis, with a mean of 3.4 decompressed levels (range: 3-4). Cervical lordosis improved significantly from 12.2° preoperatively to 20.7° at final follow-up (p < 0.001). Mean JOA score increased from 11.1 to 16.3 (p = 0.003), while NDI improved from 35.6 to 10.1 (p < 0.001). VAS scores decreased from 5.8 (arm) and 6.3 (neck) to 1.1 and 1.9, respectively (both p < 0.001). CT scans showed three cases of suspected delayed fusion; however, no loosening or implant failure was detected, and solid fusion was ultimately confirmed in all patients. No pseudoarthrosis was observed. Conclusion: The modified posterior cervical double-rod technique facilitates easier connection between pedicle and lateral mass screws, reduces the need for extensive rod contouring and soft tissue dissection, and avoids the use of lateral connectors. With excellent fusion rates, significant improvements in clinical scores, and a low complication profile, this technique represents a safe and effective option for the surgical management of multilevel cervical spinal stenosis.

ID: AGS26/974

A198: Developmental cervical spinal stenosis in degenerative cervical myelopathy: distinct clinical features, surgical patterns, and recovery ceiling effect

Rui Chen 1 , Xiaoxiong Yang 1 , Feifei Zhou 1

1 Peking University Third Hospital, Beijing, China

Introduction: Developmental cervical spinal stenosis (DCS) is increasingly recognized as an important modifier of degenerative cervical myelopathy (DCM). However, whether DCS influences baseline severity, surgical strategies, and postoperative recovery remains incompletely understood. This study aimed to compare clinical characteristics, surgical patterns, and functional outcomes between DCS and non-DCS (nDCS) patients, and to explore the potential “ceiling effect” of recovery in DCS. Material and Methods: A retrospective cohort of cervical surgery patients was analyzed. Baseline comorbidities, symptoms/signs, functional scores [modified Japanese Orthopaedic Association (mJOA), Neck Disability Index (NDI), Visual Analogue Scale (VAS), and SF-36], surgical approaches, and follow-up outcomes (3-24 months) were compared between DCS and nDCS groups. Chi-square/Fisher tests and logistic regression were applied, with odds ratios (ORs) and 95% confidence intervals (CI) calculated. Results: DCS was significantly associated with ossification-related pathologies [OPLL: OR = 1.65, 95%CI 1.28-2.13; DISH: OR = 3.76, 95%CI 1.14-12.40], and hypertension (OR = 1.44, 95%CI 1.07-1.94). Symptomatically, DCS patients more frequently exhibited limb numbness, lower limb weakness, band-like sensation, and “stepping-on-cotton” gait (all p < 0.05), but less neck pain. Baseline mJOA scores were lower in DCS (12.6 vs. 13.4, p < 0.001), while NDI and VAS showed no difference. DCS patients more often required posterior (44.9% vs. 26.9%) or combined approaches (20.4% vs. 10.5%, both p < 0.001). During follow-up, DCS patients showed greater absolute mJOA improvement, but a smaller proportion achieved ≥ 75% recovery, especially at 12 months (35.8% vs. 46.9%, p = 0.035), reflecting a ceiling effect. Conclusion: DCS represents a distinct clinical entity within DCM, characterized by ossification-related comorbidities, sensory-motor dominant symptomatology, and greater need for extensive surgery. Although DCS patients can achieve marked postoperative improvement, their likelihood of reaching the highest recovery thresholds is limited, likely due to baseline low scores and structural constraints. Recognizing this recovery ceiling may inform surgical decision-making and prognostic counseling, while highlighting DCS as a target for refined imaging and predictive modeling

OP23: Thoracic/Cervical MIS

ID: AGS26/1071

A199: Risk factors for reoperation after minimally invasive tubular transforaminal lumbar interbody fusion: a cohort study of 756 patients

Aaron Lerch 1,2 , Martin Wood 1 , Anthony Chau 1

1 Brisbane Clinical Neuroscience Centre, Brisbane, Australia , 2 University of Queensland, Brisbane, Australia

Introduction: There is a paucity of high-volume series identifying the complication profile and risk factors for reoperation after minimally invasive spinal (MIS) surgery. We analysed our department’s experience with short-term (< 30 days) and long-term (> 30 days) reoperation for lumbar MIS tubular transforaminal lumbar interbody fusions (TLIFs). Material and Methods: All tubular TLIF procedures from 2011 to February 2024 were retrospectively reviewed, with incidence of reoperation determined from patient follow-up until March 2024. Eight neurosurgeons performed operations at our single centre. Two case-controls per complication were manually selected. Each control had undergone the same primary procedure within six months of the index case, at the same levels and, if possible, by the same surgeon. Parameters assessed included age, sex, BMI, frailty (measured by the Modified Charlson Comorbidity and mFI-5), comorbidities (including diabetes, hypertension, osteoporosis/osteopenia, depression/anxiety and smoking/drinking status), private/public status, workplace injury, preoperative antithrombotics, preoperative platelet, haemoglobin, cholesterol and fasting glucose levels, the symptomatic indications, prior spinal surgery, preoperative imaging findings, side of approach, operative time, use of postoperative drains/antibiotics, length of stay, time to follow-up, and clinical outcome. All variables were assessed with a series of univariate logistic regressions, with a broad statistical threshold of p < 0.2 for identifying candidates as per convention. Multivariate logistic regressions were then conducted to reveal true associations. Results: From our database of 756 patients who underwent tubular TLIF, there was a 6.6% reoperation rate (n = 50) at a mean follow-up of 1.5 years (SD 1.7 years). Short-term reoperation was predominantly for cage migration and had no statistical risk factors. Long-term reoperation was also mainly for cage migration and was associated with private status, workcover status, osteoporosis/osteopenia, and previous lumbar surgery (at a different level). Combined timeframe reoperations were associated with workcover status, osteoporosis/osteopenia, diabetes, and previous lumbar surgery (at a different level). Patients who had reoperations fared worse than their matched controls. Additionally, a worse patient outcome was associated with diabetes, a higher Modified Charlson Comorbidity Index, and previous lumbar surgery (at the same level). Conclusion: This retrospective case-control series reports our institution’s complication profile and risk factors for reoperation and poor patient outcomes after tubular TLIFs. Our results suggest reoperation is associated with osteoporosis/osteopenia, diabetes, private-funding, being a workcover patient, and previous lumbar surgery at another spinal level.

ID: AGS26/101

A200: Endoscope-assisted anterior cervical discectomy and fusion (ACDF) outperforms microscope-assisted ACDF in improving postoperative swallowing function with the use of zero-P cage

Yi-Hao Liang 1 , Jialin Liu 1 , Shudong Chen 1

1 Minimally Invasive Spine Center, Orthopaedic Hospital of Guangdong Provincial Hospital of Chinese Medicine, Guangzhou University of Chinese Medicine, Guangzhou, China., Guangzhou, China

Introduction: This study aimed to compare the efficacy of Endoscopic-assisted Anterior Cervical Discectomy and Zero-P Fusion (ACDF) and Microscope-assisted Zero-P ACDF in treating single-segment Cervical spondylotic myelopathy and their effects on postoperative swallowing function recovery using clinical and imaging evaluations. Material and Methods: The retrospective study included 68 patients with single-segment CSM who had surgical treatments from April 2020 to August 2023. Among these, 34 got Endoscope-assisted ACDF (EA-ACDF) with the use of Zero-P fusion cage, 34 got Microscope-assisted Anterior Cervical Discectomy and Zero-P Fusion (MA-ACDF) with the use of Zero-P fusion cage. Age, gender, BMI, surgical duration and blood loss were documented. mJOA assessed functional recovery. DSQ and EAT-10 evaluated swallowing function, and PSTSR change rate measured soft tissue structure changes before and after surgery. Results: 68 patients were equally split into EA-ACDF and MA-ACDF groups. No statistical differences were found in age, gender, BMI, surgical segment and follow-up time between them. EA-ACDF had a significantly shorter hospital stay than MA-ACDF (p < 0.05). Its operation time and intraoperative blood loss showed no significant differences compared with the ACDF (p = 0.07, 0.72). The mJOA scores of both groups improved significantly after surgery (p < 0.05). There were no significant differences in Nurick scores and mJOA between the two groups. The EA-ACDF’s EAT-10 and DSQ was better than that of the ACDF at each post-surgery time point, yet not statistically significant. The EA-ACDF group’s PSTSR was significantly better than the MA-ACDF group’s from 3 days to 3 months after surgery (p<0.001). Conclusion: Endoscope-assisted ACDF technique, based on the use of zero-profile fusion cage, has the same reliable clinical efficacy as Microscope-assisted ACDF, and is superior to Microscope-assisted ACDF in the early recovery of swallowing function after surgery, which can reflect the advantages of minimally invasive spinal surgery technique.

Keywords: Anterior Cervical Discectomy and Fusion, Endoscopy, Dysphagia, Eating Assessment Tool, Dysphagia Severity Questionnaire, Radiographic Image Interpretation

ID: AGS26/1266

A201: Comparative analysis of posterior cervical endoscopic discectomy and posterior cervical microscopic discectomy

Atif Ali 1 , Muhammad Farrukh Bashir 1

1 Ghurki Trust Teaching Hospital Lahore, Lahore, Pakistan

Introduction: Cervical disc herniation in the posterior cervical discectomy is a well-accepted treatment. Posterior cervical endoscopic discectomy (PCED) and posterior cervical microscopic discectomy (PCMD) are some of its techniques that are frequently used. The aim of this study was to compare clinical outcomes, safety, and the efficiency of the procedure of PCED with PCMD. Materials and Methods: A retrospective cohort study involved 40 participants that had single-level cervical disc herniation surgery between January 2024 and December 2024. The patients were allocated to PCED group (n = 20) or PCMD group (n = 20). Outcome measures were operation time, intraoperative bleeding, length of stay, pain treatment (Visual Analog Scale [VAS]) level, functional recovery (Neck Disability Index [NDI]) level, rate of complication, and recurrence rate. p = .05 was selected as a statistical significance level. Results: Both PCED and PCMD significantly improved VAS (7.8 ± 1.2 to 2.1 ± 0.9 vs. 7.9 ± 1.3 to 2.3 ± 1.0, p < 0.01) and NDI scores (38.5 ± 5.4 to 12.3 ± 3.2 vs. 39.1 ± 5.6 to 13.1 ± 3.5, p < 0.01). Compared with PCMD, PCED demonstrated shorter operative time (45 ± 10 min vs. 62 ± 12 min, p < 0.05), reduced blood loss (30 ± 8 mL vs. 50 ± 12 mL, p < 0.05), and lower complication rates (3.3% vs. 10%, p < 0.05). Hospital stay was comparable (2.1 ± 0.4 vs. 2.3 ± 0.5 days, p > 0.05), as were recurrence rates (5% vs. 6.7%, p > 0.05). Conclusion: Compared to PCMD, PCED has clear benefits, such as shorter operative period, less blood loss, and less complications, but the same functional results, and recurrence rates. These results indicate that PCED is a safe and effective minimally invasive intervention in cervical disc herniation, but the results of long-term follow-up studies are necessary to ensure its long-term safety.

ID: AGS26/1686

A202: Erector spinae plane block augmentation following unilateral biportal endoscopic TLIF: a randomized controlled trial

Songwut Sirivitmaitree 1

1 Bandung Crown Prince Hospital, Udon, Thailand

Introduction: Unilateral biportal endoscopic transforaminal lumbar interbody fusion (UBE TLIF) is a minimally invasive technique for lumbar spine stabilization. Despite its advantages, postoperative pain remains a challenge. The erector spinae plane block (ESPB) has emerged as a promising regional anesthesia technique for spine surgery. This study evaluates the efficacy of ESPB in improving postoperative analgesia and recovery following UBE TLIF. Material and Methods: In this randomized controlled trial. A total of 52 patients undergoing single-level UBE TLIF were randomly assigned to either the ESPB group (n = 26) or control group (n = 26). The ESPB group received a single-shot injection mixture of 0.5% bupivacaine 10ml + 1%lidocain 10ml + normal saline 10ml under fluoroscopy guidance prior to incision. Primary outcomes included total opioid consumption in 24hours and visual analog scale (VAS) pain scores at 2,6, 12, 24 and 48 hours. Secondary outcomes intraoperative opioid consumption, first time to ambulation after surgery, Quality of Recovery-15 (QoR-15) scores, incidence of postoperative nausea and vomiting (PONV), and length of hospital stay. Results: Compared to controls, the ESPB group demonstrated significantly lower opioid requirements (mean fentanyl consumption: 0 µg vs. 81.6 ± 18.7 µg, p < 0.001) and reduced VAS scores at all time points (p < 0.01). Shorten the time to ambulation after surgery (4.3 ± 1.2 hours vs. 8.4 ± 0.5 hours P = p < 0.01). QoR-15 scores were higher in the ESPB group at 24 hours (132.4 ± 6.1 vs. 118.7 ± 7.3, p < 0.001). The ESPB group doesn’t lower incidence of PONV (12% vs. 18.3%, p = 0.09) and shorter hospital stays (2.9 ± 0.4 days vs. 3.6 ± 0.5 days, p < 0.001). Conclusion: Fluoroscopy-guided ESPB significantly improves postoperative pain control and recovery in patients undergoing UBE TLIF. It is a safe and effective adjunct to multimodal analgesia in minimally invasive lumbar spine surgery.

ID: AGS26/1783

A203: 3D-navigation-guided endoscopic lumbar medial branch rhizotomy for failed back surgery syndrome

Sheng Jia Huang 1 , Chien-Min Chen 1

1 Changhua Christian hospital, Changhua, Taiwan

Introduction: Failed back surgery syndrome (FBSS), characterized by persistent pain after spinal procedures, remains diagnostically challenging and therapeutically demanding. Facet joint pain is a frequent but often under-recognized etiology. This study retrospectively evaluated the outcomes of full endoscopic lumbar medial branch rhizotomy in FBSS patients with facetogenic pain. Material and Methods: Material: Adults with FBSS derived from facet joint pain treated between 2020-07 and 2025-06 were reviewed. Eligibility required ≥ 6 weeks of unsuccessful conservative management and positive response on medial branch block. Baseline characteristics, including age, sex, body mass index (BMI), prior surgery and intraoperative detail were summarized. The primary endpoint was 12-month pain-free survival (PFS), defined as the interval from surgery to recurrence of significant pain (VAS ≥ 4 for ≥ 7 days), escalation of analgesics, or re-intervention. Patients without such events were censored at last follow-up. Subgroups were classified as Instrumented-MIS, Instrumented-Open, and Non-instrumented. Analgesic use was standardized to morphine milligram equivalents (MME) for opioids and defined daily doses (DDD) for non-opioids. Kaplan-Meier analysis with log-rank testing compared subgroup PFS, and Cox proportional hazards regression estimated hazard ratios (HR) for instrumented versus non-instrumented patients. Methods: Procedures were performed under local anesthesia in a hybrid operating room with 3D navigation. A 6-mm incision allowed insertion of 30° endoscope toward the target-level base of transverse process. Under irrigation, bipolar cautery achieved hemostasis, and soft tissues were dissected to expose the medial branch. Nerves were transected with endoscopic punches; if obscured by scarring, dissection followed anatomic landmarks such as the mamillo-accessory ligament or pedicle screws. Residual stumps were ablated until pain resolved. For facet pain at a given level, both the targeted branch and the immediately cranial branch were ablated to ensure complete denervation. Results: Twenty patients (mean age 57 years, mean BMI 24.2 kg/m 2 ) were included: Instrumented-MIS (n = 5), Instrumented-Open (n = 8), and Non-instrumented (n = 7). Median PFS follow-up was 5.6 months; median overall follow-up was 12.1 months. Twelve-month PFS rates were 20%, 54%, and 71%, respectively. No significant differences were observed across subgroups (log-rank p = 0.88). Cox analysis indicated a non-significant trend toward inferior PFS in instrumented patients (HR 2.49, 95% CI 0.66-9.43, p = 0.18). Overall, 50% of patients achieved durable 1-year PFS. Conclusion: Half of the patients maintained pain-free survival at 1 year, supporting full endoscopic lumbar medial branch rhizotomy as an effective and advantageous minimally invasive strategy for FBSS with facetogenic pain. Patients with prior instrumentation exhibited a trend toward poorer satisfaction, although not statistically significant. Study limitations include small sample size and short-term follow-up; further multicenter investigations with more large sample size are warranted to validate these findings.

ID: AGS26/1823

A204: Lumbar epiduroneurolysis: a prospective observational study on its efficacy as a minimally invasive treatment for lumbar canal stenosis in patients refuse for surgery

Muhamad Aulia Rahman 1 , Fabianto Santoso 1 , Muhammad Andika Wibisono 1 , Paul Atmaja 1 , Arief Purwodito 1

1 RISEandSPINE, Jakarta, Indonesia

Introduction: Lumbar canal stenosis (LCS) frequently causes neurogenic claudication and radiculopathy in older adults. Although decompressive or fusion are considered definitive treatment, some patients decline surgery due to medical risk or personal preference. Lumbar epiduroneurolysis offers a minimally invasive alternative targeting epidural fibrosis and inflammation. Our aim is to evaluate the effectiveness of lumbar epiduroneurolysis in improving pain, function, and quality of life in LCS patients who refuse surgical intervention. Material and Methods: Thirty patients with MRI-confirmed LCS underwent lumbar epiduroneurolysis (ABEL, GS Medical, Korea). Pain intensity (radicular and low back) was measured using the Numerical Rating Scale (NRS), disability via Oswestry Disability Index (ODI), and quality of life using SF-36 physical and mental component scores. Outcomes were recorded at baseline, immediately post-procedure, 1 month, 3 months, and 6 months. The minimal clinically important difference (MCID) thresholds were defined as ≥ 2-point reduction for NRS, ≥ 10-point reduction for ODI, and ≥ 5-point improvement in SF-36. Paired t-tests were used to determine statistical significance. Results: Radicular pain NRS decreased significantly from a baseline of 7.8 ± 1.1 to 6.1 immediately (p < 0.001), 5.6 at 1 month, 5.4 at 3 months, and 5.2 at 6 months. Low back pain NRS declined from 6.5 ± 1.0 to 5.2 immediately, 4.9 at 1 month, 4.7 at 3 months, and 4.5 at 6 months (all p < 0.001). ODI scores improved from 56.3 ± 9.5% to 49.0 immediately, 45.2 at 1 month, 43.1 at 3 months, and 41.2 at 6 months (p < 0.0001). SF-36 physical scores increased from 31.5 ± 6.0 to 35.2 at 6 months (p < 0.0001), while mental scores improved from 39.2 ± 7.0 to 43.3 (p < 0.0001). MCID was achieved in 93.3% of patients for radicular NRS, 56.7% for low back NRS, and 96.7% for ODI. Fewer patients met MCID thresholds for SF-36 scores, despite statistical significance. Conclusion: Lumbar epiduroneurolysis produced immediate and sustained improvements in radicular pain, back pain, and functional disability over a 6-month period in patients with LCS who declined surgery. While quality of life improved modestly, the procedure demonstrated clinically meaningful and statistically significant benefit in pain and function, supporting its use as a non-surgical alternative in selected patients.

ID: AGS26/2178

A205: Open fixation with arthrodesis versus percutaneous fixation without arthrodesis in pure ligamentous B2 thoracolumbar fractures: a comparative non-inferiority study

Oscar Bravo 1 , José Via Dorado 2 , Guillermo Izquierdo 2 , Felipe Novoa Tonda 2 , Manuel Valencia Carrasco 2

1 Hospital Clínico Mutual de Seguridad ChCh, Hospital Clínico de la Fuerza Aérea de Chile, Santiago, Chile

2 Hospital Clínico Mutual de Seguridad ChCh, Santiago, Chile

Introduction: Thoracolumbar type B2 fractures with ligamentous disruption are potentially unstable injuries, with open fixation with arthrodesis traditionally regarded as the gold standard. Minimally invasive fixation without arthrodesis has emerged as an alternative in trauma management; however, its role in this specific scenario remains controversial. This study aims to compare clinical and radiological outcomes between open fixation with arthrodesis and percutaneous fixation without arthrodesis in pure ligamentous type B2 thoracolumbar fractures. Material and Methods: This non-inferiority study included the totality of consecutive cases of patients surgically treated for pure ligamentous type B2 thoracolumbar fractures between 2017 and 2025, divided according to surgical technique: open fixation with arthrodesis or percutaneous fixation without arthrodesis. Parametric and non-parametric tests were applied according to data distribution. For angular variables, a clinical non-inferiority margin of 5° was established. For all other variables, a non-inferiority margin of 10% was adopted, defined as a clinically acceptable threshold to support therapeutic equivalence. Non-inferiority was declared when the 95% confidence interval of the difference between techniques remained above the predefined margins. The study was approved by the institutional ethics committee. Results: A total of 39 cases were analysed: 21 treated with open fixation with arthrodesis and 18 with percutaneous fixation without arthrodesis. Demographic characteristics were similar between groups. Percutaneous fixation without arthrodesis demonstrated non-inferiority in angular radiological variables (regional and vertebral kyphosis), with no loss of regional kyphosis correction, in contrast to open fixation with arthrodesis, where loss occurred in 14.3% of cases. Moreover, percutaneous fixation without arthrodesis was associated with reduced intraoperative blood loss (100 cc vs. 200 cc; p < 0.0001) and shorter operative time (90 vs. 120 minutes; p = 0.0037), both statistically significant differences compared with open fixation with arthrodesis. Regarding clinical outcomes, the percutaneous technique demonstrated non-inferiority in total length of hospital stay (mean 14.8 days [95% CI: 9.9-19.7] vs. 14.3 days [95% CI: 7.0-21.6]) and in medical leave duration (mean 3.1 months [95% CI: 2.4-3.7] vs. 3.2 months [95% CI: 2.5-4.0]). A favourable trend was also observed in postoperative hospital stay (median 9 vs. 11 days), although this did not reach statistical significance (p = 0.18). Conclusion: Although open fixation with arthrodesis has historically been the reference treatment for pure ligamentous type B2 thoracolumbar fractures, our findings indicate that, in appropriately selected patients, percutaneous fixation without arthrodesis represents a valid therapeutic alternative. This technique proved non-inferior in terms of clinical and radiological safety, while offering significant advantages in reduced intraoperative blood loss and shorter operative time compared with open fixation with arthrodesis.

ID: AGS26/993

A206: The “one-and-a-half” minimally invasive transforaminal lumbar interbody fusion: a single-center retrospective case series

Minaam Farooq 1 , Ahmet Kartal 1 , Chibuikem Ikwuegbuenyi 1 , Mousa Hamad 1 , Anthony Robayo 1 , Evan Wang 1 , Noel Manalil 1 , Noah Willett 1 , Andreas Kramer 2 , Rodolfo Villalobos-Díaz 3 , Ibrahim Hussain 1 , Roger Härtl 1

1 Department of Neurological Surgery, Och Spine at New York Presbyterian Hospital, Weill Cornell Medicine, New York, United States, 2 Department of Neurosurgery, LMU University Hospital, LMU Munich, Munich, Germany, 3 Department of Neurosurgery at National Institute of Neurology and Neurosurgery “Manuel Velasco Suárez”, Mexico City, Mexico

Introduction: Symptomatic lumbar spondylolisthesis with multilevel spinal stenosis poses a complex surgical challenge, often managed through multilevel decompression and fusion. We evaluated a hybrid surgical approach termed “One-and-a-Half” TLIF, combining minimally invasive transforaminal lumbar interbody fusion (mTLIF) at the unstable segment with a unilateral laminotomy for bilateral decompression (ULBD) at an adjacent stenotic level without fusion. Our objective is to assess the reoperation rate, specifically fusion extension to adjacent decompressed segments, in patients undergoing One-and-a-Half mTLIF. Methods: We retrospectively analyzed 36 patients who underwent One-and-a-Half mTLIF between January 2016 and February 2023, evaluating revision surgery rates along with radiographic and clinical outcomes. Results: 33 patients met the inclusion criteria for the One-and-a-Half mTLIF cohort, with a median follow-up of 26 months. Fusion was achieved in 96.2% of cases. The reoperation rate was 9.1% - two patients (6.1%) required subsequent fusion at the previously decompressed (ULBD-treated) level and one patient (3.0%) at the index level. The perioperative complication rate was 12.1%; neurological complications were not observed. Conclusion: The One-and-a-Half mTLIF technique is a promising alternative to traditional multilevel fusion in patients with lumbar spondylolisthesis and concurrent multilevel stenosis. This hybrid approach offers a biomechanically stable alternative to multilevel fusion by selectively fusing the unstable segment while preserving spinal motion through decompression alone at the adjacent stenotic level. Our findings highlight its long-term effectiveness in reducing the necessity for reoperation and fusion extension, suggesting it as a less invasive yet durable solution for managing complex lumbar spine pathology with promising clinical and radiographic outcomes.

ID: AGS26/907

A207: Comparative study between MIS TLIF and OLIF in management of lumber degenerative spondylolisthesis: clinical and radiological outcome

Apurba Gorain 1 , Prashant Baid 1

1 Apollo Multispeciality Hospitals, Kolkata, India

Introduction: Degenerative spondylolisthesis (DS) is a common spinal pathology often managed surgically. Minimally invasive transforaminal lumbar interbody fusion (MIS TLIF) and oblique lateral interbody fusion (OLIF) are two contemporary fusion techniques gaining popularity due to their purported advantages over open approaches. However, a direct comparative analysis of their efficacy in treating DS remains limited. This study aims to compare the clinical and radiological outcomes of MIS TLIF and OLIF in the management of degenerative spondylolisthesis. Material and Methods: This retrospective comparative study included patients with degenerative spondylolisthesis, age matched and preop ODI matched who underwent MIS TLIF or OLIF at our institution between May 2019 and May 2024. Clinical outcomes were assessed using the Visual Analog Scale (VAS) for back and leg pain, Oswestry Disability Index (ODI) at baseline, 6 months, 1 year, and final follow-up. Intra operative time, blood loss, days of hospital admission, Radiological outcomes included segmental lordosis, restoration of total lumber lordosis, and fusion rates and presence of complications such as early post operative leg pain, neurological injury, late in form of cage subsidence, pseudoarthrosis, or adjacent segment disease. Statistical analysis was performed using [appropriate statistical tests, e.g., independent t-tests, chi-square tests, ANOVA] to compare the two groups. Results: A total of 50 patients were included in the study, with 25 in the MIS TLIF group and 25 in the OLIF group. Both groups demonstrated significant improvements in VAS and ODI scores from baseline to final follow-up. Preliminary findings suggest comparable pain relief, better lordosis restoration with OLIF. Complication rates were similar between the two groups. OLIF group had significant less peri operative blood loss, and lesser duration of hospital stay. Conclusion: Both MIS TLIF and OLIF are effective surgical options for the treatment of degenerative spondylolisthesis, leading to favorable clinical and radiological outcomes. While both techniques offer significant benefits, OLIF may provide superior restoration of sagittal alignment, earlier pain relief quicker return to normal activities. Both have similar fusion rates in our study. Further prospective and randomized controlled trials with larger cohorts are warranted to validate these findings and establish definitive guidelines for surgical approach selection in degenerative spondylolisthesis.

OP24: Adolescent Scoliosis Care

ID: AGS26/100

A208: Average 20 year clinical and radiographic follow up of short segment anterior spinal fusion for adolescent idiopathic scoliosis

Neel Thakkar 1 , Theodore Choma 1 , Devin Stclair 1 , Don Moore 1

1 University of Missouri-Columbia, Columbia, United States

Introduction: Adolescent idiopathic scoliosis (AIS) affects up to 4% of adolescents, with surgical treatment rates increasing by 193% between 1997 and 2012. While posterior segmental fusion is the standard surgical approach, anterior short segment fusion offers a potential advantage by achieving similar curve correction with fewer fused segments, thereby preserving more spinal mobility. This study evaluates long-term clinical and radiographic outcomes - averaging 20 years postoperatively - of a consecutive cohort of AIS patients treated at a single center using an anterior short segment “bone-on-bone” fusion technique. Material and Methods: Patients with adolescent idiopathic scoliosis who underwent anterior short segment instrumented fusion at the University of Missouri between 1994 and 2012 were identified. PROMIS and SRS-22R surveys were administered, and a subset also completed in-person clinical and radiographic follow-up. Of 101 patients contacted, 62 completed outcome measures and 22 completed full in-person evaluations. Descriptive statistics and two-factor ANOVA were used to analyze outcomes and satisfaction. Results: Over 90% of patients were female, and ∼50% had no family history of scoliosis; half had a mental health diagnosis. Mean age at surgery was 16 years, with an average follow-up of 21 years. Forty of 55 respondents (73%) were satisfied or very satisfied with surgery, and 69% would choose it again. Mean SRS-22R score was 4.01; mean PROMIS physical and mental health scores were 51.61 and 48.07, respectively. On average, 4.5 spinal levels were fused (highest in Lenke I, lowest in Lenke V); 12% underwent revision surgery for adjacent segment disease. Satisfaction and radiographic outcomes did not differ by fusion level or mental health status. Mean initial correction was 28° (56%), with a mean loss of 13° (9%) at long-term follow-up. Conclusion: This is the largest reported 20-year follow-up of anterior bone-on-bone fusion for adolescent idiopathic scoliosis, demonstrating favorable long-term subjective and objective outcomes.

ID: AGS26/1190

A209: Optimal tranexamic acid dosing for adolescent idiopathic scoliosis surgery: a network meta-analysis

Paweł Łajczak 1 , Ayesha Ayesha 2 , Aisha Ahmed 3 , Paweł Chochoł 4 , Enzo Quednow 5 , Fabio Rocha 6 , Oguz Sahin 7 , Numa Rajab 8 , Martin Kotochinksy 9 , Yasmin Silva 10 , Yan Silva 11 , Walter Fagundes 12

1 Faculty of Medical Sciences in Zabrze, Medical University of Silesia, Katowice, Poland, Katowice, Poland

2 Shifa College of Medicine, Islamabad, Pakistan., Islamabad, Pakistan

3 Jinnah Medical and Dental College, Karachi, SINDH, Pakistan., Karachi, Pakistan

4 University of Warmia and Mazury, Olsztyn, Olsztyn, Poland

5 Department of Clinical Neurophysiology, Albacete University Hospital Complex, Spain, Albacete, Spain

6 Department of Neurosurgery, Federal University of Espírito Santo, (UFES), Vitoria, Brazil., Vitoria, Brazil

7 Edremit state hospital, Balikesir, Turkey, Balikesir, Turkey

8 Sulaiman AlRajhi University, Qassim, Saudi Arabia

9 National University of Cuyo, Argentina., Cuyo, Argentina , 10 Healthcare Institution of South Iceland, Selfoss, Iceland, Selfoss, Iceland

11 Spine Surgery, Hospital Ortopédico do Estado (Soc. Ben. Israelita Albert Einstein), Salvador, Bahia, Brazil., Salvador, Brazil

12 Department of Neurosurgery, Professor at the Federal University of Espirito Santo. (UFES), Vitória, Brazil

Introduction: TXA is an antifibrinolytic agent widely used in spine surgery. However, the optimal TXA dose remains unclear, especially for younger patients. We conducted a systematic review and network meta-analysis (NMA) to evaluate the clinical efficacy and safety of various doses of Tranexamic acid (TXA) for adolescent idiopathic scoliosis (AIS) instrumentation. Methods: We performed a systematic search across PubMed, Scopus, Web of Science, and the Cochrane Library up to March 2025. We included studies that compared different TXA doses or patients without TXA in AIS surgery. We evaluated five TXA doses: TXA 0 (control without TXA); TXA 1 (10 mg/kg bolus + 1 mg/kg/hr infusion); TXA 2 (15-20 mg/kg bolus + 1-10 mg/kg/hr infusion); TXA 3 (30-50 mg/kg bolus of TXA + 5-10 mg/kg/hr infusion); and TXA 4 (100 mg/kg bolus of TXA + 10 mg/kg/hr infusion). This meta-analysis compared bleeding volume, surgical time, transfusion rates, length of hospital stay, and complications. Random-effects frequentist NMA was employed. Treatments were ranked with P-scores. Results: A total of 1,523 patients, mainly aged 13 to 18 years from 16 studies were included in this analysis (eight retrospective studies). TXA 4 outperformed other doses in reducing intraoperative bleeding, operation time, and cell saver volume. TXA 3 also showed significant benefits but was less effective than TXA 4. There was no difference in complication rates, although these data were not uniformly reported in the studies. No evidence of publication bias was found. Quality concerns were observed. Conclusion : The tranexamic acid dose of 100 mg/kg bolus + 10 mg/kg/hr infusion (TXA 4) was the most effective in bleeding control during AIS instrumentation in teenagers. While these results are promising, more randomized trials are needed to directly compare various TXA doses.

Keywords: Tranexamic acid, adolescent idiopathic scoliosis, network meta-analysis, blood loss, spinal fusion.

ID: AGS26/1269

A210: A step closer to bringing idiopathic scoliosis monitoring at home: a smartphone app accurately generates digital twins in 624 pediatric patients

Paolo Brigato 1,2 , GianMario Sangiovanni 3 , Sabina Fehric 4 , Kevin Smit 5 , Andrew Tice 5 , Firoz Miyanji 6 , Stefan Parent 7 , Thierry Odent 8 , Neil Saran 9 , Anne Tabard-Fougère 10 , Nathaly Gavira 10 , Romain Dayer 10 , Jean Ouellet 2

1 University Hospital Campus Bio-Medico of Rome, Rome, Italy,

2 McGill University, Montreal, Canada,

3 Sapienza University, Rome, Italy,

4 University of Montréal, Montréal, Canada,

5 Children's Hospital of Eastern Ontario (CHEO), Ottawa, Canada,

6 BC Children's Hospital, Vancouver, Canada,

7 CHU Sainte-Justine, Montréal, Canada,

8 CHRU de Tours, Tours, France,

9 McGill University, Montréal, Canada,

10 HUG - Hôpitaux universitaires de Genève, Geneva, Switzerland

Introduction: Idiopathic scoliosis (IS) accounts for about 80% of pediatric spinal deformities, with Adolescent Idiopathic Scoliosis (AIS) affecting 1-4% of individuals aged 10-16 years. Standard radiographic monitoring exposes patients to repeated radiation, despite only ∼10% requiring intervention. Surface topography (ST) offers a non-invasive, radiation-free alternative by correlating torso asymmetry with spinal deformity. This study evaluates the accuracy and reproducibility of Cobb angle measurements using a smartphone-based 3D imaging application, compared with conventional radiography. Material and Methods: This prospective cohort study was conducted between November 2022 and March 2025 across five sites in three countries. Patients aged 5 to 21 years with confirmed or suspected IS were consecutively enrolled. At each clinical visit, participants underwent both standard biplanar radiographs and a smartphone-based 3D surface scan. Exclusion criteria included radiographs obtained in brace, pregnancy, and additional deformities potentially affecting topographic assessment (e.g., limb length discrepancy > 2 cm). Collected demographic and clinical data included age, sex, Body Mass Index (BMI), skin type (Fitzpatrick scale), bracing status, primary curve location, Risser stage, and relevant medical history. Radiographic parameters included manually calculated primary curve Cobb angle, thoracic kyphosis (T1-T12), and lumbar lordosis (T12-L5). Each surface scan was graded on a 0-2 image quality scale. All statistical analyses were performed in R (v4.4.3). Agreement between surface scan-derived and manual Cobb angle measurements was assessed using Spearman’s correlation and Bland-Altman analysis. Mean Absolute Error (MAE) and Root Mean Squared Error (RMSE) were also calculated to quantify prediction accuracy. Factors influencing discrepancies were explored with Kruskal-Wallis and correlation tests, with statistical significance set at α = 0.05. Results: Among 624 patients with IS (76.7% female; mean age 13.6 ± 2.4 years), a total of 624 paired surface scans and radiographs were analyzed. The mean manually measured primary curve Cobb angle was 32.1° ± 16.1°. Surface scan-derived values showed strong correlation with these measurements (r = 0.775, p < 0.001). Bland-Altman analysis indicated a mean bias of -3.6° with limits of agreement from -16.9° to 24.0°. Prediction errors were significantly associated with bracing, study site, Cobb angle class, skin type, and image quality (p < 0.05), while no significant relationships emerged with age, sex, weight, height, or BMI. Scan quality strongly influenced prediction accuracy: errors were highest with poor-quality images (MAE 11.87°, RMSE 15.95°), intermediate with fair quality (MAE 10.02°, RMSE 13.06°), and lowest with high-quality scans (MAE 6.98°, RMSE 8.89°). Conclusions: The smartphone-based 3D imaging system demonstrated good agreement with manual Cobb angle measurements, with only a modest systematic underestimation. Prediction accuracy was influenced by factors such as image quality, bracing status, skin type, and study site, with higher-quality images yielding the most reliable results. These findings support the potential of this tool as a complementary aid in scoliosis assessment, while also underscoring areas where further refinement and validation could enhance clinical utility.

ID: AGS26/1496

A211: Association between correction index and patient reported outcome scores in severe spine deformity

Arthur Sackeyfio 1 , Alexander Upfill-Brown 2 , Sarah Obiri 1 , Kwadwo Yankey 1 , Derrick Owusu 1 , Francis Addai 1 , Irene Wulff 1 , Liudmila Gross 1 , Oheneba Boachie-Adjei 1

1 FOCOS Orthopedic Hospital, Accra, Ghana

2 UCLA Health, Los Angeles, United States

Introduction: A significant component of the success of complex spine deformity surgery is the patient reported outcome measures. These corrective spine surgeries in curves above 100° are considered risky, as they may be complicated by neurologic deficit, cardiopulmonary complications and implant related complications. In the event where these complications do not occur, patient reported outcomes become a significant measure of success of such surgeries. We seek to see if there is an association between correction index and patient reported outcomes in severe spine deformity correction surgery. Material and Methods: Patient with pre-operative coronal or sagittal curves greater than or equal to 100 degrees retrospectively reviewed at a single institution in Ghana between 2015 and 2017. Inclusion criteria included pre-operative sagittal or coronal curve ≥ 100 degrees and follow-up of two-years or greater. Exclusion criteria included revision surgeries or growing rods. Correction index defined as ratio of post-operative cobb measurement over pre-operative measurement. When patient had sagittal and coronal cobb angles ≥ 100, the larger CI was selected. Analysis was also performed separately for sagittal CI and coronal CI. Outcomes included the change in SRS total score and component scores between pre-op and post-op. Results: A total of 41 patient identified. Fifteen patients were diagnosed with adolescent idiopathic kyphoscoliosis, 8 with congenital kyphoscoliosis, 5 with neurofibromatosis and 13 with tuberculosis. Median age was 15 years (IQR 13 to 18). 37 patients were treated with halo traction. Twenty-three patients were treated with VCR, 1 had one-level, 11 had two-level and 6 patients had a three-level VCR. Twenty-two patients had a coronal cobb ≥ 100 degrees, with a median of 138 (range 100-174). Of these, median coronal CI was 0.51 (range 0.15 to 0.80). All 41 patients had a sagittal cobb ≥ 100 degrees, with a median of 141 (range 100-240). Of these, the median sagittal CI was 0.59 (range 0.27 to 0.90). The median overall CI was 0.60 (range 0.27-0.90). The total SRS score improved significantly in all patients (mean pre-op 3.5, mean post-op 4.1, p < 0.0001). Of the sub-scores, SRS Self Image (2.4 to 3.9, p < 0.0001) and Mental Health (4.0 to 4.3, p = 0.039) improved significantly. Change in SRS score significantly associated with correction index (ß = 2.15, p = 0.006, Figure 1A ) and sagittal correction index (ß = 2.22, p = 0.0039), but not coronal correction index (p = 0.24). Change in sub-scores were not independently associated with CI. Analysis of the relationship of CI and change in total SRS score as non-linear resulted in substantial improvement in model fit (Adj R 2 0.246 vs 0.159, Figure 1B ). On threshold analysis, a threshold of CI > 0.40 was identified using global CI, resulting in the best model fit (Adj R 2 0.267). This was found to be significantly associated with an improvement in total SRS scores of 1.4 for global CI (p = 0.00036) and 1.1 for sagittal CI (p = 0.0027). Conclusion: In patients presenting with large kyphoscoliotic deformities (> 100°), we find that the degree of correction is associated with improvement in SRS scores, with a critical CI of 0.4. We also find that excessive corrections beyond a CI of 0.4 may not be necessary for improved patient reported outcomes, and to mitigate high surgical risk, surgical options which may result in satisfactory alignment within a CI of 0.4 should be opted for. This implies that for a kyphotic curve of 120°, a correction to 70° degrees will result in similar outcomes compared to a correction to 30°. This is similar to work done in less severe deformities, showing a CI of approximately 0.50 resulting in reliable improvement in outcomes.

ID: AGS26/1590

A212: Wiltse approach and inter-apophisary fusion for adolescent idiopathic scoliosis: analysis of complications at the 2-year follow-up

Georgios Bakaloudis 1 , Lorenzo D'Apice 1 , Alice Baroncini 1

1 Casa di Cura Humanitas San Pio X, Milano, Italy

Introduction: In the past years, posterior instrumentation utilizing a Wiltse approach and inter-apophisary fusion has been introduced to reduce the invasiveness of surgical management for adolescent idiopathic scoliosis (AIS). However, the safety and efficacy of this approach have not yet been fully investigated. This work aimed to evaluate the radiographic results and complication rate in a single-surgeon case series. Material and Methods: Retrospective analysis of prospectively collected data. All consecutive AIS patients managed with Wiltse approach and inter-apophisary fusion between Sept 2016 and Jan 2023, who had a minimum follow-up of two years, were included. The amount of correction at the last follow-up and the rate of complications were recorded. Results: The study included 123 patients (88 female), mean age 15.6 ± 2.6 years, follow-up 26-75 months. Most patients had a Lenke 1 or 3 curve, mean Cobb angle 69 ± 7°. The skin incision was a single, median incision in 62 cases, while 61 patients had three separate midline incisions. Screws were placed through a paramedian, multilevel Wiltse approach in all cases; screw placement was guided fluoroscopically in 84 cases, with spinal navigation in 15 cases, and with robotic neuronavigation in 24 cases. 6mm titanium rods were used in almost all cases. On average, 9.8 ± 2.6 levels had been instrumented, with a mean implant density of 58% (range 53-67%). None of the patients required a blood transfusion. The length of hospital stay was 3-5 days, opioid analgesia was suspended within 15 days from surgery, and patients returned to school 2-3 weeks after surgery. At the last available follow-up, a mean correction by 60.8 ± 9.3% (47.3 ± 6.6°) was observed, and the mean Cobb angle measured 23.3 ± 5.4°. In terms of complications, 5 cases of wound dehiscence were observed, all in patients with separate midline incisions. Two cases healed secondarily, and two required implant removal. The latter did not show bacterial growth on the implant, and no loss of correction was observed at the 2-year follow-up. Three distal screw breakages, one implant exchange due to allergy (with revision with different materials), one proximal screw pullout and one rod mobilization were observed, three patients required revision surgery. No rod breakages were observed. Conclusion: The Wiltse approach with inter-apophisary fusion is a safe and effective technique for posterior spinal instrumentation in AIS. The revision rate is in line with that for open posterior fusion. The observation of wound dehiscence with the use of separate incisions suggests that the use of a single midline incision might be a safer approach.

ID: AGS26/1639

A213: Beyond bending films: estimating scoliosis flexibility from standing and supine radiographs

Altug Yucekul 1 , Feyzi Kilic 2 , Suha Aktas 3 , Elif Gizem Carus 2 , Tais Zulemyan 2 , Caglar Yilgor 4 , Ahmet Alanay 1

1 Acibadem Mehmet Ali Aydinlar University School of Medicine, Istanbul, Turkey

2 Comprehensive Spine Center, Acibadem Maslak Hospital, Istanbul, Turkey

3 Istanbul Physical Therapy and Rehabilitation Training and Research Hospital Orthopedics and Traumatology Clinic, Istanbul, Turkey

4 Koc University Hospital Spine Center, Istanbul, Turkey

Introduction: Flexibility assessment is a fundamental component in the preoperative evaluation of patients with adolescent idiopathic scoliosis (AIS). The degree of curve reducibility is crucial for determining surgical strategy, selecting the levels of instrumentation, and anticipating postoperative alignment. Traditionally, flexibility is measured using special radiographs such as side-bending, fulcrum, and TRUGA (traction radiograph under general anesthesia) radiographs. These radiographs are time-consuming, technician-dependent, and expose the patient to additional radiation. The present study aimed to formulate a simple radiographic predictor of maximum curve flexibility (SUFLEX) using only standard standing and supine Cobb angles, and to test whether SUFLEX can accurately identify curves that bend below 25°. Materials and Methods: A retrospective review was conducted on 142 consecutive AIS patients for whom a complete set of flexibility radiographs (standing, supine, side-bending, fulcrum, and TRUGA) were available. Patients with a main curve magnitude ≥40° on standing radiographs were included in the SUFLEX formulation, yielding 102 main thoracic (MT) and 89 thoracolumbar/lumbar (TL/L) curves. Maximum flexibility (%) for each curve was defined as the greatest percentage reduction from standing Cobb angle observed on any of the special flexibility radiographs. Two separate multiple linear regression models were constructed, one for MT and one for TL/L, with standing and supine Cobb angles as predictors. Model performance was assessed using five-fold cross-validation, reporting R 2 and mean absolute error (MAE). Logistic regression analyses using the same predictors were performed to estimate the probability that a given curve would bend to less than 25°. Model discrimination was evaluated by receiver operating characteristic (ROC) analysis with out-of-fold probabilities, and the area under the curve (AUC) was calculated. Classification thresholds were determined using the Youden index. Results: The study cohort consisted of 142 patients, of whom 118 were female and 24 were male. The mean age was 14.5 ± 2.1 years. The mean Risser stage was 2.8 ± 1.3, and the mean Sanders stage was 4.9 ± 1.2. Curve classification according to Lenke type was type 1 (54 patients), type 2 (17 patients), type 3 (24 patients), type 4 (15 patients), type 5 (22 patients), and type 6 (10 patients). Lumbar modifier distribution was A (59 patients), B (52 patients), and C (31 patients). For main thoracic curves (n=102), cross-validated model performance was R 2 = 0.50 ± 0.09 with MAE 9.34% ± 1.29. For thoracolumbar/lumbar curves (n=89), performance was R 2 = 0.30 ± 0.18 with MAE 10.04% ± 1.42. Using the SUFLEX formula, prediction for minimum Cobb angles for MT curves were within ± 5° in 51.0% of cases, within ± 5-10° in 32.4%, and within ± 10-13° in 9.8% ( Figure 1 ). For TL/L curves, the proportions were 59.6%, 27.0%, and 5.6%, respectively. In logistic regression analyses, the discrimination of SUFLEX for identifying curves that would bend below 25° was strong (AUC of 0.921 (95% CI 0.864-0.966) for MT and 0.853 (95% CI 0.645-0.981) for TL/L. Accuracy was 87.3% for MT and 96.6% for TL/L. Conclusions: A SUFLEX estimator derived from standing and supine Cobb angles provides a practical and validated estimate of maximum curve flexibility, accurately identifying curves likely to bend below 25°. This simple approach can streamline preoperative assessment, minimize the need for additional flexibility films, and reduce radiation exposure. By reliably predicting curves below 25°, SUFLEX may also facilitate Lenke classification without routine bending radiographs, offering a low-cost and efficient adjunct for surgical planning in AIS.

ID: AGS26/1742

A214: Selective thoracic fusion with postoperative bracing versus non-selective fusion in Lenke 1C adolescent idiopathic scoliosis: a propensity score-matched analysis

Giovanni Viroli 1 , Matteo Traversari 1 , Marco Manzetti 1 , Alberto Ruffilli 1 , Cesare Faldini 1

1 Istituto Ortopedico Rizzoli, Bologna, Italy

Introduction: The surgical treatment of Lenke 1C adolescent idiopathic scoliosis (AIS) remains controversial. While non-selective fusion (NSF) addresses both thoracic and lumbar curves, selective thoracic fusion (STF) aims to preserve lumbar mobility but may risk coronal imbalance. The use of postoperative corrective bracing has been proposed to enhance lumbar curve control after STF. This study compared radiographic and patient-reported outcomes of STF+brace versus NSF in Lenke 1C AIS using a propensity score-matched cohort. Materials and Methods: A retrospective series of 76 Lenke 1C AIS patients was reviewed: 38 underwent STF with postoperative bracing until skeletal maturity, and 38 underwent NSF. Propensity scores were estimated using age, sex, skeletal maturity (Risser, CVM), curve magnitudes, lumbar flexibility, sagittal parameters, and baseline PROMs (SRS-22, QLPSD). One-to-one optimal matching yielded 30 well-balanced pairs (n = 60). Balance was assessed using standardized mean differences (|SMD < 0.10). Post-matching comparisons were performed using paired t-tests for continuous variables and McNemar’s test for categorical variables. All patients had a minimum of 24 months follow-up (range 24-41 months). Results: At last follow-up, radiographic outcomes showed no significant differences between groups. Thoracic Cobb correction was comparable (STF+brace 67 ± 9% vs. NSF 69 ± 8%, p = 0.42). Lumbar Cobb angles at follow-up were slightly lower in the NSF group (18 ± 6° vs. 21 ± 7° in STF+brace), but this trend was not statistically significant (p = 0.09). PROMs demonstrated no significant differences in overall SRS-22 total (STF+brace 4.3 ± 0.3 vs. NSF 4.4 ± 0.4, p = 0.28) or QLPSD total scores (STF+brace 42 ± 6 vs. NSF 41 ± 5, p = 0.31). However, analysis of subdomains revealed distinct patterns: SRS-22 function was significantly better in the STF+brace group (4.5 ± 0.3 vs. 4.2 ± 0.4, p = 0.02), while SRS-22 satisfaction was slightly but significantly lower (4.1 ± 0.4 vs. 4.3 ± 0.3, p = 0.04). Regarding QLPSD domains, back flexibility was significantly better in STF+brace (2.2 ± 0.8 vs 3.3 ± 0.5, p = 0.01), whereas body image was worse (1.8 ± 0.9 vs. 1.4 ± 0.7, p = 0.04). Conclusion: In Lenke 1C AIS, selective thoracic fusion with postoperative bracing provides equivalent global radiographic and overall quality-of-life outcomes compared with non-selective fusion, while preserving lumbar mobility. Subdomain analyses highlight trade-offs: STF+brace improves function and back flexibility but may modestly reduce patient satisfaction and body image perception. These findings suggest STF+brace as a viable motion-preserving strategy, although longer-term prospective studies are warranted.

ID: AGS26/1813

A215: Multicenter retrospective validation study of artificial intelligence-assisted bone age estimation from left-hand radiographs: comparison of idiopathic scoliosis patient groups and healthy controls

Bilge Yilmaz 1 , Soheil Sabet 1 , Yunus Akman 1 , Huseyin Ozturk 1 , Cem Sever 1 , Huseyin Coskun 1 , Mutlu Cobanoglu 1 , Celaleddin Bildik 1 , Gokce Guven 2 , Deniz Karasahin 2 , Sinan Kahraman 1 , Tunay Sanli 1

1 Anatolian Spine Study Group, Istanbul, Turkey

2 Osteoid Health Technologies, Istanbul, Turkey

Introduction: Accurate skeletal maturity assessment is essential for predicting curve progression and optimizing treatment timing in idiopathic scoliosis (IS). While traditional bone age methods (Greulich-Pyle, Tanner-Whitehouse) and scoliosis-specific scales are widely used, deep learning models have recently shown promising accuracy. This study aimed to validate an Artificial Intelligence (AI)-assisted program (Osteoid-BAA 0,1) for bone age estimation in the Turkish population and to compare skeletal maturity patterns between IS subgroups and healthy controls. Material and Methods: A multicenter retrospective review was conducted across 8 centers. The study cohort comprised IS patients aged 8-18 years with scoliosis who underwent left hand-wrist radiography, together with age-matched healthy controls. Subjects were categorized into: Group 1 (IS, Cobb 10-40°), Group 2 (IS, Cobb > 40°, surgical candidates), and Group 3 (Age-matched healthy controls). All bone age was assessed by AI and an experienced radiologist. Accuracy was evaluated using Pearson correlation, mean absolute error (MAE), and bias. Group differences were tested with ANOVA/Kruskal-Wallis, including subgroup analyses by sex and age (8-12, 13-15, 16-18 years). Results: A total of 208 individuals (237 left hand-wrist radiographs; mean age 13.6 ± 2.3 years, 157F/51M) were analyzed. All individuals were categorized into three groups: G1 (n = 103, 88 females/15 males, mean age 13.3 years), G2 (n = 63, 53 females/10 males, mean age 13.8 years), and G3 (n = 71, 39 females/32 males, mean age 12.7 years). AI showed strong correlation with chronological age (r = 0.74, MAE = 15.6 months) as well as with the radiologist (r = 0.87, MAE = 11.1 months). The radiologist achieved a correlation of r = 0.89 with chronological age and a MAE of 9.9 months. Between-group comparisons revealed the following biases: AI bias was +3.6 ± 16.6 months in G1, +6.9 ± 19.2 months in G2, and -2.5 ± 19.8 months in G3 (ANOVA p = 0.009; G2 vs. G3 p = 0.0087), whereas radiologist bias was +5.8 ± 11.0 months in G1, +6.1 ± 13.0 months in G2, and +5.4 ± 14.8 months in G3 (p = 0.945). AI predicted more advanced bone age in surgical IS patients (p = 0.0099), suggesting accelerated maturation in severe curves. In sex-specific analyses, AI showed a bias of -7.7 months in females and +13.4 months in males (p < 0.001), while radiologist bias was -6.5 months in females and -3.6 months in males (p = 0.18). In age subgroup analyses, AI bias was -10.2 months in the 8-12 years group, -6.7 months in the 13-15 years group, and +16.6 months in the 16-18 years group (p < 0.001), whereas radiologist bias was -4.6 months, -9.8 months, and +1.3 months across the same age groups, respectively (p < 0.001). Conclusion: AI-assisted bone age estimation demonstrated accuracy comparable to expert evaluation and revealed maturity differences between IS subgroups and controls. AI predicted advanced bone age in the surgical IS group, suggesting accelerated skeletal maturation in patients with severe curves. This approach holds promise as a scalable, objective decision-support tool in IS management. Further external validation and clinical outcome studies are warranted.

ID: AGS26/3166

A216: Enhanced recovery after surgery in adolescent spinal deformities: from experimental pathway to routine practice

Konstantinos Martikos 1 , Bruna Maccaferri 1 , Emanuela Asunis 1 , Francesco Vommaro 1 , Chiara Cini 1 , Francesca Salamanna 1 , Alessandro Ricci 1 , Aristide Morigi 1 , Renata Bacchin 1 , Lesley De Pietri 1 , Milena Fini 1 , Alessandro Gasbarrini 1

1 IRCCS Istituto Ortopedico Rizzoli, Bologna, Italy

Introduction: In recent years, there has been growing evidence supporting Enhanced Recovery After Surgery (ERAS) protocols in orthopedic surgery. Four years ago, at the national congress, we presented a comparative study between ERAS and conventional postoperative care in idiopathic scoliosis. In the present work, we report our extended case series managed with ERAS protocols, further supported by additional patients, to advocate their adoption as part of daily clinical practice. Materials and Methods: A retrospective evaluation was conducted on 110 adolescent patients (aged 11-18 years) with spinal deformities (scoliosis and hyperkyphosis) who underwent thoracolumbar spinal fusion with accelerated postoperative recovery. Inclusion criteria: idiopathic scoliosis, traditional instrumented posterior fusion, and all curve types according to the Lenke classification. Outcomes included radiographic parameters, postoperative pain questionnaires, and complication rates. Results: Mean hospital stay was 4.5 days (range 2-10). Admission to intensive care was required in only 4.8% of cases. The incidence of infections was 1.7%. Pain scores decreased from 8.2 (VAS, postoperative day 1) to 4.2 at discharge. The presentation will detail our intra- and perioperative management protocol, including anesthetic and orthopedic strategies for pain control. Conclusions: Our experience with ERAS protocols in adolescent spinal deformity surgery demonstrates improved functional recovery, effective pain management, reduced hospitalization costs, and lower complication rates.

OP25: Tumor Reconstruction

ID: AGS26/1010

A217: Surgical management of multiple myeloma vertebral column lesions - Results from an international survey distributed to the AO Spine Knowledge Forum Tumor

Vanessa Hubertus 1,2 , Emily von Bronewski 1 , Lucius Fekonja 1 , Anton Früh 1 , Christian Entenmann 1 , Hannah Miller 1 , Charlotte Buhre 1 , Michael Fehlings 3 , Ilya Laufer 4 , Peter Vajkoczy 1 , Julia Onken 1,5

1 Charité University Hospital Berlin, Department for Neurosurgery, Berlin, Germany,

2 Berlin Institute of Health, BIH Charité Clinician Scientist Program, Berlin, Germany,

3 Division of Neurosurgery and Spinal Program, Toronto Western Hospital, Krembil Neuroscience Center, and University of Toronto, Toronto, Canada,

4 Department of Neurosurgery, New York University Langone Health, New York, United States,

5 German Consortium for Translational Cancer Research, German Cancer Research Centre, Berlin, Germany

Introduction: Vertebral fractures and epidural infiltration are common complications in Multiple Myeloma (MM), causing vertebral pain and neurological deficits. Non-surgical management is universally preferred, but internationally accepted guidelines are lacking. The aim of this study was to assess the current international treatment standards and clinical conditions guiding surgical decision-making in MM vertebral lesions. Material and Methods: A survey was distributed to the members of the AO Spine Knowledge Forum Tumor, an international, interdisciplinary key opinion leader group specialized on the treatment of oncologic spine disease. The survey, consisted of 25 questions, of which 15 assessed the participant`s background, clinical expertise, and experienced treatment standards regarding MM vertebral lesions, followed by ten case examples with seven possible treatment scenarios. The distribution was supported by AO Spine Network Clinical Research and facilitated via the web-based application REDCap®. Results: 51 international experts, from 14 countries and five continents completed the survey, with 51% being of orthopaedic, and 44% of neurosurgical background, while 6% were radio-oncologists. 87% of the participants were university-associated, and 96% part of interdisciplinary cancer centers. 84% of the participants stated that they “see vertebral lesions in MM in general as a non-surgical disease”. As strongest indicators to nonetheless perform surgery, neurological deficits (74%), and potentially unstable lesions (20%) were chosen. Clinical and radiological follow-up is performed by 83%, mostly by MRI (62%), however only in 46% at defined intervals. Overall, 89% would choose “less invasive” surgical strategies for MM than similar lesions in metastatic spine disease. Conclusion: The results of this international survey display the current international treatment standards for MM vertebral column lesions. While the participating experts agreed towards a more restrained management of MM patients, the applicability of surgical scores, standards for clinical and radiological follow-up, and indications as well as surgical strategies for MM vertebral lesions varied widely, illustrating the need for international guidelines standardizing treatment.

ID: AGS26/1768

A218: Evaluation of adverse events and the impact on health-related outcomes in patients undergoing surgery for metastatic spine tumors: analysis of the Metastatic Tumor Research And Outcomes Network (MTRON) Registry Dataset

Giovanni Barbanti Bròdano 1 , Cristiana Griffoni 1 , Francesca Salamanna 2 , Luigi Emanuele Noli 1 , Annalisa Monetta 1 , Alessandro Luzzati 3 , Alexander Disch 4 , Aron Lazary 5 , Ori Barzilai 6 , Ilya Laufer 7 , Ziya Gokaslan 8 , Michael Fehlings 9 , Jorrit-Jan Verlaan 10 , Dean Chou 11 , Laurence Rhines 12 , John Shin 13 , William Teixeira 14 , Daniel Sciubba 15 , Chetan Bettegowda 16 , Raphaële Charest-Morin 17 , Stefano Boriani 18 , Tony Goldschlager 19 , Michael Weber 20 , Charles Fisher 17 , Alessandro Gasbarrini 1 21

1 Complex Spinal Surgery Unit, IRCCS Istituto Ortopedico Rizzoli, Bologna, Italy

2 Surgical Sciences and Technologies, IRCCS Istituto Ortopedico Rizzoli, Bologna, Italy

3 IRCCS Istituto Ortopedico Galeazzi, Milano, Italy

4 University Hospital Carl Gustav Carus at the TU Dresden, Dresden, Germany

5 National Center for Spinal Disorders, Budapest, Hungary

6 Memorial Sloan-Kettering Cancer Center, New York, United States

7 New York University Langone Health, New York, United States

8 Brown University, Warren Alpert School of Medicine, Providence, United States

9 Division of Neurosurgery and Spine Program, University of Toronto, Toronto, Canada

10 University Medical Center Utrecht, Utrecht, The Netherlands

11 Columbia University Irving Medical Center, New York, United States

12 The University of Texas MD Anderson Cancer Center, Houston, United States

13 Massachusetts General Hospital, Boston, United States , 14 Instituto do Câncer do Estado de São Paulo, Sao Paulo, Brazil

15 Northwell Health, Manhasset, New York, United States

16 Johns Hopkins University School of Medicine, Baltimore, United States

17 University of British Columbia, Vancouver, Canada

18 University of Bologna, Bologna, Italy

19 Monash University, Melbourne, Australia

20 University of Connecticut, Department of Orthopedic Surgery, Farmington, United States

21 Department of Biomedical and Neuromotor Sciences, University of Bologna, Bologna, Italy

Introduction: This study is part of the AO Spine Metastatic Tumor Research and Outcomes Network [MTRON], an international multicenter prospective observational registry including patients with spinal metastases. This study aims to elucidate the incidence of surgical complications, their risk factors and consequent effects on survival outcomes, hospital length of stay, and overall health-related quality of life (HRQOL) parameters in a large cohort of patients affected by spinal metastases who were surgically treated. Material and Methods: Available data from February 2017 to July 2023 were analyzed. The primary outcome of this study was the evaluation of the incidence of intraoperative and postoperative adverse events (AEs). The secondary outcomes included the assessment of risk factors for surgery-related AEs and the impact of AEs on survival, length of hospital stay and quality of life. Results: Among the 1267 patients analyzed, 6.9% experienced intraoperative AEs and 19.3% experienced at least one postoperative AE. Several factors resulted to be associated to the occurrence of postoperative AEs: age, smoking habit, poor Eastern Cooperative Oncology Group ( ECOG) Performance status, previous radiation therapy at the index target, duration of surgery, number of instrumented levels, simultaneous anterior and posterior approach, presence of metastases at other sites, multiple spinal metastases. Postoperative AEs were associated with reduced survival rates, increased hospital length of stay and poorer HRQOL outcomes, particularly in domains such as neurological function and mental health. In general, surgery substantially improves HRQOL across multiple domains, with these benefits persisting over time despite the occurrence of AEs. However, patients with preoperative risk factors, including comorbidities, smoking, neurological impairment, and prior radiation therapy, experienced less improvement. Conclusion: The negative impact of AEs on overall survival and HRQOL could be associated with the presence of some preoperative parameters of frailty that are detected as risk factors for AEs occurrence. This finding emphasizes the need for personalized preoperative assessments and optimized perioperative care strategies.

The trial is registered at ClinicalTrials.gov with ID number NCT02830451 .

ID: AGS26/2246

A219: Bone graft substitutes and spinal fusion outcomes in spine oncology

Esteban Ramirez-Ferrer 1 , Juan Zuluaga 1 , Romulo Augusto Andrade de Almeida 1 , Francisco Call-Orellana 1 , Gil Kimchi 1 , Christopher Alvarez-Breckenridge 1 , Rob North 1 , Laurence Rhines 1 , Claudio Tatsui 1

1 The University of Texas MD Anderson Cancer Center, Houston, United States

Introduction: Surgery for spine tumors prioritizes oncologic control spinal cord decompression, and structural stabilization while achieving spinal fusion remains a secondary goal. This study aimed to assess fusion rates across three different bone grafts materials andtheir association with hardware failure. Material and Methods: We performed a retrospective cohort study of patients who underwent tumor resection and spinal instrumentation from January 2018 toJanuary 2024. Patients with overall survival (OS) of less than 6 months or incomplete follow-up data were excluded. Descriptive statistics summarized the cohort characteristics. Group comparison test, Kaplan-Meier curves estimated fusion probabilities over time, and Cox proportional hazard models identified predictors of spinal fusion. Results: A total of 159 patients were included: 93 treated with demineralized bone matrix (DBM), 30 with biphasic calcium phosphate (BCP), and 36 with bioactive glass fiber (BGF). Groups showed no significant differences in age, sex, BMI, KPS, neurological status, Spinal Instabitily Neoplastic Scores (SINS), bone graft volume, radiation therapy, hardware failure, or pseudoarthrosis rates. Mean time to spinal fusion was 54, 14, and 22 months for the DBM, BCP, and BGF, respectively. On multivariate analysis all grafts materials were positively associated with spinal fusion, with BCP showing the strongest effect. Prior radiation therapy was negatively associated with fusion rates, while BMI > 25 kg/m 2 was associated with a higher risk of hardware failure. Conclusion: Bone grafts substitutes improve spinal fusion rates in spine oncology, with ceramic-based grafts demonstrating the highest and earliest fusion rates. Conversely, postoperative radiation therapy negatively affects fusion, and further prospective studies are needed to determine the optimal graft selection, volume, and timing with adjuvant radiation therapy to balance oncologic controland mechanical stability.

ID: AGS26/243

A220: Hybrid spine reconstruction using an expandable cage combined with vascularized free fibula autograft after en bloc spondylectomy for primary bone tumors of the mobile spine

Gil Kimchi 1 , Garrett Walsh 1 , Reza Mehran 1 , Alex Mericli 1 , Paul Shay 1 , Edward Chang 1 , David Adelman 1 , Christopher Nguyen 1 , Francisco Call-Orellana 1 , Juan Zuluaga 1 , Laurence Rhines 1

1 MD Anderson Cancer Center, Houston, United States

Introduction: En bloc spondylectomy for primary spinal tumors creates extensive bony defects that require durable anterior column reconstruction. Unlike metastatic cases, these patients are often young and potentially curable, necessitating long-term bony fusion. Conventional reconstruction with cages and allograft frequently fails to achieve solid arthrodesis and is associated with high failure rates. The objective of this study was to compare outcomes of a novel hybrid reconstruction using expandable cages combined with vascularized autologous fibula grafts versus traditional cage/allograft-only techniques. Material and Methods: We retrospectively reviewed patients who underwent en bloc spondylectomy for primary spinal bone tumors by a single surgeon (2002-2024). Patients were grouped by reconstruction type: hybrid (expandable cage packed with allograft with adjacent vascularized fibula with microvascular anastomosis) or control (cage with allograft alone). Clinical, radiographic, surgical, and functional data were analyzed. Results: Forty-two patients were included (20 hybrid, 22 control). Fusion was achieved in 100% of hybrid cases versus 52.4% of controls (p < 0.001). Mean time to fusion was significantly shorter with the hybrid technique (9.1 ± 4.9 vs. 24.9 ± 25.43 months, p < 0.001). Time to ambulation was similar (median 5 days). No hybrid group patients required reoperation for hardware failure, compared to four in the control group (p = 0.45). The hybrid group demonstrated lower complication rates across all categories. Conclusion: Hybrid reconstruction with an expandable cage and vascularized fibula autograft significantly improves fusion rates and time to fusion compared to conventional methods, without increased morbidity. This approach may offer a more durable long-term solution in patients undergoing curative surgery for primary spinal tumors.

ID: AGS26/2571

A221: Local disease control and survival analysis after surgical treatment of spinal metastases: comparison between carbon fiber-reinforced polythertherketone and titanium implants for posterior stabilization

Luigi Falzetti 1 , Cristiana Griffoni 1 , Simona Rosa 2 , Silvia Terzi 1 , Chiara Alcherigi 1 , Mariarita Intagliata 1 , Stefano Bandiera 1 , Giovanni Barbanti Bròdano 1 , riccardo ghermandi 1 , Giuseppe Tedesco 1 , Gisberto Evangelisti 1 , Valerio Pipola 1 , Marco Girolami 1 , Stefano Pasini 1 , Giovanni Tosini 1 , Chiara Cini 1 , Alessandro Gasbarrini 1,2

1 Complex Spinal Surgery Unit, IRCCS Istituto Ortopedico Rizzoli, Bologna, Italy

2 Department of Biomedical and Neuromotor Sciences, University of Bologna, Bologna, Italy

Introduction: Metastatic involvement of the spine can lead to significant morbidity, including intractable pain, pathological fractures, spinal instability, and neurological deficits due to spinal cord compression. While systemic therapies and radiotherapy remain central in the treatment of metastatic cancer, surgery plays a pivotal role in selected cases, particularly when mechanical instability or neurological compromise is present. Carbon fiber-reinforced PEEK (CRFP) spinal instrumentation has emerged as a promising alternative in the oncological setting: CFRP is radiolucent, producing minimal artifacts on CT and MRI, and it interacts less with radiation therapy, resulting in reduced backscatter and shielding effects, thereby allowing for superior postoperative imaging and radiotherapy planning. Considering the lack of comparative clinical trials, we proposed analyzing differences in terms of survival, local disease control and quality of life in a population of spinal metastatic patients surgically treated with CFRP or titanium implants. Methods: We performed a prospective monocentric cohort clinical study on patients with vertebral metastases who underwent spinal surgery from August 2018 to October 2022. Demographic, clinical and surgical data were collected, as well as adverse events, clinical outcomes and patient-reported outcomes (PROMs) during the follow up period. Results: 160 patients were included in this study: 129 patients were treated with titanium instrumentation and 31 patients were treated with CFRP instrumentation. A local disease progression was recorded in 36 patients (22.5%), 27 patients with titanium instrumentation (20.9%) and 9 patients with CFRP instrumentation (29.0%). Kaplan Meier analyses: The median survival time of the entire population was 24.3 months (95% CI 15.0-32.1), no statistically significant difference was observed between the two groups (Log- rank test, p = 0.637). The type of instrumentation had no significant impact on the local progression (Log-rank test, p = 0.577). Local progression during the first year after surgery had no significant impact on the long-term risk of death (Log rank test, p = 0.531) and on the risk of death during the first 12 months (Log- rank test, p = 0.363). After the surgery, 31.9% of patients underwent radiotherapy at the index spinal lesion: 13 cases (41.9%) in the group treated with CFRP and 38 cases (29.5%) in the group treated with titanium. Cox model showed that adjuvant radiotherapy performed within the first year after surgery reduced the risk of local progression by 69% with a statistically significant protective result (p = 0.035). Adjuvant radiotherapy also decreased the risk of death by 46% (p = 0.011). Seven cases of mechanical complications (5.4%) were detected in the group treated with titanium implants, while no cases were registered in the group treated with CFRP, however, the difference was not statistically significant (p = 0.347). Concerning PROMs, no statistically significant difference appeared between the two groups of patients (CFRP implants versus titanium implants) during the follow up period. Conclusion: The use of CFRP instrumentation can improve the adjuvant radiotherapy treatment after surgery for spinal metastases, which is demonstrated to have a protective effect on the local progression of the disease and on the risk of death, if carried out in the first year after surgery.

ID: AGS26/2580

A222: Morphological descriptors associated with biomechanical deterioration and need for stabilization in spinal metastases - Results from a modified Delphi Study from the AO Spine Knowledge Forum (KF) Tumor

Aron Lazary 1 , Marie-Laure Vial 2 , Kristof Koch 1 , C. Rory Goodwin 3 , Arjun Sahgal 4 , Laurence Rhines 5 , Cordula Netzer 6 , Jorrit-Jan Verlaan 7 , Alessandro Gasbarrini 8 , Jeremy Reynolds 9 , Ori Barzilai 10 , Chetan Bettegowda 11 , Stefano Boriani 12 , Charles Fisher 13 , Ziya Gokaslan 14 , Ilya Laufer 15 , Nicolas Dea 13

1 National Center for Spinal Disorders, Budapest, Hungary

2 AO Foundation, AO Spine, Davos, Switzerland

3 Department of Neurosurgery, Spine Division, Duke University, Durham, United States

4 Department of Radiation Oncology, Sunnybrook Health Sciences Centre, University of Toronto, Toronto, Canada

5 Department of Neurosurgery, Division of Surgery, The University of Texas MD Anderson Cancer Center, University of Texas, Houston, United States

6 Department of Spine Surgery, University Hospital of Basel, Basel, Switzerland

7 Department of Orthopaedic Surgery, University Medical Center Utrecht, Utrecht University, Utrecht, Netherlands

8 Department of Spine Surgery, IRCCS Istituto Ortopedico Rizzoli, Bologna, Italy

9 Department of Spinal Surgery, Oxford University Hospitals NHS Foundation Trust, Oxford, United Kingdom

10 Department of Neurosurgery, Memorial Sloan Kettering Cancer Center, New York, United States

11 Department of Neurosurgery, Johns Hopkins University School of Medicine, Baltimore, United States

12 ost graduate program of orthopedics, University of Bologna, Bologna, Italy

13 Combined Neurosurgical and Orthopedic Spine Program, Department of Orthopedics Surgery, University of British Columbia, Vancouver, Canada

14 Department of Neurosurgery, Brown University, Warren Alpert School of Medicine, Providence, United States

15 Department of Neurological Surgery, New York University Grossman School of Medicine, New York, United States

Introduction: The Spinal Instability Neoplastic Score is a widely used and cited referral tool to identify patients with potentially unstable spinal metastases; however, treatment decision (ie. need for stabilization) may also be influenced by various morphological and clinical factors. The aim of this study is to develop an evidence-based decision-making tool for surgical treatment in patients with spinal metastases. In this first stage, we identified morphological descriptors associated with biomechanical deterioration. Materials and Methods: Fifty standardized case studies, including CT and MRI images, were evaluated by members of the AO Spine KF Tumor. Key Opinion Leader assessments were obtained regarding 1) risk of biomechanical deterioration, 2) need for stabilization, 3) preferred type of stabilization. These ratings were used to determine the relative severity score for each case. Based on group consensus, 17 different morphological parameters were defined and subsequently evaluated by two experienced spine tumor surgeons using categorical scales. The resulting dataset of morphological descriptors was analyzed in relation to the mean severity scores using univariate and multivariate linear regression models, as well as hierarchical clustering. Results: A total of 45 KF Tumor members participated in the Delphi study. The case collection demonstrated a harmonic distribution of the three severity scores. Risk of biomechanical deterioration was strongly correlated with both the need for surgery (r = 0.994) and the type of stabilization surgery (r = 0.983). The three severity scores were used as dependent variables in three multivariate linear regression models to evaluate the influence of the morphological descriptors. The final models were highly significant (p < 0.0001) and identified segmental alignment, extension of vertebral body fracture, lesion type, facet joint involvement and paravertebral tumor extension as significant predictors. These models explained 76.6%, 75.4% and 73.5% of the variance in the three severity scores. Using these five parameters, hierarchical clustering identified two distinct patient groups with significantly different levels of biomechanical severity (p < 0.0001). Conclusion: Based on our analysis, five morphological parameters were identified as the most important imaging features that may influence a surgeon’s decision-making process on the need for stabilization surgery in patients with spinal metastases. Clinical parameters as well as other factors may modify the therapeutic decision, thus the identification of these key parameters and their inclusion into a comprehensive decision-support system is essential.

ID: AGS26/2951

A223: First use of cesium-131 embedded collagen tiles for brachytherapy in the spine: combined surgical resection and brachytherapy for recurrent rectosigmoid adenocarcinoma metastasis to the sacrum with 1-year follow up

Anisse Chaker 1 , Heegook Yeo 1 , Jessie Huang-Vredevoogd 2 , Anthony Doemer 2 , Lisa Scarpace 1 , Alexandra Moceri 3 , Mira Shah 2 , Ian Lee 1 , Steven Kalkanis 1 , Salim Siddiqui 2 , Adam Robin 1

1 Department of Neurosurgery

2 Department of Radiation Oncology, Henry Ford Hospital Detroit, United States

3 Henry Ford Innovations, Henry Ford Health System, Detroit, United States

Introduction: Cesium-131 embedded collagen tiles (Cs131CT) for high-dose rate brachytherapy has thus far primarily been utilized for intracranial applications. Advantages of Cesium-131 include less tissue penetration thus lowering the risk of damage to nearby organs and neural elements. Utilization of brachytherapy allows for a surgically targeted radiation source that takes effect immediately upon intraoperative implantation. We aim to present the first utilization of Cs131CT brachytherapy in the spine with 1-year follow up. Material and Methods: Patient selection was conducted via multidisciplinary discussion amongst our institutional Spine Tumor Board. Inclusion criteria were limited to patients with metastatic cancer resistant to multiple lines of chemotherapy, biologic therapy, and radiation with symptomatic metastasis to the lumbosacral spine. A 55 year-old male was selected who presented with symptomatic lumbosacral compressive polyradiculopathy from locally advanced and recurrent rectosigmoid adenocarcinoma involving the sacrum and lumbosacral nerve roots. After Spine Tumor Board review, determined options included continued observation with expected disease progression and neurological decline, resection via sacrectomy, or off-label compassionate Cs131CT brachytherapy with tumor debulking and partial sacrectomy. Given the associated morbidity with a high sacrectomy, patient declined this option. A shared-decision was made to proceed with a L5-S3 decompression and partial sacrectomy with tumor debulking and placement of Cs131CT brachytherapy. Patient was followed post-operatively with surveillance imaging for 1-year. Results: In order to simulate the operation in a lab setting prior to surgery, a 3D model of the patient’s spine with the tumor included was first created. An important aspect of the operative simulation utilizing the 3D model was to place the Cs131CT in the surgical cavity and estimate the local radiation dose delivered. This would allow for development of an intra-operative Cs131CT placement strategy that avoids radiation damage to the cauda equina. Once a safe Cs131CT placement plan was achieved in the 3D model, surgery was performed on the patient with successful debulking of the tumor and circumferential decompression. This was followed by placement of the Cs131CT with adequate distance from the neural elements. Post-operative dosimetry calculation based on the patient’s imaging was 106.4Gy - within 10% of the predicted Cs131CT dosimetry. Three-, six-, and 1-year post-operative MRIs were completed without apparent evidence of recurrence at the area of intended brachytherapy. Conclusion: Through the utilization of a pre-surgical model for feasibility and radiation safety planning, we present the first ever use of Cs131CT brachytherapy in the spine. Given the lack of radiographic recurrence at 1 year, this promising technique may warrant further investigation in appropriately selected patients. Strategic planning and establishment of a multi-disciplinary team were critical in providing a potential new alternative for patients faced with limited treatment options.

ID: AGS26/333

A224: Neurological recovery after delayed surgery for MESCC: outcomes in patients with paralysis exceeding 48 hours

Pan-pan HU 1 , Yan Li 1 , Feng Wei 1 , Zhongjun Liu 1 , Ben Wang 1

1 Peking University Third Hospital, Beijing, China

Introduction: Metastatic epidural spinal cord compression (MESCC) represents one of the most debilitating sequelae of advanced cancer, often resulting in rapid neurological decline and loss of ambulation. Current guidelines emphasize decompression within 24-48 hours, yet many patients present outside this window due to delayed recognition or the need for systemic oncologic evaluation. The efficacy of surgery in patients with paralysis exceeding 48 hours remains uncertain. This study assessed neurological and functional recovery in MESCC patients who underwent delayed decompressive surgery. Materials and Methods: A retrospective review was conducted of consecutive MESCC patients treated surgically between 2012 and 2020 at a single tertiary spine center. Inclusion criteria comprised severe neurological deficit (Frankel A-C), non-ambulatory status for > 48 hours, life expectancy >3 months, and pathologically confirmed spinal metastasis. Clinical variables, operative details, complications, neurological outcomes (Frankel grade), ambulatory status, Karnofsky Performance Score (KPS), and survival were analyzed. The primary endpoint was postoperative neurological recovery. Results: Thirty-seven patients met inclusion criteria, with a median preoperative paralysis duration of 9 days (range, 2-35 days). The cohort included 24 males and 13 females, mean age 55.2 years. Lung, prostate, and kidney cancers were the most frequent primary tumors. All patients underwent circumferential decompression with stabilization, and cement augmentation was applied when appropriate. Postoperatively, 30 patients (81.1%) improved at least one Frankel grade. Ambulation was regained in 22 cases (59.5%), including 28.6% of Frankel A, 57.1% of Frankel B, and 75.0% of Frankel C patients. Significant improvements were observed in pain scores and KPS (p < 0.001). Complications occurred in 13 patients (35.1%), but no neurological deterioration was recorded. Median survival was 19 months, with 1-year and 2-year survival rates of 58.9% and 40.3%, respectively. No single clinical factor significantly predicted neurological recovery. Conclusion: Decompressive surgery for MESCC patients with prolonged paralysis (> 48 hours) can still achieve substantial neurological and functional recovery. Nearly 60% of non-ambulatory patients regained walking ability, demonstrating that meaningful benefit remains possible beyond the conventionally recommended surgical window. These findings highlight the value of surgery even in delayed presentations and support a proactive surgical approach in appropriately selected patients.

ID: AGS26/569

A225: Impact of bowel and bladder dysfunction on patient reported outcomes in patients undergoing a sacrectomy for a primary bone tumor

Taryn Ludwig 1 , Laurence Rhines 2 , Chetan Bettegowda 3 , Áron Lazáry 4 , Nicolas Dea 5 , Alessandro Luzzati 6 , Alessandro Gasbarrini 7 , Michelle Clarke 8 , Elizabeth Lord 9 , Praveen Mummaneni 10 , Daniel Tobert 11 , Raphaele Charest Morin 5

1 University of Calgary, Calgary, Canada

2 The University of Texas MD Anderson Cancer Center, Houston, United States

3 Johns Hopkins University School of Medicine, Baltimore, United States

4 National Center for Spinal Disorders, Buda Health Center, Budapest, Hungary

5 University of British Columbia, Vancouver, Canada

6 Department of Orthopaedics, Istituto Ortopedico Galeazzi, Milan, Italy

7 Department of Spine Surgery, IRCCS Istituto Ortopedico Rizzoli, Department of Biomedical and Neuromotor Sciences, University of Bologna, Bologna, Italy

8 Department of Neurosurgery, Mayo Clinic, Rochester, United States

9 Department of Orthopaedic Surgery, UCLA health, Los Angeles, United States

10 Department of Neurosurgery, University of California San Francisco, San Francisco, United States

11 Department of Orthopaedic Surgery, Massachusetts General Hospital, Harvard Medical School, Boston, United States

Introduction: Reports of outcomes after sacrectomy have demonstrated increasing complications, ambulatory difficulties, bowel and bladder issues with more cranial sacral resections. The primary objective was to compare patient reported outcomes (PROs) between patients with normal and impaired bowel and bladder function at 12 months after having undergone a sacrectomy for a primary bone tumor. Secondary objective was to describe bowel and bladder impairment in relation to nerve root sacrifice. Material and Methods: This registry-based observational study included patients from the Primary Tumor Research and Outcomes Network (PTRON) database who underwent sacrectomy (between S1-S5) for a primary bone tumor. To be included in the analysis, patients were required to complete at least one PRO (SOSGOQ; EQ-5D; SF36) completed at baseline and 12 months. Bowel, bladder and sexual function were recorded by the investigators at both baseline and 12 months. Patient demographics, medical history, comorbidities and tumor characteristics were collected as were surgical and perioperative variables including nerve root sacrifices. This study was initiated by the AO Spine Knowledge Forum Tumor. Results: Eighty-one patients were included in this study. Information regarding baseline bowel and bladder function was available for 71 patients. At baseline, 13 (18%) patients presented with impaired bowel and bladder function and 58 (82%) with normal bowel and bladder function. Of these with normal baseline bowel and bladder function, 15 (26%) had impaired bowel and bladder function at 12 months and 12 (92%) with impaired bowel and bladder function remained with impaired bowel and bladder at 12 months. At 12 month post-operatively, EQ-5D was statistically significantly lower in those with impaired function compared to those with normal function (p < 0.05). The SOSGOQ and the SF-36 Physical Component Score (PCS) and Mental Component Score (MCS) at 12 months were similar between those with maintained bowel and bladder and those with impaired function. When comparing to those who had roots caudal to S3 sacrificed, patients who had nerves roots from S3 and cranial sacrificed were more likely to experience impaired bowel and bladder function (p < 0.05). Conclusion: This study provides valuable addition to the limited existing information on how impairment in bowel and bladder function impact PROS after a sacrectomy. Patients with impaired bowel and bladder function had worse EQ5D at 12 months but the SF-36 PCS, MCS and SOSGOQ were similar compared to patients with normal function.

OP26: Patient-Specific Innovation

ID: AGS26/1346

A226: Towards digital outcome measures in spine care: psychometric properties of movemed for degenerative cervical myelopathy

Ali Gharooni 1,2 , Tatiana Houhou 2 , Alvaro Yanez 2 , Benjamin Davies 1

1 University of Cambridge, Cambridge, United Kingdom

2 MoveMed Ltd, Cambridge, United Kingdom

Introduction: Degenerative cervical myelopathy (DCM) suffers from subjective outcome measures with low granularity contributing to imprecise measurements, leading to delayed care in clinics and inflated sample sizes in trials. These limitations have spurred interest in the development of novel assessment tools as a top 10 research priority in DCM. MoveMed app is a smartphone based battery of patient-performed performance outcomes (PerfOMs) targeting dexterity, upper-limb stability, and gait via fast tap, typing, hold, and stand-and-walk tasks administered in natural environments under standardized instructions. It allows high-frequency, remote and objective measurement of DCM. The aim of this abstract is to synthesize validity and reliability evidence and outline a pragmatic roadmap for adoption, cost-effectiveness evaluation, and real-world deployment in spine pathways and research. Material and Methods: Prospective, decentralized observational studies enrolled adults with DCM who completed at least weekly MoveMed in-app battery of tests and periodic ePROMs over a period of 12 weeks. Primary endpoints for MoveMed were fast-tap count (dexterity) and hold stability score (upper-limb stability); typing speed and stand-and-walk cadence were secondary, with COSMIN risk-of-bias appraisal. Construct validity was assessed by testing a priori hypotheses about directionality and magnitude of correlation of MoveMed tests to patient-derived modified Japanese Orthopaedic Association score (P-mJOA) and WHOQOL-Bref. Reliability analyses used a clinically stable subset with repeated measures to compute intraclass correlation (ICC) of agreement and measurement error (standard error of mean(SEM), and smallest detectable change (SDC) for each MoveMed task endpoint, with sufficiency judged at ICC ≥ 0.70 per Criteria from the Consensus-Based Standards for the Selection of Health Measurement Instruments (COSMIN). Results: Construct validity (n = 27) was rated “very good” overall as per COSMIN criteria: 74% (67/90) of unidimensional correlations (fast tap, hold, typing) and 47% (8/17) of multidimensional correlations (stand-and-walk) met a priori hypotheses, with greatest convergence to neuromuscular P-mJOA domains and expected divergence from WHOQOL-Bref. Reliability (n = 7) was “very good” overall as per COSMIN criteria: ICCs were moderate-to-excellent for unidimensional tests (fast tap, hold, typing), with coefficients generally ≥ 0.70 across MCID subgroups, while the multidimensional gait task trended lower and is undergoing optimization. Measurement error thresholds supported individual-level interpretation: ±3-7 taps (fast tap), ±4-10% stability points (hold), ±0.2 keys/s (typing), and ±27 steps/min (gait), enabling discrimination of real change beyond noise. Standardized, smartphone-based PerfOMs demonstrated feasibility for decentralized, high-frequency assessment, complementing PROMs and improving sensitivity to clinically meaningful change in DCM. Conclusion: MoveMed, a smartphone based digital outcome measure, demonstrates good construct validity and test-retest reliability in adults with DCM, supporting remote monitoring and in spine care pathways and trials. Next steps: 1) hybrid implementation-effectiveness studies for adoption in clinics and peri-operative pathways, with electronic healthcare record integration; 2) health-economic modeling and trial simulations to quantify cost-effectiveness by reduced sample sizes and visit burden when using MoveMed; 3) multicenter real-world evidence cohorts embedding MoveMed into registries and 4) further trials assessing psychometric properties (i.e. responsiveness and MCID). Use cases span in-clinic triage and risk stratification, post-operative recovery tracking, decentralized interventional and observational studies, and registry-based surveillance across spine populations.

ID: AGS26/1511

A227: From CT to 3D-printable guide in minutes: automated design of patient-specific pedicle screw templates via point transformer

Qingzhi Xiang 1,2 , Xiao Xia 1,2 , Yan Yu 1,2 , Liming Cheng 1,2

1 Division of Spine, Department of Orthopaedics, Tongji Hospital, Tongji University School of Medicine, Shanghai, China , 2 Key Laboratory of Spine and Spinal Cord Injury Repair and Regeneration (Tongji University), Ministry of Education, Shanghai, China

Introduction: Pedicle screw insertion is a cornerstone of spinal fixation procedures, requiring sub-millimeter positional accuracy and precise angular alignment to avoid neurovascular injury and ensure biomechanical stability. Conventional freehand techniques are highly operator-dependent and associated with variability in accuracy. Manual computer-aided design (CAD) workflows for patient-specific drill guides can improve precision but remain labor-intensive, requiring 1-2 hours of interactive modeling by trained engineers. Surgical robots represent another solution, offering high precision and intraoperative navigation. However, their clinical adoption is constrained by high capital costs, substantial setup time, and the requirement for a specialized operating-room environment. The aim of this research is to develop a fully automated design pipeline that converts spinal CT data into a 3D-printable, patient-specific surgical guide within minutes. Material and Methods: A standardized nnU-Net segmentation pipeline was established to ensure robust vertebral reconstruction across variable image quality and anatomical deformities. Following segmentation, vertebral surface meshes were generated for subsequent processing.A multi-task Point Transformer network was trained to simultaneously (i) localize optimal screw entry points and (ii) segment guide-plate contact regions on the vertebral surface. This architecture leverages both global anatomical context and local geometric features, enabling accurate and reproducible feature extraction without manual annotation.Candidate screw trajectories were computed using a geometric optimization framework with a multi-objective cost function. The function balanced: Safety (cortical breach avoidance, pedicle wall thickness), Biomechanical criteria (trajectory length, maximal bone purchase), Clinical alignment with established surgical guidelines. Patient-specific guides were automatically generated by conforming a base plate to the segmented vertebral surface and embedding guiding channels along optimized screw trajectories. Results: The mean localization error was 2.90 ± 1.16 mm, with 92.5% of entry points falling within 3 mm of expert-defined references. Contact-region segmentation achieved a mean IoU of 0.742 ± 0.050. Optimized screw trajectories avoided cortical breach in 98.2% of vertebrae, and the mean minimal pedicle wall distance was 2.4 ± 0.8 mm. In vitro validation with 3D-printed templates yielded a mean angular deviation of 1.53 ± 0.76° and a translational deviation of 1.82 ± 0.94 mm compared with planned trajectories. These values are comparable to, or in some cases better than, accuracies reported for commercial robotic systems (1.5-2.0° angular deviation, 1.5-2.5 mm translational error). The automated pipeline completed guide generation in under 3 minutes per vertebra, representing a >95% reduction in design time compared with conventional manual CAD workflows (1-2 hours). Conclusion: This study demonstrates a automatic framework for surgical guide design achieved through the integration of deep learning and geometric optimization. The proposed system eliminates labor-intensive manual CAD steps, standardizes guide quality, and offers a cost-effective alternative to surgical robots without sacrificing accuracy. By integrating deep-learning based feature recognition with geometric optimization, our framework provides efficiency, reproducibility, and precision comparable to robotic systems while maintaining the simplicity and low cost of guide-based approaches.

ID: AGS26/1514

A228: Cervical spinal cord stimulation can have promising outcomes: early results of a clinical trial

Harvinder Singh Chhabra 1 , Sunpreet Singh Hanspal 1 , Ankush Gera 1 , Ankit Sharma 1

1 Sri Balaji Action Medical Institute, Delhi, India

Introduction: Despite advancements in management, there have been no effective interventions for restoring voluntary motor control in individuals with SCI. Recent breakthroughs in neuromodulation, particularly spinal cord epidural stimulation (SCES), have shown encouraging potential in facilitating motor recovery and enhancing functional independence in individuals with SCI. This study presents early findings from a clinical trial investigating the effects of SCES when combined with task-specific rehabilitation in individuals with cervical SCI. To our knowledge this is the largest such case series so far. Material and Methods: A total of five participants with chronic tetraplegia (C2, C2-C3, C4, C4, C5 vertebral levels) with AIS-A neurology underwent surgical implantation of a 16-electrode array epidural stimulator, strategically positioned over the cervical spinal cord. Following implantation spinal mapping was done and participants were engaged in an intensive rehabilitation program, incorporating motor retraining and neuromodulation-assisted task-specific training. The study’s primary outcome measures included ISNCSCI upper limb motor scores, spinal cord independence measure (SCIM-III), and capabilities of upper extremity questionnaire (CUE-Q). Results: 1. Improvement in UEMS with Stimulation. o Median improvement in motor score was found to be from 0 (range 0-16) to 16 (range 0- 21). SCES facilitated notable movements in upper limb, particularly in wrist extension, elbow flexion, and hand grasp strength in two out of 5 participants. 2. Functional Independence Gains. o Median improvement in scores of SCIM-III was found to be from 9 (range 0-12) to 12 (range 0-23). Also, median of CUE-Q scores improved from 0 (range 0-11) to 8 (range 0-47). One out of five participants demonstrated improved ability to perform self-care tasks, such as grooming, feeding, and object manipulation when combined with SCES. Another participant could initiate feeding. 3. Neuroplasticity and Retention of Motor Gains. o Notably, one participant retained functional gains even after stimulation was turned off, indicating activity-dependent neuroplasticity and possible synaptic reorganization in spared neural pathways. Conclusion: These early findings highlight SCES as a transformative intervention when combined with neuro-rehabilitation with the potential to enhance motor function and improve independence in individuals with severe cervical SCI. As research advances, a deeper exploration into tailored SCES protocols, optimal SCES parameters, integration with functional training, and long-term neuroplastic adaptation will be crucial in maximizing its clinical impact. Studies with larger number of participants and randomized controlled trials are required to validate and refine SCES-based rehabilitation outcomes.

ID: AGS26/1530

A229: Extended reality-enhanced spine consultations positively impact patient experience

Richard Price 1 , Gabriel Urreolla 1 , Giselle Ghabussi 2 , Jose Castillo 1 , Michael Le 1 , Alan Harris 3 , Safdar Khan 1

1 University of California Davis, Sacramento, United States

2 Lincoln Memorial University, Harrogate, United States

3 University of Louisville, Louisville, United States

Introduction: Spinal disorders, including degenerative disc disease and stenosis, are among the leading causes of global disability. Traditional consultation appointments with patients, which rely on verbal explanations and 2D imaging, can fail to meet the educational needs of patients, particularly those with limited health literacy. When surgical intervention is required, these shortcomings may contribute to increased anxiety, misunderstanding of procedures, and suboptimal surgical outcomes. Augmented reality and virtual reality, collectively termed extended reality (XR), merges digital images with the physical world to create immersive, interactive experiences. XR can be employed to develop 3D models of patient imaging that offers an interactive, patient-centered solution that could enhance comprehension and engagement. Material and Methods: This study evaluated the impact of XR-enhanced consultations on patient understanding, satisfaction, anxiety, and conversion to surgery. Adult patients undergoing evaluation for spinal surgery at a tertiary academic center were enrolled. Each consultation incorporated patient-specific, patient specific 3D spinal models presented through the Surgical Theater ConveyXR platform. Interactive 3D models based on MRI and/or CT were professionally developed with the ConveyXR software. Patients viewed and interacted with the models on a monitor or through a VR headset with the attending surgeon describing the pathology and discussing the planned intervention. Post-consultation surveys assessed patient understanding of their pathology and overall experience with MR. Anxiety related to the procedure was measured using the Amsterdam Preoperative Anxiety and Information Scale (APAIS). Descriptive statistics were calculated using R statistical software. Results: Nineteen patients (mean age 68.9 ± 9.8 years) with surgical pathology were identified by the attending surgical based imaging and consented to participate. Patients reported high comprehension of their condition and treatment (injury: 4.05; location: 4.37; plan: 4.74 on a 5-point scale). Moderate understanding was observed for anatomical models and procedures, though comprehension of complications was lower (2.95). Anxiety scores averaged 16.7/30, with low concern about incisions or inadequate information. Eighty four percent of patient (16/19) proceeded with surgical intervention. Overall, XR consultations were well received, with patients expressing moderate interest in future XR use and reduced desire for second opinions. Ultimately this led to a higher conversion of proceeding with surgery. Conclusion: This proof of concept study demonstrates the potential of XR to improve patient education and engagement in spinal surgery consultations. XR tools enhanced understanding of spinal conditions and treatment plans as well as decreasing anxiety regarding proceeding with spinal surgery. These findings support further evaluation and implemntation of XR tools in spine surgery to improve patient-centered care.

ID: AGS26/2372

A230: The use of a custom 3D-printed patient-specific native PEEK implant for complex en bloc spinal sarcoma resection and reconstruction

Shalin Patel 1 , Karthik Tappa 1 , Tyler Hodge 2 , Joseph Fullerton 2 , Devarsh Vyas 2 , Elsa Arribas 1 , Paul Shay 1 , Christopher Nguyen 1 , Justin Bird 1

1 The University of Texas MD Anderson Cancer Center, Houston, United States

2 3D Systems Inc., Rock Hill, United States

Introduction: Primary spinal sarcomas require complex en bloc resection and reconstruction. Anterior column reconstruction is commonly achieved with vertebral body replacement (VBR), either with titanium or more recently, carbon fiber-reinforced polyetheretherketone (PEEK) implants. However, conventional off-the-shelf VBRs do not account for unique patient anatomy or sacral slope, cannot incorporate vascularized structural autograft to biologically augment fusion, may require additional stabilization to prevent subsidence or implant displacement, and often cause CT/MRI imaging artifacts that make radiation treatment planning and oncologic surveillance challenging. Native PEEK is a biomaterial with a modulus of elasticity similar to cortical bone, is radiolucent, and is now amenable to additive manufacturing allowing for custom 3D printing. To our knowledge, we report the first case of a custom 3D-printed patient-specific native PEEK implant used for complex vertebral reconstruction. Materials and Methods: The patient is a 60-year-old male with a localized rhabdomyosarcoma involving the caudal portion of the L4 vertebral body and a majority of the L5 vertebral body. Staging workup demonstrated no evidence of metastatic disease. The patient was evaluated by our institution’s multidisciplinary sarcoma tumor board. Consensus recommendation was to proceed with neoadjuvant systemic therapy, en bloc surgical resection, and adjuvant radiation therapy. The planned resection included the L4 vertebral body caudal to the L4 pedicles, the L4-L5 disc, and the L5 vertebral body. Additionally, the decision was made to preserve the anterior and right lateral vertebral body walls and disc annulus to limit the risk of major neurovascular injury during retroperitoneal dissection and to maintain the anterior longitudinal ligamentous anatomy. A custom, patient-specific native PEEK implant was then designed in conjunction with our engineering and surgical colleagues with the following goals in mind: to maximize anterior column biomechanical support, incorporate a free vascularized fibula autograft to augment biologic fusion, accommodate the patient’s sacral slope, provide secure fixation of the implant to the L4 and S1 vertebral bodies, and to facilitate MRI-based 1) treatment planning for adjuvant radiation therapy and 2) oncologic surveillance. Institutional IRB approval, FDA compassionate use approval, and patient consent was obtained for the use of this implant. Results: A two-stage en bloc resection of the lumbar sarcoma was successfully performed with negative margins. The custom 3D-printed patient-specific native PEEK implant was successfully implanted as planned along with the free vascularized fibula autograft. The patient tolerated surgery and postoperative recovery well. He received adjuvant radiation therapy without issue and routine sarcoma surveillance MR imaging demonstrates the patient has no evidence of disease. Conclusions: This case illustrates the feasibility of using a custom, patient-specific 3D-printed native PEEK implant in complex spinal tumor resection and reconstruction. This novel reconstructive approach highlights the potential of additive manufacturing in improving anatomical fit, optimizing biomechanical performance, facilitating integration of vascularized structural autograft, and limiting imaging artifact for postoperative radiation treatment and oncologic surveillance. Further studies are warranted to assess long-term outcomes in more patients, implant durability, and cost-effectiveness.

ID: AGS26/28

A231: The validity and educational impact of virtual reality in training spine surgeons: a systematic review

Matthew Cecere 1 , Luke Reda 1,2,3 , Jeremie Larouche 1,2,3

1 Temerty Faculty of Medicine, University of Toronto, Toronto, Canada

2 Sunnybrook Health Sciences Centre, Division of Orthopaedic Surgery, Toronto, Canada

3 Division of Orthopaedic Surgery, University of Toronto, Toronto, Canada

Introduction: Virtual reality (VR) simulation is an emerging educational tool in spine surgery, offering a risk-free, immersive environment for surgical skill development. However, its validity and educational impact in spine-specific training remains to be comprehensively evaluated. We aimed to systematically assess the validity evidence and educational outcomes of VR simulation in training spine surgeons. Material and Methods: A systematic review was conducted using MEDLINE, EMBASE, and SCOPUS to identify English-language studies published between 2000 and 2024. Eligible studies evaluated VR simulation for orthopedic spine surgery training. Studies focusing on patient education, rehabilitation, augmented reality, or non-spine procedures were excluded. Of 7,456 articles screened, 58 underwent full-text review, and 13 met final inclusion criteria. Extracted data included simulator features, spine procedures simulated, participant characteristics, and outcomes related to face, content, and construct validity, as well as performance differentiation and correlation with validated assessment tools. Results: Thirteen studies involving over 300 participants were included. Simulated procedures included pedicle screw placement, spinal decompression, balloon kyphoplasty, and minimally invasive fusions. Eleven studies demonstrated face and content validity, often using Likert-based realism scores and procedural fidelity checklists. Construct validity was supported in 10 studies through performance stratification by training level. Several studies reported statistically significant improvements in post-training performance. For example, one study demonstrated increased OSATS and PPDIS scores in 93% of trainees. While validated metrics such as OSATS and GRS were used, their application was inconsistent across studies. VR-based training improved surgical competency in four core domains: anatomical understanding, purposeful motion, complex coordination, and force/vector-based kinematics. Conclusion: VR simulation demonstrates consistent validity evidence and measurable performance benefits in spine surgery training, particularly among early-stage learners. Its integration into competency-based curricula shows promise, though broader adoption will require greater standardization of assessment tools and validation protocols.

ID: AGS26/3483

A232: Descriptive study of artificial intelligence-assisted preoperative evaluation in patients with lumbar spinal stenosis

Sebastian Moreno Herrera 1 , Hugo Vilchis Sámano 2 , Luis Herrera Alcaraz 2

1 Consejo Mexicano de Ortopedia y Traumatología, AO Spine Latin America, Ciudad de Mexico, Mexico,

2 Consejo Mexicano de Ortopedia y Traumatologia, Asociacionismo Mexicana de Cirugia de Columna, Ciudad de Mexico, Mexico,

Introduction: Lumbar spinal stenosis is one of the leading causes of low back pain and neurogenic claudication in the adult population, with a high incidence in patients over 60 years of age. This condition is characterized by a progressive narrowing of the spinal canal and/or intervertebral foramina, which causes compression of the nerve roots with significant functional impairment. Preoperative analysis is a fundamental step in therapeutic decision-making, as it allows for the identification of critical radiological parameters such as the lumbar lordotic angle, pelvic incidence, pelvic tilt, and sagittal imbalance, which are associated with a patient's surgical prognosis. However, these measurements are often manual, require time and specialized expertise, and can present interobserver variability. In recent years, artificial intelligence (AI) has emerged as an innovative tool in the healthcare field. The use of AI in preoperative analysis could optimize surgical planning, reduce diagnostic variability, and promote greater accuracy in predicting outcomes. Material and Methods: An observational, descriptive, and cross-sectional study was conducted in patients with lumbar spinal stenosis treated at the Spine Service of the Lomas Verdes Traumatology and Orthopedics Hospital between March and September 2025. Clinical records and digital images (X-rays and MRIs) in DICOM format were collected and processed using artificial intelligence software for the automatic detection of radiological parameters (PI, PT, SS, LL, sagittal balance), and surgical treatment was proposed based on the AI analysis. Clinical variables, radiological findings, and AI performance metrics (analysis time, success rate) were recorded. Results: Thirty-five patients were analyzed (mean age 64.8 ± 8.4 years; 48.6% men). The IA quantified PI 53.8°, PT 23.0°, SS 34.6°, LL 45.2°; a PI-LL mismatch > 10° was observed in 45.7%. Roussouly morphology was: Type 3 48.6%, Type 1 22.9%, Type 2 20.0%, Type 4 8.6%. The most affected levels were L5-S1 (60.0%) and L4-L5 (45.7%); multilevel stenosis occurred in 42.9%. The dominant Schizas was B 48.6%, C 28.6%, D 20.0%, and A 2.9%. In the clinical indication, Decompression was proposed in 15/35 (42.9%) and Decompression + Fusion in 20/35 (57.1%). In the fusion subgroup (n = 20): TLIF 65.0%, ALIF 25.0%, LLIF/OLIF 10.0%; MIS 70.0% vs open 30.0%; TLIF 1 level 75.0% and 2 levels 25.0%. In the exploratory analysis, Decompression showed PI 55.8°, LL 43.8° and |PI-LL| > 10° in 53.3%; Fusion had PI 52.2°, LL 46.3° and |PI-LL| > 10° in 40.0%. Conclusion: AI-assisted preoperative evaluation of lumbar spinal stenosis was feasible and efficient, with a high rate of automatic measurement of spinopelvic parameters and short processing times. Systematic characterization of the sagittal profile (PI, PT, SS, LL, PI-LL) and severity/radiology (Schizas, levels) provided useful information to guide the indication (decompression vs. decompression + fusion) and define the most common management strategy in practice. However, AI should complement, not replace, comprehensive clinical judgment. Prospective studies with larger sample sizes and comparisons with manual measurements are needed to validate the accuracy, reproducibility, and impact on clinical outcomes and surgical planning.

ID: AGS26/374

A233: Evaluating surgeon proficiency in the novel robotic-assisted muscle-preserving (RAMP) decompression

Giuseppe Loggia 1 , Fedan Avrumova 1 , Marco Burkhard 1 , Michael Kelly 1 , Franziska Altorfer 1 , Maksym Polt 1 , Jiaqi Zhu 1 , J. Levi Chazen 1 , Ek Tsoon Tan 1 , Darren R. Lebl 1

1 Hospital for Special Surgery, Balgrist Zurich, New York, United States

Introduction: Robotic-assisted systems have enhanced precision in spinal instrumentation, but their use in decompression procedures remains underexplored. Robotic-assisted muscle-preserving (RAMP) decompression is a novel technique designed to achieve controlled laminar bone removal while preserving paraspinal musculature. The impact of prior surgical experience on proficiency with this technique has not been systematically evaluated. The objective of this study was to evaluate whether prior surgical experience influences the accuracy and efficiency of RAMP decompression in human cadaveric specimens. Material and Methods: RAMP decompressions were performed from T8 to L5 in eight cadavers by a non-fellowship-trained surgeon (NFTS) and a fellowship-trained surgeon (FTS). Computed tomography (CT) images were used for preoperative robotic planning and registration. Operative durations were documented. Postoperative CT imaging evaluated decompression accuracy and quantified anterior cortical bone removal (ACBR). Results: A total of 80 levels underwent RAMP decompression (40 thoracic, 40 lumbar). Median single-level operative times were 2.6 min (IQR 1.9-3.7) for NFTS and 2.1 min (IQR 1.6-3.0) for FTS, with no significant difference. Decompression accuracy showed submillimeter deviation from planned margins, with posterior deviation measuring 0.6 mm (IQR 0.5-1.5) for NFTS and 0.7 mm (IQR 0.4-1.0) for FTS, and anterior deviation measuring 0.3 mm (IQR 0.1-0.5) in both. ACBR did not differ significantly between NFTS and FTS. Operative times declined with experience in both surgeons. Initial single-level RAMP decompressions showed greater variability (NFTS: 3.13-8.10 min; FTS: 1.39-10.56 min), while later cases converged to narrower ranges (NFTS: 1.18-2.46 min; FTS: 1.16-2.11 min). Conclusion: This first comparative study evaluating surgeon experience in the novel RAMP decompression demonstrates high accuracy and incrementally improving efficiency regardless of prior robotic experience. These findings support the integration of decompression into robotic workflows and pave the way for a broader applicability of high-precision, muscle-preserving techniques in spine surgery.

ID: AGS26/654

A234: Comparative analysis of machine learning models for osteoporosis prediction using lumbar CT-derived Hounsfield Units

Akihiko Hiyama 1 , Nobuaki Hattori 1 , Daisuke Sakai 1 , Masahiko Watanabe 1

1 Department of Orthopaedic Surgery, Tokai University School of Medicine, Isehara, Kanagawa, Japan

Introduction: Osteoporosis is a major risk factor for implant loosening, nonunion, and postoperative complications following spine surgery. Dual-energy X-ray absorptiometry (DXA) is widely used as the standard diagnostic tool; however, its availability is often limited, and degenerative spinal changes can lead to overestimation of bone mineral density (BMD). Recent studies have demonstrated the potential of lumbar CT-derived Hounsfield Unit (HU) values as a surrogate marker for BMD, offering the advantage of opportunistic assessment without additional imaging. Machine learning (ML) techniques have further enabled the development of predictive models that can improve diagnostic accuracy. Nevertheless, a comprehensive comparison of different ML algorithms using CT-derived HU values for osteoporosis screening remains limited. The present study aimed to evaluate and compare the diagnostic performance of seven commonly used ML models using mean HU values from L1-L4 vertebrae in patients undergoing spine surgery. Material and Methods: We retrospectively analyzed 172 patients (mean age 71.2 ± 9.8 years; 68 males, 104 females) who underwent both lumbar CT and DXA between 2016 and 2025 prior to spine surgery. Osteoporosis was defined as a T-score < -2.5 at either the lumbar spine or femoral neck. The mean HU value across L1-L4 vertebrae was measured on preoperative CT scans and used as the sole feature to train seven ML models: logistic regression, random forest, support vector machine (SVM), k-nearest neighbors (KNN), naïve Bayes, decision tree, and gradient boosting. Fivefold cross-validation was performed to evaluate accuracy, precision, recall, F1 score, and area under the receiver operating characteristic curve (AUC). Optimal HU thresholds for osteoporosis classification were determined using Youden’s Index. Class imbalance was addressed using model-based weighting strategies. Learning curves were constructed to assess model generalizability and risk of overfitting. Statistical significance was set at p < 0.05. Results: Among the 172 patients, 59 were classified into the osteoporosis group, and 113 into the non-osteoporosis group. The osteoporosis group exhibited significantly lower HU values compared with the non-osteoporosis group (105.1 ± 47.5 vs. 140.1 ± 61.3, p < 0.01). KNN demonstrated the most balanced diagnostic performance, achieving an F1 score of 0.466 ± 0.073, accuracy of 0.714 ± 0.048, and AUC of 0.705 ± 0.046. Logistic regression (AUC = 0.785) and SVM (AUC = 0.745) also showed high discriminatory ability, although with lower recall, while tree-based models such as decision tree and gradient boosting tended to overpredict osteoporosis cases, leading to reduced specificity. Conclusion: ML models utilizing lumbar CT-derived HU values offer a practical and cost-effective approach for opportunistic osteoporosis screening in preoperative spinal fusion surgery settings. Among the evaluated algorithms, KNN achieved the best balance between sensitivity and specificity, suggesting its suitability as a screening tool when DXA is unavailable or impractical. Logistic regression and SVM also demonstrated strong classification performance and may be preferable when higher diagnostic precision is required. These findings support the integration of ML-based HU analysis into routine preoperative assessment protocols to optimize surgical planning and reduce osteoporosis-related complications.

OP27: Outcomes for Cervical Myelopathy

ID: AGS26/2208

A235: Dynamic MRI in cervical myelopathy: factors associated with new compression findings

Juan Ignacio Perez Abdala 1 , Gonzalo Kido 1 , Ivan Huespe 1 , Marcelo Gruenberg 1 , Matias Petracchi 1

1 Orthopedic Department, Hospital Italiano de Buenos Aires, Buenos Aires, Argentina

Introduction: Cervical spondylotic myelopathy is the compression of the spinal cord secondary to degenerative narrowing of the canal. Dynamic cervical magnetic resonance imaging (MRI) can reveal new compressions that are not evident in neutral cervical MRI. However, there is currently no consensus on the indications for dynamic MRI. This study aimed to identify radiographic and MRI factors associated with increased spinal cord compression when comparing neutral and dynamic MRI. Material and Methods: We conducted a retrospective, cross-sectional study from 2017 to 2024, encompassing all patients with symptoms of spondylotic cervical myelopathy who underwent both dynamic and neutral cervical MRI and cervical X-rays. We collected alignment and stability variables from neutral and dynamic lateral cervical X-rays. Static MRI variables included paraspinal muscle quality, assessed using the Goutallier classification. The canal diameter at C7 and the spinal cord-to-canal ratio (CCR) were measured on axial and sagittal MRI, respectively. Additionally, we recorded disc degeneration according to the Pfirrmann classification, intramedullary T2 high-intensity signal (HIS), facet joint degeneration, and the presence of facet effusion. Anterior and posterior spinal cord compression were graded using the Muhle scale. A cervical compression change was defined as an increase of ≥ 2 grades in the Mulhe classification from neutral to dynamic MRI. Therefore, levels graded as Mulhe 2 or 3 on neutral MRI were not considered suitable for further compression changes. We calculated the proportion of cervical compression changes between neutral and dynamic cervical MRI. Subsequently, we analyzed the risk of cervical compression change based on neutral and dynamic X-rays and regional parameters obtained from neutral cervical MRI. Finally, we compared each cervical level by clustering them into groups with and without compression changes. We performed mixed-effects logistic regression analyses with random intercepts for patients to account for multiple cervical levels per subject. Odds ratios with 95% confidence intervals were calculated, and model fit was evaluated using likelihood ratio tests. Results: 82 patients were included, and 57 (69.5%) had a cervical compression change at some level. The C4-5 level showed the highest number of changes, which were more frequent in extension than in flexion dynamic MRI. No significant associations were observed between demographic, radiological, or MRI factors and the presence of cervical compression changes. When comparing cervical levels, Pfirrmann grades 4 and 5 were significantly associated with compression changes, with ORs of 5.90 (95% CI: 1.08-32.22; p = 0.041) and 6.97 (95% CI: 1.23-39.45; p = 0.028), respectively. The CCR also showed a strong association (OR = 1.08; 95% CI: 1.04-1.13; p < 0.01). However, we did not find an association with HIS and the degree of facet arthropathy. Conclusion: Dynamic cervical MRI detected new levels of spinal cord compression in nearly 70% of cases. Higher Pfirrmann grades and increased CCR were significantly associated with these changes. These findings underscore the value of dynamic MRI in detecting hidden compression and guiding clinical decision-making in cervical myelopathy.

ID: AGS26/1365

A236: The CHANGE criteria for screening patients with suspected degenerative cervical myelopathy

Lindsay Tetreault 1 , Lance Burn 2 , David Anderson 3 , Konstantinos Margetis 4 , Julio Furlan 5,6 , Tanzil Rujeedawa 7 , Ricardo Rodrigues-Pinto 8,9 , Brian Kwon 10 , Mark Kotter 11 , Lianne Wood 12 , Rohil Chauhan 13,14 , Carl Zipser 15 , Jamie Wilson 16 , Caroline Treanor 17,18 , Noam Harel 19 , Michael Fehlings 20,21 , James Harrop 22 , James Milligan 23 , Ligia Onofrei 24 , Michael Lee 25,26 , Mike Hutton 27,28 , Justin M. Lantz 29 , Ratko Yurac 30 , Sybil Stacpoole 31 , Benjamin Davies 32

1 Department of Neurology, Massachusetts General Brigham, Boston, Massachusetts, United States,

2 North West Anglia NHS Foundation Trust, Huntingdon, United Kingdom,

3 Faculty of Medicine and Health, School of Health Sciences, University of Sydney, New South Wales, Australia,

4 Department of Neurosurgery, Icahn School of Medicine at Mount Sinai, New York, New York, United States,

5 KITE Research Institute and Lyndhurst Centre, Toronto Rehabilitation Institute, University Health Network, Toronto, Ontario, Canada,

6 Department of Medicine, Division of Physical Medicine and Rehabilitation, University of Toronto, Toronto, Ontario, Canada,

7 Royal Victoria Infirmary, Newcastle Upon Tyne, United Kingdom,

8 Spinal Unit (UVM). Department of Orthopaedic Surgery, Centro Hospitalar Universitário de Santo António, ULS de Santo António, Porto, Portugal,

9 ICBAS School of Medicine and Biomedical Sciences, Universidade do Porto, Porto, Portugal,

10 Vancouver Spine Surgery Institute, International Collaboration on Repair Discoveries (ICORD), University of British Columbia, Vancouver, British Columbia, Canada,

11 Department of Clinical Neurosciences, University of Cambridge, Cambridge, United Kingdom,

12 Department of Public Health and Sports Science, University of Exeter, Exeter, United Kingdom,

13 Department of Orthopaedic Physiotherapy, Auckland Spine Surgery Centre, Aukland, New Zealand,

14 Active Living and Rehabilitation: Aotearoa New Zealand, Health and Rehabilitation Research Institute, Auckland University of Technology, Aukland, New Zealand,

15 Spinal Cord Injury Center and Department of Neurology and Neurophysiology, Balgrist University Hospital, University of Zurich, Zurich, Switzerland,

16 Department of Neurosurgery, Medical University of South Carolina, Charleston, South Carolina, United States,

17 Department of Physiotherapy & Neurosurgery, Beaumont Hospital, Dublin, Ireland,

18 School of Physiotherapy, Royal College of Surgeons in Ireland, Dublin, Ireland,

19 Departments of Neurology and of Rehabilitation and Human Performance, Icahn School of Medicine at Mount Sinai, New York, NY, United States,

20 Division of Neurosurgery and Spine Program, Department of Surgery, University of Toronto, Toronto, Ontario, Canada,

21 Division of Neurosurgery, Krembil Neuroscience Centre, Toronto Western Hospital, University Health Network, Toronto, Ontario, Canada,

22 Department of Neurological Surgery, Thomas Jefferson University, Philadelphia, PA, United States,

23 Department of Family Medicine, McMaster University, Hamilton, ON, Canada,

24 Department of Neurology, University of Utah, Salt Lake City, UT, United States,

25 The Hong Kong Neurosurgical Society, Hong Kong, Hong Kong,

26 Department of Neurosurgery, Pamela Youde Nethersole Eastern Hospital, Hong Kong, Hong Kong,

27 Consultant Spine Surgeon, New Hall Hospital, Salisbury, United Kingdom,

28 National Clinical Lead Spine Services, NHS England, London, United Kingdom,

29 Division of Biokinesiology and Physical Therapy, University of Southern California, Los Angeles, United States,

30 Department of Orthopedics and Traumatology, University del Desarrollo, Clinica Alemana de Santiago, Santiago, Chile,

31 University of Cambridge, Cambridge, United Kingdom,

32 Academic Neurosurgery Unit, Department of Clinical Neurosurgery, University of Cambridge, Cambridge, United Kingdom

Introduction: The diagnosis of degenerative cervical myelopathy (DCM) is often delayed due to its heterogeneous presentation and the lack of standardized diagnostic and screening criteria. Delayed diagnosis of DCM has significant consequences, including the development of more permanent or severe neurological impairment, incomplete or suboptimal postoperative recovery and increased dependence on caregivers. The AO Spine RECODE-DCM (Research Objectives and Common Data Elements for DCM) initiative identified developing diagnostic criteria as one of the top three research priorities. Objective: The objectives of this study were to (i) develop pragmatic, evidence-based screening criteria to help frontline healthcare providers identify patients at risk for DCM and (ii) define key clinical scenarios in which the application of the screening criteria would be most appropriate. Methods: A three-step process was used to develop screening criteria for DCM. First, two systematic reviews were conducted to determine the frequency and diagnostic accuracy of DCM-related symptoms and signs. Second, an international survey was distributed to a multidisciplinary group of clinicians asking them to rank thirty-nine symptoms and thirty-four signs in terms of their importance for diagnosing DCM. Third, an Expert Working Group engaged in a structured consensus process to refine and rank candidate criteria. The group also identified key clinical scenarios for application through discussion. Results: Fifteen symptoms and twelve signs identified as having high diagnostic utility in an international expert survey were presented to an Expert Working Group for consensus discussion. The group comprised 21 specialists, including neurologists (n = 6), spine surgeons (n = 10), rehabilitation specialists or physiotherapists (n = 4), and a primary care physician (n = 1). Following consensus discussions, the top eight features were consolidated into six core criteria: C lumsy or weak hands, H offmann or Tromner sign, A taxia or imbalance, N umb or tingling hands, G ait abnormalities, and E xaggerated reflexes, forming the mnemonic “CHANGE.” The group recommended applying these criteria to any patient with motor or sensory complaints of the extremities or gait dysfunction, with special emphasis on five clinical scenarios: suspected carpal tunnel syndrome, neck pain with or without cervical radiculopathy, lumbar radiculopathy or spinal stenosis, recurrent falls, and incidental cervical cord compression on MRI. Conclusion: By integrating evidence from the literature with multidisciplinary and geographically diverse expert input, we developed a clinically relevant tool to support frontline healthcare providers in the early recognition of DCM. The screening criteria integrate common symptoms and easily elicited signs of DCM into a simple mnemonic “CHANGE” to support early recognition by non-specialists. By prompting targeted neurological assessment, this tool aims to reduce diagnostic delays and expedite surgical referral when indicated. Widespread adoption of CHANGE has the potential to improve detection, prevent progression, and enhance outcomes for patients with DCM.

ID: AGS26/1483

A237: Development of a predictive model for the use of dynamic cervical MRI in patients with degenerative cervical myelopathy

Juan Ignacio Perez Abdala 1 , Gonzalo Kido 1 , ivan huespe 1 , Marcelo Gruenberg 1 , Matias Petracchi 1

1 Hospital Italiano de Buenos Aires - Orthopedic Department, Buenos Aires, Argentina

Introduction: The use of dynamic cervical magnetic resonance imaging (MRI) for the diagnosis and management of degenerative cervical myelopathy remains a subject of debate. Our objective is to develop a predictive model based on radiographs and static MRI to identify patients who would benefit from dynamic cervical MRI. Materials and Methods: We conducted a cross-sectional study from 2017 to 2024, including adults (≥ 18 years) with clinical features of cervical myelopathy and available static and dynamic cervical MRI, as well as lateral radiographs in neutral, flexion, and extension positions. Exclusion criteria for patients were trauma, prior cervical surgery, inflammatory disease, congenital anomalies, or poor-quality imaging. We collected alignment and stability variables from neutral and dynamic lateral cervical radiographs. Static MRI variables included paraspinal muscle quality, assessed using the Goutallier classification. The canal diameter at C7 and the spinal cord-to-canal ratio (CCR%) were measured on axial and sagittal MRI, respectively. Additionally, we recorded disc degeneration according to the Pfirrmann classification, intramedullary T2 high-intensity signal (HIS), facet joint degeneration, and the presence of facet effusion. Anterior and posterior spinal cord compression were graded using the Muhle scale. Cervical levels with anterior or posterior compression ≥ grade 2 on static MRI were excluded from the final analysis. The outcome was new dynamic compression, defined as a ≥ 2-point increase in Muhle grade-both anterior and posterior-between static and dynamic cervical MRI. Predictors included variables explained previously. Each cervical level was analyzed, with clustering accounted for within individual patients. Predictor variables were evaluated using a log-odds analysis. We developed a predictive model and assessed its discrimination and calibration. Results: We included 81 patients, of whom 47% were male, with a mean age of 55.7 years. Among 491 cervical levels from 81 patients, 284 levels without baseline compression were included in the analysis. New dynamic compression was identified in 56 levels (19.7%). Disc degeneration grade and the CCR% showed strong and statistically significant associations with the outcome, with p-values of 0.055 and < 0.001, respectively. After testing multiple predictive models, we selected a final model that incorporated these two variables. Model performance was strong, with an area under the ROC curve (AUROC) of 0.896 (95% CI: 0.846-0.946) and good calibration (Brier score: 0.117). To assess clinical utility, we applied a sensitivity-focused probability threshold of 0.19. Using this cutoff, the model achieved 100% sensitivity, 59.0% specificity, and an overall accuracy of 80.3%. Conclusion: The predictive model we developed, based on just two variables from static MRI, demonstrated strong accuracy in identifying patients at risk of dynamic spinal cord compression. This approach allows for the safe reduction of dynamic MRI use in low-risk patients, optimizing resource utilization and enhancing clinical efficiency. Additionally, this tool may assist in identifying previously unrecognized sites of compression and support clinical decision-making.

ID: AGS26/1709

A238: Recovery patterns from C5 palsy after anterior cervical decompression and fusion, posterior cervical decompression and fusion, and laminoplasty for degenerative cervical myelopathy: systematic review and meta-analysis of 748 C5 palsy cases

Sathish Muthu 1,2 , Vibhu Krishnan Viswanathan 1,3 , Guna Pradeep 1,4

1 Department of Spine Surgery, Orthopaedic Research Group, Coimbatore, India

2 Department of Orthopaedics, Government Medical College, Karur, India

3 Department of Spine Surgery, Devadoss Multispeciality Hospital, Madurai, India

4 Department of Orthopaedics - Spine Division, Meenakshi Mission Hospital, Madurai, India

Introduction: Despite the favorable postoperative prognosis of C5 palsy (C5P), a certain proportion of these patients have less satisfactory outcomes. The current systematic review and meta-analysis thus aimed to comprehensively evaluate existing literature and identify the onset, recovery patterns, and outcomes of C5P following diverse surgical approaches. Materials and Methods: Five different databases (Google Scholar, Embase, PubMed, Web of Science, and Cochrane Library) were thoroughly searched for relevant literature on October 15, 2024. Studies reporting on incidences of C5P following surgery for degenerative cervical conditions with recovery data published until 2024 were scrutinized. Narrative or systematic reviews, opinions, letters to the editor, and manuscripts published in non-English languages were excluded. Results: A total of 30 articles involving 8,116 patients who underwent undergoing surgery for degenerative cervical myelopathy with 748 reported C5P cases were included for analysis. The overall time to palsy reported in the included studies was 3 days (95% confidence interval [CI], 2.56-3.60). Palsy occurred earliest with anterior cervical decompression and fusion (ACDF) at 2 days (95% CI, 0.35-4.54), followed by laminoplasty (LP) at 3.2 days (95% CI, 2.02-4.34) and posterior cervical decompression and fusion (PCDF) at 3.6 days (95% CI, 2.81-4.37). Patients with palsy showed improved recovery with time. At the 1-year follow-up, the reported recovery rates were 100%, 52.9%, and 50% for ACDF, LP, and PCDF, respectively. Conclusions: C5P demonstrated a delayed presentation, with mean onset of 3 days after surgery, which can range from 2 days for ACDF to 3.6 days for PDCF. Recovery improved progressively with time and varied for different surgical procedures, with ACDF showing the best recovery and PDCF for cervical myelopathy showing the poorest recovery.

ID: AGS26/2621

A239: Is patient satisfaction static at 5 years after surgery for cervical spondylotic myelopathy? A spine CORe QOD review

Aidan Rezner 1 , Kathleen Hannon 1 , Vincent Alentado 1,2 , Nicholas Tippins 1 , Andrew Chan 3 , Paul Park 4 , Erica Bisson 5 , Regis Haid 6 , Mohamad Bydon 7 , Anthony Asher 8 , Domagoj Coric 8 , Kevin Foley 4 , Dean Chou 3 , Kai-Ming Fu 9 , Michael Wang 10 , Michael Virk 9 , Timothy Yee 11 , Praveen Mummaneni 12 , John Knightly 13 , Christopher Shaffrey 14 , Oren Gottfried 14 , Cheerag Upadhyaya 15 , Luis Tumialan 16 , Chun-Po Yen 17 , Eric Potts 1 2

1 Goodman Campbell Brain and Spine, Carmel, IN, United States

2 Ascension St. Vincent Hospital, Indianapolis, IN, United States

3 Department of Neurosurgery, Columbia University, New York, NY, United States

4 Semmes Murphey Neurologic and Spine Institute, Memphis, TN, United States

5 Department of Neurosurgery, University of Utah, Salt Lake City, UT, United States

6 Atlanta Brain and Spine Care, Atlanta, GA, United States

7 Department of Neurological Surgery, University of Chicago, Chicago, IL, United States

8 Carolina Neurosurgery and Spine Associates, Carolinas Medical Center, Charlotte, NC, United States

9 Department of Neurological Surgery, Weill Cornell Medicine, New York, NY, United States

10 Department of Neurological Surgery, University of Miami, Miami, FL, United States

11 Department of Neurosurgery, University of Michigan, Ann Arbor, MI, United States

12 Department of Neurological Surgery, University of California, San Francisco, San Francisco, CA, United States

13 Atlantic Neurosurgical Specialists, Altair Health Spine & Wellness Center, Morristown, NJ, United States

14 Department of Neurosurgery, Duke University Medical Center, Durham, NC, United States

15 Department of Neurosurgery, University of North Carolina, Chapel Hill, NC, United States

16 Department of Neurosurgery, Barrow Neurological Institute, St. Joseph's Hospital and Medical Center, Phoenix, AZ, United States

17 Department of Neurosurgery, University of Virginia, Charlottesville, VA, United States

Introduction: Surgical management of cervical spondylotic myelopathy aims to improve neurologic deterioration. However, long-term predictors of outcome are unknown. The present study examined highest and lowest patient-reported satisfaction at 5-year follow-up. We hypothesized that most satisfied patients experience durable long-term improvement and have distinct characteristics compared to least satisfied. Material and Methods: A prospectively collected quality outcomes database (QOD) from the Spine CORe Study Group was retrospectively reviewed. Patients were stratified by NASS satisfaction scores of 1 (“most satisfied”) and 4 (“least satisfied”). Demographics, surgical characteristics, and patient-reported outcomes metrics (PROMs) were compared between cohorts at baseline, 24-month, and 60-month follow-up. A mixed-effects logistic regression assessed independent predictors of 60-month most and least satisfaction. Univariate and multivariable analyses were repeated in the 2-year most satisfied sub-sample. Univariate analysis was completed for the 2-year least satisfied sub-sample. Results: Of 1,085 patients with CSM, 976 were alive at 5-year follow-up and 785 (80.4%) provided a 5-year satisfaction score. Of the 621 patients meeting inclusion criteria, 558 (90.1%) were most satisfied and 61 (9.9%) were least satisfied. The most satisfied cohort included more bachelor’s degrees (25% vs. 13%, p = 0.011), more ACDFs (61% vs. 48%, p = 0.038), shorter hospitalizations (1.7 ± 1.8 vs. 2.8 ± 2.6 days, p = 0.002), fewer laminectomies (23% vs. 39%, p = 0.004), and fewer smokers (p = 0.036). Most satisfied patients reported higher baseline EQ-VAS (60.8 ± 21.3 vs. 51.5 ± 24.8, p = 0.008), but other baseline PROMs were similar. In multivariable analysis, postsecondary education (OR = 2.54, p = 0.004), preoperative depression (OR = 2.75, p = 0.043), higher baseline EQ-VAS (OR = 1.02, p = 0.009), and shorter hospitalization (OR = 0.81, p = 0.003) independently predicted 60-month most satisfaction. Among 60-month least satisfied patients, 28% were most satisfied at 24 months. Among 60-month most satisfied patients, 2.9% were least satisfied at 24 months. Among 24-month most satisfied patients, maintenance of most satisfaction at 60 months was independently predicted by older age (OR = 1.09, p = 0.045), preoperative pain (OR = 9.28, p = 0.013), and higher 24-month NP-NRS (OR = 1.58, p = 0.047). NDI and mJOA-based independence measures correlated with bidirectional satisfaction shift between 2 and 5 years. Conclusion: Highest satisfaction at 5 years is predicted by education level, depression, baseline EQ-VAS, and length of stay. Younger age, lack of preoperative pain, and lower 2-year NP-NRS predict movement from highest to lowest satisfaction. Most patients report highest satisfaction at 5 years, but delayed satisfaction reversals exist and warrant monitoring.

ID: AGS26/496

A240: Performance on tandem gait does not predict the presence of cervical myelopathy or stenosis

Brody Fitzpatrick 1 , Jakub Ratkowski 1 , Nathaniel Luketich 1 , Kyle Holmberg 1 , Julio Rodriguez Jr. 1 , Felicity Fisk 1 , Jonathan Karnes 1 , Scott Daffner 1 , John France 1

1 West Virginia University, Morgantown, United States

Introduction: The diagnosis of degenerative cervical myelopathy (DCM) is based on the combination of subjective complaints and clinical findings in the setting of radiographically-confirmed cervical stenosis. Patients often report a history of gait disturbance, which is a major component of the Nurick and Japanese Orthopaedic Association (JOA) scales. These scales, however, are limited by the subjectivity of their assessments. Clinically, the tandem gait test is utilized to evaluate gait disturbance; however, there is limited evidence in its ability to predict the presence of myelopathy. The purpose of this study is to determine whether performance on a graded 10-step tandem gait test predicts the presence of myelopathy. Material and Methods: Patients referred to our spine clinic for evaluation of cervical pathology over a six-week period were prospectively enrolled. Inclusion criteria were age ≥ 18 and cervical MRI obtained within two years of the clinic visit date. Exclusion criteria included alternative pathology contributing to gait disturbance, chronic non-ambulators, prior cervical surgery, and inadequate quality of MRI precluding accurate assessment of cervical stenosis. Data collected included patient demographics, medical history, subjective symptoms, and physical examination. Our primary and secondary outcomes were the correlations between performance on tandem gait and: 1) signs and symptoms of myelopathy, 2) Nurick and JOA classifications, and 3) severity of cervical stenosis on MRI. All patients completed a standardized 10-step tandem gait assessment, which was performed twice. To complete the assessment, patients ambulated 10 steps in a heel-to-toe fashion. The number of steps completed before the first misstep was counted. The test with the higher score was collected and graded using a 0-4 scale (see Figure). Severity of cervical stenosis on MRI was graded using the Kang scale. Results: Performance on tandem gait was inversely correlated with age (p < 0.0001), history of falls (p = 0.0059), balance/gait disturbance (p < .0001), and hyperreflexia (p = .0278). Performance on tandem gait otherwise did not correlate with other signs or symptoms of DCM. Performance on tandem gait was correlated with JOA and inversely correlated with Nurick classifications (p < 0.0001 and p = .0009, respectively). However, it was not correlated with severity of cervical stenosis. Conclusion: In this study, a standardized 10-step tandem gait test was correlated with Nurick and JOA, validating its utility as a clinical screening tool. Tandem gait was correlated with history of falls, balance/gait disturbance, and hyperreflexia, but was not correlated with signs or symptoms of DCM or degree of radiographic stenosis. Notably, patient age may be a confounding factor when utilizing tandem gait in the assessment of DCM. The utility of tandem gait in predicting the presence of DCM remains controversial. Although a useful objective screening tool for DCM, definitive diagnosis continues to require a comprehensive assessment of history, physical examination, and imaging studies.

ID: AGS26/637

A241: A retrospective comparative study of unilateral biportal endoscopy versus anterior cervical decompression and fusion for cervical spondylotic radiculopathy: an analysis of 166 cases

Hu Dong 1

1 Beijing Tsinghua Changgung Hospital, Beijing, China

Introduction: Surgical treatment for cervical spondylotic radiculopathy (CSR) focuses on nerve root decompression. Anterior cervical decompression and fusion (ACDF) remains the gold standard, while unilateral biportal endoscopic (UBE) surgery is an emerging minimally invasive technique whose efficacy requires further evidence. This study retrospectively compares perioperative indicators and early clinical outcomes between UBE and ACDF for CSR to inform clinical decision-making. Material and Methods: Data from patients who underwent surgery for CSR between January 2023 and August 2024 were retrospectively analyzed. According to the surgical technique, patients were divided into a UBE group (n = 72) and an ACDF group (n = 94). All patients were followed for at least 12 months. Baseline characteristics, perioperative parameters, and surgical outcomes were compared. Total cervical range of motion (C2-C7 ROM) was measured on dynamic radiographs at 3 and 12 months postoperatively. Complications were recorded. Results: No significant differences were observed in baseline data between the two groups (p > 0.05). The UBE group showed significantly less intraoperative blood loss (45.2 ± 15.8 mL vs. 58.5 ± 20.1 mL, p < 0.01) and shorter hospital stay (4.2 ± 1.5 days vs. 7.8 ± 2.2 days, p < 0.01) than the ACDF group. Operation time was slightly longer in the UBE group (95.6 ± 25.4 min vs. 88.3 ± 20.7 min, p < 0.05). At 12 months postoperatively, both groups showed significant improvements in VAS and JOA scores (p < 0.01), with no significant differences in final JOA scores (UBE: 14.8 ± 1.2 vs. ACDF: 15.0 ± 1.1) or JOA recovery rates (UBE: 85.3% ± 6.5% vs. ACDF: 86.1% ± 5.9%) (p > 0.05). Cervical ROM was significantly greater in the UBE group than in the ACDF group at both 3 months (39.5° ± 6.2° vs. 34.8° ± 5.7°, p < 0.01) and 12 months (40.2° ± 5.9° vs. 35.1° ± 5.5°, p < 0.01). Complications in the ACDF group included 2 cases of dysphagia and 1 case of symptomatic cage subsidence; the UBE group had 1 minor intraoperative dural tear (resolved with conservative treatment) and no severe neurovascular injuries. Conclusion: Both UBE and ACDF provide excellent pain relief and neurological recovery. UBE demonstrates advantages including less bleeding, shorter hospitalization, and effective preservation of cervical motion. Long-term outcomes and adjacent segment degeneration require further follow-up.

ID: AGS26/692

A242: Safety and effectiveness of anterior corpectomy surgery for cervical myelopathy due to the ossification of posterior longitudinal ligaments

Soobin Im 1 , Je Hoon Jeong 1 , Jeng Hoon Kang 1

1 Soonchuhyang University Bucheon Hospital, Bucheon, South Korea

Purpose: Anterior decompression and reconstruction is one of the standard surgical treatments for cervical OPLL. “Floating the ossified PLL” is the traditional method, which is described in the textbook for the anterior decompression. However, the detailed method was not described and we encounter the postoperative neurologic aggravation despite using the “floating method” with maximum attention during the decompression. This disastrous cord injury happens in especially severe OPLL cases. The “floating” method can not radically eliminate the pressure on the cord during decompression. Another fact is that even if we fully decompress the surrounding structure of the main compressing lesion, the ossified PLL will not easily float because of its thickness. This presentation will describe the typical pattern of ossified PLL and the detailed decompression method of ossified PLL. The presentation also introduces the technical tips for the more safe method of lifting PLL up method for anterior OPLL surgery. Materials and Methods: Between 2003 and 2024, 245 patients with cervical OPLL underwent anterior corpectomy and reconstruction using these technical tips. Drill the posterior part of the vertebrae until the soft part of the ventral cord is exposed. Usually the main compressing lesion forms a cruciate shape at the ventral surface of the dura. The most compressive part is usually located at the level of the bony endplate and disc level. Technical Tip: To prevent postoperative neurological aggravation, the pressure on the ossified PLL should be eliminated during decompression with a drill. After exposure of the ossified PLL by drilling the posterior part of the vertebra, drilling on the ossified PLL of the critical area will put pressure on the cord, which is very prone to being damaged. In order to eliminate the pressure on the cord, drilling of the PLL should be done by lifting it up and separating it from the cord. The bent sonopet tip can be used for undermining the PLL. The sharp tip of the hook will be able to hook up the hard edge of the main compressing lesion and detach from the cord. If we elevate the edge of OPLL, the CSF spreads behind OPLL. This makes safe drilling and can avoid disastrous cord damage. As the operator needs both hands to hold the drill and arachnoid knife, suction or irrigation should be done with assistance. Results: Among the 245 corpectomy patients with OPLL, one patient aggravated neurologically due to the fault of implant installation, not by this lifting ossified PLL up method just after the operation. 24 cases (10%) of overt CSF leakage on the operative field were managed by sealant application or direct suture. If CSF came out through the operative wound through the skin, the lumbar drain was inserted postoperatively. Lumbar drain is maintained for 5∼7 days. If the skin integrity is well maintained, we don’t insert a lumbar drain even if the CSF collection is detected in the subcutaneous tissue. Conclusion: To prevent neurological aggravation in OPLL surgery, we have to understand the detailed structure of ossified PLL. Anterior corpecomy provides a wide and clear visualization. Combining with the lifting PLL method, anterior cervical corpectomy is an effective and safe tool for the myelopathy due to cervical OPLL.

ID: AGS26/973

A243: Adverse impact of frailty on postoperative outcomes of cervical laminoplasty

Ataollah Shahbandi 1 , Pegah Ghamasaee 1 , Abdul Mounnem Yassin Kassab 1 , Saman Shabani 1

1 Department of Neurosurgery, Medical College of Wisconsin, Milwaukee, United States

Introduction: As the global population ages, frailty is increasingly acknowledged as a crucial factor in predicting health outcomes in the elderly population. Although prior studies have examined the impact of frailty on surgeries for degenerative cervical spine conditions, these studies often involved heterogeneous sample populations regarding surgical approaches. Resultantly, there is a lack of large database studies focusing on cervical laminoplasty. This study aimed to examine the impact of frailty and its severity on 30-day postoperative outcomes after cervical laminoplasty. Material and Methods: This study examined the data from the 2007-2022 American College of Surgeons National Surgical Quality Improvement Program. Patients who underwent cervical laminoplasty were queried and grouped according to the modified Frailty Index (mFI-5) into four categories: non-frail, pre-frail, frail, and severely frail. The primary outcome was the occurrence of at least one complication within 30 days, excluding blood transfusions. Secondary outcomes were major and minor complication rates, readmission rates, and the length of hospital stay. Results: A total of 2,250 patients were included, categorized as 829 non-frail, 873 pre-frail, 474 frail, and 74 severely frail patients. Frail patients (7.8%, p = 0.049) and severely frail patients (24.3%, p < 0.001) experienced at least one complication (excluding blood transfusions) significantly more frequently compared to non-frail patients (4.9%). Severely frail patients experienced major complications (18.9% versus 2.6%, p < 0.001) and minor complications (14.9% versus 4.5%, p = 0.001) at significantly higher rates than those observed in non-frail patients. Severely frail patients also showed a significantly increased occurrence of pneumonia (4.1% versus 0.5%, p = 0.014), urinary tract infections (5.4% versus 0.8%, p = 0.009), postoperative renal insufficiency (1.4% versus 0%, p = 0.031), postoperative dialysis (2.7% versus 0%, p = 0.007), vein thrombosis requiring therapy (4.1% versus 0.4%, p = 0.009), and cardiac arrest requiring cardiopulmonary resuscitation (2.7% versus 0.2%, p = 0.036) compared to non-frail patients. Furthermore, frail patients (5.8%, p = 0.039) and those who were severely frail (12.3%, p = 0.002) had significantly higher readmission rates compared to non-frail patients (3.1%). The presence of frailty (23.7%, p < 0.001) and severe frailty (46.6%, p < 0.001) were associated with a significantly increased risk of extended hospitalization compared to non-frail patients (11.2%). After controlling for confounders, frailty (odds ratio [OR] 1.661, 95% confidence interval [CI] 1.122-2.463; P = 0.011) and severe frailty (OR 4.328, 95% CI 2.271-8.253; p < 0.001) were independently associated with prolonged hospitalization. Additionally, severe frailty was independently correlated with a heightened risk of readmission (OR 3.976, 95% CI 1.392-10.959; p = 0.008), increased overall complication rates (OR 2.831, 95% CI 1.282-6.12; p = 0.009), and a higher likelihood of experiencing at least one major complication (OR 4.879, 95% CI 1.947-12.229; p = 0.001). Conclusion: Frailty and its severity may be associated with a higher rate of unfavorable perioperative outcomes after cervical laminoplasty. As a result, implementing preoperative strategies such as strict serum glucose control and diligent management of other existing comorbidities might help lower surgical risks and associated costs.

OP28: Management of Osteoporotic Fractures

ID: AGS26/112

A244: More than a spinal injury: the clinical significance of sternal fractures in thoracic spine trauma - A retrospective analysis of 4th column fractures of the spine

Josina Straub 1 , Melanie Ardelt 2 , Jan Reinhard 3 , Fabian Katzhammer 1 , Addisu Mesfin 4 , Jonas Krueckel 1 , Volker Alt 1 , Daniel Popp 1 , Siegmund Lang 1

1 University Hospital Regensburg, Department for trauma surgery, Regensburg, Germany

2 Division of Orthopaedics and Traumatology, University Hospital Krems, Krems, Austria

3 Clinic of Orthopaedics, University Hospital Regensburg, Regensburg, Germany

4 Strong Memorial Hospital, University of Rochester Orthopaedics, , Rochester, United States

Introduction: The thoracic spine and sternum are pivotal for the thoracic stability. High-energy trauma can compromise both, resulting in complex injury patterns. However, the prevalence and associated factors of concurrent thoracic spinal and sternal fractures remain poorly defined. This study aimed to assess the prevalence, the associated factors and risk factors for increased mortality of combined thoracic spine and sternal fractures. Material and Methods: Patients admitted to our level 1 trauma centre between 2006 and 2022 with thoracic vertebral fractures were retrospectively analysed and stratified into two cohorts based on the presence or absence of a concurrent sternal fracture. Results: Overall, N = 366 patients were included: N = 311 with isolated thoracic vertebral fracture and N = 55 with concomitant sternal fracture (4 th column fractures). In the total cohort, 76.2% of patients were male, with a mean age of 48.7 years and a mean Injury Severity Score of 32.6. In the group with combined sternal fracture, 72.7% of spinal fractures were classified as AO Spine Type A, 5.5% as AO Spine Type B and 21.8% as AO Spine Type C. The overall in-hospital mortality was 16.7%, with 18.0% of deaths occurring in patients with sternal fracture (OR = 1.24). Spinal cord injury was more frequently observed in patients with sternal fracture (24% vs. 14.4%). Intubation duration was significantly longer in patients with sternal fracture (10.7 ± 11.7 vs. 7.2 ± 9.9 days; p = 0.018), while hospital length of stay did not differ significantly (22.6±16.3 with sternal fracture vs. 20.9 ± 18.3 without; p = 0.526). Overall, 129 (35.2%) patients required surgical stabilization of thoracic vertebral fracture, including 27 (49.0%) with a combined sternal fracture. Upper thoracic spine fracture (T1-7, p = 0.018), AO Spine Type C fractures (p = 0.003), thoracoabdominal injuries (p = 0.001), additional surgeries performed by other specialties (p = 0.011), initial surgical stabilization as part of damage control management (p = 0.012) and dorsal instrumentation (p =0.020) were significantly more common in patients with concurrent sternal fractures, with more segments being fused (2.2 ± 2.8 vs. 1.3 ± 2.2; p = 0.008). Mortality was significantly higher in patients with thoracoabdominal injuries (p = 0.044) and those older than 65 years (p < 0.001). Conclusion: Concomitant sternal fractures in case of thoracic spine trauma are markers of high-energy mechanisms and are significantly associated with luxation fractures and longer intubation times. Frequent coexistence of thoracoabdominal injuries underscores the complexity of these trauma cases, contributing to increased morbidity and mortality, particularly in elderly patients. These findings highlight the need for heightened clinical vigilance and a multidisciplinary approach in managing patients with combined thoracic spine and sternal injuries.

ID: AGS26/1977

A245: Spinopelvic dissociation in polytrauma: challenges, complications and the role of triangular fixation

Alicia Collado Gosalvez 1 , Rafael Luque 2 , Eduardo Gómez 1 , Jose Luis Perez Gonzalez 1 , Pedro Lalueza 1 , Ignacio Dominguez 1

1 Hospital Clínico San Carlos, Madrid, Spain

2 Hospital Clinico San Carlos, Madrid, Spain

Introduction: Spinopelvic dissociations (SD) are high-energy injuries with unstable fracture patterns that invariably require surgical fixation. These lesions are associated with severe morbidity and often present with neurological impairment. They represent a major challenge for trauma centers in terms of surgical decision-making and postoperative management. The aim of this study was to analyze our institutional experience in the surgical treatment of SD and to discuss the outcomes of individualized fixation strategies aimed at restoring mechanical stability. Material and Methods: We conducted a retrospective review of polytrauma patients with SD admitted to a Level I trauma center between 2020 and 2024, with a minimum follow-up of 12 months. Data collected included demographics, mechanism of injury, associated lesions, neurological status, time to surgery, ICU and hospital stay, and complications. Fractures were classified according to Roy-Camille and AO/OTA systems. Surgical variables included number of vertebrae instrumented, use of iliac or transiliac-transsacral screws (TTS), decompression, and open versus percutaneous approach. Follow-up data included time to weight-bearing, fracture consolidation, neurological sequelae, and hardware removal. Results: Ten patients were included (median age 38.5). The predominant mechanism was fall from height (9/10). Associated injuries included pneumothorax (40%), intra-abdominal hematoma (1), traumatic brain injury (3), lower-limb fractures (70%), upper-limb fractures (3), and spinal fractures (4). Fracture classification revealed Roy-Camille type II (4) and III (3). AO type C3 was seen in 6 patients and type B3 in 4. Neurological deficit was presented in 70%, all involving sacral root compromise. The median number of instrumented vertebrae was 3; patients with associated spinal fractures required longer constructs (6-10 levels). Triangular fixation with TTS was achieved in 4 patients; lumbopelvic fixation with iliac or S2AI screws was used in 90%. A fully percutaneous approach was feasible in 3 cases, while decompression was performed in 4 patients. Mean ICU stay was 10.9 days, with a mean total hospital stay of 76 days. The most frequent postoperative complication was surgical site infection (50%). No correlation was found between infection rate and open surgery. Weight-bearing was initiated at a mean of 52 days, delayed in patients with lower-limb fractures. Radiographic consolidation was achieved in 7 cases. Hardware removal was performed in 2 patients, both at 20 weeks postoperatively. At final follow-up, outcomes were heterogeneous. 4 patients achieved full recovery. Three patients remained unable to return to work due to pain (VAS 3-4) or sphincter dysfunction. Overall, urinary incontinence persisted in 4 patients, and moderate-to-severe disability was observed in 5. Conclusion: The surgical management of SD remains complex due to their heterogeneous fracture patterns and frequent association with pelvic injuries. Early restoration of vertical and horizontal stability is essential, with triangular fixation representing the most standardized approach. However, achieving triangular constructs is not always feasible, as shown in our cohort (4/10), due to anatomical and technical limitations. These patients present high morbidity, consume considerable resources, and experience frequent complications. Early and appropriate fixation may help reduce these burdens. Development of standardized treatment algorithms is warranted to guide surgical decision-making, while maintaining flexibility for individualized strategies.

ID: AGS26/2038

A246: Management of AO Classification Type B2 thoracolumbar fractures: results of classic open arthrodesis versus percutaneous posterior fixation

Juan José Zamorano 1 , José Gengler Prieto 1 , Esteban Peña 1 , Nicolas Wolff hoepke 1 , Ignacio Cirillo 1 , Iris Castillo 1 , Marcos Gimbernat 1 , Ignacio Farías 1 , Carlos Tapia 1

1 Hospital del Trabajador, Santiago, Chile

Introduction: Thoracolumbar fractures represent one of the most common spinal injuries, with the majority located at the thoracolumbar junction (T10-L2). While the therapeutic consensus for patients with neurological deficits is decompression and instrumented fixation, the management of fractures without neurological compromise remains a topic of debate. Among these, flexion-distraction fractures (FDI), classified by AO as type B, account for up to 15% of cases and are considered intrinsically unstable. In particular, B2 injuries, which are predominantly ligamentous, exhibit a poor potential for spontaneous consolidation, historically motivating the recommendation for instrumented arthrodesis. In recent decades, percutaneous fixation has emerged as an attractive alternative to open surgery, offering advantages such as reduced tissue dissection, decreased intraoperative bleeding, less postoperative pain, and lower infection rates. This study presents a change in surgical technique for patients with AO classification type B2 thoracolumbar fractures, showing the results of patients treated with percutaneous posterior fixation compared to those operated on using classic posterior arthrodesis in the same workers' compensation hospital. Material and Methods: A clinical record and imaging analysis were performed on 40 patients with AO classification type B2 thoracolumbar fractures treated in a workers' compensation hospital between 2023 and 2025, with 20 patients treated using classic open posterior arthrodesis and 20 patients treated with percutaneous posterior technique. Results were analyzed for operative time, intraoperative bleeding, return to work, and reoperation rates with a follow-up of 9 months. Results: A total of 40 cases of AO classification type B2 thoracolumbar fractures were recorded, with 20 managed via percutaneous posterior approach (Group A) and 20 cases via open posterior arthrodesis (Group B). The average intraoperative bleeding was 46.75 ml in Group A and 234 ml in Group B. The average operative time was 121 minutes in Group A and 166 minutes in Group B. The number of reoperations in Group A was 4 patients, while in Group B it was 8 patients. Return to work was limited in both groups due to comorbid conditions present at the time of the accident. Conclusion: The treatment via percutaneous posterior approach presents shorter surgical times, less intraoperative bleeding, and lower reoperation rates, yielding favorable results for patients. No significant differences were demonstrated regarding return to work, given the high variability due to concomitant injuries at the time of the accident. The percutaneous posterior approach is an optimal therapeutic option, with favorable outcomes compared to the classic open posterior arthrodesis technique. This study aims to provide information for future research regarding the surgical management of this type of fracture.

ID: AGS26/2165

A247: Can we all use the OF score? A validation study in a Chilean cohort

Oscar Bravo 1 , Lucas Rifo 1 , Joaquín Cortés 1 , Dusan Ancic 1 , Vicente Jimenez 2 , Alvaro Silva González 1,2 , Andres Lisoni 1,2

1 Hospital Clínico de la Fuerza Aérea de Chile, Santiago, Chile

2 Clinica Alemana de Santiago, Santiago, Chile

Introduction: Osteoporotic vertebral fractures (OVF) are a major cause of morbidity, mortality, and reduced quality of life with rising incidence due to population aging. In Chile, this represents a growing clinical and public health challenge. Despite their impact, there are no validated predictive tools for treatment selection in the Chilean population. The Osteoporotic Fracture Score (OFS) has been proposed to guide management, but its external validity remains uncertain. Our objective is to assess the utility of the OFS in predicting therapeutic strategies, particularly its ability to identify patients requiring surgical management. Material and Methods: We performed a single-center, retrospective observational study approved by the institutional ethics board. Patients > 50 years with OVF between 2021-2025 were included. The follow up period was 3 to 60 months, with a mean of 13.9 months. They were classified according to treatment strategy (conservative vs surgical) and OFS. Univariate logistic regression was used to evaluate associations with surgical treatment. The discriminative ability of the OFS and a modified version was tested using receiver operating characteristic (ROC) curves, with calculation of area under the curve (AUC), sensitivity, specificity, and accuracy. Results: Eighty-four patients were included. Most (86.9%) received conservative management, while 13.1% underwent surgery. Pain relief rates exceeded 90% according to the Visual Analog Scale (VAS) in the conservative treatment group, despite their high OFS. Both OFS and modified OFS were significantly associated with surgical treatment (OR 1.96; 95% CI: 1.39-3.12, and OR 2.05; 95% CI: 1.42-3.43; both p < 0.001). ROC analysis showed good discriminative capacity: for the OFS, AUC was 0.87 with cut-off > 6.5 yielding sensitivity 72.7%, specificity 83.6%, and accuracy 82.1%. For the modified OFS, AUC was 0.88, with cut-off > 5.5 yielding sensitivity 90.9%, specificity 71.2%, and accuracy 73.8%. Conclusion: Findings show a strong institutional preference for conservative treatment, even in patients with high OFS values suggesting surgery. Despite this, pain relief was good to excellent, supporting the effectiveness of conservative approaches in selected patients. While the OFS demonstrated value in stratifying patients, it failed to reflect the complexity of clinical decision-making in our setting. Important variables such as frailty, comorbidities, functional demands, bone quality, and patient preferences likely influence outcomes and should be incorporated into future predictive models. Adjusting cut-off points or integrating clinical modifiers may improve the tool’s applicability across diverse populations. In this Chilean cohort, the OFS algorithm did not demonstrate sufficient reliability as a stand-alone tool for identifying patients requiring surgical management of OVF. Refinement of predictive systems through incorporation of clinical and patient-centered variables is needed to optimize individualized treatment strategies.

ID: AGS26/241

A248: Clinical comparation of two surgical methods in the treatment of re-fracture of the cemented vertebrae: re-percutaneous vertebroplasty or posterior fixation?

Chaoyuan Ge 1 , Wenlong Yang 1 , Zhengwei Xu 1

1 Xi'an Honghui Hospital, Xi'an, China

Introduction: Percutaneous vertebral augmentation has been widely used for osteoporotic vertebral compression fracture (OVCF) in the elderly. However, re-fracture or re-collapse of the cemented vertebrae after initial surgery are common. Currently, surgical treatment for re-fracture of the cement vertebrae mainly includes re-percutaneous vertebroplasty (PVP) or posterior open reduction, fixation and fusion using cement augmented screws (CAS), and which surgery is better is still controversial. This study aims to compare the efficacy of the two surgical methods in treatment of cement vertebral re-fracture after percutaneous vertebral augmentation. Material and Methods: The clinical data of patients with cemented vertebral refracture after percutaneous vertebral augmentation treated with re-PVP or CAS in our hospital from January 2020 to January 2024 were retrospectively analyzed. A total of 50 patients with OVCF were included in the study, including 20 males and 30 females, aged 68-82 years [(72.6 ± 5.8) years]. All patients presented with low back pain and varying degrees of local kyphosis. The distribution of cemented vertebrae: T10 in 1 case, T11 in 10 cases, T12 in 18 cases, L1 in 20 cases, and L2 in 1 case. According to the different surgical methods, they were divided into the re-PVP group (n = 28 cases) under local anesthesia and the CAS group (n = 22 cases) under general anesthesia. The clinical efficacy indicators including the visual analogue scale (VAS) and Oswestry disability index (ODI) before surgery, 6 days and 6 months after surgery and at the last follow-up were recorded and compared between the two groups. The imaging indicators including the vertebral body index (VBI), the kyphosis angle (KA) of the injured vertebra and the local kyphosis Cobb angle were also recorded. The intraoperative complications and the occurrence of new vertebral fractures during the follow-up period were recorded between the two groups. Results: There was no statistically significant difference in the preoperative general information between the two groups (p > 0.05). All patients successfully completed the operation. There were 2 cases of asymptomatic bone cement leakage in the re-PVP group and 1 case of asymptomatic bone cement leakage in the CAS group. No serious complications such as nerve or blood vessel injury occurred in both groups. The clinical efficacy indicators and imaging indicators of the two groups after surgery and during the follow-up period were significantly improved compared with those before surgery (p < 0.05). In terms of clinical efficacy indicators, the VAS scores of the patients in the re-PVP group improved better than those in the CAS group 6 days after surgery (p < 0.05), but there was no statistically significant difference at 6 months after surgery and at the last follow-up (p > 0.05); There was no statistically significant difference in ODI after surgery and during the follow-up period; in terms of imaging indicators, both the re-PVP group and the CAS group could significantly improve the height of the injured vertebra and correct local kyphosis. There was no statistically significant difference in VBI, KA and local Cobb of kyphosis between the two groups after surgery and during the follow-up period (p > 0.05). During the follow-up period, no adjacent vertebral body or other vertebral body fractures occurred in both groups. Conclusion: For OVCF patients with bone cemented vertebral refracture after percutaneous vertebral augmentation, re-PVP or posterior open reduction, fixation and fusion using CAS can effectively relieve low back pain, improve quality of life, effectively restore the height of the injured vertebra and correct local kyphosis. Compared with CAS, re-PVP has better analgesic effect in the early stage (1 week after surgery) due to its minimally invasive characteristics, but there is no difference in analgesic effect between the two groups in the later stage.

ID: AGS26/2778

A249: Surgical management of osteoporotic vertebral fractures: current techniques, indications, and knowledge gaps

Martin Holas 1 , Gaston Camino-Willhuber 2 , Julian Scherer 3,4 , Vishal Kumar 5 , Said Sadiqi 6 , Masahiko Takahata 7 , Emiliano Vialle 8 , Ratko Yurac 9 , Eugen Cezar Popescu 10 , Matthias Pumberger 11 , Sebastian Bigdon 12 , Mohammad El-Sharkawi 13 , Richard Bransford 14 , Andrei Joaquim 15 , Klaus Schnake 16,17 , Ulrich Spiegl 18 , Gregory Schroeder 19

1 Department of Orthopaedics and Trauma Surgery, Slovak Medical University, F.D. Roosevelt University Hospital, Banská Bystrica, Slovakia

2 Department of Orthopedics, Policlinica Gipuzkoa, San Sebastian, Spain

3 Orthopaedic Research Unit, University of Cape Town, Cape Town, South Africa

4 Department of Traumatology, University Hospital of Zurich, Zürich, Switzerland, Zürich, Switzerland

5 PGIMER, Chandigarh, India

6 Department of Orthopaedics, University Medical Center Utrecht,, Utrecht, Netherlands

7 Department of Orthopaedic Surgery Dokkyo Medical University, Japan, Dokkyo, Japan

8 Spine Surgery Group, Department of Orthopaedics, Cajuru University Hospital, Catholic University of Parana, Curitaba, Brazil, Curitaba, Brazil

9 University del Desarrollo, Clinica Alemana de Santiago, Chile, Santiago, Chile

10 Emergency Hospital, Iasi, Romania, Iasi, Romania ,

11 Center for Musculoskeletal Surgery, Charité - Universitätsmedizin Berlin, Germany, Berlin, Germany

12 Department of Orthopaedic Surgery and Traumatology, Inselspital, University Hospital, University of Bern, Bern, Switzerland, Bern, Switzerland

13 Department of Orthopaedic and Trauma Surgery, Faculty of Medicine, Assiut University, Assiut, Egypt, Assiut, Egypt

14 Department of Orthopaedics and Sports Medicine, Harborview Medical Center, University of Washington, Seattle, WA, USA, Seattle, United States , 15 Neurosurgery Division, Department of Neurology, State University of Campinas, Campinas-Sao Paulo, Brazil, Sao Paulo, Brazil

16 Center for Spinal and Scoliosis Surgery, Malteser Waldkrankenhaus St. Marien Erlangen, Erlangen, Germany, Erlangen, Germany

17 Department of Orthopedics and Traumatology, Paracelsus Private Medical University Nuremberg, Nuremberg, Germany, Nuremberg, Germany

18 Unfallchirurgie und Orthopädie, München Klinik Harlaching, Germany, München, Germany

19 Department of Orthopaedic Surgery, Rothman Institute, Thomas Jefferson University Hospital, Philadelphia, PA, USA, Philadelphia, United States

Introduction: Osteoporotic vertebral fractures (OVF) represent the most frequent fragility fractures worldwide and pose major clinical and socioeconomic challenges. Their management differs substantially from traumatic fractures in younger patients due to poor bone quality, higher comorbidity burden, and lack of universally accepted treatment guidelines. Recent developments, including the AO Spine-DGOU Osteoporotic Fracture (OF) classification and OF score, have provided a framework for decision-making. This review summarizes current surgical techniques for thoracic and lumbar OVF, highlights their indications and limitations, and outlines key knowledge gaps. Materials and Methods: A narrative review was performed through a non-systematic search of PubMed, Scopus, and Google Scholar, complemented by manual screening of reference lists. Both prospective and retrospective studies focusing on surgical techniques for OVF were included, while case reports and cadaveric studies were excluded. Selection was guided by critical appraisal and author expertise. Results: Minimally invasive cement augmentation techniques, such as vertebroplasty and kyphoplasty, are recommended for less severe OVF (OF-1 to OF-3), with kyphoplasty showing superior vertebral height restoration and reduced cement leakage. Short segmental fixation (SSF) is suitable for moderately unstable fractures but carries a higher risk of implant failure unless combined with screw or vertebral augmentation. Long segmental fixation (LSF) offers greater stability in severe cases (OF-4), particularly in the mid-thoracic spine, but increases surgical invasiveness and the risk of adjacent segment disease. Pedicle screw augmentation significantly enhances implant stability in osteoporotic bone. Hybrid stabilization, combining posterior instrumentation with vertebral augmentation, provides promising mid-term results, reducing implant failure with less invasiveness than combined anterior-posterior (AP) fusion. AP surgery remains the most robust option for highly unstable or collapsed fractures with neurological deficits or rigid kyphosis but is often limited by surgical morbidity in elderly or frail patients. Recent minimally invasive AP approaches show potential to expand indications. Conclusion: Surgical management of OVF requires tailored decision-making that balances mechanical stability, invasiveness, and patient frailty. Cement augmentation is appropriate for less severe fractures, while SSF and LSF are reserved for unstable patterns depending on severity and location. Screw augmentation improves fixation strength, and hybrid stabilization represents a less invasive alternative to combined AP surgery with comparable outcomes. Combined AP reconstruction remains the gold standard for highly unstable fractures but should be reserved for select patients due to its invasiveness. Despite recent advances, evidence gaps remain regarding optimal management of OF-4 fractures, the role of prophylactic augmentation, and the necessity of sagittal balance restoration in elderly patients. Future research should aim to refine treatment algorithms and align global practice standards.

ID: AGS26/322

A250: A new classification of Kummell disease based on imaging features and clinical manifestations of 168 cases

Chaoyuan Ge 1 , Wenlong Yang 1 , Zhengwei Xu 1

1 Xi'an Honghui Hospital, Xi'an, China

Introduction: The aim of this study is to propose a new system for classifying Kummell disease and evaluates its feasibility for guiding clinical treatment. Material and Methods: A retrospective study of 168 cases of Kummell disease with complete follow-up data was conducted at our institute from January 2019 to December 2022. Patients were divided into types I-V based on a combination of clinical manifestations and different morphological changes detected by imaging. Type I is characterized by bone non-union imaging; percutaneous vertebroplasty (PVP) is the primary treatment option for this basic imaging morphological change in Kummell disease. Type II is characterized by imaging findings indicating instability; posterior fixation and fusion is the primary treatment option. Type III is characterized by imaging findings indicating spinal stenosis, accompanied by symptoms of neurological damage; the primary treatment option is posterior decompression fixation and fusion. Type IV is characterized by imaging showing kyphosis. These patients exhibit malunion, often with a kyphotic angle ≥ 30°, and no change in kyphotic angle as assessed by dynamic radiography; the primary treatment option is posterior osteotomy, fixation, and fusion. Type V is characterized by a combination of 3 or 4 of the above imaging signs; the treatment provided is related to the specific types responsible. Treatment outcomes were evaluated based on postoperative visual analogue scale (VAS) and Oswestry Disability Index (ODI) scores. During the follow-up period, we observed whether the damaged vertebra exhibited compression or local secondary kyphosis and whether neurological function improved in patients accompanying with neurological symptoms. Results: All patient received average 38.4 ± 7.8 months of follow-up care. Patients of each type had statistically significant improvements in postoperative VAS and ODI scores compared with preoperative scores. The height of the damaged vertebra in all patients with Kummell disease and the local kyphotic angle of patients with kyphosis improved postoperatively. During the follow-up period, 19 patients (11.3%) exhibited compression of the damaged vertebra or local secondary kyphotic deformity. At the final follow-up of patients with neurological deficits, the American Spinal Injury Association (ASIA) grades of D in 3 patients and E in 12 patients were seen postoperatively, compared with preoperative ASIA grades of C in 12 patients and D in 3 patients. Conclusion: This study shows that this new classification of Kummell disease into types I-V can effectively guide clinical treatment.

ID: AGS26/74

A251: Compression rate progression of osteoporotic vertebral compression fracture in conservative treatment

Jaewan Soh 1 , Jae Chul Lee 2

1 Hanyang University Guri Hospital, College of Medicine,, Guri, South Korea

2 Soonchunhyang University Seoul Hospital, College of Medicine,, Department of Orthopedic Surgery, Seoul, South Korea

Introduction: When osteoporotic vertebral compression fractures (OVCF) occur, the purpose of treatment is to reduce pain and prevent progression of compression, and many results have been reported. In OVCF, vertebral body compression often progressed. We analyzed whether there were any factors affecting the progression of the compression rate and also investigated the period when the compression rate progressed the most, and analyzed the correlation with clinical outcomes. Material and Methods: One hundred twenty-two patients with OVCF, followed for more than six months, were included. Factors such as age, gender, fracture level, bone mineral density (BMD), body mass index (BMI), osteoporosis medication, initial compression rate, fracture location, and pelvic incidence were assessed. Compression rates were measured initially and at 1, 2, 3, and 6 months. The correlation between these factors and compression rate progression of more than 50% was analyzed using univariate and multivariable Cox regression analysis. Compression rate progression over time was evaluated using a repeated measures general linear model, and clinical outcomes were assessed via paired t-test. Results: Univariate analysis showed that an initial compression rate of less than 50% was significantly associated with progression (p = 0.001). Multivariable Cox regression confirmed this association (p = 0.013, Hazard ratio = 3.018, 95% Confidence Interval 1.265-7.200). The most significant compression rate progression occurred within the first month of follow-up (p = 0.038). At the final follow-up, both VAS for back pain (p = 0.009) and ODI (p = 0.026) were significantly higher in the compression rate progression group. Conclusion: An initial compression rate of less than 30% was associated with a higher correlation of compression rate progression, particularly within the first month after injury. Compression rate progression was correlated with poorer clinical outcomes in terms of pain and quality of life.

ID: AGS26/808

A252: Impact of paravertebral sarcopenia on long-term survival and functional recovery after percutaneous kyphoplasty in older adults with osteoporotic thoracolumbar compression fractures: a prospective dual-center cohort study

Wenjie Wu 1 , Xiaojing Si 1 , Junxiong Cao 1 , Boxing Wang 1 , Chuanqi Zou 1

1 The People's Hospital of Banan District, Chongqing, China

Introduction: Percutaneous kyphoplasty (PKP) is a widely used minimally invasive procedure for osteoporotic thoracolumbar compression fractures (OTLCF). While effective, a subset of elderly patients continues to experience poor functional recovery and elevated mortality. Sarcopenia, particularly of the paravertebral muscles-key stabilizers of the spine-may be a critical yet understudied predictor of long-term outcomes. This study investigates whether pre-operative paravertebral sarcopenia independently predicts 12-month all-cause mortality and functional recovery after PKP. Material and Methods: A prospective, non-randomized, observational cohort study was conducted across two centers in Chongqing, China. We enrolled 200 patients aged ≥ 65 years with acute or subacute single-level OTLCF (T10-L2) scheduled for PKP. Paravertebral sarcopenia was assessed preoperatively on routine T2-weighted MRI by measuring the cross-sectional area (CSA) of the erector spinae and multifidus muscles at the level above the fractured vertebra. Sex-specific thresholds were applied to define sarcopenia. Primary outcome was 12-month all-cause mortality. Secondary outcomes included pain intensity (VAS) and functional disability (ODI) assessed at pre-op, 1, 3, 6, and 12 months post-op. Statistical analyses included Kaplan-Meier survival curves, multivariable Cox regression, and linear mixed-effects models. Results: A total of 194 patients completed follow-up (mean age 74.2 ± 6.1 years; 78% female). Preoperative paravertebral sarcopenia was identified in 36.1% (70/194) of patients. The sarcopenia group showed significantly higher 12-month all-cause mortality (8.6% vs. 2.4%, p = 0.018) and a significantly increased risk of death (adjusted HR = 3.84, 95% CI: 1.21-12.18, p = 0.022) after controlling for age, sex, BMI, number of comorbidities, and baseline ODI. Functional recovery was also poorer in the sarcopenia group, with less improvement in ODI scores at 12 months (mean change: -15.3 ± 8.4 vs. -24.6 ± 9.1, p < 0.001) and higher VAS scores (2.8 ± 1.5 vs. 1.9 ± 1.2, p = 0.003). The incidence of adjacent vertebral fractures was also higher in the sarcopenia group (11.4% vs. 4.8%, p = 0.049). Conclusion: Preoperative paravertebral sarcopenia is an independent predictor of increased 12-month mortality and inferior functional recovery following PKP in older adults with OTLCF. Routine MRI-based assessment of paravertebral muscles should be integrated into preoperative risk stratification to identify high-risk patients who may benefit from targeted prehabilitation and intensified postoperative support.

Keywords: Percutaneous Kyphoplasty, Osteoporotic Fracture, Paravertebral Muscles, Sarcopenia, Mortality, Functional Outcomes, Elderly, Prospective Cohort

OP29: Complex Deformity Tactics

ID: AGS26/1184

A253: Low molecular weight heparin (LMWH) prophylaxis is safe in patients undergoing staged thoracolumbar fusion

Harsh Wadhwa 1 , Austin Stoner 1 , Jayme Koltsov 1 , James Mok 1 , John Kleimeyer 1 , Atman Desai 1 , Corinna Zygourakis 1 , Jon Park 1 , Kirkham Wood 1 , Anand Veeravagu 1 , John Ratliff 1 , Todd Alamin 1 , Serena Hu 1

1 Stanford University, Stanford, United States

Introduction: Longer thoracolumbar fusions sometimes include anterior approaches, which, due to retraction of the great vessels, can carry higher risk of deep vein thrombosis (DVT) and pulmonary embolism (PE). Low Molecular Weight Heparin (LMWH) prophylaxis may ameliorate these risks. Materials and Methods: Retrospective review of adults undergoing combined thoracolumbar fusion of ≥ 5 levels (single institution 2014-2024). Anticoagulation was driven by surgeon choice. Patients receiving LMWH versus no VTE prophylaxis were compared. Patient demographics, comorbidities, perioperative data (EBL, transfusions, etc.), and postoperative data (90-day complications, readmissions, reoperations) were collected. Analyses were performed separately for staged and unstaged patients, with differences with LMWH assessed via chi-squared and Fisher’s exact tests (α = 0.05). Results: Overall, 191 staged [131 (68.6%) female, 66.7 ± 9.2 years] and 216 unstaged [146 (67.6%) female, 67.7 ± 9.8 years] patients were included. LMWH was given in 133 (70%) of staged surgeries and 107 (50%) of unstaged surgeries. Patients receiving LMWH had higher BMI (29.1 [28.2, 30.1] vs. 27.0 [25.7, 28.2], p = 0.011). Among staged patients, rates of PE were 3.1% [95%CI 1.2, 7.8] with LMWH vs. 9.1% [4.0, 19.6] without LMWH (p = 0.132), and rates of DVT were 4.7% [2.2, 9.9] with LMWH vs. 5.5% [1.9, 14.9] without LMWH (p > 0.999). Among unstaged patients, rates of PE were 3.9% [1.5, 9.6] with LMWH vs. 3.1% [1.1, 8.8] without LMWH (p = 0.185) and rates of DVT were 2.9% [1.0, 8.2] with LMWH vs. 4.2% [1.6, 10.2] without LMWH (p = 0.338). In both staged and unstaged cohorts, transfusion rates (and amounts) were comparable with vs. without LMWH (Staged: 96.4% [87.7, 99.0] vs. 92.1% [86.1, 95.7], (p = 0.352); Unstaged: 73.8% [64.6, 81.3] vs. 76.0% [66.6, 83.5], p = 0.387). Complications, readmission, and 6 month reoperation were also comparable with vs. without LMWH. In the staged cohort, wound complication (9.4% [5.5, 15.8] vs. 3.6% [1.0, 12.3], (p = 0.234)) and deep or superficial infection rates (6.3% [3.2, 11.9] vs. 3.6% [1.0, 12.3], (p = 0.726)) were comparable with vs. without LMWH. In the unstaged cohort, wound complications (8.3% [4.3, 15.6] vs. 1.9% [0.5, 6.8], (p = 0.052)) were comparable, but deep or superficial infections (8.3% [4.3, 15.6] vs. 1.9% [0.5, 6.8], (p = 0.042)) were more common in the LMWH group. Conclusion: In the largest study to date directly comparing combined thoracolumbar fusion with vs. without LMWH prophylaxis, LMWH did not increase bleeding or transfusions. LMWH may provide benefits, particularly in PE reduction; however, infection rates were greater in unstaged surgeries, and wound complications may trend higher. Large population-based study is needed for confirmation.

ID: AGS26/1518

A254: Is surgery for AIS always successful? A Delphi consensus study including patient participation to define poor outcomes and subsequent identification of incidence in a large multicenter database

Raf Mens 1,2 , Pepijn Bisseling 1 , Marinus De Kleuver 2 , Peter Newton 3 , Michael Kelly 3 , Michelle Marks 4 , Miranda van Hoff 2

1 Sint Maartenskliniek, Nijmegen, The Netherlands

2 Radboudumc, Nijmegen, The Netherlands

3 Rady Children’s Hospital, San Diego, United States

4 Setting Scoliosis Straight Foundation, El Cajon, United States

Introduction: Surgery for adolescent idiopathic scoliosis (AIS) is considered successful for the majority of patients. To identify those patients who currently least benefit from surgery, a definition of treatment failure is needed. The aim of this study was to draft a definition of treatment failure for AIS-surgery, and identify its incidence in a large multicenter database. Material and Methods: A three-round modified Delhi-procedure with experts (physicians (n = 14) and physical therapists (n = 8)), preceded by a patient survey (n = 33). Consensus threshold was 70%. Incidence of failure according to the resulting definition was analyzed in the multicenter Harms study group database (HSGdb, N = 4974, mean follow-up: 3.6 yrs [0-23]). Results: Failure of AIS-surgery was defined as: 1) Clinical deterioration: - clinically relevant decrease in condition-specific quality-of-life compared to preoperative scores, - Permanent disability due to neurologic symptoms, - Mortality related to index surgery, and/or. 2) Revision surgery for following indications: - Deterioration of instrumented curve, - Deterioration of non-instrumented motion segments, - Overcorrection, - Pseudoarthrosis, and/or. 3) Radiologic deterioration: - Progression of instrumented curve > 5 0 , Incidence in HSGdb: - Clinical: i. Relevant decrease on SRS-22r*: 8.7% [428/4894], ii. Neurologic: 2/4974, iii. Mortality: 0, - Revision surgery: 1.3% [64/4974], - Radiologic deterioration: 17.2% [817/4751], * Minimal detectable measurement difference was taken as threshold for clinical relevancy. Conclusion: This is the first attempt to define ‘treatment failure’ of AIS-surgery. Utilizing patient-participation, expert consensus was reached that failure can be considered: Clinical deterioration and/or revision surgery and/or radiologic deterioration. With the presented Delphi-consensus definition a clinical- or revision failure-rate of 1.3%-8.7% was identified in a large multicenter database. A radiologic failure-rate of 17% does not seem in line with clinical experience. Clinical relevancy of this criterium needs further investigation. A solid, evidence-based and uniformly accepted definition can be used to uniformly evaluate unfavorable outcomes and compare them between different cohorts and registries, ultimately helping to improve outcomes of surgery.

ID: AGS26/1267

A255: False postoperative PI-LL mismatch due to altered hip coverage: implications onthe surgical outcome of adult spinal deformity

Li Jie Li Jie 1 , Zhen Liu 1

1 Nanjing Drum Tower Hospital, Nanjing, China

Introduction: Exploring the correlation between changes in hip joint coverage and postoperative PI in adult spinal deformity (ASD) patients after fixation with S2 alar iliac screws (S2AI), and whether there is a link between PI-LL mismatch and mechanical complications immediately after surgery and during follow-up, in order to provide reference for clinical treatment evaluation. Material and Methods: Retrospective analysis of clinical and imaging data of 91 ASD patients who underwent posterior S2AI screw fixation surgery in the Orthopedics Department of Nanjing Gulou Hospital from January 2019 to January 2021. According to the change in hip joint coverage (HJC) immediately after surgery, patients were divided into two groups: HJC-C group (with immediate postoperative Δ HJC at the upper half level, > Q2) and HJC-NC group (with immediate postoperative Δ HJC at the lower half level, < Q2). In the HJC-C group, further divided into the HJC-R group (follow-up period Δ HJC > Q2) and the HCJ-NR group (follow-up period Δ HJC < Q2) based on whether the hip joint coverage rate rebounded during the 2-year follow-up. During preoperative and postoperative first standing, as well as 2-year follow-up, the patient underwent standing position X-ray imaging of the entire spine. The following spinal parameters were measured and recorded: Cobb angle, proximal lumbar lordosis (PLL), distal lumbar lordosis (DLL), lumbar lordosis (LL), Lordosis distribution index (LDI), sagittal vertical axis (SVA), coronal balance distance (CBD), thoracic kyphosis (TK), and T1 pelvic angle (T1). Pelvic angle (T1PA), pelvic incidence angle (PI), pelvic tilt angle (PT), sacral slope (SS), PI-LL, and proximal junction angle (PJA); Related parameters of hip joint coverage include HJC, Lateral Center Edge angle (LCE angle), Acetabular index (AI), Sharp angle, and hip protrusion index (EI). Analyze and compare the differences in preoperative, postoperative, and final follow-up imaging indicators between two groups of patients. Results: In the overall cohort, ASD patients showed an increasing trend in HJC immediately after S2AI fixation surgery (2.2 ± 3.9%), accompanied by a rebound during follow-up (-2.1 ± 3.8%), and a positive correlation was observed between the changes in Δ PI and Δ HJC immediately after surgery and during follow-up (r = 0.587, p = 0.008). Although the immediate postoperative PI-LL of patients in the HJC-C group (n = 46) was significantly lower than that of patients in the HJC-NC group (n = 45) (-22.4 ± 10.8 vs. -10.3 ± 9.3, p = 0.023). However, at the last follow-up, the HJC-C group showed more rebound in PI, and there was no significant statistical difference in PI-LL between the two groups of patients at the last follow-up (-14.6 ± 11.2 vs. -9.7 ± 8.5, P = 0.135).And there was a significant difference in the incidence of PJK between the HJC-C group and the HJC-NC group at the last follow-up [26.1% (6/23) vs. 23.8% (5/21), p = 0.298]. Although there was no significant difference in immediate postoperative PI-LL between the HJC-R group and the HCJ-NR group (-21.1 ± 11.2 vs. 22.6 ± 10.1, p = 0.675). But at 2-year follow-up, the PI-LL of patients in the HJC-R group was significantly higher than that of patients in the HCJ-NR group (-8.3 ± 6.9 vs. -15.9 ± 13.2, p = 0.038). The incidence of PJK in the HJC-R group (n = 12) was significantly lower than that in the HCJ-NR group (n = 11) at the last follow-up [8.7% (2/23) vs. 43.5% (10/23), p = 0.008]. Conclusion: The changes in PI after S2AI surgery in ASD patients are closely related to changes in hip joint coverage. An increase in hip joint coverage after surgery can lead to immediate PI-LL mismatch, but during follow-up, hip joint coverage and PI rebound gradually match the hip pelvic spinal balance. However, for patients with poor hip joint coverage during follow-up, the persistent presence of PI-LL mismatch during follow-up can affect the maintenance of sagittal balance and lead to complications such as PJK.

ID: AGS26/1533

A256: Best in class prediction of PJK with a novel T1-T9 pelvic angle mismatch

Nuri Demirci 1 , Yasemin Yavuz 2 , Munish Gupta 1

1 Department of Orthopaedic Surgery, Washington University in St. Louis, Saint Louis, United States

2 Department of Biostatistics, Faculty of Medicine, Ankara University, Ankara, Turkey

Introduction: Proximal junctional kyphosis (PJK) is a frequent complication after adult spinal deformity (ASD) surgery. Traditional sagittal alignment parameters such as thoracic kyphosis (TK), PI-LL mismatch, and C7 sagittal vertical axis (SVA) have been widely studied, while more recent targets including the GAP score and T4L1 pelvic angle (T4L1PA) have also been proposed. However, their ability to independently predict PJK remains uncertain. We defined the T1-T9 pelvic angle (T1T9PA) mismatch as a novel sagittal alignment parameter to quantify proximal thoracic alignment and investigated whether it independently predicts PJK. Materials and Methods: A prospective cohort of 207 ASD patients who underwent thoracic-to-pelvis fusion with ≥ 7 instrumented levels and pelvic fixation, all with ≥ 2 years of follow-up was analyzed. PJK was defined as a ≥20° increase in proximal junctional angle from baseline or revision for a PJK. Predictors included demographic factors, surgical approach, UIV level, and radiographic parameters (PI-LL, lumbar lordosis (LL), TK, C7-SVA, GAP, T4L1PA, and T1T9PA). TK and LL were continuous; all others categorical: PI-LL (± 10°), SVA (< 5 vs ≥ 5 cm), GAP (proportioned vs disproportionate), T4L1PA (within vs outside -1° to +3°), T1T9PA (ROC-defined cutoff), UIV (upper ≤ T6 vs lower ≥ T7), and approach (posterior-only vs combined). Univariate logistic regression identified predictors; those with p < 0.25 entered a forward stepwise multivariate model. Results: Mean age was 62.1 ± 11.1 years, BMI 29.7 ± 5.7, and 77.8% were female; 72.9% had upper thoracic UIVs, 27.1% lower thoracic UIVs, PSO was performed in 20.8% and VCR in 6.3%. Over a mean 46-month follow-up, 32 patients (15.5%) developed PJK. ROC analysis identified 10° as the optimal cutoff for T1T9PA. On univariate analysis, TK was significant as a continuous variable (p = 0.011), T1T9PA ≥ 10° was strongly predictive (p < 0.001), and T4L1PA outside its literature-defined safe zone was predictive (p = 0.034) for PJK. Surgical approach (p = 0.057), UIV level (p = 0.060), and GAP score (p = 0.22) met the candidate threshold (p < 0.25). Other demographic and radiographic factors were not predictive. Although age, sex, and BMI were not below the p < 0.25 threshold, they were included in the multivariate model due to clinical relevance. The multivariate model included T1T9PA, T4L1PA, TK, surgical approach, UIV level, GAP, age, sex, and BMI. In forward stepwise regression, only T1T9PA ≥ 10° remained independently significant (OR 4.64, 95% CI 2.0-10.0, p < 0.001). Conclusion: Only T1T9PA was an independent risk factor in predicting PJK. No traditional TK, PI-LL mismatch, C7-SVA or recent GAP, T4L1PA parameters independently predicted PJK. A 10° threshold was established, with patients above this cutoff having a 4.6-fold increased risk of PJK (OR 4.64, 95% CI 2.0-10.0) and reduced PJK-free survival. These findings support T1T9PA as a novel thoracic alignment target to mitigate proximal junctional failure.

ID: AGS26/157

A257: Efficacy of halo pelvic traction versus halo gravity traction in treating severe rigid spinal deformity - A matched retrospective study

Jianqiang Wang 1 , Yong Hai 1

1 Beijing Chaoyang Hospital, Beijing, China

Introduction: Severe rigid spinal deformity (SRSD)-defined by > 100° Cobb angle and < 30% flexibility-causes restrictive pulmonary dysfunction, cardiopulmonary compromise, and reduced quality of life. Advanced techniques like vertebral column resection enable significant correction albeit with substantial neurologic risks. Preoperative halo traction serves as a crucial adjunct to reduce osteotomy dependency. Halo gravity traction (HGT) and halo pelvic traction (HPT) both allow gradual correction with functional improvements, yet comparative efficacy data remain scarce. This matched retrospective study evaluates HGT versus HPT in optimizing deformity correction, pulmonary/nutritional status, and surgical safety to inform evidence-based preoperative planning for SRSD. Material and Methods: We retrospectively reviewed 104 severe rigid spinal deformity patients treated with preoperative Halo traction and posterior correction surgery (2016.1-2022.12). Patients were matched 1:1 by main curve Cobb angle and age into HGT and HPT groups (33 participants in each group). Radiographic parameters (main curve, kyphosis, coronal/sagittal balance, apical vertebral translation) were assessed before and after traction, preoperatively, and at final follow-up. Surgical planning for osteotomy and fusion segments used pre-traction imaging, compared with actual outcomes. Traction duration, surgical parameters, and complications were recorded. Results: HGT patients had 4.5 ± 1.5 months of traction, improving main curve Cobb angle from 138.4° to 94.9° (23.5%) and kyphosis from 87.9° to 73.1° (22.6%), with postoperative values of 51.1° and 39.7°. HPT patients had 5.1 ± 1.1 months of traction, improving main curve from 143.2° to 69.6° (51.3%) and kyphosis from 117.7° to 56.8° (49.1%), with postoperative values of 56.1° and 34.5°. HPT showed significantly better improvement rates (p < 0.05). HPT achieved 50% deformity improvement after 3 months, vs. 5 months for HGT. By 7 months, > 60% of HPT vs. < 10% of HGT patients reached this threshold. Planned three-column osteotomy use dropped from 90.9% to 51.5% (actual 36.4%) in HGT and 84.8% to 21.2% (actual 9.1%) in HPT. HPT reduced osteotomy grade (p < 0.05) and fusion segments (p = 0.02) more effectively. Complication rates were similar. Conclusion: Halo gravity traction (HGT) and halo pelvic traction (HPT) effectively correct severe rigid spinal deformity and imbalance. Both reduce high-grade osteotomy needs compared to initial plans. HPT outperforms HGT in deformity correction, osteotomy grade reduction, and fusion segment minimization.

ID: AGS26/158

A258: Risk stratification and clinical classification for postoperative neurological complications in post-tuberculosis kyphosis: a retrospective cohort study

Jianqiang Wang 1 , Yong Hai 1

1 Beijing Chaoyang Hospital, Beijing, China

Introduction: Spinal tuberculosis, affecting over a billion people in developing countries, is a leading cause of severe kyphotic deformity. Untreated cases progress to rigid angular kyphosis, with significant risk of spinal cord compression, neurological deficits (occurring in 23%-76% of cases), and paraplegia (15%-42.5%). This results in chronic pain, spinal imbalance, and reduced quality of life. Conservative management often fails, necessitating surgery for neurological compromise or spinal instability. Modern techniques like pedicle screw instrumentation and posterior osteotomy improve deformity correction and fusion outcomes. However, surgical challenges include rigid deformities, structural abnormalities, and prolonged cord compression, increasing neurological complication risks (reported up to 30%). Overcorrection may cause cord stretching or ischemia-reperfusion injury, while abrupt decompression can induce neuronal apoptosis, potentially leading to paralysis. Current strategies lack robust preoperative risk identification. This study analyzes risk factors for postoperative neurological complications in tuberculous kyphosis patients and proposes a clinical classification system to identify high-risk individuals, guiding preoperative interventions to minimize surgical risks. Material and Methods: We retrospectively analyzed 51 patients with post-tuberculosis kyphosis who underwent single-stage posterior osteotomy and correction at our institution. Radiographic parameters, including the kyphotic angle, cross-sectional area ratio of the spinal cord (CSAR), and spinal cord angle (SCA), alongside surgical factors such as intraoperative blood loss and osteotomy grade, were evaluated. Postoperative neurological complications were recorded. Univariate and multivariate logistic regression analyses were employed to identify potential influencing factors. Based on the interaction between these factors, a clinical classification system for post-tuberculosis kyphosis was established to stratify the risk of postoperative neurological complications. Results: A minimum 2-year follow-up was conducted for the 51 patients with post-tuberculosis kyphosis. Multivariate logistic regression analysis revealed that the Baltalimani sign, spinal cord MRI type, Rajasekaran classification, three-column osteotomy, C-reactive protein (CRP), and SCA were significant risk factors for postoperative neurological complications. Two risk prediction models were developed accordingly. Synergistic interactions were identified between spinal cord MRI type and Rajasekaran classification, as well as between CRP and SCA. These findings informed the development of two clinical classification systems: the Mechanical Classification (AUC = 0.751) and the Imaging-Biomarker Classification (AUC = 0.883). The area under the curve (AUC) for both classifications demonstrated good predictive performance, with the DeLong test indicating superior efficacy for the Imaging-Biomarker Classification (p = 0.039). Conclusion: Spinal structural instability, spinal cord compression, osteotomy grade, and elevated CRP levels may all contribute to an increased risk of postoperative neurological complications in patients with post-tuberculosis kyphosis. The clinical classification systems established herein facilitate precise risk stratification, enabling targeted preoperative interventions to mitigate surgical risks and enhance surgical outcomes.

ID: AGS26/2127

A259: Impact of pelvic fusion on walking performance following scoliosis correction surgery

Sara McMahan 1 , Bethany Wilson 1 , Kyle Robinson 1 , Isador Lieberman 2

1 Texas Back Institute, Plano, TX, United States

2 Texas Back Institute, Frisco, TX, United States

Introduction: Scoliosis correction surgery is increasingly involving fusion to the pelvis in order to stabilize the construct, providing improved outcomes for appropriately selected patients with spinal deformity (ASD). For patients undergoing these procedures, it is important to consider how the patient’s day-to-day function will be impacted. Maintaining walking ability is vital for independent function. It is presently not thoroughly investigated how pelvic fusion in a corrective fusion construct affects patient walking performance. The purpose was to investigate how fusion to the pelvis affects function in patients with ASD. Material and Methods: This was a post-hoc analysis of a prospective study with 45 ASD patients receiving corrective alignment surgery without fusion to the pelvis (N) (35F/10M, 56.9 ± 19.1y, 27.3 ± 4.4 kg/m 2 , SVA: 28.7 ± 54.2 cm, PI-LL: 10.6 ± 20.1 o , Cobb: 36 ± 16.5 o ) and 35 receiving fusion to the pelvis (Y) (28F/7M, 65.9 ± 7.1y, 28.6 ± 6.5 kg/m 2 , SVA: 48.5 ± 47.8 cm, PI-LL: 24.3 ± 20.8 o , Cobb: 40.4 ± 16.5 o ) and 35 asymptomatic adult controls (C) (20F/15M, 39 ± 13.3y, 24.7 ± 3.6 kg/m 2 ). Patient reported outcome measures included Visual Analog scale for midback (VASmb) low back (VASlb), and leg (VASleg) pain, Oswestry Disability Index (ODI). Walking measures included Cadence, Single support time, double support time, step length, width, height, and time, and gait deviation index. ASD subjects were divided into two cohorts based on whether their pelvis was fused during their procedure (PF), or not (NPF). Evaluations were conducted preoperatively (P0), 3 months post-surgery (P3), and 1-year post-surgery (P12). At each visit patients were assessed with VAS and ODI and performed three over ground walking tests to assess spatiotemporal parameters of the gait cycle. Results: PF were significantly older with significantly larger PI-LL than NPF but were comparable in all other demographics. PF also reported significantly higher preoperative ODI than NPF. C were significantly younger and had lower BMI’s than both PF and NPF. All reported pain and disability improved significantly following surgery at all follow ups. At P3, all walking measures were comparable between PF and NPF and remained significantly different to C with the exception of single support time and step width which were comparable between PF, NPF and C at all follow ups. At P12, significant differences were observed in cadence (PF:95.4 ± 15.6step/min; NPF:103.5 ± 8.4step/min), step time (PF:0.6 ± 0.1s; NPF:0.6 ± 0.1s), and double support time (PF:0.4 ± 0.2s; NPF: 0.3 ± 0.1s) for treatment groups. Cadence in NPF normalized to C and step height normalized to C in PF while all other measures continued in the same pattern (all p < 0.05). Conclusion: Patients who were not fused to the pelvis exhibited delineation from patients who were fused to the pelvis by improving gait measures at one year postop. Interestingly, NPF showed cadence normalized to controls at one year postop despite significant age, height, and BMI differences, while PF showed similar normalization to C in step height despite the same demographic differences. These findings suggest that pelvic fusion may have a differential impact on specific aspects of gait dynamics over time. Further research is needed to explore the long-term functional outcomes and compensatory mechanisms that may develop in patients with pelvic fusion to optimize recovery strategies and improve quality of life.

ID: AGS26/2703

A260: Dystrophinopathy in the paravertebral muscle of adolescent idiopathic scoliosis: a prospective case-control study in China

Junyu Li 1 , Xiang Zhang 1 , Zekun Li 2 , Hanwen Zhang 1 , Y Zhang 3 , Miao Yu 1

1 Department of Orthopaedics, Peking University Third Hospital, Beijing Key Laboratory of Spinal Disease Research, Engineering Research Center of Bone and Joint Precision Medicine, Beijing, China

2 Peking University Health Science Center, Beijing, China

3 Department of Neurology, Peking University Third Hospital, Peking University, Beijing, China

Introduction: Adolescent idiopathic scoliosis (AIS) features vertebral rotation with lateral curvature, affecting 2%-4% of children under 16, with 0.04%-0.3% progressing to > 40°. Its cause is unclear despite genetic, epigenetic, endocrine, and developmental clues. Accumulating evidence points to the paraspinal muscles: fiber-type imbalance, multifidus ultrastructural changes, core-myopathy-like findings in some AIS cases, and parallels with neuromuscular disorders. Protein dysregulation, including altered vitamin D-binding protein, may also influence spinal morphology. We therefore conducted a prospective case-control study using histology and immunohistochemistry to characterize paraspinal muscle pathology and relate it to curve severity, testing a neuromuscular/protein-abnormality hypothesis for AIS. Material and Methods: This study enrolled 40 patients with AIS, 20 patients with congenital scoliosis (CS), and 20 patients with spinal degenerative disease (SDD). All patients underwent open posterior surgery in our hospital, and a paravertebral muscle (multifidus muscle) biopsy was performed intraoperatively. This study included many indexes that describe muscle, especially dystrophin staining. The above pathological results were compared among the AIS, CS, and SDD groups. The correlation between the Cobb angle and Nash-Moe classification and the above pathological results was analyzed in patients with AIS. Results: Significant reductions in the dystrophin staining of dystrophin-1 (p < 0.001), dystrophin-2 (p < 0.001), and dystrophin-3 (p < 0.001) were observed in the AIS group than in the CS and SDD groups. The higher the Nash-Moe classification in the AIS group, the more significant the loss of dystrophin-2 (p = 0.042) in the convex paraspinal muscles. Additionally, a significantly positive correlation was observed between the reductions of dystrophin-2 on the concave side of the AIS group and Cobb angle (p = 0.011). Conclusion: Dystrophin protein deficiency in the paraspinal muscles plays a crucial role in AIS formation and progression. The severity of scoliosis in patients with AIS is correlated with the extent of dystrophin loss in the paravertebral muscles. Therefore, dystrophin dysfunction may be relevant to AIS occurrence and development.

ID: AGS26/3323

A261: PI-LL should be replaced by PI-L4S1 to determine optimal lordosis in lumbar fusions

Fatemeh Alavi 1 , Christopher J. Nielsen 1 , Michael Kelly 2 , Jeffrey Hills 3 , Stephen Lewis 1

1 University of Toronto, University Health Network (UHN), Toronto, Canada

2 Rady Children’s Hospital, University of California San Diego, San Diego, United States

3 UT Health San Antonio, San Antonio, United States

Introduction: Understanding the correlations within the normal spine is essential to properly plan alignment targets that minimize the risk of mechanical complications following spinal deformity surgery. The PI-LL relationship has been used as a prominent measure in the past, however its inability to account for distribution of segmental lordosis has led to its inaccuracies in achieving ideal outcomes. Among sagittal parameters, the L1–pelvic angle (L1PA) has been shown to be a more reliable clinical measure that integrates pelvic morphology with lumbar curvature. However, its correlation with segmental sagittal parameters is unclear. The aim of this study is to investigate the relationship between L1PA and segmental parameters of spinal alignment in a normal population. Material and Methods: A cross-sectional radiographic analysis was conducted on international volunteers with full body standing radiographs. Study subjects with ODI >20, and radiographic disc degeneration or scoliosis were excluded. Spinopelvic parameters such as PI, L4-S1 lordosis (L4S1), and L1PA were measured. Linear regression analysis was used to quantify the relationship between sagittal parameters. Results: 320 healthy individuals with mean PI of 52° ± 10 (60% female) were available for analysis. The mean value of L1-S1 lordosis was 58° ± 10 and L4S1 was 37° ± 7. There were no correlations between PI-LL and PI, as well as L4S1 and PI. However, a strong correlation was found between PI-L4S1 and PI, using the following formula: PI-L4S1=0.9 × PI-30±7.4 (r2=0.62). The mean value of L1PA was 4.3° ± 6.5. A highly significant association was observed between PI-L4S1 and L1PA: L1PA=0.56 × (PI-L4S1)-4 (r2=0.95). Combining the proposed relation and the suggested L1PA=0.5 × PI-20 results in ideal value for L4S1: L4S1=0.12 × PI+30. Conclusion: PI–L4S1 is a novel sagittal alignment parameter, revealing several important correlations within the sagittal alignment of the spine. A significantly strong association between L1PA and PI–L4S1 demonstrated that upper lumbar alignment is closely linked to the orientation of the lower lumbar segments. Incorporating PI–L4S1 rather than relying solely on L1PA provides a more accurate and physiologically relevant reference for spinal alignment, particularly when planning surgical realignment or fusion procedures involving L4 to S1. This approach improves the generalizability of L1PA-based alignment targets to lower lumbar fusion constructs and may help optimize postoperative sagittal balance and mechanical loading, potentially reducing the risk of mechanical complications.

OP30: Thoracic Disc Surgery Focus

ID: AGS26/1180

A262: Patients with chronic low back pain without advanced disk degeneration exhibit gut microbiome dysbiosis: evidence from an age, sex and BMI matched study

Stone Sima 1 , Thomas Jeffries 2 , Alisha Sial 1,3,4 , Suhani Sharma 1 , Neha Chopra 1,3 , Fan Zhang 5 , Georgina Hold 5 , Victor Chen 1 , Lauren Barber 6 , Ashish Diwan 1,3,4

1 Spine Labs, Department of Orthopaedic Surgery, St George and Sutherland Clinical Campuses, School of Clinical Medicine, UNSW, Sydney, Sydney, Australia

2 School of Science, Western Sydney University Hawkesbury Campus, Richmond, New South Wales, Sydney, Australia

3 Spine Service, Department of Orthopaedic Surgery, St George Hospital, School of Clinical Medicine, UNSW, Sydney, Australia

4 Spinal Unit, Department of Orthopaedic Surgery, Royal Adelaide Hospital, University of Adelaide, Adelaide, Australia

5 Microbiome Research Centre, St George and Sutherland Clinical Campuses, Faculty of Medicine and Health, UNSW, Sydney, Australia

6 Department of Orthopaedic Surgery, Virginia Commonwealth University, Richmond, United States

Introduction: Chronic low back pain (LBP) is the leading cause of disability worldwide, yet its underlying pathophysiology remains poorly understood. Current diagnostic techniques rely heavily on imaging, which is insufficient for identifying inflammatory or systemic contributors. Emerging evidence suggests a role for gut microbiome dysbiosis in inflammatory pain pathways, yet its specific contribution to chronic LBP remains unclear. The aim of this study is to investigate gut microbiome alterations in chronic LBP patients compared to healthy controls. Materials and Methods: Twenty-eight patients with chronic LBP (> 3 months) without advanced disk degeneration (Pfirrmann grade <4) and 28 age-, sex-, and BMI-matched healthy controls were recruited. Stool samples were analyzed using 16S rRNA V4 gut microbiome analysis. Alpha diversity and beta diversity were assessed to determine microbial diversity differences. Taxonomic composition was compared at the phylum and genus levels. Results: There were no significant differences in age, sex, BMI, and race between the two groups. Patients with chronic LBP exhibited significantly reduced alpha diversity compared to controls (U = 517, p = 0.003, r = 0.395). Beta diversity analysis demonstrated distinct clustering between groups, suggesting significant shifts in microbial composition (PERMANOVA, p < 0.05). At the phylum level, Proteobacteria (U = 91, p < 0.001, r = 0.659) and Desulfobacterota (U = 73, p < 0.001, r = 0.718) were significantly elevated in chronic LBP patients, while Bacteroidota was significantly reduced (U = 524, p = 0.031, r =0.289). At the genus level, Prevotella (U = 272, p = 0.045, r = 0.268) and Faecalibacterium (U = 254, p = 0.024, r = 0.302) were increased in chronic LBP patients, whereas Parabacteroides was depleted (U = 621, p < 0.001, r = 0.500). Conclusion: These findings indicate that gut microbiome dysbiosis may play a key role in chronic LBP among patients without identifiable surgical pathologies, likely through systemic inflammation, reinforcing the “gut-disc axis” hypothesis. Given the limitations of conventional diagnostic tools and management strategies, microbiome profiling may provide novel biomarkers for LBP and inform microbiome-targeted therapies such as probiotics, dietary modifications, and fecal microbiota transplantation. Future research should explore causal relationships and strain-specific microbial effects to refine microbiome-based interventions for chronic LBP.

ID: AGS26/1293

A263: Bilateral transforaminal microdiscectomy (TFMD) combined with ponte osteotomy in thoracic disc surgery

Buse Sarigul 1 , Mesut Yilmaz 2 , Sedat Dalbayrak 3

1 Umraniye Research and Training Hospital, Department of Neurosurgery, Istanbul, Turkey

2 Medicana Atasehir Hospital, Norospinal Akademi, Istanbul, Turkey

3 Medicana Atasehir Hospital, Department of Neurosurgery, Norospinal Akademi, Istanbul, Turkey

Introduction: Surgical management of thoracic disc herniations, particularly those with calcified or hard disc morphology, remains challenging. Posterior approaches are associated with neural complications, whereas anterior approaches carry considerable morbidity and mortality. In this study, we aimed to evaluate the clinical and radiological outcomes of a combined technique of Ponte osteotomy and bilateral transforaminal microdiscectomy (PO+TFMD) in patients with thoracic disc herniation. Material and Methods: Medical records of 18 patients in whom thoracic disc herniation was detected and underwent PO+TFMD were evaluated retrospectively. The surgery was performed in a prone position with a median skin incision. Cranial and caudal pedicles and ascending neural roots were exposed using a Ponte osteotomy. The operating table was adjusted to extension to reduce spinal cord tension. The herniated disc was approached bilaterally via the foramina using microsurgical techniques. Disc material, including calcified central fragments, was excised with a high-speed drill and osteotome. A transforaminal lumbar interbody fusion (TLIF) cage, filled with local autograft from facets, was inserted, and compressive segmental instrumentation was applied in extension. All patients underwent thoracic MRI, CT and AP/lateral scoliosis X-rays preoperatively and postoperatively. Duration of surgery, intraoperative blood loss, complications, Visual Analog Scale (VAS) and Oswestry Disability Index (ODI) scores, segmental and thoracic kyphosis (T2-T12) were recorded preoperatively, on postoperative day 1, and at 6-month follow-up. Results: Of the 18 patients, 7 were female (38.9%) and 11 were male (61.1%), with a mean age of 41.37 ± 9.26 years. All patients presented with back pain, and 12 patients (66.7%) exhibited paraparesis preoperatively. The mean operative time was 111.8 ± 17.6 minutes, with an average intraoperative blood loss of 457.4 ± 142.3 mL. The mean preoperative VAS score was 7.42 ± 1.13, which decreased to 4.18 ± 0.87 on postoperative day 7 (p < 0.001) and further to 0.76 ± 0.71 at 6 months (p < 0.001). The mean preoperative ODI was 47.6 ± 10.4%, which improved significantly to 9.3 ± 6.4% at 6 months (p < 0.001). Thoracic kyphosis (T2-T12) improved from 47.83 ± 7.62° preoperatively to 39.27 ± 6.84° on postoperative day 1 (p < 0.001) and was maintained at 40.14 ± 6.91° at 6 months (p < 0.001). Segmental kyphosis corrected from 18.36 ± 4.73° preoperatively to 9.58 ± 3.82° postoperatively (p < 0.001) and stabilized at 10.21 ± 3.87° at 6 months (p < 0.001). No intraoperative complications were observed. Two patients (11.1%) developed transient postoperative paresthesia, which resolved spontaneously within 2 weeks. At the 6-month follow-up, all patients were pain-free, ODI ≤ 20%, and neurologically intact. Conclusion: The combination of Ponte osteotomy and bilateral transforaminal microdiscectomy provides a safe and effective posterior-only approach for thoracic disc herniation, with low complication rates, significant pain relief, and favorable kyphotic correction at mid-term follow-up. This technique may represent a viable alternative to conventional anterior or posterior strategies, particularly in cases involving calcified discs.

ID: AGS26/585

A264: Clinical and economic outcomes of open, microscopic, and endoscopic techniques for thoracic discectomy: a single institution study

Garrison Bentz 1 , Dillon Stone 1 , John Ogunlade 2 , Salim Yakdan 2 , Karan Joseph 2 , Mark Lambrechts 1

1 Washington University in St. Louis Department of Orthopaedic Surgery, St. Louis, United States

2 Washington University in St. Louis Department of Neurosurgery, St. Louis, United States

Introduction: Symptomatic thoracic disc herniations are a rare and difficult to treat pathology. Historically, open thoracic discectomy and fusion (TDF) was considered the primary treatment, however the introduction of minimally invasive options such as the anterolateral retropleural thoracic microdiscectomy (TAD) and posterolateral transforaminal endoscopic thoracic discectomy (TETD) with less soft tissue disruption and neural complications led to a decline in TDF use. Few studies have directly compared TDF, TAD, and TETD within a single institution. Materials and Methods: 37 patients over 10 years from a single institution were reviewed. Surgical outcomes, complications, cost, and other operative variables were compared between TDF, TAD, and TETD procedures used to treat single-level, symptomatic thoracic disc herniations. A minimum follow-up of 1 year was required. Cases with indications of trauma, malignancy, or infection were excluded. The Kruskal-Wallis test and the Chi-square test were used to analyze continuous and categorical variables, respectively. Significance was p < 0.05. Results: 37 patients were included (TDF: 15, TAD: 10, TETD: 12). TETD had a mean estimated blood loss (EBL) of 9 ml, lower than both TAD and TDF (p < 0.001). TETD has displayed a shorter length of stay at 1.75 days than both TAD (3.3 days) and TDF (4.67 days) (p = 0.008). TETD and TAD had a similar index cost of surgery at $13,122.33 and $14,816.78, however both were lower than the TDF procedures at $34,518.51 (p < 0.001). TDF also required intervention on a greater number of levels to achieve posterior fusion, whereas TETD and TAD only involved the single level (p < 0.001). There was no difference between the three cohorts in operative duration, rate of dural tears, post-operative complications, or emergency department visits. Conclusion: Minimally invasive techniques such as TETD and TAD have multiple advantages over traditional TDF in the treatment of thoracic disc herniations including EBL, length of stay, and cost. TETD was shown to have lower mean estimated blood loss and length of stay with a similar index cost when compared to TAD procedures. While both TAD and TETD have the benefits of a minimally invasive approach, TETD can provide similar outcomes with a similar cost, less blood loss, and a shorter hospital stay at the expense of a significant learning curve. Ultimately, the type and location of the disc herniation will play a large role in the optimal treatment approach, however TETD provides a valuable option for the non-centrally calcified symptomatic thoracic disc herniation.

ID: AGS26/485

A265: Stem cell therapy for degenerative disc disease: a systematic review of preclinical evidence, clinical translation, and future directions: systematic review

Vinicius Ricieri Ferraz 1 , Carlos Goulart 2 , Maria Ricieri Ferraz Franco de Souza 3 , Philippe A. Mercier 4 , Tobias Mattei 4 , Marcelo Ochoa Coelho de Souza Furlan 5

1 Dr. José de Carvalho Florence Hospital, The Regional Hospital of São José dos Campos - Dr. Rubens Savastano, São José dos Campos, São Paulo, Brazil

2 Division of Neurological Surgery, Essentia Health, Duluth, Minessota, United States

3 Division of Neurology, Irmandade da Santa Casa de Misericordia de Santos, Santos, São Paulo, Brazil

4 Division of Neurological Surgery, Saint Louis University School of Medicine,, Saint Louis, Missouri, United States

5 Pontifical Catholic University of Paraná, Curitiba, Brazil

Introduction: Degenerative disc disease (DDD) represents a predominant cause of chronic low back pain and associated disabilities on a global scale. Traditional treatment modalities, such as analgesics, physical therapy, and spinal fusion procedures, are primarily palliative and do not facilitate the restoration of disc structure or physiological function. The field of regenerative medicine has gained prominence as a promising alternative, with stem cell-based therapies demonstrating potential to reestablish disc functionality and to prevent or delay mechanical complications. This systematic review endeavors to synthesize the evidence derived from preclinical and clinical investigations concerning the efficacy, safety, and translational challenges associated with stem cell-based interventions for DDD. Material and Methods: A systematic review of the literature was conducted across PubMed/MEDLINE, Embase, Scopus, and the Cochrane Library from January 2000 to July 2025, in accordance with PRISMA guidelines. Eligible studies encompassed both preclinical and clinical trials assessing intradiscal injection or implantation of stem cells (including mesenchymal stem cells, adipose-derived stem cells, induced pluripotent stem cells, and stem cell-derived products) in degenerative disc disease (DDD). The outcomes evaluated comprised disc regeneration (via histology, imaging, and disc height measurements), patient-reported measures (such as Visual Analog Scale [VAS] and Oswestry Disability Index [ODI]), and adverse events. Data extraction and synthesis were performed independently by two reviewers. Results: Thirty-one studies met the inclusion criteria: 18 preclinical (animal models) and 13 clinical (1,299 patients, including five randomized controlled trials). Preclinical studies, all conducted in animal models, demonstrated consistent disc regeneration evidenced by increased proteoglycan content, enhanced disc hydration, and restoration of disc height. In human subjects, stem cell therapies yielded modest yet statistically significant improvements in pain (measured by VAS) and disability (assessed by ODI), although radiographic evidence of genuine disc regeneration remained inconsistent. The largest randomized controlled trial (Beall et al., 2025; n = 404) did not achieve its primary endpoint; however, subgroup analyses suggested potential benefits in specific populations. The safety profiles were favorable, predominantly indicating mild, transient adverse events; serious complications such as discitis were infrequent. No reports of tumorigenesis or severe immune reactions were documented. Conclusion: Stem cell-based therapies for DDD demonstrate encouraging regenerative effects in preclinical animal studies and provide modest clinical benefits in early human studies with an acceptable safety profile. However, consistent structural regeneration in humans remains unproven. Translational barriers include the hostile disc microenvironment, limited durability of transplanted cells, and heterogeneous patient selection. Future directions emphasize multimodal strategies-combining stem cells with biomaterials, exosomes, or genetic modifications-and the need for large, multicenter randomized trials to establish efficacy and optimize patient-specific applications.

ID: AGS26/590

A266: Benchmarking surgical approaches to thoracic disc herniation: a retrospective database study

Manish Kasliwal 1

1 University Hospitals Cleveland Medical Center, Case Western reserve University School of Medicine, Cleveland, United States

Introduction: Thoracic disc herniations are a relatively uncommon form of degenerative disc disease, comprising an estimated 0.1-3 % of all spinal disc herniations. Surgical treatment may utilize an anterior, lateral, or posterior/posterolateral approach, and approach selection is based on a number of factors. This study investigated national trends in surgical approach utilization as well as patient characteristics predictive of receiving a specific surgical approach. Material and Methods: This was a retrospective study querying the TriNetX national database. Patients were identified through specific ICD-9 coding for degenerative thoracic disc herniation and CPT coding for thoracic spinal procedures. Results: Of 3009 total patients meeting inclusion criteria, 2703 patients (89.8%) underwent decompression through a posterior or posterolateral approach, 222 (7.4%) via anterior approach, and 84 (2.8%) via lateral extracavitary approach. The number of posterior or posterolateral operations notably increased from 43 cases in 2010 to over 240 cases per year from 2022 to 2024. In contrast, anterior approach procedures reached a peak of 27 cases in 2017 and declined to 13 or fewer cases per year since. Conclusion: Most thoracic disc herniation patients in the TriNetX database received surgical decompression via a posterior or posterolateral approach. Although the annual number of surgeries performed for thoracic disc herniation patients increased after 2017, there was a decreased utilization of anterior approaches. This study demonstrates the widespread utilization of posterolateral-based approaches in the surgical management of thoracic disc herniations.

ID: AGS26/632

A267: Progression of disc degeneration over 23 years from adolesence to adulthood

Teija Lund 1 , Anni Aavikko 2 , Leena Ristolainen 3 , Hannu Kautiainen 4 , Martina Lohman 5 , Dietrich Schlenzka 3

1 Helsinki University, Helsinki University Hospital, Helsinki, Finland

2 Päijät-Häme Central Hospital, Lahti, Finland

3 Research Institute Orton, Helsinki, Finland

4 Folkhälsan Research Center, Helsinki, Finland

5 Helsinki University Hospital, Helsinki University, Helsinki, Finland

Introduction: Disc degeneration (DD) is commonly observed in both symptomatic and asymptomatic individuals and tends to increase with age. In adults, the most rapid progression occurs before the age of 50, with a strong genetic influence at younger age. This study aimed to investigate the longitudinal progression of DD and its relationship to lifetime low back pain (LBP) in healthy volunteers from adolescence into adulthood. Material and Methods: This longitudinal observational study included healthy volunteers examined at ages 11, 19 and 34 years. Each assessment comprised a structured interview, clinical examination, and lumbar spine magnetic resonance imaging (MRI). The interview addressed lifetime occurrence of LBP without specific trauma. Mid-sagittal T2-weighted MRI images were independently analyzed by a musculoskeletal radiologist and two spine surgeons. Intervertebral discs from L1/L2 to L5/S1 were graded using the 5-point Pfirrmann classification, with grade 3 or higher indicating degeneration (DD). A Pfirrmann Summary Score (PSS, range 5-25) was calculated by summing the scores across discs. Generalized Estimating Equation (GEE) model was applied to establish the rate of discs with Pfirrmann grade 3 or higher and the annual change in PSS. Results: Forty subjects had complete data sets at all three time points. The proportion of participants with DD was 12% at age 11, 48% at age 19, and 72% at age 34. At ages 19 and 34, no significant demographic or clinical differences were observed between participants with and without DD. PSS increased significantly in both groups over time, with significant between-group differences at ages 19 and 34. Across the cohort, the annual PSS increase was significantly greater from age 11 to age 19 (0.55, 95% CI 0.48 to 0.63) than from age 19 to age 34 (0.08, 95% CI 0.05 to 0.11). During adolescence, individual PSS increase varied considerably (range 1-7 grades), with no significant relationship to sex, sex-adjusted Body Surface Area (BSA) growth, or physical activity at age 19. After adjustment for sex, BMI, smoking and physical activity, participants with DD in at least one lumbar disc at age 19 or 34 reported significantly more lifetime LBP than those without DD. Conclusion: Progression of DD is strongly age dependent, with adolescence emerging as the critical phase marked by high inter-individual variability. The occurrence of DD increased steeply from early adolescence to adulthood. The presence of DD at ages 19 and 34 was associated with a higher lifetime report of LBP compared with those without DD.

ID: AGS26/796

A268: Recurrent lumbar disc herniation: a systematic review and meta-analysis of risk factors, surgical timing, and outcomes of revision discectomy versus fusion

Sydney Rucker 1 , Aidan Gillespie 1 , Robert Ferdon 2 , Warren Roth 1 , Jason Silvestre 1 , Charles Reitman 1 , James Lawrence 1 , Robert Ravinsky 1

1 Medical University of South Carolina, Charleston, United States

2 University of South Carolina - School of Medicine, Columbia, United States

Introduction: Recurrent lumbar disc herniation (RLDH) occurs in up to 15% of patients following primary discectomy and poses significant challenges in surgical decision-making. Two common reoperative strategies include revision discectomy (RD) and spinal fusion (SF), yet the optimal treatment approach remains a subject of debate. Additionally, the ideal timing for reoperation is poorly defined, with limited data available. This study aimed to systematically review the risk factors associated with RLDH, evaluate the impact of surgical timing, and compare outcomes between RD and SF. Materials and Methods: A systematic review and meta-analysis were performed in accordance with PRISMA guidelines and registered with PROSPERO. Literature searches were conducted through December 2024 using PubMed, Scopus, and CINAHL databases. Studies were included if they reported on patients undergoing revision surgery for RLDH and provided data on surgical techniques, timing, and clinical outcomes. Risk of bias was assessed using the ROB-2 tool for randomized controlled trials and the Newcastle-Ottawa Scale for observational studies. A random-effects meta-analysis was used to evaluate 12-month improvements in back pain (VAS back), leg pain (VAS leg), and disability (ODI) scores. Secondary outcomes included perioperative parameters, complication rates, recurrence rates, and patient satisfaction. Results: Ten studies comprising a total of 939 patients met inclusion criteria. Commonly reported risk factors for RLDH included male sex, obesity, smoking, diabetes, and large annular defects. Surgical timing was inconsistently reported across studies, preventing meaningful pooled analysis on this variable. When comparing surgical approaches, spinal fusion was associated with significantly greater intraoperative blood loss (p = 0.039), while no significant differences were observed in operative time or hospitalization duration. Recurrence rates were not significantly different between RD and SF (p = 0.262), although open fusion showed a trend toward reduced recurrence. No significant differences were found in 12-month VAS or ODI scores between the two groups (p > 0.3). Minimally invasive techniques such as MIS-TLIF and PELD were associated with fewer complications and higher patient satisfaction. Conclusion: Both RD and SF are viable surgical options for the treatment of RLDH, each presenting unique risk-benefit considerations. While fusion may reduce recurrence in some cases, it is associated with greater blood loss. Importantly, the inconsistent reporting of surgical timing emerged as a major limitation across studies. To better inform surgical decision-making, future research should focus on high-quality studies with standardized definitions of recurrence and consistent reporting of reoperation timeframes, supported by long-term follow-up data.

ID: AGS26/838

A269: National database analysis of trends in thoracic disc herniation surgical management

Pranav Kumar 1 , Eric Herring 2 , rohit mauria 2 , Collin Labak 2 , Jack Xiao 2 , Manish Kasliwal 2

1 Case Western Reserve University School of Medicine, University Hospitals Cleveland Medical Center, Cleveland, United States

2 University Hospitals Cleveland Medical Center, Cleveland, United States

Introduction: Thoracic disc herniation is a relatively uncommon form of degenerative disc disease, comprising an estimated 0.1% to 3% of all spinal disc herniations. Surgical treatment may utilize an anterior, lateral, or posterior/posterolateral approach, and approach selection must consider direction of herniation, level, presence of calcification, comorbidities, visualization required, and surgeon preference. The anterior approach is typically associated with less neurologic morbidity but involves more manipulation of cardiac and pulmonary structures. This study investigated the number of patients receiving surgical treatment of thoracic disc herniation through anterior, lateral, or posterior/posterolateral approaches to identify national trends in surgical approach utilization, as well as patient characteristics predictive of receiving a specific surgical approach. Material and Methods: This was a retrospective cohort study querying all patients in TriNetX. Patients were identified through specific ICD-9 coding for degenerative thoracic disc herniation and CPT coding for thoracic spinal procedures. Further exclusion criteria were diagnosis of spinal infection, neoplasm, pathologic fracture, or trauma. Patients were further divided through approach-specific CPT codes. Of 111,942 patients in the TriNetX database with thoracic disc herniation, 3,009 total patients met inclusion criteria. The primary outcome compared was the number of patients receiving the posterior/posterolateral, anterior, or lateral approach. The number of surgeries utilizing each approach was recorded by year to evaluate trends of utilization for each approach. Other data points compared between approach cohorts were patient geographic region, sex, and race. Results: Of 3,009 total patients meeting inclusion criteria, 2,703 patients (89.8%) underwent decompression through a posterior or posterolateral approach, 222 patients (7.4%) via anterior approach, and 84 patients (2.8%) via lateral extracavitary approach (LECA). The number of posterior or posterolateral approach-coded decompression surgeries notably increased from 43 cases in 2010 to over 240 cases per year over 2022 - 2024. In contrast, the number of anterior approach procedures reached a peak of 27 cases in 2017 and declined to 13 or fewer cases per year since. Compared to patients receiving a posterior or posterolateral approach, anterior approach patients were significantly more likely to be female (50% vs 42%, p = 0.02) and white (82% vs 75%, p = 0.009). There was no significant difference in approach utilization by geographic region within the United States. Conclusion: Most thoracic disc herniation patients in the TriNetX database received surgical decompression via a posterior or posterolateral approach. Although the annual number of surgeries performed for thoracic disc herniation patients increased after 2017, there was a decreased utilization of anterior approaches.

ID: AGS26/859

A270: Outcomes of endoscopic vs. traditional lumbar microdiscectomy: a comparitive cohort analysis

Hanna von Riegen 1 , Erica Greiner 1 , Jason Pittman 1 , Sapan Gandhi 1

1 Department of Orthopedics, Beth Israel Deaconess Medical Center, Harvard Medical School in Boston, MA, Boston, United States

Introduction: Endoscopic lumbar discectomy is increasingly adopted as a minimally invasive alternative to traditional (i.e.open or tubular) lumbar microdiscectomy. While single-center studies suggest reduced perioperative complications with endoscopic approaches, real-world evidence across broader healthcare systems remains limited. We compared short- and long-term outcomes between endoscopic and traditional lumbar microdiscectomy in a large international database. Material and Methods: We queried the TriNetX Global Collaborative Network (> 150 healthcare organizations, ∼20 countries) for adults (≥ 18 years) undergoing traditional lumbar microdiscectomy (CPT 63030; excluding CPT 63035) or endoscopic lumbar discectomy (CPT 62380) between 2005-2025. The index event was the first qualifying procedure. Outcomes were analyzed in three time windows: immediate (day 1), short-term (1-90 days), and long-term (20-730 days). To ensure sufficient follow-up, only patients with index procedures up to August 2023 were included in 2-year analyses. Propensity score matching (1:1) was performed across all available demographic, comorbidity, and laboratory variables. Risk differences, ratios, and odds ratios were calculated; patients were censored at their last recorded fact. Sensitivity analyses including later index dates and excluding patients with pre-existing outcomes yielded similar results. Results: After matching, 1,987 patients per cohort were analyzed for 90-day outcomes and 1,263 per cohort for 2-year outcomes. At day 1, dural tears were more frequent after open discectomy (1.0% vs. 0.5%, n.s.), while nerve injury was equal (0.5%). At 90 days, wound complications, urinary tract infections, and readmissions were significantly higher after traditional procedures, whereas persistent pain was more common following endoscopic discectomy (6.5% vs. 3.4%, p < 0.001). Neurostimulator implantation was also more frequent after open surgery (0.5% vs. 0%, p = 0.002). At 2 years, repeat endoscopic procedures were more common after endoscopic discectomy (2.1% vs. 0.8%, p = 0.007), while revision open surgery, laminectomy, and posterior fusion rates were comparable. Conclusion: In this large real-world cohort, endoscopic discectomy was associated with fewer short-term complications and readmissions compared with traditional microdiscectomy, but also with higher rates of persistent pain and repeat endoscopic procedures at long-term follow-up. These findings complement and partly diverge from single-center reports, highlighting the need for further studies.

OP31: Recovery Influencers in Lumbar Surgery

ID: AGS26/269

A271: Prevalence of radiculitis after TLIF with expandable cage: our experience and comparison with the literature

Diego Prado 1 , Andres Chahín 1 , Cristian Ricardo Correa Valencia 1 , Álvaro Soteres 1 , Javiera Babarovich 1 , Sofia Chahin 1 , Andrés Leiva Rivera 2

1 Clinica Las Condes, Instituto de Neurocirugía Doctor Asenjo, Santiago, Chile

2 Clinica Meds, Santiago, Chile

Introduction: Transforaminal lumbar interbody fusion (TLIF) with expandable cages has gained widespread acceptance for the treatment of degenerative lumbar spine pathology. This technique enables restoration of disc height and foraminal space with less manipulation of neural structures compared with traditional static cages. These advantages may reduce the risk of postoperative radiculitis, a complication characterized by new or worsened radicular pain after surgery. However, international literature reports highly variable prevalence rates with static cages, ranging from 2.8% to 57%, reflecting heterogeneity in definitions, implant designs, and patient populations. Material and Methods: We performed a retrospective descriptive review of consecutive patients who underwent TLIF with expandable cages between January 2023 and March 2025. Inclusion criteria were patients with degenerative lumbar spine pathology treated with TLIF using an expandable interbody device. Exclusion criteria included infection, trauma, tumor, or incomplete clinical records. Clinical and demographic variables (age, sex, comorbidities), surgical characteristics (operative level, duration, estimated blood loss), and hospital course were collected. The primary outcome was the presence of postoperative radiculitis, defined as new or persistent radicular pain within the first postoperative month not attributable to residual stenosis, hematoma, or implant malposition. Secondary variables included severity of symptoms, duration of hospital stay, and need for surgical reintervention. Results: A total of 102 patients met inclusion criteria. The mean age was 57.9 years, and 72.8% were female. The prevalence of postoperative radiculitis was 17.6% (n = 18). Among these cases, symptoms were self-limited or managed conservatively in the majority, with 88% not requiring reoperation. Only two patients (12%) required revision surgery due to persistent or severe radiculitis. The mean length of hospital stay was comparable between patients with and without radiculitis. The observed prevalence in this cohort lies within the range reported in contemporary literature. Conclusion: In our two-center experience, the prevalence of postoperative radiculitis following TLIF with expandable cages was 17.6%, consistent with previously published rates. Although expandable cages provide technical and radiographic benefits, such as restoration of disc height and foraminal enlargement, they do not appear to significantly reduce the incidence of postoperative radiculitis compared with traditional static devices. These findings suggest that factors beyond implant design-such as surgical technique, patient-specific anatomy, and biological responses-may play a more decisive role in the development of this complication. Further prospective studies with standardized definitions and larger sample sizes are warranted to better clarify risk factors and potential preventive strategies.

ID: AGS26/3029

A272: Paraspinal muscle degeneration and spinal mobility: evidence from the Berlin Back Study

Lea Hühren 1 , Lukas Schönnagel 1 , Lea Cordes 1 , Ali Firouzabadi 1 , Rosa-Maria Berndt 1 , Leo Lemminger 1 , Lukas Moedl 1 , Luis Becker 1 , Bernhard Hoehl 1 , Hendrik Schmidt 1 , Matthias Pumberger 1

1 Charite Universitätsmedizin Berlin, Berlin, Germany

Introduction: Chronic low back pain (cLBP) is considered a multifactorial condition in which structural and functional changes of the paraspinal musculature are increasingly considered important contributors as they play a crucial role in stabilization of the spine. While numerous studies have examined the association of morphological changes with pain, clinical evidence linking muscle morphology to spinal mobility remains limited. Methods: This study is a sub analysis of the “Berlin Back Studie” in which patients with and without cLBP were prospectively enrolled from the general population. Paraspinal muscles (multifidus, erector spinae, psoas major) were segmented at L1-S1 on axial T2-weighted MRI using ITK-SNAP; average values were used to represent overall muscle status. Atrophy was measured by the percentage of fatty infiltration (FI%) and the height adjusted functional cross-sectional area (fCSA). Sagittal and coronal spinal mobility was assessed using a gyroscopic skin surface device. Multivariable linear regression was used to analyze the association between FI and spinal mobility, adjusted important confounders: for age, sex, BMI, physical activity (measured by the IPAQ) and disc degeneration. In the cLBP group, pain intensity was also accounted for. Results: A total of 721 patients were included in the study, after exclusions. 226 without LBP and 495 with LBP. The mean Age was 40.75 (13.06) and 42.37 (11.58), respectively. In the multivariable analysis in the cLBP group, FI of the multifidus showed the largest effect size among all muscle parameters (Eta 2 = 0.054, p < 0.001) in its association with sagittal lumbar mobility, also exceeding that of other confounders such as age, sex, BMI, and disc degeneration, physical activity and even pain intensity (Eta 2 = 0.022, p = 0.002). However, no significant association with multifidus fCSA existed. For or the erector spinae, fCSA - but not FI - showed a significant association with mobility (Eta 2 = 0.017, p = 0.004). No significant associations of muscle and lumbar sagittal mobility were found in the asymptomatic group or for coronal lumbar mobility in either group. Conclusion: This study provides clinical evidence that paraspinal muscle degeneration is independently associated with reduced spinal mobility, particularly in individuals with chronic low back pain, demonstrating a larger independent effect than age, disc degeneration or pain intensity. In contrast to previous studies that primarily examined associations with strength or pain, our analysis focused on unloaded spinal mobility, capturing a distinct functional domain relevant to everyday activities. Notably, associations differed between cLBP and controls and also differed by muscle, evaluated morphological characteristics and direction of movement, highlighting the need for comprehensive measurements in future research.

ID: AGS26/2830

A273: Forearm bone mineral density predicts screw loosening after lumbar fusion similar to lumbar Hounsfield unit value in patients with lumbar spondylolisthesis

Qian Chen 1 , Youwei Ai 2 , Hong Ding 2 , Limin Liu 2

1 Affiliated Hospital of North Sichuan Medical College, Nanchong, China

2 West China Hospital, Sichuan University, Chengdu, China

Introduction: Preoperative bone density assessment is necessary to predict screw loosening. The forearm BMD is a useful predictor of BMD-related complications after lumbar operation. Our results show that the forearm BMD is as effective a predictor of screw loosening as the lumbar average HU value. Measurement of the forearm BMD may be a useful adjunct in predicting screw loosening following lumbar fusion. To determine the relationship between forearm bone mineral density (BMD) and the risk of pedicle screw loosening in patients with lumbar spondylolisthesis. Material and Methods: We retrospectively evaluated 270 patients who underwent posterior lumbar interbody fusion for lumbar spondylolisthesis. The patients were divided into two groups on the basis of the with or without loose screws: the loosening group and the non-loosening group. The patient’s gender, age, BMI, smoking and diabetes histories, and the operative segment were recorded as the basic information. The Hounsfield unit (HU) value for the BMD of the L1-4 lumbar was measured using computed tomography. The patient’s distal one-third of the length of the radius and ulna of the non-dominant forearm was chosen as the site for dual-energy X-ray (DXA) bone density testing. Results: The rate of screw loosening was 13% at a minimum 12 months follow-up. Average forearm BMD (0.461 ± 0.1 vs 0.577 ± 0.1, p < 0.001) and mean HU value (L1-4) (121.1 ± 27.3 vs 155.6 ± 32.2, p < 0.001) were lower in the screw loosening group than those in the non-loosening group. In multivariate logistic regression analysis, the forearm BMD (OR 0.840; 95%CI 0.797-0.886) and HU value (L1-4) (OR 0.952; 95%CI 0.935-0.969) were independent risk factor for screw loosening. The area under the curve (AUC) for the forearm BMD and HU value for prediction of pedicle screw loosening was 0.802 and 0.811. The forearm BMD cut-off for predicting pedicle screw loosening was 0.543 (sensitivity, 0.800; specificity, 0.864). Conclusion: The forearm BMD was an independent risk factor for loosening of the lumbar pedicle screws. The forearm BMD was a valid predictor of pedicle screw loosening in patients undergoing lumbar fusion, as was the CT HU value.

ID: AGS26/3373

A274: ALIF standalone vs ALIF + posterior instrumentation: precise indications and long-term outcomes. A systematic review

Oscar Molina-Romero 1 , Santiago Vallejo-Puerta 2 , Julio Fonnegra-Pardo 3 , Andres Segura-Hernandez 3 , Andrés Fonnegra-Caballero 3 , Juan Carlos Diez-Palma 3 , Jorge Aristizabal-Maya 2

1 Neurosurgery program, Universidad el bosque, Bogotá, Colombia, Colombia

2 Fundación Cardioinfantil, Bogotá, Colombia

3 Clinica Shaio, Bogotá, Colombia

Introduction: The optimal surgical approach for lumbar degenerative pathology remains controversial. While ALIF with posterior instrumentation (ALIF+PI) provides maximum stability, standalone ALIF (sa-ALIF) offers reduced morbidity with preservation of posterior elements. The present review aims to compare long-term clinical and radiographic outcomes between sa-ALIF and ALIF+PI, establishing precise indications for each technique. Material and Methods: Systematic review following PRISMA guidelines. Comprehensive search of PubMed, Embase, and Cochrane databases (2000-2024) for comparative studies evaluating sa-ALIF versus ALIF+PI. Inclusion criteria: adult patients with lumbar degenerative pathology, minimum 2-year follow-up, data on clinical and radiographic outcomes. Data extraction included fusion rates, clinical improvement (ODI, VAS), complications, and reoperation rates. Results: Comparative analysis revealed similar fusion rates: sa-ALIF 89.3% vs ALIF+PI 92.1% (p = 0.18). Clinical outcomes showed comparable improvements: ODI reduction sa-ALIF -21.6 points vs ALIF+PI -16.0 points (p = 0.12), with sa-ALIF demonstrating superior SF-36 physical component improvement (+12.6 vs +8.8 points, p < 0.03). Sa-ALIF showed consistent operative advantages: reduced surgical time (87-114 min vs 118-164 min, p < 0.001), decreased blood loss (131-212 mL vs 341-413 mL, p < 0.008), and faster recovery with earlier mobilization. However, sa-ALIF exhibited higher subsidence rates (11-32% vs 4-8%) and increased reoperation rates for supplemental fixation (5.8% vs 0-2%). Lordosis restoration was superior with ALIF+PI (+8-12° vs +5-8°). For specific pathologies, sa-ALIF showed excellent results in degenerative disc disease but limited evidence in spondylolisthesis, where ALIF+PI demonstrated greater stability and lower failure rates. Conclusion: Sa-ALIF provides comparable clinical outcomes to ALIF+PI with significant perioperative advantages but higher reoperation rates. Sa-ALIF is indicated in patients with degenerative disc disease, preserved facet joints, adequate bone density, and minimal segmental instability. ALIF+PI is preferred for spondylolisthesis, severe sagittal malalignment, osteoporosis, and facet degeneration. High-quality randomized controlled trials are needed to obtain treatment algorithms.

ID: AGS26/3408

A275: Risk of nerve injury in LLIF: a propensity-matched assessment of retractor time and postoperative neurologic outcomes

Kyle Dann 1 , Michael McDermott 2 , Alfred-John Bayaton 2 , Ashish Patel 1

1 The Spine Center at Duly Health and Care, Naperville IL, United States

2 Kettering Health Grandview Hospital Orthopedic Surgery Residency Program, Dayton OH, United States

Introduction: For many lateral surgeons, a psoas retraction time under 20 minutes has been considered a surrogate marker for minimizing nerve injury risk since Bendersky first suggested it in 2015. However, this threshold was merely an observation of that study, not the product of multivariate analysis. This has led to an emphasis on minimizing retractor time in conventional teaching, often adhering to a “20-minute rule.” Our recent multi-center analysis examined the retractor time theory with a study of 658 patients, finding no relationship between retractor time and injury, highlighting that emphasizing speed over technical precision may not be what is best for our patients. This study aims to re-examine the relationship between retraction time and nerve injury in Lateral Lumbar Interbody Fusion (LLIF) using a rigorously propensity-matched cohort to provide a more robust evaluation of the “20-minute rule.” Material and Methods: This retrospective cohort study of 658 patients undergoing LLIF was propensity matched based on gender, age, surgical positioning, and number of operative levels, with a matching tolerance of 0.05. After matching, the cohort was split into groups based on retractor time above and below the 20-minute threshold, yielding two comparable groups: 172 patients with retractor times < 20 minutes and 170 patients with retractor times > 20 minutes. Following a statistical comparison of the groups, a Receiver Operating Characteristic (ROC) curve analysis was performed to assess the predictive value of using retractor time as a threshold for injury. Results: The overall injury rate in the propensity-matched cohort was 2.9% (10/342). The average retractor time for injured patients was 19.8 ± 12.4 minutes compared to 21.2 ± 9.2 minutes for non-injured patients (p = 0.634). There was no statistical difference in the injury rates between patients above or below the 20-minute threshold (3.5% vs 2.3%, p = 0.541). The ROC curve analysis yielded an Area Under the Curve (AUC) of 0.498, indicating that using a retractor time as a surrogate marker for injury had no predictive power, performing equivalently to a random guess. Conclusion: This propensity-matched analysis strongly indicates that postoperative quadriceps motor palsy after LLIF is not primarily caused by prolonged retractor time alone, and the “20-minute rule” is a poor predictor of injury. There is a need for a more sophisticated understanding of the injuries in LLIF and a modality that can identify injuries in real time. Until then, surgeons should prioritize meticulous technique over adherence to a strict time limit.

ID: AGS26/124

A276: Discharge hemoglobin impacts functional recovery after lumbar fusion

Mitsuhiro Nishizawa 1 , Steven Glassman 1 , Mladen Djurasovic 1 , Charles Crawford, III 1 , Jeffrey Gum 1 , John Dimar 1 , Kirk Owens 1 , Justin Mathew 1 , Leah Carreon 1

1 Norton Leatherman Spine Center, Louisville, Kentucky, United States

Introduction: Acute blood loss anemia is common following lumbar spine surgery. Current protocols emphasize limiting transfusions resulting in more patients discharged with lower hemoglobin. The purpose of this study is to identify the impact of anemia at discharge on surgical outcomes and functional recovery following lumbar fusion. Material and Methods: Records of a consecutive series of patients who underwent one- or two-level transforaminal lumbar interbody fusion (TLIF) from 2013 to 2022 were reviewed. Standard demographics, surgical parameters, and complications within a year after discharge were recorded. Oswestry Disability Index (ODI) and Numeric Rating Scale (NRS, 0-10) for back and leg pain, were collected pre-op, and at 3-month and 1 year postoperatively. Patients were categorized into three groups based on discharge hemoglobin: Moderate (8-10g/dL), Mild (10-13g/dL for males; 10-12g/dL for females), and No anemia (> 13g/dL for males; > 12g/dL for females). Complications and PROMs were compared. Results: Of 576 patients, 253 were included in the analysis. The average hemoglobin at discharge was 11.1 ± 1.6 mg/dL. Fifty-eight patients (23%) were discharged with moderate, 144 (57%) with mild, and 51 (20%) with no anemia. Complications rates were higher in patients discharged with lower hemoglobin, with significant differences observed in 90-day readmission (14%, 4%, 0%, p = 0.002), infections (14%, 4%, 4%, p = 0.022), stroke (5%, 0%, 2%, p = 0.021), and gastrointestinal complications (13%, 0%, 8%, p < 0.001). Patients with moderate anemia at discharge showed the highest ODI and NRS score of leg pain at all timepoints. There were significant differences in the trajectories of ODI (p = 0.021) and leg pain (p = 0.018) among the three groups. Among 180 patients with no complications, either during hospitalization or after discharge, the significant differences remained in leg pain. Conclusion: Anemia at discharge was significantly associated with complication rate and worse functional recovery following lumbar fusion surgery.

ID: AGS26/846

A277: Lateral decubitus anterior lumbar interbody fusion at the L4-5 level is a safe alternative to lateral lumbar interbody fusion in lateral decubitus single position surgery

Aaron Buckland 1 , Nam Hunyh 1 , Brett Braly 2 , Cristiano Menezes 3 , Emily Oettinger 1 , Hannah Herrick 4 , CJ Tyson 1 , J. Alex Thomas 5

1 Spine and Scoliosis Research Associates Australia, Windsor, Australia

2 Healthcare Partners Investments, Inc, Oklahoma Sports, Science and Orthopaedics, Oklahoma City, United States

3 Instituto Columna, Belo Horizonte, MG, Brazil

4 Medical Scribe at UNC Health, Chapel Hill, United States

5 Atlantic Neurosurgical and Spine, Wilmington, NC, United States

Introduction: The safety of L4-5 access in lateral lumbar interbody fusion (LLIF) may be limited by psoas, iliac crest and vascular anatomy. Alternatively, lateral decubitus anterior lumbar interbody fusion (L-ALIF) can be used for the L4-5 level during lateral single position surgery (L-SPS). This study aims to assess differences in efficiency, safety and alignment between L-ALIF and LLIF at the L4-5 level in L-SPS. Material and Methods: Multicentre, consecutively, prospectively enrolled 1-2 level L-SPS patients including L4-5 were analysed from 7 institutions across 3 countries (Australia, Brazil, USA). Patients were groups by the L4-5 anterior column access technique: L-ALIF or LLIF. Demographics, procedural characteristics, complications and sagittal alignment were analysed. Results: Of 511 patients, 72 underwent L4-5 ALIF and 439 L4-5 LLIF. Demographics were similar except for sex (36.10% vs. 55.30% males, p = 0.003) and diabetics (13.90% vs. 7.10%, p = 0.048). L-ALIF patients had more levels fused (1.65 vs.1.25, p < 0.001), longer operative time (115.78 vs. 89.08 minutes, p < 0 .001) and length of stay (LOS) (2.41 vs. 1.71days, p < 0.001). Propensity matching for demographics and levels fused (72 per group), demonstrated similar operative time (115.78 vs. 115.77 min, p = 0.998), LOS (2.41 vs. 1.98 days, p = 0.069) and blood loss (138.68 vs. 165.00mL, p = 0.408). L-ALIF had less postoperative- (4.20% vs. 21.60%, p < 0.001) and minor complications (1.40% vs. 14.40%, p = 0.002), mainly attributable to transient thigh dysaesthesias amongst LLIF (0.00% vs. 6.20%, p = 0.031). Intraoperative (2.80% vs. 2.50%, p = 0.892) and major postoperative complications (2.80% vs. 8.90%, p = 0.077) was similar. Reoperations within 90-days (2.80% vs. 7.10%, p = 0.170) and cage revision (1.40% vs. 1.10%, p = 0.855) were similar. L-ALIF better restored lumbar lordosis (58.54° vs. 53.34°, p = 0.007), L4-5 segmental lordosis (6.62° vs. 1.76, p < 0.001) and L4-S1 lordosis (36.14° vs. 29.73°, p = 0.017). Conclusion: L-ALIF at L4-5 is a safe and effective alternative to LLIF in L-SPS. Additionally, L4-5 L-ALIF may more effectively improve L4-5 segmental- and lumbar-lordosis, and reduce complications.

ID: AGS26/847

A278: Preoperative opioid prescribing prior to lumbar decompression surgery markedly increases the risk of chronic opioid use at 6- and 12-month postoperatively

Nam Hunyh 1 , Darcy Fay 1 , Aaron Buckland 1

1 Spine and Scoliosis Research Associates Australia, Windsor, Australia

Introduction: Lumbar decompression surgeries are common procedures for lumbar canal stenosis and disc herniation. Opioids are commonly prescribed for lumbar back pain pre- and postoperatively. However they carry the risk ongoing usage and associated morbidity. This study aims to analyse the effects of preoperative opioid prescribing on chronic opioid use following 1-2 level lumbar decompression. Material and Methods: Consecutively, prospectively enrolled 1-2 level lumbar decompression patients were analysed. Patient demographics, complications and patient reported outcome measures (PROMs) were compared amongst opioid-sensitised and opioid-naïve patients. Opioid dosage was measured in morphine milligram equivalents (MME). Results: Of 254 patients, 77 were opioid-sensitised and 177 opioid-naïve. Demographics and complication rates were similar. Six-month postoperative MME was 3x higher amongst opioid-sensitised patients (1931 vs. 641 mg, p < 0.001) and correlated with MME prescribed 1-month (r = 0.428, p < 0.001) and 3-months preoperatively (r = 0.594, p < 0.001). More opioid-sensitised patients continued opioids at 6-weeks (36.4% vs. 11.3%, p < 0.001), 6-months (31.5% vs. 8.5%, p < 0.001) and 12-months (33.8% vs. 13.6%, p < 0.001) postoperatively. Binary logistic regression analyses found opioid use > 1-month preoperatively increased the risk of postoperative use at 6-weeks (OR = 6.7 [95%CI 3.4-13.3], p < 0.001), 6 months (OR = 11.6 [95%CI 4.8-28.3], p < 0.001) and 12-months (OR = 15.5 [95%CI 6.3-37.2], p < 0.001). Opioid-sensitised patients had poorer preoperative ODI (50.1 vs. 39.8, p < 0.001) and EQ-5D-3L index (0.48 vs. 0.57, p = 0.002), but similar VAS-back (5.02 vs. 5.18, p = 0.820) and VAS-leg scores (7.41 vs. 6.93, p = 0.404). Six-week, 6-months and 12-months postoperative ODI, VAS-back, VAS-leg and EQ-5D-3L were similar. At 12-months postoperatively, opioid-sensitive patients scored worse for anxiety/depression (EQ-5D-3L) (1.44 vs. 1.21, p = 0.033) and social life (ODI) (1.33 vs. 0.68, p = 0.017). Conclusion: Preoperative opioid prescribing predisposes lumbar decompression patients to increased perioperative opioid requirements and higher rates of continued use postoperatively, particularly when prescribed for > 1 month preoperatively, with no differences in postoperative PROMs. Minimising preoperative opioid use and/or timely surgical referral can help minimise long-term opioid use.

ID: AGS26/929

A279: Lateral interbody fusion at l4-5: safety and outcomes from a series of over 100 cases

Puya Alikhani 1 , Diego Soto Rubio 1 , Nikhil Dholaria 2 , Katherine Baker 1 , Sawyer Bauer 1 , Joshua Vignolles-Jeung 1 , Dana Saleh 1 , Mohamed Emara 1 , Mohsen Rostami 1 , Patrick Kim 1

1 Department of Neurosurgery, Brain and Spine, University of South Florida, Morsani College of Medicine, Tampa, United States

2 University of Arizona College of Medicine - Phoenix, Phoenix, United States

Introduction: The lateral retroperitoneal approach has been widely adopted as a minimally invasive option for degenerative lumbar spondylolisthesis and adult spinal deformities, offering high fusion rates while preserving the posterior tension band. A major concern during lateral lumbar interbody fusion (LLIF/XLIF) is the proximity of the femoral nerve, which varies by disc level and may increase the risk of postoperative motor and sensory deficits. Because knee extension is directly dependent on femoral nerve function, it serves as a reliable marker of nerve injury. Notably, as the femoral nerve descends caudally, it lies closer to the lumbar spine, and the safe working corridor narrows at L3-L4 and L4-L5, raising concern for higher complication rates at these levels. Material and Methods: We retrospectively reviewed patients who underwent single-level LLIF at our tertiary center between 2014 and 2022. Demographic, operative, and complication data were collected, including intra- and postoperative outcomes. Descriptive statistics were generated using SPSS. Only patients with a minimum six-month follow-up were included. Results: A total of 133 patients underwent LLIF: L1-L2 (n = 8), L2-L3 (n = 18), L3-L4 (n = 32), and L4-L5 (n = 75). Patients treated at higher levels were older (71.6 ± 5.0 years at L1-L2 vs. 59.3 ± 9.4 years at L4-L5, p < .001). Sex, BMI, and hospital length of stay did not differ significantly. Percutaneous screw use was highest at L4-L5 (52%) and absent at L1-L2 (p < .001). Standalone procedures were most common at L2-L3 (83.3%) and least common at L4-L5 (41.3%, p = .007). Neurological complications did not differ significantly across levels. Knee extension and hip flexion weakness were rare, each affecting one patient at L4-L5 (1.3%, p = .854). Transient weakness occurred in 16.7% at L2-L3, 15.6% at L3-L4, and 9.3% at L4-L5 (p = .489). Pain- or effort-limited weakness was observed in 32% at L4-L5 and 46.9% at L3-L4 (p = .251). Acute thigh paresthesia was most frequent at L3-L4 (21.9%) but not statistically significant (p = .182). Transient paresthesia occurred in 15.6% at L3-L4 and 6.7% at L4-L5 (p = .157). Conclusion: LLIF is an effective, minimally invasive fusion technique with an acceptable safety profile. Although anatomical considerations suggest greater risk at L4-L5, due to the more anterior location of the lumbar plexus, we found no significant increase in neurological complications at L4-L5. Transient motor and sensory alterations were similarly distributed across levels. These findings indicate that femoral nerve-related complications are not uniquely elevated at L4-L5, though minimal traction time remains essential.

OP32: Endoscopic MIS

ID: AGS26/1677

A280: Comparative analysis of uniportal and biportal (contralateral and ipsilateral) endoscopic spine surgery for cervical foraminal stenosis: clinical and radiographic outcomes

Borriwat Santipas 1

1 Department of Orthopaedic Surgery, Faculty Medicine Siriraj Hospital, Mahidol University, Bangkok, Thailand

Introduction: Cervical radiculopathy, a common consequence of cervical foraminal stenosis, often necessitates surgical intervention after conservative treatments fail. Among the available surgical options, posterior cervical foraminotomy (PCF) is a less invasive alternative to anterior cervical discectomy and fusion, aiming to relieve nerve root compression effectively. The evolution of minimally invasive techniques, including uniportal and biportal endoscopic approaches, has further refined this procedure, potentially offering reduced muscle dissection, less postoperative pain, and faster recovery. However, despite the advancements, a critical gap remains in the current understanding-specifically, the comparative effectiveness of the contralateral versus ipsilateral approaches during uniportal and biportal endoscopic PCF. Preliminary findings suggest that approach angles and the extent of facet resection might significantly impact clinical outcomes such as pain relief, functional recovery, and complication rates. The contralateral approach, potentially allowing a more favorable angle for decompression with minimal facet resection, might reduce the risk of postoperative instability and preserve the structural integrity of the cervical spine. Therefore, this study aims to conduct a systematic comparison of the clinical and radiographic outcomes of the contralateral and ipsilateral approaches in uniportal and biportal endoscopic PCF. By examining variables such as foraminal expansion rates, facet joint preservation, and postoperative clinical scores, this research could illuminate optimal surgical techniques and influence practice guidelines, thereby improving patient outcomes in the treatment of cervical radiculopathy due to foraminal stenosis. Material and Methods: This retrospective cohort study included 147 patients with cervical foraminal stenosis who underwent endoscopic spine surgery from single private specialized hospital. The patients were divided into three groups: uniportal (n = 54), contralateral biportal (n = 44), and ipsilateral biportal (n = 49). Clinical outcomes were assessed using operative time, complications, Macnab grading, and VAS scores for neck and arm pain. Radiographic outcomes included midforaminal diameter (MFD), distal foraminal diameter (DFD), midforaminal height (MFH), facet joint resection rate, and sagittal alignment change. Statistical analysis was performed to compare the outcomes between the groups using appropriate tests. Results: All surgical procedures resulted in similar clinical outcomes in terms of arm and neck pain reduction, operative time, and postoperative Macnab grading. Complication rates were low and comparable across all groups (p = 0.553). The mean reduction in VAS scores for arm pain was 4.81 ± 1.06 in the uniportal group, 4.55 ± 1.02 in the contralateral biportal group, and 4.73 ± 1.00 in the ipsilateral biportal group (p = 0.368). The mean reduction in VAS scores for neck pain was 4.04 ± 1.46 in the uniportal group, 3.30 ± 1.69 in the contralateral biportal group, and 3.84 ± 1.52 in the ipsilateral biportal group (p = 0.107). Most radiographic parameters, including foraminal expansion ratios for MFD, DFD, and MFH, were similar among the groups. However, the contralateral biportal group had a significantly narrower facet joint width postoperatively (p = 0.014) and a higher, though not statistically significant, facet joint resection rate (p = 0.054). Conclusion: All three endoscopic surgical approaches provided similar clinical outcomes and radiographic improvements for cervical foraminal stenosis. The contralateral biportal approach resulted in a significantly narrower facet joint width postoperatively and a higher facet joint resection rate compared to the uniportal and ipsilateral biportal approaches. Long-term outcomes should be followed to further assess the benefits and potential risks of these surgical techniques.

ID: AGS26/1732

A281: Endoscopic discectomy reduces subsequent fusion: a multinational, real-world comparative study

Sung Huang Laurent Tsai 1

1 Taipei Medical University Hospital, Taipei, Taiwan

Introduction: The choice between percutaneous endoscopic lumbar discectomy (PELD) and open lumbar discectomy (OLD) for managing lumbar disc herniation (LDH) remains debated. While both techniques achieve comparable short-term outcomes, their long-term effects, particularly regarding the progression to spinal fusion, are underexplored. This study evaluates the risk of subsequent spinal fusion associated with PELD versus OLD and examines how patient demographics and comorbidities influence these outcomes. This study evaluates the long-term risk of spinal fusion following PELD versus OLD, with a focus on subgroup-specific differences, including age, race, BMI, and comorbidities. Material and Methods: Using the TriNetX Research Network, we analyzed data from 123,405 patients diagnosed with LDH between 2000 and 2023, including 93,853 PELD and 29,552 OLD cases. After propensity score matching, a 1:1 cohort was created to ensure baseline comparability. Patients were followed for up to 20 years to assess the risk of subsequent spinal fusion. Kaplan-Meier survival analyses and Cox proportional hazard models were employed to estimate hazard ratios (HR) and evaluate subgroup outcomes. Results: PELD significantly reduced the risk of spinal fusion compared to OLD over a 20-year follow-up (HR 0.706; 95% CI: 0.656-0.760). Subgroup analysis revealed consistent benefits across younger patients (20-45 years), high-BMI individuals, and White and African American populations. However, PELD demonstrated limited effectiveness in Asian patients and those with alcohol-related diseases. Mechanistically, PELD's preservation of paraspinal tissues and reduction of inflammation may explain its long-term advantages in minimizing adjacent segment degeneration and instability. Conclusion: PELD offers substantial long-term benefits in reducing the need for spinal fusion, particularly for younger, obese, and racially diverse patients. These findings support PELD as a personalized surgical approach that minimizes invasive burden while optimizing outcomes. Further research should investigate mechanisms underlying subgroup-specific differences and evaluate additional long-term endpoints, including functional recovery and quality of life.

ID: AGS26/2339

A282: Validation of a novel grading scale (VIEW Score) for intraoperative visualization in endoscopic spine surgery: a multi-center reliability and clinical utility study

Vit Kotheeranurak 1 , Surachat Jaroenwareekul 1 , Wicharn Yingsakmongkol 1 , Worawat Limthongkul 1 , Jin-Sung Kim 2 , Don Park 3 , Christoph Siepe 4 , Javier Quillo-Olvera 5 , Weerasak Singhatanadgige 1

1 Department of Orthopaedics, Faculty of Medicine, Chulalongkorn University and King Chulalongkorn Memorial Hospital, Bangkok, Thailand

2 Seoul St. Mary’s Hospital, College of Medicine, The Catholic University of Korea, Seoul, South Korea

3 Department of Orthopaedic Surgery, UC Irvine School of Medicine, Irvine, CA, United States

4 Centre for Spinal Surgery, Schoen Clinic München, Harlaching, Munich, Germany

5 The Brain and Spine Care, Minimally Invasive Spine Surgery Group, Queretaro, Mexico

Introduction: The clarity of the surgical field is a critical determinant of safety in endoscopic spine surgery, yet its assessment remains subjective and unstandardized. This lack of objectivity hinders reproducible research, standardized training, and quality assurance. To address this gap, we developed and aimed to validate a novel grading scale, the Visualization in Intraoperative Endoscopy Workspace (VIEW) Score, for the objective assessment of the surgical field. Material and Methods: The VIEW Score, a 4-point ordinal scale (0 = Excellent to 3 = Poor), was developed through expert consensus to evaluate bleeding, clarity of key anatomical structures, and impact on surgical progress. In this multi-center reliability study, five experienced endoscopic spine surgeons independently scored 40 de-identified, 5-second video clips from lumbar procedures. After a 4-week washout period, the evaluation was repeated to assess intra-rater reliability. A two-way random effects model for absolute agreement was used to calculate the Intraclass Correlation Coefficient (ICC). Results: The VIEW Score demonstrated excellent inter-rater reliability for the average score of the five raters (ICC = 0.963; 95% Confidence Interval: 0.940 - 0.979; p < 0.001). The reliability for a single rater was found to be moderate (ICC = 0.697). Furthermore, intra-rater reliability was good, with an average ICC of 0.889 across all raters. Conclusion: The VIEW Score is a simple and highly reliable tool for the objective assessment of visualization in endoscopic spine surgery. Its excellent inter-rater and good intra-rater reliability confirm its utility as a standardized metric for surgical training, quality assurance, and future clinical research comparing surgical techniques and technologies.

ID: AGS26/2486

A283: Comparison of unilateral biportal endoscopic lumbar interbody fusion (UBE-LIF) and conventional OPEN TLIF in degenerative lumbar spine disease: a radiological, clinical and laboratory study

Yunus Şamil Gönüldenk 1 , Mustafa Özyıldıran 2 , Abdullah Merter 1

1 Ankara University, Ankara, Turkey

2 Sandıklı State Hospital, afyon, Turkey

Background: Parallel to the advancement of endoscopic techniques, unilateral biportal endoscopic lumbar interbody fusion (UBE-LIF) has become a popular surgical method in the treatment of degenerative lumbar instability. This study aimed to compare the clinical and radiological outcomes of UBE-LIF and conventional open TLIF techniques. Materials and Methods: Patients with single-level degenerative lumbar spondylolisthesis who underwent either UBE-LIF or conventional open TLIF at our institution between January 2021 and May 2024 were included in this single-center study. Clinical outcomes were assessed using Visual analogue scale (VAS) scores, Japanese Orthopaedic Association (JOA) scores, and Oswestry Disability Index (ODI), both preoperatively and postoperatively. Operative time, length of hospital stay, laboratory markers of muscle injury (hemoglobin, CRP, ESR, WBC, myoglobin), and radiological fusion rates were also evaluated. Results: Among the 100 patients who met the inclusion criteria, UBE-LIF was performed in 76 and conventional open TLIF in 24. JOA, VAS-leg, VAS-low back pain, and ODI scores significantly improved in both groups postoperatively (p < 0.001). Clinical scores did not differ significantly between the groups at any time point (p > 0.05). Postoperative blood transfusion was required in 3 patients (11.5%) in the conventional group, whereas none was needed in the UBE-LIF group (p = 0.003). Length of hospital stay was also significantly shorter in the UBE-LIF group (45.4 vs. 98.6 hours, p < 0.001). The difference in 12-month postoperative fusion rates between UBE-LIF (100%) and conventional open TLIF (92.3%) was not significant (p = 0.07). Postoperative progression of Goutallier grade (paraspinal muscle degeneration) was not observed in the UBE-LIF group, whereas it occurred in 21 patients (80.8%) in the conventional open TLIF group (p < 0.001). 1 Conclusion: Clinical outcomes improved similarly with both techniques. However, postoperative muscle degeneration, intraoperative blood loss, and length of hospital stay were more favorable in the UBE-LIF group. UBE-LIF appears to be a strong alternative to conventional open TLIF.

Keywords: Degenerative lumbar instability, degenerative lumbar spine disease, unilateral biportal endoscopy (UBE), lumbar interbody fusion (LIF), UBE-LIF, conventional open surgery, conventional open TLIF

ID: AGS26/3186

A284: Comparison of one-year clinical outcomes between unilateral biportal and full endoscopic lumbar interbody fusion: a single-surgeon retrospective cohort study

Abhirat Suepsing 1 , Akarawit Asawasaksakul 2 , Khananut Jaruwanneechai 1 , Warayos Trathitephun 3 , Boonserm Pakdeenit 4 , Siravich Suvithayasiri 5

1 Warinchamrab Hospital, Ubonratchathani, Thailand

2 Ramkhamhaeng Hospital, Bangkok, Thailand

3 Chulabhorn Hospital, Bangkok, Thailand

4 Bangkok Pattaya Hospital, Pattaya, Thailand

5 Thonburi Hospital, Bangkok, Thailand

Introduction: Endoscopic lumbar interbody fusion has become a popular minimally invasive approach for treating degenerative lumbar spine disease. Among available techniques, unilateral biportal endoscopic lumbar interbody fusion (UBE-LIF) and full-endoscopic lumbar interbody fusion (FE-LIF) are commonly performed. However, direct comparisons between these two techniques remain limited, especially in studies conducted under a single-surgeon setting. This study aimed to compare the clinical outcomes and fusion results between UBE-LIF and FE-LIF at one-year follow-up. Material and Methods: This retrospective study included patients who underwent single-level UBE-LIF or FE-LIF performed by a single surgeon. Clinical outcomes were assessed using the Oswestry Disability Index (ODI) and Visual Analog Scale (VAS) for back and leg pain. Assessments were conducted preoperatively and postoperative at day 1, 1 month, 6 months, and 12 months. Perioperative parameters including operative time, estimated blood loss (EBL), drain output, and length of stay (LOS) were recorded. Radiographic fusion at 12 months was evaluated using the CT scan and the Bridwell classification, with grades I and II were considered fusion. Results: A total of 92 patients were included, with 49 patients in the UBE-LIF group and 43 in the FE-LIF group. At 12 months ODI scores significantly improved in both groups, without statistical difference between UBE-LIF (10.49 ± 2.06) and FE-LIF (10.28 ± 2.82; p = 0.57). VAS scores for back pain also showed comparable improvement in both groups. Fusion rates based on Bridwell grade I-II were 100% for UBE-LIF and 95.35% for FE-LIF, without statistical significance (p = 0.20). Perioperative outcomes including EBL, drain output, and LOS were not difference, although operative time was slightly longer in the FE-LIF group (143.98 ± 11.68 and 139.39 ± 7.68 minutes, p = 0.001). Conclusion: Both UBE-LIF and FE-LIF have demonstrated comparable and excellent clinical and radiological outcomes at the one-year follow-up mark. Within the limitations of this retrospective, single-surgeon study, our findings suggest that patients in both cohorts achieved similar, significant improvements in functional scores and pain reduction. With high fusion rates observed in both groups, we assert that both techniques are effective and reliable. Therefore, the selection of either UBE-LIF or FE-LIF may be judiciously guided by the specific nuances of the case and the surgeon's individual preference and expertise.

ID: AGS26/418

A285: Comparative analysis of full endoscopic unilateral laminotomy with bilateral decompression using 3D versus 2D spine endoscopy for the treatment of spinal stenosis: a retrospective study

Yuhsuan Chung 1

1 Show Chwan Memorial Hospital, Taiwan, Changhua, Taiwan

Introduction: Lumbar spinal stenosis (LSS) is a common degenerative disease characterized by the narrowing of the spinal canal, leading to nerve compression and neurogenic claudication. This study aimed to compare the efficiency, safety, and surgical outcomes of unilateral laminotomy with bilateral decompression (ULBD) using a novel three-dimensional (3D) endoscopic technology versus the conventional two-dimensional (2D) endoscopic system. Material and Methods: This study included 62 consecutive patients who underwent full endoscopic ULBD for lumbar spinal stenosis at a tertiary medical center between January 2023 and December 2024. Patients were retrospectively assigned to either the 3D Endoscopy Group or the 2D Endoscopy Group. Data collected included demographic information, comorbidities, Charlson comorbidity index (CCI), operative details (surgery time, time to drill, time to yellow ligament, blood loss), and functional outcomes (Visual Analog Scale for pain, Oswestry Disability Index, patient satisfaction scores) evaluated at 1, 3, 6, and 12 months postoperatively. Statistical analysis was performed using SPSS 26.0, with p < 0.05 considered statistically significant. Results: In terms of surgical efficiency, the 3D group had significantly shorter times from skin incision to drill positioning (p = 0.047) and from skin incision to yellow ligament exposure (p < 0.001), demonstrating superior procedural navigation and speed. While intraoperative blood loss was lower in the 3D group, this difference was not statistically significant (p = 0.079). The 3D endoscopic system also demonstrated significant advantages over the traditional 2D system in clinical outcome measures. The 3D group showed significantly lower back pain VAS scores at 6 months postoperatively compared to the 2D group (1.935 vs. 2.355, p = 0.050). Additionally, at 12 months postoperatively, the 3D group exhibited significantly lower ODI percentages (0.194 vs. 0.232, p = 0.050), reflecting better long-term pain relief and functional recovery. Although leg VAS scores in the 3D group were slightly higher than in the 2D group at 1, 3, and 12 months postoperatively, these differences did not reach statistical significance (p > 0.05). Conclusion: The novel 3D endoscopic system significantly improves surgical efficiency by reducing the time from skin incision to drill positioning and yellow ligament exposure, demonstrating enhanced navigational accuracy during ULBD procedures. This optimization of surgical exposure and procedural workflow leads to better long-term pain relief and functional outcomes compared to conventional 2D endoscopy. By expediting the critical early stages of surgery and improving decompression processes, the 3D endoscopic system not only enhances surgical efficiency and safety but also effectively promotes postoperative pain relief and functional recovery, demonstrating its potential for widespread application in endoscopic spine surgery.

ID: AGS26/634

A286: Clinical comparative study of video-assiste thoracoscopic surgery and uniportal endoscopic spine surgery in the treatment of thoracic disc herniation

Kunfeng Song 1 , Daniel Rosenthal 2 , Honggang Zhou 1 , Hejun Yang 1 , Mingkui Shen 1

1 Third People's Hospital of Henan Province, Zhengzhou, China

2 Hochtaunus-Kliniken, Bad Homburg, Germany

Introduction: Minimally invasive treatment techniques for thoracic disc herniation are constantly evolving. Video-Assisted Thoracoscopic Surgery (VATS) has become a mature option, while Uniportal Endoscopic Spine Surgery (UESS), as a new minimally invasive technique, lacks sufficient comparative studies on its clinical efficacy with VATS. This study aimed to systematically compare the surgical efficacy, safety, and economic benefits of VATS and UESS in the treatment of thoracic disc herniation, clarify the clinical advantages and application scenarios of the two techniques, explore the potential direction of technical integration, and verify the application value of the Enhanced Recovery After Surgery (ERAS) concept in minimally invasive thoracic spine surgery. Material and Methods: A multicenter, prospective randomized controlled study was conducted. From January 2019 to December 2023, 520 patients with thoracic disc herniation from 5 tertiary hospitals nationwide were enrolled and randomly divided into the VATS group (260 cases) and UESS group (260 cases) at a 1:1 ratio. A stratified randomization method was used, with lesion segments (upper thoracic T1-T4, middle thoracic T5-T8, lower thoracic T9-T12), herniation types (soft herniation, calcified herniation), patient age (< 45 years, 45-60 years, > 60 years), and comorbidities (hypertension, diabetes) as stratification factors to ensure balanced baseline data between groups. The evaluation indicators included: ① Surgery-related indicators (operation time, intraoperative blood loss, incision length); ② Clinical efficacy (VAS pain score, mJOA neurological function score at 1 week, 3 months, and 1 year postoperatively); ③ Long-term prognosis (1-year postoperative recurrence rate, complication rate); ④ Economic indicators (hospitalization cost, postoperative rehabilitation cost); ⑤ Patient-reported outcomes (SF-36 quality of life score). Innovatively, three-dimensional kinematic analysis was introduced to evaluate spinal stability, advanced intraoperative neuromonitoring system (transcranial motor evoked potentials + somatosensory evoked potentials) was used to reduce the risk of nerve injury, and diffusion tensor imaging was applied to assess spinal cord microstructure changes. Meanwhile, both groups systematically implemented the ERAS protocol (including preoperative nutritional support, intraoperative temperature protection, postoperative multimodal analgesia, and early ambulation guidance). Results: All patients in both groups completed follow-up (follow-up rate 100%) with no loss to follow-up. ① Surgery-related indicators: The average operation time in the UESS group (92.3 ± 15.7 min) was significantly shorter than that in the VATS group (108.6 ± 18.2 min) (t = 8.72, p < 0.001); the average intraoperative blood loss in the UESS group (35.6 ± 12.8 mL) was significantly less than that in the VATS group (89.2 ± 21.5 mL) (t = 26.35, p < 0.001); the incision length in the UESS group (1.2 ± 0.3 cm) was significantly smaller than that in the VATS group (3.5 ± 0.8 cm) (t = 38.16, p < 0.001). ② ERAS application effect: After implementing the ERAS protocol, the average length of hospital stay was reduced from 7.2 days (in the preoperative pilot study) to 4.3 days in the UESS group, and from 8.1 days to 5.1 days in the VATS group, with significant reductions in both groups (all p < 0.001); the ERAS protocol had a more significant effect on elderly patients (> 60 years old), with the postoperative complication rate in this subgroup decreasing from 12.5% (16/128) to 7.2% (9/125) (χ 2 = 4.23, p < 0.05), and the incidence of major complications (pulmonary infection, pleural effusion) all decreased. ③ Learning curve and long-term efficacy: After completing 30 surgical operations, the operation time in the UESS group shortened from the initial (115.4 ± 18.6) min to a stable level (88.2 ± 10.5) min, with a flatter learning curve than that in the VATS group; at 1 year postoperatively, the mJOA scores of both groups significantly improved (UESS group: 7.8 ± 1.6 points preoperatively → 11.2 ± 1.3 points postoperatively; VATS group: 7.6 ± 1.5 points preoperatively → 10.8 ± 1.4 points postoperatively), with no significant difference between groups (p > 0.05). However, the scores of “bodily pain” and “physical function” dimensions in the SF-36 score of the UESS group were significantly higher than those of the VATS group (all p < 0.05); there was no significant difference in the recurrence rate between the two groups (UESS group: 1.5% (4/260), VATS group: 1.9% (5/260), p > 0.05). Conclusion: The UESS technique breaks the traditional perception that “higher minimally invasive degree leads to longer operation time”. It is significantly superior to VATS in terms of operation time, intraoperative blood loss, and incision trauma, with a flat learning curve that is easier to promote. The systematic application of the ERAS concept can further shorten the length of hospital stay for patients in both groups, especially reducing the risk of postoperative complications in elderly patients and significantly improving prognosis. The two techniques have comparable effects in long-term neurological function recovery and recurrence rate, but UESS has more advantages in improving patients' subjective quality of life. In the future, the technical integration model of “UESS precise decompression + VATS spinal stability reconstruction” can be explored to provide a more optimized plan for the minimally invasive treatment of thoracic disc herniation.

ID: AGS26/698

A287: Intraoperative local methylprednisolone in single-position lateral decubitus oblique lumbar interbody fusion (OLIF) for postoperative anterior thigh pain prevention: a randomized controlled trial

Kantadon Prateepswangwong 1 , Worawat Limthongkul 1 , Vit Kotheeranurak 1 , Weerasak Singhatanadgige 1 , Wicharn Yingsakmongkol 1

1 Department of Orthopedics, Faculty of Medicine, Chulalongkorn University, Bangkok, Thailand

Introduction: Oblique lumbar interbody fusion (OLIF) is a minimally invasive lumbar fusion surgery in which interbody cage is inserted through an oblique lateral approach, positioned anterior to psoas major muscle (PMM). Although this approach reduces the risk of complications from direct injury to PMM or lumbar plexus, anterior thigh pain (ATP) remains one of the most common postoperative problems. Hypothesis of this study is that locally administered steroid, methylprednisolone, may reduce the inflammation of PMM and consequently prevent ATP. Material and Methods: This study was a prospective, double-blinded, randomized controlled trial in single institute. Patients with degenerative spine disease who underwent single level OLIF using single lateral position with percutaneous pedicle screw fixation were randomized into two groups. A 3x3 cm sterile, absorbable gelatin sponge soaked with 2 cc of 80mg methylprednisolone (intervention; I group) or 2 cc of saline (control; C group), was placed anterior to PMM (OLIF cage corridor). The primary outcome was postoperative ATP, measured using the visual analog scale (VAS) at nine time points: 0, 1, and 2 days; 1 and 2 weeks; 1, 3, and 6 months; and 1 year postoperatively. Secondary outcomes consist of muscle damage cytokines: creatinine (Cr), creatinine phosphokinase (CK), aspartate aminotransferase (AST) and lactate dehydrogenase (LDH); inflammatory cytokines: erythrocyte sedimentation rate (ESR), C-reactive protein (CRP), interleukin-1β (IL-1β), interleukin-6 (IL-6) and tumor necrosis factor (TNF-α); cross-sectional area of psoas major muscle (PMM-CSA) from magnetic resonance imaging (MRI); and morphine consumption during admission. Results: Forty patients were included, with 20 in each group. The most common surgical level was L4/5 (95% in I group and 85% in C group). There were no statistically significant differences in demographic data, operative time, or psoas retraction time. Both groups showed significant clinical improvement, with no significant difference between them. No statistically significant differences were found in any biomarkers or morphine consumption. PMM-CSA was slightly lower in I group (955.8 mm 2 ) compared to C group (1046.7 mm 2 ), albeit not statistically significant. However, when comparing participants with and without ATP, PMM retraction time was identified as the only significant factor (37.5 min in the ATP group vs. 29.1 min in the no-ATP group; p = 0.023). Conclusion: Local methylprednisolone in the OLIF corridor was not effective as an intervention to prevent ATP. Additionally, serum muscle damage markers and inflammatory cytokines could not be used as indicators to predict ATP. Based on our study, the factor that related to ATP is PMM retraction time of more than 30 minutes.

ID: AGS26/77

A288: Minimally invasive lateral lumbar interbody fusion showed a lower incidence of radiologic adjacent segment pathology than conventional posterior lumbar interbody fusion

Jaewan Soh 1 , Jae Chul Lee 2

1 Hanyang University Guri Hospital, College of Medicine, Guri, South Korea

2 Department of Orthopedic Surgery, Soonchunhyang University Seoul Hospital, College of Medicine, Seoul, South Korea

Introduction: Radiologic adjacent segment pathology (R-ASP) is a significant consequence following lumbar spinal fusion, potentially resulting from compromised integrity of posterior structures. Few studies have directly compared the incidence of R-ASP between minimally invasive lateral lumbar interbody fusion (LLIF) and conventional posterior lumbar interbody fusion (PLIF). To analyze risk factors for R-ASP and compare clinical outcomes between minimally invasive LLIF supplemented with percutaneous pedicle screw fixation and open conventional PLIF. Material and Methods: This study included 107 patients who underwent one- or two-segment spinal fusion for degenerative lumbar disease. Fifty-one patients underwent minimally invasive LLIF, while 56 received conventional PLIF. Factors related to the occurrence of R-ASP were investigated by analyzing demographic profiles, radiological results, and clinical outcomes. The correlation between clinical results was analyzed based on fusion methods and the presence of R-ASP. We analyzed patient-related factors, preoperative spinal diagnosis, number of fused segments, and radiologic evaluations. Clinical outcomes were also assessed. Cox regression survival analysis was performed to determine risk factors for R-ASP. The annual incidence and cumulative survival rate of R-ASP were calculated using the life-table method and Kaplan-Meier survival curve. Results: Cox proportional hazards regression analysis identified three significant risk factors for R-ASP: PLIF over LLIF (p = 0.028, Hazard Ratio (HR) = 2.321, 95% Confidence Interval (CI) = 1.096-4.913), postoperative PI-LL mismatch ≥ 10° (p = 0.022, HR = 2.280, 95% CI = 1.126-4.617), and preoperative facet arthropathy grade ≥ 2 (p = 0.016, HR = 3.491, 95% CI = 1.266-9.629). The predicted incidence of R-ASP was 48.7% (95% CI, 42.1-55.2) at 5 years post-fusion, and 80.7% (95% CI: 73.0-88.5) at 8 years. Clinical outcomes showed that the final VAS for lower back pain was significantly lower in patients who underwent the LLIF and in the patients who did not develop R-ASP. Conclusion: Minimally invasive fusion techniques that preserve posterior structures may slow the progression of degenerative changes in adjacent segments. To reduce R-ASP, preoperative assessment of facet degeneration and adequate restoration of sagittal balance during surgery are crucial considerations.

OP33: Patient Factors That Affect Complications

ID: AGS26/1189

A289: Efficacy of proximal junctional tethering in spinal fusion surgery for preventing proximal junctional kyphosis and proximal junctional failure: A meta-analysis

Yu-Chi Hsu 1 , Hsuan-Tung Lee 1 , Ying Fong Su 1 , Yang-Ching Chen 2 , Serena Hu 3 , Ching-Chi Hsu 4 , Pei-I Tsai 5 , Wei-Bin Hsu 6 , Den-Tai Lin 5 , Ching-Yu Lee 7 , Tsung-Jen Huang 7 , Tan Nguyen 8 9 , Max Meng-Huang Wu 7

1 School of Medicine, College of medicine, Taipei Medical University, Taipei, Taiwan

2 Department of Family Medicine, Wan Fang Hospital, Taipei Medical University, Taipei City, Taiwan , 3 Department of Orthopedic Surgery, Stanford University School of Medicine, Taipei, Taiwan

4 Department of Mechanical Engineering, National Taiwan University of Science and Technology, Taipei, Taiwan

5 Biomedical Technology and Device Research Labs, Industrial Technology Research Institute, Hsinchu, Taiwan

6 Department of Orthopedics, Taipei Medical University Hospital, Taipei City, Taiwan

7 Department of Orthopedics, Taipei Medical University Hospital, Taipei, Taiwan

8 Department of Orthopedics, Faculty of Medicine, Can Tho University of Medicine and Pharmacy, Can Tho, Viet Nam, Can Tho, Viet Nam

9 The International Ph.D. Program in Medicine, College of Medicine, Taipei Medical University, Taipei, Taiwan

Introduction: Spinal fusion surgery is effective for treating various adult spinal deformities. However, spinal fusion surgery is associated with the risk of adjacent segment disease (ASD; 5%-30%), particularly proximal junctional kyphosis (PJK) and proximal junctional failure (PJF). Proximal junctional tethering (PJT) has become a popular technique owing to increasing evidence that it can decrease the rate of PJK or PJF. Material and Methods: A literature search was conducted using PubMed, Embase, and Cochrane Library. Ten eligible studies were identified. These studies were predominantly retrospective in nature and compared the incidence of PJK or PJF in adults undergoing spinal fusion surgery with or without PJT. Risk of bias was assessed using the Newcastle-Ottawa scale. All outcomes were analyzed using R software (version 4.4.1). Results: We included 9 retrospective cohort studies and 3 propensity-score-matched analyses; these studies comprised 1500 patients. PJT was associated with a significant decrease in the rates of PJK (risk ratio [RR] = 0.58, 95% CI = 0.46-0.73) and PJF (RR = 0.46, 95% CI = 0.28-0.75) compared with control. Subgroup analysis results revealed no significant difference in ASD rates between geographical locations, between surgical indications, between tethering with and without crosslinks, and between specific tethering techniques. Conclusion: PJT significantly reduces the rates of both PJK and PJF in adults undergoing spinal fusion surgery.

ID: AGS26/1268

A290: Comparing the prognostic value of standard, modified, calibrated, and T2-weighted vertebral bone quality scores for cage subsidence following single-level anterior cervical discectomy and fusion: a retrospective observational study

Omar Lubbad 1 , Laila Lubbad 2 , Suzanne Murphy 3 , Nektarios K Mazarakis 4

1 Brighton and Sussex Medical School, Brighton, United Kingdom

2 Royal College of Surgeons in Ireland Medical School, Manama, Bahrain

3 Department of Surgery, Royal College of Surgeons in Ireland, Brighton, Ireland

4 Department of neurosurgery, Royal Sussex County Hospital, Brighton, United Kingdom

Introduction: Magnetic resonance imaging (MRI)-derived vertebral bone quality (VBQ) scores have emerged as potential predictors of cage subsidence after anterior cervical discectomy and fusion (ACDF). Several modifications of the VBQ method have been proposed, but their relative predictive performance remains uncertain. Material and Methods: We retrospectively analyzed 114 patients undergoing single-level ACDF (71 without subsidence, 43 with subsidence). Demographic, clinical, surgical, and imaging variables were recorded. Bone quality was assessed using VBQ, calibrated VBQ (cVBQ), modified VBQ (MVBQ), posterior VBQ variants, and T2-weighted signal-based measures (T2CSF, T2FAT). Logistic regression was used to assess associations with subsidence. Predictive performance was evaluated with ROC analysis, DeLong’s test, calibration slopes and intercepts, McNemar’s test, net reclassification improvement (NRI), and Bland-Altman agreement. Results: Baseline demographic and comorbidity profiles were comparable between groups, although subsidence occurred more frequently at C3/4 (p = 0.014). On univariate regression, higher VBQ-derived scores and T2CSF were all significantly associated with subsidence (odds ratios 1.7-5.1, all p < 0.05), with VBQ showing the strongest effect (OR 5.14, 95% CI 2.42-10.94, p < 0.001). Multivariate analysis adjusting for age, sex, cage material, and plate use did not identify independent demographic or surgical predictors. ROC analysis showed that cVBQ achieved the highest discrimination (AUC 0.824, 95% CI 0.744-0.903), followed by cPVBQ and cMVBQ (AUCs 0.806-0.814), while T2CSF and T2FAT performed poorly (AUC 0.672-0.663). Pairwise DeLong testing confirmed that VBQ-based scores significantly outperformed T2CSF (p ≤ 0.04), but not each other. McNemar’s test demonstrated significant classification differences between VBQ, cVBQ, and MVBQ (all p < 0.05). NRI showed modest improvement for cVBQ over VBQ (+0.26), whereas T2CSF consistently reduced classification performance. Calibration analysis indicated acceptable slopes for VBQ-derived metrics (1.20-1.64) but poor calibration for T2CSF (slope 4.76). Bland-Altman analysis confirmed close agreement between VBQ, cVBQ, and MVBQ, with wide bias and poor concordance for T2CSF. Conclusion: MRI-derived bone quality scores, particularly cVBQ, demonstrate strong predictive performance for cage subsidence after ACDF, independent of patient or surgical factors. In contrast, T2-based measures show poor discrimination and calibration. These findings support cVBQ as the most robust MRI-derived predictor of bone quality-related implant failure.

ID: AGS26/1354

A291: Preoperative malnutrition is associated with increased postoperative complications following lumbar fusion: a propensity-matched analysis

Ali Mehaidli 1 , Muaaz Wajahath 2 , Jad Lawand 3 , Omar Abdalla 4 , Haytham Alqasmi 4 , Ahmad Almaat 4 , Alqasim Elnaggar 4 , Lorenzo Deveza 5 , Noah Hodson 1 , Kevin Taliaferro 1 , Ryan Berger 1

1 Department of Orthopaedic Surgery, Henry Ford Hospital, Detroit, United States

2 Michigan State University College of Human Medicine, East Lansing, United States

3 Department of Orthopaedic Surgery, John Peter Smith Health Network, Fort Worth, United States

4 Wayne State University School of Medicine, Detroit, United States

5 Department of Orthopaedic Surgery, Baylor College of Medicine, Houston, United States

Introduction: Lumbar fusion (LF) is a common surgical intervention used to manage a variety of spinal conditions, including degenerative disc disease, spondylolisthesis, and spinal instability. While effective for improving pain and function in select patients, LF is also associated with significant short- and long-term risks, including surgical site infection, pseudoarthrosis, readmission, and the need for revision procedures. One increasingly recognized yet under-addressed modifiable risk factor is preoperative malnutrition, which has been previously associated with impaired wound healing and increased susceptibility to infection. This study aimed to evaluate the association between preoperative malnutrition and postoperative outcomes after LF using a large, multicenter dataset with long-term follow-up. Material and Methods: A retrospective cohort analysis was performed using the TriNetX US Collaborative Network, including adults undergoing LF between 2003 and 2023. Malnutrition was defined as serum albumin < 3.5 g/dL or leukocyte count < 1,500/mm 3 within one year before surgery. Patients were matched 1:1 by propensity scoring for demographics, comorbidities, body mass index, and surgical approach, yielding two cohorts of 20,693 each. Outcomes were assessed at 90 days and 5 years postoperatively. Hazard ratios (HRs) with 95% confidence intervals (CIs) were estimated using Cox proportional hazards models. Results: Malnourished patients experienced significantly higher 90-day rates of transfusion (8.4% vs 4.5%; HR 1.87, 95% CI 1.73-2.02), wound complications (2.6% vs 1.5%; HR 1.70, 95% CI 1.48-1.95), infection (3.2% vs 2.3%; HR 1.37, 95% CI 1.22-1.54), and hospital readmission (17.3% vs 13.8%; HR 1.26, 95% CI 1.20-1.32). At 5 years, malnourished patients had higher rates of pseudoarthrosis (25.8% vs 20.5%; HR 1.32, 95% CI 1.26-1.38), revision surgery (16.2% vs 11.9%; HR 1.41, 95% CI 1.34-1.49), and compression fractures (5.8% vs 3.7%; HR 1.60, 95% CI 1.46-1.76). All-cause mortality was lower in the malnourished cohort (4.3% vs 5.8%; HR 0.75, 95% CI 0.69-0.82), likely reflecting selection bias. Conclusion: Preoperative malnutrition is independently associated with significantly higher short- and long-term complication rates after LF, including infection, transfusion, readmission, pseudoarthrosis, and revision surgery. These findings underscore the importance of routine nutritional screening and optimization as part of preoperative planning in spine surgery. Incorporating nutritional assessment into perioperative risk stratification may reduce morbidity, improve surgical success, and decrease healthcare utilization.

ID: AGS26/1693

A292: Presentations, management strategies, and outcomes of intracardiac cement embolism following spinal vertebral augmentation: a systematic review of case reports

Omar Lubbad 1 , Laila Lubbad 2 , Guiseppe Morassi 3 , Nektarios K Mazarakis 3

1 Brighton and Sussex Medical School, Brighton, United Kingdom

2 Royal College of Surgeons in Ireland Medical School, Manama, Bahrain

3 Department of neurosurgery, Royal Sussex County Hospital, Brighton, United Kingdom

Introduction: Intracardiac cement embolism (ICE) is a rare, but potentially life-threatening complication of percutaneous vertebroplasty (PV), percutaneous kyphoplasty (PK), and cement augmented pedicle screw (CAPS) fixation. This systematic review aims to summarise and compare patient presentations, management approaches, and clinical outcomes of ICE. Material and Methods: PRISMA guidelines were followed. Standard medical databases such as MEDLINE, Embase, Pubmed, and Cochrane Library were searched to find case reports or series on patients with ICE following PV, PK, or CAPS. Data on patient demographics, embolus location, diagnostic modalities, management strategies, and outcomes were recorded. Results: 67 case reports and 2 case series were included with 71 patients in total. 67 patients presented symptomatically with the most common complaints being chest pain and dyspnoea, the other 4 cases were incidentally discovered. Emboli were predominantly found in the right heart and were primarily discovered via echocardiography or computed tomography. Open-heart surgery was the most adopted management strategy (46 cases, 100% retrieval success), especially in complicated symptomatic cases. Endovascular (11 cases, 72% retrieval success) or conservative approaches (14 cases, 2 mortalities) were reserved for asymptomatic cases with smaller emboli or patients not deemed fit for surgery. Reported mortality was low across all patients, but delayed diagnosis was associated with more severe cardiopulmonary complications. Notably, 28 patients presented within the first day postoperatively, indicating an early peak in presentation. Conclusion: Open-heart surgery seems to be the preferred, most effective intervention for ICE, while endovascular and conservative management were viable in select cases. However, due to the potential for delayed complications, conservative management should be avoided. Earlier diagnosis is imperative for improving patient outcomes and preventing severe complications.

ID: AGS26/2491

A293: Complications after anterior lumbar interbody fusion in patients with prior abdominal surgery

Hannah Shelby 1 , Sarah Bergren 1 , Aidan Lindgren 1 , Daniel Rusu 1 , Mirbahador Athari 1 , Jeffrey C. Wang 1 , Raymond Hah 1 , Kiran Alluri 1

1 Department of Orthopaedic Surgery, Keck School of Medicine of the University of Southern California, Los Angeles, United States

Introduction: Anterior lumbar interbody fusion (ALIF) is a widely utilized surgical technique for the treatment of numerous spinal pathologies. The ALIF approach necessitates mobilizing the intra-abdominal viscera and vasculature, understandably, there is a unique morbidity profile associated with this approach. Adhesions, scarring, and altered anatomy following procedures such as laparotomy, cesarean section, colectomy, or other major abdominal operations may increase operative difficulty and elevate the risk of complications. Existing literature often cites significant prior abdominal surgery with adhesions as a relative contraindication to ALIF. There is a paucity of data regarding the rates of ALIF post-op complications following specific abdominal surgeries and the nature of these complications. This study aims to investigate this clinical relationship. Material and Methods: A retrospective cohort study was performed using the PearlDiver Mariner national insurance claims database. Of the 1,123,841 ALIF patients from 2010-2024, patients were stratified according to history of prior cesarean section (CSEC; n = 5,085), prior non-obstetric abdominal surgery (ABD; n = 49,734), any prior abdominal surgery (ANYABD; n = 54,819), and no prior abdominal surgery (NOABD; n = 1,069,022). Outcomes included complications within 30 days, 90 days, and one year. Complication subtypes included vascular injury, gastrointestinal complications (e.g., ileus, fistula), venous thromboembolism, infection, reoperation, and mechanical complications. Univariate and multivariate logistic regression were performed with predictor variables of age, sex, and Elixhauser Comorbidity Index (ECI). Results: Among 1,123,841 ALIF patients (57% male), total complications occurred in 28,698 (2.6%) at 30 days, 39,858 (3.5%) at 90 days, and 60,824 (5.4%) at 1 year. Univariate analyses demonstrated that prior abdominal surgery significantly increased the odds of complications in nearly all sub-groups. Patients with prior non-obstetric abdominal surgery (ABD) or any abdominal surgery (ANYABD) had more than double the odds of complications compared to those without abdominal surgery (NOABD) at all time points (1-year OR 2.3, 95% CI 2.2-2.4; p < 0.0001). On the other hand, patients with a history of cesarean section (CSEC) showed no significant increase in risk compared to NOABD (1-year OR 1.16, 95% CI 0.99-1.35; p = 0.07). However, after adjusting for age, sex, and comorbidity burden (ECI) the differences between groups were no longer significant (ABD vs NOABD at 1 year: OR 1.00, 95% CI 0.97-1.04; p = 0.82). The ECI in the adjusted model was the strongest predictor of complications (OR 1.08, 95% CI 1.07-1.09, p < 0.0001). Conclusion: This study investigates the importance of surgical history in predicting ALIF complications and specific post-operative outcomes. Patients with prior abdominal surgery represent a vulnerable subgroup and would benefit from comorbidity optimization rather than solely surgical history guiding perioperative risk assessment and surgical planning.

ID: AGS26/354

A294: Prevention of revision surgery through wound drainage in elderly patients undergoing posterior spinal fusion surgery - A prospective randomized multicenter study

Klaus Schnake 1 , Denis Rappert 1 , Olga Cheremina 1 , Matthias Pumberger 2 , Thilo Khakzad 2 , Lars Klapproth 3 , Jorg Franke 3 , Torsten Kluba 4 , Alexander Hammer 1

1 Center for Spinal and Scoliosis Surgery, Malteser Waldkrankenhaus St. Marien, Erlangen, Germany

2 Center for Musculoskeletal Surgery, Charité - University Medicine Berlin, Berlin, Germany

3 Department of Orthopedics, Klinikum Magdeburg, Magdeburg, Germany

4 Department of Orthopaedic Surgery, Städt. Klinikum Dresden, Dresden, Germany

Introduction: The purpose of this study was to examine whether wound drainage prevents the need for revision surgery in elderly patients undergoing posterior thoracolumbar spinal fusion surgery involving up to four segments. Material and Methods: Multicenter, prospective, randomized study including patients (age > 65 years) with degenerative diseases, deformities, or fractures. The drainage group (DG) received two drains, which were removed after at least two and no more than four days. The exclusion criteria were: dural tears, infections (e.g. spondylodiscitis), inflammatory diseases, tumours, multiple injuries, polytrauma, unstable fractures and known coagulation disorders. Categorical variables were analyzed using the chi-square test and Mann-Whitney U-test. Logistic regression analysis was performed to identify factors associated with revision surgery after three months. Linear regression analysis was performed to detect factors modifying LOS. Results: The evaluation included 123 patients, with an average age of 74.55 years (range 66-89 years). Revision rate was significantly higher in the non-drainage group (NDG) (19 of 64 patients; 29.7%), compared to DG (7 of 59 patients; 11.9%) (p = 0.016; odds ratio (OR) 3.14; 95% confidence interval (CI) 1.21 - 8.14). There were no revisions due to hematoma in the DG compared with NDG (7 of 64; 10.9%) (p = 0.0087). Multivariate analysis revealed that the use of drainage (p = 0.015; OR 0.25; 95% CI 0.08-0.77), low body mass index (BMI) (p = 0.012; OR 1.15; 95% CI 1.03-1.28), and a greater number of operated segments (p = 0.024; OR 0.45; 95% CI 0.22-0.90) were significant predictors of avoiding revision surgery in this group. The application of a wound drainage had no significant impact on length of stay (LOS) (LOS no drain: 13.13 days ± 7.79 days; LOS with drain: 12.59 days ± 7.58 days; p = 0.67). In multivariate analysis LOS was not significantly predicted by the use of a drainage system (p = 0.25). Conclusion: Using wound drainage significantly decreases the early revision rate in elderly patients undergoing posterior thoracolumbar spinal fusion surgery. This is especially true for revisions caused by hematoma.

ID: AGS26/408

A295: Extended oral antibiotic prophylaxis after anterior cervical spinal fusion is associated with increased infection and sepsis rates

Michael Kim 1 , Melissa Romoff 1 , Emily Tse 1 , Courtney Spitzer 1 , Andy Ton 1 , Vivian Chen 1 , Ryan Le 1 , Abhinav Sharma 1 , Justin Chan 1 , Emily Mills 1 , Hao-Hua Wu 1 , Don Park 1 , Yu-po Lee 1 , Nitin Bhatia 1 , Sohaib Hashmi 1

1 Univeristy of California, Irvine, Orange, United States

Introduction: Extended oral antibiotic prophylaxis, commonly employed in hip and knee arthroplasty beyond the standard 24-hour perioperative period 1 , has gained popularity as a preventive measure 2 . However, recent large-scale matched cohort studies have reported no reduction in prosthetic joint infection rates and, in some cases, an increase in complications. Despite limited evidence, the practice remains widespread, particularly among high-risk patients 3,4 . In spine surgery, the efficacy of extended oral antibiotics in reducing postoperative infection, sepsis, or revision rates has not been evaluated in large, multi-institutional studies. Given the rising concerns over antibiotic resistance and microbiome disruption, this study investigates the clinical utility of extended prophylaxis following anterior cervical spinal surgery. Material and Methods: A retrospective cohort study was performed using data from the TriNetX United States Collaborative Network. Adult patients (≥ 18 years) who underwent anterior cervical spinal surgery were identified via CPT codes. All patients were assumed to have received standard intravenous antibiotic prophylaxis within 24 hours postoperatively. Cohorts were defined by the presence or absence of extended oral antibiotic prescriptions administered between 24 hours and 7 days following surgery. Propensity score matching (1:1) was used to control for age, sex, BMI, smoking status, and relevant comorbidities (e.g., renal failure, diabetes mellitus). The primary outcome was 90-day postoperative spinal surgical site infection. Secondary outcomes included reoperation and sepsis. Outcomes were analyzed using Kaplan-Meier survival estimates, hazard ratios, and cumulative incidence statistics. Results: A total of 4,964 patients who received extended oral antibiotics following anterior cervical spinal fusion were propensity score matched 1:1 to 4,964 patients who did not. The 90-day incidence of surgical site infection was not statistically significant between cohorts (16.277% vs. 15.008%; RR: 1.10, p = 0.0818). Reoperation rates were also similar, with 3.71% in the extended antibiotic group and 3.10% in controls (p = 0.0969). However, incidence of sepsis was significantly higher in the extended antibiotic group (2.56% vs. 1.23%; RR: 2.11, p < 0.0001). Wound infections were also more common in the extended antibiotic group (4.029% vs. 2.115%; RR 1.943, p < 0.0001). Conclusion: Extended oral antibiotic prophylaxis following cervical spinal fusion was not associated with a significant reduction of surgical site infection or reoperation. However, patients who received extended antibiotics experienced significantly higher rates of sepsis and wound infection compared to matched controls. These findings, while accounting for potential selection bias, challenge the presumed protective benefit of extended antibiotic use in this context. In light of the increased complication rates and the broader implications for antimicrobial resistance and microbiome disruption, the routine use of extended prophylaxis warrants critical reevaluation in anterior cervical spinal surgery.

ID: AGS26/923

A296: Should venous thromboembolism chemoprophylaxis be utilized following spinal fusions? A clinical decision analysis

Sofia Perez 1 , Owen Dean 1 , Jackson Pophal 1 , Emmanuel Fatodu 1 , Jack Eichman 1 , Ilyas Aleem 1,2

1 University of Michigan Medical School, Ann Arbor, United States

2 Department of Orthopedic Surgery, Michigan Medicine, Ann Arbor, United States

Introduction: Venous thromboembolism (VTE), which includes deep vein thrombosis (DVT) and pulmonary embolism (PE), is a life-threatening complication that can occur after major surgery. Despite increasing rates of spinal fusions nationally, the efficacy of VTE chemoprophylaxis remain unknown, and chemoprophylaxis comes with the risk of bleeding, epidural hematoma, and possible reoperation. The current literature is limited by heterogeneous patient populations, surgical techniques, and inconsistent timing and dose of chemoprophylaxis. As direct comparisons of each intervention may not reflect health utility values for the different outcomes and associated probabilities, we sought to synthesize the available evidence and construct a clinical decision tree model to better guide clinicians in the use of VTE prophylaxis following spinal fusion. Material and Methods: We constructed a decision analytic model representing potential outcomes after spinal fusions with and without VTE prophylaxis. Probabilities of outcomes (epidural hematoma, VTE event) were derived from a systematic review of randomized trials and observational studies. Five reviewers independently screened abstracts and full texts, with data extracted to derive probabilities and health utility values. A clinical decision tree was constructed using the weighted averages of each outcome probability and utility values among the studies collected. Weighted probabilities and utility values were combined to conduct a rollback analysis and subsequent sensitivity analysis to determine the strength of utilities. Results: A total of 8 studies with 2,177 patients were included in the final analysis. The base-case analysis demonstrated that VTE Chemoprophylaxis (n = 1113) yielded an expected utility of 0.79 compared to 0.89 for No VTE Chemoprophylaxis (n= 1064). Results from one-way sensitivity analyses indicated that the model conclusions were robust to plausible variations in key parameters, including utility values and probabilities associated with VTE and epidural hematomas. Across tested ranges, the expected utility values varied minimally (range: 0.85 to .94 for No VTE Chemoprophylaxis versus 0.75 to 0.84 for VTE Chemoprophylaxis), and No VTE Chemoprophylaxis remained the preferred option at all threshold changes for both utility values and probabilities. Conclusion: When outcomes and their health utilities are evaluated in a systematic manner, VTE chemoprophylaxis does not show any added benefit after spinal fusions, even over a wide range of outcome probabilities and health utility values.

ID: AGS26/981

A297: The use of the psoas-to-lumbar vertebrae index and modified frailty index in predicting postoperative complications in degenerative spine surgery: can sarcopenia or frailty be underestimated?

Marco Manzetti 1,2 , Alberto Ruffilli 1,2 , Giovanni Viroli 1,2 , Matteo Traversari 1 , Marco Ialuna 1 , Cesare Faldini 1

1 Istituto Ortopedico Rizzoli, Clinica Ortopedica e Traumatologica I, Bologna, Italy

2 Department of Biomedical and Neuromotor Science - DIBINEM, Bologna, Italy

Introduction: Degenerative spine disease (DSD) is increasingly prevalent due to aging populations, leading to higher surgical interventions and associated complications. This necessitates a comprehensive preoperative assessment, evaluating frailty through tools such as the modified Frailty Index 5 and modified Frailty Index 11 (mFI-5 and mFI-11). Despite the utility of mFI-5 and mFI-11 in predicting postoperative complications, these indices do not account for sarcopenia, a syndrome related to but distinct from frailty, which is associated with higher complication rates. This paper aims to retrospectively evaluate the influence of sarcopenia and frailty on postoperative adverse events in a cohort of patients who underwent posterior spine fusion for degenerative disease of the lumbar spine. Material and Methods: We retrospectively reviewed 286 patients aged 50-85 years who underwent posterior lumbar fusion of up to three levels for DSD between 2010 and 2023. Exclusion criteria included trauma, tumor, revision surgery, scoliosis, rheumatologic disease, and malabsorptive disorders. Frailty was assessed with mFI-5 and mFI-11, while sarcopenia was measured on preoperative MRI using the psoas-to-lumbar vertebral index (PLVI). Primary outcomes were adverse events within 30 days, including surgical site infection, reoperation, cardiopulmonary, neurological, and systemic complications. Univariate and multivariate analyses were performed to identify independent predictors. Results: Higher frailty scores were significantly associated with increased postoperative complications (p < 0.05). Both mFI-5 and mFI-11 were predictive, with strong correlation between the two (r = 0.916). mFI-11 provided slightly higher discrimination, but receiver operating characteristic curves demonstrated comparable accuracy, supporting mFI-5 as a practical alternative. Complication rates rose progressively with higher frailty categories, including infections, reoperations, and systemic events. Sarcopenia, as measured by PLVI, was not associated with overall complication rates. A significant association emerged only with pulmonary complications (p = 0.031), while no correlations were observed for other outcomes. Multivariate logistic regression confirmed frailty as a strong independent predictor of postoperative complications, whereas sarcopenia did not retain significance. Odds ratios for frailty remained robust across different complication subgroups, but PLVI showed limited prognostic contribution and was excluded as an independent factor. Conclusion: Frailty, as measured by mFI-5 or mFI-11, is a reliable and independent predictor of early postoperative complications following lumbar fusion for DSD. In contrast, sarcopenia assessed by PLVI has limited predictive power, with relevance only to pulmonary outcomes. These findings highlight the superiority of frailty indices for preoperative risk stratification, offering a comprehensive reflection of biological reserve and vulnerability beyond muscle mass alone. Incorporating frailty screening into routine surgical planning may improve patient counseling, perioperative management, and outcome prediction. Further research should refine sarcopenia assessment methods and explore integrated models combining frailty, muscle quality, and systemic biomarkers.

OP34: Biomechanics in Your Practice

ID: AGS26/1034

A298: Dynamic assessment of sagittal balance in patients with degenerative spinal deformity: a controlled biomechanical study

Armand Dominik Škapin 1 2 , Janez Vodičar 3 , Matej Supej 3 , Miha Vodičar 1 2

1 University Medical Centre Ljubljana, Ljubljana, Slovenia

2 Faculty of Medicine, University of Ljubljana, Ljubljana, Slovenia

3 Faculty of Sport, University of Ljubljana, Ljubljana, Slovenia

Introduction: Spinal degeneration is becoming increasingly common within the aging population and often leads to progressive deformity and worsening of sagittal balance. The resulting persistent pain and disability may necessitate surgical intervention. Surgical planning to restore optimal spinal alignment typically relies on preoperative standing full spine radiographs. However, this approach has limitations as it captures the spine only during a brief period of forced upright stance and does not always reflect the dynamic sagittal profile of the patient. This occurs because compensatory mechanisms may maintain normal sagittal balance during radiographic imaging but can sometimes fail during gait as the postural muscles become fatigued. To further explore this discrepancy, a controlled clinical biomechanical study was conducted to analyze sagittal balance dynamics during walking and compare it with sagittal balance during upright stance in patients with adult spinal deformity. Material and Methods: A total of 33 patients with pathological sagittal balance parameters on standing radiographs and clinical complaints of pain and disability, together with 31 matched controls, were enrolled. Standing full spine x-rays were obtained before and immediately after six minutes of walking. All participants underwent three-dimensional motion capture analysis during a six-minute walking assessment. Additionally, we analyzed body composition, muscle strength, and patient-reported outcome measures (VAS, ODI, EQ-5D-5L). Results: Radiological measurements after walking revealed statistically significant deterioration in all sagittal balance parameters, such as SVA (68.4 ± 52.1 mm to 80.4 ± 60.9 mm), as well as lumbar lordosis and thoracic kyphosis in the study group, while no such changes were observed in the control group. Measurements during three-dimensional motion capture showed substantial deterioration of sagittal balance during the transition from standing to walking, with significantly larger changes observed in the study group (for example, an increase in C7-L5-sagittal trunk shift from 69.1 ± 50.6 mm to 138.7 ± 51.9 mm) compared with the control group (20.9 ± 18.8 mm to 69.2 ± 19.7 mm). No additional deterioration occurred on average during the walk itself, although notable worsening was observed in three individual patients in the study group. Predictive factors for sagittal balance deterioration during walking included higher VAS scores, a subjective feeling of forward trunk tilt while walking, an increased extracellular to total body water ratio, and specific radiological parameters (PI, SS, LL, PI-LL, SVA, CAM-HA, OD-HA, GT). Conclusion: Sagittal balance compensatory mechanisms are significantly more effective in a standing position than during walking. The most significant deterioration occurs during the transition from standing to walking and mostly returns to baseline levels after walking. While sagittal balance remains stable in most patients during the walking phase, noticeable deterioration occurred in several individuals. The severity of spinal deformity appears to be the main factor influencing sagittal balance deterioration during walking.

ID: AGS26/1720

A299: Intervertebral disc degeneration: age-related mechanical changes and possible influence on adjacent vertebrae

Margherita Pasini 1 , Gaia Youssef 1 , Davide Alberti 1 , Luca Cristofolini 1 , Marco Palanca 1

1 Department of Industrial Engineering, Alma Mater Studiorum - University of Bologna, Bologna, Italy

Introduction: Intervertebral disc (IVD) degeneration is a common age-related process and a leading cause of low back pain. Despite the crucial role of IVDs in maintaining spinal stability and flexibility, IVD degeneration is currently only partially considered by clinicians when assessing spine-related complications, such as osteoporotic (T-score, FRAX) or metastatic (SINS) fractures. Indeed, the IVD condition is typically assessed by providing morphological information through the Pfirrmann scale [1], which does not capture the associated mechanical changes. A comprehensive characterization of the IVDs biomechanical behaviour is essential to advance the current clinical knowledge. This study aimed to characterise mechanical changes in human IVDs across different ages and degeneration conditions to better understand their interaction with adjacent vertebrae. Material and Methods: Ten T12-L1 IVD specimens were harvested and prepared from human donors: 5 young (33 ± 2) and 5 elderly (63 ± 1) (Prot. 113043). Specimens were imaged using 3T-Magnetic Resonance (sagittal and axial T2-weighted sequences) and Computed Tomography to assess the degeneration of the IVDs (using the Pfirrmann scale [1]) and exclude pathological bone conditions. The specimens were mechanically tested using a spine simulator (MIB4.0-MAS2-S) under axial compression (up to 0.5% of the body weight), clockwise/counterclockwise torsion (up to 4Nm) and flexion/extension (up to 4Nm) loading in the elastic regime . A four-camera 3D-Digital Image Correlation system (Aramis Adjustable, Zeiss) was used to measure the full-field displacements and strains on the surface of the IVDs. The stiffness was calculated as the slope of the load-displacement curve between 20% and 80% of the maximum load. Deformations were measured and compared between the two age groups. Results: The young specimens showed no/mild IVD degeneration (Pfirrmann 1/2 [1]), while the elderly ones had moderate degeneration (Pfirrmann 3 [1]). Elderly IVDs were stiffer in axial compression (+111%), in flexion (+102%) and extension (+94%), while no differences were observed in torsion. Elderly IVDs showed larger deformations compared to young IVDs. Elderly IVDs also exhibited higher variability (+53%). The direction of the deformations was consistently oriented in young IVDs and scattered in elderly IVDs. Conclusion: The higher stiffness of degenerated/elderly IVDs represents a contributing factor to low back pain. It also increases the physical effort required to achieve certain spinal positions and restore the neutral posture, making these movements more difficult to perform. The deformation maps on the surface of the IVDs revealed localised zones of high strain in the elderly IVDs, reflecting their reduced ability to withstand high loads. Moreover, such deformation concentrations may be transmitted to adjacent vertebrae [2, 3], suggesting that degenerated IVDs may further compromise vertebral mechanical behaviour, potentially increasing the risk of fracture in case of vertebrae affected by pathologies. This study provides essential data to advance clinical and scientific understanding of the effects of IVD degeneration and ageing on spine biomechanics. Such insight could help identify predictive markers to support clinical decision-making and improve the quality of life for patients suffering from vertebral pathologies.

References

1. Pfirrmann. Spine 2001

2. Pasini. JMBBM 2025

3. Cavazzoni. Acta 2025

Acknowledgements: HEU-Funded METASTRA project (grant 101080135) and Fondazione del Monte (MULTIDISC project).

ID: AGS26/2419

A300: Early adjacent segment degenration and disc biochemical markers

Theodore Choma 1 , Jacob Kramer 1 , Aaron Stoker 1 , Dan Hogan 1 , Quayd Robertson 1 , Don Moore 1 , Nikhil Jain 1 , Abdul Buyuk 1

1 University of Missouri, Columbia, United States

Introduction: Adjacent segment degeneration (ASD) is a known complication following spinal fusion. Biomechanical mechanisms have been implicated but it remains unclear whether these patients share common underlying biochemical mechanisms. This study aims to investigate the biochemical differences in intervertebral disc cells in patients who experienced subsequent early ASD by analyzing tissue markers and cellular behavior in the discs that were surgically explanted, using these cells as a proxy for those involved in the ASD segment. Material and Methods: Intervertebral disc samples were routinely collected from 211 patients undergoing one- or two-level cervical and lumbar fusion surgeries between 2018 and 2022. Patients were followed for a minimum of one year (mean follow-up: 27 months) and stratified based on the presence of radiographic ASD. Tissue samples were analyzed for biochemical markers. Secondary outcomes included in vitro analysis of cultured disc cells, evaluating the expression of relevant factors. Results: Disc cells from patients with early ASD exhibited significantly lower levels of sclerostin (p = .029) compared to unaffected segments. Additional biochemical differences were observed in tissue samples, with variations in key molecular markers suggesting altered disc homeostasis. Cultured disc cells showed lower expression of VEGF (p = .043), MCP-1 (p = .039), MMP-2 (p = .02), MMP-3 (p = .042), TIMP-1 (p = .002), TIMP-2 (p = .001) and TIMP-4 (p = .038). Conclusion: Biochemical processes may be associated with developing early ASD following spinal fusion surgery. Lower sclerostin levels in affected segments suggest a potential role in disease progression. Lower levels of VEGF, MCP-1, MMP-2, MMP-3 suggest altered disc homeostasis further supporting a distinct biochemical environment in ASD spines. Perhaps the most striking is the substantially lower levels of TIMP (-1, -2, and -4) expression in the early ASD patients. It is unclear if these differences are based in genetic variations or epigenetics. These findings contribute to a basic understanding of ASD and may inform future strategies to mitigate its development in spine surgery.

ID: AGS26/2720

A301: Spino-pelvic sagittal changes during growth: result from a longitudinal retrospective study on 134 patients

Matteo Traversari 1 , Alberto Ruffilli 1 , Giovanni Viroli 1 , Marco Manzetti 1 , Chiara Ricci Petitoni 2 , Cesare Faldini 1

1 IRCCS - Istituto Ortopedico Rizzoli, Dipartimento di Scienze Biomediche e Neuromotorie - DIBINEM, Bologna, Italy

2 IRCCS - Istituto Ortopedico Rizzoli, Bologna, Italy

Introduction: Spinopelvic sagittal alignment evolves during growth, influenced by both morphological and biomechanical maturation. While few cross-sectional studies have described age-related changes in pelvic and spinal parameters, the lack of longitudinal data following the same individuals throughout skeletal development remains a major gap in the literature. A better understanding of these changes is essential to improve the characterization and management of adolescent patients with spinal deformities, both in conservative and surgical settings. The aim of the present retrospective longitudinal study was to characterize the evolution of spinopelvic morphological parameters during growth in a cohort of subjects treated conservatively for spinal deformities involving the cervical and/or thoracic spine, based on serial radiographic evaluations performed throughout skeletal maturation. Material and Methods: This was a retrospective longitudinal study based on full-spine standing bi-planar x-rays of 134 children or adolescent patients (mean age at baseline: 11.6 ± 1.8 years; mean follow-up: 2.6 ± 1.6 years), each undergoing between 2 and 8 spinal radiographic evaluations (mean: 2.88 x-rays, IQR: 1 x-ray) throughout skeletal maturation. A total of 386 radiographs were analyzed. All subjects were treated conservatively for spinal deformities involving the cervical and/or thoracic spine. Exclusion criteria included prior spinal surgery, neuromuscular or syndromic etiology, and any pathology affecting the lumbopelvic region. For each radiograph, spinopelvic morphologic parameters such as pelvic incidence (PI), pelvic thickness (distance between the center of the femoral heads and the midpoint of the S1 endplate), anteroposterior length of the S1 superior endplate, and relative pelvic thickness (pelvic thickness divided by S1 plateau width) were measured using Surgimap by two independent observers blinded to age and skeletal maturity stage. Skeletal maturity was assessed by a separate expert in pediatric orthopaedics using CVM stage, Risser sign, triradiate cartilage status, and Pesenti maturity classification (a five-stage system based on a combination of chronological age, triradiate cartilage closure, and Risser grade: stage 1 = prepuberty; stage 2 = early pubertal growth spurt; stage 3 = late pubertal growth spurt; stage 4 = deceleration; stage 5 = skeletal maturity). Patients were then stratified into subgroups representing distinct stages of skeletal development. Based on published data on the intra-observer measurement error of pelvic incidence, which typically ranges from 2° to 6° depending on imaging modality and observer experience, we considered as “true variation” only those changes greater than 3.5° between two timepoints. This threshold was chosen as a conservative approximation of the minimal detectable change. Changes in spinopelvic morphologic parameters were annotated in both absolute and relative terms across different skeletal maturity stages. The probability of observing a subsequent significant change in PI was calculated for each Pesenti maturity stage. Results: During growth, patients showed an average individual increase in PI, across multiple timepoints, resulting in a transition from 40.3° at Pesenti maturity stage 1 to a peak of 47.8° at stage 4-5. Patients who began longitudinal radiographic follow-up during Pesenti maturity stages 1 and 2 commonly experienced a significant absolute increase in PI over time. In contrast, such increases became progressively less frequent starting from stage 3, with the probability of observing further significant increases in PI dropped significantly after stage 2 (the first part of the pubertal growth spurt), suggesting a reduced potential for morphological spinopelvic changes once mid-pubertal development is reached. The probability of a true absolute increase in PI between two timepoints was 64.7% for transitions starting from Pesenti stage 1 (95% CI: 47.9%-78.5%), 60.0% from stage 2 (95% CI: 45.5%-73.0%), 14.3% from stage 3 (95% CI: 7.4%-25.7%), and 17.2% from stage 4 (95% CI: 7.6%-34.5%). Similar proportions were observed when considering relative PI changes, confirming that most spinopelvic morphological changes occur during the early stages of skeletal maturation, and that further modifications become unlikely once the triradiate cartilages are closed. Conclusion: Our results show that PI undergoes significant increase during skeletal maturation, with the most pronounced changes occurring between Pesenti maturity stages 1 and 3. In conclusion, only limited modifications in PI were seen once the triradiate cartilage was closed.

ID: AGS26/3273

A302: Greater cage angle ≠ greater alignment: rethinking TLIF planning

Schahin Salmanian 1 , Kiana Yeganeh 1 , Ryan Screven 1 , Diego Soto Rubio 1 , Mohammad Hassan A. Noureldine 1 , Jay I. Kumar 1 , Mohsen Rostami 1 , Mark S. Greenberg 1 , Puya Alikhani 1

1 University of South Florida, Department of Neurosurgery, Brain, and Spine, Tampa, United States

Introduction: Transforaminal lumbar interbody fusion (TLIF) aims to arrest motion, restore spinal stability, and improve sagittal alignment. A common assumption in surgical planning is that interbody cages with greater lordosis will create larger changes in lumbar lordosis. However, the extent to which cage angle directly influences sagittal alignment remains poorly defined. This study assessed the relationship between the cage angle and the degree of segmental and global lordosis correction achieved following TLIF. Material and Methods: A retrospective review was performed on 106 adult patients who underwent TLIF between 2021 and 2024. Pre- and postoperative lateral radiographs were analyzed to measure segmental and global lumbar lordosis. Cage angle (in degrees) was recorded for each level. Correlation and linear regression analyses were performed to evaluate the association between cage angle and achieved lordosis. Results: There was no statistically significant correlation between greater cage angles and lordosis correction at any spinal level. A slight inverse trend was observed overall (r = -0.173, p = 0.2058). L2-L3 showed the strongest negative correlation (r = -0.566, p = 0.434), while L3-L4 had a weak positive trend (r = 0.183, p = 0.550). L4-L5 and L5-S1 both demonstrated non-significant inverse correlations (r = -0.187, p = 0.340and r = -0.344, p = 0.331, respectively). Linear regression further supported these results, showing no significant association between cage angle and either segmental (R 2 = 0.012, p = 0.21) or global (R 2 = 0.008, p = 0.234) lordosis correction. Conclusion: This study challenges the assumption that larger cage angles lead to greater lordosis correction. No significant correlation was found across spinal levels, and higher cage angles may even be counterproductive. By avoiding hyperlordotic cages, surgeons may reduce the risk of subsidence while focusing on factors like technique, placement, and patient anatomy that more reliably impact alignment.

ID: AGS26/3398

A303: Biomechanical comparison of realignment strategies in patients with lumbar fusion to the pelvis

Christopher J. Nielsen 1 , Fatemeh Alavi 1 , Stephen Lewis 1

1 University of Toronto, University Health Network (UHN), Toronto, Canada

Introduction: Inadequate restoration of spinopelvic parameters following surgery has been associated with elevated mechanical loads, junctional failure, and worse clinical outcomes. While several realignment strategies have been proposed to optimize postoperative balance, the biomechanical implications of each strategy remain incompletely understood. The objective of this study is to compare different realignment strategies and clarify how specific alignment targets influence spinal forces at the proximal junction in patients undergoing lumbar fusion to the pelvis. Material and Methods: 41 patients (46 % female) with mean age of 68.9 ± 7.2 with lumbar fusion to the pelvis were retrospectively reviewed. UIV in 90% was L2 and 10% was L1. The mean PI was 52.9 o ± 11.9. Five realignment strategies including Age-adjusted, GAP, Segmental, Roussouly, T4-L1-axis were simulated. Sex-matched generic musculoskeletal (MSK) models were individualized for each suggested alignment. In total, 205 models were evaluated in standing posture, with spinal shear and compressive forces computed. Alignment strategies resulting in the lowest spinal forces were identified as optimal. Results: Across all patients, spinal forces varied significantly between simulated alignments. The optimal strategy with the lowest proximal junction compression was T4–L1-axis in 53.5% of cases, followed by Segmental, Roussouly, and GAP. In contrast, Age-adjusted, Roussouly, Segmental alignments most often minimized shear. On average, inter-alignment compression differences reached 149.8 N, with optimal strategies reducing forces by 15.6% compared with the next-best alternative. In several subgroups, reductions exceeded 20%, underscoring the clinical relevance of alignment selection. Stratified analysis showed that higher PI and older age were associated with greater inter-alignment variability. Conclusion: Realignment strategies have a substantial impact on proximal junction biomechanics after lumbar fusion to the pelvis. No single strategy was universally optimal, as the best alignment depended on individual pelvic incidence and age. However, the T4–L1-axis most consistently minimized compression, while Age-adjusted often reduced shear. These findings highlight the need for individualized alignment targets to optimize biomechanical efficiency, reduce junctional loading, and consequently mechanical complications.

ID: AGS26/3439

A304: Biomechanical analysis of long-segment cervical spine stabilization: finite element analysis of pedicle, pars, translaminar and hybrid screw fixation models

Mohamad Bakhaidar 1,2 , Ataollah Shahbandi 2 , Pegah Ghamasaee 2 , Abdul Mounnem Yassin Kassab 2 , Balaji Harinathan 2 , Narayan Yoganandan 2 , Karthik Devaraj 2 , Saman Shabani 2

1 Ottawa University, Ottawa, Canada

2 Medical college of Wisconin, Milwaukee, United States

Introduction: Screw placement in the C2 vertebra plays a pivotal role in providing a robust anchor point for cervical fusion. This study aimed to evaluate the biomechanical performance of pars, pedicle, and translaminar screw techniques, alongside various hybrid constructs using finite element (FE) analysis. Material and Methods: A validated FE model of the C2-T1 vertebrae was employed, with lateral mass screws from C3 to C6, and a pedicle screw for T1 vertebra to simulate posterior cervical fusion. A physiological loading condition was simulated by applying a 2 Nm force across all scenarios. The range of motion (ROM) and rod stress were compared among pedicle, pars, and translaminar screw fixation. Additionally, the performance of various hybrid constructs was assessed. Results : In flexion, pars screws exhibited 38.5% increase in motion compared to pedicle screws and a 65.5% increase compared to translaminar screws. In extension, pars screws showed a 36.1% increase in motion versus pedicle screws and 72.9% versus translaminar screws. In lateral bending, translaminar screws presented a 8.9-fold greater motion compared to pedicle screws and nearly double the ROM in axial rotation. Translaminar screws exhibited an increase in rod stress (19%) compared to pedicle screws. Among the hybrid constructs, the bilateral pedicle-bilateral translaminar (BiPed-BiTrans) construct demonstrated the least ROM in flexion, rendering it the stiffest. In contrast, the unilateral pedicle-unilateral translaminar model exhibited the least stability in flexion, with 2.2 times greater ROM versus the Biped-BiTrans construct. Conclusion: Translaminar screws provided robust fixation in a C2-T1 construct during flexion and extension but were inferior to pars and pedicle screws in axial rotation and lateral bending and associated with greater rod stress. The Biped-BiTrans construct was the most rigid hybrid construct, demonstrating the smallest ROM in flexion.

ID: AGS26/397

A305: Comparison of posterior atlantoaxial spondylodesis procedures: a biomechanical cadaver study

Adrian Cavalcanti Kußmaul 1 , Titus Kuehlein 1

1 Musculoskeletal University Center Munich (MUM), Munich, Germany

Introduction: Fractures of the atlas and axis remain the primary indication for C1-C2 fusion, largely due to demographic shifts increasing fracture incidence and the decline of rheumatic disease-related instability. While anterior screw osteosynthesis is often feasible in younger patients with minor instability, dorsal stabilisation remains the gold standard in complex fractures and severe dislocation. Among dorsal procedures, rigid screw fixation according to Harms is widely applied, consisting of a combination of atlas massa lateralis screws and axis isthmus srews. Recently, Harati et al. proposed a modified approach combining transarticular C1-C2 screws with atlas massa lateralis screws in a screw-rod system, generating three-point fixation. Biomechanical studies suggest this provides superior stability, but cadaveric validation has been lacking. This study aimed to compare the primary stability of the Harati and Harms techniques in human C1/C2 cadaver specimens. Material and Methods: Sixteen human cadaver cervical spines were collected, of which thirteen met inclusion criteria after radiological and structural assessment. Each specimen was tested in native, injured (Type II dens fracture), Harms, and Harati states. Testing followed established biomechanical protocols using flexion/extension, lateral bending, and axial rotation at 1 Nm with a axial preload of 50 N. Specimens were mounted in custom-designed pots with physiological alignment, and motion was recorded using a 3D optical measurement system. Statistical analysis employed Friedman tests and Wilcoxon signed-rank post-hoc comparisons. Results: Both the Harms and Harati methods significantly reduced motion compared to native and injured states. Mean ROM values (°) were: flexion/extension, Harms 0.67 ± 0.25, Harati 0.59 ± 0.23; lateral flexion, Harms 0.95 ± 0.48, Harati 0.67 ± 0.47; and rotation, Harms 0.92 ± 0.24, Harati 0.55 ± 0.19. Direct comparison showed no significant differences between Harms and Harati in flexion/extension (p = 0.649). However, in rotation, the Harati technique achieved significantly greater stability (p = 0.048) and also in lateral flexion (p = 0.004). Conclusion: Both the Harms and Harati techniques demonstrated effective stabilization of C1/C2 instability. The Harati method provided superior stability regarding rotation and lateral flexion compared with the Harms procedure, while sagittal stability was comparable. These findings suggest that the Harati three-point fixation may offer biomechanical advantages in clinical scenarios requiring enhanced primary stability.

ID: AGS26/948

A306: Positioning matters: preoperative imaging as a predictor of positional correction

Diego Soto Rubio 1 , Chloe Chose 1 , Albert Brotgandel 2 , Dana Saleh 1 , Mohsen Rostami 1 , Daniel Colome 3 , Patrick Kim 1 , Puya Alikhani 1 , Jay I. Kumar 1

1 Department of Neurosurgery, Brain and Spine, University of South Florida, Morsani College of Medicine, Tampa, United States

2 USF Health Morsani College of Medicine, Tampa, United States

3 Florida Atlantic University Charles E. Schmidt College of Medicine, Boca Raton, FL, USA, Boca Raton, United States

Introduction: Prone positioning alone has been shown to increase lumbar lordosis before any surgical intervention. However, the ability to predict the degree of correction achieved from positioning has not been well established. This study evaluates whether preoperative imaging can reliably predict intraoperative positional correction in adult spinal deformity patients. Material and Methods: We retrospectively reviewed adult patients with degenerative thoracolumbar deformity undergoing fusion from L2 or above to the pelvis. Inclusion required preoperative standing scoliosis films, supine CT with scout X-ray, dynamic flexion/extension radiographs, and intraoperative prone lateral fluoroscopy with measurable L1-S1 angles. L1-S1 lordosis was measured on all studies, and differences were calculated between imaging modalities. Correlation and regression analyses were performed to identify predictors of intraoperative correction. Results: A total of 40 patients met the inclusion criteria. Prone positioning significantly increased lumbar lordosis: mean preoperative CT L1-S1 = 47.2°, intraoperative L1-S1 = 56.8°, with a mean gain of 9.6° (p < 0.001) when compared to preoperative scoliosis films. Supine CT L1-S1 demonstrated the strongest correlation with intraoperative lordosis (r = 0.62, p = 0.003), followed by extension (r = 0.56), standing (r = 0.56), and flexion (r = 0.47). Linear regression showed intraoperative L1-S1 lordosis = 13.62 + 0.90 × (Supine CT L1-S1), with R 2 = 0.38 (adjusted R 2 = 0.35). This indicates that nearly 1° of intraoperative correction can be expected for every degree measured on supine CT. Conclusion: Supine CT is the most reliable preoperative predictor of intraoperative positional correction of lumbar lordosis in adult spinal deformity patients. These findings provide a quantitative framework to estimate the degree of sagittal correction achievable from positioning alone, informing surgical planning, the anticipated need for osteotomies, and preoperative patient counseling.

OP35: Diagnostics and Predictive Analytic in Spine

ID: AGS26/275

A307: Erector spinae could be the game changer in surgical decision-making in patients with lumbar spondylolisthesis: a cross-sectional analysis of an age-, sex-, subtype-, level-matched patients with similar spinopelvic parameters received surgical or conservative management

Murat Ekşi 1 , Emel Özcan-Ekşi 2

1 İstinye University, Istanbul, Turkey, 2 Private Practice, Istanbul, Turkey

Background: In this study we aimed to identify whether paraspinal muscle morphology could play a role in surgical decision-making in patients with lumbar spondylolisthesis. Methods: We conducted a cross-sectional analysis of a prospectively collected database between January 2013 and May 2023. Consecutive women and men, who visited our outpatient clinics with chronic LBP, neurogenic claudication, and had lumbar spine magnetic resonance imaging (MRI) for their complaints were included into the preliminary dataset. We compared the patients who had conservative management (conservative group) or underwent surgery for lumbar spondylo−listhesis (surgical group) in terms of intervertebral disc degeneration, end-plate changes, fatty infiltration in the paraspinal muscles and spinopelvic parameters. Results: Conservative and surgical groups were similar in terms of severe IVDD and Modic changes at any lumbar level. Surgical group had significantly fattier erector spinae compared to the conservative group. Regression analysis and ROC analysis revealed an OR of 1.088 and a cut-off value of 17 points for fatty infiltration in the erector spinae to predict which patient could undergo surgery for lumbar spondylolisthesis. Conclusion: Each 1-point increment in fatty infiltration in the erector spinae at any lumbar level increased the likelihood of surgery by 8%. Lumbar spondylolisthesis patients with fatty infiltration score for erector spinae at or above 17 were more likely to have surgery. We recommend clinicians to focus on improving erector spinae muscles in patients with lumbar spondylolisthesiss.

ID: AGS26/121

A308: Distribution characteristics of vertebral CT Hounsfield unit values in patients with degenerative lumbar scoliosis

Fei Xu 1 , Weishi Li 1

1 Orthopaedic Department, Peking University Third Hospital, Beijing, China

Introduction: This study aims to identify how CT HU values vary in degenerative lumbar scoliosis (DLS) patients with different scoliotic apexes and in those with different degree of scoliosis. Material and Methods: We included 222 DLS patients and 140 lumbar spinal stenosis (LSS) patients, dividing the DLS patients into two groups based on scoliotic apex location. Patients were further categorized by T scores into osteoporotic and non-osteoporotic groups, and by Cobb angle into mild (< 20°) and severe (> 20°) scoliosis. We analyzed CT value distribution in these groups and compared the area under the curve (AUC) for predicting osteoporosis based on HU values from different vertebrae. Results: CT HU values for L1 and L2 were significantly lower in the DLS group compared to LSS group (p < 0.05). The lowest HU values in patients with a scoliotic apex at L2 or L2-3 were observed at the scoliotic apex region, while those with the scoliotic apex at L3, L3-4, and L4 had the lowest values at L3. This pattern was more pronounced in patients with a Cobb angle > 20°. In the mild scoliosis group, the lowest CT value was found at L3 when the scoliotic apex was between L2 and L4. Additionally, L4 CT HU thresholds were better at distinguishing osteoporosis than other levels (p < 0.05), when the scoliotic apex was located at L2 or L2-3. Conclusion: CT HU values in DLS patients, particularly with a Cobb angle > 20°, decrease at the scoliotic apex. Caution is required when using HU values from L1, L2, and L3 to assess osteoporosis in patients with a scoliotic apex at L2 and L2-3.

ID: AGS26/2716

A309: Radiological patterns and surgical strategies in degenerative lumbar scoliosis with shoulder imbalance: a novel classification approach

Junyu Li 1 , Junjie Ma 2 , Hanwen Zhang 2 , Zhuo Ran Sun 1 , Yongqiang Wang 1 , Miao Yu 1 , Weishi Li 1

1 Department of Orthopaedics, Peking University Third Hospital, Beijing Key Laboratory of Spinal Disease Research, Engineering Research Center of Bone and Joint Precision Medicine, Beijing, China

2 Peking University Health Science Center, Beijing, China

Introduction: Degenerative lumbar scoliosis (DLS) is a prevalent spinal condition in the elderly, with an incidence of approximately 2%-68%. Numerous studies have verified the significant effect of SI on appearance and satisfaction of patients, and have regarded SI as one of the main determinants of evaluating the outcome of orthopedic surgery. Currently, there is only few systematic studies on the radiographic characteristics of DLS patients with SI. For DLS patients with SI, whether postoperative SB is important and how to achieve postoperative SB is unclear. By radiological observation, we identified two subtypes of SI in DLS patients, each with distinct clinical traits requiring tailored surgical approaches. The objectives of this study were to propose a novel preoperative SI classification in DLS patients and explore how this classification can guide scoliosis corrective surgery. Material and Methods: A cohort of 297 adult patients who underwent posterior scoliosis correction surgery was reviewed. Among them, 71 patients were identified as having coronal and shoulder imbalance (imbalanced group, defined as clavicle angle (CA) ≥ 2° and T1 plumb line-central sacral vertical line (T1PL-CSVL) ≥ 20 mm), while 117 patients constituted a balanced control group (CA < 2° and T1PL-CSVL < 20 mm). The analysis of the imbalanced group involved a two-tiered classification: First, the 71 imbalanced patients were divided into two major groups according to the Obeid classification (Obeid type I vs. II). Subsequently, within each Obeid group, patients were further subdivided based on the shoulder imbalance type, defined as Type A (the higher clavicle aligns with the T1PL offset) or Type B (the higher clavicle is opposite to the T1PL offset). Statistical comparisons were performed using t-tests or chi-square tests. Results: This study identified 71 patients with shoulder imbalance and 117 balanced controls from a cohort of 297. Analysis revealed that within the Obeid type I subgroup, patients with shoulder imbalance type II had significantly higher postoperative T1PL-CSVL (p = 0.014) and lower CRIA (p = 0.009) and Cobb angle (p = 0.041) at the last follow-up. Within the Obeid type II subgroup, type I patients exhibited higher postoperative AVT (p = 0.037) and a higher T1 tilt angle at the last follow-up (p = 0.018). Compared to the balanced group, shoulder imbalance type I patients had larger pre- and postoperative Cobb angles (p = 0.027) and a lower correction rate (p = 0.024). Type II patients had higher preoperative AVT (p = 0.01), postoperative Cobb (p < 0.001) and AVT (p = 0.005), as well as higher last-follow-up T1 tilt (p = 0.008), CRIA (p < 0.001), and Cobb angle (p < 0.001), alongside a lower correction rate (p = 0.011). Direct comparison between the two shoulder imbalance types showed that type II was associated with higher postoperative Cobb (p = 0.032) and AVT (p = 0.035), higher last-follow-up T1 tilt (p = 0.044), CRIA (p = 0.015), and Cobb angle (p = 0.012), and greater Cobb angle progression (p < 0.001). Conclusion: The Obeid classification stratifies prognosis: shoulder imbalance type II in Obeid I patients predicts poor coronal shift correction, while type I in Obeid II patients indicates residual apical translation. Preoperative identification is crucial for surgical planning.

ID: AGS26/2129

A310: AI-assisted prediction of RLDH using preoperative MRI and radiographic parameters

Sebastiano Mazzotta 1 , Francesco Certo 1 , Giulio Bonomo 1 , Giuseppe Barbagallo 1

1 A.O.U. Policlinico G. Rodolico - San Marco, Catania, Italy

Introduction: Recurrent lumbar disc herniation (rLDH) is a major complication after surgical discectomy, with an incidence ranging from 5% to 15% in published series. rLDH can worsen symptoms and necessitate reoperation, with substantial impact on patients’ quality of life. Numerous studies have explored the role of radiological parameters in predicting biomechanical changes of the spine and the risk of recurrence, but definitive conclusions remain lacking. Therefore, the aim is to analyze the association between preoperative radiological parameters and rLDH in order to identify potential predictors of recurrence and, based on these, optimize the surgical strategy. Material and Methods: We conducted a retrospective analysis of 186 patients who underwent lumbar discectomy using microsurgical or endoscopic techniques between 2021 and 2025. Eighteen patients (9.68%) developed recurrences, classified as early (≤ 6 months) or late (> 6 months). For each patient, the following radiological parameters were assessed-on the preoperative MRI and, for recurrent cases, on the MRI documenting recurrence: Pfirrmann grade, Modic changes score, facet joint angle, facet joint hyperintensity, and the presence of micro-instability (evaluated on flexion-extension radiographs). Each parameter was analyzed both at the index level and at the immediately adjacent levels . Results: Multivariate regression and machine-learning-based predictive analyses of preoperative imaging for the entire cohort, compared with the imaging of patients reoperated for recurrence, showed that the radiological variables most frequently associated with recurrence were the facet joint angle (R = 0.54) and facet joint hyperintensity (R = 0.61). The machine-learning analysis further indicated that combining these features with a Pfirrmann grade > 4 of the index disc or the superior disc constitutes a predictive marker of late recurrence that is already identifiable on the initial MRI. No reliable predictors of early recurrence were identified. Conclusion: Facet joint angle and facet joint hyperintensity are the main radiological factors associated with rLDH, particularly in late recurrences. Their combination with a Pfirrmann grade > 4 at the superior level enables the identification of higher-risk patients on the initial MRI. If confirmed in larger cohorts, these findings could support a more aggressive primary surgical strategy in this subgroup of patients.

ID: AGS26/2386

A311: Paraspinal fat index for assessment of sarcopenia in spine research

Murat Ekşi 1 , Emre Unal 2 , Mehmetzeki Yıldız 2 , Akın Akakın 2 , Türker Kılıç 1 , Deniz Konya 2 , Emel Özcan-Ekşi 3

1 İstinye University, Istanbul, Turkey

2 Bahçeşehir University, Istanbul, Turkey

3 Private Practice, Istanbul, Turkey

Introduction: The global population is aging at an unprecedented rate, increasing the importance of geriatric care. Age-related spinal deformities are expected to become more prevalent, and frailty represents a critical concern in elderly patients undergoing spine surgery. Previous studies have demonstrated a strong correlation between fat-infiltrated or atrophied paraspinal muscles and frailty. The Charlson Comorbidity Index (CCI) has long been used to predict clinical outcomes in spine-related disorders and surgeries, preceding frailty assessments. However, no prior studies have evaluated the relationship between CCI and de novo spinal or paraspinal muscle degeneration. This study aimed to examine whether CCI is associated with spinal degeneration and fatty infiltration of the paraspinal muscles. Furthermore, if such an association existed, we aimed to develop a paraspinal fatty infiltration index to predict patients with higher CCI scores. Material and Methods: A previously established CCI cut-off value of 2.5 was used to stratify patients into two groups: CCI ≥ 2.5 and CCI < 2.5. The groups were compared in terms of intervertebral disc degeneration (IVDD), Modic changes, intradiscal vacuum phenomenon (IVP), intrafacet vacuum phenomenon (IfVP), and fatty infiltration of the paraspinal muscles across all lumbar levels. Significant associations between fatty infiltration scores and CCI ≥ 2.5 were further analyzed using binary logistic regression and receiver operating characteristic curve analyses to define the most appropriate Paraspinal Fat Index (PFI). Results: A total of 208 participants were included, of whom 72.6% (n = 151) were women. The mean CCI score was 2.77 ± 2.08 (range: 0-11). Overall, 110 patients (52.9%) had a CCI score ≥ 2.5. Severe IVDD at upper lumbar levels and IfVP at lower lumbar levels were more common in the higher CCI group, whereas no significant differences were observed regarding Modic changes or IVP. Fatty infiltration scores for both multifidus and erector spinae were significantly associated with higher CCI scores. The association was consistent for the multifidus across all lumbar levels whereas it was limited to L1-L2, L2-L3, and L4-L5 for the erector spinae. A combined fatty infiltration score of 2.25 for the multifidus and erector spinae at L2-L3 yielded a sensitivity of 59.3% and a specificity of 62.7% for predicting patients with CCI ≥ 2.5. Conclusion: A combined fatty infiltration score of 2.25 for the multifidus and erector spinae at L2-L3 represents the most appropriate index (PFI) for identifying patients with CCI > 2.5. PFI may serve as a practical tool for detecting comorbid patients prior to spinal interventions. Clinicians should consider increased comorbidity risk in patients with fatty paraspinal muscles. Further validation of PFI is warranted in clinical practice and spine research.

ID: AGS26/2640

A312: Comparison of diagnostic accuracy of CT and SPECT in pseudarthrosis diagnosis: a systematic review

Miguel Cartagena 1 , Elyette Lugo 2 , Nicholas Schiller 1 , Diego Lopes de Lima 1 , Brett Schlifka 1 , Milad Alam 1

1 HCA Florida JFK, Atlantis, United States

2 Johns Hopkins University, Baltimore, United States

Introduction: Pseudoarthrosis is a significant complication of fusion surgery, often causing pain, instability, and the need for revision surgery. CT scanning (CT) remains the gold standard for detecting bony nonunion due to its high resolution, but it can still fail to accurately confirm the presence of solid fusion regardless of metallic artifact. SPECT-CT has emerged as an alternative, offering better material differentiation and reduced artifact interference, yet its diagnostic accuracy remains unclear. This study compares SPECT-CT and CT in detecting pseudoarthrosis using surgical exploration as the gold standard. Material and Methods: This systematic review was conducted in accordance with PRISMA guidelines. Studies were included if they evaluated CT or SPECT-CT as a diagnostic tool for pseudoarthrosis using surgical exploration as the reference standard. A comprehensive literature search was performed across PubMed, Web of Science, Scopus, and the Cochrane Library using the following keywords : imaging modalities, pseudoarthrosis, CT scan, single photon emission CT, thoracolumbar fusion . Pooled estimates for sensitivity, specificity, PPV, and NPV were calculated using a random-effects model, with subgroup analysis performed for each imaging modality. Between-study and between-group heterogeneity were assessed using Cochran’s Q test and the I 2 statistic. Results: A total of 674 patients from 13 studies were included in this analysis. The pooled analysis showed that CT had higher sensitivity (0.89, 95% CI: -22.67 to 24.45) compared to SPECT-CT (0.68, 95% CI: -32.83 to 34.19), but SPECT-CT demonstrated superior specificity (0.83, 95% CI: -35.03 to 36.60) vs. CT (0.57, 95% CI: -22.99 to 24.13). Regarding predictive values, SPECT-CT had a slightly higher PPV (0.74, 95% CI: -32.77 to 34.25) vs. CT (0.57, 95% CI: -22.99 to 24.13), however, CT had a superior NPV (0.88, 95% CI: -22.68 to 24.44) compared to SPECT-CT (0.69, 95% CI: -32.82 to 34.20). Cochran’s Q test confirmed no significant heterogeneity between modalities (p > 0.99). Conclusion: Our results suggest that CT remains the preferred imaging modality for pseudoarthrosis detection, though SPECT-CT may offer advantages in specific clinical scenarios where specificity is prioritized.

ID: AGS26/2836

A313: Assessment of bone mineral density in patients with degenerative spinal disease by MRI-based vertebral bone quality score at different lumbar vertebral levels: an observational prospective study

Youwei Ai 1 , Hong Ding 1 , Qian Chen 2 , Limin Liu 1 , Yueming Song 1

1 West China Hospital, Sichuan University, Chengdu, China , 2 Affiliated Hospital of North Sichuan Medical College, Nanchong, China

Introduction: Osteoporosis is highly prevalent in the elderly population and often remains asymptomatic until complications occur, leading to increased intraoperative risks and poor prognosis. Current bone mineral density (BMD) assessment methods such as DEXA and QCT have inherent limitations and suboptimal compliance in screening. The vertebral bone quality (VBQ) score, an MRI-based radiation-free method, has demonstrated good correlation with DEXA and QCT and shows potential in predicting postoperative complications in spine surgery. This study aimed to prospectively evaluate the effectiveness of VBQ score in assessing BMD and compare its performance with DEXA and QCT to validate its clinical utility. Material and Methods: This prospective cohort study included patients over 45 years of age who underwent single- or two-level lumbar fusion surgery for degenerative spinal disease at our center. All participants underwent MRI, CT, or X-ray within one week before surgery and were followed for at least six months. VBQ scores were calculated on T2-weighted MRI images from L1-S1 and compared with DEXA T-scores and QCT-vBMD. Statistical analyses included correlation tests, intergroup comparisons, and ROC curve analyses to evaluate the diagnostic accuracy and reliability of VBQ score in detecting osteoporosis. Results: A total of 198 patients were enrolled, with a mean age of 58.94 ± 11.84 years; 56.6% were female. Based on DEXA and QCT classifications, patients were stratified into normal, osteopenia, and osteoporosis groups, with significant differences observed in age and sex distribution among groups. Segmental VBQ scores correlated significantly with both DEXA and QCT. The correlation between DEXA and VBQ ranged from r = -0.361 (L4) to -0.500 (L1-L4), while correlation with QCT was stronger (r = -0.412 to -0.563), with the L1 VBQ score showing the highest correlation. In ROC analysis based on DEXA classification, VBQ score showed the highest AUC at S1 (0.703) for differentiating normal from osteopenia, with relatively higher sensitivity at L1 (64.4%) and specificity at L2 (86.7%). For distinguishing osteopenia from osteoporosis, the highest AUC was at L1 (0.697), with sensitivity at L2 (72.6%) and specificity at L1-L4 (80.5%). Based on QCT classification, VBQ scores significantly predicted osteoporosis at most levels except L4 and L5. The L1 VBQ score achieved the highest AUC (0.714), with sensitivity at L1-L4 (68.5%) and specificity at S1 (89.7%). For distinguishing osteopenia from osteoporosis, the highest AUC was at L2 (0.726), with the highest sensitivity at L3 (62.9%) and specificity at L2 (92.1%). Conclusion: This prospective study demonstrates that the VBQ score effectively reflects lumbar BMD, showing significant correlation with both DEXA and QCT and moderate predictive performance. VBQ has promising potential as a preoperative screening tool for osteoporosis in patients undergoing lumbar fusion surgery.

ID: AGS26/2955

A314: Pediatric developmental spondylolisthesis may be the first presenting symptom of hereditary connective tissue disorders

Kexin Xu 1,2,3 , Xiangjie Yin 1,2,3 , Jialuo Han 1,2,3 , Sen Zhao 1,2,3 , Jianguo Zhang 1,2,3 , Nan Wu 1,2,3

1 Department of Orthopedic Surgery, Peking Union Medical College Hospital, Peking Union Medical College and Chinese Academy of Medical Sciences, People’s Republic of China, Bejing, China

2 Beijing Key Laboratory of Big Data Innovation and Application for Skeletal Health Medical Care, People’s Republic of China, Bejing, China

3 Key Laboratory of Big Data for Skeletal Deformities, Chinese Academy of Medical Sciences, People’s Republic of China, Beijing, China

Introduction: Spondylolisthesis affects 4%-6% of children. It may be associated with genetic syndromes or occur as “clinically isolated” developmental spondylolisthesis of unclear etiology. The genetic landscape of developmental spondylolisthesis has never been characterized. Material and Methods: This retrospective study included patients who: (1) were diagnosed with developmental spondylolisthesis before the age 18; (2) had an indication for surgical intervention; (3) were not suspected or diagnosed with a genetic syndrome on preoperative assessment; and (4) underwent genome or exome sequencing (GS or ES). Medical records were reviewed, and genetic data were interpreted according to the ACMG/AMP guidelines. Patients with diagnostic results were categorized into Syndromic Group, while others were categorized into Isolated Group. Results: A total of 19 patients were included. None were suspected or diagnosed with a genetic syndrome during preoperative assessment. After genetic testing, eight patients were diagnosed with genetic syndromes, reaching a diagnostic rate of 42%. Notably, six of them had hereditary connective tissue disorders (HCTDs), including Loeys-Dietz syndrome (LDS, n = 4) and Marfan syndrome (n = 2), indicating that 32% of patients with “clinically isolated” developmental spondylolisthesis actually had HCTDs. Additionally, one patient was diagnosed with Kabuki syndrome, a disorder also associated with connective tissue defects. To investigate whether clinical features could distinguish syndromic cases from isolated cases, we next compared spinal and extra-spinal features between the groups. Age at preoperative assessment and sex distribution were comparable: 12 ± 4 years and 12% male in the Syndromic Group, while 11 ± 2 years and 10% male in the Isolated Group. Body mass index was slightly lower in the Syndromic Group (18 kg/m 2 vs. 20 kg/m 2 ). Patients with genetic syndromes had a lower prevalence of spondylolysis (50% vs. 64%) but were more likely to present with extra-spinal skeletal features (63% vs. 18%) and extra-skeletal features (50% vs. 27%). However, all extra-spinal features in both groups were mild and non-specific, such as pediatric generalized joint hypermobility and highly arched eyebrows. Without genetic testing, it was not possible to determine whether these features represented manifestations of genetic syndromes, normal phenotypic variation, or environmental influences. Notably, one patient underwent surgical reduction for “clinically isolated” developmental spondylolisthesis at the age of 5. She experienced a cascade of major complications, including multiple implant- and alignment-related complications requiring five revision surgeries, two episodes of intraoperative anaphylactic shock, and one postoperative cerebral infarction. Prior GS performed at an outside institution was negative. Mild features suggestive of HCTDs were not apparent until the age of 19, and reanalysis of the GS data led to a definitive diagnosis of LDS. The previous false-negative result may have been due to incomplete phenotypic presentation at a young age and the underrecognized association between spondylolisthesis and HCTDs. Conclusion: Approximately 40% of pediatric developmental spondylolisthesis cases have an underlying genetic etiology. Pediatric developmental spondylolisthesis may represent the initial presenting feature of HCTDs. Spine surgeons are the first specialists these patients visit, and preoperative genetic testing is essential for achieving a timely and accurate diagnosis.

ID: AGS26/755

A315: Survival prediction based on machine-learning in severe traumatic cervical spinal cord injury patients: a multicenter study

Quanchang Tan 1 , Tao Jiang 1 , Zixiang Wu 1

1 Xijing Hospital, Xi'an, China

Introduction: Traumatic cervical spinal cord injury (TCSCI) is recognized as one of the most severe traumatic conditions with a high mortality rate. However, clinical tool for assessing mortality risk in TCSCI patients is still lacking. This study aimed to design and validate a tool to evaluate death risk in patients with motor complete TCSCI by using an artificial intelligence (AI) algorithm and aimed to developer an online application for predicting patient mortality risk to offer valuable insights for treatment. Material and Methods: A total of 670 patients were included, with 505 and 165 patients in the development and external validation cohorts, respectively. Patients within the development cohort were randomly allocated in a 7:3 ratio to training and internal validation cohorts. Various machine-learning survival models were used, including random survival forest (RSF), gradient boosting machine (GBM), extra survival tree (eST) and Cox proportional hazards (CPH) models. Results: The mortality rate in the development cohort was 38.04%. The final model included fifteen predictors (sex, age, Charlson Comorbidity Index, injury severity score, time of injury, energy of damage, spinal cord injury severity classification, neurological level of impairment, important organs damage, transfusion, cervical fracture level, surgical timing, ventilator support, nutritional approach, complications). The GBM model showed the best integrated area under the curve (iAUC) in internal validation (0.901, 95%CI: 0.879-0.923), and external validation showed RSF model with the best iAUC (0.84, 95%CI: 0.813-0.867). The comprehensive performance from internal and external validation indicates that the RSF model achieved the highest score (49), followed closely by the GBM model (45). Therefore, the RSF model was identified as the optimal model and online visualization was implemented as an AI tool. Conclusion: This study developed and validated an AI application to evaluate death risk in patients with motor complete TCSCI. This application can offer healthcare professionals more valuable insight to help clinical decision-making. This advancement holds significant potential for improving outcomes in patients with motor complete TCSCI.

OP36: More Complex Tumor

ID: AGS26/1762

A316: Outcomes of intradural extramedullary tumor resection in the elderly: a 10-year multicenter analysis

John Zhang 1 , A Aravin Kumar 1 , Zheting Zhang 1 , Shane Zaw 1 , David Mak 1 , Damian Lee 1 , Choo Heng Tan 1 , Nishal Primalani 1 , Krishan Kumar Sharma 1 , Robin Pillay 1 , Dinesh Shree Kumar 2 , Ji Min Ling 1 , Lester Lee 1

1 Department of Neurosurgery, National Neuroscience Institute, Singapore, Singapore

2 Department of Orthopaedic Surgery, Changi General Hospital, Singapore, Singapore

Introduction: Intradural extramedullary (IDEM) tumors account for two-thirds of all primary intraspinal tumors, and their incidence increases with age. As most of these tumors are benign, surgery is offered primarily for control of symptoms or neurological deficits. Outcomes of IDEM tumor resection in the elderly are not well studied, and the risk-benefit ratio is challenging to assess in older patients with perceived higher surgical risks. We aimed to evaluate the outcomes of IDEM tumor resection in our institutions to determine the safety of such operations in the elderly population. Material and Methods: Patients who underwent IDEM tumor resection at three tertiary neurosurgical centers from January 2015 to December 2024 were retrospectively analyzed. Data collected included preoperative (demographics, comorbidities, Charlson Comorbidity Index [CCI], symptoms), intraoperative (pathology, details of surgery, duration of surgery) and postoperative variables. Primary outcome measures included postoperative morbidity or mortality either within 30 days or within the same admission, whichever was longer. Complications were classified using the Clavien-Dindo (CD) classification, where grades I and II comprise medically-managed complications, grade III comprises complications requiring procedural or surgical intervention, grade IV comprises complications involving single or multi-organ failure, and grade V refers to patient death. Secondary outcome measures included length of stay and discharge location. Results: 211 patients comprising 79 males and 132 females were included. Median age was 61 years (interquartile range [IQR] 50-70, range 18-94). There were 53 (25.1%) and 13 (6.2%) patients above the ages of 70 and 80 years, respectively. Overall, complications occurred in 36 patients (17.1%). These included 28 (13.3%) CD I or II, 4 (1.9%) CD III, 2 (0.9%) CD IV and 2 (0.9%) CD V complications. Age and CCI were not found to be significant risk factors of postoperative complications (p = 0.454 and p = 0.209, respectively). Patients above the age of 80 years old had similar complication rates both in general and across various CD grades compared to those below 80 (p = 0.554 for all complications, p = 0.817 for CD I or II, p = 0.994 for CD III, p = 0.053 for CD IV, and p = 0.996 for CD V complications). Likewise, complication rates were similar between patients above 70 years old and those below 70 (p = 0.391 for all complications, p = 0.345 for CD I or II, p = 0.994 for CD III, p = 0.437 for CD IV, and p = 0.437 for CD V complications). Age was not significantly correlated with length of hospital stay (p = 0.239). However, patients with older age were more likely to discharge to rehabilitation facilities or nursing homes (p < 0.001). Conclusion: While advanced age predicted the need for eventual discharge to rehabilitation facilities or nursing homes, age was not a significant predictor of postoperative complications both in general and across various CD grades. The decision for IDEM tumor resection in elderly patients should encompass considerations on eventual care needs and discharge destination. Age alone should not be a contraindication to IDEM tumor resection surgery in elderly patients.

ID: AGS26/2350

A317: Pre-operative embolization does not reduce blood loss in surgery for hypervascular spinal tumors

Mohammad Daher 1 , Ethan Cottrill 1 , Amer Sebaaly 2 , Peter Passias 1 , Alan Daniels 3 , William Eward 1

1 Duke University, Durham, United States

2 Hotel Dieu de France, Beirut, Lebanon

3 Brown University, Providence, United States

Introduction: Hypervascular tumors present a surgical challenge, due to their substantial intraoperative blood loss. Although preoperative embolization is often employed to mitigate intraoperative bleeding, its consistent advantage has not been conclusively demonstrated across existing studies. The purpose of this meta-analysis is to appraise the evidence comparing surgical outcomes with and without preoperative embolization. Material and Methods: Following the PRISMA guidelines, PubMed, Cochrane, and Google Scholar (pages 1-20) were accessed and explored until May 2025. Articles were included if they reported comparative studies evaluating perioperative outcomes of preoperative embolization (E) versus no embolization (NE) in the surgical treatment of hypervascular spinal tumors. A sub-analysis was performed based on whether studies controlled for operative characteristics between the two groups. Results: Fifteen studies met the inclusion criteria, including 225 patients in group NE and 340 in group E. When all studies were analyzed collectively, no significant differences were observed between NE and E for any of the outcomes. However, in the subgroup of studies that did not control for specific surgical variables, the NE group showed greater blood loss (mean difference = 270.60 ml; 95% CI: 21.52- 519.67, p = 0.03) and longer operative time (mean difference = 48.00 minutes; 95% CI: 16.67- 79.33, p = 0.003). In contrast, studies that controlled for these variables showed no significant differences in blood loss (p = 0.75) or operative time (p = 0.41) between the NE and E groups. Conclusion: Pre-operative embolization was not associated with consistent benefits in surgical outcomes for hypervascular spinal tumors. While some studies reported reduced blood loss and operative time with embolization, these findings were limited to analyses lacking control for specific surgical characteristics. This suggests that the perceived advantages may reflect confounding factors rather than the true effect of embolization. However, surgical equipoise remains and the decision to embolize should be individualized based on surgical and patient-related factors.

ID: AGS26/2382

A318: Intralesional surgery with adjuvant charged-particle therapy in the treatment of intact and recurrent primary spinal tumors

Valerio Pipola 1 , Emanuela Cosentino 1 , Cristiana Griffoni 1 , Monica Cosentino 2 , Laura Maria Beatrice Belotti 2 , Riccardo Ghermandi 1 , Marco Girolami 1,3 , Stefano Bandiera 1 , Giovanni Barbanti Brodano 1 , Silvia Terzi 1 , Giuseppe Tedesco 1 , Gisberto Evangelisti 1 , Stefano Pasini 1 , Luigi Falzetti 1 , Giovanni Tosini 1 , Chiara Cini 1 , Maria Rosaria Fiore 4 , Marco Cianchetti 5 , Alessandro Gasbarrini 1,6

1 Complex Spinal Surgery Unit, IRCCS Istituto Ortopedico Rizzoli, Bologna, Italy,

2 Scientific Direction, IRCCS Istituto Ortopedico Rizzoli, Bologna, Italy,

3 Spine Unit, Hospital Universitario Fundacion Jimenez Diaz, Madrid, Spain,

4 Radiation Oncology Unit, Clinical Department, CNAO Centro Nazionale di Adroterapia Oncologica, Pavia, Italy,

5 Proton Therapy Unit, Azienda Provinciale per i Servizi Sanitari, Trento, Italy,

6 Department of Biomedical and Neuromotor Sciences, University of Bologna, Bologna, Italy

Introduction: The management of primary tumors of the mobile spine has long been a challenge in oncological spinal surgery. Historically, en bloc resection with wide or marginal margins has been considered the gold standard for achieving local control and reducing recurrence rates, particularly for aggressive benign and malignant tumors. Recent advancements in adjuvant radiotherapy, particularly charged-particle therapies such as proton therapy and carbon ion radiotherapy, have begun to shift this paradigm. Complementing these advancements in radiotherapy, the development of carbon fiber implants has further enhanced the feasibility and safety of intralesional surgery. Unlike traditional metallic implants, carbon fiber-reinforced polyether ether ketone (CFR-PEEK) implants are radiolucent, minimizing artifacts on postoperative imaging and allowing for more accurate radiation planning and delivery. Methods: The study is a retrospective analysis of data prospectively collected. It includes patients affected by primary tumors of the mobile spine who were treated by intralesional surgery at a tertiary center specialized for spinal diseases between March 2015 and February 2024. Demographic and clinical information, surgical data, complications, and local recurrences were collected. After the intralesional surgery, all patients underwent charged- particle radiotherapy, Carbon ion radiation therapy (CIRT) or Proton Beam Therapy (PBT), at specialized external centers. Patients were followed up longitudinally through clinical visits and imaging studies. Local recurrences (LR) after surgery were monitored and recorded. Results: A total of 36 patients were included in the study: 24 males and 12 females, with a mean age of 57.5 years at the time of surgery (range 11-83 years). The most frequent tumor type was chordoma (26 cases, 72.2%), and the most frequent localization was lumbar (44.4%). Most patients (27/36, 75%) had already undergone prior surgeries for the same lesion: 20 patients (58.3%) were re-operated because of one or more local recurrences before the index surgery; 7 patients were re-operated because of a previous Enneking-inappropriate (EI) treatment without LR. Nine patients (25%) had virgin tumors at the time of index surgery. Patients underwent intralesional excision (19, 52.8%) or intralesional vertebrectomy (6, 16.7%) with CFR-PEEK posterior instrumentation for most patients (27, 75%). All patients underwent charged-particle therapy after the intralesional surgery: 27 patients (75%) were treated by PBT and 9 patients (25%) were treated by CIRT. The survival probability at 3 years of follow-up was 87.7% (95% CI: 77.1-99.8), decreasing to 57.6% (95% CI: 40.7-81.4) at 5 years. Potential risk factors for 5-year mortality were also evaluated, and only the occurrence of LR at any point during follow-up emerged as a significant predictor (HR = 10.6 [95% CI 2.26,49.3]). Fifteen patients (41.7%) had one or more post-operative local recurrences after the index surgery and 12 of them were re-operated. Thirteen out of 15 patients (86.7%) with post-operative LR had also local recurrence before the index surgery. Univariate regression analysis did not identify pre-operative LR as a significant predictor of LR occurring during the follow up period. Conclusion: Intralesional surgery followed by charged-particle radiotherapy provides good results in terms of local control and survival rates, even in cases of recurrent tumors.

ID: AGS26/2567

A319: Symptom-to-surgeon interval in metastatic spinal cord compression: a single-center retrospective analysis

Tiago Ribeiro da Costa 1,2 , António Pinto 1 , Filipe Vaz da Silva 1 , Gonçalo Figueiredo 1 , Eduardo Cunha 1 , Luís Rocha 1 , Joaquim Reis 1,2 , Alfredo Calheiros 1,2

1 Unidade Local de Saúde de Santo António, Porto, Portugal

2 Instituto de Ciências Biomédicas Abel Salazar, University of Porto, Porto, Portugal

Introduction: Metastatic spinal cord compression (MSCC) is a frequent neurosurgical emergency with potentially catastrophic neurological outcomes if not managed promptly. Although early surgical evaluation and decompression-within 24 to 48 hours of symptom onset-are generally considered critical to optimize functional outcomes, the actual timeframes achieved in clinical practice are not well characterized. Clinical perception suggests that significant delays in referral and assessment may occur. This study aimed to objectively evaluate the interval between the onset of neurological symptoms and the first contact with a spine surgeon, in patients undergoing urgent decompression for metastatic epidural spinal cord compression. Material and Methods: We retrospectively reviewed all patients in whom surgical decompression for MSCC was considered at our institution between January 2018 and September 2025. Demographic data, tumor type, neurological status at presentation, and referral pathway were collected. The main outcome was the time from symptom onset (defined as first neurological deficit compatible with acute myelopathy) to first contact with a spine surgeon (neurosurgeon or orthopedic spine surgeon). Patients presenting with complete paraplegia (ASIA A) and with symptom-to-contact intervals exceeding 48 hours were not included, as surgery was not considered in this subgroup. Results: A total of 60 patients were identified, of whom 2 had insufficient data to determine symptom-to-surgeon interval. Among the remaining 58 patients, 42 (72.4%) presented with an interval longer than 72 hours, while 16 (27.6%) were assessed within 72 hours of symptom onset. In this subgroup, the mean interval was 26.7 hours (range 0-60h). Nine patients (15.5%) were seen within 24 hours, and twelve (20.7%) within 48 hours. Conclusion: This study provides an objective characterization of referral times in MSCC. Fewer than one-third of patients considered for surgery were assessed by a spine surgeon within 72 hours, and only one in five within 48 hours. The exclusion of patients with complete paraplegia after delayed referral represents a methodological limitation, likely leading to an underestimation of the true burden of delayed pathways. Future prospective and multicentric registries should aim to capture all patients with MSCC, and to analyze referral determinants such as triage processes and first-point recognition, to optimize timely referral and preserve neurological outcomes.

ID: AGS26/1257

A320: Mapping surgical invasiveness in oncologic surgery for spinal metastases: a scoping review

Rafael De la Garza Ramos 1 , C. Rory Goodwin 2 , Michael Weber 3 , Markian Pahuta 4 , Shalin Patel 5 , Mark Maclean 6 , Laurence Rhines 5 , Daniel Sciubba 7 , Cordula Netzer 8 , Nicolas Dea 9 , Jorrit-Jan Verlaan 10 , Alessandro Gasbarrini 11 , Jeremy Reynolds 12 , Ori Barzilai 13 , Chetan Bettegowda 14 , Stefano Boriani 15 , Charles Fisher 16 , Ziya Gokaslan 17 , Áron Lazáry 18 , Ilya Laufer 19 , John Shin 20 , Raphaële Charest-Morin 16

1 Montefiore Einstein, New York, United States

2 Duke University, Durham, United States

3 University of Connecticut, Farmington, United States

4 McMaster University, Hamilton, Canada

5 MD Anderson Cancer Center, Houston, United States

6 University of Pittsburgh School of Medicine, Pittsburgh, United States

7 Northwell Health, Manhasset, United States

8 University Hospital of Basel, Basel, Switzerland

9 Combined Neurosurgical and Orthopedic Spine Program, Vancouver, Canada

10 University Medical Center Utrecht, Utrecht, Netherlands

11 IRCCS Istituto Ortopedico Rizzoli, Bologna, Italy

12 Oxford University Hospitals NHS Foundation Trust, Oxford, United Kingdom

13 Memorial Sloan Kettering Cancer Center, New York, United States

14 Johns Hopkins Hosital, Baltimore, United States

15 University of Bologna, Bologna, Italy

16 University of British Columbia, Vancouver, Canada

17 Brown University, Providence, United States

18 National Center for Spinal Disorders, Budapest, Hungary

19 New York University Grossman School of Medicine, New York, United States

20 University of Pennsylvania, Philadelphia, United States

Introduction: Surgery for spinal metastases encompasses a broad spectrum of procedures, from percutaneous fixation to en bloc resection in select cases. However, no standardized framework exists to describe surgical invasiveness in this population. Existing indices were not designed for metastatic spine disease and do not incorporate modern minimally invasive, lateral, or hybrid approaches. Prior attempts at invasiveness scores remain unvalidated and not widely adopted. This review sought to map how invasiveness has been defined and measured in the current literature on metastatic spinal tumor surgery (MSTS). Material and Methods: A literature search was performed by searching the PubMed/Medline database for papers published up to February 29, 2023. The search algorithm included terms related to operative variables, technique, and factors and their association with complications and/or adverse events in MSTS. We evaluated which operative variables were studied, which outcomes they were linked to, and how these associations were reported. An exploratory data analysis and narrative synthesis were conducted. Results: A total of 78 articles representing 86,353 patients met our final inclusion criteria. Eleven different groups of covariates were found, with the most common being surgical technique, number of levels, approach, and operative time. The most common examined outcomes included complications, length of stay, and operative time. Two-thirds of studies (68%; 53 of 78) found a positive association between operative variables and the studied outcome(s). Most of these studies supported an association between more aggressive/invasive techniques (en bloc resection, corpectomy, chest wall resection), longer operations (both in number of levels and operative time), combined approaches, and emergent procedures with complications after spine tumor surgery. However, only 53% of these studies (28 of 53) performed a multivariable analysis. Conclusion: Surgical invasiveness in MSTS is inconsistently defined and measured. Operative variables are frequently associated with adverse events, but no validated framework exists. These findings underscore the need to develop a modern, comprehensive invasiveness index that incorporates both open and minimally invasive techniques and provides a common language for describing surgical extent in this population.

ID: AGS26/2796

A321: Statistical description of relevant mechanical parameters of metastatic vertebrae

Silvia Casadei 1 , Giulia Cavazzoni 1 , Margherita Pasini 1 , Luca Cristofolini 1 , Saulo Martelli 2 , Marco Palanca 1

1 Department of Industrial Engineering-University of Bologna, Bologna, Italy

2 Faculty of Engineering, School of Mechanical, Medical & Process Engineering-Queensland University, Brisbane, Australia

Introduction: Bone metastases compromise the stability of the spine, increasing the risk of fracture [1]. Currently, the mechanisms of failure in metastatic vertebrae and the influence of metastasis type or location are still unclear. The Spine Instability Neoplastic Score (SINS) is the most widely used clinical tool for fracture risk assessment in metastatic spine disease, but it presents several limitations [2]. This study had two aims: (1) to identify the principal sources of variation in shape and material properties in vertebrae with lytic metastases, by using statistical methods, and (2) to evaluate if and how these components may explain the mechanical behaviour under compressive loading conditions. Material and Methods: Twenty-eight human cadaveric vertebrae with lytic metastases were mechanically tested under axial compression in the elastic regime, and full-field surface displacements and strains were measured with a four-camera 3D Digital Image Correlation system (Aramis Adjustable, Zeiss). The same specimens were CT-scanned (in-plane resolution = 0.39 mm, slice thickness = 0.625 mm). The vertebral body and the metastatic regions were segmented to generate surface meshes. Volumetric meshes were mapped to the original CT images to extract local density values from the Hounsfield Units, which were subsequently converted into elastic modulus values [3]. The lytic regions were also mapped onto the volumetric meshes. Principal component analysis (PCA) was applied to capture the main source of variation in vertebral shape and material properties. PCA is a statistical technique that reduces the complexity of the data by maximizing the variance explained and making possible to highlight the “main associations” of variability of bone shape and quality. These main associations were then used in a linear regression model to predict the minimum compressive strain and the percentage strain difference respect to control vertebrae under compressive loading. Model generalization was assessed using leave-one-out cross-validation, and performance was quantified by the root mean square error (RMSE) and the coefficient of determination (R 2 ). Results: Principal component analysis identified 25 parameters that accounted for 95% of the variation in shape and density. Among the 25 principal components, only 15 were found to accurately predict the mechanical behaviour under compressive loading conditions, by minimizing the RMSE. The linear regression model predicting minimum compressive strain showed a median RMSE below 100 µε (strain range -1000 -9400 µε, median percentage error of 5%) and a R 2 was 0.95. The percentage compressive strain difference median RMSE error was 34% (percentage range -39 -1135 % and a median percentage error of 14%) and R 2 was 0.93. Conclusion: This study demonstrates that statistical models of vertebral shape and material properties can extract principal components to capture the main sources of variation, while accurately predicting compressive strain in vertebrae with lytic metastases. Since strain is one indicator of fracture risk [1], these results highlight that a larger set of specific factors, both morphological and material, provide stronger predictive power than the limited parameters considered in currently clinical assessments.

References

1. Palanca. Bone (2023)

2. Palanca. Bone (2021)

3. Garavelli. PLoS One (2022)

Acknowledgements: HEU-Funded METASTRA project (grant 101080135).

ID: AGS26/382

A322: Pseudarthrosis risk in patients undergoing spinal fusion for multiple myeloma-associated vertebral body lesions

Catherine Ravikumar 1 , Nishanth Krishnan 1 , Alysha Jamieson 1 , Arohi Chiprutkar 2 , Hunter Yamada 1 , Alexander Aabedi 1 , Nathan Han 3 , Niall Buckley 1 , Jay Kumar 1 , Praveen Mummaneni 1 , Blake Taylor 1

1 University of California, San Francisco, San Francisco, United States

2 University of California, Berkeley, Berkeley, United States

3 Hoag Hospital, Irvine, United States

Introduction: Vertebral lesions are a common finding in patients with occult or newly diagnosed multiple myeloma (MM), and approximately 20% of MM-associated compression fractures present with spinal instability. Spinal fusion may be considered for management, but MM patients are thought to have more risk factors for failed fusion, including poor bone quality and advanced age at diagnosis. Understanding the true long-term incidence of re-operation and pseudarthrosis may inform optimal surgical decision-making. We sought to assess the true incidence of wound healing complications and pseudarthrosis rate in MM patients undergoing spinal fusion. Material and Methods: We conducted a single-institution, retrospective review of patients with MM undergoing spinal fusion between 2014 and 2024. Inclusion criteria were adult patients with MM diagnosis confirmed by pre-operative biopsy or intraoperative pathology and a 6-month minimum follow-up (12-month minimum for pseudarthrosis subgroup analysis). Pertinent patient features, including baseline clinical characteristics, pre/post-operative chemotherapy and radiation therapy regimen, and surgical parameters, were obtained. Pseudarthrosis was evaluated by a neurosurgical fellow or attending using CT radiographs. Results: 46 patients were reviewed (mean age ± SD: 60.94 ± 13.48, female: 37.0%). MM-associated lesions were most commonly observed in the thoracic spine (41.3%). Mean SINS score was 11.3. Corpectomy was performed in 67.4% of cases, and the mean number of instrumented levels was 5.4 ± 2.5. Three patients had re-operation within 6 months (two cases of wound infection and one case of adjacent segment disease). Patients with at least one-year radiographic follow up (n= 28) were evaluated for pseudarthrosis (median follow-up time: 17.5 months). Four patients were observed to have non-clinically significant screw lucencies on CT follow-up. Two patients had rod fractures requiring re-operation, 20 and 21 months post-operatively. Conclusion: In this single institution case series, we provide the first estimate of clinically significant pseudarthrosis in MM patients as 7.1%.

ID: AGS26/430

A323: Vertebral artery trapping in metastatic cervical spine tumors: propensity-matched analysis of surgical outcomes

Wei-Lin Pan 1 , Yen-Heng Lin 1 , Fon-Yih Tsuang 1

1 National Taiwan University Hospital, Taipei, Taiwan

Introduction: Vertebral artery (VA) trapping during metastatic cervical spine tumor surgery is technically demanding, and its safety profile remains debated. While it facilitates more extensive tumor resection, evidence regarding perioperative outcomes compared with non-VA trapping procedures is limited. Material and Methods: We retrospectively analyzed 73 patients who underwent surgery for metastatic cervical spinal tumors at our institution, including 12 with VA trapping and 61 without. Surgical methods, intraoperative parameters, and perioperative outcomes were compared. To minimize selection bias, 1:1 propensity score matching (PSM) was performed using Tomita, SINS, Tokuhashi, and ESCCnew scores to balance preoperative tumor burden and spinal stability. Results: Debulking surgery was significantly more frequent in the VA trapping group than in the non-VA trapping group. Despite this, no significant differences were observed in estimated blood loss, operative time, hospital stay, 30-day complications, reoperation, massive bleeding events (> 1000 mL), or 90-day mortality. After PSM, patients in the VA trapping cohort continued to undergo more debulking procedures despite similar preoperative characteristics, yet perioperative outcomes remained comparable to the matched non-VA trapping cohort. A trend toward greater use of combined surgical approaches was also observed in the VA trapping group, though not statistically significant. Conclusion: VA trapping allows surgeons to pursue more aggressive resection strategies in metastatic cervical spine tumor surgery without compromising perioperative safety. These findings suggest that VA trapping can be safely considered in selected patients when radical tumor control is indicated.

ID: AGS26/446

A324: Predictive factors for spinal deformity following resection of intramedullary spinal cord tumors: a 15-year retrospective cohort study

Vitor Viana Bonan de Aguiar 1 , Caio Perret 1 , Ricardo de Amoreira Gepp 1

1 Sarah Network of Rehabilitation Hospitals, Brasilia, Brazil

Introduction: Spinal deformity is a significant long-term complication that can occur after the resection of intramedullary spinal cord tumors (IMSCT), especially in younger patients and in cases where surgical instrumentation is not used. Although there are various reports identifying potential risk factors, the specific roles of these factors in the development of deformities-such as kyphosis or scoliosis-in both the sagittal and coronal planes remain underexplored. The objective of this study is to clearly identify the clinical, anatomical, and surgical predictors of spinal deformity following IMSCT resection in patients who do not undergo routine fusion. Material and Methods: We conducted a retrospective cohort study involving consecutive patients who underwent surgical resection of histologically confirmed intramedullary spinal cord tumors at a single tertiary neurosurgical center between January 2000 and June 2025. Eligible participants were aged 21 years or younger, had complete preoperative and postoperative imaging, and had at least 12 months of follow-up data. Posterior decompression was carried out through either laminectomy or laminoplasty, without prophylactic fusion. We analyzed demographic, clinical, surgical, and radiographic variables. The primary outcome measured was the development of new radiographic coronal or sagittal deformity. Both univariate and multivariable penalized logistic regression analyses were employed to identify independent predictors. Results: A total of sixty-eight patients met the inclusion criteria for the study. Postoperative coronal deformity was observed more frequently in younger patients, with a mean age of 10.8 ± 5.3 years compared to 13.4 ± 6.3 years in older patients (p = 0.001). Additionally, the occurrence of postoperative deformity was linked to preoperative sagittal deformity (p = 0.002), the location of the tumor in the thoracic region (p = 0.009), and the performance of a laminectomy (p = 0.015). In the multivariable analysis, younger age (odds ratio [OR] 0.80; 95% confidence interval [CI] 0.76-0.84; p = 0.001), preoperative sagittal deformity (OR 12.44; 95% CI 1.68-91.95; p = 0.001), and thoracic tumor location (OR 4.53; 95% CI 1.02-20.09; p = 0.047) were identified as independent predictors of postoperative deformity. Conclusion: In this 25-year, single-center study, several factors were found to be independently associated with the development of postoperative coronal deformity following the resection of intramedullary spinal cord tumors, without preventive fusion. These factors include younger age, preoperative sagittal deformity, and the location of thoracic tumors. The findings underscore the importance of conducting preoperative risk assessments and ensuring long-term radiographic monitoring, particularly for skeletally immature patients and those with high-risk anatomical characteristics.

OP37: Navigation and Robotics in Spine

ID: AGS26/1199

A325: Clinical outcomes of dual-pin reference frame placement in MI-TLIF: a retrospective review of 266 pins using a three-zone grading system

Chibuikem Ikwuegbuenyi 1 , Noah Willett 1 , Mousa Hamad 1 , Andreas Kramer 1 , Rachel Bratescu 1 , Natasha Kharas 1 , Galal Elsayed 1 , Ibrahim Hussain 1 , Osama Kashlan 1 , Roger Härtl 1

1 Department of Neurological Surgery, Och Spine at New York-Presbyterian/Weill Cornell Medical Center, New York, United States

Introduction: Percutaneous dual-pin reference frames enable image-guided minimally invasive transforaminal lumbar interbody fusion (MI-TLIF) but carry risks of pin-related morbidity, notably sacroiliac (SI) joint dysfunction. While single-pin systems have been studied, clinical data on dual-pin systems remain limited. We report our experience identifying complications related to dual-pin placement and evaluating their distribution using a three-zone anatomical grading system that defines optimal and suboptimal pin trajectories. Materials and Methods: We retrospectively reviewed 133 consecutive patients who underwent MI-TLIF with intraoperative CT-guided dual-pin placement between January 2019 and July 2023, with at least one month of clinical follow-up. Pin-related complications assessed included new SI joint pain, persistent pin-site pain, hematoma, infection, fracture, and visceral injury. Two independent, blinded spine surgeons graded 266 total pins into one of three anatomical zones: Green zone: medial PSIS/ilium; Yellow zone: abutting or breaching the SI joint or with inadequate bony purchase; Red zone: sacrum, neural foramen, or pelvic cavity. Interrater reliability was calculated using quadratic-weighted Cohen’s κ. Results: Twelve patients (9.0%) experienced pin-related morbidity: 11 (8.3%) developed new SI joint pain, but only 2 had pins that penetrated the SI joint on the symptomatic side. One patient (0.8%) reported persistent pin-site pain. No hematomas, infections, fractures, or visceral injuries were observed. All complications occurred exclusively in the yellow or red zones. Among the 266 reference frame pins evaluated, 192 (72%) were in the green zone, 64 (24%) in the yellow zone, and 10 (4%) in the red zone. Interrater reliability for zone assignment was excellent (quadratic-weighted Cohen’s κ = 0.89, p < 0.001). Conclusion: Dual-pin RF placement in MI-TLIF demonstrated a low complication rate, with SI joint dysfunction nearly exclusively associated with suboptimal (yellow or red zone) placement. The green-zone placement was complication-free. Strict adherence to green-zone anatomical guidelines may reduce morbidity and improve the safety of navigated minimally invasive spine surgery.

ID: AGS26/3370

A326: 2D vs 3D computer-assisted navigation in thoracolumbar pedicle screw placement. updated systematic review with meta-analysis of comparative clinical studies

Waeel Hamouda 1 , Stipe Corluka 2 , Mohamed Soliman 3 , Stjepan Dokuzovic 2 , Alexander Aguirre 4 , Sathish Muthu 5 , Mohamed Ali 6 , Samuel Cho 7 , Amit Jain 8 , AO Spine Knowledge Forum Degenerative 9

1 Neurosurgery department - Security Forces Hospital Dammam - Saudi Arabia, Neurosurgery department - Faculty of Medicine - Cairo University - Egypt, Dammam, Saudi Arabia

2 Spinal Surgery Division, Department of Traumatology, University Hospital Centre Sestre Milosrdnice, Zagreb, Croatia., St. Catherine Specialty Hospital, Zagreb, Croatia., Department of Anatomy and Physiology, University of Applied Health Sciences, Zagreb, Croatia, Zagreb, Croatia

3 Neurosurgery Department, Cairo University, Cairo, Egypt, Neurosurgery Department, University at buffalo, Buffalo, NY, United states, Cairo, Egypt

4 Neurosurgery Department, University at Buffalo, Buffalo, NY, United states, Buffalo, United States

5 Department of Spine Surgery, Orthopaedic Research Group, Coimbatore, Tamil Nadu, India, Central Research Laboratory, Meenakshi Medical College Hospital and Research Institute, Meenakshi Academy of Higher Education and Research, Chennai, India, Tamil Nadu, India

6 Department of Neurosurgery, Prince Mohamed ben Abdulaziz Hospital, Riyadh, Saudi Arabia, Riyadh, Saudi Arabia

7 Department of Orthopedic Surgery, Icahn School of Medicine at Mount Sinai, New York, NY, USA, New York, United States

8 Department of Orthopaedic Surgery, Johns Hopkins University, Baltimore, USA, Baltimore, United States

9 AO Spine International, AO Foundation, Davos, Switzerland

Introduction: Computer-assisted navigation (CAN) for thoracolumbar pedicle screw (TL-PS) placement resulted in increased placement accuracy and lower complication rates. Two-dimensional (2D) navigation uses C-arm-acquired intraoperative anteroposterior and lateral images, while Three-dimensional (3D) navigation uses 360 images obtained from preoperative CT or intraoperative 3D fluoroscopy. The aim of this research is to analyze the accuracy and safety of 2D vs 3D CAN for TL-PS reported in published comparative studies. Material and Methods: A systematic analysis was conducted to examine TL-PS accuracy in clinical studies comparing 2D to 3D CAN. PubMed and 5 other databases were searched through March 2024 for studies published since 2000 in English. Prevalence P and 95% confidence intervals CI were used to report TL-PS accuracy as segregated by distance of screw deviation outside the pedicle cortex using a 2mm grading criteria (Grade I no breach, Grade II less than 2mm, Grade III from 2 to 4 mm, and Grade IV more than 4 mm). Results: Four studies were identified reporting collectively on 428 screws placed in 85 patients using 2D CAN compared to 441 screws placed in 83 patients using 3D CAN. In comparison to 3D cases, 2D cases showed 62 misplaced screws [vs 32, (P: 1.09- 95%CI: 0.67,1.76) with 33 G II breach [vs 25 (P: 0.76- 95%CI: 0.44,1.33) and 19 G III breach [vs 7 (P: 2.45- 95%CI: 0.96,6.29)], with no G IV breach. No complications were reported related to the misplaced screws. Conclusion: The results of this meta-analysis showed that there is no statistical significance in pedicle screw placement accuracy in the thoracolumbar spine if either 2D or 3D computer-assisted navigation systems are utilized although 3D navigation tends to reduce occurrence of misplacements, especially of higher grades.

ID: AGS26/3085

A327: Trends in navigation-guided spinal instrumentation surgeries in the United States

Maxwell Sahhar 1 , Ahmed Sulieman 2 , Ishandeep Singh 3 , Sang Hun Lee 2

1 Brown University Warren Alpert Medical School, Providence, United States

2 Johns Hopkins University Department of Orthopedic Surgery, Baltimore, United States

3 University of East Anglia Norwich Medical School, Norfolk, United Kingdom

Introduction: Navigation-guided instrumentation has been increasingly applied in spine surgery globally as well as in the United States. This study evaluates national adoption trends and reports descriptive, population-level outcomes across pathology types and spine regions. Material and Methods: A retrospective analysis was performed using the MarketScan Claims Database on 158,168 spine fusion surgeries between 2011 and 2020, including 15,382 navigated and 142,786 non-navigated cases. Demographics were summarized, and adoption trends were compared before and after 2016. Outcomes were assessed using unadjusted 2-year reoperation rates and mean hospital length of stay, stratified by pathology and spinal region indicated by ICD-10 Codes. Results: The mean age was 49.6 years with 46.5% being female in navigated cases and 48.9 years with 44.7% being female in controls. Navigation adoption increased significantly after 2016 across all pathologies (p < 0.001). Deformity cases rose from 6.2% pre-2016 to 14.6% post-2016, degenerative from 6.2% to 15.9%, trauma from 6.5% to 17.2%, and neoplastic from 9.2% to 25.1%. By spinal region, adoption increased from 4.3% to 12.0% in cervical, 7.0% to 20.0% in thoracic, and 6.3% to 16.4% in lumbar procedures. Crude outcome comparisons showed no significant differences in 2-year reoperation rates between navigated and non-navigated cohorts. In deformity surgery, reoperation occurred in 3.86% of navigated cases versus 4.32% of controls (p = 0.41), with mean length of stay of 4.0 versus 4.2 days (p = 0.32). Degenerative surgery showed reoperation in 3.62% versus 3.87% (p = 0.47), with mean stay of 3.5 days in both groups (p = 0.88). Trauma cases had reoperation in 2.23% versus 2.43% (p = 0.56) and mean stay of 5.8 versus 6.5 days (p = 0.04). Neoplastic procedures showed reoperation in 2.53% versus 2.92% (p = 0.38), with mean stay of 5.8 versus 6.8 days (p = 0.02). Kyphosis cases had reoperation in 5.19% versus 5.68% (p = 0.44) and scoliosis 4.23% versus 4.47% (p = 0.51), with mean stays of 4.9 versus 5.0 (p = 0.62) and 4.2 versus 4.5 days (p = 0.21), respectively. Conclusion: Between 2011 and 2020, computer navigation in spine surgery increased significantly across all pathology groups and spinal regions, with the largest growth observed in neoplastic, thoracic, and lumbar cases. At the population level, crude reoperation rates were not significantly different between navigated and non-navigated cohorts. Length of stay was shorter for navigated trauma and neoplastic cases, while differences in other groups were not statistically significant. These findings demonstrate the growing role of navigation in spine surgery, especially in surgeries that require precise localization in highly unpredictable anatomy such as neoplastic or trauma cases.

ID: AGS26/2595

A328: Trajectory mismatch between navigation-guided and anatomically ideal C3-6 cervical pedicle screws: clinical implications

Yoshitaka Nagashima 1

1 Department of Neurosurgery, Nagoya University Graduate School of Medicine., Nagoya, Japan

Introduction: Navigation systems have facilitated safer placement of C3-6 cervical pedicle screws. However, the ideal anatomical trajectory remains poorly defined, and the extent to which actual screw paths deviate from this ideal has not been systematically evaluated. Material and Methods: We retrospectively analyzed patients who underwent C3-6 pedicle screw placement with either preoperative CT-based navigation or intraoperative O-arm navigation between 2019 and 2024 at three institutions. Cases with alternative skin entry were excluded. Pre- and postoperative CT scans were fused on a workstation to reconstruct the actual screw trajectory and the anatomically ideal trajectory. Entry point distance from the lateral border of the lateral mass, medial placement ratio, and screw angle were compared between trajectories. Screw perforation was graded according to the Neo classification, and parameters were compared between screws with and without perforation. Results: A total of 127 screws were placed in 33 patients (66 with CT-based navigation and 61 with O-arm navigation). Screw distribution was C3: 29, C4: 31, C5: 35, and C6: 32. Perforation grading was Grade 0: 79.5%, Grade 1: 15.7%, Grade 2: 4.7%, and Grade 3: 0%. Compared with the ideal trajectory, actual screws demonstrated significantly more medial entry (median 6.4 mm vs. 2.3 mm), higher medial ratio (0.480 vs. 0.183), and smaller insertion angle (28.0° vs. 44.3°) (all p < 0.001). Screws with perforation (Grades 1-2) had greater medial entry distance (8.0 mm vs. 6.1 mm) and medial ratio (0.582 vs. 0.452) than those without perforation (both p < 0.001), while insertion angle did not differ significantly (p = 0.20). Conclusion: Navigation-guided C3-6 cervical pedicle screws were consistently placed more medially and at shallower angles than the anatomically ideal trajectory. Excessively medial entry points, especially those medial to the midpoint of the lateral mass, increased the risk of perforation. These findings underscore a specific pitfall of navigation-guided screw placement that must be considered to enhance accuracy and safety.

ID: AGS26/2799

A329: New innovation for spine navigation

Ufuk Aydinli 1 , Zlatko Kolic 2 , Duje Vukad 3 , Scott Hook 4 , Emir Kamaric 4

1 Acibadem Hospital, Bursa, Türkiye

2 Hospital Pula, Pula, Croatia

3 Deparment Neurosurgery, Rijeka, Croatia

4 CDI, Houston, United States

Introduction: Image-guided navigation in spinal interventions and related procedures commonly relies on fluoroscopy or CT imaging, which are costly, expose patients and clinicians to radiation, and are often limited by complex intraoperative registration challenges. Material and Methods: We present a novel computer navigation system designed to overcome these limitations through real-time, radiation-free, and contactless surgical guidance. This new system integrates proprietary computer vision algorithms with streamlined hardware to enable accurate anatomical landmark identification, patient registration, and trajectory correction. The system supports hybrid electromagnetic (EM) and computer vision (CV) navigation and includes optional real-time ultrasound verification. Its compact design, intuitive interface, and minimal training requirements make it suitable for widespread clinical adoption. Results: Compared with conventional navigation technologies that depend on bulky equipment and prolonged calibration, New system provides precise guidance for implants, needles, and instruments, while reducing operating room time, cost, and radiation exposure. The system is particularly impactful in spinal interventions such as vertebroplasty, minimally invasive screw placement, and needle-guided drug delivery. Conclusion: This new system represents a transformative, low-cost, and portable solution that enhances safety, accuracy, and efficiency in minimally invasive spine and pain management procedures. Its integration of EM and CV technologies with contactless registration addresses key unmet needs in the field and supports scalable adoption across diverse healthcare settings.

ID: AGS26/2920

A330: Thoracolumbar pedicle screws misplacement using 2D computer-assisted navigation. A systematic review and meta-analysis with reported avoidance techniques

Waeel Hamouda 1 , Stipe Corluka 2 , Mohamed Soliman 3 , Stjepan Dokuzovic 4 , Alexander Aguirre 5 , Sathish Muthu 6 , Mohamed Ali 7 , Samuel Cho 8 , Amit Jain 9 , AO Spine Knowledge Forum Degenerative 10

1 Neurosurgery department - Security Forces Hospital Dammam - Saudi Arabia, Neurosurgery department - Faculty of Medicine - Cairo University - Egypt, Dammam, Saudi Arabia

2 Spinal Surgery Division, Department of Traumatology, University Hospital Centre Sestre Milosrdnice, Zagreb, Croatia., St. Catherine Specialty Hospital, Zagreb, Croatia., Department of Anatomy and Physiology, University of Applied Health Sciences, Zagreb, Croatia, Zagreb, Croatia

3 Neurosurgery Department, Cairo University, Cairo, Egypt, Neurosurgery Department, University at buffalo, Buffalo, NY, United states, Cairo, Egypt

4 Department of Traumatology, University Hospital Centre Sestre ilosrdnice, Zagreb, Croatia., Zagreb, Croatia

5 Neurosurgery Department, University at Buffalo, Buffalo, NY, United states, Buffalo, United States

6 Department of Spine Surgery, Orthopaedic Research Group, Coimbatore, Tamil Nadu, India, Central Research Laboratory, Meenakshi Medical College Hospital and Research Institute, Meenakshi Academy of Higher Education and Research, Chennai, India, Tamil Nadu, India

7 Department of Neurosurgery, Prince Mohamed ben Abdulaziz Hospital, Riyadh, Saudi Arabia, Riyadh, Saudi Arabia

8 Department of Orthopedic Surgery, Icahn School of Medicine at Mount Sinai, New York, NY, USA, New York, United States

9 Department of Orthopaedic Surgery, Johns Hopkins University, Baltimore, USA, Baltimore, United States

10 AO Spine International, AO Foundation, Davos, Switzerland., Davos, Switzerland

Introduction: Computer-assisted navigation (CAN) for pedicle screw placement in the thoracolumbar (TL) spine resulted in increased accuracy and lower complication rates. Two-dimensional (2D) navigation uses C-arm-acquired intraoperative anteroposterior and lateral images. This study aim to analyze the accuracy and safety of thoracolumbar pedicle screw (TLPS) placement using 2D CAN reported in literature, with review of the reported surgical techniques aimed to increase accuracy. Material and Methods: PubMed, Embsae, ClinicalTrials.gov, and the Cochrane library were searched through February 1st, 2025, for studies on TLPS accuracy with 2D CAN published in 2000 or later in English language. Prevalence (P) and 95% confidence intervals (CI) were used to report pedicle screw accuracy as segregated by distance of screw deviation outside pedicle cortex using a 2mm grading criteria (Grade I no breach, Grade II less than 2mm, Grade III from 2 to 4 mm, and Grade IV more than 4 mm) and by direction of violation whether lateral or medial to pedicle. Results: Six studies were included for analysis with 153 patients and 953 TLPS placement data. 12% (n = 113) of the screws were misplaced (P: 9.36- 95%CI: 5.92,14.49) of which 76% Grade II (n = 85; P: 5.73- 95%CI: 2.64,11.99) followed by 23% Grade III (n = 27; P: 2.33- 95%CI: 1.09,4.88), and 1% Grade IV (n = 1, P: 0.10- 95%CI: 0.01,0.74). There were 54% lateral violations (n = 61; P: 5.18- 95%CI: 1.94,13.15) of which Grade II, III and IV were 72%, 26%, and 2% respectively. There were 34% medial violations (n = 38; P: 4.50- 95%CI: 2.74,7.29) of which Grade II, and III were 84% and 16% respectively. No complications were reported related to the misplaced screws. Conclusion: This is the first available systematic review to investigate and demonstrate that TLPS insertion using 2D CAN appears to have acceptable reliability and safety, with most misplacements being of low grade and lateral to the pedicle and no reported misplacement-related complications.

ID: AGS26/3089

A331: Navigation-guided spinal instrumentation: an exact-matched analysis of outcomes

Maxwell Sahhar 1 , Ahmed Sulieman 2 , Sang Hun Lee 2

1 Brown University Warren Alpert Medical School, Providence, United States , 2 Johns Hopkins University Department of Orthopedic Surgery, Baltimore, United States

Introduction: Previous research has reported modest benefits of navigation in spine surgery reoperation rates but has not adequately addressed confounding factors or examined whether outcomes have improved over time. This study evaluates the association of navigation with postoperative outcomes using a cohort matched exactly on primary procedure, pathology, and spinal region. Material and Methods: A retrospective analysis was performed using the MarketScan Claims Database on patients undergoing spine fusion between 2011 and 2020. Navigated surgeries were identified using the navigation CPT code and matched exactly to non-navigated surgeries by primary procedure, spine region, and pathology type (deformity, degenerative, trauma, and neoplastic, non-mutually exclusive) using ICD-10 and CPT codes. Outcomes included 2-year reoperation, 90-day reoperation, and hospital length of stay (LOS). Logistic and linear regression models controlling for age and sex were also performed to evaluate whether the odds of 2-year reoperation and LOS varied over time. Results: Propensity matching produced 10,847 navigated and 10,847 non-navigated cases. Mean age was 50.0 ± 13.8 years in navigated cases and 49.8 ± 14.0 years in controls, with 45.4% and 44.2% female, respectively. A total of 4,535 navigated cases were excluded for having insufficient data or lacking an exact match. In deformity cases (n = 6,214), 2-year reoperation was 3.57% versus 4.18% (p = 0.21), 90-day reoperation 0.55% versus 0.84% (p = 0.17), and LOS 3.92 versus 4.05 days (p = 0.14). Trauma (n = 880) showed 2-year reoperation 2.27% versus 6.36% (p < 0.001), 90-day reoperation 0.23% versus 1.36% (p = 0.06), and LOS 5.39 versus 5.21 days (p = 0.57). Degenerative cases (n = 19,152) had 2-year reoperation 3.83% versus 4.05% (p = 0.44), 90-day reoperation 0.41% versus 0.57% (p = 0.10), and LOS 3.43 versus 3.31 days (p = 0.001). Neoplastic cases (n = 292) showed 2-year reoperation 2.74% versus 3.42% (p = 0.73), 90-day reoperation 2.05% versus 1.37% (p = 0.65), and LOS 5.75 versus 7.32 days (p = 0.008). By region, cervical (n = 1,738) had 2-year reoperation 5.75% versus 4.49% (p = 0.23), 90-day reoperation 0.69% versus 0.92% (p = 0.59), and LOS 3.79 versus 3.60 days (p = 0.23). Thoracic (n = 944) showed 2-year reoperation 6.14% versus 6.36% (p = 0.89), 90-day reoperation 1.91% versus 1.27% (p = 0.43), and LOS 5.67 versus 5.99 days (p = 0.45). Lumbar (n = 17,436) demonstrated 2-year reoperation 3.57% versus 4.45% (p = 0.12), 90-day reoperation 0.34% versus 0.56% (p = 0.03), and LOS 3.39 versus 3.26 days (p = 0.001). Logistic regression relating time of surgery to odds of reoperation showed no significant association in either navigated (p = 0.436) or non-navigated (p = 0.457) models. Conclusion: In an exact-matched national cohort of 21,694 spine surgeries, navigation was associated with lower reoperation rates in trauma surgery, and with markedly shorter hospital stays in neoplastic cases. Although these benefits were not consistent across all groups, the findings suggest that navigation may confer advantages in specific clinical contexts and demonstrates the need for further investigation into scenarios where navigation technology provides the greatest benefit.

ID: AGS26/377

A332: Accuracy of augmented reality navigation versus freehand pedicle screw placement

Giuseppe Loggia 1 , Aron Alakmeh 1 , Fedan Avrumova 1 , Franziska Altorfer 1 , Jiaqi Zhu 1 , Frank Cammisa 1 , Andrew Sama 1 , Mazda Farshad 2 , Darren R. Lebl 1

1 Hospital for Special Surgery, Balgrist Zurich, New York, United States

2 Balgrist, Zurich, Switzerland

Introduction: Accurate pedicle screw placement is crucial in spinal fusion surgery to ensure optimal fixation and avoid complications. Augmented Reality (AR) navigation is currently emerging as a novel radiation-sparing alternative to conventional freehand (FH) techniques in pedicle screw placement, with potential benefits including improved accuracy, workflow efficiency, and surgical ergonomics. This study aims to compare the accuracy of AR-guided versus conventional FH techniques in lumbosacral pedicle screw placement. Material and Methods: All patients who underwent pedicle screws placement from L1 to S1 using either Xvision® AR navigation or conventional FH technique, with available postoperative CT imaging for screw accuracy assessment, were analyzed. Demographic and surgical characteristics were compared between both cohorts. Generalized linear mixed models were used to examine the association between screw placement technique (AR vs. FH) and breach outcomes, accounting for clustering of screws within patients. Results: A total of 202 patients were analyzed, comprising 1,109 pedicle screws placed (AR: n = 548; FH: n = 561). The proportion of clinically acceptable screws (GRS Grade A or B) was significantly higher in the AR group compared to the FH group (99.3% vs. 96.4%; p = 0.013). The rate of clinically unacceptable breaches (GRS Grade C-E) was significantly lower in the AR group than in the FH group (0.7% vs. 3.6%; p = 0.013). Level-specific directional summary showed that the highest number of breaches in the AR group occurred at L4, predominantly lateral (n = 24), while in the FH group, breaches were most common at L5 and primarily medial (n = 22). No screw-related adverse events or revision surgeries were observed in either group. Demographic and surgical characteristics demonstrated no significant differences between the cohorts. Conclusion: AR-guided navigation was associated with significantly higher pedicle screw placement accuracy compared to conventional FH technique. As the largest study to date directly comparing these approaches, this analysis reinforces the notion that AR may offer more consistent accuracy in screw placement. Further prospective studies are needed to determine its long-term clinical implications.

ID: AGS26/549

A333: Clinical observation of traditional UBE vs. robot-assisted navigated UBE in the treatment of severe lumbar spinal stenosis

Yu Zhang 1

1 The Fifth Clinical Medical College of Guangzhou University of Chinese Medicine, Guangzhou city, China

Introduction: Unilateral Biportal Endoscopy (UBE) surgery, when applied in the treatment of severe lumbar spinal stenosis, relies on the surgeon's experience for operation. It is associated with issues such as insufficient decompression accuracy, limited decompression range, and high radiation exposure. This study compared the efficacy of traditional UBE and robot-assisted navigated UBE to provide evidence for clinical practice. Material and Methods: A total of 72 patients diagnosed with severe lumbar spinal stenosis in the Department of Spinal Orthopedics of our hospital from March 2022 to April 2024 were enrolled and divided into two groups: the UBE group (33 cases) and the robot-assisted navigated UBE group (39 cases). The traditional group underwent conventional UBE decompression; the navigated group received preoperative CT-based planning and intraoperative robot-assisted positioning for decompression. Observation indicators included: perioperative indicators (operation time, intraoperative blood loss, intraoperative effective radiation dose); clinical function scores Visual Analogue Scale (VAS), Oswestry Disability Index (ODI), Japanese Orthopaedic Association (JOA). lumbar score before surgery and at 3 and 12 months after surgery; imaging indicators (sagittal diameter of the spinal canal and nerve root canal volume at the lesioned segment at 6 months after surgery). Results: In the navigated group: The operation time for two-level segments was (109.2 ± 15.2) minutes, which was shorter than that of the traditional group (124.5 ± 17.8) minutes; The intraoperative blood loss was (43.9 ± 11.5) mL, lower than that of the traditional group (67.5 ± 16.3) mL;The effective radiation dose was (3.6 ± 1.2) mSv, lower than that of the traditional group (7.9 ± 2.2) mSv (all p < 0.05). At 6 months after surgery: The sagittal diameter of the spinal canal in the navigated group was (13.6 ± 1.3) mm, higher than that of the traditional group (12.3 ± 1.4) mm (p < 0.05); The volume of the nerve root canal in the navigated group was (182.4 ± 19.5) mm 3 , higher than that of the traditional group (148.5 ± 17.8) mm 3 (p < 0.05). At 12 months after surgery: The Visual Analogue Scale (VAS) score of the navigated group was (1.1 ± 0.3) points, lower than that of the traditional group (1.8 ± 0.6) points; The Oswestry Disability Index (ODI) score of the navigated group was (16.8 ± 4.3) points, lower than that of the traditional group (26.5 ± 4.8) points; The Japanese Orthopaedic Association (JOA) lumbar score of the navigated group was (24.6 ± 2.3) points, higher than that of the traditional group (18.7 ± 2.4) points (all p < 0.05). Conclusion: Robot-assisted navigated UBE is superior to traditional UBE in the treatment of severe lumbar spinal stenosis, as it shortens the operation time, reduces blood loss and radiation exposure, and enhances therapeutic efficacy and decompression effect. With high safety, it can be used as a preferred treatment option.

OP38: Lumbar Stenosis and Spondylolisthesis

ID: AGS26/134

A334: Retrospective review on the trend of lumbar spinal decompression surgery in Singapore

Kai Lin Lee 1 , Yilun Huang 2 , John Thng 1 , Dhivakaran Gengatharan 2 , Athanasios Sivridis 3

1 national university of Singapore, Singapore, Singapore

2 Sengkang General Hospital, Singapore, Singapore

3 Sofia University St. Kliment Ohridski, Sofia, Bulgaria

Introduction: Spinal stenosis and degenerative spinal disorders are becoming increasingly common and have a substantial impact on patients’ quality of life. Surgical decompression, performed through either open microscopic or endoscopic approaches, remains a cornerstone in the management of these conditions. While microdecompression has consistently demonstrated favorable outcomes, endoscopic techniques are gaining traction due to their minimally invasive nature. This study aims to investigate the evolving trends in single-level lumbar spinal decompression procedures at a tertiary academic hospital in Singapore. Material and Methods: A retrospective observational study involving 588 patients (unilateral biportal endoscopic: n=364; microdecompression: n = 224) undergoing single-level spinal decompression surgery between 2021 and 2024 was conducted. Primary outcome measures included changes in SF-36 and Visual Analogue Scale (VAS) scores at 3 months, 6 months, and 2 years postoperatively. Secondary outcomes were length of hospital stay, reoperation rates, and operative time. Collected data encompassed patient demographics, level and anatomical distribution of pathology, surgical techniques employed, and postoperative outcomes. Difference-in-differences (DID) analysis compared outcomes between both groups. Results: Both groups showed significant improvements in SF-36 and VAS scores postoperatively. At 1 month, endoscopic surgery resulted in greater early improvement in physical functioning (SFPF mean difference (MD): +11.2, p < 0.001 vs +7.0, p < 0.005) and VAS leg pain (MD: -3.64, p < 0.001 vs -3.50, p < 0.001). At 6 months, VAS back pain remained improved (MD: -3.30 vs -3.17, DID p = 0.82). At 2 years, SFPF improved (MD: +18.6, SD = 21.7, p = 0.064) in the endoscopic group and (MD: +36.7, SD = 20.9, p = 0.007) in the open group, with a non-significant DID of -18.1 (p = 0.155). Overall, no DID comparisons across domains reached statistical significance. Operative time for endoscopic cases decreased over the years (2022: mean 249 min vs 2024: mean 145 min), reflecting a learning curve. Surgeons with higher endoscopic caseloads demonstrated greater improvements in functional outcomes. Conclusion: Both endoscopic and open microscopic decompression achieve comparable short- and long-term outcomes. Endoscopy surgery offers comparable outcomes with faster recovery and shorter hospital stays. However, the presence of a subset of patients who do not achieve full symptom relief or require revision surgery suggests the need for further investigation into factors influencing surgical success. Future research with long-term follow-up is needed to further assess the durability and comparative efficacy of different surgical methods.

ID: AGS26/1355

A335: Reactive sagittal realignment drives low back pain improvement following fusionless decompression for degenerative lumbar spinal stenosis: a prospective analysis

Mahmoud A.F Amin 1 , Ahmed Shawky Abdelgawaad 1,2 , Ahmed M. Sayed 1 , Mahmoud saleh 1 , Mohamed El-Meshwawy 1

1 Department of Orthopaedic and Trauma Surgery, Assiut University Hospitals, Assiut, Egypt

2 Department of Spinal Deformities, Helios Hospitals Erfurt, Erfurt, Germany

Introduction: Lumbar spinal stenosis (LSS) is one of the most common causes of pain and disability in the aging population, frequently presenting with neurogenic claudication, radicular symptoms, and low back pain (LBP), and usually associated with sagittal spinal imbalance. Surgical decompression reliably alleviates neurogenic claudication and radicular pain. Traditionally, fusion surgery is considered necessary for associated sagittal imbalance correction and LBP relief. Emerging evidence suggests that fusionless decompression may induce reactive sagittal realignment , allowing restoration of more physiological global and spinopelvic harmony once neural compression, compensatory stooping and longstanding muscle strain resolve. However, prospective studies evaluating alignment changes following lumbar spinal canal decompression without fusion and their clinical correlates are scarce. This study aimed to examine the short term impact of fusionless decompression surgery on sagittal spinal alignment, explore its association with LBP improvement, and identify predictors of residual postoperative pain in patients with degenerative LSS. Material and Methods: In this prospective cohort, 55 consecutive patients with symptomatic degenerative LSS refractory to ≥ 3 months of conservative therapy underwent decompression without fusion. Patients with high-grade spondylolisthesis, instability, or deformity were excluded. Clinical outcomes included Visual Analog Scale (VAS) for LBP and leg pain, Oswestry Disability Index (ODI), and EuroQol-5D-5L. Radiographic evaluation comprised pelvic incidence (PI), pelvic tilt (PT), sacral slope (SS), lumbar lordosis (LL), thoracic kyphosis (TK), sagittal vertical axis (SVA). Patient’s sagittal profiles were evaluated and classified preoperatively and at the final follow-up using Roussouly’s system. Patients were stratified into satisfactory pain relief (Group S) and residual pain (Group R) based on achieving the minimal clinically important difference (MCID) for VAS-LBP and achieving end VAS-LBP score of ≤ 3. Follow-up averaged 6 months. Results: The whole cohort’s PROMs significantly improved following decompression surgery [LBP VAS (3.01 vs 6.06), LLP VAS (2.36 vs 6.4), ODI (19.26 vs 48.05), and EQ 5D 5L (0.788 vs 0.212 all p < 0.001]. Radiographically, PI-LL mismatch significantly decreased postoperatively (9.25° ± 11.4° vs 14.15° ± 15.7°, p = 0.007), and SVA significantly decreased (15.59 mm ± 42 vs 37.15 mm ± 61, p = 0.029), reflecting significant sagittal realignment. Roussouly type changed in 32.7% of patients, most commonly from hypolordotic type 2 to more physiological type 3, although this trend did not reach statistical significance (p = 0.141). No significant association was found between preoperative Roussouly type and pain relief (p = 0.255). On subgroup analysis,70.9% of patients achieved satisfactory LBP relief (Group S) vs 29.1% with residual pain (Group R). Group S demonstrated greater sagittal alignment improvement than Group R, as evidenced by reduction of SVA, PT, and PI-LL and increase of SS. Residual pain correlated with older age (57.1 vs 51.1 years) and multilevel surgery (2.07 vs 1.38 levels, p < 0.01). Conclusion: Fusionless lumbar spinal decompression surgery not only relieves symptoms of LSS but also promotes reactive sagittal realignment , marked by restoration of lumbar lordosis and reduction of sagittal imbalance. Patients achieving greater realignment experienced superior improvement in low back pain and quality of life. Younger age and fewer operated levels are associated with better outcomes.

ID: AGS26/3412

A336: Physiologic decompression of lumbar spinal stenosis through anatomic restoration using trans-kambin oblique lateral posterior lumbar interbody fusion (OLLIF): a retrospective analysis

Amir Sharif 1,2 , Hamid Reza Abbasi 1

1 Inspired Spine, Burnsville, United States

2 Sana Regio Klinikum Pinneberg, Pinneberg, Germany

Introduction: Introduction Lumbar spinal stenosis (LSS) is one of the most common indications for spinal surgery. Traditionally, decompression is achieved by removing bony and ligamentous structures through open surgery. However, recent studies have shown that symptomatic relief can be accomplished in many patients by increasing intervertebral and interpedicular height using fusion alone. In this study, we evaluate whether trans-Kambin oblique lateral lumbar interbody fusion (OLLIF) can effectively and safely relieve symptoms of LSS when an indication for fusion is present. Material and Methods: This is a retrospective single surgeon cohort study of 187 patients with LSS who underwent 189 OLLIF procedures between 2012 and August 2, 2019. Inclusion criteria for this study were age > 18 years with symptoms of LSS, including pain, sensory, and motor deficits, and an additional indication for fusion, which included spondylolisthesis, degenerative disk disease, disk herniation, or scoliosis. Exclusion criteria were the bony obstruction of the approach, osteogenic spinal canal stenosis, large facet hypertrophy, and listhesis grade II or greater. The primary outcome was a change in the Oswestry Disability Index (ODI) one year after surgery. Secondary outcomes were the resolution of radiculopathy at the first follow-up visit and one year after surgery, complication rates, surgery time, blood loss, and hospital stay. Results: Results ODI improved from 52% pre-op to 37% at the one-year follow-up. At the first follow-up, radiculopathy had resolved in 39% of patients, and 72% of patients experienced improvement of 50% or greater. One year after surgery, radiculopathy had resolved in 52% of patients and 74% experienced improvement of 50% or greater. Single-level surgeries required 56.4 ± 21.5 minutes, with a mean hospital stay of 1.6 - ± 2.4 days. Nerve irritation occurred in 12% of patients at the first postoperative follow-up and persisted in 6.8% of patients one year after surgery. There was one case each of persistent weakness at one year, infection, and cage subsidence. Conclusion: Trans-Kambin OLLIF delivers anatomic restoration of intradiscal and interpedicular distance, which results in physiologic decompression of lumbar spinal stenosis in patients undergoing lumbar fusion for degenerative or herniated disk disease, spondylolisthesis, or scoliosis. Amongst patients with LSS, OLLIF results in significant improvement of radiculopathy and patient-reported disability in the majority of patients with low rates of long-term complications. Unlike other minimally invasive surgery (MIS) fusions, OLLIF can be safely used from T12-S1.

ID: AGS26/2815

A337: Risk of asymmetric cage overhang in standalone LLIF: a patient-specific finite element analysis

Samuel Wakelin 1 , Armani Manov 1 , Ben Carnovale 1 , Andrew Legarreta 1 , Joseph Hudson 1 , Kojo Hamilton 1,2,3 , Nitin Agarwal 1,2,3

1 Medical Center, University of Pittsburgh, Pittsburgh, United States

2 University of Pittsburgh School of Medicine, Pittsburgh, United States

3 Veterans Affairs Pittsburgh Healthcare System, Pittsburgh, United States

Introduction: Standalone LLIF is a minimally invasive option to indirectly decompress the neural elements. Postoperative subsidence is a risk following standalone procedures, and cage characteristics and placement profiles are associated with subsidence risk. Though previous finite element (FEA) studies have examined subsidence following standalone LLIF, most prior FEA models are not patient-specific and do not consider the effect of asymmetric graft overhang. Objective: To develop a patient-specific FEA model incorporating Hounsfield Unit (HU) based bone density calculation to analyze the effect of asymmetric cage overhang on subsidence risk. Methods: A non-pathologic CT abdomen and pelvis was obtained from a publicly available database. A Python pipeline was developed using previously developed deep learning segmentation tools. This pipeline was used for segmentation, surface meshing, and HU to Elastic modulus conversion for patient-specific bone material generation. The vertebrae surface meshes were imported to FEBio studio version 2.10 (University of Utah, Salt Lake City, UT), for volumetric meshing and FEA. LLIF cages were created in fusion based on publicly available specifications for PEEK and titanium cages. A 500 N compressive force was applied to L2, with lateral bending and flexion/extension ROM elicited by 7.5 N·m moment. Von-Mises stress was extracted from the L3 inferior and L4 superior endplates, using the 99.9 th percentile value to limit the effect of single-element outliers. Results: For PEEK cages, the symmetrically and asymmetrically overhung cages caused the greatest von-Mises stress in all degrees of freedom at the L3 endplate. The symmetrically placed 55 mm cage caused the highest von-Mises stress in pure compression (8.49 MPa), left lateral bending (8.10 MPa), flexion (7.59 MPa), and extension (7.87 MPa). The asymmetrically placed 55 mm cage caused the highest stress in right lateral bending (13.70 MPa). At the L4 endplate, the asymmetrically placed 55 mm cage produced the highest von-Mises stress in all ROM with 7.22 MPa in pure compression, 16.68 MPa in right lateral bending, 11.33 MPa in left lateral bending, 5.79 MPa in flexion, and 10.68 MPa in extension. For titanium cages, at the L3 endplate, the 40 mm cage produced the highest von-Mises stress in pure compression (4.23 MPa), right lateral bending (7.87 MPa), left lateral bending (4.20 MPa), and flexion (4.07 MPa), while the asymmetrically placed 55 mm cage produced the highest stress in extension (5.01 MPa). Stresses were relatively stable with lower von-Mises stress values across all ROM at the L3 inferior endplate. At the L4 superior endplate, the 55 mm cages caused the highest von-Mises stress in all ROM. The symmetric cage caused the highest stress in pure compression (3.99 MPa) and flexion (3.83 MPa), while the asymmetric cage caused the highest stress in right lateral bending (13.88 MPa), left lateral bending (5.26 MPa), and extension (6.02 MPa). Conclusions: In this patient-specific LLIF model, symmetric and asymmetric cage overhang consistently increased endplate stress in PEEK cages. The results were more varied in the titanium model, suggesting a possible impact of overall cage geometry on subsidence risk.

ID: AGS26/2828

A338: Machine learning-driven clinical and imaging clustering of degenerative lumbar spondylolisthesis: implications for stratified surgical care

Karlo Pedro 1 , Aazad Abbas 2 , Chris Bailey 3 , Jennifer Urquhart 3 , Greg McIntosh 4 , Andrew Glennie 5 , Charles Fisher 6 , Philippe Phan 7 , Raja Rampersaud 2,8 , CSORN Investigators

1 Division of Neurosurgery and Spine Surgery, Toronto Western Hospital, Toronto, Canada

2 Division of Orthopaedic Surgery, University of Toronto, Toronto, Canada

3 Combined Orthopaedic and Neurosurgical Spine Program, London Health Sciences Centre, London, Canada

4 Canadian Spine Outcomes and Research Network, Toronto, Canada

5 Department of Surgery, Dalhousie University, Halifax, Nova Scotia, Canada, Halifax, Canada

6 Combined Neurosurgical and Orthopedic Spine Program, Department of Orthopaedic Surgery, University of British Columbia, Vancouver, British Columbia, Canada, Vancouver, Canada

7 Division of Orthopaedic Surgery, The Ottawa Hospital, Ottawa, Canada

8 Schroeder Arthritis Institute, Division of Orthopaedic Surgery, Toronto Western Hospital, Toronto, Canada

Introduction: Recent randomized trails for the surgical management of degenerative lumbar spondylolisthesis (DLS) report non-inferiority of decompression alone. Yet decompression and fusion remains the most common procedure for DLS patients. Previous work from the Canadian Spine Outcomes and Research Network (CSORN) including a modified Delphi process, suggest that the significant clinical and radiographic heterogeneity demonstrated by DLS patient creates substantial variability in surgeon reported decision-making. In this study, we applied unsupervised clustering techniques to identify distinct DLS groups, aiming to inform clinical decision-making in selecting the right patients for decompression or decompression and fusion. Material and Methods: We performed a sub-analysis of an ongoing prospective observational CSORN DLS study from 2015 to 2022 at eight Canadian sites. The CSORN-DLS study aims to ultimately develop a consensus-based surgeon decision aid to guide surgical management of DLS in a stratified manner to optimize outcome and resource utilization (i.e. value). Consensus cluster analysis was used based on demographic, clinical, and radiological characteristics of patients enrolled in phase one (baseline current surgeon practice). We selected clinically relevant variables to guide feature selection and evaluated the optimal number of clusters. Euclidean distance metric was used to determine the cluster number and structure, with up to seven clusters considered and 50 sampling iterations. Key characteristics of each cluster were identified, and post-surgical outcomes were compared using the Oswestry Disability Index (ODI) with pairwise statistical tests. Results: Consensus cluster analysis of 486 DLS patient (mean standard deviation [SD] age, 55.4 [9.2] years, 180 men (37%)) identified four distinct clusters. Cluster 1 (n = 175) included patients with moderate disability, low comorbidity, low depression symptoms, and radiographically balanced spine. Cluster 2 (n = 85) comprised predominantly female (94%) patients with severe disability, high comorbidity, high depression symptoms, greatest slip % and sagittally imbalanced spine (mean sagittal vertical axis (SVA) = 76mm, pelvic incidence - lumbar lordosis (PI-LL) mismatch = 23 degrees, Pelvic Tilt (PT) = 30 degrees). Cluster 3 (n = 91) mainly consisted of male (97%) patients with moderate disability, low comorbidity, low depression symptoms, and imbalanced (mean SVA = 57mm, PI-LL mismatch = 18 degrees, PT = 24 degrees) spine; while Cluster 4 (n = 135) included elderly patients with severe disability, high comorbidity, high depression, and balanced spine. Clusters 1 and 3 showed the most favorable final ODI outcomes at 1- and 2-years follow-up (Figure 1). Cluster 2 had the highest rate of fusion surgery (84%) and the greatest ODI change at 1 year (ΔODI mean [SD] 27.97 (19.32)). Decompression alone provided sustained benefit in Cluster 1 and 3, while improvement significantly regressed in Cluster 2 and to a lesser degree cluster 4 at two years. Conclusion: Unsupervised machine learning identified four clinically distinct clusters of DLS patients with varying outcome and responses to surgical interventions, highlighting the importance of stratified care strategies to optimize patient outcomes. Furthermore, our results confirm the clinical and radiographic heterogeneity of the surgical DLS population and inform the need for stratified recruitment strategies in future trial design for the evaluation of surgical management of DLS.

ID: AGS26/339

A339: Stand-alone OLIF: a viable treatment option for adjacent segment disease

Shubham Kadam 1 , Ayush Sharma 1

1 Dr Babasaheb Ambedkar Memorial Hospital, Mumbai, India

Introduction: Adjacent segment disease (ASD) remains a challenging complication following spinal fusion surgery. Stand-alone Oblique Lateral Interbody Fusion (OLIF) has emerged as a promising minimally invasive solution. Material and Methods: A total of 25 consecutive patients who underwent stand-alone OLIF for the treatment of adjacent-segment disease were reviewed. Visual analog scale (VAS) of back pain and leg pain and the Oswestry Disability Index (ODI) before surgery and at last postoperative clinic visits were obtained. Clinical outcomes, radiographic parameters, and complications were assessed. Results: During the study period, 25 cases were successfully treated with stand-alone OLIF. The mean follow-up was 12.7 ± 8.3 months. The back pain VAS improved from 7.2 ± 1.4 to 2.0 ± 1.2 (p < 0.01), and the leg pain VAS improved from 8.0 ± 1.8 to 1.4 ± 0.9 (p < 0.01). ODI improved from 29.0 ± 7.5 to 9.8 ± 3.0 (p < 0.01). The disc height (DH) increased from 8 ± 2 to 13 ± 2 mm (p < 0.01), the cross-sectional area (CSA) of spinal canal increased from 82 ± 24 to 138 ± 24 mm 2 (p < 0.01), the foraminal height increased from 16 ± 2 to 22 ± 2 mm (p < 0.01) and the CSA of foramen increased from 93 ± 25 to 158 ± 33 mm 2 (p < 0.01). Cage subsidence was observed in 1 case. Conclusion: Stand-alone OLIF demonstrates promising results in treating ASD, offering: - Reduced operative time and blood loss, - Minimally invasive approach, - Improved clinical outcomes, - Low complication rates.

ID: AGS26/3459

A340: To determine the optimal extent of fusion in anterior lumbar interbody fusion (ALIF) for degenerative disc disease. A retrospective analysis based on the university of Michigan Hospital experience

Jawad Saad 1,2 , Peyton Goethe 2 , Noojan Kazemi 2

1 Central Michigan University, Saginaw, United States

2 University of Michigan, Ann Arbor, United States

Introduction: Anterior lumbar interbody fusion (ALIF) provides durable pain relief and functional improvement for degenerative disc disease but carries increasing approach related and perioperative risks with additional fused levels. Prior studies suggest higher estimated blood loss (EBL), operative time, complication rates, and reoperation risk with multi-level ALIF. This study comprehensively evaluates preoperative, intraoperative, and postoperative risks of one-, two-, and three-level ALIFs, using estimated blood loss and operative time to quantify the incremental surgical burden of each additional level fused. Methods: A single-institute retrospective analysis of 141 patients undergoing ALIF (112 single-level, 22 two-level, 7 three-level) was performed. Intraoperative metrics (EBL and operative time), postoperative outcomes (length of stay, reoperations, readmissions, and complications), and patient demographics were analyzed. Statistical analyses used Kruskal–Wallis tests for continuous variables and Fisher–Freeman–Halton exact tests for categorical variables. Significant omnibus tests were followed by post-hoc Mann–Whitney U tests with Bonferroni correction or Fisher’s exact tests with Holm–Bonferroni adjustment. Statistical significance was defined as p < 0.05, with a Bonferroni-corrected threshold of p < 0.0167. Results: Baseline characteristics were comparable across all groups. Operative time increased significantly with fusion levels (single: 232.6 min; two-level: 311.6 min; three-level: 350.0 min; p< 0.0001), with significant differences between single and two-level procedures (p=0.0001). EBL differed across groups (p=0.0325) on omnibus analysis, though no individual pairwise comparisons were significant after correction. Mean EBL was 92.8 cc (single-level), 213.9 cc (two-level), and 146.4 cc (three-level). LOS showed a non-significant increasing trend (single: 2.98±1.85 days; two-level: 3.18±1.07 days; three-level: 4.57±2.82 days; p=0.110). Postoperative complications occurred in all groups, with the highest absolute number observed in the single-level cohort, including radicular pain and weakness. Although not significant, reoperations during any readmission occurred exclusively in single-level patients. Conclusion: Multi-level ALIF procedures demonstrated expected increases in intraoperative demands, with operative time differing significantly across fusion levels, increasing from 232.6 min in single-level to 311.6 min in two-level to 350.0 min for three-level procedures (p < 0.0001). EBL shows an increasing trend with each additional fused level, though it lacked statistical significance following correction for multiple comparisons. Contrary to existing literature, our study found complications predominantly in the single-level cohort, with no reoperations or readmissions in the two- or three-level groups. While multi-level procedures require greater surgical effort as measured by operative time, the absence of statistically significant postoperative complications in our multi-level cohorts suggests that higher intraoperative demands may not necessarily translate to worse clinical outcomes. However, our study's findings warrant further investigation to optimize surgical selection while mitigating the risks and benefits of a multilevel ALIF.

ID: AGS26/348

A341: Impact of preoperative sagittal malalignment on patient-reported outcomes following decompression for lumbar spinal stenosis

Shota Tamagawa 1 , Hidetoshi Nojiri 1 , Hiromitsu Takano 1 , Arihisa Shimura 1 , Juri Teramoto 1 , Yuta Sugawara 1 , Muneaki Ishijima 1

1 Juntendo University Faculty of Medicine, Tokyo, Japan

Introduction: Although the relevance of sagittal alignment in adult spinal deformity is well-established, its role in lumbar spinal stenosis (LSS) remains controversial. T1 pelvic angle (TPA) and composite indices, such as the Global Alignment and Proportion (GAP) score and Sagittal Age-Adjusted Score (SAAS), may better characterize global balance and compensatory mechanisms than sagittal vertical axis (SVA) alone. The purpose of this study was to investigate the impact of preoperative sagittal malalignment (MA) on patient-reported outcome measures (PROMs) following decompression with or without short-segment fusion for LSS. Material and Methods: This retrospective cohort study included LSS patients with predominant neurogenic claudication who underwent decompression ± interbody fusion at ≤ 3 levels between March 2021 and July 2024 and completed 1-year follow-up. Patients were divided into MA (TPA ≥ 20°) and non-MA (TPA < 20°) groups. Whole-body EOS radiographs were obtained preoperatively and 1-year postoperatively to measure spinopelvic parameters and calculate GAP and SAAS. PROMs included Visual Analog Scale (VAS) for low back pain (LBP), leg pain, and leg numbness; Japanese Orthopaedic Association Back Pain Evaluation Questionnaire (JOABPEQ); and Oswestry Disability Index (ODI). Group comparisons were performed using chi-square or Mann-Whitney U tests. Multivariable logistic regression was performed to identify independent predictors of achieving the minimal clinically important difference (MCID) in ODI, adjusting for age, sex, body mass index, American Society of Anesthesiologists classification, spondylolisthesis, fusion, number of surgical levels, baseline ODI, and TPA ≥ 20°. Results: Of the 176 patients included in this study (mean age 70.1 ± 10.7 years, 57.4% male), 97 patients (55.1%) were classified as MA. The MA group demonstrated significantly worse baseline VAS LBP (6.5 ± 2.6 vs. 5.4 ± 2.9, p = 0.009), JOABPEQ walking ability (23.1 ± 23.1 vs. 32.7 ± 28.7, p = 0.032), and ODI (46.4 ± 14.9 vs. 40.8 ± 17.9, p = 0.012). Radiographically, the MA group had greater pelvic incidence-lumbar lordosis mismatch, pelvic tilt, SVA, and TPA (all p < 0.001). At 1 year, both groups improved significantly, and ODI MCID achievement rates were comparable (52.1% vs 53.8%, p = 0.817). Multivariable analysis identified younger age (OR 0.95; 95% CI 0.92-0.98; p = 0.002) and higher baseline ODI (OR 1.04; 95% CI 1.02-1.07; p < 0.001) as independent predictors of MCID achievement; TPA ≥ 20° was not significant (OR 1.18; 95% CI 0.60-2.31; p = 0.638). GAP and SAAS stratification showed non-significant trends toward poorer outcomes in severely disproportioned and over-corrected groups. Conclusion: Preoperative sagittal MA was associated with worse baseline LBP and function but did not adversely affect 1-year PROMs. Decompression ± short fusion yields satisfactory outcomes even in patients with sagittal imbalance, suggesting deformity-based thresholds may not directly apply to LSS with neurogenic symptoms.

ID: AGS26/597

A342: Enhanced recovery after spine fixation surgery: an international survey on current practice

Hein Jonkman 1,2 , Floris van Tol 2 , Tim Govers 1,3 , Silje Debets 1 , Tom Wainwright 4 , Bertrand Debono 5 , Michael Yang 6 , Hans De Boer 7 , Maroeska Rovers 1 , Jorrit-Jan Verlaan 2,8

1 Department of Medical Imaging, Radboud University Medical Center, Nijmegen, The Netherlands

2 SentryX, Utrecht, The Netherlands

3 Medip Analytics, Nijmegen, The Netherlands

4 Orthopaedic Research Institute, Bournemouth University, Bournemouth, United Kingdom

5 Centre Francilien Du Dos, Clinique des Franciscaines, Versailles, France ,

6 Department of Clinical Neurosciences, Division of Neurosurgery, University of Calgary, Calgary, Canada

7 Department of Anesthesiology, Pain Medicine and Procedural Sedation and Analgesia, Martini General Hospital Groningen, The Netherlands, Groningen

8 Department of Radiotherapy, University Medical Center Utrecht, Utrecht, The Netherlands

Introduction: The demand for spine fixation surgery is rising globally, mostly due to an ageing population and corresponding increase in degenerative spinal conditions. Meanwhile, perioperative care delivery faces growing pressure in terms of cost and capacity. Enhanced Recovery After Surgery (ERAS) aims to standardize and improve perioperative care to enhance patient outcomes, reduce complications, and shorten hospital stays. Although well-established in other surgical fields, its application in spine fixation surgery is relatively new, and it remains unclear to what extent recommendations have been adopted in clinical practice. The ERAS-DSF (Degenerative Spine Fixation) study is the first international survey to assess spine surgeons’ perspectives on the importance, implementation, and barriers related to ERAS in spine surgery. Material and Methods: A cross-sectional, web-based survey was distributed in May 2024 to members of AOSpine, the global academic network of spine professionals. The survey focused on instrumented procedures for degenerative spine disease and was structured around eight ERAS components adapted from the published 2021 consensus statement for perioperative care in lumbar spinal fusion. The current work reports one section of the survey, in which respondents were asked to identify which ERAS components had the biggest impact on patient recovery, which ones could be improved the most, what barriers to implementation exist in their hospital, and what support is needed to facilitate adoption. Responses were analyzed using descriptive statistics. Results: The survey was accessed by 400 individuals, filled in by 322 (80.5% response rate), and completed by 247 (76.7% completion rate). Respondents were spine surgeons from a range of clinical settings, including academic (50.6%), private (28.9%), and local/community hospitals (20.5%) from all over the world. The ERAS components most often perceived as having the biggest impact on patient recovery were multimodal opioid-sparing analgesia (73.3%), early mobilization (63.8%), and preoperative education & counselling (58.6%). These same three components were also most frequently selected as leaving the greatest room for improvement, albeit in a different order: preoperative education & counselling (62.6%), early mobilization (41.4%) and multimodal opioid-sparing analgesia (39.4%). Key barriers to ERAS implementation were the absence of clear protocols and guidelines (56.2%), staff shortages (53.8%), and difficulties in coordinating implementation and adherence (43.0%). In terms of support, respondents expressed the need for clear (inter)national recommendations (75.3%), tools for compliance monitoring (60.3%), online training and education (56.3%), and more evidence on the cost-effectiveness (44.5%). Conclusion: Spine surgeons worldwide recognized the ERAS components of multimodal opioid-sparing analgesia, early mobilization, and preoperative education & counselling as having the biggest impact on patient recovery. Meanwhile, these same three areas could also be improved the most, suggesting a gap between scientific evidence and clinical practice. To close this gap, clinicians require more practical guidance, training resources, and institutional support. Future research should focus on developing clear and actionable treatment protocols, defining reliable indicators of patient recovery, providing more robust evidence on the cost-effectiveness of ERAS in spine fixation surgery, and establishing best practices for implementation and multidisciplinary collaboration.

OP39: Spinal Pelvic Considerations

ID: AGS26/653

A343: The role of pelvic compensation in sagittal balance and imbalance: the impact of pelvic compensation on spinal alignment and clinical outcomes following adult spinal deformity surgery

Xiaolong Chen 1 , Zheng Wang 1 , Shibao Lu 1

1 Xuanwu Hospital Capital Medical University, Beijing, China

Introduction: The Scoliosis Research Society (SRS)-Schwab system does not include a pelvic compensation (PC) subtype, potentially contributing to gaps in clinical characteristics and treatment strategy for deformity correction. It also remains uncertain as to whether PC has differing roles in sagittal balance (SB) or imbalance (SI) status. To compare radiological parameters and SRS-22r domains between patients with failed pelvic compensation (FPC) and successful pelvic compensation (SPC) based on preoperative SB and SI. Methods: A total of 145 adult spinal deformity patients who received deformity correction were analyzed. Radiographic and clinical outcomes were collected for statistical analysis. Patients were classified into 4 groups based on the median value of PT/PI ratio (PTr) and the cutoff value of SB. Patients with low PTr and high PTr were defined as FPC and SPC, respectively. Radiographic and clinical characteristics of different groups were compared. Results: Patients with SPC exhibited significantly greater improvements in lumbar lordosis, pelvic tilt, PTr, and T1 pelvic angle as compared to patients with FPC, irrespective of SB or SI. No apparent differences in any of SRS-22r domains were observed at follow-up when comparing the SB-FPC and SB-SPC patients. However, patients with SI-SPC exhibited significantly better function, self-image, satisfaction, and subtotal domains at follow-up relative to those with SI-FPC. When SI-FPC and SI-SPC patients were subdivided further based on the degree of PI-LL by adjusting for age, the postoperative function and self-image domains were significantly better in the group with overcorrection of PI-LL than under correction of PI-LL in SI-FPC patients. However, no differences in these SRS-22r scores were observed when comparing the subgroups in SI-SPC patients. Conclusion: Flexible pelvic rotation is associated with benefits to the correction of sagittal parameters, irrespective of preoperative SB or SI status. However, PC is only significantly associated with clinical outcomes under SI. Patients with SI-FPC exhibit poorer postoperative clinical outcomes, which should be recommended to minimize PI-LL.

ID: AGS26/1741

A344: Long-term impact of total hip arthroplasty on subsequent lumbar spine surgery in patients with hip-spine disease: a nationwide analysis of risk factors and outcomes

Ali Issani 1 , William Karakash 1 , Henry Avetisian 1 , Aidan Lindgren 1 , Daniel Rusu 1 , Mirbahador Athari 1 , Dil Patel 1 , Raymond Hah 1 , Kiran Alluri 1 , John Liu 2 , Jeffrey C. Wang 1

1 Department of Orthopaedic Surgery, Keck School of Medicine of the University of Southern California, Los Angeles, United States

2 Department of Neurological Surgery, Keck School of Medicine of the University of Southern California, Los Angeles, United States

Introduction: Degenerative hip and spine pathologies often coexist, with total hip arthroplasty (THA) and lumbar fusion addressing localized disease but showing complex biomechanical interplay. Although THA was once thought to relieve low back pain, recent studies suggest limited benefit, raising questions about its effect on lumbar procedures in patients with HOA and LDD. Prior evidence is conflicting and often lacks adjustment for comorbidities or unilateral versus bilateral THA. This study aims to evaluate whether THA alters the risk of subsequent lumbar decompression or fusion and to identify patient-specific risk factors. Materials and Methods: This retrospective study used the PearlDiver Mariner Database to identify adults with LDD and HOA who underwent THA between 2010 and 2023, with exclusions for prior lumbar surgery, trauma, malignancy, or infection; 1,000,000 matched patients without THA served as controls. The primary outcome was subsequent lumbar decompression or fusion, with secondary analyses assessing comorbidities, lumbar pathologies, and unilateral versus bilateral THA within the THA cohort. Multivariable logistic regression adjusted for age, gender, comorbidities, and scoliosis, with significance set at p < 0.05. Results: Among 1,194,269 patients with LDD and HOA, 194,269 (16.3%) underwent THA. Compared to controls, the THA cohort was older, had higher comorbidity burden, and showed significantly lower rates of subsequent lumbar decompression (1.4% vs. 4.2%) and fusion (0.7% vs. 2.4%), with multivariable regression confirming reduced odds for both (OR 0.42, p < 0.001). In the THA cohort, decompression risk was increased with male sex (OR 1.7, p < 0.001), obesity (OR 1.3, p < 0.001), opioid use (OR 1.4, p < 0.001), fibromyalgia (OR 1.3, p < 0.001), diabetes (OR 1.2, p < 0.001), depression (OR 1.2, p < 0.001), scoliosis (OR 2.97, p < 0.001), spondylolisthesis (OR 1.96, p < 0.001), myelopathy (OR 2.65, p < 0.001), radiculopathy (OR 1.98, p < 0.001), and spinal stenosis (OR 2.1, p < 0.001); Within the THA cohort, bilateral THA was protective (decompression OR 0.9, 95% CI 0.77 -0.99, p = 0.05; fusion OR 0.8, 95% CI 0.7 -0.96, p = 0.02), compared to unilateral THA (decompression OR 1.1, 95% CI 1.0 -1.29, p = 0.05; fusion OR 1.2, 95% CI 1.03 -1.5, p = 0.02). Predictors for fusion were similar, with strongest associations for scoliosis (OR 4.5, p < 0.001) and spondylolisthesis (OR 2.95, p < 0.001), while osteoporosis (OR 0.75, p < 0.001) and bilateral THA (OR 0.8, p = 0.02) were associated with reduced subsequent surgery. Patients aged 60 -69 had the highest odds of decompression (OR 1.4, p < 0.01) and fusion (OR 1.3, p < 0.001), whereas age > 69 was protective for both (OR 0.8 and 0.6, p < 0.001). Time to subsequent surgery was longer in THA patients (6.5 years) compared to controls (∼3 years). Conclusion: THA in patients with LDD and HOA was associated with reduced risk of lumbar decompression (OR 0.42, p < 0.001) and fusion (OR 0.42, p < 0.001), with bilateral THA protective and unilateral THA increasing risk. Scoliosis, spondylolisthesis, myelopathy, radiculopathy, and spinal stenosis were the strongest predictors of surgery, while comorbidities including opioid use, fibromyalgia, depression, diabetes, and obesity further elevated risk. THA also delayed time to lumbar surgery, suggesting timely hip intervention may influence spinopelvic biomechanics. These findings highlight the need for multidisciplinary management of hip -spine syndrome with careful consideration of patient-specific risk factors.

ID: AGS26/2351

A345: Impact of prior joint arthroplasty on outcomes of lumbar spinal fusion: 30% of patients with prior total hip arthroplasty require revision lumbar fusion within 2-years of surgery

Mohammad Daher 1 , Jonathan Liu 1 , Amer Sebaaly 2 , Mouhanad El-Othmani 1 , Bassel Diebo 1 , Alan Daniels 1

1 Brown University, Providence, United States ,

2 Hotel Dieu de France, Beirut, Lebanon

Introduction: The prevalence of patients requiring lumbar spinal fusion (LSF) with prior total joint arthroplasty (TJA) (total knee and hip arthroplasty (TKA, THA)) is increasing. However, the impact of prior both THA and TKA, or either TKA/THA on the outcomes of LSF is still incompletely understood. This study aims to fill this gap by comparing the surgical and medical complications after elective lumbar fusion outcomes between patients with prior TKA, THA, both, or none. Material and Methods: This is a retrospective review of the PearlDiver Mariner Database. Patients undergoing elective lumbar spinal fusion (LSF) for spinal stenosis between 2010 and 2020 were identified and grouped based on prior arthroplasty history into four cohorts: No-TJA, Spine-TKA, Spine-THA, and Spine-TJA (both TKA and THA). The four groups were propensity score matched in a 1:1:1 ratio based on age, gender, obesity, and Charlson Comorbidity Index (CCI). Results: Patients in the No-TJA group had the lowest readmission (7.6%; p < .001) and ED visits rates (18.9%; p < .001) at 90 days post-operatively. Patients in the spine-TJA group had the highest ED visits rate (31.1%, p < .001). Furthermore, the group with no previous TKA/THA had the lowest rate of revisions at 1 year (5.6%, p < .001) and 2 years post-operatively (10.6%, p < .001) while the spine THA group had the highest rate at both 1 (15.6%, p < .001) and 2 years (32.5%, p < .001). Conclusion: This study highlights the considerable challenges and complications faced by patients with prior hip and knee replacement undergoing LSF. Patients with prior TKA and THA had higher rates of ED visits at 90 days postoperatively and higher revision spinal fusion rates. Patients undergoing LSF with prior THA suffered a 30% revision spinal fusion rate within 2 years.

ID: AGS26/2843

A346: A comparison of gait kinematic and muscular activity between walking cane and a walker in the elderly with sagittal imbalance

Thamrong Lertudomphonwanit 1 , Panyajarn Laohapornsvan 2 , Pongsiri Dooduang 1

1 Ramathibodi Hospital, Faculty of Medicine, Mahidol University, Bangkok, Thailand

2 Ramathibodi Hospital, Faculty of Medicine, Mahidol University, Nopparat Rajathanee Hospital, Bangkok, Thailand

Introduction: Adult spinal deformity (ASD) patients demonstrate an altered gait pattern. A walking stick and walker are often used to improve patient mobility and maintain balance. There is a lack of information about the benefits of walking sticks versus a walker on the trunk and lower extremity muscular control in ASD patients who have sagittal imbalance. The purpose of this study is to compare back muscle activity of ASD patients with sagittal imbalance who use walker and walking stick while walking. We hypothesized that walking stick may help ASD patients with sagittal imbalance to promote their back and leg muscle activity better than walker during walking. Material and Methods: A cross-over randomized controlled trial was conducted among adult patients diagnosed with spinal deformity (ASD) and sagittal imbalance. The study included patients who had a sagittal vertical axis (SVA) greater than 5 cm, a pelvic tilt of more than 25 degrees, or thoracic kyphosis of over 60 degrees. 20 patients were recruited and were assigned to three different walking conditions: 1) no device (ND), 2) walking sticks (WS), and 3) a walker (WR). The peak muscle activation, timing of peak muscle activity, muscle duration, muscle onset, and integrated electromyography (iEMG) of the trunk and lower extremities were recorded and compared during ground walking under each condition. Results: The use of WS resulted in greater activation leg muscles of right external oblique muscle in comparison to WR (WS 48.26 vs WR 30.36 % Root Mean Square (RMS), p = 0.044), and semitendinosus muscle (WS 30.88 vs WR 28.43 %RMS, p = 0.03) There was also a similar trend on left side of left external oblique muscle although this did not statistical significance. (WS 36.20 vs 33.27 %RMS, p = 0.057). When using WR, there was a significant difference in the iEMG of both legs and back muscles including Erector spinae at T12 and Gluteus maximus at WS 0.35 vs WR 0.31 and WR 0.14 vs WR 0.12, p = 0.009, 0.033 respectively). Conclusion: The findings demonstrated that the use of walking sticks may be more effective in promoting lower extremity neuromuscular control compared to that of walker. In patients with adequate lower limb strength, this advantage can potentially translate into greater functional benefit. Therefore, the use of walking sticks may be more beneficial to that of a walker in patients with ASD requiring walking aids with preserved lower extremity strength.

ID: AGS26/2938

A347: Sagittal alignment as a major prognostic determinant in degenerative spine surgery: from 3D biomechanical planning to spinal instrumentation optimization through robotic assistance and intraoperative navigation

Grouz Khalil 1 , Amine Selmene 1 , Elyes Zoghlami 2

1 Centre de Traumatologie et Des Grands Brulées de Ben Arous, Tunis, Tunisia

2 Hopital Militaire, Tunis, Tunisia

Introduction: Degenerative spine surgery has evolved toward a comprehensive reconstructive approach emphasizing sagittal balance restoration. This study aims to demonstrate that sagittal alignment represents the primary prognostic determinant of therapeutic success and to quantify the contribution of computer-assisted surgical technologies in optimizing biomechanical and clinical outcomes. Materials and Methods: Prospective single-center study including 85 consecutive patients undergoing lumbosacral arthrodesis for symptomatic degenerative pathologies (degenerative spondylolisthesis, adult scoliosis, spinal stenosis with instability). Systematic three-dimensional preoperative planning incorporated spinopelvic morphometric analysis (pelvic incidence, sacral slope, pelvic tilt) to determine individualized target lumbar lordosis according to Roussouly's formula. Pedicle screw instrumentation was performed under robotic guidance (Mazor X™ system) coupled with O-arm™ navigation. Statistical analysis (Pearson coefficient, multivariate regression) assessed correlations between postoperative sagittal correction and evolution of validated functional scores (ODI, lumbar and radicular VAS, SF-36) with minimum 24-month follow-up. Results: A highly significant positive correlation (r = 0.847, p < 0.001) was observed between spinopelvic harmony restoration and functional improvement. Patients achieving optimal alignment (Δ(LL-PI) < 10°, SVA < 50 mm) demonstrated mean ODI improvement of 65% versus 28% for patients with residual imbalance (p < 0.001). Mean postoperative ODI scores were 16.2 ± 8.4 versus 42.7 ± 12.1, respectively (p < 0.001). Robotic assistance ensured implant positioning accuracy of 98.7% (angular deviation < 2°, translation < 2 mm) with 47% reduction in intraoperative radiation exposure. Surgical revision rates were significantly lower in the optimal alignment group (4.2% vs 18.6%, p = 0.008). Conclusion: Sagittal homeostasis restoration represents the fundamental therapeutic paradigm in degenerative spine surgery, directly conditioning long-term functional prognosis. The synergistic integration of personalized 3D planning technologies and robotic assistance transforms biomechanical conceptualization into reproducible surgical precision, establishing a new quality standard. This holistic approach, centered on therapeutic individualization and technical excellence, should constitute the contemporary benchmark for optimizing clinical outcomes in degenerative spinal pathologies.

ID: AGS26/1691

A348: Normal spino-pelvic alignment in healthy adolescents: defining the PI-SS relationship and laying the foundations for pre-operative sagittal planning

Matteo Traversari 1 , Alberto Ruffilli 2 , Giovanni Viroli 2 , Marco Manzetti 2 , Alessandro Cargeli 1 , Cesare Faldini 1

1 IRCCS - Istituto Ortopedico Rizzoli, Bologna, Italy

2 IRCCS - Istituto Ortopedico Rizzoli, Dipartimento di Scienze Biomediche e Neuromotorie - DIBINEM, Bologna, Italy

Introduction: Surgical treatment of adolescent spinal deformities requires preoperative planning. Few studies confirmed that sagittal alignment restoration is crucial to reduce the risk of early degeneration, yet sagittal alignment targets are lacking in adolescents. No predictive formula for ideal SS has been developed specifically for surgical planning in adolescents. The aim of the present study was to investigate the relationship between pelvic incidence (PI) and sacral slope (SS) in healthy adolescents aged 11-18 years to identify the most accurate model for SS prediction. Material and Methods: Full-spine standing biplanar X-rays were analyzed. PI, SS and pelvic tilt (PT) were measured. Three regression models (linear, polynomial, and logarithmic) were generated to describe the PI-SS relationship. The adult-derived model by Yilgor et al. was also applied to the study cohort. Accuracy and precision of all models were assessed globally and within PI-based subgroups. External validation was performed using a previously published dataset. Results: A total of 122 patients (mean age 14 years) were included. Logarithmic and polynomial models achieved the best fit within the study cohort (R 2 = 0.63 and 0.63). The Yilgor et al. model showed substantially lower fit (R 2 = 0.39). Residual and absolute error pairwise analyses confirmed significantly lower errors for the polynomial and logarithmic models compared to the Yilgor model (p < 0.01), which exhibited systematic underestimation of SS in intermediate and high PI subgroups. The linear model performed best within the external validation cohort with the logarithmic model performing similarly. The Yilgor model again showed systematic SS underestimation and higher errors. Conclusion: Our results suggest that the formula SS ideal = 0.61 x PI + 11.5 can be reliably employed in the pre-operative planning of adolescent spinal deformities. However, logarithmic model of PI-SS relationship showed high performances in study cohort, particularly in patients with low PI.

ID: AGS26/3335

A349: The association between spinopelvic parameters and thoracolumbar spinal forces

Fatemeh Alavi 1 , Christopher J. Nielsen 1 , Michael Kelly 2 , Jeffrey Hills 3 , Stephen Lewis 1

1 University of Toronto, University Health Network (UHN), Toronto, Canada

2 Rady Children’s Hospital, University of California San Diego, San Diego, United States

3 UT Health San Antonio, San Antonio, United States

Introduction: Sagittal malalignment is a well-established risk factor for mechanical complications following adult spinal deformity correction surgery. Excessive spinal force has been proposed as one of the key mechanisms linking malalignment to these complications. However, little is known about how spinopelvic parameters influence spinal forces in individuals without deformity. The present study aims to investigate the association between spinopelvic alignment and spinal forces at the thoracolumbar junction in a normal population. Methods: Full-body low dose stereoradiographic evaluation was performed in an international multi-ethnic cohort of 319 asymptomatic adult volunteers. Demographic data and spinopelvic parameters such as pelvic incidence (PI), T1-pelvic angle (TPA), L1-S1 lordosis (LL), L4-S1 lordosis (L4S1), L1 pelvic angle (L1PA), and L4 pelvic angle (L4PA) were measured. Individualized musculoskeletal (MSK) models based on sex, weight, height, and spine curvature were created and spinal forces (shear and compression) were calculated using MSK analysis. The associations between demographic data, spinopelvic alignment parameters, and spinal forces were evaluated using multivariate regression analyses. Results: 192 out of 319 individuals (60%) were female. The mean weight was 62 ± 13kg, height was 1.6 ± 0.8m. The mean PI was 52° ± 10°, pelvic tilt (PT) was 12° ± 6°, and LL was 58° ± 10°. The spinal forces were calculated using MSK analysis. The mean compressive forces at L2 and L1 were 346 ± 100N and 369 ± 106N, while mean shear forces were 55 ± 27N and 70 ± 36N. The shear forces at L1 and L2 levels were associated with PI, PT, L1PA, weight, L4-S1 horizontal distance, and the Cobb angle between these levels and S1 (r2=0.86). The compressive forces at these levels were associated with PI, PT, L1PA, weight, L4S1, and L1-sacrum horizontal distance (r2=0.78). Conclusion: Understanding the association between spinopelvic parameters and spinal forces in asymptomatic individuals provides essential baseline data that may help distinguish compensatory mechanisms from pathological loading patterns. These insights could ultimately contribute to improved pre-planning, risk stratification and preventative strategies for mechanical complications in spinal deformity surgery.

ID: AGS26/444

A350: Impact of lumbar disc arthroplasty on spinopelvic parameters and sagittal balance

Anthony Nguyen 1 , Daniele Morelli 1 , Henri-Benjamin Pouleau 1 , Nicolas Massager 1 , Alexander Jodaitis 1

1 University Hospital of Tivoli, La Louvière, Belgium

Introduction: Lumbar disc arthroplasty has emerged as an alternative to fusion in the treatment of refractory discogenic low back pain, aiming to preserve motion and reduce adjacent segment degeneration. However, its effect on sagittal balance parameters remains insufficiently documented. Material and Methods: We conducted a retrospective monocentric study including 81 patients (mean age 43.8 ± 6.6 years; 68% female) who underwent lumbar arthroplasty between 2017 and 2024. Patients with less than 6 months of follow-up or lacking postoperative full-spine radiographs were excluded. Radiographic parameters analyzed pre- and postoperatively included pelvic incidence (PI), pelvic tilt (PT), sacral slope (SS), lumbar lordosis (LL), segmental lordosis, and segmental range of motion (ROM). Sagittal balance was defined according to Schwab’s formula (LL = PI ± 9°). Results: Arthroplasty significantly increased LL (52.0° ± 11.1° to 58.2° ± 8.9°, p < 0.001), segmental lordosis (5.9° ± 4.0° to 11.7° ± 4.1°, p < 0.001), and ROM (5.2° ± 4.2° to 13.6° ± 4.9°, p < 0.001). PT and SS remained stable. The proportion of patients with sagittal balance rose from 37% to 85% (p < 0.001). At last follow-up (mean 38 months), 93.8% reported clinical satisfaction. Adverse events included adjacent disc degeneration in 6.1% and facet syndrome in 9.9%. Conclusion: Lumbar arthroplasty significantly restores lumbar lordosis and disc mobility, thereby improving sagittal alignment in most patients. These results support its role as a motion-preserving alternative to fusion, with promising radiological and clinical outcomes.

ID: AGS26/875

A351: Validation of pelvic inclination angle as a novel sagittal parameter in adult spinal deformity surgery: association with postoperative radiographic and clinical results

Se-Jun Park 1 , Jin-Sung Park 1 , Dong-ho Kang 1

1 Samsung Medical Center, Seoul, South Korea

Introduction: Pelvic tilt (PT) is a widely used parameter to assess pelvic compensation in adult spinal deformity (ASD), but its normal values are influenced by pelvic incidence (PI), limiting its universal applicability. The pelvic inclination angle (PIA) is a novel sagittal parameter reportedly independent of PI, with a normal value near 0°, but its clinical relevance in surgically treated patients has not been established. This study aims to validate the clinical utility of PIA in patients with sagittal spinal malalignment (SSM) undergoing corrective surgery, and to compare its correlation with clinical outcomes to established sagittal parameters. Material and Methods: Retrospective cross-sectional study was performed. A total of 211 consecutive patients (mean age, 69.6 years; 88.2% female) who underwent ≥ 6-level fusion including the sacrum or pelvis for symptomatic SSM between 2015 and 2022 and completed 2-year follow-up. Radiographic parameters including PIA, PI-LL (lumbar lordosis), PT, T1 pelvic angle (T1PA), and sagittal vertical axis (SVA) and clinical outcome including Oswestry Disability Index (ODI) and Scoliosis Research Society-22r (SRS-22r) questionnaire were evaluated. Correlations between PIA and other sagittal parameters, as well as with ODI and ODI change (ΔODI), were assessed. In addition to PROM scores, the proportion of patients achieving minim clinically important difference (MCID) was evaluated. PIA thresholds corresponding to mild and severe disability were calculated. Linear regression analysis was performed to identify factors affecting the postoperative PIA values. Results: Postoperative PIA showed stronger correlation with ODI (cc = 0.43) than PI-LL (cc = 0.22), PT (cc = 0.23), T1PA (cc = 0.30), or SVA (cc = 0.27). ΔPIA correlated more strongly with ΔODI (cc = 0.24) than changes in other parameters (cc ranging from 0.02 to 0.12). PIA thresholds for mild and severe disability were calculated as 8.0° and 12.0°, respectively. Low PIA (< 8.0°) group demonstrated the most favorable postoperative ODI, SRS-22r scores, and highest MCID achievement. Multivariate linear regression analysis revealed that postoperative PIA was significantly affected by postoperative LL and PT, but not affected by PI. Conclusion: PIA demonstrated significant correlations with established sagittal parameters and a stronger association with disability than conventional measures. Postoperative PIA was unaffected by PI, supporting its universal applicability. A threshold of 8.0° was linked to optimal outcomes, while values > 12.0° were associated with greater disability. PIA may serve as a reliable, PI-independent metric for assessing pelvic compensation and guiding surgical targets.

OP40: Endoscopy vs Open

ID: AGS26/1334

A352: A decade of lumbar tubular minimally invasive spine surgery: analyzing complication trends through a systematic review and meta-analysis

Chibuikem Ikwuegbuenyi 1 , Sean Inzerillo 1 , Eesha Gurav 1 , Noah Willett 1 , Mousa Hamad 1 , Galal Elsayed 1 , Osama Kashlan 1 , Roger Härtl 1

1 Department of Neurological Surgery, Och Spine at New York-Presbyterian/Weill Cornell Medical Center, New York, United States

Introduction: Tubular Minimally Invasive Spine Surgery (MISS) is an effective approach for treating lumbar spine disorders. This study aimed to comprehensively quantify the incidence of overall and specific complications associated with tubular lumbar MISS through a systematic review and proportional meta-analysis of studies published over the last decade. Material and Methods: A systematic search of PubMed, Medline, Embase, and the Cochrane Library (January 2013 to March 2024) was conducted following PRISMA guidelines. Studies were included if they involved ten adult patients undergoing tubular lumbar MISS and provided extractable complication data. A random-effects model was used to pool complication rates, and study quality was assessed using the Cochrane Risk of Bias Tool and Newcastle-Ottawa Scale. All analyses were done using R studio. Results: Seventy-five studies involving approximately 12,600 patients were included in the analysis. The complication rate was 10% (95% CI: 8%-14%, I 2 = 93%). Specific complication rates were: dural tears 4% (95% CI: 3%-5%, I 2 = 69%) in 56 studies (6,651 patients); nerve injuries 1% (95% CI: 1%-3%, I 2 = 70%) in 41 studies (5,278 patients); postoperative hematoma 1% (95% CI: 1%-2%, I 2 = 31%) in 19 studies (2,454 patients); surgical site infections 1% (95% CI: 0%-1%, I 2 = 27%) in 46 studies (10,439 patients); revision surgeries 2% (95% CI: 2%-3%, I 2 = 77%) in 43 studies (8,948 patients); and disc re-herniation 3% (95% CI: 1%-7%, I 2 = 84%) in 14 studies (1,928 patients). Conclusion: This meta-analysis provides a comprehensive overview of complication rates in tubular lumbar MISS, revealing generally low rates but significant heterogeneity across studies. These findings offer valuable insights for patient counseling and surgical planning, though individual patient factors and surgeon experience should be considered. Standardized complication reporting and research into outcome variability are needed to improve future studies.

ID: AGS26/1474

A353: Oblique lumbar interbody fusion with anterolateral screw-rod fixation versus posterior pedicle screws for adjacent segment degeneration: a two-year comparative study

Khanathip Jitpakdee 1 , Weerasak Singhatanadgige 2 , Vit Kotheeranurak 2 , Worawat Limthongkul 2 , Wicharn Yingsakmongkol 2

1 Department of Orthopaedics, Queen Savang Vadhana Memorial Hospital, Chonburi, Thailand

2 Department of Orthopaedics, King Chulalongkorn Memorial Hospital, Bangkok, Thailand

Introduction: Surgical management of adjacent segment disease (ASD) often involves extending the previous fusion to include the degenerated adjacent segments. Oblique lumbar interbody fusion (OLIF) is a minimally invasive option for treating ASD following primary lumbar fusion. Traditionally, following retroperitoneal approach for OLIF, a separate extended posterior approach needs to be performed to revise posterior construct and extend fusion with posterior pedicle screws (OLIF-PPS). However, OLIF with anterolateral screw-rod fixation (OLIF-ASR) allows single-position fixation through the same retroperitoneal approach with less invasiveness, however, its comparative outcomes remain unclear. This study compares the two-year clinical and radiographic outcomes of OLIF-ASR versus OLIF-PPS for ASD. Material and Methods: Fifty-five patients undergoing single-level OLIF for symptomatic ASD at L2-3 or L3-4 between January 2018 and September 2023 were reviewed. Twenty patients received OLIF-ASR, and 35 received OLIF-PPS. Clinical outcomes included Oswestry Disability Index (ODI) and visual analog scale scores for back (VAS-B) and leg pain (VAS-L) at 1 month, 6 months, 1 year, and 2 years postoperatively. Radiographic outcomes included disc height, foraminal height, segmental and lumbar lordosis. Fusion status and cage subsidence were assessed on CT scan. Operative time, blood loss, hospitalization, and complications were also analyzed. Results: Both groups showed significant improvement in ODI, VAS-L, and VAS-B at all follow-up intervals compared with baseline (p < 0.01). VAS-L improvement was comparable between groups. However, VAS-B reduction was significantly greater in the OLIF-ASR group at all time points, with a mean 2-year score of 1.9 ± 0.6 versus 2.6 ± 0.7 in OLIF-PPS (p < 0.05). ODI improvements did not differ significantly at final follow-up. Both groups demonstrated significant postoperative improvement in disc height, foraminal height, and sagittal alignment, but no significant differences between groups were observed. Fusion was achieved in 88% of OLIF-ASR and 93% of OLIF-PPS cases (p = 0.53). Subsidence rates were similar (13% vs 8.6%, p > 0.05). Perioperatively, OLIF-ASR was associated with significantly shorter operative time (95 ± 22 min vs. 143 ± 29 min, p < 0.01), less blood loss (50 ± 24 mL vs. 97 ± 46 mL, p < 0.01), and shorter hospitalization (3.0 ± 0.9 vs. 4.3 ± 1.1 days, p < 0.01). The overall complication rate was not significantly different between groups; no major vascular complications or major neurological injury were observed. There was one patient with cage migration following OLIF-ASR which subsequently needed revision surgery. Conclusion: Both OLIF-ASR and OLIF-PPS provide favorable clinical and radiographic outcomes for ASD at two years. OLIF-ASR offers comparable fusion and safety profiles while offering the advantages of reduced operative time, blood loss, hospital stay, and superior back pain improvement. These findings support OLIF-ASR as a less invasive alternative to posterior fixation in selected patients requiring revision surgery for ASD.

ID: AGS26/149

A354: Learning curve and CUSUM analysis of the interlaminar approach in full-endoscopic spine surgery: a phase-specific evaluation

Takeshi Kaneko 1

1 Inanami Spine and Joint Hospital, Tokyo, Japan

Introduction: Full-endoscopic spine surgery (FESS) via the interlaminar approach has been increasingly adopted as a minimally invasive and effective technique. However, its mastery requires stepwise skill acquisition. This study aimed to analyze the learning curve of FESS by dividing the procedure into operative phases and applying Cumulative Summation (CUSUM) analysis to determine phase-specific proficiency thresholds. Materials and Methods: This retrospective study included 123 consecutive single-level lumbar FESS cases performed by three novice surgeons (Surgeon A: 52 cases, Surgeon B: 41 cases, and Surgeon C: 30 cases) between June 2023 and August 2025. Each procedure was segmented into three phases: • Phase 1: Skin incision to ligamentum flavum exposure • Phase 2: Flavum exposure to flavum removal and nerve root visualization, • Phase 3: Nerve root visualization to herniated disc removal and wound closure, Operative time was recorded for each phase. Learning curves were assessed using CUSUM analysis to evaluate stabilization and proficiency thresholds. Results: CUSUM analysis revealed distinct patterns of proficiency across the three surgeons. In Phase 1, proficiency was achieved relatively early, with Surgeon A reaching the decision limit at the 12th case, Surgeon B at the 13th case, and Surgeon C at the 8th case. Phase 2 showed greater variability: Surgeon A achieved proficiency at the 3rd case and Surgeon C at the 4th case, while Surgeon B required more experience, reaching proficiency only at the 23rd case. Phase 3 demonstrated the steepest learning curve, with proficiency achieved at the 21st case for Surgeon A, the 32nd case for Surgeon B, and the 8th case for Surgeon C. Overall, Phases 1 and 2 tended to stabilize earlier, whereas Phase 3 consistently demanded more cases, highlighting its technical complexity and the need for accumulated surgical experience in herniated disc removal. Conclusion: The learning curve of interlaminar FESS is phase-dependent. Phase 1 and Phase 2 can be mastered within a relatively small number of cases, although variability exists among surgeons. In contrast, Phase 3, which involves herniated disc removal, consistently required more operative experience before proficiency was achieved, underscoring its technical complexity. These findings suggest that accumulated surgical experience plays a decisive role in mastering advanced steps of FESS. While cadaveric training may be sufficient for Phase 1 and Phase 2, complementary strategies such as structured video-based learning and observation of expert-performed surgeries may be necessary to accelerate proficiency in Phase 3 and ensure safe and effective adoption of endoscopic spine surgery.

ID: AGS26/1566

A355: Optimizing OLIF strategy for degenerative spondylolisthesis: robot-assisted single-position technique versus combined fixation approaches

Xilin Liu 1

1 Department of Orthopedics, Sichuan Provincial People's Hospital, School of Medicine,University of Electronic Science and Technology of China, Chengdu, China

Introduction: Degenerative lumbar spondylolisthesis is a leading cause of axial back pain and radiculopathy in adults and frequently necessitates instrumented interbody fusion to restore segmental alignment and achieve neural decompression. Oblique lumbar interbody fusion (OLIF) has emerged as a minimally invasive technique that enables indirect decompression and reliable disc-height restoration while minimizing exposure-related morbidity. The optimal choice of supplemental fixation after OLIF remains contentious: posterior percutaneous pedicle screw constructs afford strong posterior support and slip reduction, whereas lateral vertebral or plate fixation avoids a posterior approach but raises concerns about cage subsidence and long-term maintenance of disc height. Robot-assisted single-position OLIF may combine OLIF’s advantages with enhanced implant accuracy and procedural efficiency. This retrospective study compares clinical, radiographic, and perioperative outcomes among RA-OLIF, OLIF+VF, and OLIF+PPS to guide evidence-based surgical selection. Material and Methods: We retrospectively reviewed the clinical data of 85 patients with single-level lumbar degenerative spondylolisthesis who underwent RA-OLIF (n = 35), OLIF+PPS (n = 28), or OLIF+VF (n = 22) between March 2017 and January 2024 in the Department of Orthopedics, Sichuan Provincial People’s Hospital,Chengdu,China. Clinical outcomes and complications were assessed, along with radiographic parameters at 6 and 12 months postoperatively, including disc height (DH), slip ratio, cage subsidence, lumbar lordosis (LL), fused segmental angle (FSA), and fusion rate. Statistical analyses were performed using SPSS 26.0, with one-way ANOVA and LSD-t tests applied for group comparisons. A P-value < 0.05 was considered statistically significant. Results: All three groups achieved satisfactory clinical outcomes, with no significant differences in postoperative and follow-up VAS scores for back and leg pain or ODI scores (p > 0.05). No major complications such as vascular or visceral injury were observed. Transient thigh numbness and mild hip flexion weakness occurred in each group but resolved within one month, with no difference in incidence (p > 0.05). Intraoperative blood loss was minimal across groups, with the OLIF+VF group showing the least blood loss due to the absence of a posterior incision (p < 0.05). Operative duration was shortest in the OLIF+VF group, followed by RA-OLIF, and longest in OLIF+PPS (all pairwise p < 0.05). At 6- and 12-month follow-ups, RA-OLIF and OLIF+PPS demonstrated comparable improvements in DH restoration and slip correction (p > 0.05), both superior to OLIF+VF (p < 0.05). Cage subsidence was significantly higher in the OLIF+VF group than in the other two groups (p < 0.05), while RA-OLIF and OLIF+PPS showed no difference (p > 0.05). Improvements in LL, FSA, and fusion rate were similar across all groups (p > 0.05). Conclusion: RA-OLIF, OLIF+VF, and OLIF+PPS are all effective surgical options for lumbar degenerative spondylolisthesis. RA-OLIF and OLIF+VF offer notable advantages in surgical efficiency. Compared with OLIF+VF, RA-OLIF provides superior maintenance of postoperative disc height, better slip reduction, and reduced risk of cage subsidence. Further validation with multicenter, large-sample, and long-term studies is warranted.

ID: AGS26/1391

A356: Evaluation of students perception of the impact of a endoscopic post-graduation program in their professional skills to perform endoscopic surgery

Rafael Lindi Sugino 1 , Weby Mizael 2 , Fernanda Petreche 1 , Jullyene Pokorny 3 , Gabriel Pokorny 3

1 Einstein Hospital Israelita, São Paulo, Brazil

2 Weby Mizael, São Paulo, Brazil

3 Instituto de Patologia da Coluna, São Paulo, Brazil

Introduction: Endoscopic spine surgery (ESS) is a rapidly evolving field that offers significant advantages over traditional open surgery. However, its learning curve requires appropriate training, courses, and continuous professional development for spine surgeons. The present study aims to obtain quantitative and qualitative information from graduates of a one-year Post-Graduate program in Endoscopic Spine Surgery to quantify and qualify the advantages and disadvantages observed by the subjects after completing the course. Methods: A quantitative, prospective, pre- and post-test study was conducted with surgeons participating in a post-graduate program in spinal endoscopy. Data were collected through a questionnaire that assessed self-perception of safety (0-10 scale) and clinical practices (Yes/No). Paired numerical variables were compared using the Wilcoxon signed-rank test, and paired categorical variables were evaluated with McNemar's test (p < 0.05). Results: A statistically significant increase (p < 0.001) in self-perceived safety was demonstrated for all procedures, with substantial gains in lumbar cases. Additionally, a significant reduction in the need for a proctor/assistance (p < 0.001), an increased ability to complete surgeries (p < 0.001), and a important adoption of advanced techniques were revealed, including cervical and thoracic cases, which were not performed by any participant before the course (p = 0.002). Conclusion: The post-graduate program in endoscopic spine surgery proved to be a highly effective training tool. The program not only significantly increased the surgeons' confidence but also directly translated into a measurable expansion of their technical skills, enabling the adoption of complex procedures. It is concluded that structured education is fundamental to accelerating the learning curve and promoting autonomy in the practice of advanced surgical techniques.

ID: AGS26/611

A357: Analysis of imaging findings, comprehensive endoscopic evaluation, and pathological characteristics of a novel interlaminar ligament

Xiaofeng Le 1

1 Ministry of Health holding, Khoo Teck Puat Hospital (KTPH), Singapore, Singapore

Introduction: This study aimed to characterize a previously unnamed interlaminar ligament (ILL) using magnetic resonance imaging (MRI), endoscopic visualization, and histopathological examination. Material and Methods: A retrospective analysis was conducted on patients who underwent posterior endoscopic lumbar surgery for disc herniation or spinal stenosis between January 2021 and February 2022. The height, width, and cross-sectional thickness of the ILL were measured on T2-weighted MRI. Morphological and histological features were compared with those of the ligamentum flavum (LF) to identify structural differences. Results: Forty-three patients (27 males, 16 females; mean age 46.6 ± 12.1 years) were included, with 20 cases of disc herniation and 23 of spinal stenosis. In the herniation group, mean ILL width, height, and thickness were 17.7 ± 3.5 mm, 4.3 ± 1.3 mm, and 18.3 ± 3.5 mm; LF thickness was 5.3 ± 1.9 mm; surgical time averaged 53.2 ± 14.5 minutes. In the stenosis group, corresponding values were 16.0 ± 3.1 mm, 4.1 ± 1.6 mm, 17.6 ± 4.8 mm (ILL), 6.3 ± 0.8 mm (LF), and 61.8 ± 12.4 minutes. LF thickness (p = 0.02) and surgical time (p = 0.04) differed significantly. Endoscopic assessment showed distinct differences in anchor points and gross appearance. Histology revealed differences in fibrous density and orientation. ILL thickness grading showed a trend toward significance based on pathology (p = 0.09). Conclusion: The ILL is an unclassified anatomical structure with distinct morphological and histological differences from the LF. It serves as a key landmark in posterior endoscopic lumbar surgery. Variations in LF and ILL thickness across pathologies highlight its clinical importance.

ID: AGS26/685

A358: Visualized reamer foraminaloplasty in transforaminal endoscopic lumbar discectomy: controlling fluoroscopy times and optimizing surgical workflow

Honglin Liu 1 , Yuxiang Hu 1 , Zhuoxuan Zhang 1 , Zhinxin Kang 1 , Zibo Gao 1 , Dingkun Lin 1,2 , Yi-Hao Liang 1,2 , Yongjin Li 1,2

1 The Second Clinical School of Guangzhou University of Chinese Medicine, Guangzhou, China

2 Orthopaedic Hospital of Guangdong Provincial Hospital of Chinese Medicine, Guangzhou, China

Introduction: Foraminaloplasty is a critical step in transforaminal endoscopic lumbar discectomy (TELD). The traditional TESSYS technique performs foraminaloplasty under X-ray fluoroscopy guidance, which has a relatively steep learning curve and poses risks of nerve and dural sac injury during the stepwise foraminaloplasty. In recent years, the visualized foraminaloplasty technique proposed by Chinese scholars integrates the advantages of open and minimally invasive surgeries, enabling targeted articular process resection/plasty with clear visualization of anatomical structures. This study aims to compare the clinical efficacy and perioperative indicators of visualized reamer foraminaloplasty versus fluoroscopy-guided foraminaloplasty in TELD, and to explore the clinical value of the fully visualized technique in reducing intraoperative fluoroscopy times and optimizing surgical workflow. Material and Methods: A retrospective study was conducted on 54 patients who underwent TELD from January to December 2023. Patients were divided into the visualized reamer group (27 cases, 18 males and 9 females) and the TESSYS group (27 cases, 16 males and 11 females) according to the foraminaloplasty method. There were no significant differences in baseline characteristics between the two groups. All patients were followed up for at least 12 months postoperatively. Clinical efficacy was evaluated using the Visual Analogue Scale (VAS) for low back and leg pain at preoperative and follow-up visits. Functional outcomes were assessed by the Oswestry Disability Index (ODI) and McNab criteria. Perioperative indicators (including intraoperative fluoroscopy times, surgical duration, and intraoperative blood loss) were recorded for both groups. Results: Both groups showed significant improvement in VAS scores for low back and leg pain at postoperative and all follow-up time points compared with preoperative baseline (all p < 0.05). No significant differences were observed between the two groups in VAS, ODI, or McNab criteria at follow-up. Compared with the TESSYS group, the visualized reamer group had fewer intraoperative fluoroscopy times [(6.19 ± 0.74) times vs. (12.22 ± 0.97) times, p < 0.05], shorter surgical duration [(71.26 ± 10.74) min vs. (82.95 ± 14.51) min, p < 0.05], and less intraoperative blood loss [(6.44 ± 2.68) ml vs. (9.07 ± 4.17) ml, p < 0.05]. Conclusion: Visualized reamer foraminaloplasty and fluoroscopy-guided foraminaloplasty yield comparable postoperative clinical outcome in TELD. However, the fully visualized technique significantly reduces intraoperative fluoroscopy times, shortens surgical duration, and reduces intraoperative blood loss. For beginners in minimally invasive spine surgery, with the assistance of visualized reamer spinal endoscopic technology, it is easier to identify anatomical structures in the foraminal region, accurately determine the location and extent of articular process plasty, and gradually establish a standardized and highly reproducible surgical workflow.

ID: AGS26/750

A359: Expanding endoscopic application to neuromodulation: novel biportal endoscopic insertion technique for surgical paddle lead

Young San Ko 1 , Yong Jin Park 2 , Dae-Chul Cho 1

1 Kyungpook National University, School of Medicine, Kyungpook National University Hospital, Daegu, South Korea

2 Suncheon Hana Hospital, Suncheon, South Korea

Introduction: Surgical paddle-lead insertion for spinal cord stimulation (SCS) is a cornerstone therapy for chronic refractory pain, offering lower impedance and reduced battery usage compared with percutaneous leads. However, conventional surgical implantation is invasive and associated with procedural morbidity. Uniportal endoscopic approaches remain limited by visualization and instrumentation. We evaluated a novel unilateral biportal endoscopic (UBE) technique for SCS paddle-lead implantation, combining minimal invasiveness with stable lead positioning. Material and Methods: A retrospective analysis was performed of patients who underwent UBE-SCS between 2023 and 2025 at a single tertiary center. Clinical, surgical, and radiological data were reviewed from medical records and intraoperative videos. The UBE technique employed to perform partial laminotomy, flavectomy, and paddle-lead placement, followed by a trial period and permanent implantation in responders. Operative parameters, complications, and clinical outcomes were assessed. Results: A total of 19 patients (6 females, 13 males; mean age 57.3 years) were included. Mean operative time was 85.6 ± 21 minutes, with minimal intraoperative blood loss (< 50 mL). After the trial period, 16 patients (94.1%) achieved successful implantation. Complications occurred in 3 patients (15.7%): two transient thoracic radiculopathies and one asymmetrical lead position, all resolving without sequelae. No lead migrations were observed. At a median follow-up of 6 months, mean VAS improved significantly from 8.0 ± 1.1 to 3.1 ± 1.3 (p < 0.01). Conclusion: UBE-SCS is a feasible and effective minimally invasive alternative to conventional paddle-lead implantation, achieving high technical success, significant pain reduction, and acceptable complication rates. By combining the stability of surgical leads with reduced invasiveness, this technique expands the role of endoscopy in neuromodulation. Further prospective, large-scale, long-term studies are warranted to validate durability and refine patient selection.

Keywords: spinal cord stimulation, biportal endoscopy, failed back surgery syndrome, chronic pain, neuromodulation

ID: AGS26/169

A360: Mid-term clinical outcomes of unilateral biportal endoscopic discectomy via the contralateral sublaminar approach for very high-grade lumbar disc migration

Yong Jin Park 1 , Sang Kyu Son 2 , Young San Ko 3

1 Suncheon Hana Hospital, Suncheon, South Korea

2 Good Moonhwa Hospital, Busan, South Korea

3 Kyungpook National University Hospital, Daegu, South Korea

Introduction: Lumbar disc herniation (LDH) with very high-grade migration presents a significant surgical challenge due to the difficulty in accessing migrated fragments without excessive bone resection or neural manipulation. The unilateral biportal endoscopic (UBE) contralateral sublaminar approach offers a minimally invasive alternative, enabling decompression through an existing sublaminar corridor, thereby minimizing bone removal and preserving spinal stability. This study aimed to evaluate the mid-term clinical and radiological outcomes of this technique in a larger patient cohort. Material and Methods: This multicenter retrospective study included 30 patients who underwent UBE discectomy via the contralateral sublaminar approach for LDH with very high-grade upward or downward migration between June 2022 and July 2024. Patients with at least 12 months of clinical and radiological follow-up were included. Clinical outcomes were assessed using the visual analog scale (VAS) for back and leg pain, Oswestry Disability Index (ODI), and modified MacNab criteria. Radiological assessments included postoperative MRI to confirm adequate decompression and evaluate for residual disc fragments, as well as serial lateral flexion-extension radiographs to assess segmental stability. Results: The mean follow-up period was 15.6 months. The average operation time was 46.3 ± 7.4 minutes, and the mean hospital stay was 4.3 ± 0.7 days. VAS scores significantly improved for back pain (from 4.8 ± 0.6 to 1.7 ± 0.5) and leg pain (from 8.1 ± 0.7 to 1.1 ± 0.3), respectively (p < 0.001). ODI scores decreased from 75.7 ± 3.9 to 9.1 ± 1.6 (p < 0.001). Postoperative MRI confirmed complete decompression in all cases. Two patients experienced transient postoperative hypesthesia, which resolved spontaneously without further intervention. No iatrogenic instability or major complications were observed during follow-up, and no cases of recurrent disc herniation were reported. Conclusion: This study suggests that the contralateral sublaminar UBE approach provides effective and durable decompression for very high-grade migrated lumbar disc herniation, with favorable mid-term clinical and radiological outcomes. Further studies are warranted to validate its long-term efficacy and define its optimal indications.

OP41: AI, AR, and Robotics in Spine

ID: AGS26/1007

A361: Evidence-based AI decision support system for spine surgery: a novel approach using dynamic PubMed integration

Begüm Aslantaş Kaplan 1 , Ali Kaplan 1 , İlker Solmaz 2

1 Tokat Erbaa State Hospital, Tokat, Turkey , 2 Private Practice, Ankara, Turkey

Introduction: Decision-making in spine surgery is a complex process defined by evolving evidence and nuanced patient factors. While evidence-based medicine is the standard of care, the rapid expansion of literature makes it challenging for clinicians to remain constantly updated. Large Language Models (LLMs) offer potential for information synthesis, but their clinical use is limited by risks of hallucination and generic responses. This study introduces a novel evidence-based AI decision support system that dynamically integrates real-time PubMed database queries to provide contemporary, literature-supported clinical recommendations. Material and Methods: We developed an AI system using Model Context Protocol (MCP) methodology, enabling ChatGPT-5 to access PubMed API in real-time. Unlike fine-tuning, which freezes knowledge at a point in time, or static RAG systems that query a fixed local database, our approach establishes a direct, real-time query-response loop with the PubMed API for every clinical prompt. This eliminates the need for manual model updates and guarantees the system remains perpetually up-to-date with the latest medical publications. A senior spine surgeon curated twelve challenging clinical scenarios from anonymized past cases, covering trauma, infection, tumor, and degenerative pathologies, each with confounding variables to challenge the decision-making process. The AI system generated comprehensive reports for each scenario. These AI recommendations were then independently evaluated by three spine surgeons, blinded to the MCP methodology: (1) diagnostic accuracy (1-10 scale), (2) treatment approach appropriateness (conservative vs. surgical), (3) surgical technique selection rationale, (4) evidence quality and citation accuracy, (5) absence of hallucinations or fabricated references, (6) clinical reasoning coherence, and (7) risk-benefit analysis completeness. Results: The evidence-based AI system demonstrated high performance across all evaluation domains. Mean diagnostic accuracy scores ranged from 9.2-9.8 across the three evaluators (ICC = 0.82). Treatment approach appropriateness achieved consistently high ratings (mean 8.5, range 7.9-9.1), with 91.7% agreement between AI recommendations and expert consensus. Surgical technique selection showed moderate agreement (mean 7.8), reflecting inherent variability in surgical approaches among experts. Critically, no hallucinated references were identified, with 100% of cited literature verified as authentic and accurately represented. Evidence quality scores averaged 8.4, indicating robust literature synthesis. Clinical reasoning coherence and risk-benefit analysis completeness both exceeded 8.0 mean scores across all evaluators. The system successfully avoided generic recommendations, with all outputs directly traceable to specific PubMed articles. Conclusion: This novel evidence-based AI decision support system represents a significant methodological advancement from traditional AI applications by ensuring real-time literature integration through MCP, which is distinct from static RAG systems and fine-tuning approaches that require manual updates and become progressively outdated. The high expert validation scores demonstrate the system's potential for reliable clinical decision support in complex spine surgery cases. The MCP-enabled dynamic querying provides transparent, traceable evidence sources while maintaining perpetually current knowledge without manual intervention, achieving clinical accuracy comparable to expert recommendations. This methodology offers a promising framework for developing trustworthy, automatically up-to-date AI tools in surgical specialties, addressing key limitations of conventional AI systems through evidence-based, real-time knowledge synthesis.

ID: AGS26/1062

A362: Distance-aware augmented reality rendering platform with enhanced accuracy through user-specific eye calibration

Sepehr Madani 1,4 , Carlo Santaguida 2 , Robert Turcotte 1,3,4 , Amir Hooshiar 4 , Ahmed Aoude 1,3,4

1 Department of Orthopedic Surgery, McGill University

2 Department of Neurology and Neurosurgery, McGill University

3 Department of Surgical Oncology, Gerald Bronfman Department of Oncology, McGill University

4 McGill SuPER Center for Robotic Surgery

*At the time of submission of this abstract for GSC2026 the complete version of this work where we have discussed all the mathematical models and technical implementations has been submitted to the IEEE Transactions on Visualization and Computer Graphics and is under review.

Introduction: Augmented reality(AR) in minimally invasive spine surgery(MISS) offers real-time enhanced visualization of target anatomy and critical structures such as boundaries of a bone tumor or vertebrae submerged under the skin. However, clinically significant parallax-induced projection errors (particularly during close AR goggle–anatomy distances) drastically deteriorate its projection accuracy. These optical distortions are amplified by individual user eye variations. Objective: To develop and validate a device-agnostic, indifferent to inter-user optical variabilities, active parallax-effect compensated (APC) framework that minimizes projection error across variable surgeon-to-target distances without reliance on fiducial markers. Methods: Major AR projection system components are as follows: Optical See-through Head-Mounted Display(OSTHMD) and Target Anatomy Tracked by Infrared Surgical Navigation System(NAV). We devised a fiducial-free OSTHMD–NAV non-linear optimization-based spatial registration framework. Afterwards, two canonical working distances were sampled: “near”(≈350 mm) and “far”(≈800 mm). Using an ASTM-F2554-22 reference block, the operator first probes 4 physical landmarks with an IR-tracked stylus, then acquires the corresponding virtual points rendered by the OSTHMD in the workspace. The system automatically calibrates itself using our newly devised APC algorithm. Five clinicians participated in a prospective validation. Our user-specific calibration protocol was performed prior to real-anatomy visualization, followed by AR projection accuracy assessments on a 3D-printed lumbar spine model. Six anatomical landmarks were used: L4 spinous process base(p1), L3 right transverse process(p2), L3 spinous process(p3), L2 left transverse process(p4), L2 spinous process(p5), and L1 mid-right transverse process(p6). The devised real-time algorithm dynamically adjusted rendering based on goggle-to-anatomy distance using interpolated transformation matrices derived from six individual hyperbolic tangent functions in tune-space. Projection accuracy was quantified in APC ON vs. OFF conditions at near and far viewing distances using stratified sphere indicators on both ASTM block and spine model anatomic landmarks. Results: Without APC, projection errors averaged 8.50 ± 2.98 mm(near) and 6.26 ± 2.54 mm(far). Enabling APC reduced these to 2.10 ± 1.35 mm and 1.72 ± 1.27 mm, respectively. Our method successfully represents a 75.2%(near) and 72.5%(far) spatial projection accuracy enhancement. Our projection accuracy is unprecedented in literature. Two one-sided t-tests(TOST) confirmed statistical equivalence of near and far projection errors(p < 0.05) within submillimeter equivalency margins. Conclusion: This study demonstrates that user-calibrated APC enables millimeter-scale AR accuracy for both near and far operating distances, addressing a critical barrier to clinical precision AR-assisted MISS deployment. Our fiducial-free, distance-aware system supports precise visualization in spine and orthopedic-oncology surgery and may advance AR integration into our robotic workflows.

ID: AGS26/1099

A363: New treatment approaches for early-onset scoliosis: early results of the combined use of magnetically controlled growing rod and active apex control technique

Mehmet Balioğlu 1 , Muhammet Işık 1 , Kadir Abul 1

1 Basaksehir Cam and Sakura City Training and Research Hospital, Department of Orthopedics and Traumatology, University of Health Sciences, Hamidiye Faculty of Medicine, Department of Orthopedics and Traumatology, İstanbul, Turkey

Introduction: While various growth methods have been described for the surgical treatment of early-onset scoliosis (EOS), studies on hybrid techniques are limited. Active apical control (APC) hybrid surgical technique is one of the newer methods that allows growth while correcting the deformity. This study evaluated changes in radiological parameters associated with deformity and spinal lengthening over time using APC. This study aimed to evaluate the effectiveness of magnetically controlled growing rods (MCGR) on the concave side of the curve and APC on the convex side in the treatment of EOS. This study aimed to maintain convex apical growth, unlike previously described convex apical fusion and compression. Material and Methods: The study included patients with a Cobb angle > 50°, a bone age of < 10 years, a minimum follow-up of 12 months, and no previous surgery for scoliosis. Twenty patients (14 female) who met these criteria were retrospectively evaluated. The distribution of cases according to their etiologies was 10 Neuromuscular, 4 Congenital, 3 Syndromic and 3 Idiopathic. The system was designed to allow distraction with the MCGR on the concave side and growth with the APC on the convex side. For APC, growth was allowed between the convex apical screws, compression and fusion were not applied. Major Cobb angle (MCA), apicall vertebral translation (AVT), T1-S1 and T1-T12 heights, kyphosis and lordosis were measured preoperatively, early postoperatively, and at final follow-up. Descriptive statistics were performed on the obtained data, and measurements at three time points were compared. Statistical significance was considered at p< 0.05. Results: The mean age of the patients at the time of surgery was 95 months (51-136), and the mean follow-up period was 21 months (12-31). At preoperative and final follow-up, there was a significant decrease in mean MCA and AVT values (40.8% and 49.8%), and a significant increase in T1-S1 and T1-T12 lengths (21% and 19%). In the sagittal plane, there was a significant decrease in kyphosis (52%), while no significant change was observed in lordosis (8%). When comparing early postoperative and final follow-up measurements, no significant change was observed in MCA and AVT values, but there was a significant increase in T1-S1 and T1-T12 heights. There was no significant change in kyphosis or lordosis angles. The mean operative time was 366 minutes, and the mean bleeding volume was 451 ml. Blood transfusion were required in 10 cases. Complications and unplanned surgery were required in 5 patients, 7 required additional surgery. Conclusion: The hybrid technique was concluded to be a promising method for both deformity control and preservation of growth potential, making it a useful early-stage treatment for EOS. Multicenter studies with longer follow-up periods and larger series are needed to better evaluate the hybrid technique compared to other methods

ID: AGS26/1398

A364: Democratizing the adoption of artificial intelligence: a global survey of spine surgeons' needs and priorities for AI

Luke Jouppi 1,2 , Clifford Pierre 1,2 , Beth Stella 2 , Jeremiah Laichena 3 , Oliver Mrowczynski 1,2 , Mohamad Nabih El Housheimy 1,2 , Hagay Amster Kahn 1,2 , Mohsen Karami 1,2 , Arnav Gambir 1,2 , Tara Heffernan 1,2 , Mark Kraemer 1,2 , Amir Abdul-Jabbar 1,2 , Rod Oskouian 1,2 , Jens Chapman 1,2

1 Swedish Neuroscience Institute, Swedish Medical Center, Seattle, United States

2 Seattle Science Foundation, Seattle, United States

3 Mount Kenya University, Nairobi, Kenya

Introduction: The application of artificial intelligence (AI) to spine surgery has experienced a dramatic rise in interest in recent years. A critical gap persists between existing proof-of-concept models and our understanding of the real-world needs of practicing surgeons. This study aims to identify global spine surgeons’ priorities, perceived benefits, and barriers for AI deployment. Secondarily, we aim to support the ethical deployment of AI across diverse practice settings by elucidating whether any regional differences exist. Material and Methods: On August 23, 2025, the authors distributed a 9-question Qualtrics survey through the Seattle Science Foundation’s electronic mailing list. The questionnaire began with demographics and information about respondents’ practice settings before inquiring whether respondents had used AI in their practice. Next, Likert scales assessed physicians’ interest in several applications for AI, ranging from 0 to 5: 0 was presented as “not useful at all” and 5 was equivalent to “essential for my practice.” Respondents were then asked to select their top three barriers to implementing AI in their practice. Results: As of September 11, 2025, we received 134 responses. Regarding experience, 45 respondents (34%) identified as spine surgeons in training, 15 respondents (11%) identified as having practiced for under 5 years, 18 (15%) identified as having practiced between 5-10 years, 18 respondents (13%) reported between 10-20 years, and 41 respondents (31%) reported over 20 years in practice. Seventy physicians (52%) reported current use of AI in their practice, 9 (7%) endorsed previous use of AI, and 55 (41%) reported never having used AI in their practice. On the Likert scale ranging from not useful (0) to essential (5), physicians mean ratings for radiological AI were 3.38 ± 1.40, 3.61 ± 1.17 for predictive modeling, 3.95 ± 1.04 for administrative AI and practice optimizations, 3.46 ± 1.14 for intraoperative guidance, 3.07 ± 1.39 for decision-making support, and 3.94 ± 1.22 for patient education. Analysis of variance (ANOVA) revealed significant differences in physicians’ priorities for AI (p < 0.0001). Administrative and practice optimization technologies and patient education assistance presented the greatest interest. Subsequent analysis via Tukey’s Honestly Significant Difference revealed administrative AI received significantly greater interest than radiology AI models (p<0.01) and decision-making support models (p < 0.001). Models supporting patient education also received significantly greater interest than radiological AI (p = 0.01) and AI for decision-making support (p < 0.001). Respondents identified technological maturity, ethical and legal concerns, and familiarity with these technologies as the most impactful to their adoption with 56%, 50%, and 48% of respondents reporting these concerns as the most impactful barriers to adopting AI, respectively. Conclusion: Many physicians and a majority of respondents have already begun integrating AI into their practice. Ensuring AI best supports physicians’ and patients’ needs may prove essential as this technology continues to mature. Spine surgeons expressed the strongest interest in AI to decrease the administrative burdens of clinical practice and to support patient education. Our study also underscored the need for further validation of these technologies, updated regulations on their implementation, and educational materials to familiarize physician-colleagues with AI.

ID: AGS26/1390

A365: Objective measurement of surgical forces and jerks for a more analytical future in spinal surgical training and practice - A study with novel sensorised force-gloves and interpretable machine learning

Chan Hee Koh 1,2,3 , James Booker 1,2 , Carmen Salvadores Fernandez 4 , Yang Xue 4 , David Choi 1,2 , Parag Sayal 2 , Danail Stoyanov 5 , George Prezerakos 1,2,3 , Manish Tiwari 4 , Hani Marcus 1,2

1 Department of Translational Neuroscience and Stroke, Queen Square Institute of Neurology, University College London, London, United Kingdom

2 Department of Neurosurgery, National Hospital for Neurology and Neurosurgery, London, United Kingdom

3 Neurosciences Institute, Cleveland Clinic London, London, United Kingdom

4 Nanoengineered Systems Laboratory, Department of Mechanical Engineering, University College London, London, United Kingdom

5 Department of Computer Science, Hawkes Institute, University College London, London, United Kingdom

Introduction: The ability to apply appropriate surgical forces is crucial for mastery in spinal surgery. However, these have traditionally been difficult for an external assessor to measure and objectively assess, more often relying on subjective visual assessments and feedback. We have developed universal sensorized force gloves that were able to reliably measure force applied at the hand-tool interface. This also allows the measurement of “jerk”, which is the instantaneous change in force over time. This study leveraged interpretable machine learning and distributional analysis of surgical forces and jerks in spinal surgical simulations to assess surgical proficiency, with a view to advancing a more objective and analytical approach to surgical training and practice. Material and Methods: 373 force segments were collected from 28 novice and expert surgeons. High-fidelity simulated lumbar decompressions were performed while wearing sensorized surgical gloves. The jerk was calculated by taking the derivative of force with respect to time. These segments were used to train 510 interpretable machine learning algorithms to predict proficiency in lumbar decompressions. The distributional properties of forces and jerks were used as input features for machine learning. These were: (1) the first raw moment of the distribution (the mean); (2) the second unstandardized central moment (the variance i.e. variability in force applied); (3) the third standardized central moment (the skewness i.e. asymmetry in distribution of forces applied); (4) the fourth standardized central moment (the kurtosis i.e. propensity to outliers). Nested crossvalidation schemes were used to tune and train interpretable machine learning algorithms, and then to evaluate the performance on unseen data. The outputs of the best performing algorithm was inspected to investigate how the application of forces and jerks impact on surgical proficiency. Results: Objective and quantitative thresholds of surgical expertise, as it pertains to surgical forces and jerks, were derived from the machine learning algorithms. These thresholds were able to classify surgical proficiency with an excellent balanced accuracy of 95.9% (CI: 74.0% to 100.0%). The excellent overall discriminative utility of the model was demonstrated by AUC-ROC of 96.4% (CI: 84.0% to 100.0%). The outputs of machine learning algorithms showed that strongest predictors of surgical proficiency were: (1) greater mean force applied (log-odds - weak hand: 1.4 /N; strong hand: 0.4 /N); (2) smaller variance of force (log-odds - weak hand: -0.5 /N; strong hand: 0.2 /N); (3) smaller kurtosis of force (log-odds - weak hand: -0.02 /unit; strong hand: -0.01 /unit); and (4) smaller variance of jerk (log-odds - strong hand: -3.0 /N ms -1 ; weak hand: 0.3 /N ms -1 ). Conclusion: The attainment of surgical proficiency requires the ability to (1) apply greater force but (2) within tighter margins of variability and control, and (3) avoiding outliers in force application, while (4) also reducing sudden changes in force (jerks). These properties can be objectively measured using sensorised surgical gloves, and interpretable machine learning algorithms can be used to set objective and quantitative thresholds of surgical proficiency. These technologies may enable more objective and analytical spinal surgical training, assessments, and practice in the future.

ID: AGS26/2004

A366: Spatial spinal fixation: a new image-guided robotic technique for enhanced implant fixation in compromised vertebral bone

Yash Kulkarni 1 , Daniyal Maroufi 1 , Mobina Tavangarifard 1 , Maryam Tilton 1 , Justin Bird 2 , Shalin Patel 2 , Laurence Rhines 2 , Zachary Lawson 2 , Jeffrey Siewerdsen 2 , Farshid Alambeigi 1

1 University of Texas, Austin, United States

2 MD Anderson Cancer Center, Houston, United States

Introduction: Traditional spinal fixation techniques rely on rigid drills and straight screws to stabilize the spine. However, the complex anatomy of the spine often limits the surgeon’s ability to place implants along optimal trajectories. This challenge is especially critical in osteoporotic patients and patients with metastatic cancer of the spine, where low bone mineral density and/or lytic lesions increase the risk of screw loosening, pullout, and implant failure. To address these limitations, there is a growing need for steerable drilling systems and more adaptable implant designs that enable safer and more effective fixation in compromised bone. Material and Methods: To overcome the limitations of rigid instruments and linear drilling, we developed an image-guided minimally invasive Steerable Drilling Robotic system capable of creating straight and curvilinear trajectories, both in-plane and out-of-plane, within the vertebral body. To enable fixation along these customized paths, we also have designed, and 3D printed, Flexible Pedicle Screws using titanium and stainless steel. We refer to this technique as Spatial Spinal Fixation, as it enables secure placement of flexible implants along patient-specific, non-linear paths that would not be feasible with conventional tools. Results: Steerable drilling and flexible implant placement were guided and verified via 3D-2D registration across straight and J-shaped trajectories with radii from 120 mm to 25 mm, both in-plane and out-of-plane, in synthetic bone, animal models, and cadavers. Drilling channels 5-8 mm in diameter and 50 mm in length were created in < 35 s across Sawbones densities from PCF 5-30, with trajectory errors < 10%. Screws (6-9 mm OD) were placed and verified under C-arm imaging. Pullout testing showed FPS achieved fixation strengths up to 435 N, comparable to rigid screws. These results confirm the accuracy and robustness of Spatial Spinal Fixation for complex spinal trajectories. Conclusion: The Spatial Spinal Fixation approach shows strong potential to improve spinal fixation, especially in bone disease states such as osteoporosis, multiple myeloma, and metastatic cancer of the spine. By enabling screw placement along curved and off-axis trajectories, Spatial Spinal Fixation overcomes key limitations of rigid, linear fixation systems, enabling the surgeon to provide patient specific stabilization solutions, targeting higher bone density regions within the vertebrae.

ID: AGS26/301

A367: The performance and clinical translation of artificial intelligence in preoperative spine surgical planning: a systematic review and meta-analysis

Vinicius Ricieri Ferraz 1 , Guilherme Santos Piedade 2 , Carlos Goulart 3 , Maria Ricieri Ferraz Franco de Souza 4 , Marcelo Ochoa Coelho de Souza Furlan 5 , Philippe A. Mercier 6 , Tobias Mattei 6

1 Dr. José de Carvalho Florence Hospital, The Regional Hospital of São José dos Campos - Dr. Rubens Savastano, São José dos Campos, São Paulo, Brazil

2 Department of Neurosurgery, University of Utah, Salt Lake City, Utah, United States

3 Division of Neurological Surgery, Essentia Health, Duluth, Minessota, United States ,

4 Division of Neurology, Irmandade da Santa Casa de Misericordia de Santos, Santos, São Paulo, Brazil

5 Pontifical Catholic University of Paraná, Curitiba, Brazil

6 Division of Neurological Surgery, Saint Louis University School of Medicine,, Saint Louis, Missouri, United States

Introduction: Artificial intelligence (AI) has seen growing application in preoperative planning for spine surgery, enhancing image analysis, optimizing pedicle screw placement, simulating sagittal alignment, and predicting potential adverse events. Despite significant progress in academic research, the integration of AI into clinical practice remains limited. This systematic review and meta-analysis evaluate the effectiveness of AI systems in preoperative spine surgical planning and identify barriers to their clinical implementation. Material and Methods: Following PRISMA guidelines, we conducted a comprehensive search of MEDLINE, EMBASE, Scopus, Web of Science, IEEE Xplore, the Cochrane Library, and arXiv from 2010 through July 2025. We included studies that developed or validated artificial intelligence tools for preoperative spine surgical planning, such as CT/MRI segmentation and 3D reconstruction, pedicle screw trajectory planning, alignment simulation, risk stratification for mechanical complications, and outcome prediction. The quality and potential bias of studies were evaluated using QUADAS-2, PROBAST-AI, and ROBINS-I. When applicable, meta-analyses were performed; otherwise, results were synthesized narratively and aligned with implementation frameworks. Results: Twenty-four studies met the inclusion criteria. The majority were retrospective, single-center investigations, while a minority were prospective studies. Two randomized controlled trials were identified among the included research. Additionally, several proof-of-concept preclinical studies using animal models were incorporated. In technical tasks, artificial intelligence (AI) demonstrated high performance, particularly in image segmentation, reconstruction, and pedicle screw planning, often achieving strong accuracy metrics. Comparative clinical studies indicated that AI- or navigation-assisted workflows could improve screw placement accuracy and reduce fluoroscopy time or radiation exposure, although sometimes at the cost of longer operative durations. Conversely, models designed to predict complex outcomes such as proximal junctional kyphosis, implant failure, or reoperations showed only modest to moderate discrimination capabilities. Furthermore, the generalizability of these models outside their original datasets remains limited. The heterogeneity in datasets, reporting practices, and endpoints prevented meta-analysis in several areas. Conclusion: AI in preoperative spine surgical planning demonstrates strong technical performance and early clinical benefits in specific tasks such as pedicle screw placement accuracy and radiation dose reduction. However, evidence for reliably predicting complex postoperative outcomes remains limited, and its adoption is hindered by problems such as dataset bias, lack of external validation, workflow integration challenges, and medico-legal considerations. Human-in-the-loop deployments, standardized reporting protocols, and multicenter prospective trials are needed to validate efficacy and define patient populations most likely to benefit.

ID: AGS26/3196

A368: Augmented reality surgical navigation in transforminal lumbar interbody fusion (TLIF) procedure: a comparative study

Giacomo Cammarata 1 , Francesco Certo 1 , Giuseppe Sotera 1 , Angelo Basile 1 , Sara Pappalardo 1 , Giuseppe Barbagallo 1

1 Department of Neurosurgery, Universitary Hospital G. Rodolico-San Marco“,Catania, Italy

Introduction: Transforaminal lumbar interbody fusion (TLIF) is a well-established procedure for treating degenerative lumbar spine conditions. With technological advance, Augmented Reality (AR) has emerged as a tool to improve precision, workflow, and reduce complications. Traditional TLIF relies heavily on fluoroscopy or neuronavigation to guide surgeons during critical phases such as pedicle screw placement. However, recent studies suggest that AR may improve outcomes by offering surgeons a more intuitive visual overlay that facilitates real-time guidance throughout the procedure. This study aims to evaluate the role of AR in minimally invasive (mini-open) TLIF compared to fluoroscopy-only and neuronavigation-assisted techniques. Material and Methods: We retrospectively reviewed 59 patients (17 male, 42 female; mean age 64.4 years, range 35-87) who underwent mini-open TLIF between January 2024 and June 2025. Patients were divided into three groups: fluoroscopy-only (17), neuronavigation (30), and AR-supported (12). Data collected included patient demographics, surgery duration, radiation exposure, pedicle screw placement accuracy, postoperative complication rates, and hospital stay. Postoperative imaging was segmented using the Elements Smartbrush software by BrainLab (BrainLab, Munich, Germany) to evaluate surgical accuracy, alignment, and osteotomy volume. Statistical analysis was performed to compare operative times, radiation exposure, and clinical outcomes among the groups. Results: Radiation exposure differed significantly across groups: 195.8 Cy·cm 2 in the fluoroscopy group, 12.1 Cy·cm 2 with navigation, and 3.4 Cy·cm 2 with AR (p < 0.0001). Clinical outcomes improved significantly within each group. Mean NRS decreased from baseline to follow-up in all groups (all p < 0.005 vs baseline): 8.6 versus 5.9 (Fluoro), 8.5 versus 4.9 (Nav), and 7.3 versus 4.3 (AR). Intergroup comparison showed no significant differences (p = 0.685), although AR patients demonstrated a greater numerical reduction. Similarly, ODI improved significantly in all groups (all p < 0.01 vs baseline): 62.5 versus 41.2 (Fluoro), 59.3 versus 38.5 (Nav), and 58.2 versus 35.7 (AR), with no significant intergroup differences (p = 0.504). Spinopelvic parameters remained stable across groups. At follow-up, PI was comparable (≈57-60°, p = 0.843), PT values were within normal compensation ranges (14-20°, p = 0.257), and SS tended to be higher in Fluoro and lower in Nav, with AR intermediate (p = 0.140). No significant differences were found, confirming alignment preservation across techniques. No group differences emerged in screw accuracy or complication rates, although AR showed a trend toward fewer revisions. Mean hospital stay was 6.1 days, with differences between groups (p = 0.0378). Surgeons reported that AR improved anatomical orientation and intraoperative confidence, particularly in complex or revision cases. Conclusion: Augmented reality in mini-open TLIF significantly reduces radiation exposure compared with fluoroscopy and neuronavigation. All groups experienced significant clinical improvement in pain and disability, with AR showing the greatest numerical reduction, although intergroup differences were not statistically significant. Spinopelvic parameters were preserved in all groups, confirming the safety of the approach. AR therefore represents a promising adjunct to conventional guidance systems, with potential to refine surgical precision and improve outcomes. Larger prospective studies are warranted to confirm long-term benefits and clarify its role in complex and revision cases.

ID: AGS26/527

A369: The diagnostic efficacy of MRI-based vertebral bone quality score in assessing bone mineral density in adolescent idiopathic scoliosis: a retrospective cohort study

Shahzaib Riaz Baloch 1 , Mohammad Sohail Rafi 1 , Raghad Barri 2

1 Dr. Ziauddin Hospital, Clifton Campus, Ziauddin University, Karachi, Pakistan

2 Dr. Soliman Fakeeh Hospital Riyadh, Riyadh, Saudi Arabia

Introduction: Adolescent idiopathic scoliosis (AIS) is frequently associated with low bone mineral density (BMD), which is crucial to assess pre-surgically. Current diagnostic methods like QCT involve ionizing radiation, highlighting the need for alternative screening tools. This study evaluates the diagnostic accuracy of a radiation-free, MRI-based Vertebral Bone Quality (VBQ) score for assessing BMD in patients with AIS. The objective is to evaluate the diagnostic accuracy of the MRI-based vertebral bone quality (VBQ) score in assessing bone mineral density (BMD) in patients with adolescent idiopathic scoliosis (AIS), and to compare its performance against existing modalities. Material and Methods: In this retrospective investigation, the hospital database (EMR) was searched for resources of tertiary care institution to identify AIS patients who received lumbar MRI scans between January 2020 and December 2024. Following approval of the ethical review committee, 200 patients with simultaneous MRI and quantitative computed tomography (QCT) data were enrolled. VBQ scores were derived from T1-weighted MRIs. Patients were divided into normal BMD (QCT Z-score > -2.0) or low BMD (QCT Z-score ≤ -2.0) groups. The univariate analysis was used to compare the variables between the groups. Pearson's test was used to evaluate the relationship between VBQ score and QCT Z-score as well. Independent factors for low BMD were determined by a multivariate logistic regression model. Receiver operating characteristic (ROC) curve analysis was performed to assess the diagnostic value of the VBQ score. Results: Out of the 200 patients with AIS (average age 14.9± 2.2 years) there were 68 patients with low BMD. Within the low BMD cohort there was a statistically significant difference between the VBQ scores of the low and normal BMD cohorts (3.59 ± 0.81 vs 2.58 ± 0.59, p < 0.001). A weak negative correlation was noted between the VBQ score and the QCT Z-score (r = -0.492, p < 0.001). After adjusting for other variables, VBQ score low BMD became an independent associated factor of low BMD (OR: 5.221, 95% CI 2.845-9.583, p < 0.001). According to ROC analysis the VBQ score was able to predict low BMD with an AUC score of 0.86. The optimal VBQ score of 3.24 had a sensitivity of 80.9% and a specificity of 84.8%. Conclusion: The VBQ score, derived from standard MRI, is a highly accurate and non-invasive tool for identifying low BMD in AIS patients. Its diagnostic performance is robust, making it a powerful opportunistic screening tool for pre-surgical evaluation and a valuable complementary measure to QCT.

OP42: Degenerative Lumbar Techniques

ID: AGS26/1072

A370: Accuracy and feasibility of S2AI screw placement using fluoroscopic guidance: a prospective clinical study

Nilesh Barwar 1

1 ALL India Institute of Medical Sciences, Bathinda, India

Introduction: S2 alar-iliac (S2AI) screw placement provides robust sacropelvic anchorage, especially in cases requiring lumbopelvic fixation. Traditional S1 pedicle and alar screws engage primarily cancellous bone, making them prone to failure, particularly in osteoporotic patients, and contributing to complications such as L5-S1 pseudoarthrosis. Given the junctional biomechanics and compromised bone quality in such cases, S2AI screws offer enhanced fixation strength. This study evaluates the feasibility, accuracy, and outcomes of fluoroscopy-guided S2AI screw placement in a clinical setting. Materials and Methods: A prospective analysis was conducted on 16 patients (total of 32 S2AI screws) who underwent fluoroscopy-assisted S2AI screw placement between 2020 and 2025. A non-radiolucent operating table was used, with intraoperative C-arm imaging utilizing both lateral and pelvic outlet-obturator views. To improve visualization, patients were manually shifted away from the table’s side bar, enhancing access to the hemipelvis targeted for screw insertion. Postoperative CT scans were performed to assess screw trajectory accuracy and potential breach. Results: The cohort had a mean age of 46 years (range: 23-57 years). The most common indication was adult degenerative scoliosis and lumbar spondylosis with endplate degeneration (7/16, 44%) followed by tuberculous spondylodiscitis at the lumbar/lumbosacral junction (4/16, 25%). Mean screw insertion time was 22 minutes (range: 15-30 minutes), with an overall mean operative duration of 4 hours (range: 3-6 hours). One intraoperative dural tear occurred (1/16, 6.2%). Postoperative CT revealed a mean transverse plane angulation of 35° anteriorly (range: 25°-38°), and a mean caudal angulation of 28° in the coronal plane (range: 20°-34°). The average distance from the screw to the anterior sacral cortex was 3 mm (range: 1-4 mm). No cortical breaches or neurovascular complications were observed. No wound-related complications were reported at 3-month follow-up. Conclusion: S2AI screw placement using standard fluoroscopic guidance is a reliable and accurate technique for lumbopelvic fixation, even in osteoporotic bone. The use of lateral and pelvic outlet-obturator views provides sufficient intraoperative visualization. Adjusting patient positioning to optimize imaging on non-radiolucent tables significantly enhances screw placement accuracy. This method offers a practical solution for centers without advanced navigation systems.

ID: AGS26/1214

A371: Comparison of muscle fat infiltration index of lumbar paraspinal muscles in people with chronic low back pain and healthy individuals

Dipak Shrestha 1 , Jagadish Thapa 1 , Bikash Parajuli 1

1 Dhulikhel Hospital, Kathmandu University Hospital, Dhulikhel, Nepal

Introduction: Low back pain is one of the common cause of significant morbidity. Among the various causes of low back pain, fatty infiltrations of para spinal muscles have also been attributed for the cause of back pain. However, consensus is lacking for clear correlation between fatty infiltration of lumbar para spinal muscle and chronic low back pain. The present study compares lumbar paraspinal muscle fat infiltration in MRI in patients with chronic low back pain with healthy volunteers. Material and Methods: Standardized MRI examination of patients with persistent low back pain for at least for three months and healthy volunteers were evaluated for fatty infiltration of Multifidus and Erector spinae muscle at the level of L3/L4, L4/L5 and L5/S1 using and Image J software to calculate muscle fat infiltration index (MFI) which was defined as a ratio of relative cross-sectional area (rCSA) pixel intensity of the muscle and extra muscular fat pixel intensity. Relative muscle cross-sectional area (rmCSA) was calculated in T1-weighted MRI image by using formula rmCSA = (1 - MFI) *rCSA. Right and left side values were averaged for each of the independent measurements for the analysis. Correlation of MFI was also calculated for numeric pain scale measured by Visual Analogue Scale (VAS), Oswestry disability index (ODI) and body mass index (BMI). Results: Of the 190 participants, 97 (Male: 53; mean age 41.09 ± 10 yrs, Female: 44; mean age 39 ± 11.47 yrs) cases were included in the chronic low back pain group and 93 (Male: 65; mean age 40.42 ± 8.46 yrs, Female: 28; mean age 31.61 ± 8.86 yrs) individuals were in the healthy volunteers group. MFI of multifidus (0.14 ± 0.07 vs 0.12 ± 0.05) and erector spinae (0.13 ± 0.66 vs 0.11 ± 0.04) at the level of L3/L4 were significantly different (p = 0.01, 0.03 respectively) between two groups but not significant at L4/L5 and L5/S1 level. There was no correlation between VAS pain score and muscle fatty infiltration index and cross-section area of the paraspinal muscles at all three levels except MFI of multifidus at the L3 and L4 levels (p = 0.005, 0.007). At the L3 level, both MFI and rmCSA of multifidus and erector spinae were correlating with ODI. At the L4 level, MFI and rmCSA of ES only had correlation with ODI. Whereas at the L5 level there was no correlation between the muscle parameters and ODI. With an increase in BMI, there was an increase in the muscle fatty infiltration index of the multifidus at L3 level and the erector spinae at L3, L4 and L5 levels. Conclusion: Muscle fat infiltration index of multifidus and erector spinae at the L3 and L4 levels, but solely of multifidus at the L5 level, was substantially different between patients with chronic low back pain and healthy volunteers. The cross-section area and the muscle fat infiltration index of the erector spinae at both L3 and L4 and the cross-section area of the multifidus at only the L3 level strongly predicted disability from chronic back pain.

ID: AGS26/1251

A372: Redefining the learning curve in UBE-XTLIF: why endplate preparation matters most

Young San Ko 1 , Dae-Chul Cho 1 , Yong Jin Park 2 , Subin Kweon 3

1 Kyungpook National University, School of Medicine, Kyungpook National University Hospital, Daegu, South Korea

2 Suncheon Hana Hospital, Suncheon, South Korea

3 Armed Forces Daegu Hospital, Daegu, South Korea

Introduction: Although numerous studies have investigated the learning curve of unilateral biportal endoscopic transforaminal lumbar interbody fusion (UBE-TLIF), the operative phase that most strongly determines proficiency remains uncertain. For UBE-assisted TLIF utilizing a large XLIF-type cage (UBE-XTLIF), no phase-specific learning curve analysis has been reported. This study aimed to identify the procedural step that drives the learning curve for UBE-XTLIF and to establish a pragmatic threshold for surgical proficiency. Materials and Methods: We retrospectively analyzed 62 consecutive UBE-XTLIF procedures performed via a posterolateral approach by a single surgeon. Operative times were prospectively recorded for two key steps: (1) decompression and (2) endplate preparation with cage insertion. Complications and patient-reported outcomes (VAS and ODI) were collected. The learning curve was assessed using cumulative sum (CUSUM) analysis of endoscopic time, with the elbow point demarcating Early (cases 1-30) and Late (cases 31-62) phases. Between-phase comparisons were conducted using standard statistical tests. A two-way repeated-measures ANOVA evaluated the Step × Phase interaction, and Pearson correlation coefficients examined associations between step times and total operative time. Results: The CUSUM curve demonstrated an inflection point near 30 cases, after which operative times stabilized, and adverse events declined. Efficiency gains were not uniform across steps: the endplate preparation with cage insertion phase demonstrated the greatest reduction in time and a significant Step × Phase interaction, indicating that proficiency was primarily achieved during this stage (p < 0.001). This step also showed a stronger correlation with total operative time compared with decompression (r = 0.95, p < 0.001). The overall complication rate was 11.3% (7/62). Most complications clustered in the Early phase and were primarily implant-related, including cage rotation failure, vertebral endplate violation, and nerve root injury during cage insertion.VAS and ODI improved from baseline with similar trajectories across phases. Conclusion: The learning curve for UBE-XTLIF exhibits a distinct inflection point at approximately 30 cases. Endplate preparation with cage insertion is the primary driver of operative efficiency and represents a practical surrogate marker of procedural proficiency. Focused cadaveric training emphasizing standardized contralateral endplate preparation, preservation of the subchondral plate, and controlled cage manipulation may shorten the learning curve and reduce implant-related complications.

ID: AGS26/2279

A373: Global practice patterns in the surgical management of primary lumbar disc herniation: an AO Spine Knowledge Forum Degenerative Global Cross-Sectional Study

Luca Ambrosio 1 , Sathish Muthu 2 , Patrick Hsieh 3 , S. Tim Yoon 4 , Jeffrey C. Wang 3 , Gianluca Vadalá 1 , Hans-Jörg Meisel 5 , Stipe Corluka 6 , Zorica Buser 7 , AO Spine Knowledge Forum Degenerative 8

1 Fondazione Policlinico Campus Bio-Medico, Rome, Italy

2 Government Medical College, Karur, India

3 University of Southern California, Los Angeles, United States

4 Emory University, Atlanta, United States

5 BG Klinikum Bergmannstrost Halle, Halle, Germany

6 University Hospital Centre Sestre Milosrdnice, Zagreb, Croatia

7 NYU Grossman School of Medicine, New York, United States

8 AO Spine, Davos, Switzerland

Introduction: Lumbar disc herniation (LDH) is among the most prevalent spinal disorders encountered in clinical practice, frequently resulting in significant morbidity and imposing substantial socioeconomic burdens worldwide. Despite considerable advancements in diagnostic imaging and surgical interventions over recent decades, no universal consensus remains on the optimal management strategy for LDH. This ongoing debate is reflective of the diverse array of clinical presentations and disability leading to different treatment modalities offered to patients, from conservative approaches, such as physical therapy and pharmacological management, to more invasive options, including various surgical techniques. The heterogeneity in treatment approaches can be attributed to multiple factors, including differences in surgeon training, regional medical practices, and, primarily, the variable natural history of the disease This study aimed to characterize global practice patterns in the surgical management of primary LDH among AO Spine surgeons. Material and Methods: The AO Spine Knowledge Forum Degenerative developed an online questionnaire to assess the global practice patterns related to the treatment of LDH. The survey was distributed in September 2024 to AO Spine members. The questionnaire collected detailed information on demographic characteristics, surgical indications, preferred techniques, and approaches for primary LDH treatment. Data on specialty, practice setting, fellowship training, and surgical case volume were analyzed using univariate and multivariate logistic regression to identify significant associations between surgeon characteristics and treatment preferences. Results: A total of 714 surgeons participated, representing diverse regions: North America (9.0%), Latin America (18.7%), Europe & Southern Africa (34.7%), Middle East & Northern Africa (12.8%), and Asia Pacific (24.8%). Neurological status was the most critical factor influencing early operative treatment in nonurgent cases, while bladder/bowel dysfunction and severe motor deficits were the primary indications in urgent scenarios. The majority of respondents (54.2%) preferred a mini-open technique-using either a surgical microscope or loupes-with partial laminotomy (58.1%) and partial discectomy (63.2%) being the most frequently performed procedures. Regional variations and differences in surgeon training were significantly associated with the choice of surgical approach and overall case volume. Conclusion: The findings reveal substantial variability in the surgical management of primary LDH across regions and specialties. This comprehensive dataset underscores the need for standardized, evidence-based guidelines to harmonize treatment strategies and optimize patient outcomes.

ID: AGS26/2331

A374: Interbody fusion techniques for L5-S1 degenerative spondylolisthesis: a comparison of radiographic correction and postoperative complications

Gabriel Gonzalez 1 , Manjot Singh 1 , Rhea Rasquinha 1 , Nicolas Carayannopoulos 1 , Zvipo Chisango 1 , Michael Farias 1 , Joseph Nassar 1 , John Czerwein 2 , Eren Kuris 2 , Bryce Basques 2 , Bassel Diebo 2 , Alan Daniels 2

1 The Warren Alpert Medical School of Brown University, Providence, United States

2 Brown University, Providence, United States

Introduction: Degenerative lumbar spondylolisthesis is a common spinal pathology, especially in the caudal levels, that can contribute to severe pain and functional limitations. Surgical interventions are frequently pursued to decompression the foramen, stabilize the vertebral bodies, and restore sagittal alignment. However, a comparison of the available interbody techniques to manage L5-S1 degenerative spondylolisthesis is lacking. Material and Methods: Adults who underwent one-level interbody fusion for L5-S1 degenerative spondylolisthesis were included in this retrospective cohort study. Patients were stratified by surgical technique: transforaminal lumbar interbody fusion (TLIF), standalone anterior lumbar interbody fusion (ALIF), and combined ALIF + posterior lumbar fusion (ALIF+PLF). Demographics, six-week radiographic spinopelvic alignment, and two-year postoperative complications were compared using univariate and multivariate regression modeling adjusting for age, sex, body mass index (BMI), Charlson Comorbidity Index (CCI), history of spinal fusion, practicing surgeon, surgeon experience, and preoperative L5-S1 slip. Results: Among 123 included patients, mean age was 55.1 years, 57.7% were female, mean BMI was 30.3kg/m2, and mean CCI was 1.9. After stratification by procedure, mean age (TLIF = 59.2, ALIF = 48.0, ALIF+PLF = 54.3years, p = 0.006), BMI (32.3, 28.9, 28.8, p = 0.017), and CCI (2.3, 0.8, 1.9, p = 0.002) differed across cohorts. Radiographically, preoperative L5-S1 slippage (12.4%, 8.4%, 19.8%, p = 0.007) and six-week postoperative ΔPI-LL (0.9°, 3.9°, -4.7°, p = 0.002), ΔL5-S1 lordosis (2.9°, 6.2°, 9.1°, p < 0.001), and ΔL5-S1 slippage (-8.7%, -12.5%, -22.2%, p < 0.001) correction differed across cohorts. Multivariate linear regressions revealed that ALIF patients had comparable correction but ALIF+PLF patients had higher six-week postoperative ΔPI-LL (coeff = -5.8°, p = 0.006), ΔL5-S1 lordosis (coeff = 4.9°, p = 0.001), and ΔL5-S1 slippage (coeff = -7.9%, p < 0.001) correction with respect to TLIF patients. Two-year postoperatively, reoperation rates (1.9%, 16.7%, 2.1%, p = 0.011) differed across cohorts. Multivariate logistic regressions revealed that ALIF patients had higher odds of pseudoarthrosis/subsidence (OR = 14.2, p = 0.032) and reoperations (OR = 15.7, p = 0.50) but ALIF+PLF patients had comparable odds of complications with respect to TLIF patients. Conclusion: In this investigation, ALIF+PLF was shown to be the most effective at correcting spondylolisthesis slip magnitude and segmental alignment. Standalone ALIF and TLIF, on the other hand, demonstrated comparable correction, with standalone ALIF patients experiencing higher rates of pseudoarthrosis/subsidence and reoperations. Combined anterior-posterior support appears beneficial to achieve optimal radiographic and clinical outcomes in patients with L5-S1 degenerative spondylolisthesis.

ID: AGS26/3391

A375: Does adding direct decompression impact indirect decompression in one to two level lumbar fusion?

Rowen Lin 1 , Kevin Mo 2 , Justin Chung 1 , William Fang 2 , Shanthan Challa 2 , Daniel Lee 2 , Hugh Bassewitz 2

1 Touro University Nevada, College of Medicine, Henderson, United States

2 Valley Hospital Medical Center, Department of Orthopaedic Surgery, Las Vegas, United States

Introduction: In one to two-level lumbar fusion, indirect decompression increases the height of intervertebral discs and can relieve neural compression. While direct decompression can further enlarge canal dimensions, it can also increase operative time and increase blood loss. The aim of this study was to determine whether adding direct decompression alongside an initial indirect decompression would impact outcomes following lumbar spinal fusion. Material and Methods: Adult patients undergoing one to two-level indirect decompression were identified from a single surgeon database from 2018-2022. Patients were then sub-stratified into two groups: 1) Patients who received only indirect decompression, 2) patients who received both a direct and indirect decompression. All patients received an indirect decompression through interbody realignment. Pre-operative, intraoperative, perioperative, 6 weeks, and 6 months post-operative data were gathered. Bivariate and multivariable analysis was conducted using STATA 17.0 College Station, TX. Penalized logistic regression was utilized when variables had perfect fit with direct decompression. Results: 125 adult patients were identified who underwent a one to two-level robotic-assisted lumbar fusion. 14 patients received a direct plus indirect decompression. The average age of all patients was 54 (SD 13), 60 were female (478.0%), 42 were smokers (33.6%), and 73 were opiate users (58.4%). Bivariate analysis only showed statistical significance with increased operative time (246.972 ± 65.4 mL vs 292.5 ± 73.5 mL; p = 0.017). However, there was no significance between both cohorts in respect to hospital length of stay (3.12 ± 1.42 days vs 3.42 ± 1.74 days; p = 0.47), post-operative day 1 mobility (168.1 ± 200.5 m vs 87.1 ± 125.2 m; p = 0.14), or VAS scores at 6 weeks (5.7 ± 3.2 vs 6.43 ± 3.8; p = 0.45). On multivariable analysis, when controlling for age, laminectomy, workers' compensation status, preoperative smoking status, robotic assisted, and intercept, direct decompression was not a statistically significant predictor of 6 month satisfaction (OR: 2.44 [-0.48-5.36; p = 0.1021]. Conclusion: While direct decompression added to indirect decompression prolonged operative time, it did not have a statistically significant association with intra-operative blood loss, length of stay, early ambulation, or patient-reported pain at 6 weeks and 6 months. Multivariable regression also showed direct decompression had no statistical impact on patient satisfaction at 6 months. These findings suggest that a direct decompression may be unnecessary when performing indirect decompression to achieve similar outcomes. Further multi-institutional and prospective studies should be performed to confirm these findings.

ID: AGS26/2829

A376: Preoperative geriatric nutritional risk index is useful factor for predicting postoperative delirium among elderly patients with degenerative lumbar diseases

Qian Chen 1 , Youwei Ai 2 , Hong Ding 2 , Limin Liu 2

1 Affiliated Hospital of North Sichuan Medical College, Nanchong, China

2 West China Hospital, Sichuan University, Chengdu, China

Introduction: Post-operative delirium (POD) complicates up to one-third of lumbar fusions in older adults, prolonging stay, increasing morbidity and mortality. Malnutrition is an emerging, modifiable risk factor, and the Geriatric Nutritional Risk Index (GNRI)-a simple score derived from serum albumin, weight and height-predicts adverse outcomes in elderly surgical patients. While GNRI correlates inversely with POD after cardiac surgery, its utility after lumbar fusion remains untested; we therefore evaluated whether pre-operative GNRI reliably forecasts POD in geriatric patients undergoing lumbar fusion. Material and Methods: A retrospective study was conducted to assess the outcomes of TLIF surgery in elderly patients with lumbar degenerative disease between the years 2016 and 2022. Delirium was diagnosed by reviewing postoperative medical records during hospitalization, utilizing the Confusion Assessment Method. The geriatric nutritional risk index was calculated using the baseline serum albumin level and body weight. Multivariate logistic regression analysis was employed to identify the association between preoperative GNRI and postoperative delirium (POD). Additionally, a receiver operating characteristic curve was utilized to determine the optimal GNRI cutoff for predicting POD. Results: POD was observed in 50 of the 324 patients. The GNRI was visibly reduced in the delirium group. The mean GNRI was 93.0 ± 9.1 in non-delirium group and 101.2 ± 8.2 in delirium group. On multivariate logistic regression, Risk of POD increases significantly with low GNRI and was an independent factor in predicting POD following TLIF (OR 0.714; 95% CI 0.540-0.944; p = 0.018). On receiver operating characteristic curve, the area under curve (AUC) for GNRI was 0.738 (95% CI 0.660-0.817). The cutoff value for GNRI according to the Youden index was 96.370 (sensitivity: 66.0%, specificity: 70.4%). Conclusion: Our study indicated that lower GNRI correlated significantly with POD after TLIF. Performing GNRI evaluation prior to TLIF may be an effective approach of predicting the risk for POD among elderly patients with degenerative lumbar diseases.

ID: AGS26/877

A377: Static vs. expandable cages in posterior lumbar interbody fusion: effects on spinopelvic alignment and adverse events

Enrico Giordan 1 , Nicola Bresolin 1 , Alessandro Trento 1 , Jacopo Del Verme 1 , Roberto Zanata 1

1 AULSS 2 Marca Trevigiana, Treviso, Italy

Introduction: Posterior lumbar interbody fusion techniques (PLIF and TLIF) have a known limitation in their relatively modest ability to restore and maintain physiological lumbar lordosis compared with anterolateral approaches (ALIF). In recent years, expandable cages have been increasingly adopted as an alternative. However, evidence regarding their effects on spinopelvic parameters, sagittal balance, and particularly lumbar lordosis restoration remains limited. The present study aims to compare changes in spinopelvic parameters in patients undergoing transforaminal lumbar interbody fusion (TLIF / PLIF) using oblique static lordotic cages versus expandable cages. Specifically, we investigate whether expandable cages promote greater overall and single-level improvements in lumbar lordosis and spinal alignment, and whether their safety profile and incidence of adverse events are comparable to those of static cages. Material and Methods: We conducted a retrospective study on adult patients who underwent posterior lumbar interbody fusion at one or more lumbar or lumbosacral levels between 2015 and 2024, comparing outcomes between expandable cages and oblique static lordotic cages. Patient data collected included age, sex, bone quality, degree of disc degeneration (Pfirrmann classification), and disc height before and after the procedure. In both cohorts, several radiological parameters were analyzed, including spinopelvic alignment before and after surgery, as well as the occurrence of mechanical adverse events or revisions in the postoperative period (such as cage migration or subsidence, adjacent segment disease, pseudoarthrosis). Patients who met the inclusion criteria were divided into two groups: those who underwent posterolateral fusion with static cages (S group) and those who underwent fusion with expandable cages (E group). The mean differences in spinopelvic parameters (i.e., lumbar lordosis, pelvic tilt, pelvic incidence, and pelvic mismatch) were compared between different groups and within groups by categorizing the analysis by surgical approach and fusion level. Analyses were adjusted for age, and sex. Results: Of 257 reviewed patients, 177 met the inclusion criteria (129 static vs. 48 expandable cages). Median age was 56 years (IQR 48-66), with a female predominance (54.8%). The most common fusion levels were L4-L5 (46.0%) and L5-S1 (36.7%); 36 patients underwent multilevel fusion. At last follow-up, revision was required in 5.2% of cases (p = 0.704). Subsidence occurred in one expandable cage (0.48% overall, p = 0.585), with no loosening recorded. Adjacent segment disease occurred in 1.9% (p = 0.918). There were no significant differences between groups. Overall, the mean difference in PT values was 1.6° ± 0.9 between static and expandable cages (p-value: 0.0621), PI 0.8° ± 1.21° (p-value: 0.694), SS 1.7° ± 0.8° (p-value: 0.036), LL 0.37 ± 0.92 (p-value: 0.689) and PI-LL 0.9° ± 0.8°. After stratifying each group by level of fusion we found that the mean difference in PT values in L4-L5 was 2.7° ± 1.1 between static and expandable cages (p-value: 0.187), PI 3.7° ± 1.6° (p-value: 0.023), SS 2.2° ± 1.21° (p-value: 0.072), LL 1.34 ± 1.58 (p-value: 0.339) and PI-LL 0.5° ± 0.81° (p-value: 0.541). In the L5-S1 subgroup we found the mean difference in PT values for was 1.7° ± 1.4 between static and expandable cages (p-value: 0.222), PI 0.1° ± 2.14° (p-value: 0.964), SS 1.7° ± 1.3° (p-value: 0.183), LL 0.76 ± 1.38 (p-value: 0.577) and PI-LL 1.70° ± 1.7° (p-value: 0.313). We observed a significant difference in the distribution of disc degeneration grades, with the expandable cage group showing a significantly lower Pfirrman grade (p-value: 0.001) and a notably higher ability to restore disc height (p-value: 0.001). Conclusion: Our analysis indicates that well-positioned static and expandable cages show no significant differences in their ability to restore lordosis or influence spinopelvic parameters. However, expandable cages offer advantages in surgical access and demonstrate superior restoration of disc height, and can cope with a higher degree of degeneration. These features make them particularly effective for managing more advanced discopathies through a posterior approach.

ID: AGS26/898

A378: Multidimensional burden of low back pain: a prospective cross sectional study of patient-reported outcomes and sociodemographic factors at a tertiary neurosurgical center

Pavlina Lenga 1,2 , Sebastian Ille 1 , Martin Dugas 2 , Sandro Krieg 1

1 Department of Neurosurgery, Heidelberg University Hospital, Heidelberg, Germany

2 Institute of Medical Informatics, Heidelberg, Germany

Introduction: Low back pain (LBP) significantly impacts patients’ physical function, mental health, sexual wellbeing, and occupational status. However, comprehensive prospective assessments of these interrelated dimensions in specialized neurosurgical outpatient settings remain sparse. This study aimed to prospectively evaluate functional disability, depression, erectile dysfunction (ED), occupational impairment, and their associations with demographic and lifestyle factors among patients presenting to a neurosurgical outpatient clinic. Material and Methods: Between February and June 2025, we prospectively enrolled 247 consecutive patients (mean age 60.4 ± 16.0 years, 53.8% male) presenting to the neurosurgical outpatient clinic at Heidelberg University Hospital with degenerative lumbar spine disorders, of whom 110 (44.5%) subsequently underwent decompression surgery. Patients completed validated patient-reported outcome measures (PROMs), including pain intensity (Numeric Rating Scale [NRS]), functional disability (Oswestry Disability Index [ODI]), depression (PHQ-9), and erectile dysfunction (IIEF-5). Correlation analyses and regression models assessed associations among these outcomes and relevant demographic and lifestyle variables. Results: Patients reported high pain intensity (NRS 7.0 ± 2.2), severe functional disability (ODI 57.4 ± 16.5; 82.2% severely disabled or worse), frequent moderate-to-severe depressive symptoms (42.2%), and prevalent erectile dysfunction among males (55%). Multivariate analyses identified depression severity (PHQ-9; B = 1.30, p < 0.001), older age (B = 0.20, p = 0.045), and surgery (B = 6.13, p = 0.049) as significant predictors of greater disability. Erectile dysfunction severity in males was independently associated with higher disability (B = -0.18, p = 0.002) and older age (B = -0.32, p < 0.001). Undergoing surgery was significantly predicted by higher baseline disability (ODI; OR = 1.04, p = 0.032) and private insurance (OR = 22.39, p < 0.001). Among working-age patients (≤ 65 years, n = 142), occupational disability was notably high, with 39.4% currently work-disabled and an additional 15.5% having experienced disability within the past year. Conclusion: This prospective study highlights the significant multidimensional burden of low back pain among neurosurgical outpatients, characterized by substantial functional disability, high prevalence of depression, frequent erectile dysfunction, and marked occupational impairment. Our results underscore the critical need for routine psychological screening, proactive sexual health assessment, and targeted occupational rehabilitation within specialized neurosurgical care pathways. Implementing these comprehensive approaches may improve patient outcomes, support return to work, and mitigate potential socioeconomic impacts associated with degenerative lumbar spine disorders.

OP43: AIS Navigator

ID: AGS26/1084

A379: Patient counselling with educational videos may reduce postoperative pain scores in AIS patients

Yağız Akcan 1 , Ahmet Özaslan 2 , Asim Ahmadov 2 , Aliekber Yapar 3 , Alim Baymurat 2 , Alpaslan Senkoylu 2

1 Çorum Erol Olçok Training and Research Hospital, Çorum, Turkey

2 Gazi University, Ankara, Turkey

3 Antalya Training and Research Hospital, Antalya, Turkey

Introduction: Due to the increasing use of smartphones and social media today, adolescents experience a rapid loss of attention while reading or listening while they tend to focus on audiovisual resources more efficiently. Aim of the study is to evaluate the impact of educational videos, educational brochures, and routine verbal counseling on preoperative anxiety and postoperative outcome. Material and Methods: Participants were randomly assigned into three groups: educational video (n = 14), educational brochure (n = 14), and verbal counseling (n = 14). Patients were given the SRS-22 questionnaire, the Spinal Appearance Questionnaire (SAQ), the State-Trait Anxiety Index, the Rosenberg Self-Esteem Scale, and the Pain Catastrophizing Scale three days before the surgery. Patients were informed about the surgical process according to their assigned group. At the same day before the surgery, the STAI was administered to assess the patients' situational anxiety levels. During postoperative follow-up, the same scales and questionnaires were administered at 3-month follow-up. Results: Among the cohort (n = 42), those who received educational videos or brochures during preoperative counseling had significantly less reduction in their SRS-22 pain scores at the third postoperative month compared to the verbally informed group (p = 0.03). The group that received video education showed a significantly higher increase in morning STAI scores compared to the verbally informed group (p < 0.001), indicating heightened anxiety. No significant difference was observed between brochure group and control group (p = 0.118). Age was not found to have a significant effect on changes in STAI scores (p = 0.558). These findings suggest that video education may increase anxiety by raising morning STAI scores. Conclusion: Preoperative anxiety is a common response experienced by AIS patients undergoing surgical intervention, with one of its primary causes being fear of the unknown. It has been demonstrated that sharing these unknown elements with patients through videos may increase anxiety; however, it has also been shown to have a positive effect on pain management and the recovery process in the postoperative period.

ID: AGS26/1280

A380: Does preoperative and postoperative thoracic kyphosis affect the cervical and cervicothoracic alignment in adolescent idiopathic scoliosis?

Ajoy P Shetty 1 , Juan Esteban Muñoz 1 , Karthik Ramachandran 1 , S. Rajasekaran 1

1 Ganga Medical Centre and Hospital, Coimbatore, India

Introduction: Pelvic incidence (PI) is a key morphological parameter governing sagittal spinal alignment and serves as the foundation for the theoretical Roussouly classification in adults. However, its applicability in pediatric populations remains uncertain due to growth-related variability. This study aimed to evaluate PI stability across developmental stages and assess its utility in classifying sagittal profiles in adolescents with idiopathic scoliosis (AIS). Material and Methods: A retrospective cross-sectional analysis was conducted on AIS patients aged 12-17 years. Whole-spine radiographs were obtained at three time points: preoperatively, 6 months postoperatively, and 2 years postoperatively. Global, regional, and pelvic sagittal parameters were measured. PI variability was assessed using repeated measures ANOVA, with post hoc analysis via Tukey’s HSD test to evaluate statistical significance. Results: A total of 100 AIS patients were included, predominantly female (66%). Curve types 5 (34%) and 3 (24%) were most prevalent. Across all time points, Roussouly type 3 was the predominant sagittal profile. Mean PI values were 48.6° ± 11.6° preoperatively, 48.4° ± 11.2° at 6 months, and 49.7° ± 10.8° at 2 years. ANOVA revealed no statistically significant changes in PI across the three intervals (p = 0.69), confirming its relative stability during adolescence. Conclusion: Pelvic incidence remains stable across preoperative, postoperative, and follow-up periods in AIS patients aged 12-17, supporting the use of Roussouly classification in this group. Further research is needed to assess its role in surgical planning and its potential complementarity with the Lenke system.

ID: AGS26/1525

A381: Clinical and radiological outcomes of hybrid technique in the management of AIS with double major curves over a minimum 5 years follow up

Burak Abay 1 , Sepehr Asadollahmonfared 1 , Mehmet Zamanoglu 1 , Hamisi Mraja 1 , Sinan Kahraman 2 , Meric Enercan 2 , Selhan Karadereler 1 , Azmi Hamzaoglu 1

1 Scoliosis and Spine Center - Istanbul, Istanbul, Turkey

2 Department of Orthopedics and Traumatology, Demiroglu Bilim University, Istanbul, Turkey

Hypothesis: Hybrid technique; Combination of posterior pedicle screw fixation-fusion for thoracic curve and VBT with Double Screw Double Cord (DS-DC) fixation for TL/L will restore physiologic thoracic kyphosis and preoperative cervical sagittal malignment and will preserve spinal flexibility and lumbar motion in AIS pts with double major curves (DMC). DS-DC fixation will minimize the tether breakage (TB) problem. Introduction: Hybrid technique has become a viable option for the surgical treatment of AIS with double major curves. In this study, we aimed to evaluate the efficacy of the Hybrid Technique in restoring ideal sagittal and coronal alignment in pts with DMC and assess tether integrity over 5 yrs f/up period. Methods: AIS pts treated with HT with a minimum 5 yrs f/up were included. Coronal and sagittal parameters were measured on preop, first erect and f/up x-rays. > 5° interscrew angle change is accepted as tether breakage (TB) for VBT. SRS22r was used for clinical assessment. Results: 45 (34F,11M) AIS pts with mean age 14 (11-17) yrs and f/up was 69 (62-88) m. Mean MT 49° was corrected to 9° at f/up (84%). Mean TL/L of 51° was corrected to 15° at f/up (84%). TK of 25° restored to 9°. 30 (66%) pts had cervical kyphotic alignment preoperatively. Among these 30 pts, 26 (87%) of them had improved cervical alignment postoperatively. Ponte osteotomy was performed in 6 pts with preoperative lordotic thoracic alignment. According to T10-L2 sagittal alignment 12 pts presented with TL kyphosis. T10-L2 kyphosis improved from 16° to 2°. Tether breakage (TB) occurred in 14 patients (29%) at a mean of 39 months (28-58), resulting in TL/L curve progression to 24° and loss of sagittal alignment; T10-L2 kyphosis declined to 5°; lumbar lordosis returned to preoperative levels. Patients with TB had larger preoperative TL/L curves (56° vs 49°) and higher lumbar lordosis (60° vs 53°) than non-TB patients. 2 pts (4%) required revision surgery due to TB at a mean of 35.3 months. Overcorrection occurred in 6 pts (13%) but did not require any additional surgery. SRS-22r total scores were similar between TB and non-TB groups (4.1 vs. 4.3), but satisfaction scores were lower in the TB group (3.8 vs. 4.2). Conclusion: In our study, 87% of the pts with preoperative cervical kyphosis had improved postop cervical alignment with Hybrid technique. Tether breakage rate was 29% and overcorrection rate was 13% over a minimum 5 yrs f/up. TB occurred at a mean of 3 yrs or later following surgery. SRS22r total scores did not show any difference but tether breakage lowers SRS satisfaction score (3.8 vs 4.2).

ID: AGS26/1534

A382: Selective vs. non-selective thoracolumbar/lumbar fusion in Lenke 6C AIS: impact on thoracic curve correction and clinical outcomes

Korawish Mekariya 1 , Thanase Ariyawatkul 1 , Monchai Ruangchainikom 1 , Sirichai Wilartratsami 1 , Ekkapoj Korwutthikulrangsri 1 , Chatupon Chotigavanichaya 1 , Kasidit Laorkanjanakul 1 , Surin Thanapipatsiri 1

1 Department of Orthopaedic Surgery, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok, Thailand

Introduction: Selective thoracolumbar/lumbar fusion (SLF) remains a challenge in patients with adolescent idiopathic scoliosis (AIS), particularly those with Lenke 6C curves, as both the thoracic and TL/L curves are structural. Limited studies have directly compared the outcomes of SLF versus extended fusion to include the thoracic curve. This study aims to evaluate the clinical and radiographic outcomes of SLF in patients with Lenke 6C AIS compared with non-selective lumbar fusion. Materials and Methods: This retrospective analysis utilized a prospectively collected database. Patients with Lenke 6C curves who underwent posterior spinal fusion and had at least 2 years of follow-up were included. Radiographic parameters were assessed preoperatively, early postoperatively, and at the final follow-up. SRS-22 scores were also collected and compared between the SLF and non-SLF groups. Results: A total of 48 patients were included (SLF: 23, non-SLF: 25), with a mean age of 14.6 years and an average follow-up of 6.4 years. Demographic and preoperative radiographic parameters were comparable between groups, except for a lower preoperative thoracic Cobb angle (42.2° vs. 51.0°, p = 0.024) and greater thoracic curve flexibility (24.6% vs. 15.1%, p = 0.019) in the SLF group. At final follow-up, thoracic curve correction was significantly lower in the SLF group (45.0% vs. 61.3%, p = 0.019). However, the residual mean thoracic Cobb angle was not significantly different (22.9° vs. 19.8°, p = 0.118). No significant differences were observed in final lumbar Cobb angle, coronal balance, shoulder height, T1 tilt angle, or sagittal parameters, including thoracic kyphosis, between groups. Operative time and blood loss were significantly less in the SLF group (both p < 0.001). SRS-22 scores at the final follow-up were comparable between groups across all domains. Conclusion: Selective thoracolumbar/lumbar fusion is a viable treatment option in carefully selected Lenke 6C patients, providing acceptable spontaneous thoracic curve correction and comparable functional outcomes while minimizing perioperative morbidity and preserving motion segments.

ID: AGS26/1571

A383: Radiation exposure in posterior fusion for adolescent idiopathic scoliosis: a comparison between low-dose, 2D fluoroscopy, navigation on preoperative CT scan and navigation on intraoperative CT scan

Georgios Bakaloudis 1 , Lorenzo D'Apice 1 , Alice Baroncini 1

1 Casa di Cura Humanitas San Pio X, Milano, Italy

Introduction: The limitation of radiation exposure for the surgical team and the patients is one of the key advantages of enabling technologies. However, currently available data that compares the radiation exposure with different enabling technologies is limited. The aim of this work was thus to evaluate the radiation exposure with three different technologies in the setting of posterior fusion for adolescent idiopathic deformity (AID). Material and Methods: Retrospective study on all consecutive cases operated by a single surgeon. The instrumentation was performed with one or more midline skin incisions and a paramedian Wiltse approach. Screw placement was guided with either low-dose 2D fluoroscopy - 2DF, robotic neuronavigation on preoperative CT scan - preCT, or navigation based on intraoperative CT scan - intraCT. The surgical time (skin-to-skin), the time required for screw placement and the radiation exposure were compared among the three techniques. Results: Between Sept 2016 and Dec 2024, 146 consecutive AID patients (101 female) underwent posterior fusion and were included in the study. The mean age was 15.6 ± 2.6 years. 2DF was used for 104 cases, preCT for 24 and intraCT for 18. There was no significant difference between the groups in term of number of instrumented levels, screw density or number of placed screws. The average time for screw placement was significantly lower in the 2DF group compared to preCT and intraCT (60 min vs 180 vs 140, respectively; r = 0.34 and 0.56, p < 0.0001). The surgical time was also lower in the 2DF group compared to preCT and intraCT (140 ± 20 min vs. 250 ± 25 vs. 210 ± 10, p = 0.0003). In the 2DF group, the radiation exposure was 132 ± 8 mGy with a mean radiation time of 5min; in the preCT group, the radiation exposure was 400 mGy for the preoperative scan plus 20-30 intraoperative X-ray shots for image matching; in the intraCT the mean dose was 80 ± 5Gy. None of the patients experienced major complications, no screw misplacement was recorded. Conclusion: Navigation based on an intraoperative CT scan is the enabling technology that can most effectively reduce radiation exposure in posterior fusion for ASD. However, at present, these technologies still require a longer surgical time and longer time for screw placement.

ID: AGS26/1729

A384: Preoperative planning with gap score and virtual simulation improves sagittal alignment in adolescent idiopathic scoliosis surgery

Giovanni Viroli 1 , Matteo Traversari 1 , Marco Manzetti 1 , Alberto Ruffilli 1 , Cesare Faldini 1

1 Istituto Ortopedico Rizzoli, Bologna, Italy

Introduction: Adolescent idiopathic scoliosis (AIS) surgery, despite continuous evolution, still faces challenges in restoring the sagittal profile. Extensive corrections in the coronal and axial planes do not always result in optimal sagittal balance, potentially impacting quality of life and complication rates. This study aimed to evaluate whether the integration of the Gap Score into preoperative planning, together with sagittal correction simulation using dedicated surgical planning software, improves radiographic outcomes in AIS patients. We hypothesized that precise planning based on objective parameters and virtual simulation may optimize surgical results in terms of sagittal alignment. Materials and Methods: A retrospective case-matched study was conducted on AIS patients undergoing corrective surgery. Fifty patients operated between 2018 and 2020 (Group 1) were compared with 50 patients treated between 2020 and 2022 (Group 2). Groups were matched according to curve pattern (Lenke classification), magnitude of the main curve, curve flexibility as measured on side-bending radiographs, and preoperative sagittal alignment parameters including thoracic kyphosis, lumbar lordosis, and pelvic parameters. This strict matching procedure ensured comparability between the two cohorts and minimized potential confounding variables. All patients underwent the same surgical technique consisting of high-density pedicle screw constructs, concave translation, and direct vertebral derotation maneuvers, performed by the same surgical team. In addition, Group 2 received a dedicated preoperative planning protocol: the surgical target was determined using the Gap Score, and the planned correction was virtually simulated with specialized software to predict postoperative sagittal alignment and optimize implant positioning. Patients were between 10 and 18 years of age at the time of surgery and had a minimum follow-up of two years. Standardized radiographic parameters (Cobb angle, thoracic kyphosis, cervical lordosis, sacral slope, and junctional angles) were assessed on standing anteroposterior and lateral radiographs both preoperatively and at the latest follow-up. Results: The main Cobb angle decreased from 64° to 18° in Group 1 and from 63° to 20° in Group 2. Thoracic kyphosis increased from 9° to 16° in Group 1 and from 10° to 26° in Group 2 (p < 0.05). Cervical alignment changed from 11° kyphosis to 3° kyphosis in Group 1, and from 10° kyphosis to -14° lordosis in Group 2 (p < 0.05). Sacral slope remained stable in Group 1 (44° to 43°), but significantly decreased in Group 2 (43° to 36°, p < 0.05). Three junctional complications (PJK, DJK, adding-on) were recorded in Group 1, while none occurred in Group 2 (p > 0.05). Conclusion: Preoperative planning strategies based on target values derived from the Gap Score, combined with surgical simulation using dedicated software, allow for improved restoration of physiological sagittal alignment in AIS patients. Significant improvements in thoracic kyphosis and reversal of compensatory mechanisms (cervical kyphosis, pelvic anteversion) were observed. This approach may also contribute to a reduction in junctional complications, although this did not reach statistical significance.

ID: AGS26/2612

A385: Outcomes of posterior spinal fusion in AIS: a minimally invasive vs. open technique analysis

Mario Ropelato 1 , Jacopo Vitale 1 , Anne Mannion 1 , Jani Puhakka 1 , Markus Loibl 1 , Tamas Fekete 1

1 Schulthess Clinic, Zurich, Switzerland

Introduction: The most popular surgical treatment method for adolescent idiopathic scoliosis (AIS) is posterior spinal fusion. Recent innovations in implant designs have facilitated the development of minimally invasive (MISS) approaches in spinal surgery. However, these techniques have not yet achieved widespread adoption in the treatment of spinal deformities due to various challenges. Although MISS techniques are associated with reduced tissue damage, they present greater intraoperative challenges due to limited visualization. Few previous studies have analysed surgical/patient outcomes in patients with AIS undergoing MISS; even fewer have compared MISS with the open approach. This study aimed to compare the results of AIS patients treated with MISS with those of a historical cohort treated using traditional open methods. Material and Methods: This was a single-center retrospective analysis of data collected prospectively within the EUROSPINE Spine Tango Registry framework. Inclusion criteria were as follows: idiopathic scoliosis; posterior spinal fusion as surgical treatment; major coronal Cobb < 80°; and patient age 10-30 years. A total of 71 patients operated on using an conventional approach (OPEN) between 2005 and 2015 were compared with 39 patients treated with a minimally invasive posterior transmuscular approach (MINI-OPEN) between 2016 and 2021. For both groups a free-hand pedicle screw insertion technique was utilized. Complications, blood loss, operation time, and length of stay were documented on Tango surgery forms. Preoperative and postoperative standing whole-spine X-rays were analyzed. Patients completed the Core Outcome Measures Index (COMI) preoperatively and at 3, 12, and 24 months postoperatively. Results: The preoperative main curve was 55.8 ± 8.5° in the MINI-OPEN group and 58.1 ± 8.8° in the OPEN group (p = 0.21). Blood loss was significantly lower in the MINI-OPEN group than in the OPEN group (439.1 ± 272.6mL vs 1184.5 ± 704.8mL, respectively; p < 0.0001). Conversely, operating time was significantly longer in the MINI-OPEN group (452 ± 104 minutes vs 328 ± 84 minutes in the OPEN group; p < 0.0001). Length of stay from surgery to discharge was almost half as long in the MINI-OPEN group as in the OPEN group (5.1 ± 1.3 days vs 9.3 vs 9.2 ± 2.2 days, respectively; p < 0.0001). The percentage of coronal correction of the main curve did not differ significantly between groups (75.7 ± 14.3% for MINI-OPEN vs 74.5 ± 10.5% for OPEN; p = 0.37). There was no significant difference in COMI score reduction between groups, and improvements were maintained for up to 2 years postoperatively. Conclusion: Compared with the conventional open approach, the minimally invasive approach using a freehand pedicle screw insertion technique was associated with a significantly lower blood loss and shorter hospital stay and similar patient rated outcomes. Refinements in surgical technique and reductions in surgical invasiveness can help minimize patient suffering immediately after surgery while maintaining similar midterm outcomes to conventional methods.

ID: AGS26/1610

A386: Balancing motion and correction: mid-term outcomes of hybrid fixation in adolescent idiopathic scoliosis

Feyzi Kilic 1 , Altug Yucekul 2 , Gokhan Ergene 3 , Sahin Senay 4 , Elif Gizem Carus 1 , Tais Zulemyan 1 , Caglar Yilgor 5 , Ahmet Alanay 2

1 Comprehensive Spine Center, Acibadem Maslak Hospital, Istanbul, Turkey, 2 Acibadem Mehmet Ali Aydinlar University School of Medicine, istanbul, Turkey, 3 Department of Operating Room Services, Acibadem Mehmet Ali Aydinlar University Vocational School of Health Sciences, Istanbul, Turkey, 4 Department of Cardiovascular Surgery, Acibadem Mehmet Ali Aydinlar University School of Medicine, Istanbul, Turkey, 5 Koc University Hospital Spine Center, Istanbul, Turkey

Introduction: Vertebral body tethering (VBT) provides growth-modulated correction for scoliosis while preserving segmental motion. In patients with double major curves or thoracic deformities accompanied by thoracolumbar/lumbar curves, selective thoracic fusion alone may not provide sufficient control. Extending fusion into the lumbar spine ensures correction but at the cost of motion loss. To balance correction and mobility, a hybrid approach-thoracic posterior fusion combined with thoracolumbar/lumbar VBT-has recently gained popularity as a motion-preserving alternative. However, comprehensive series reporting mid-term outcomes of this technique remain limited. This study reports radiographic, patient-reported, pulmonary, and implant-survival outcomes of a consecutive hybridx fixation cohort. Material and Methods: Prospectively maintained data were retrospectively analyzed at a single center. Outcomes included Cobb angles for the upper thoracic (UT), main thoracic (MT), and thoracolumbar/lumbar (TLL) curves at preoperative, first-erect, and latest follow-up. Patient-reported outcomes comprised SRS-22 Function, Pain, Self-image, and Mental Health domains from preoperative to latest follow-up. The minimal clinically important difference (MCID) was defined as 0.5× the preoperative SD. Pulmonary function tests (PFT) included percent-predicted FVC and FEV1. Complications, readmissions/reoperations, and time to first tether break were captured. Results: Twenty-five patients were analyzed (mean age 13.9 ± 2.5 years; 84% female). Preoperatively, median Sanders, CVM, and Risser were 5.5 [IQR 3-7], 4.0 [4-5], and 1.0 [0-3], respectively. Median follow-up was 25.3 months [24.2-29.2]. Lenke types were: Type 1: 43.5%, Type 2: 17.4%, Type 4: 4.3%, Type 5: 26.1%, and Type 6: 8.7%; lumbar modifiers: A: 4.3%, B: 26.1%, C: 69.6%. Preoperative flexibility was 48.8 ± 20.5% for UT, 56.8 ± 20.5% for MT, and 80.7 ± 18.9% for TLL. UT, MT, and TLL Cobb angles were 36.0 ± 11.1°, 61.8 ± 17.3°, and 50.9 ± 13.9° preoperatively; improved to 8.1 ± 5.0°, 7.8 ± 5.0°, and 12.8 ± 6.6° at first erect; and measured 8.8 ± 6.3°, 11.0 ± 8.7°, and 12.3 ± 8.3° at latest follow-up. SRS-22 self-image improved significantly (? +0.84, MCID achievement 67%, p. < 0.001), while Function (Δ -0.16), Pain (Δ -0.16), and Mental Health (Δ +0.02) remained unchanged (all p > 0.05). PFT percent-predicted changes were -4.25 for FVC% and +0.21 for FEV1% (p > 0.05). Two mechanical complication occurred (tether break in 8%); no readmissions or reoperations were recorded Conclusion: In this consecutive hybrid fixation cohort with median follow-up exceeding two years, UT, MT, and TLL Cobb angles improved substantially and were maintained. Patient self-image improved with preservation of pulmonary function. The mid-term implant profile was favorable, with low mechanical complication rates and survival without tether break >95% at one year and >80% at two years. These results highlight hybrid fixation as a promising motion-preserving alternative, warranting confirmation in larger cohorts with longer follow-up.

ID: AGS26/2882

A387: Assessment of changes in curve patterns in idiopathic scoliosis: a single-center longitudinal cohort study

Anastasios Charalampidis 1 , Kasim Abul-Kasim 2 , Acke Ohlin 3

1 Department of Clinical Science, Intervention and Technology (CLINTEC), Karolinska Institutet, Stockholm, Sweden

2 Neuroradiology section, Radiology Department Skåne University Hospital Lund-Malmö, Malmö, Sweden

3 Clinical and Molecular Osteoporosis Unit, Department of Clinical Sciences, Malmö, Lund University, Lund, Sweden, Malmö, Sweden

Introduction: Changes in curve patterns over time in adolescent idiopathic scoliosis (AIS) may reflect different progression types. While several classification systems exist to categorize curve patterns and guide surgical planning, their application to non-surgical curves remains limited. This study aimed to determine whether AIS curve types, as defined by the Lenke classification system, change over time and to identify demographic or radiographic factors associated with these changes. Material and Methods: We performed a retrospective review of 165 consecutive AIS patients (age 10-25 years at surgery). Baseline demographic and radiographic data were retrieved from an institutional database. The earliest available standing posteroanterior (PA) radiograph was classified using the modified Lenke (mLenke) system, while the last standing PA and bending radiographs prior to surgery were classified using the Lenke system. Curve magnitudes were measured using the Cobb method. Results: At baseline, the most common curve type was Lenke type 3 (34%), followed by type 1 (29%). Patients with types 1 and 5 had smaller curves and were younger than those with types 2, 3, 4, or 6. The mean interval from baseline to preoperative radiographic assessment was 41 (26) months. Preoperatively, Lenke type 1 predominated (51%), with type 3 being second most frequent (17%). Overall, 47% of patients experienced a change in curve type. The most frequent transitions were from type 3 (58%) and type 2 (36%) to type 1, and from type 4 (33%) to type 1 or 2. Age and Cobb angles at baseline did not differ significantly between patients with stable versus changing curve types. Conclusion: Less than half of AIS patients exhibit curve pattern changes over time, with thoracic-dominant curves remaining stable and lumbar-dominant curves stable in approximately 74% of cases.

OP44: Regenerating the Disc

ID: AGS26/1166

A388: Cartilaginous endplate and disc resorption failure: evidence from histological and immunohistochemical analysis

Samina Ashraf 1 , Huma Qamar 1 , Muhammad Saad Ilyas 2 , Amer Aziz 2 , Uruj Zehra 1

1 University of Health Sciences, Lahore, Pakistan

2 Ghurki Trust Teaching Hospital, Lahore, Pakistan

Introduction: Herniated discs are a common cause of low back and leg pain; spontaneous resorption often facilitate recovery under conservative management. This process is mediated by inflammatory responses and neovascularization driven by cytokines and matrix-degrading enzymes such as TNF-α, IL-1β, IL-6, MMPs, and ADAMTS. Resorption is favored in hydrated, less-degenerated nucleus pulposus, whereas cartilaginous endplate tissue may slow the resorption process. This study explores key inflammatory markers to better understand and predict disc resorption in the presence of CEP. Material and Methods: Lumbar intervertebral disc samples were collected from adult patients undergoing disc excision surgery for herniated discs. The relevant demographic and clinical data were also obtained in detail including age, sex, and duration of low back pain, body mass index (BMI) and history of smoking. Visual Analog Scale (VAS) scores were recorded by using professional numerical rating scales for evaluating back pain pre surgery. Paraffin embedded five-micron thick sections were stained with hematoxylin and eosin, Alcian blue periodic acid-Schiff stain; examined for histological degeneration scores (HDS; 0-15), inflammation (0 [absence]-3 [severe]), and presence of cartilage endplate (CEP). Three-micron thick sections were immunostained with anti TNFα antibody and IL-1β expression. Disc samples with and without CEP were compared for the expression of these immunostains. Results: Total of 109 patients (69M/40F) with average age & BMI of 39.4 ± 9.9 years & 25.3 ± 5.1kg/m 2 were included in the study after informed consent and ethical approval of the institution. Spinal levels ranged from L2 to L5 and most (52%) of the patients had L4 disc herniation. Histological degeneration scores ranged from 4-13, with a mean ± SD of 9.0 ± 2.0. Age, gender, spinal level and VAS did not show any relation with histological degeneration scores. Presence of inflammatory infiltrates was found in almost 73.5 % of the samples while proteoglycan loss was found in 95% of samples. CEP was detected in 69.7% samples, age, sex, BMI, spinal level and VAS score did not show any significant relation with the presence and absence of cartilage endplate. Samples with cartilage endplate showed increased mucous degeneration (p = 0.001), granular changes (p = 0.02) and total histological degeneration scores (p = 0.01) as compared to the samples without cartilage endplate. However, factors such as inflammation and proteoglycan loss did not display any significant association with the presence of cartilage endplate in herniated disc samples. Interestingly, the samples with the presence of CEP had significantly decreased expression of IL-1β (p = 0.03) and TNF-α (p = 0.01). Conclusion: Our findings suggest that while CEP contributes to structural degeneration of the disc, its suppressive effect on inflammatory mediators such as TNF-α and IL-1β may impair the natural resorption process. This paradox highlights that degeneration alone does not guarantee resorption; rather, a robust inflammatory and angiogenic response is required for spontaneous regression of herniated disc material.

ID: AGS26/2048

A389: Combining engineering precision with clinical relevance: a novel dual framework for assessing pedicle screw accuracy in spine surgery

Cordemans Virginie 1 , Arnaud Delafontaine 2 3 , Cartiaux Olivier 4

1 Department of Orthopedic and Traumatological Surgery, CHU Brugmann, Bruxelles, Belgium

2 Department of Orthopedic, Traumatological and Reconstructive Surgery, Prosthetic Replacement Unit of the Spine, Hip and Knee, Antoine Béclère Hospital, APHP, University Paris Saclay, Clamart, France, CIAMS, University Paris-Saclay, Orsay, France, CIAMS, University of Orléans, Orléans, Clamart, France

3 CIAMS, University Paris-Saclay, Orsay, France

4 Haute École ICHEC-ECAM-ISFSC, ECAM Brussels Engineering School, Bruxelles, Belgium

Introduction: Accurate pedicle screw placement is essential in spine surgery, as malposition can result in neurological, vascular, or visceral injury and compromise construct stability. Traditional qualitative classifications, such as Gertzbein-Robbins or Heary, remain observer-dependent and poorly reproducible, limiting cross-study comparisons and objective assessment. To overcome these limitations, we introduce a dual quantitative framework: (1) coaxiality , an engineering-derived metric summarizing the geometrical accuracy of screw insertion, and (2) pedicle wall distance (dpw) , a novel parameter providing surgeons with direct, millimetric, clinically actionable feedback. This approach bridges the gap between engineering precision and operative applicability, offering both rigorous assessment and immediate clinical relevance. Material and Methods: Eight operators (four orthopedic senior surgeons, four residents) performed a total of 240 pedicle screw insertions on synthetic lumbar spine models, using three techniques: freehand, cone-beam CT (CBCT)-assisted, and navigation-assisted. Preoperative trajectories were defined in 3D, and actual screw positions were digitized with sub-millimetric precision using a coordinate measuring system. Insertion errors (entry point, target point, orientation angle), coaxiality, and dpw were computed. Correlations assessed whether coaxiality captured overall insertion accuracy. dpw was validated against intraoperative CBCT grading according to Gertzbein and Heary classifications. Agreement coefficients (Kappa, Gwet), sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated. Results: Of 236 analyzable screws, coaxiality correlated strongly with entry point (ρ = 0.41), target point (ρ = 0.85), and orientation angle errors (ρ = 0.48), confirming its robustness as a standardized engineering metric. However, as a composite measure, coaxiality is difficult to interpret directly during surgery. In contrast, dpw quantified screw proximity to the pedicle wall in millimeters, providing immediate, actionable feedback. Compared with CBCT grading, dpw demonstrated almost perfect agreement (Kappa = 0.86, Gwet = 0.96), with sensitivity 0.96, specificity 0.97, PPV 0.79, and NPV 0.99. dpw additionally detected breaches not captured by qualitative classifications, including superior pedicle violations, highlighting its complementary clinical value. Conclusion: This study presents a dual-metric, ISO-based framework for pedicle screw evaluation in spine surgery. Coaxiality offers a rigorous, standardized measure of geometrical accuracy suitable for research, multicenter comparisons, and reproducibility. Pedicle wall distance, in contrast, delivers direct, millimetric, and clinically interpretable feedback, enabling surgeons to identify and correct potential breaches in real-time. Together, these complementary parameters bridge the traditional gap between engineering precision and intraoperative utility, establishing a new standard for quantitative, reproducible, and clinically actionable assessment of pedicle screw placement. Implementation in clinical practice and multi-institutional studies could enhance surgical safety, reduce complication rates, improve educational outcomes, and ultimately optimize patient care. This dual approach represents a transformative step in combining engineering rigor with tangible surgical relevance.

ID: AGS26/2079

A390: Pre-transfusion testing in cervical spine surgery: are we over-prepared?

Jorge Nuñez 1,2 , Anna Taberner 1 , Juan Diego Montenegro 1 , Victoria San Gil 1 , Berta Escudero 1

1 Hospital Universitari MútuaTerrassa, COT & CARE Institute SL, Terrassa, Spain

2 Clinica Teknon, Barcelona, Spain

Introduction: Previous studies have shown that routine pre-transfusion testing is unnecessary for most patients undergoing cervical spine surgery, except for those with preoperative hemoglobin (Hb) levels below 12 g/dL. However, the applicability of this recommendation in patients undergoing anterior cervical arthrodesis (ACA) remains uncertain. This study aimed to evaluate the necessity of pre-transfusion testing in this specific patient subgroup. Material and Methods: A prospective study was conducted including patients undergoing ACA for degenerative conditions between 2014 and 2023 at our center. Out of 672 patients, 318 fulfilled the inclusion and exclusion criteria. Demographic, clinical, and radiological data were analyzed. The variables collected included: age at the time of surgery, sex, ASA (American Society of Anesthesiologists) classification, need for transfusion, number of fused levels, and preoperative and postoperative laboratory values, particularly hemoglobin (Hb) and hematocrit (Hct). Bivariate analyses were performed to identify predictive factors for transfusion based on preoperative Hb (< or ≥ 12 g/dL). The crossmatch-to-transfusion ratio (C/T), likelihood of transfusion (%T), and transfusion index (TI) were also assessed. Results: The overall transfusion rate was 0.6% (2/318 patients). No cases of emergency transfusion or same-day transfusions at the time of surgery were recorded. Regarding laboratory values, mean preoperative Hb and Hct were 14.2 ± 1.5 g/dL and 42.3% ± 4.6%, respectively, while mean postoperative values were 12.8 ± 1.6 g/dL for Hb and 35.1% ± 4.4% for Hct. A mean decrease of 1.4 g/dL (standard deviation [SD] 1.1) in Hb and 7.2% (SD 4.1) in Hct was observed between the preoperative and postoperative measurements. The difference in hemoglobin decrease between patients who required transfusion and those who did not was 0.6 g/dL. Thirteen patients underwent surgery with preoperative Hb < 12 g/dL. Bivariate analysis showed that patients with preoperative Hb < 12 g/dL had an odds ratio (OR) of 25.33 (95% CI: 1.49-429.8; p = 0.02) for requiring transfusion compared with those with Hb ≥ 12 g/dL. Moreover, statistically significant differences were identified between the groups according to preoperative Hb levels, specifically in age (p = 0.002), ASA classification (p < 0.001), and transfusion rate (p = 0.001), the latter being 7.7% in the Hb < 12 g/dL group versus 0.3% in the Hb ≥ 12 g/dL group. In the group of patients with Hb < 12 g/dL, suboptimal values were found in transfusion efficiency indicators: a C/T ratio of 13.0, a %T of 7.69%, and a TI of 0.08. The TI was notably different between groups (0.08 in patients with Hb < 12 g/dL versus 1 in those with Hb ≥ 12 g/dL), although this difference did not reach statistical significance (p = 0.14). Conclusion: Routine blood reservation and pre-transfusion testing are unnecessary in patients undergoing ACA, regardless of their preoperative hemoglobin level. These findings support the optimization of preoperative protocols by eliminating routine pre-transfusion testing in this surgical subgroup.

ID: AGS26/2278

A391: Expression of LBX1 and TGFβ1 in paraspinal muscles: correlation with curve severity and muscle remodeling in adolescent idiopathic scoliosis

Kinza Fatima 1 , Ayesha Ahmad 1 , Muhammad Saad Ilyas 2 , Amer Aziz 2 , Uruj Zehra 1

1 University of Health Sciences, Lahore, Pakistan , 2 Ghurki Trust Teaching Hospital, Lahore, Pakistan

Introduction: Adolescent idiopathic scoliosis (AIS) is a complex three-dimensional spinal deformity, with genetic and molecular factors thought to play central roles in its pathogenesis. Genome-wide association studies have consistently implicated LBX1 and TGFβ1 in AIS susceptibility, yet their functional role in paraspinal muscle pathology remains poorly understood. Since the concave and convex sides of scoliotic curves exhibit distinct structural and histological changes, exploring the differential expression of LBX1 and TGFβ1 in these muscles may provide valuable insights into mechanisms driving curve severity, muscle remodeling, and disease progression. Material and Methods: Paraspinal muscle samples were collected bilaterally from idiopathic scoliosis patients (n = 22) and from degenerative disc disease patients (n = 21, controls) undergoing surgery, following informed consent and detailed clinical documentation. Radiographic parameters including coronal and sagittal Cobb angles, vertebral rotation, apical vertebral translation (AVT), Risser’s sign, and coronal balance were measured using RADIANT DICOM viewer. Muscle specimens were analyzed for TGF-β1 and LBX1 mRNA expression, and histological evaluation of necrosis, atrophy, hypertrophy, and fatty involution was performed on H&E-stained sections. Statistical analysis was conducted using one-way ANOVA and linear regression. Results: Scoliotic patients (n = 22; mean age 14.1 ± 2.1 years; mean coronal Cobb angle 81.6° ± 17.7) included 61% thoracic and 39% double major curves. TGFβ1 mRNA expression was significantly higher on the concave side compared to controls (p = 0.04). Both concave and convex paraspinal muscles showed increased necrosis, atrophy, hypertrophy, and fatty involution versus controls (p < 0.001), with greater fatty involution on the concave side (p = 0.004). Fatty involution correlated positively with TGFβ1 expression (r = 0.28, p = 0.02), and expression also trended with hypertrophy (r = 0.21, p = 0.08). LBX1 expression on the convex side correlated significantly with coronal Cobb angle (p = 0.03) and apical vertebral translation (p = 0.006). Conclusion: Paraspinal muscles in scoliosis exhibit marked histopathological changes, with greater fatty involution on the concave side. Elevated TGFβ1 expression correlates with fatty degeneration, suggesting a role in pathological muscle remodeling, while convex-side LBX1 expression associates with curve severity and vertebral translation, linking it to deformity progression. These findings highlight distinct molecular pathways contributing to both muscle degeneration and spinal curve progression in scoliosis.

ID: AGS26/270

A392: Extent of fatty infiltration in lumbar paraspinal muscles in adults with degenerative scoliosis: segmental differences in muscle involvement

Emel Özcan-Ekşi 1 , Murat Ekşi 2

1 Private Practice, Istanbul, Turkey , 2 İstinye University, Istanbul, Turkey

Introduction: With increased life expectancy, age-related spinal disorders such as adult degenerative scoliosis (ADS) are becoming more prevalent. However, few studies have comprehensively evaluated the role of lumbar paraspinal muscles in the pathogenesis of ADS. This study aimed to (1) systematically review current literature on the association between paraspinal muscles and ADS and (2) compare lumbar paraspinal muscle fatty infiltration between age- and sex-matched individuals with and without ADS, stratified by treatment type. Material and Methods: For the first part of the study, a systematic literature review was conducted using PubMed/MEDLINE databases, following PRISMA guidelines, to identify studies on the role of paraspinal muscles in ADS. For the second part, a total of 145 participants were evaluated: 40% (n = 58) were controls, 32.4% (n = 47) had conservatively managed ADS, and 27.6% (n = 40) had surgically treated ADS. Fatty infiltration of the multifidus, erector spinae, and psoas muscles was assessed via lumbar spine magnetic resonance imaging at three curve locations: apex, above the apex, and below the apex. Results: Compared with controls, patients with ADS exhibited significantly higher fatty infiltration in the multifidus, erector spinae, and psoas muscles on the concave and convex sides of the curve. Logistic regression analysis revealed that each one-point increase in fatty infiltration score of the concave-sided multifidus and psoas at the apex was associated with a 2.3-fold and 4.1-fold increased likelihood of having ADS, respectively. Additionally, a one-point increase in fatty infiltration in the concave-sided psoas above the apex and the convex-sided psoas below the apex was associated with 6.9-fold and 2.7-fold increased odds of ADS, respectively. Conclusion: Fatty infiltration in the psoas muscle may contribute to the early development of ADS, while increased fatty infiltration in the concave-sided multifidus at the apex may influence surgical decision-making. Comprehensive assessment of lumbar paraspinal muscles is essential for optimizing management strategies in patients with ADS.

ID: AGS26/2782

A393: Mechanically matched injectable photocrosslinkable materials loaded with exosomes repair combined annulus fibrosus-nucleus pulposus defect

Qiang Yang 1 , Yiming Dou 2

1 Capital Medical University Affiliated Beijing Friendship Hospital, Tianjin Hospital, Tianjin, China ,

2 Children's Hospital Zhejiang University School of Medicine, Zhejiang, China

Introduction: Annulus fibrosus-nucleus pulposus complex defect are known to precipitate severe intervertebral disc degeneration and disc herniation. Hydrogels are promising materials for disc regeneration. We comprise novel mechanically matched injectable photocro-sslinkable poly(lactide-co-propylene glycol-co-lactide) dimethacrylates (P7L2DMA) and methacrylated hyaluronic acid (MeHA), which are loaded with bone marrow mesenchymal stem cells-derived exosomes (Exos), and evaluate the therapeutic efficacy on intervertebral disc degeneration in vivo and in vitro. Material and Methods: Exosomes isolated via differential ultracentrifugation were characterized by TEM, NTA, and WB. Hydrogels P7L2DMA and MeHA underwent FTIR, ^1H NMR, and SEM characterization. Their mechanical properties were evaluated through compression tests. P7L2DMA@Exo and MeHA@Exo constructs were assessed for exosome release and internalization. Effects of MeHA@Exo, MeHA, and PBS on NPC regeneration and RAW 264.7 cell inflammation were analyzed using PCR, ELISA, flow cytometry, and immunofluorescence. Animal study groups: Sham, PME, PM, and Defect groups, in a rabbit disc defect model, with radiological and histological evaluations at 2, 4, and 12 weeks. Results: Characterization confirmed exomes and hydrogels met specifications. MeHA@Exo showed concentration-dependent effects in restoring type II collagen and reducing MMP3 in LPS-induced NPCs, reducing apoptosis and pyroptosis. PCR, ELISA, and immunofluorescence showed MeHA@Exo decreased IL-6 and TNF-α, and increased IL-4 and IL-10 in LPS-induced RAW 264.7 cells, with reduced M1 macrophage marker CD86. MRI and histological assessments showed that, compared to the Defect group, both the PME Group and the PM Group significantly alleviated intervertebral disc degeneration within 4 weeks. Moreover the PME Group continued to exhibit recovery in type II collagen expression and reduced apoptosis of NPCs up to 12 weeks. Conclusion: The combined application of P7L2DMA@Exo and MeHA@Exo demonstrates significant potential in repairing the annulus fibrosus-nucleus pulposus complex defect,alleviating inflammation,and decelerating IVDD.

ID: AGS26/1200

A394: DeVa (Decay Variance): a novel score calculated via postprocessing the changes in signal intensity of an intervertebral disc in a T2* multi-echo magnetic resonance image can quantify painful and degenerate lumbar vertebral discs

Stone Sima 1 , Alisha Sial 1,2,3 , Suhani Sharma 1 , Dheera Ananthakrishnan 4 , Victor Chen 1 , Lauren Barber 5 , Ashish Diwan 1,2,3

1 Spine Labs, Department of Orthopaedic Surgery, St George and Sutherland Clinical Campuses, School of Clinical Medicine, UNSW, Sydney, Sydney, Australia

2 Spine Service, Department of Orthopaedic Surgery, St George Hospital, School of Clinical Medicine, UNSW, Sydney, Sydney, Australia

3 Spinal Unit, Department of Orthopaedic Surgery, Royal Adelaide Hospital, University of Adelaide, Adelaide, Australia

4 Merunova, Sydney, Australia

5 Department of Orthopaedic Surgery, Virginia Commonwealth University, Richmond, VA, Richmond, United States

Introduction: Low back pain (LBP), a global disability leader, is often linked to intervertebral disc (IVD) degeneration. Traditional diagnostics like T2-weighted MRI provide qualitative but imprecise evaluations. A novel post-processing MRI technique, Decay Variance (DeVa), has shown promise in differentiating degenerate from healthy discs in animal studies.DeVa quantifies IVD degeneration by analyzing variations in signal intensities within each voxel in a T2* 2D FLASH multi-echo MRI sequence. This study aimed to validate DeVa clinically and explore its correlation with pain severity. Methods: A cross-sectional study included 77 chronic LBP patients and 8 controls, who underwent T2-weighted and T2* 2D FLASH MRI. DeVa scores (worst and sum of all disks) were recorded, alongside traditional assessments like disc bulge, stenosis, high-intensity zones, and Pfirrmann grade. Pain severity was measured with a numerical rating scale (NRS). Statistical analyses included Pearson correlation, t-tests, and Gardner-Altman plots to evaluate relationships between DeVa scores, degeneration, and pain. Results: DeVa scores correlated strongly with Pfirrmann grade (r = 0.692, p < 0.001) and were significantly higher in discs with bulge, stenosis, or high-intensity zones (p < 0.001). Moderate correlations was observed between worst DeVa scores, (r = 0.296, p < 0.01), total DeVa scores (r = 0.323, p < 0.005) and pain severity. Patients with chronic LBP without severe degeneration (Pfirrmann ≤ 3 with no stenosis observable on standard MRI) had significantly higher worst (1.38 ± 0.26 vs. 1.10 ± 0.29, p < 0.005) and total (5.39 ± 0.75 vs. 4.65 ± 0.61, p < 0.0.1) DeVa scores compared to controls. Discussion: DeVa offers a quantitative, non-invasive approach to assessing IVD degeneration, showing strong correlations with disc health and pain. It demonstrates enhanced sensitivity over traditional MRI, enabling the identification of pain-generating discs and informing personalized treatment strategies for chronic LBP. Further validation in larger populations is needed.

ID: AGS26/615

A395: Unexpected predictors of adjacent disc degeneration following balloon kyphoplasty: the role of vitamin D and fracture morphology

Musa Sezer 1 , Inan Uzunoglu 2 , Baran Taşkala 3 , Atakan Bayir 4 , Gökhan Gürkan 5 , Ezgi Suat Bayraktar 6 , Kazım Ayberk Sinci 7

1 Department of Neurosurgery, Ünye State Hospital, Ordu, Turkey, Ordu, Turkey

2 Department of Neurosurgery, Ankara Etlik City Hospital, Ankara, Turkey

3 Department of Neurosurgery, Başakşehir Çam and Sakura City Hospital, Istanbul, Turkey

4 Department of Radiology, İzmir Katip Çelebi University Atatürk Training and Research Hospital, İzmir, Turkey

5 Department of Neurosurgery, Medikalpoint İzmir Hospital, İzmir, Turkey

6 Department of Radiology, Aydın State Hospital, Aydın, Turkey

7 Department of Radiology, Istanbul Kanuni Sultan Süleyman Training and Research Hospital, Istanbul, Turkey

Introduction: Balloon kyphoplasty is a widely used minimally invasive technique for treating osteoporotic vertebral compression fractures, providing rapid pain relief and restoration of vertebral height. However, the impact of kyphoplasty on adjacent disc degeneration (ADD) remains controversial. While some studies have reported increased rates of ADD following cement augmentation, others have suggested that age-related degeneration may play a more dominant role. Proposed mechanisms include thermal and chemical effects of polymethylmethacrylate (PMMA), altered biomechanical load distribution, and endplate injury. Recent literature highlights intradiscal cement leakage and preexisting advanced disc degeneration as major predictors of ADD. Nevertheless, data on additional risk factors remain limited. This study aimed to investigate the incidence and potential risk factors of ADD within one year after kyphoplasty in a cohort of osteoporotic fracture patients. Material and Methods: We retrospectively analyzed 62 patients who underwent balloon kyphoplasty for osteoporotic vertebral compression fractures between 2021 and 2023. All patients had a minimum of one-year MRI follow-up. Adjacent disc degeneration was defined as a Pfirrmann grade increase of ≥ 1 at the level above (N+1) or below (N-1) the treated vertebra, as well as the second adjacent discs (N+2, N-2). Recorded variables included age, sex, body mass index (BMI), bone mineral density (BMD), vitamin D level, time from trauma to operation, injected cement volume, cement leakage pattern (including intradiscal leakage), fracture type (AO Spine A1-A4), and cement viscosity. Statistical analyses comprised independent t-tests, chi-square tests, and assumption checks for normality and variance. A p-value < 0.05 was considered significant. Results: During one-year follow-up, 23 out of 62 patients (37.1%) developed significant adjacent disc degeneration. Analysis of N+1 levels revealed that vitamin D levels were significantly higher in patients with degeneration compared to those without (43.7 vs. 27.1 ng/mL, p = 0.013). No association was observed with age, sex, BMI, BMD, time to surgery, cement volume, or cement leakage. At N-1 levels, none of the evaluated parameters reached statistical significance. At N-2 levels, fracture morphology demonstrated a significant association: patients with AO type A3 fractures exhibited a higher rate of degeneration compared to other fracture types (p = 0.023). At N+2 levels, no significant predictors of degeneration were identified. Overall, classical risk factors such as osteoporosis severity, cement leakage, and BMI did not show a consistent impact in this cohort. Conclusion: Approximately one-third of patients developed adjacent disc degeneration within one year after kyphoplasty. While classical predictors such as osteoporosis severity and cement leakage were not significant, two notable associations were identified: higher vitamin D levels were linked with degeneration at the superior adjacent disc, and AO type A3 fractures predisposed to degeneration at the inferior second adjacent level. These findings suggest that factors beyond traditional biomechanical explanations may contribute to ADD following kyphoplasty. Recognition of such unexpected risk profiles could refine patient counseling and postoperative monitoring. Further prospective, large-scale studies are required to validate these observations and clarify the pathophysiological mechanisms involved.

ID: AGS26/1805

A396: Age-related metabolic signatures in structurally normal intervertebral discs

S. Rajasekaran 1 , Chitraa Tangavel 2 , Divya Arunachalam 2 , Nandakumar M 2 , Gnanaprakash Gurusamy 1

1 Ganga Hospital and Medical Centre, Coimbatore, India

2 Ganga Research Centre, Coimbatore, India ,

Background: Aging, primary etiological factor in intervertebral disc degeneration (IVDD) contributing to chronic low back pain and spinal disability, particularly in older adults. While transcriptomic and proteomic changes in IVDD has been extensively characterized, metabolic alterations occurring during physiological aging of intervertebral discs remain poorly understood. Investigating these changes is essential to identify early biochemical shifts that may precede or accompany degeneration. Methods: Our study untargeted ultra high pressure liquid chromatography-tandem mass spectrometry (UHPLC-MS/MS) to analyze the metabolomic profile of nucleus pulposus (NP) tissues obtained from 21 brain-dead organ donors stratified into three age groups: young-aged (YAD, 20-30 years), middle-aged (MAD, 31-50 years), and old-aged (OAD, > 50 years). All samples were classified as Pfirrmann grade I on MRI, indicating absence of degeneration. Data preprocessing included quality control normalization, log transformation, and auto scaling. Statistical analysis included Partial Least Squares Discriminant Analysis (PLS-DA), unpaired t-tests, and pathway enrichment using MetaboAnalyst v5.0. Results: Metabolomic profiling yielded 20,585 spectral matches, of which 3,871 unique metabolites were retained after quality filtering. Among these, 831 metabolites showed statistically significant differences across age groups. Annotation using HMDB classified these into 17 superclasses, with lipids and lipid-like molecules (39%) being the most abundant, followed by organoheterocyclic compounds and organic acids. Distinct clustering patterns in PLS-DA and hierarchical analysis confirmed age-associated metabolic shifts. In particular, in old-aged groups showed elevated levels of sphingolipids, steroid derivatives, branched-chain amino acids, and omega-3 polyunsaturated fatty acids, while saturated and omega-6 fatty acids declined. Pathway enrichment highlighted dysregulation in amino acid metabolism, fatty acid biosynthesis, nucleotide metabolism, and energy pathways. Conclusion: This study presents the first comprehensive metabolomic profiling of human intervertebral disc tissues across physiological aging. Age-associated metabolic shifts were evident even in structurally normal discs, suggesting that metabolic alterations may serve as early indicators or drivers of disc aging. Clinical Significance: Establishing age-specific metabolic baselines in non-degenerated discs enhances our understanding of disc biology and aging. These insights might assist the development of early biomarkers for disc degeneration and enable targeted interventions to maintain disc health with aging.

OP45: Lateral / Oblique Fusion

ID: AGS26/1144

A397: Full endoscopic lumbar discectomy: 10-year single-centre experience with focus on readmission trends and surgical outcomes

Sasidharan MDS 1

1 Gleneagles HealthCity, Chennai, India

Introduction: Minimally invasive spine surgery (MIS), particularly full endoscopic lumbar discectomy, has transformed the management of lumbar disc herniation by enabling shorter hospitalization, faster recovery, and reduced tissue trauma compared to traditional open surgery. While initial studies have reported favourable outcomes, data on postoperative readmission rates remain limited, especially in the context of surgical experience beyond the learning curve. This study provides a comprehensive analysis of a decade of single-centre experience, emphasizing readmission trends and short-term clinical outcomes. Material and Methods: A retrospective review was conducted on adult patients who underwent full endoscopic lumbar discectomy between 2014 and 2025 at a single centre. A total of 699 cases were analysed, comprising an initial cohort of 578 cases (2014-2023) and a recent cohort of 121 consecutive cases (2023-2025) performed. The primary outcome was the rate of readmission within 30 and 90 days postoperatively. Secondary outcomes included reasons for readmission and clinical outcomes, assessed using the Visual Analog Scale (VAS) for back and leg pain, and the Oswestry Disability Index (ODI). Results: In the initial cohort of 578 cases, the overall 30- and 90-day readmission rate was 4.32%, with recurrent disc prolapse, residual compression/incomplete discectomy, and spondylodiscitis being the primary causes. These readmissions were more frequent during the early phase of the learning curve. Notably, in the subsequent 121 cases performed during the most recent 2 years, the readmission rate was 0% at both 30 and 90 days. There were no cases of residual compression, recurrent disc herniation, or spondylodiscitis observed in this advanced cohort. Patients across all cases showed statistically significant improvements in VAS and ODI scores, indicating effective symptom relief and functional recovery. Conclusion: This 10-year single-centre experience confirms that full endoscopic lumbar discectomy is a safe, effective, and reproducible surgical option for lumbar disc herniation, with consistently low readmission rates. Importantly, the data demonstrate that as surgical expertise advances, readmission rates decrease substantially, reaching zero in the most recent series. These findings highlight the pivotal role of surgical experience in optimizing patient outcomes and minimizing complications, and strongly support the wider adoption of full endoscopic techniques in modern spinal surgery practice.

ID: AGS26/1254

A398: Two portals, two paths, one goal: who reaches proficiency first?

Yanting Liu 1 , Yi-Hao Liang 2 , Kun Wu 3 , Jin-Sung Kim 4

1 Department of Neurosurgery, CHA Bundang Medical Center, Seoul, South Korea

2 Orthopaedic Hospital of Guangdong Provincial Hospital of Chinese Medicine, Guangzhou University of Chinese Medicine, Guangzhou, China

3 Department of Orthopedics, The Second Hospital of Jilin University, Changchun, China

4 Department of Neurosurgery, Seoul St. Mary's Hospital, College of Medicine, The Catholic University of Korea, Seoul, South Korea

Introduction: The principal barrier to widespread dissemination of ESS lies not in equipment availability, but in the steepness of its learning curve. How quickly surgeons reach proficiency, and what helps them get there, remains inconsistently defined across uniportal and biportal approaches. This study compares learning-curve dynamics head-to-head and distills a competency-based, stage-matched training roadmap. Material and Methods: A PRISMA-guided systematic review of PubMed, Embase, Web of Science, and Cochrane (inception-Aug 2025) identified original studies reporting learning-curve metrics for uniportal or biportal ESS. Eligible designs included RCTs and observational cohorts/series (≥ 10 cases) with outcomes such as operative time, complications, conversions, radiation exposure, and CUSUM/risk-adjusted CUSUM (RA-CUSUM) thresholds. Risk of bias was assessed using RoB-2 (RCTs) and NOS (non-randomized). Where feasible, random-effects meta-analysis was performed; predefined subgroups examined prior surgical experience, case complexity, and enabling technologies (navigation/simulation/AR). Results: From 1,532 unique records screened, 59 full texts were assessed; 42 studies were synthesized qualitatively and 32 contributed to meta-analysis. Evidence from decompression cohorts shows that interlaminar endoscopic approaches, whether uniportal or biportal, generally stabilize after a modest number of cases, with biportal surgery reaching stable performance earlier and uniportal procedures requiring somewhat longer exposure. These trajectories are well represented by a four-stage model: surgeons begin as novices, practicing basic docking under supervision while susceptible to trajectory deviations and inadvertent neural irritation; progress as beginners to independent handling of simple herniations, where incomplete decompression and durotomy remain risks; advance to competence with reliable management of stenosis and ligamentum flavum hypertrophy, challenged by bleeding control and facet preservation; and finally achieve proficiency when capable of consistently treating multilevel or revision cases with reproducible outcomes and low complication rates. In contrast, the transforaminal uniportal pathway demonstrates greater variability and technical demand. To capture this trajectory, a six-stage model is warranted. Novices achieve safe access in protrusion-type herniations but remain vulnerable to nerve irritation and excessive fluoroscopic exposure. Beginners consolidate basic foraminoplasty and cannula docking. Competence, typically reached in standard cases, allows stable outcomes but complications persist in migrated or extruded fragments, leading to a consolidator stage that requires repeated exposure to adhesions, engorged vessels, and recurrent disease. Surgeons who progress to proficiency demonstrate reliable outcomes in multilevel and revision pathology, while expert performance is defined not by case numbers alone but by the consistent reproducibility of results in anatomically demanding scenarios and the ability to innovate and mentor others. Conclusion: Overall, biportal surgery offers a shallower learning curve slope during the novice and beginner phases, facilitating safer early adoption, whereas uniportal techniques, particularly transforaminal, require prolonged consolidation but ultimately reward mastery with durable efficiency, reduced tissue disruption, and sustained clinical benefits. A competency-based, stage-specific curriculum that aligns case selection with anticipated complication risks, supported by mentorship, simulation, and enabling technologies, provides the most pragmatic and safe pathway to shorten the ESS learning curve and optimize surgical outcomes.

ID: AGS26/2554

A399: Combined lateral interbody fusion and balloon kyphoplasty without vertebral body replacement for osteoporotic vertebral fractures: a retrospective case series

Shogo Tahata 1 , Seiichiro Naruo 1 , Toru Fujimoto 1 , Takahiro Ushimaki 1 , Takafumi Inoue 1

1 Naruo Orthopedic Hospital, Kumamoto, Japan

Introduction: Osteoporotic vertebral fractures (OVF) in the elderly often present with instability that is difficult to manage through conventional approaches. Vertebral body replacement carries a high surgical burden in fragile patients. We investigated the feasibility of a less invasive anterior-posterior reconstructive strategy using lateral interbody fusion (LIF) combined with balloon kyphoplasty (BKP), without vertebral body replacement. Material and Methods: We retrospectively reviewed 10 consecutive OVF patients (5 men, 5 women; mean age 76.7 years) who underwent LIF-based anterior-posterior surgery between March 2023 and April 2025. In 10patients, LIF was combined with BKP. Operative time, blood loss, perioperative complications, postoperative pain, and neurological outcomes were evaluated. Results: Mean operative time was 195 minutes, with mean blood loss of 71 mL. All patients demonstrated postoperative low back pain relief and neurological improvement. One pleural injury occurred intraoperatively, but no severe perioperative complications were observed. LIF provides minimally invasive reconstruction of intervertebral instability, while BKP stabilizes the vertebral body itself. Their combination allows for robust spinal reconstruction without osteotomy or vertebral body replacement, reducing surgical invasiveness in elderly patients. Conclusion: Combined LIF and BKP without vertebral body replacement is a safe, effective, and minimally invasive surgical option for OVF, even in elderly and fragile patients.

ID: AGS26/303

A400: Full-endoscopic vs microsurgical decompression for lumbar disc herniation: one-year outcomes from a prospective comprehensive cohort trial (FEMT-LDH-Trial)

Frank Hassel 1 , Saravi Babak 2 , Alisia Zink 1 , Sara Ülkümen 1 , Sébastien Couillard-Després 3 , Gernot Michael Lang 2

1 Department of Spine Surgery, Loretto-Krankenhaus, Freiburg, Germany

2 Department of Spine Surgery and Trauma Surgery, University of Freiburg, Freiburg, Germany

3 Institute of Experimental Neuroregeneration, Spinal Cord Injury and Tissue Regeneration Center Salzburg (SCI-TReCS), Paracelsus Medical University, Salzburg, Austria

Introduction: Whether full-endoscopic decompression (FED) achieves patient-reported outcomes comparable to the long-established microsurgical decompression (MSD) for lumbar disc herniation remains debated. We prospectively compared both techniques with a focus on 1-year disability, pain, quality of life, perioperative events, and biomarker kinetics. Material and Methods: We conducted a single-centre, prospective comprehensive cohort study with two parallel arms: a randomized arm (1:1 FED:MSD) for patients accepting allocation and a contemporaneous preference arm for decliners, all treated under one protocol and follow-up schedule. Adults with single-level symptomatic herniation were eligible; conversions from FED to MSD were permitted and counted as FED complications under intention-to-treat in the randomized arm. The primary endpoint was the Oswestry Disability Index (ODI) at 1 year. Secondary outcomes comprised VAS back/leg, SF-36 domains, operative time, length of stay, complications (including readmission/revision), and perioperative laboratory indices. Patient-reported measures were analyzed as change scores (Δ) across prespecified timepoints-preoperative baseline, immediate postoperative, 3 weeks, and 1 year-using the contrasts post-pre, 3-weeks-pre, 1-year-pre, 3-weeks-post, and 1-year-3-weeks. Laboratory markers were evaluated as Δs at preoperative baseline, immediate postoperative, and 48 h (contrasts: post-pre, 48 h-pre, and 48 h-post). Analyses were prespecified for the pooled cohort and, separately, the randomized and non-randomized subsets. Continuous data were summarized as median (IQR); Welch’s t-test was used if both groups were approximately normal by Shapiro-Wilk, otherwise Mann-Whitney U; categorical data used χ 2 /Fisher’s exact tests; p < 0.05 two-sided. All analyses were performed in Python. Results: We enrolled 127 patients (FED 87, MSD 40); 37 accepted randomization (FED 23, MSD 14) and 90 entered the preference arm (FED 64, MSD 26). Operative time, length of stay, complications, and readmission/revision did not differ between techniques across analytic sets. PROMs improved substantially in both groups. At 1 year, ODI, VAS, and SF-36 outcomes were comparable between FED and MSD in pooled, randomized-only, and non-randomized-only analyses. Several early differences around 3 weeks favored MSD-e.g., greater additional ODI reduction from post-op to 3 weeks (pooled p < 0.001; randomized p = 0.003; non-randomized p = 0.005), higher SF-36 physical functioning at 3w-post (pooled p = 0.017; randomized p = 0.040), and lower VAS back at 3w-post in pooled and non-randomized cohorts (both p = 0.002)-but these did not persist at 1 year. Laboratory kinetics indicated a stronger 48-h inflammatory response after MSD (CRP and IL-6, p < 0.001 in pooled analyses), while other hematologic and metabolic changes were largely similar. Conclusion: FED and MSD yielded equivalent, clinically meaningful 1-year improvements in disability, pain, and health-related quality of life, with transient early advantages for MSD that resolved by 1 year and biomarker evidence of less systemic inflammation after FED. Both procedures are safe, effective options; longer-term multicentre data will clarify durability and refine patient selection.

ID: AGS26/335

A401: Minimally invasive versus open transforaminal or posterior lumbar interbody fusion for degenerative lumbar disease and the long-term risk of adjacent segment pathology: a systematic review and meta-analysis

Farzin Farahbakhsh 1,2 , Faramarz Roohollahi 1,2 , Mohammad Mohammadi 3 , Shervin Molavi 1,2

1 Neuroscience Institute, Sports Medicine Research Center, Tehran University of Medical Sciences, Tehran, Iran

2 Spine Center of Excellence, Yas Hospital, Tehran University of Medical Sciences, Tehran, Iran

3 School of Medicine, Iran University of Medical Sciences, Tehran, Iran

Introduction: Minimally invasive surgery (MISS) for lumbar fusion offers well-established short-term benefits over traditional open approaches, including reduced blood loss and shorter hospital stays. However, its long-term effect on adjacent segment pathology (ASP) remains a subject of debate. It is hypothesized that by preserving the posterior tension band, MISS techniques may reduce biomechanical stress on adjacent levels. The aim of this systematic review and meta-analysis was to compare the long-term incidence of ASP-specifically radiographic adjacent segment degeneration (ASDeg), adjacent segment disease (ASDis), and operative ASP (OASP)-between MISS and open transforaminal/posterior lumbar interbody fusion (TLIF/PLIF). Material and Methods: A comprehensive search was performed on August 6, 2025, across four electronic databases: PubMed, Web of Science, Scopus, and Cochrane CENTRAL to identify relevant randomized controlled trials and cohort studies. Data from eligible studies were synthesized by calculating pooled odds ratios (ORs) and 95% confidence intervals (CIs) using a common-effects model. The methodological quality of included studies was assessed using the Newcastle-Ottawa Scale for observational studies and the Cochrane Risk of Bias 2 (RoB 2) tool for RCTs. All analyses were performed using R version 4.2.3. We performed subgroup analyses based on follow-up duration: < 2 years, 2 to < 5 years, 5 to < 10 years, and ≥ 1 0 years. Results: Twenty-five studies (2 RCTs, 23 cohorts) comprising 3,227 patients (1,408 MIS, 1,819 open) met the inclusion criteria. The pooled meta-analysis demonstrated that the MISS approach was associated with a significantly lower incidence of all forms of adjacent segment pathology compared to the open approach. • ASDeg: OR 0.39 (95% CI 0.27-0.56; p < 0.001). A subgroup analysis showed a consistent benefit for MIS with an OR of 0.36 (95% CI 0.18-0.71) for a 2-5 year follow-up and an OR of 0.41 (95% CI 0.27-0.62) for a 5-10 year follow-up. • ASDis: OR 0.54 (95% CI 0.34-0.85; p = 0.008). This protective effect was also seen in subgroup analyses for follow-ups of 2 to 5 years (OR 0.37, 95% CI 0.15-0.90) and 5 to 10 years (OR 0.43, 95% CI 0.23-0.82). • OASP: OR 0.55 (95% CI 0.39-0.79; p = 0.001). For follow-ups between 2 and 10 years, the results were consistently in favor of MISS. Low statistical heterogeneity was observed across all analyses (I^2 < 25%). The two RCTs had “some concerns” for the overall risk of bias. Studies with follow-up beyond 10 years for OASP showed no statistical significance, likely due to small sample sizes. Conclusion: This meta-analysis provides strong evidence that MISS TLIF/PLIF is associated with a significantly lower long-term risk of adjacent segment pathology compared to the open approach. These findings suggest that the muscle- and ligament-preserving nature of MISS techniques may confer a durable protective effect on adjacent spinal levels. The significant reduction in OASP, a hard clinical endpoint, is particularly impactful.

ID: AGS26/389

A402: Patient-reported outcome measures following endoscopic decompression for cauda equina syndrome: an 18-month prospective cohort study

Joel Stoy 1 , Santhosh G. Thavarajasingam 2 , Samira Akmal 3 , Darren Lui 4 , Jin-Sung Kim 5 , Salvatore Russo 3

1 Imperial College School of Medicine, London, United Kingdom

2 Ludwig-Maximilians-Universität München, Munich, Germany

3 Imperial College Healthcare NHS Trust, London, United Kingdom

4 St. George’s University Hospitals NHS Foundation Trust, London, United Kingdom

5 The Catholic University of Korea, Seoul, South Korea

Introduction: While open laminectomy has traditionally been the standard surgical approach for cauda equina syndrome (CES), endoscopic lumbar decompression is increasingly utilized. This endoscopic approach minimizes collateral tissue damage, which we hypothesize will reduce long-term morbidity while achieving immediate postoperative clinical outcomes equivalent to open laminectomy. To date, no prospective studies have evaluated long-term patient-reported outcome measures (PROMs) following endoscopic decompression for CES; this study assesses these outcomes over an 18-month period. Material and Methods: This prospective, longitudinal cohort study was conducted at Imperial College Healthcare NHS Trust between November 2023 and September 2025. We enrolled 21 patients with CES who underwent endoscopic lumbar decompression. Patients were selected based on clinical and radiological confirmation of CES. Baseline data included symptom duration, neurological deficits, and preoperative Visual Analogue Scale (VAS) scores for leg and back pain. Surgical details were also recorded. Postoperative follow-ups at 3, 6, 12, and 18 months included VAS (leg/back), Oswestry Disability Index (ODI), and EQ-5D-5L scores collected via questionnaires. Postoperative MRI, along with follow-up imaging, was performed in all patients to assess decompression and monitor for recurrence. We used paired Wilcoxon tests to assess changes in pain scores and analyzed correlations between ODI and EQ-5D-5L using Spearman’s rank. Results: The mean age of the cohort was 47.7 years. Bilateral sciatica was the most frequent presentation (76.2%), with motor and sensory deficits each present in 81% and autonomic dysfunction in 91%. Mean symptom duration was 11.7 days, operative time 112.8 minutes, and hospital stay 1.3 days. Complications occurred in 2 patients (9.5%), comprising two new neurological deficits. Leg pain (VAS) improved significantly from 6.9 ± 2.3 preoperatively to 0.6 ± 1.3 postoperatively (p < 0.0001), with sustained reductions at 6 M (2.2 ± 1.5, p = 0.0049), 12 M (2.4 ± 2.4, p = 0.0020), and 18 M (2.4 ± 2.3, p = 0.0156). Back pain VAS decreased from 7.4 ± 1.2 preoperatively to 1.9 ± 1.7 postoperatively (p < 0.0001), improved at 6 M (3.8 ± 2.3, p = 0.0020), but showed no significant change at 12 M (6.1 ± 2.8, p = 0.156) or 18 M (4.4 ± 2.8, p = 0.078). ODI was 37.3 ± 23.6 at 3 M, improved at 6 M (27.7 ± 17.3), but worsened at 12 M (40.2 ± 24.7) and 18 M (34.6 ± 29.8). EQ-5D-5L increased from 0.59 ± 0.24 at 3 M to 0.72 ± 0.22 at 6 M, with values of 0.66 ± 0.19 at 12 M and 0.76 ± 0.17 at 18 M. EQ-VAS rose from 41.0 ± 17.3 at 3 M to 66.3 ± 11.7 at 6 M, with later values of 57.5 ± 25.5 at 12 M and 59.3 ± 29.8 at 18 M. Conclusion: Endoscopic decompression for CES provided sustained leg pain relief over 18 months. Early improvements in PROMs, such as ODI and EQ-5D-5L, showed variable trajectories beyond 12 months, reflecting small cohort limitations. At 18 months, outcomes were comparable to, and in some cases slightly better than, those reported in the literature for open laminectomy.

ID: AGS26/1492

A403: Impact of lateral plate and screw fixation on long term radiographic subsidence in lateral lumbar interbody fusion

Joseph Hudson 1 , David McCarthy 1 , Andrew Legarreta 1 , Thomas Buell 1 , Nitin Agarwal 1 , David Hamilton 1 , David Okonkwo 1 , Vincent Miele 1

1 University of Pittsburgh Medical Center, Pittsburgh, United States

Introduction: Lateral Lumbar Interbody Fusion (LLIF) can be performed in a stand-alone manner, with lateral plate and screw fixation, or with supplemental posterior instrumented fusion. In vitro biomechanical data suggests lateral plate fixation limits segmental motion. This study aims to evaluate the impact of lateral screw and plate fixation on the rate and grade of long-term radiographic implant subsidence after LLIF. Material and Methods: A review of patients who underwent LLIF between 2014 and 2022 at two quaternary referral centers was conducted. All patients had CT scans at a minimum of one year after surgery. Patients with posterior instrumented fusion were excluded. Age, sex, BMI, smoking status, cage material, and operative level bone density (Hounsfield units) were collected. Post-operative implant subsidence at long-term follow-up was graded according to Marchi criteria by two separate spine surgeons. A stepwise multivariate logistic regression was performed. Results: 146 patients (219 surgical levels) met inclusion criteria. LLIF with lateral plate and screw fixation was performed in 65 patients (45%) and 81 patients (55%) underwent standalone LLIF. No intraoperative subsidence was observed. On univariate analysis, age, sex, smoking status, operative level bone density, and BMI were not significantly associated with radiographic subsidence on a per level basis (p > 0.05). Multivariable logistic regression revealed 3D printed porous titanium cages was moderately associated with reduced subsidence (OR 0.401, 95% CI 0.20 to 1.069, p = 0.07). Lateral plate fixation was significantly associated with (OR 0.484, 95% CI 0.099 to 0.652, p = .006) reduced subsidence. All 4 cases (3.5%) of grade two (severe) subsidence occurred in stand-alone LLIF patients. One delayed vertebral body fracture occurred in a lateral plate patient. All patients with lateral plates received 18 mm wide interbody grafts, while all but two stand-alone patients received 22mm wide implants. Conclusion: LLIF with lateral plate and screw fixation is significantly associated with a lower rate of radiographic subsidence at long term follow up when compared with stand-alone LLIF. There was no significant difference in reoperation rate. Our results trend towards the superiority of 3D printed porous titanium cages over PEEK cages. No cases of reoperation were identified in patients receiving 3D printed porous titanium stand-alone implants. The results of this study may be interpreted to suggest that lateral plate and screw fixation has potential to protect against severe (grade two or worse) radiographic subsidence.

ID: AGS26/500

A404: Learning curve and risk profile in thoracoscopic spine surgery: a 10-year single-center experience

Kunfeng Song 1,2 , Daniel Rosenthal 2 , Zizhuo Jin 2

1 Third People's Hospital of Henan Province, Zhengzhou, China

2 Hochtaunus-Kliniken, Bad Homburg, Germany

Introduction: Thoracoscopic spine surgery is a technically demanding procedure with a steep learning curve, often limiting its widespread adoption.This study analyzes the learning curve and complication rates over a 10-year period to assess how surgical experience impacts outcomes in thoracoscopic treatment of thoracic disc herniation. Material and Methods: A total of 182 thoracoscopic procedures were divided chronologically into three stages (early: cases 1-60, mid: 61-120, late: 121-182). Operative time, complication rate, and postoperative recovery were compared. Results: Mean operative time decreased significantly from 210 min (early) to 172 min (late), p < 0.01. Complications such as dural tears and pleural effusion dropped from 13.3% to 5.4%. Surgical confidence and proficiency plateaued after approximately 100 cases. Conclusion: Thoracoscopic spine surgery presents a clear learning curve, requiring at least 100 cases for stable, low-complication outcomes. Structured training and simulation-based practice are essential to safely broaden adoption and reduce surgeon-related risk.

ID: AGS26/680

A405: Safety and accuracy of fully percutaneous cervical pedicle screw instrumentation

Gonzalo Moros-Daza 1 , Feras Rawagah 1 , Bujung Hong 1 , Thomas Marx 1 , Petros Evangelou 1 , Hind Chaib 1 , Maria Kipele 1 , Ina Konietzko 1 , Ehab Shiban 1

1 University Hospital Lausitz Carl-Thiem, Cottbus, Germany

Introduction: Cervical pedicle screws offer superior biomechanics compared to lateral mass screws resulting in stronger and more durable constructs. Here, we present the case series of fully percutaneous cervical und upper thoracic pedicle screw instrumentation assisted by navigation. Material and Methods: A Prospective observational study between October 2024 and September 2025 was performed. The primary outcomes assessed were the accuracy of pedicle screw placement-radiologically evaluated using the Gertzbein and Robbins classification- and clinically defined by the lack of need for screw revision. Additional data collected included intraoperative blood loss, operative time, length of hospital stay, and the incidence of intraoperative and postoperative complications. Secondary outcomes encompassed improvements in neck pain, assessed using the Numeric Rating Scale (NRS), as well as the correlation between pedicle dimensions and pedicle breach. Results: Thirteen consecutive Patients and 42 screws were analysed. The mean patient age was 66 years. The average length of hospital stay was 15 days, with a mean operative time of 110 minutes and an average intraoperative blood loss of 350 mL. The most common underlying pathology was traumatic fracture, observed in eight patients, followed by three cases of cervical degenerative stenosis and two cases of spinal metastatic disease. Most pedicle screws were placed across three to six spinal levels. There were 42 and 10 pedicle screw in the subaxial cervical spine and thoracic spine, respectively. According to the Gertzbein and Robbins classification of screw placement accuracy, 28 cervical screws (66.6%) were classified as Grade A, 3 screws (7.1%) as Grade B, 5 screws (11.9%) as Grade C, 3 screws (7.1%) as Grade D, and 3 screws (7.1%) as Grade E. The most common side of breech was medial. One screw required revision due to malposition, resulting in a clinically acceptable screw placement rate of 97,6%. Regarding complications, one patient experienced bar dislocation and another developed occipital neuralgia, which resolved with conservative management. No cerebrospinal fluid (CSF) leak, postoperative infection, or vertebral artery injury was documented. The procedure resulted in a significant reduction in pain, with an average improvement of nearly 4 points on the VAS (p < 0.001, 95% confidence interval: [2.29; 5.40]). No statistically significant correlation was found between pedicle dimensions and the risk of pedicle breach. Conclusion: Percutaneous navigated cervical pedicle screw instrumentation is a safe and effective technique for patients with various spinal pathologies.

OP46: Prevention of Spine Surgery Complications

ID: AGS26/544

A406: Do all symptomatic adjacent segment diseases (ASD) require surgery? A prognostic classification and predictors of surgical treatment of lumbar ASD

Rishi Kanna 1 , Gnanaprakash Gurusamy 1

1 Ganga hospital, Coimbatore, India

Introduction: Symptomatic Adjacent segment disease presents in myriad ways with varying clinic-radiological presentation but the optimal management and the factors that predict need for revision surgery are unclear. Material and Methods: Consecutive patients (n = 132) who experienced significant back or leg pain after an asymptomatic period (minimum 1 year) following a lumbar fusion were evaluated with dynamic radiographs and MRI. ASD was classified anatomically as proximal (86.3%), distal (12.1%, or combined (1.5%) and, sub-classified pathologically into six sub-types: Type 1: Disc prolapse (17.4%), Type 2: Disc degeneration with intra-discal instability (15.1%), Type 3: Anterolisthesis and facet instability (6.8%), Type 4: Retrolisthesis (25%), Type 5: Stenosis (21.9%) and Type 6: Combined types (13.6%). Results: The mean duration between the index surgery and ASD was 5.9 ± 4.1 years. Surgery was required in 40.9% (54/132). Six factors were highly predictive of surgical treatment for ASD, viz., Diabetes mellitus (OR-4.3, p = 0.007), ASD - disc prolapse (OR 6.3, p = 0.018), ASD intra-discal instability (OR-16.8206, p = 0.003), ASD - anterolisthesis (OR - 8.2998, p = 0.043), Facet angle >75° (OR 37.6, p = 0.02) and radiculopathy (OR 48.0, p = 0.002). Retrolisthesis was the most benign type as 84.8% patients were managed conservatively (p = 0.016). Among proximal ASD, retrolisthesis was the most common subtype (28.9%) while disc prolapse was the most common subtype in distal ASD 37.5% (p = 0.032). Conclusion: The study classifies the versatile clinic-radiological presentation of ASD, and identifies factors that predicted the development of the different sub-types of ASD. Knowledge about the six clinic-radiological factors that predict the need for surgical treatment in ASD is beneficial.

ID: AGS26/259

A407: Independent risk factors for retrograde ejaculation after anterior versus posterior lumbar fusion in male patients: a cox proportional hazards retrospective analysis

Anthony Baumann 1 , Robjert Trager 2 , Omkar Anaspure 3 , Ethan Cottrill 4 , Peter Passias 4 , Jacob Hoffmann 1

1 Cleveland Clinic Akron General, Akron, United States

2 Department of Rehabilitation Services, University Hospitals, Cleveland, United States

3 Perelman School of Medicine, University of Pennsylvania, Philadelphia, United States

4 Department of Orthopaedic Surgery, Duke University, Durham, United States

Introduction: RE is a feared yet uncommon complication after ALF in males that can lead to infertility. However, knowledge of the incidence and risk factors for RE is sparse due to limitations of the current literature. Our purpose is to examine the incidence and independent risk factors for retrograde ejaculation (RE) after anterior lumbar fusion (ALF) versus posterior lumbar fusion (PLF) in males to enhance decision-making. Material and Methods: This pre-registered retrospective database study (TriNetX) included adult males (≥ 18 years) without prior RE who underwent primary ALF (ALF cohort) or PLF (PLF cohort). The outcome was the two-year incidence of RE after ALF or PLF, estimated using Kaplan-Meier analysis, with hazard ratios (HRs) and 95% confidence intervals (CI) calculated using a multivariable Cox proportional hazards model (p < 0.05). Results: There were 68 cases (< 0.1%) of RE among 106,005 males who underwent fusion with 26 cases (0.1%) in the ALF cohort (n = 18,437) and 42 cases (< 0.1%) in the PLF cohort (n = 87,568). Independent risk factors (p < 0.05) for RE after fusion included ALF [95% CI] (HR: 2.86 [1.75, 4.69]), use of recombinant human bone morphogenetic protein-2 (rhBMP-2, HR: 2.65 [1.25, 5.62]), and thyroid disorders (HR: 2.45 [1.28, 4.69]) whereas age was protective (HR: 0.97 [0.95, 0.99]). Variables that were not independent risk factors (p > 0.05) included diabetes and prescription of alpha-adrenergic antagonists and antipsychotics. Conclusion: In the largest study to date, the incidence of RE after ALF as determined by clinical diagnosis and/or urine analysis was rare (0.1%). ALF was an independent risk factor for RE as compared to PLF, although the clinical relevance of this risk likely varies on a case-by-case basis. Other independent risk factors for RE identified in this study included the intraoperative use of rhBMP-2 and a history of thyroid disorders, while increasing patient age was protective. Importantly, diabetes, alpha-adrenergic antagonists, and antipsychotics were not independent risk factors for RE after fusion. Spine surgeons should consider both the incidence and risk factors for RE in their decision-making process with male patients. More clinical research is needed regarding the resolution of RE after ALF to further enhance shared decision-making.

ID: AGS26/262

A408: The risk of dysphagia after short-segment posterior occipital-C2 fusion versus C1-C2 cervical fusion: a retrospective cohort study

Anthony Baumann 1 , Robjert Trager 2 , Omkar Anaspure 3 , Nicole Baumann 4 , Wyatt Ramey 5 , Jacob Hoffmann 1

1 Cleveland Clinic Akron General, Akron, United States

2 Department of Rehabilitation Services, University Hospitals, Cleveland, United States

3 Perelman School of Medicine, University of Pennsylvania, Philadelphia, United States

4 Department of Speech Language and Hearing Sciences, Purdue University, West Lafayette, United States

5 Department of Neurosurgery, Houston Methodist Hospital, Houston, United States

Introduction: Posterior O-C2 fusion is suspected of being associated with a higher risk of postoperative dysphagia relative to C1-C2 fusion, yet comparative studies on this topic are limited. To test the hypothesis that short-segment posterior occipital-C2 (O-C2) fusion is associated with a higher risk of postoperative dysphagia than short-segment posterior C1-C2 fusion through two years, to enhance preoperative education and shared decision-making. Material and Methods: This database study (TriNetX) included two cohorts of adults (≥18 years old) who underwent either posterior O-C2 or C1-C2 fusion without extension into the subaxial cervical or thoracic spine. The primary outcome was the risk ratio (RR) of dysphagia, based on incidence through two years. Dysphagia risk factors were propensity matched to minimize confounding. Secondary outcomes included RRs of potential oropharyngeal/respiratory complications and procedures. Results: There were 371 patients per cohort (mean age: 55 years). Comparing O-C2 to C1-C2 fusion through two years, there was a statistically significant increase in risk of dysphagia (RR: 1.97 [1.38, 2.82]; p < 0.001; 20.8% versus 10.5%), with a median time to 50% of events of 11.0 days and 36.3 days, respectively. For secondary outcomes, there was a statistically significant increased risk of obstructive sleep apnea (RR: 2.07 [1.11, 3.86]; p = 0.018), speech therapy (RR: 2.04 [1.32, 3.16]; p = 0.001), and swallowing evaluation (RR: 1.85 [1.25, 2.74]; p = 0.002), and small, non-significant increases in risk of dyspnea (RR: 1.30 [0.83, 2.05]; p = 0.255) and pneumonia (RR: 1.09 [0.63, 1.88]; p = 0.765). Conclusion: The findings of the current study suggest that short-segment posterior O-C2 fusion is associated with a statistically significant and clinically meaningful increase in the risk of dysphagia as compared to short-segment posterior C1-C2 fusion in adults through two years postoperatively. Furthermore, O-C2 fusion may also be associated with an increased risk of other oropharyngeal and respiratory complications or treatments as compared to C1-C2 fusion, such as OSA, speech therapy, and swallowing evaluation. Spine surgeons should cautiously incorporate these results into their decision-making process when choosing an upper cervical fusion technique on a case-by-case basis. Future research should investigate the duration and severity of dysphagia after these surgeries as well as the underlying mechanisms to enhance care.

ID: AGS26/277

A409: Pre-operative presence of anterior osteophyte at L3-L4 and intrafacet vacuum phenomenon at L2-L3 could be predictive for adjacent segment disease following lumbar spine fusion

Murat Ekşi 1 , Emel Özcan-Ekşi 2

1 İstinye University, Istanbul, Turkey

2 Private Practice, Istanbul, Turkey

Introduction: Degenerative lumbar spine disorder (DLSD) is a public health problem presenting with low back pain, radicular leg pain, and poor quality of life. Even though conservative management is the first choice of treatment, patients could require surgery due to no improvement with conservative management and progressing neurologic deficits. Lumbar instrumented fusion is a commonly preferred surgery for DLSDs. Adjacent segment disease (ASD) is a common complication following spinal fusion surgeries either at cranial or caudal side of the fusion mass. Radiological predictors for poor outcomes and complications following spinal fusion are less known even though the role of spinopelvic parameters in ASD has been well understood. In the present study, we aimed to identify the possible pre-operative radiological predictors of ASD that could require revision surgery in subjects who underwent short-segment decompression and fusion for DLSD. Material and Methods: In the present study, we cross-sectionally evaluated the radiological data of the patients that underwent short-segment decompression and spinal fusion for DLSD at a tertiary spine clinic between 2013 and 2023. Results: In 209 patients who underwent short-segment decompression and spinal fusion, 31.6% (n: 66) had radiological signs of ASD. Multivariable analysis showed that only the presence of anterior osteophytes at L3-L4 level was a significant predictor for the development of ASD (OR: 3.528 ) when the whole cohort was analyzed. Anterior osteophyte at L3-L4 level (OR: 3.912-4.663) and intrafacet vacuum phenomenon at L2-L3 level (OR: 4.269-5.962) were positive predictors, whereas Modic changes at L4-L5 level (OR: 0.284) and fatty infiltration in the multifidus muscle at L2-L3 level (OR: 0.462) were negative predictors of ASD when analyses were done according to upper and lower fusion-ends. Conclusion: Presence of anterior osteophytes at L3-L4 level and intrafacet vacuum phenomenon at L2-L3 level were the strongest positive predictive factors for development of ASD following short-segment lumbar decompression and fusion. Interestingly, Modic changes and fatty infiltration in the multifidus were protective against ASD development. This knowledge will help clinicians to plan the surgery and postsurgical follow-ups properly in patients with DLSD.

ID: AGS26/1179

A410: Intraoperative monitoring of external urethral and anal sphincter potentials in lumbar spine surgery: a novel approach to preserve sacral function

Yoshikazu Yanagisawa 1

1 Department of Orthopedic Surgery and Spinal Disease Center, Fukuoka Mirai Hospital, Fukuoka city, Japan

Introduction: Intraoperative neurophysiological monitoring (IONM) is widely used to prevent postoperative motor deficits in lumbar spine surgery. However, monitoring of sacral roots (S2-S4), which are essential for bladder and bowel control, has been insufficiently addressed. We introduced a novel technique to simultaneously record external urethral sphincter (EUS) and external anal sphincter (EAS) activity, aiming to evaluate its feasibility and clinical significance. Material and Methods: Twelve male patients who underwent lumbar spine surgery under general anesthesia between June 2025 and August 2025 were prospectively analyzed. In addition to lower-limb muscle MEPs, needle electrodes were used to record EUS and EAS responses simultaneously. Patients were divided into two groups: those with lumbar canal stenosis (n = 7) and those without (n = 5). Recording rate, amplitude, and latency were compared between groups. Results: Both EUS and EAS signals were successfully obtained in more than 90% of cases, demonstrating stable feasibility. A strong positive correlation was observed between EUS and EAS responses within individuals (r ≈ 0.85). The mean EUS amplitude was significantly higher in patients with foraminal involvement compared with those without (43.7 μV vs. 27.8 μV), whereas EAS showed no significant differences. No patient developed severe postoperative bladder or bowel dysfunction. Conclusion: Simultaneous monitoring of EUS and EAS is a feasible and promising method for intraoperative assessment of sacral root function in lumbar spine surgery. This novel approach extends the role of IONM beyond motor preservation and may contribute to functional outcomes related to bladder and bowel control.

ID: AGS26/2966

A411: Pre-operative intake of antipsychotic medications is associated with higher rates of healthcare utilization, medical comorbidity, and opioid use following anterior cervical discectomy and fusion

Ankit Hirpara 1 , Nakul Narendran 1 , Jalen Warren 1 , Christina Cheng 1 , Christopher Furey 1 , Prashant Rajan 1

1 Case Western Reserve University/University Hospitals, Cleveland, United States

Introduction: Several studies have examined the impact of psychiatric disorders on outcomes following anterior cervical discectomy and fusion (ACDF). However, literature on the impact of psychotropics, such as antipsychotic medications (APMs), is limited. This is an important gap considering 40.4% of patients with mental health disorders receive pharmacologic treatments, many of which are also used for off-label or non-psychiatric purposes. Moreover, APMs are associated with metabolic, cardiovascular, and hormonal side effects that may compromise surgical outcomes. As such, the purpose of this study was to evaluate the impact of pre-operative use of APMs on rates of healthcare utilization, post-operative complications, revision surgery, and opioid use following ACDF. Material and Methods: The TriNetX database, commonly used in the orthopedic literature, was queried to identify patients over 18 years old who underwent isolated, primary ACDF without concomitant posterior instrumentation. Patients were stratified based on pre-operative use of APMs. Cohorts underwent propensity score matching in a 1:1 ratio based on 33 characteristics, including demographic variables, medical comorbidities, substance use, indication for ACDF, and multi-level ACDF. Cohorts were also matched based on underlying psychiatric diagnoses and concomitant psychotropic use, like antidepressants, in order to attribute observed associations to the pharmacologic profile of APMs. Lastly, patients were matched on suicidal ideation and prior healthcare utilization, which has been previously used as a proxy for the severity of underlying mental illness. The following outcomes were collected: 1) medical complications and healthcare utilization within 3 months, 2) opioid use within 2 years, and 3) surgical complications and revision surgery within 2 years. Results: Following propensity score matching, both cohorts were appropriately matched with respect to all covariates. Within 3 months, patients taking APMs (n = 6,138), compared to those who were not (n = 6,138), had significantly higher rates of healthcare utilization, including readmission (12.5% vs. 9.0%, p < 0.001), outpatient services (44.7% vs. 41.9%, p = 0.002), physical or occupational therapy evaluation (43.5% vs. 41.0%, p = 0.004), and emergency department visits (17.2% vs. 13.6%, p < 0.001). Patients taking APMs also had higher rates of acute kidney injury (2.8% vs. 1.5%, p < 0.001), venous thromboembolism (2.1% vs. 1.6%, p = 0.022), pneumonia (2.9% vs. 1.8%, p < 0.001), surgical site infection (1.1% vs. 0.5%, p < 0.001), sepsis (1.9% vs. 1.0%, p < 0.001), cerebrovascular accident (1.9% vs. 1.2%, p = 0.003), pulmonary embolism (1.2% vs. 0.8%, p = 0.032), and dysphagia (10.4% vs. 8.2%, p < 0.001). In terms of opioid use, patients taking APMs received more prescriptions at all time points up to two years (all p < 0.001). In addition, more patients taking APMs were prescribed opioids than control patients at every time point after 1 month (all p < 0.005). There were no differences in rates of surgical complications and revision surgery at 2 years. Conclusion: Pre-operative use of APMs is not associated with higher rates of surgical complications or revision surgery following ACDF. However, APMs are associated with greater healthcare utilization, medical morbidity, and opioid use. These findings may be related to the antagonistic effects of APMs on histamine, muscarinic, dopamine, and alpha-adrenergic receptors. Potential risks may be mitigated through opioid stewardship, multi-disciplinary peri-operative care, and patient counseling.

ID: AGS26/3117

A412: Incidence of proximal junctional kyphosis and failure in long spinal fusions with and without sublaminar bands

Marcos Parreira 1,2 , Marcelo Wajchenberg 1,2 , Delio E. Martins 1,2 , Felipe Monteiro 1,2 , Matheus Castanheira 1,2 , Nelson Astur 1,2

1 Hospital Israelita Albert Einstein, São Paulo, Brazil

2 Instituto Cohen, São Paulo, Brazil

Introduction: Proximal junctional kyphosis (PJK) and proximal junctional failure (PJF) are major complications after long spinal fusions, with incidence ranging from 5% to 46%. PJF is the most severe form, often requiring revision surgery with high morbidity and cost. Risk factors include patient characteristics, surgical strategy, and biomechanical stress at the transition between fused and mobile segments. Recently, the soft landing concept has been proposed, aiming to create a transition zone that reduces proximal stress. Placement of a sublaminar band at UIV+ represents a practical approach to achieve this goal. This study evaluated the incidence of PJK and PJF in patients undergoing long spinal fusions with and without sublaminar bands, focusing on reoperation rates as the most clinically relevant outcome. Material and Methods: A retrospective cohort study was conducted at a single tertiary institution, including patients over 18 years who underwent posterior spinal fusion of at least three levels with a minimum follow-up of 12 months. Radiographs were assessed preoperatively, postoperatively, and at last follow-up. PJK was defined as a proximal junctional Cobb angle ≥10° and ≥10° greater than baseline. PJF was defined as PJK associated with structural failure leading to reoperation. Patients were divided into two groups: with sublaminar bands at UIV+ (n = 14) and without bands (n = 10). Fusion constructs were stratified as ending at T10 or below versus extending into the upper thoracic spine. Primary outcomes were incidence of PJK, PJF, and reoperation. Results: Twenty-four patients were included, mean age 62 years (39-80), with 60% female. In the group without bands (n = 10), most had degenerative etiologies (80%), with fusions of 3-8 levels usually ending at T10. Three patients (30%) developed PJK progressing to PJF, all requiring revision, resulting in a 30% reoperation rate. In the band group (n = 14), etiologies were more heterogeneous (10 degenerative, 2 deformity, 1 trauma, 1 tumor). Patients were older (mean 64.6 years) and had longer constructs (3-15 levels), many extending to the upper thoracic spine. Despite these higher-risk features, only two patients (14.3%) developed PJK, and none progressed to PJF or required reoperation. Overall, the incidence of PJK/PJF was halved in the band group compared with the non-band group (14.3% vs. 30%). Importantly, while three patients in the non-band group required revision due to junctional failure, no patient in the band group underwent reoperation for PJF. Conclusion: The use of sublaminar bands at UIV+ was associated with lower incidence of PJK/PJF and complete absence of reoperations for PJF, even in older patients with more extensive constructs. These findings support sublaminar bands as a potential soft landing strategy to reduce proximal stress and prevent catastrophic failure. Although limited by small sample size, the clear contrast in outcomes highlights the protective potential of this technique. Larger prospective studies are warranted to confirm these results and define the role of sublaminar bands in long spinal fusions surgery.

ID: AGS26/3316

A413: Perioperative visual loss in spine surgery: risk factors, pathophysiology, and management strategies

Sebastiano Mazzotta 1 , Giulio Bonomo 1 , Francesco Certo 1 , Giuseppe Barbagallo 1

1 A.O.U. Policlinico G.Rodolico - San Marco, Catania, Italy

Introduction: Perioperative visual loss (POVL) is a rare but serious complication associated with spine surgery under general anesthesia, often linked to ischemic optic neuropathy (ION), central retinal artery occlusion (CRAO), and cortical blindness (CB). This study aims to explore the pathophysiology, risk factors, and management strategies of POVL, with a focus on a clinical case. Material and Methods: We reviewed 38 studies on POVL, including case reports, case series, and cohort studies. A clinical case of a 59-year-old male with lumbar stenosis undergoing posterior hemilaminectomy is presented. Preoperative, intraoperative, and postoperative data were collected, including imaging, ophthalmologic assessments, and surgical parameters. A literature review was conducted using PubMed and Scopus to identify additional studies on POVL. Results: The literature review included 38 studies with a total of 72 patients (35 males and 37 females, mean age 49.9 years). Surgical sites were predominantly lumbar (54.1%), followed by cervical (12.7%), thoracic (8.5%), and thoracolumbar (11.4%), with the most common pathologies being degenerative diseases (23/72) and scoliosis (8/72). The majority of patients (87.1%) underwent surgery in the prone position, with a median operative time of 300 minutes and an estimated blood loss of 1400 mL. The identified POVL subtypes were 25 cases of PION (36.8%), 20 CRAO (30.9%), 13 ION (19.1%), 6 CB (8.8%), and 3 AION (4.4%), with early onset in 67.3% of cases. Visual loss was unilateral in 76.3% and bilateral in 23.7%. Visual recovery was absent in 64.1% of cases, partial in 9, and complete in 6. Statistical analysis showed significant associations between POVL and longer operative duration, prolonged prone positioning, higher blood loss, postoperative anemia, and elevated BMI. In particular, PION cases had the longest operative and prone times, while postoperative anemia was present in 92.3% of patients. Bilateral visual loss was associated with longer surgical times and higher BMI. Periorbital edema, observed in 65.1% of cases, correlated with extended operative time, prone positioning, and increased BMI. Conclusion: POVL is a rare but serious complication in spinal surgery, most commonly presenting as PION, and is strongly associated with prolonged operative time, prone positioning, blood loss, anemia, higher BMI, and peri-orbital edema. Visual recovery is often limited, yet our illustrative cases hawed reversible and progressive improvement, suggesting that not all cases follow classical ischemic mechanisms. Early recognition and careful intraoperative management are essential, while further studies are needed to refine risk stratification and preventive strategies.

ID: AGS26/665

A414: Propensity matched analysis of anticoagulation after spine surgery: more complications and longer stay without thromboembolic gain

Reme Arhewoh 1 , Elias Schwartz 1 , Aayush Unadkat 1 , Jack Eichman 1 , Abdel Diab 1 , Joshua Piche 1 , Rakesh Patel 1 , Ilyas Aleem 1

1 University of Michigan, Ann Arbor, United States

Introduction: The routine use of anticoagulation (AC) in elective spine surgery remains debated given the competing risks of venous thromboembolism (VTE)/pulmonary embolism (PE) and epidural hematoma. Prior retrospective studies are limited by unclear anticoagulation criteria and inherent biases. Our institution's contrasting anticoagulation protocols between Orthopedic and Neurosurgery services created a unique opportunity to evaluate outcomes using prospectively collected data. Material and Methods: A prospectively collected multicenter database was queried for all patients treated at an academic tertiary care center from 2013–2023. After excluding patients with prior blood clots, hypercoagulability, or unclear anticoagulation status, 5,014 patients were identified. Propensity score matching (PSM) was performed based on age, gender, BMI, ASA grade, and diabetes history. Composite outcomes for overall and neurologic complications were created. T-tests and Chi-square tests were used for continuous and categorical variables, respectively. Binary logistic regression was employed with Firth correction for rare events. Results: The final cohort included 2,420 patients equally divided between treatment (n = 1,210) and control (n = 1,210) groups with balanced demographics. Cohorts were approximately 56% male, 78% white, with a mean age of 60. Anticoagulated patients demonstrated higher rates of deep vein thrombosis (1.2% vs. 0.3%, p = 0.011) and wound dehiscence (2.3% vs. 1.0%, p = 0.011). New neurological deficits were more prevalent in the AC group, including weakness (12.9% vs. 8.5%, p < 0.001) and myelopathy (0.5% vs. 0%, p = 0.031). The AC group also experienced higher rates of neurogenic bowel/bladder dysfunction (p < 0.001) and urinary retention (5.6% vs. 2.9%, p = 0.002). Overall complications were more frequent in AC patients (36.9% vs. 32.4%, p = 0.021), and length of stay was prolonged (3.9 vs. 2.0 days, p < 0.001), though 90-day readmission, reoperation, and mortality were comparable. Logistic regression identified key predictors of postoperative complications. For any complication, longer time to discharge (OR 1.222, p < 0.001), coronary artery disease (OR 1.465, p = 0.018), and increasing age (OR 0.982, p < 0.001) were significant, while lumbar decompression was protective (OR 0.724, p = 0.023). For neurologic complications, time to discharge remained a strong predictor (OR 1.203, p < 0.001), with lumbar fusion (OR 0.422, p < 0.001) and lumbar decompression (OR 0.338, p < 0.001) both protective. Anticoagulant use was associated with increased DVT risk (OR 4.16, p = 0.028) but not PE (p = 0.779). Conclusion: Utilizing dichotomous anticoagulation prescribing practices and prospectively collected outcomes, this study demonstrates that routine early anticoagulation is associated with increased complications, largely neurological. Importantly, routine anticoagulation did not reduce DVT or PE incidence, underscoring the need for individualized risk assessment in elective spine surgery. Prolonged hospital stay emerged as a key predictor of adverse events, highlighting the importance of timely and safe discharge.

OP47: Decision Making for Lumbar Degenerative Surgery

ID: AGS26/1098

A415: Lumbar microdiscectomy and post-operative activity restrictions: a single blinded randomised controlled trial

Jack Roadley 1,2 , Myron Rogers 3 , Andrew Danks 4 , Jeremy Kam 4 , Idrees Sher 4 , Mendel Castlebaum-Kirschnow 4 , Scott Ayton 5 , Kylie Fryer 4 , Phillipa Risbey 3 , Tony Goldschlager 3,4

1 University of Queensland, Brisbane, Australia

2 Princess Alexandria Hospital, Brisbane, Australia

3 Cabrini Hospital, Melbourne, Australia

4 Monash Medical Centre, Melbourne, Australia

5 The Florey Institute, Melbourne, Australia

Introduction: Lumbar microdiscectomy is the most common spinal procedure in the United States. While effective in relieving radicular pain in 75-80% of cases, there is significant variability in post-operative activity restrictions prescribed by surgeons. Many surgeons impose restrictions on sitting, bending, and lifting to prevent recurrent disc herniation. However, limited evidence exists to guide these recommendations. This study aims to provide objective data on whether post-operative activity restrictions impact recovery outcomes following lumbar microdiscectomy. Material and Methods: This single-blinded randomized controlled trial was conducted across three centers in Melbourne, Australia. Patients aged 18-75 years undergoing unilateral microdiscectomy for lumbar radiculopathy with concordant MRI findings were enrolled. Patients were randomized in a 1:1 ratio to either a restricted group (n = 101) or unrestricted group (n = 99). The restricted group received specific instructions limiting sitting to 15-30 minutes per two-hour period, avoiding lifting over 5kg, refraining from heavy household tasks, and avoiding strenuous sexual activity for two weeks. The unrestricted group was encouraged to resume normal activities as tolerated. All patients wore an electronic activity monitor (Activ8) for the first post-operative month. Patients were assessed pre-operatively and at day 1, 1, 3, 6, and 12 months post-surgery using visual analogue scales (VAS) for leg and back pain, Oswestry Disability Index (ODI), and quality of life questionnaires (SF-12, EQ-5D). The primary endpoint was a composite of reduced pain scores, functional improvement, and absence of further interventions at 12 months. Secondary outcomes included reherniation rates, individual functional measures, and time to return to work. Results: At one year, the primary composite endpoint showed no significant difference between restricted and unrestricted groups (45.6% vs 47.1%, p > 0.99). Individual components of the primary endpoint, including VAS back improvement (69.1% vs 63.2%, p = 0.587), VAS leg pain improvement (72.1% vs 73.5%, p > 0.99), ODI improvement (66.2% vs 80.9%, p = 0.08), and need for reoperation (2.9% vs 5.5%, p = 0.68) showed no significant differences. Secondary outcomes were also similar between groups. Reherniation occurred in 10.1% of restricted patients versus 14.1% in the unrestricted group (p = 0.61). Both groups showed significant improvements in all functional measures from baseline to 12 months. Electronic activity monitoring revealed poor adherence to restrictions, with only 10% of patients in the restricted group adhering to sitting limitations. Notably, sitting duration did not differ significantly between groups (4,102 vs 4,140 minutes/week, p = 0.89), nor did time spent in other activities (lying, standing, walking, or running). Conclusion: This randomized trial provides robust evidence that liberalizing post-operative restrictions following lumbar microdiscectomy does not negatively impact outcomes. No significant differences were found in pain reduction, functional recovery, or complication rates between either arm. The study challenges conventional concerns about activities like sitting, as objective monitoring showed similar activity patterns regardless of assigned group, with poor adherence to restrictions in the restricted group. The findings support a patient-driven approach to recovery guided by comfort and pain levels rather than rigid restrictions. This evidence may help standardize post-operative care guidelines, potentially reducing unnecessary limitations, facilitating faster return to normal activities, and improving patient satisfaction without compromising safety or outcomes.

ID: AGS26/1155

A416: Postoperative outcomes and complications after single vs. multilevel lateral lumbar interbody fusion: a retrospective cohort study

Puya Alikhani 1 , Nikhil Dholaria 2 , Diego Soto Rubio 1 , John Lingam 1 , Mohsen Rostami 1 , Dana Saleh 1 , Ahmad Alkadri 1

1 University of South Florida, Tampa, United States

2 University of Arizona, Phoenix, United States

Introduction: Lateral lumbar interbody fusion (LLIF) is often taken into consideration for patients with deformity and degenerative conditions, yet the relationship between the number of fused levels and outcomes after surgery is still lacking complete understanding. Material and Methods: We retrospectively analyzed 546 patients who underwent LLIF procedures, categorizing cases as single-level, two-level, or ≥ 3-level fusions. Length of hospital stay (LOS) was evaluated using one-way ANOVA and Kruskal-Wallis tests. Categorical outcomes-including overall complications, infection, deep venous thrombosis (DVT), pulmonary embolism (PE), myocardial infarction (MI), adjacent segment disease (ASD), revision surgery, and subsidence-were analyzed using Chi-square or Fisher’s exact tests. Multivariable analyses (n = 267) were performed using linear models, multinomial regression for postoperative Roussouly classification, and least squares regression for LOS with demographic, surgical, and implant characteristic covariates. Statistical significance was defined as p < 0.05 and analyses were completed on Python 4.7. Results: Increasing LLIF levels strongly predicted adverse outcomes. Multilevel LLIF (≥ 3 levels) was independently associated with longer LOS (β = 3.34 days, 95% CI 1.69-4.99, p = 0.0001), higher rates of overall postoperative complications (OR 3.14, 95% CI 1.35-7.30, p = 0.008), and increased infection risk (OR 5.55, 95% CI 1.40-22.00, p = 0.015). Subsidence risk increased with age (OR 1.11 per year, 95% CI 1.06-1.17, p = 0.0001). Supplementary anterior column release and prior spine surgery also contributed to prolonged hospitalization. No significant predictors were identified for PI-LL mismatch, pelvic incidence > 50°, postoperative Roussouly classification, ASD, or revision surgery. Conclusion: Multilevel LLIF is associated with increased hospital stay, increased overall postoperative complications, and increased infection rates, with age being a predictor of subsidence. Our findings emphasize the need for personalized perioperative strategies and meticulous risk stratification in patients who undergo multilevel LLIF.

ID: AGS26/3462

A417: Delayed initiation of physical therapy following lumbar fusion can be associated with increased risk of mechanical complications, nonunion, and surgical revision: a 5-year matched cohort analysis

Joshua Wang 1 , Allen Wang 1 , Pouya Alijanipour 1

1 University of Texas Medical Branch, Galveston, United States

Introduction: Current guidelines suggest initiating postoperative physical therapy (PT) within 6 to 12 weeks after lumbar fusion. However, the long-term effects of not initiating PT within this window remain underexplored. We aimed to study the impact of delayed or absent PT on mechanical complications, nonunion, hardware failure, opioid use disorder, and need for surgical revision after lumbar fusion. Material and Methods: A retrospective matched cohort study was conducted using a large, multi-institutional electronic health record database. Adult patients who underwent lumbar spinal fusion were stratified based on whether they received PT within the first 3 months postoperatively. Patients were propensity score-matched for age, sex, comorbidities, and other baseline variables. Primary outcomes assessed at 1, 3, and 5 years included mechanical complications, nonunion, hardware failure, opioid use disorder, and surgical revision. Risk ratios (RR) with 95% confidence intervals (CI) were calculated. Statistical significance was defined as p < 0.05. Results: At 1 year, patients who did not receive PT within 3 months had significantly higher risks of mechanical complications (RR 1.41, 95% CI 1.01-1.95, p = 0.040), nonunion (RR 2.79, 95% CI 2.20-3.53, p < 0.001), surgical repair (RR 2.11, 95% CI 1.63-2.73, p < 0.001), opioid use disorder (RR 1.65, 95% CI 1.19-2.28, p = 0.002), and hardware failure (RR 1.41, 95% CI 1.01-1.95, p = 0.040). At 3 years, elevated risks persisted for mechanical complications (RR 1.43, p = 0.005), nonunion (RR 2.19, p < 0.001), surgical repair (RR 1.70, p < 0.001), and hardware failure (RR 1.43, p = 0.005), though opioid use disorder was no longer significantly different. By 5 years, patients without early PT continued to show increased risks of mechanical complications (RR 1.48, p = 0.001), nonunion (RR 2.09, p < 0.001), surgical repair (RR 1.56, p < 0.001), and hardware failure (RR 1.48, p = 0.001). No significant difference in opioid use disorder was observed at this time point. Conclusion: Failure to initiate physical therapy within 3 months following lumbar fusion was associated with significantly increased risks of nonunion, hardware failure, mechanical complications, and surgical reintervention in this cohort over a 5-year period. These findings may underscore the critical role of timely rehabilitation and support current guidelines recommending early PT to optimize fusion outcomes and reduce complications.

ID: AGS26/2680

A418: Percutaneous minimally invasive lumbar decompression versus open decompression for lumbar spinal stenosis: a multicenter randomized trial using MRI endpoints

Kingsley Chin 1 , Erik Spayde 2 , William M. Costigan 3 , Sandra Thompson 4 , Mark Jones 5 , Ajay Yeddu 6 , Rajmani Krishnan 7 , Vito Lore 1 , Hope Estevez 1 , Suraj Yeddu 1 , Sajni Yeddu 1 , Shreyas Krishnan 8 , Sukanya Chebrolu 1 , Jason A. Seale 1 , Chukwunonso Ilogu 1 , Jacob Ryu 9

1 LESS Society, LESS Clinic, Fort Lauderdale, United States

2 St. Charles Spine Institute, Thousand Oaks, CA, United States

3 Congress Orthopaedic Associates, Pasadena, United States

4 The Pain Center, Boise, Idaho, United States

5 Pain Medicine of the Sounth, Knoxville, United States

6 Desert Interventional Spine Consultants, Gilbert, United States

7 Island Interventional Pain Management, Commack, NY, United States

8 LESS Society, Fort Lauderdale, Florida, United States

9 Midwestern University Chicago - College of Osteopathic Medicine, Chicago, United States

Introduction: Lumbar spinal stenosis is common in older adults and presents with neurogenic claudication and back and leg pain. Open decompression reliably enlarges the canal and relieves symptoms but requires dissection of posterior musculature. Percutaneous minimally invasive lumbar decompression targets the hypertrophied ligamentum flavum using a percutaneous approach under fluoroscopy. Its ability to achieve objective decompression has been debated. This multicenter randomized trial tested whether percutaneous decompression achieves radiographic and clinical decompression compared with open midline decompression. Materials and Methods: Adults with magnetic resonance imaging confirmed central stenosis and a positive nerve root sedimentation sign were randomized 5:1 to percutaneous decompression or open decompression across multiple centers. Imaging endpoints were sedimentation sign reversal, ligamentum flavum thickness, and dural sac cross sectional dimensions. MRI studies were read by blinded radiologists. Reliability was assessed using intraclass correlation coefficient and kappa statistics. Clinical assessments used standard pain and function scales collected longitudinally. Between group differences were analyzed using appropriate parametric or nonparametric tests with significance at p < 0.05. Results: Postoperative MRI after percutaneous decompression demonstrated no measurable enlargement of the canal and no change in ligamentum flavum thickness. The sedimentation sign remained positive in all percutaneous cases. Open decompression achieved more than ninety percent resection of the ligamentum flavum, a marked increase in dural sac dimensions, and universal sedimentation sign reversal, p < 0.001 for between group comparisons. MRI readers showed excellent reliability (intraclass correlation coefficient 0.98; kappa 1.0). Clinical measures improved in the open group in parallel with radiographic decompression. No unanticipated device related adverse events were identified in either arm within the analysis window. Conclusion: Percutaneous minimally invasive decompression failed to produce objective decompression on MRI, while open decompression reversed the sedimentation sign and enlarged the canal. The concordance of imaging and clinical outcomes supports open decompression as the effective intervention for symptomatic lumbar spinal stenosis when surgery is indicated.

ID: AGS26/721

A419: The Suspected Cauda Equina Syndrome Score (SuCESS): development and validation of a clinical triage tool

Elie Najjar 1 , Michael Grevitt 1

1 Queen’s Medical Centre, Nottingham University Hospitals, Nottingham, United Kingdom

Background: Cauda equina syndrome (CES) is a rare but critical spinal emergency. However, over 80% of urgent MRIs for suspected CES show no cauda equina compression. This diagnostic uncertainty places strain on imaging services and delays care. We aimed to develop and validate a clinical triage tool-SuCESS (Suspected Cauda Equina Syndrome Score)-to safely rule out CES and improve triage accuracy. Methods: We retrospectively developed the SuCESS score using 259 patients referred to a UK tertiary spinal unit between 2016 and 2018. Six clinical variables-saddle anaesthesia, reduced perianal sensation, motor weakness, bilateral sciatica, bladder post-void residual volume (PVR), and urethral catheterisation-were identified via multivariable logistic regression, random forest, and XGBoost, and combined into an eight-point scoring system. Penalised regression was applied to minimise overfitting. External validation was performed using two temporally independent cohorts from 2020 and 2023 (n = 444). Primary outcomes were sensitivity and negative predictive value (NPV) for surgically confirmed CES. Model performance was assessed using area under the curve (AUC), Brier scores, calibration plots, and decision curve analysis. Results: At a threshold of ≥ 3.0, SuCESS achieved 100% sensitivity and 100% NPV across all cohorts, correctly identifying all 64 cases of confirmed CES. Specificity increased from 37% in the development cohort to 54% in validation. AUC improved from 0.70 to 0.85, and Brier scores improved from 0.26 to 0.13. Use of the score might have safely deferred 38.7% of urgent MRIs, based solely on bedside clinical data. Conclusion: SuCESS is the first validated triage tool for CES to demonstrate perfect sensitivity. It offers a pragmatic and safe framework for reducing unnecessary MRI requests and improving diagnostic accuracy in urgent spinal referrals.

ID: AGS26/2840

A420: Comparison of predictive value for cage subsidence between MRI-based endplate bone quality and vertebral bone quality scores following transforaminal lumbar interbody fusion: a retrospective propensity-matched study

Youwei Ai 1 , Hong Ding 1 , Qian Chen 2 , Limin Liu 1 , Yueming Song 1

1 West China Hospital, Sichuan University, Chengdu, China , 2 Affiliated Hospital of North Sichuan Medical College, Nanchong, China

Introduction: Cage subsidence after TLIF is closely related to low vertebral bone mineral density (BMD). Conventional BMD assessment methods such as DEXA and QCT have limitations due to radiation exposure and variable accuracy. Recently, MRI-based vertebral bone quality (VBQ) and endplate bone quality (EBQ) scores have been proposed as radiation-free measures of bone quality, both associated with cage subsidence. Theoretically, EBQ may have greater predictive value than VBQ, as it directly reflects the mechanical properties of the bone-implant interface. However, no direct comparison between the two has been reported. This study aimed to investigate whether EBQ has superior predictive value for cage subsidence after TLIF compared with VBQ, using a propensity score-matched retrospective analysis. Material and Methods: We retrospectively reviewed consecutive data from 346 patients who underwent TLIF at our institution between 2014 and 2020. Patients with and without cage subsidence were matched using propensity score matching, yielding 52 cases in each group. Conditional logistic regression was performed to identify radiographic factors associated with cage subsidence. Predictive performance of VBQ and EBQ was evaluated using receiver operating characteristic (ROC) curve analysis and area under the curve (AUC). Results: n the matched cohort, patients with cage subsidence had significantly higher VBQ scores (3.7 vs. 3.1, p < 0.001) and EBQ scores (5.0 vs. 4.3, p < 0.001) compared with those without subsidence. Regression analysis showed that higher VBQ scores (OR 4.557, 95% CI 1.076-19.291, p = 0.039) and higher EBQ scores (OR 5.396, 95% CI 1.158-25.146, p = 0.032) were associated with increased risk of subsidence. For predicting subsidence, the cutoff for VBQ was 3.4 (AUC = 0.799, sensitivity 84.6%, specificity 69.2%), while the cutoff for EBQ was 4.7 (AUC = 0.829, sensitivity 76.9%, specificity 82.7%). Conclusion: Both higher VBQ and EBQ scores are associated with an increased risk of cage subsidence after TLIF. EBQ demonstrated superior predictive performance, particularly due to its higher specificity, and may serve as a more reliable tool for preoperative risk assessment.

ID: AGS26/419

A421: Discoplasty and decompression as a novel approach for advanced lumbar degeneration and spinal stenosis: a case series

Valentino Latallade 1 , Gonzalo Kido 1 , ivan huespe 1 , Matias Petracchi 1 , Marcelo Gruenberg 1 , Carlos Sola 1

1 Hospital Italiano de Buenos Aires, Ciudad Autónoma de Buenos Aires, Argentina

Introduction: The aim of this study is to describe a novel, minimally invasive technique-percutaneous discoplasty combined with decompression-for the management of low back pain and lower limb symptoms in elderly patients with advanced degenerative disease. The procedure provides spinal stability through the injection of bone cement into a severely degenerated pneumodisc, while simultaneously decompressing the canal in patients with lumbar spinal stenosis. To the best of our knowledge, this combined approach has not previously been reported in the literature. Material and Methods: This study included twenty patients with advanced lumbar degenerative disc disease and compression, with or without degenerative scoliosis, who underwent percutaneous discoplasty combined with decompression. All patients had a minimum follow-up of two years. The variables assessed comprised clinical outcomes, including the Visual Analogue Scale (VAS) for pain and the Oswestry Disability Index (ODI), as well as radiological parameters such as lumbar lordosis and Cobb angle. Additional variables included hospital length of stay and procedure-related complications. Results: With a mean follow-up of 6 ± 2.1 years and a mean age of 75 years (range 60-82), significant pain reduction (VAS: preoperative 8.26 ± 1.24; postoperative 3.42 ± 3.02) and improvement in the ODI (preoperative 72.3 ± 11.85; postoperative 22.4 ± 23.71) were observed with partial correction of radiological parameters (5° mean increase in segmental lumbar lordosis and decrease in segmental Cobb angle). No significant difference was found in the improvement of VAS and ODI scores when comparing scoliotic and non-scoliotic patients. Mean surgical time was 125 minutes, and 84% of patients were discharged during the initial 48 hours postoperatively. Four patients (21%) developed complications. Two required subsequent decompression and conventional fusion due to persistent symptoms, one presented a cerebrospinal fluid fistula requiring reoperation, and one developed a deep infection managed with surgical debridement and intravenous antibiotics. Conclusion: Percutaneous discoplasty combined with decompression appears to be an effective option for advanced degenerative disc disease and spinal canal stenosis, improving pain and functionality, and may represent a potential alternative to conventional fusion and decompression in selected cases.

ID: AGS26/2767

A423: Patients receiving GLP-1 agonists have reduced odds of pseudoarthrosis and revision following lumbar fusion surgery

Manaav Mehta 1 , Abhinav Sharma 2 , Ryan Beyer 2 , Vivian Chen 2 , Michael Kim 2 , Paramveer Birring 2 , Hao-Hua Wu 2 , Sohaib Hashmi 2 , Yu-po Lee 2 , Nitin Bhatia 2 , Don Park 2

1 University of Massachusetts Chan Medical School, Worcester, United States

2 University of California Irvine, Orange, United States

Introduction: In the past two decades, the frequency of lumbar spine fusion surgeries has risen markedly. This increase has coincided with a rise in postoperative complications, notably pseudoarthrosis-a failure in spinal fusion manifesting as recurring pain and mechanical instability. The association between glucagon-like peptide-1 (GLP-1) agonist use and complications following lumbar fusion has yet to be explored. The goal of this study is to examine the association between GLP-1 agonist usage and the incidence of pseudoarthrosis, adjacent segment disease, revision surgery, and other perioperative complications following lumbar fusion surgery. Material and Methods: TriNetX, a global health research network, was queried from 2004-2024 for patients undergoing open lumbar fusion using a posterior/transforaminal (CPT 22633, 22630, 22612), anterior (CPT 22558), or lateral approach (CPT 22533), and who had documented GLP-1 agonist usage within 1-year post-operation. Analysis included chi-squared tests with a control cohort propensity score matched for age, sex, race, smoking, diabetes, BMI, and uremia who did not have post-operative GLP-1 agonist usage. Outcomes of interest were DVT, emergency department (ED) visits, inpatient hospitalization, opioid abuse, PE, sepsis, surgical site infection, and wound complications within 90 days, and adjacent segment disease, implant failure, pseudoarthrosis, and revision within 2 years. Subgroup analysis based on surgical approach was conducted. Results: A total of 2,972 patients who received GLP-1 were compared to propensity matched controls. After matching, patients were 61.5 ± 11 years old, 39% male, 70% white, 81% diabetic, and 69% obese. GLP-1 agonist use within 1-year post-operation was significantly associated with reduced odds of DVT (OR 0.61, CI 0.44-0.83), ED visits (OR 0.78, CI 0.62-0.97), inpatient hospitalization (OR 0.22, CI 0.20-0.25), sepsis (OR 0.47, CI 0.31-0.71), post-operative SSI (OR 0.51, CI 0.38-0.68), and wound complications (OR 0.54, CI 0.41-0.72) within 90 days, and pseudoarthrosis (OR 0.44, CI 0.38-0.50) and revision surgery (OR 0.78, CI 0.64-0.95) within 2 years post-operation. In the posterior approach subcohort, GLP-1 agonist use was significantly associated with reduced odds of DVT (OR 0.58, CI 0.42-0.80), ED visits (OR 0.64, CI 0.55-0.75), inpatient hospitalization (OR 0.22, CI 0.19-0.25), sepsis (OR 0.31, CI 0.20-0.47), post-operative SSI (OR 0.46, CI 0.34-0.62), wound complications (OR 0.56, CI 0.42-0.76) within 90 days post-operation, implant failure (OR 0.58, CI 0.36-0.94) and pseudoarthrosis (OR 0.51, CI 0.44-0.58) within 2 years post-operation. In the anterior approach subcohort, GLP-1 agonist use was significantly associated with reduced odds of DVT (OR 0.43, CI 0.20-0.90), ED visits (OR 0.57, CI 0.40-0.81), inpatient hospitalization (OR 0.21, CI 0.16-0.28), post-operative SSI (OR 0.29, CI 0.14-0.60), wound complications (OR 0.45, CI 0.22-0.92) within 90 days, pseudoarthrosis (OR 0.65, CI 0.47-0.88) and revision surgery (OR 0.33, CI 0.22-0.51) within 2 years post-operation. The study population was insufficient to assess lateral approaches independently. Conclusion: Lumbar fusion patients prescribed GLP-1 agonists in the post-operative period experienced lower rates of pseudoarthrosis, revision surgery, postoperative infections, and overall hospitalizations compared to those who did not receive GLP-1 agonists. GLP-1 agonists may offer a therapeutic benefit in improving spinal fusion outcomes, and further prospective research is needed to validate these findings and investigate the underlying mechanisms.

OP48: Outcomes of Spine Trauma Management

ID: AGS26/3157

A424: Timing of thromboprophylaxis in acute spinal cord injury patients: a TQIP study

Aazad Abbas 1 , Jay Toor 2 , Albert Yee 3 , Cari Whyne 4 , Stephen Lewis 5 , Michael Fehlings 5 , Jetan Badhiwala 3

1 University of Toronto, Division of Orthopaedic Surgery, Department of Surgery, Toronto, Canada

2 University of Manitoba, Division of Spine Surgery, Winnipeg, Canada

3 Division of Spine Surgery, Sunnybrook Hospital, University of Toronto, Toronto, Canada

4 Sunnybrook Research Institute, University of Toronto, Toronto, Canada

5 Division of Spine Surgery, Toronto Western Hospital, University of Toronto, Toronto, Canada

Introduction: Optimal timing of venous thromboembolism prophylaxis (VTEp) after acute spinal cord injury (SCI) remains is debated, with most guidelines recommending VTEp be given within 72 hours. However, these recommendations are largely consensus-based and are lacking in primary large scale clinical data. As such, the goal of this study was to evaluate an ideal timeframe for delivery of thromboprophylaxis in the setting of acute SCI. Material and Methods: The American College of Surgeons (ACS) Trauma Quality Improvement Program (TQIP) was used to identify adults with acute SCI due to blunt trauma undergoing operative treatment within 24 hours from 2017-2022. Inclusion criteria were patients receiving LMWH or UFH within five days postoperatively. Exclusion criteria were age < 16, non-blunt trauma, low-volume centers (< 10 cases/year), missing data, non-LMWH/UFH prophylaxis, VTEp before surgery or >5 days postop, early care withdrawal, and death within five days. Restricted cubic splines modeled risk-adjusted outcome probabilities versus hours from surgery to VTEp to identify an inflection point, which defined early versus late VTEp for propensity score-matched comparisons. Covariates for propensity matching included patient, injury, treatment, and hospital characteristics. Primary outcome was in-hospital VTE; secondary outcomes were return to operating room, in-hospital mortality, and hospital length of stay (LOS). Effect size was calculated odds ratio (OR) with associated 95% confidence interval (CI). Significance level was set at p = 0.05. Results: A total of 15,960 patients across 511 trauma centres met inclusion criteria (mean age 48.6 years, 23.9% female). Spline analysis indicated increasing risk after approximately 24-48 hours, motivating two cutoffs for early versus delayed thromboprophylaxis: 1) 24-hours (postoperative day 0) and 2) 48-hours (postoperative day 1). In propensity-matched cohorts for the postoperative day 0 cutoff (i.e. early < 24 hours; late ≥ 24 hours) (N = 2,807 per group), there was no difference in rates of VTE, return to operating room, and in-hospital mortality. However, there was a borderline significant difference in length of stay (15.1 vs 16.2 days; p = 0.011; OR 3.16 [95% CI: 1.31, 7.66]). In propensity-matched cohorts for the postoperative day 1 cutoff (i.e. early < 48 hours; late ≥ 48 hours) (N = 6,867 per group), early VTEp was associated with lower VTE (4.7% vs 5.9%; p = 0.002; OR 1.26 [95% CI: 1.09, 1.46]), fewer returns to operating room (2.2% vs 2.9%; p = 0.004; OR 1.38 [95% CI 1.11, 1.71]), lower mortality (4.0% vs 4.9%; p = 0.003; OR 1.28 [95% CI: 1.09, 1.51]), and shorter length of stay (15.8 vs 16.9 days; p < 0.001; OR 2.89 [95% CI: 1.6; 5.2]). Conclusion: In acute SCI patients undergoing early surgery, initiating postoperative VTE prophylaxis within 48 hours was associated with reduced VTE, return to OR, mortality, and hospital LOS, supporting a < 48-hour target i.e. postoperative day 1 as a clinically meaningful threshold for early prophylaxis. These data from a large, multicenter cohort provide outcome-based timing guidance beyond consensus opinion and suggest that delaying VTEp beyond postop day 1 may worsen postoperative outcomes.

ID: AGS26/2022

A425: Minimal clinically important difference (MCID) and long-term prospective validation of the AO Spine PROST (Patient Reported Outcome Spine Trauma)

Said Sadiqi 1,2 , Charlotte Dandurand 3 , Richard Bransford 4 , Andrei Joaquim 5 , Ulrich Spiegl 6 , Alex Vaccaro 7 , F. Cumhur Oner 2

1 St. Antonius Hospital, Utrecht, The Netherlands

2 University Medical Center Utrecht, Utrecht, The Netherlands

3 University of British Columbia, Vancouver, Canada

4 University of Washington School of Medicine, Seattle, United States

5 State University of Campinas, Campinas-Sao Paulo, Brazil

6 University of Leipzig, Leipzig, Germany

7 Rothman Institute, Thomas Jefferson University, Philadelphia, United States

Introduction: The AO Spine PROST was developed as the first comprehensive, patient-centered instrument specifically designed for spine trauma. Although prior validation has confirmed its reliability and cross-cultural utility, Minimal Clinically Important Difference (MCID) of this outcome instrument has not yet been established. The goal of this study was to determine the MCID of PROST as well as conducting a long-term prospective validation. Materials and Methods: This study utilized the data from a prospective observational international multicenter cohort study. Adults aged 18-65 acute thoracolumbar (TL) burst fractures without neurologic deficits were enrolled across multiple global regions, with follow-up extending to two years post-injury. Next to PROST, patients completed the Oswestry Disability Index (ODI), EuroQol-5D (EQ-5D), and Pain NRS. Characteristics were analyzed using descriptive statistics. MCID for PROST was assessed with distribution-based approach using the standard deviation (SD) of change in scores. Floor and ceiling effects were also assessed. Internal consistency was evaluated by Cronbach’s alpha, item-total correlation coefficient (ITCC) and pairwise Spearman correlation. Construct validity was determined through Pearson correlations (r s ) with the ODI, EQ-5D, and Pain NRS. Responsiveness was assessed by calculating effect sizes (ES) and standardized response mean (SRM). Results: A total of 93 patients were included. The MCID for a moderate change (0.5*SD) in the PROST score was 10.6. No floor or ceiling effects were identified. Internal consistency was good with Cronbach’s α ranging from 0.9-1.0 and acceptable ITCC. Construct validity showed strong to very strong correlation between PROST and ODI (r s = -0.67 to -0.89; p < 0.001), while correlations with the EQ-5D were weak to very weak (r s = -0.29 to 0.05; p > .005), except at the one-year follow-up where statistical significance was reached. No consistent pattern was found in correlations with the Pain NRS. Responsiveness was very good, with ES = 3.2 and SRM = 3.1 (p < .001). Conclusion: The AO Spine PROST identified an MCID of 10.6 as indicative of a moderate clinically meaningful change using a distribution-based approach. The instrument also showed strong internal consistency, construct validity, and excellent responsiveness in long-term follow-up, supporting its utility in evaluating outcomes in spine trauma patients.

ID: AGS26/2735

A426: Retrospective study on treatment of osteoporotic fractures validating DGOU classification: a MIOT experience

Ragul Muruganandhan 1 , Charanjit Singh Dhillon 1 , Mukhesh Thangavel 1 , Tarun Roy 1

1 MIOT Hospitals Chennai, Chennai, India

Introduction: Osteoporotic fractures represent a growing healthcare challenge with aging populations. Majority of those cases remain undiagnosed, only 1/3 rd of those fractures are clinically diagnosed. Until a decade ago every surgeon have their own protocol as because no world wide accepted treatment protocol available. The German Society for Orthopaedics and Trauma (DGOU) classification system provides a standardized approach to categorize these fractures proposed by a group of doctors in 2018. It has been validated by international forum of 320 spine surgeons by classifying 27 cases and published by AO forum in 2024. There is no validation of this DGOU classification done in the Indian sub-continent. This study aims to validate DGOU classification in Indian patients and see whether our method of treatment of osteoporotic fractures in the past is concomitant with the DGOU classification system through retrospective analysis of 372 osteoporotic fracture cases treated at MIOT Hospital for a period of 9 years. (2016-2024). Aim: Validation of DGOU Classification for treatment of Indian Osteoporotic fractures: a retrospective analysis of MIOT hospital cases. Objectives: 1. Is our treatment in the past relevant and concomitant with DGOU classification. 2. To assess the applicability of DGOU classification in an Indian patient population. 3. To identify limitations of the DGOU system if any. Methodology: Study Design: Retrospective cohort study. Sample size: 372 (2016-2024) Institutional ethical committee approval obtained Inclusion Criteria: 1) Osteoporotic vertebral fractures treated at MIOT hospital (2016-2024). 2) BMD < -2.5. Exclusion Criteria: 1)Pathological fractures (malignancy-related). 2) High-energy trauma fractures. 3) Incomplete medical records. 4) Kummels lesion. Pre op investigations of 372 patients were collected. One spine Surgeon classified the fractures twice at 4 weeks interval and its reliability measured by intra-class correlation coefficient for intra-observer reliability. Two spine Surgeons classified fractures individually and its reliability was measured by Cohen's kappa coefficient for inter-observer reliability. Any disputes arising were nullified by chief spine surgeon. Statistical Analysis: SPSS software was used, and p value of <0.05 was considered significant. Statistical tools used were: • Intra class correlation coefficient - For intra observer reliability, • Cohens kappa - For inter observer reliability, • Chi square test - to identify the correlation between DGOU and MIOT protocol in classification and treatment option. Results: • Intra observer correlation coefficient 0.75 (good reliability), • Cohens kappa for inter observer reliability 0.816 (Almost perfect), • Chi square test to access the association between DGOU classification and MIOT categorization shows p value of <0.001 - significant association, • Chi square test implies there is significant association between DGOU types and their treatment protocol we had administered with p value of < 0.001. Conclusion: Our results indicate that treatments th administered in the pre-DGOU era in our hospital were largely consistent with those recommended by the DGOU, showing a strong association . DGOU fracture typing is reproducible both over time and between raters.

ID: AGS26/2740

A427: Is time really spine? A systematic review and meta-analysis of early versus delayed intervention in spinal cord injury

Wilhelm Hansen 1 , Reaz Moolla 2 , Duval Jugganath 3 , Matthew Velcich 4 , Kylen Govender 4 , Kofi Nyarko 4 , Olusegun Ogunbanjo 5 , Badar Munir 6

1 Independent Researcher, Düsseldorf, Germany

2 Independent Researcher, Johannesburg, South Africa

3 Frere Hospital, East London, South Africa

4 Chris Hani Baragwanath Academic Hospital, Johannesburg, South Africa

5 Thelle Mogoerane Regional Hospital, Vosloorus, South Africa

6 Helen Joseph Hospital, Johannesburg, South Africa

Introduction: Spinal cord injury (SCI) results in significant disability, and the optimal timing of surgical decompression remains debated. Early intervention may mitigate secondary injury mechanisms such as ischaemia and inflammation; however, definitions of “early” vary, and implementation is often limited, particularly in low-resource settings like South Africa. Landmark studies, including Surgical Timing in Acute Spinal Cord Injury Study (STASCIS), indicate improved outcomes with decompression within 24 hours, whereas others, such as Prospective, Observational European Multicenter study on the efficacy of acute surgical decompression after traumatic Spinal Cord Injury (SCI-POEM), question the universality of this benefit. This review systematically evaluates global evidence comparing early and delayed decompression, with emphasis on feasibility in resource-constrained health systems. Material and Methods: This systematic review and meta-analysis adhered to PRISMA guidelines (PROSPERO CRD420251000074). MEDLINE, Google Scholar, Scopus, CINAHL, Web of Science, and Cochrane Library were searched to February 2025. A total of 4081 records were identified and screened on Rayyan by three reviewers. After deduplication, 330 full texts were evaluated. All authors participated in the full-text screening, resulting in 91 studies included and 202 excluded; 37 could not be retrieved. An additional 7 studies were identified through reference searches. Data were extracted using a standardised form and analysed with standard Cochrane meta-analytic methods: RoB 2 for assessing bias, random-effects pooling where applicable, and heterogeneity evaluated by I 2 and Cochran’s Q. Results: Ninety-eight full texts, primarily observational cervical SCI cohorts, were included. Early decompression (≤ 24 hours) was consistently associated with improved neurological recovery at 6-12 months, with the greatest benefits in incomplete injuries, particularly AIS (ASIA Impairment Scale) C and central cord syndrome (CCS). In contrast, benefits in complete injuries (AIS A) were modest and more variable. Ultra-early decompression (≤ 6-8 hours) was associated with additional gains in select cohorts. In cervical facet dislocation, reduction within ∼6 hours was associated with better outcomes than later intervention. Motor recovery, especially upper-extremity function in CCS, and functional outcomes (Spinal Cord Independence Measure and Functional Independence Measure) also favoured earlier surgery. Early intervention did not consistently increase complications or mortality. Instead, decompression within 72 h was associated with shorter ventilation, reduced Intensive Care Unit (ICU) stay, and decreased hospital stay. Variability reflected differences in baseline severity, injury phenotype, co-interventions such as mean arterial pressure augmentation and corticosteroid use, and centre-level logistics. Certainty remains limited by the predominance of observational data and residual confounding; however, higher-quality cohorts using multivariable adjustment support these findings. Conclusion: This systematic review and meta-analysis reaffirm the principle that “Time is Spine.” Early decompression, particularly within 24 hours, is associated with better neurological recovery and a shorter hospital stay without an added safety risk. The strongest benefits are seen in incomplete injuries and CCS. Ultra-early intervention may offer additional benefits, although supporting evidence remains limited. Effects in complete injuries are less clear, and applying these findings to resource-limited health systems is challenging. Pragmatic clinical trials and targeted system strategies are necessary to facilitate timely surgery in settings where access and capacity are limited.

ID: AGS26/288

A428: Decreased time to imaging and surgery after implementation of an acute traumatic spinal cord injury pathway

Anthony DiGiorgio 1,2,3 , Sara Cole 1 , John Kanter 1,2,3 , Abel Torres Espin 1,2,3 , Nikolaos Kyritsis 1,2,3 , Ashraf El Naga 2,4 , David Gendelberg 4 , Lawrence Chyall 2 , Malini Singh 2,5 , Jason Talbott 2,6 , Rajiv Saigal 1 , Debra Hemmerle 1,2,3 , Rebecca Plevin 7 , Steven Straube 2,5 , Rachel Perry Limon 2 , Phil Bonney 1,2,3 , Abraham Dada 1 , Swetha Atluri 1 , Jay Kumar 1 , Adam Ferguson 1 , Geoffrey Manley 1 , John Yue 1 , Phiroz Tarapore 1,2,3 , Michael Huang 1,2,3

1 Department of Neurological Surgery, University of California, San Francisco, San Francisco, United States

2 Zuckerberg San Francisco General Hospital and Trauma Center, San Francisco, United States

3 Weill Institute for Neurosciences, Brain and Spinal Injury Center, University of California, San Francisco, San Francisco, United States

4 Department of Orthopaedic Surgery, University of California San Francisco, San Francisco, United States

5 Department of Emergency Medicine, University of California, San Francisco, San Francisco, United States

6 Department of Radiology and Biomedical Imaging at the University of California, San Francisco, San Francisco, United States

7 Department of Surgery, University of California, San Francisco, San Francisco, United States

Introduction: Prompt surgical intervention for acute traumatic spinal cord injury (ATSCI) is associated with improved outcomes. However, many structural barriers delay care. Here, we report the results of an institutional implementation of an ATSCI activation pathway. Material and Methods: Stakeholders from neurosurgery, orthopedic surgery, emergency medicine, trauma surgery, radiology, anesthesia, and nursing participated in an iterative process for the development of the ATSCI activation. We compared the results of the ATSCI activations to a cohort of spinal cord injury patients prior to protocol implementation using time-to-event analysis and a Cox regression model. Results: The process resulted in an order set in the electronic health record that could be triggered by the emergency medicine physician for suspected ATSCI. This notified a number of specialists simultaneously and began the process of obtaining a stat MRI. The spine surgery team would have the option to cancel the activation prior to the MRI if they felt it was a false positive. From 6/26/24 to 7/10/25 the ATSCI activation was used 162 times. 40 of these were confirmed to be spinal cord injuries. 51 of the activations were cancelled by the spine team prior to MRI, giving a 25% specificity for the ATSCI without cancellations and 36% after evaluation by a spine surgery team. Compared to pre-protocol cohort (from 1/1/23 to 7/1/23), the time from arrival to MRI significantly decreased (HR = 3.06, p < 0.001) from a median of 6.3h [95% CI 3.9 - 8.4] to 2.6h [2.3 - 2.9]; the time to the OR significantly decreased (HR = 4.23, p < 0.001) from a median of 14.4h [7.4 - 42.1] to 4.7h [3.9 - 8.2]. Conclusion: Implementation of an ATSCI activation protocol led to expedited care for acute spinal cord injury patients, reducing time to imaging by 59% and time to operative intervention by 67%.

ID: AGS26/3073

A429: Exploring the diffusion and adoption of the AO Spine/Praxis Guidelines 2024: results from a cross-sectional survey designed and conducted by the AO Spine Knowledge Forum Spinal Cord Injury

Mario Ganau 1 , Valerie Ter Wengel 2 , Vafa Rahimi-Movaghar 3 , Alan Martin 4 , Brian Kwon 5 , Shekar N. Kurpad 6 , Michael Fehlings 7

1 Oxford University Hospitals NHS FT, University of Oxford, BAU International University, Batumi, Georgia

2 Haaglanden Medical Center, Leiden University Medical Center, Amsterdam University Medical Centers, The Hague, The Netherlands

3 Tehran University of Medical Sciences, Tehran, Iran

4 UC Davis, Sacramento, United States

5 University of British Columbia, Vancouver, Canada

6 Medical College of Wisconsin, Milwaukee, United States

7 University of Toronto, Toronto, Canada

Introduction: This study aims to assess global trends in the management of acute spinal cord injury (SCI), and to evaluate the openness of the international spinal community to adopt the AO Spine/Praxis guidelines 2024. Material and Methods: A cross-sectional online survey issued by the AO Spine Knowledge Forum on SCI was circulated among AO Spine members between August and September 2025. The survey captured surgeons’ demographics and explored their preferences in terms of timing of surgical decompression and postoperative hemodynamic management of acute traumatic SCI, as well as risk profiling and prevention of intraoperative SCI. The survey was specifically designed to test the responders’ knowledge of the AO Spine/Praxis recommendations and evaluated their management preference in three exemplificative clinical vignettes. The authors conducted a descriptive statistical analysis and a regression analysis to explore associations between surgical choices and surgeon demographics. Results: The survey was completed by a total of 766 responders from the Five Regions of Asia Pacific (37%), Europe and Southern Africa (28%), Middle East and Northern Africa (13%), Latin America (12%), and North America (10%). The majority of responders (n.575) practiced urban areas, either in Academic (41%) or Public (33%) hospitals. Of note, 30% of responders revealed that their healthcare facility lacks standard operating procedures for the management of SCI. The survey proposed five drivers for incorporating international guidelines into local formal protocols, the responders ranked them as following in order of importance: 1) Evidence Based medicine, 2) Excellence in patients care, 3) Standardization of care, 4) Professionalism and 5) Medico-Legal implications. While the clinical vignettes revealed that the responders possessed a good theoretical knowledge of the AO Spine/Praxis recommendations, the survey allowed to highlight the three most common barriers to adopt those guidelines: 1) Limited availability of specialized staff during off-hours (e.g., nights/weekends), 2) Lack of relevant clinical training or skills to implement the guidelines, and 3) Limited access to essential physical resources (e.g., fluoroscopy, MRI, IONM, ICU beds). Conclusion: The international spinal community is very knowledgeable of the AO Spine/Praxis guidelines; nonetheless in the majority of cases those guidelines have not been incorporated into formal local protocols yet. Future research should focus on the impact of those recommendations on surgical decision-making and perioperative management of SCI patients. The engagement of the Knowledge Forum SCI with the international spinal community will be pivotal to understand how the barriers to adoption can be effectively brought down.

ID: AGS26/1298

A430: The Paris method: a new decision-making algorithm for the management of sacral fractures

Danilo Casasola 1 , Christophe Travert 1 , Laura Marie-Hardy 1 , Pascal-Moussellard Hugues 1

1 Pitié-Salpêtrière University Hospital, Sorbonne University, Paris, France

Introduction: Sacral fractures are rare but complex injuries that compromise both pelvic ring and spinopelvic stability, often with neurological involvement. Existing classifications (Tile, Roy-Camille, Denis) address different aspects of these injuries but lack a unified framework to guide surgical decision-making. This study proposes and validates a stepwise algorithm integrating these three systems for comprehensive management of high-energy sacral fractures. Material and Methods: We conducted a retrospective monocentric study including all patients ≥16 years admitted to a Level I trauma center between January 2021 and October 2023 with CT-confirmed sacral fractures. Fractures were classified through a three-step algorithm evaluating (1) pelvic ring stability (Tile), (2) spinopelvic stability (modified Roy-Camille), and (3) neurological prognosis (Denis). Two independent orthopedic surgeons performed double readings to assess intra- and interobserver reproducibility. Agreement was measured using Cohen’s kappa (κ). Results: Seventy-eight patients were included (mean age 39; 60% female). Falls from height represented the leading cause (67%), including 45% suicide attempts. Neurological deficits occurred in 18% of cases, all in Denis type II-III fractures. Surgical stabilization was performed in 36% of patients. Observer agreement was substantial across all steps: Tile classification κ = 0.67 (interobserver), Roy-Camille κ = 0.64, Denis κ = 0.76. Intraobserver reproducibility was also substantial (κ range 0.69-0.75). Notably, all Tile C fractures correlated with Roy-Camille type 3, and all Denis III fractures were associated with neurological impairment. Conclusion: The proposed “Paris Method” integrates Tile, Roy-Camille, and Denis systems into a reproducible three-step algorithm for sacral fracture management. It provides a systematic assessment of pelvic ring and spinopelvic stability while anticipating neurological risk, with substantial inter- and intraobserver agreement. This unified tool may improve decision-making and surgical strategy in complex sacral fractures. Prospective multicenter validation is warranted.

ID: AGS26/708

A431: Hyperextension injuries in patients with ankylosing spine disorder: impact of pre-injury mobility status on mortality

Collin Lamba 1 , John Bartoletta 1 , Emily Hunt 1 , Karina Katchko 1 , Haitao Zhou 1 , Carlo Bellabarba 1 , Richard Bransford 1

1 University of Washington/Harborview Medical Center, Seattle, United States

Introduction: Diffuse idiopathic skeletal hyperostosis (DISH) and ankylosing spondylitis (AS) are the primary ankylosing spinal disorders (ASDs). ASDs share key clinical features, including functional spinal ankylosis, poor bone quality, advanced age, and multiple associated medical comorbidities which contribute to an increased risk of fractures, as the rigid, ankylosed spinal column amplifies the effect of even minor trauma. These fractures lead to significantly higher rates of neurologic compromise, morbidity, and mortality as compared to hyperextension injuries in the absence of ASD. This study evaluates the impact of pre-injury ambulation status on mortality rates for hyperextension injuries in patients with ASD. Material and Methods: We retrospectively reviewed ASD patients who underwent surgery for hyperextension injuries at a single, Level 1 trauma center between 2005 and 2020. Patients were stratified by pre-injury ambulatory status: independent ambulation, cane/walker use, or wheelchair dependence. Collected variables included demographics, comorbidities, ASD type, pre-operative American Society of Anesthesiologists Physical Status (ASA PS), American Spinal Injury Association (ASIA) Impairment Scale (IS), fracture location, injury type (single versus multilevel versus combined spine/pelvic), mechanism, and Surgical Invasiveness Index (SII). The primary outcome was mortality; secondary outcomes included post-operative ambulation and reoperations within 90 days. Statistical analyses included Mann-Whitney U test, Student T-test, and Chi-Squared test as appropriate; survival was estimated with Kaplan-Meier. A Cox proportional hazards model assessed predictors of mortality with age and ASA PS as covariates and pre-injury ambulation and 90-day reoperation as factors. A p-value less than 0.05 was considered statistically significant. Results: Of 467 identified patients, 387 met inclusion criteria; 72.5% were independent ambulators, 21.2% used a cane/walker, and 6.3% used a wheelchair pre-injury. Assistive device use was associated with older age, higher ASA class, and hypertension, but not with BMI, sex, ASD type, neurologic status, injury characteristics, or surgical factors. Mortality differed significantly by ambulatory status: 90-day mortality was 1.4% for independent ambulators, 39.3% for cane/walker users, and 84.0% for wheelchair users; 1-year mortality was 1.7%, 47.6%, and 96.0%, respectively (p < 0.001). Age (HR 1.05 per year, p < 0.001) and ASA class (HR 1.65 per class, p = 0.018) independently predicted mortality, while hypertension was not associated. At last follow-up, 42.0% of previously ambulatory patients required an assistive device. Reoperation occurred in 14.6% of patients, most commonly for infection, and was not associated with mortality. Conclusion: Pre-injury assistive device use, older age, and higher comorbidity burden were associated with markedly increased mortality after hyperextension injury in patients with ASD. One-year mortality exceeded 95% in wheelchair users compared with <2% in independent ambulators. Pre-injury ambulatory status is a simple yet powerful predictor of mortality and functional decline after hyperextension injuries in ASD, surpassing traditional injury-specific metrics. These data offer valuable prognostic insights, helping to facilitate improved guidance for patients and their families who are navigating treatment and recovery for these injuries.

ID: AGS26/869

A432: Percutaneous stabilization of AO b3 fractures in patients with ankylosing spinal disorders (ASD) is associated with lower complication rates compared to open stabilization

Eli Bunzel 1 , Emily Hunt 1 , Jonathan Lawson 1 , Haitao Zhou 1 , Carlo Bellabarba 1 , Richard Bransford 1

1 University of Washington/Harborview Medical Center, Seattle, United States

Introduction: Ankylosing Spine Disorders (ASD), either in the form of ankylosing spondylitis (AS) or diffuse idiopathic skeletal hyperostosis (DISH), characteristically result in autofusion of the spine with abnormal rigidity and a long level arm. Patients with ASD have a fourfold increased risk of fracture compared with the general population. Surgical management of these fractures carries high perioperative complication rates and increased one-year mortality. Minimally invasive surgery (MIS) has been associated with reduced blood loss and complications compared with open approaches. We compared outcomes between open and percutaneous fixation in patients with AO B3 thoracolumbar fractures. Materials and Methods: We performed a retrospective review of adult patients with ASD presenting with thoracolumbar fractures treated operatively between 2005 and 2020 at a single, level 1 academic trauma center. Patients were divided into open versus percutaneous fixation groups. Demographic, comorbidity, operative, hospital, and 90-day postoperative data were collected. Independent-samples t-tests and Fisher’s exact tests were used for continuous and categorical variables, respectively. Results: Two hundred cases (64 AS, 136 DISH) were identified from 2005 to 2020. 174 patients underwent open instrumentation, and 26 patients underwent percutaneous instrumentation. Overall, surgical site infection (SSI) rate was 11% (22/200). SSI infection rate was 12% (21/174) in the open group, versus 4% (1/26) in the percutaneous group. 90-day reoperation rate was 14% (24/174) in the open group versus 4% (1/26) in the percutaneous group. Mean estimated blood loss (EBL) was 575.0 ± 39.6 mL in the open group versus 184.0 ± 67.7mL in the percutaneous group. In-hospital mortality was 10% (17/174) in the open group versus 4% (1/26) in the percutaneous group. These were all significant with a p value <0.05. Length of ICU stay and discharge environment did not differ significantly between groups. Conclusion: Compared with open instrumentation, percutaneous posterior instrumentation of AO B3 fractures in patients with ASD was associated with significantly reduced blood loss, lower SSI, lower 90-day reoperation, and decreased in-hospital mortality. Percutaneous posterior instrumentation should be strongly considered for stabilization of AO B3 fractures to decrease morbidity in this high-risk population.

OP49: Cervical Degenerative

ID: AGS26/2769

A433: Patients receiving GLP-1 agonists have reduced odds of pseudoarthrosis and revision following anterior cervical discectomy and fusion surgery

Manaav Mehta 1 , Abhinav Sharma 2 , Ryan Beyer 2 , Vivian Chen 2 , Michael Kim 2 , Paramveer Birring 2 , Hao-Hua Wu 2 , Sohaib Hashmi 2 , Yu-po Lee 2 , Nitin Bhatia 2 , Don Park 2

1 University of Massachusetts Chan Medical School, Worcester, United States

2 University of California Irvine, Orange, United States

Introduction: In accordance with the aging population, a significant increase in the volume of anterior cervical fusion procedures is expected over the next two decades. This increase has coincided with a rise in postoperative complications, notably pseudoarthrosis-a failure in spinal fusion manifesting as recurring pain and mechanical instability. The association between glucagon-like peptide-1 (GLP-1) agonist use and complications following anterior cervical fusion surgery has yet to be explored. The goal of this study is to examine the association between GLP-1 agonist usage and the incidence of pseudoarthrosis, adjacent segment disease, revision surgery, and perioperative complications in patients undergoing anterior cervical discectomy and fusion (ACDF) surgery. Material and Methods: TriNetX, a global health research network, was queried from 2004-2024 for patients undergoing anterior cervical discectomy and fusion (CPT 22551) and who had documented GLP-1 agonist usage within 1-year post-operation. Analysis included chi-squared tests with a control cohort propensity score matched for age, sex, race, smoking, diabetes, BMI, and uremia who did not have post-operative GLP-1 agonist usage. Outcomes of interest were DVT, emergency department (ED) visits, inpatient hospitalization, opioid abuse, PE, sepsis, surgical site infection, wound complications within 90 days, adjacent segment disease, dysphagia, implant failure, pseudoarthrosis, and revision within 2 years. Results: A total of 2,144 patients who received GLP-1 were compared to propensity matched controls. After matching, patients were 58.1 ± 10 years old, 39% male, 66% white, 80% diabetic, and 68% obese. GLP-1 agonist use within 1-year post-operation was significantly associated with reduced odds of ED visits (OR 0.82, CI 0.68-0.99), inpatient hospitalization (OR 0.28, CI 0.23-0.33), sepsis (OR 0.51, CI 0.29-0.90) within 90 days, dysphagia (OR 0.72, CI 0.60-0.87), pseudoarthrosis (OR 0.50, CI 0.41-0.60), and revision surgery (OR 0.57, CI 0.35-0.93) within 2 years post-operation. Odds of DVT, PE, surgical site infection, wound complications, opioid abuse, adjacent segment disease, and implant failure were not significant. Conclusion: In this cohort study of patients undergoing ACDFs, patients prescribed GLP-1 agonists during the post-operative period experienced lower rates of pseudoarthrosis, dysphagia, revision surgery, sepsis, and overall hospitalizations compared to those who did not receive GLP-1 agonists. These results suggest that GLP-1 agonists may offer a therapeutic benefit in improving spinal fusion outcomes, and further prospective research is needed to validate these findings and investigate the underlying mechanisms.

ID: AGS26/1316

A434: Ligament reconstruction of the cervical spine: the anterior longitudinal ligament (ALL) And its biomechanical role in anterior surgical approaches

Jana Felicitas Schader 1,2,3 , Mauro Suter 4 , Marie-Rosa Fasser 4 , Warnecke Daniela 5 , Anna Schuler 1,4 , José Spirig 2 , Jess Snedeker 1,3 , Mazda Farshad 1,2 , Jonas Widmer 4

1 Department of Orthopedics, Balgrist University Hospital, University of Zurich, University Spine Center Zürich, Balgrist University Hospital, University of Zurich, Institute of Biomechanics, Balgrist Campus, ETH Zurich, Zurich, Switzerland

2 University Spine Center Zurich, Balgrist University Hospital, University of Zurich, Zurich, Switzerland

3 Institute of Biomechanics, ETH Zurich, Zurich, Switzerland

4 Spine Biomechanics, Department of Orthopedic Surgery, Balgrist University Hospital, University of Zurich, Zurich, Switzerland

5 Research Department, Arthrex GmbH, Munich, Germany

Introduction: Typically, degenerative conditions of the cervical spine are managed via an anterior surgical approach. Nevertheless, this approach may lead to complications such as adjacent segment degeneration or instability due to biomechanical changes following the removal of the anterior longitudinal ligament (ALL) during surgery. This study leverages orthopedic strategies focused on ligament reinforcement to stabilize joints. It examines the effects of reconstructing the ALL to provide targeted stabilization at the anterior spine, while avoiding complete immobilization of the segment. The study’s purpose was to evaluate the biomechanical impacts of removing the ALL and to assess if its reconstruction could maintain stability and range of motion akin to the cervical segment's intact state. Material and Methods: Biomechanical testing of ten spinal segments from four cadaveric human cervical spines (C2/3: 2, C3/4: 2, C4/5: 2, C5/6: 2, C6/7: 2) was performed. Spinal segments were fixed using 3D-printed clamps. ALL reconstruction was executed using surgical tape and human gracilis tendon allografts, attached with bone anchors. Testing was performed on each specimen in three phases: pre-resection (native state), post-ALL resection, and post-ALL reconstruction. ROM was evaluated in flexion-extension (FE), lateral shear (LS), lateral bending (LB), anteroposterior shear (AS), and axial rotation (AR). Results: After the ALL was removed, median ROM in FE increased to 131% [25th-75th percentile: 122%-163%] (p < 0.01) and reduced to 103% [93%-115%] (p > 0.05) after ALL reconstruction compared to the intact state (100%). Similar trends were observed in LS, LB, AS, and AR, with the least change seen in AR after ALL resection (111% [108%-125%], p < 0.05) and in LS after ALL reconstruction (116% [103%-119%], p < 0.05). Absolute ROM in FE increased progressively from intact to reconstructed to resected state (8.5° [6.2°-11.7°], 10.0° [5.7°-12.9°], 12.1° [8.3°-14.3°], respectively). Conclusion: The ALL is crucial for maintaining the stability of the cervical spine in all motion directions, particularly in flexion-extension and to a lesser extent in axial rotation. Reconstructing the ALL nearly restores the spine's ROM to its original state in most loading directions, which may help decrease the likelihood of complications from ligament resection during anterior cervical surgeries.

ID: AGS26/185

A435: Ultra-early postoperative pain reduction predicts long-term outcomes following ACDF: preliminary results from a secondary analysis of a randomized controlled trial

Jon Håvard Sommernes 1,2 , Jarle Sundseth 2 , Mirad Taso 1,2 , Siri Bjorland 2 , Are Hugo Pripp 1,2 , Frode Kolstad 2 , Jens Ivar Brox 1 , John-Anker Zwart 1,2

1 University of Oslo, Oslo University Hospital, Oslo, Norway

2 Oslo University Hospital, Oslo, Norway

Introduction: Cervical radiculopathy is a common condition caused by cervical disc herniation or spondylosis, often leading to disabling arm pain and functional impairment. When conservative treatment fails, anterior cervical decompression and fusion (ACDF) is widely performed. Although most patients improve, recovery trajectories vary considerably, and clinicians lack tools to identify early who will achieve lasting benefit. Identifying reliable predictors of long-term outcomes shortly after surgery could improve counseling, guide of follow-up allocation, and support individualized care. Previous studies in lumbar surgery suggest a prognostic role for early postoperative pain reduction, but evidence in cervical radiculopathy and especially in the ultra-early period is sparse. We investigated whether pain relief on postoperative day 1 (POD1) predicts 12-month pain and disability outcomes following ACDF. Material and Methods: We conducted a secondary cohort analysis of 83 patients from the surgical arm of a randomized controlled trial comparing ACDF with nonsurgical management for cervical radiculopathy [1]. Eligible patients were 20-65 years with ≥ 3 months of radicular arm pain and imaging-confirmed disc herniation or spondylosis at C5/C6 and/or C6/C7. Exclusion criteria included cervical myelopathy and prior cervical surgery. Pain was assessed with the Numeric Rating Scale (NRS) for arm, neck, and headache at baseline, POD1, and 12 months. Disability was measured with the NDI. Early improvement was defined as a ≥ 2-point NRS reduction on POD1. Multivariable linear and logistic regression analyses adjusted for age, sex, body mass index, smoking, ASA class, diagnosis, surgical levels, operative time, baseline intensity. Results: On POD1, 94% of patients experienced a ≥ 2-point arm pain reduction. These responders had higher odds of maintaining meaningful arm pain relief at 12 months compared with non-responders (90.9% vs. 40.0%, p = 0.0077). Similar associations were observed for neck pain (87.0% vs. 53.8%, p = 0.0028) and headache (73.9% vs. 20.0%, p < 0.001). In adjusted analyses, greater POD1 arm pain reduction predicted larger improvements in arm pain (β = 0.64, p < 0.001) and NDI (β = 0.93, p < 0.001). Absolute arm pain on POD1 independently predicted 12-month arm pain (β = 0.28, p = 0.040). By contrast, associations for neck and headache pain were less consistent, and were not robust predictors of NDI. Patients with early arm pain relief were also less likely to use analgesics at one year. Conclusion: Ultra-early reduction in arm pain after ACDF is a clinically relevant predictor of long-term recovery. A simple NRS assessment on POD1 represents a feasible tool to stratify patients’ expected trajectories. Those with marked early relief may require less intensive follow-up, whereas patients without improvement could be prioritized for closer monitoring and supportive interventions. The predictive value was most evident for arm pain and disability, underscoring the role of radicular symptoms in cervical radiculopathy. External validation in larger multicenter cohorts is warranted, and future studies should explore whether combining ultra-early PROMs with psychosocial and multidomain recovery measures can further refine prognostic accuracy.

References

1. Taso M, Sommernes JH, Sundseth J, et al. Surgical versus Nonsurgical Treatment for Cervical Radiculopathy. NEJM Evid. 2025; 4(4):EVIDoa2400404.

ID: AGS26/2423

A436: Complications of staged versus same-day circumferential cervical spine fusion for degenerative disease

Mirbahador Athari 1 , Henry Avetisian 1 , William Karakash 1 , Aidan Lindgren 1 , Daniel Rusu 1 , Jordan Gasho 1 , Matthew Gallo 1 , Jeffrey C. Wang 1 , Raymond Hah 1 , Kiran Alluri 1

1 Department of Orthopaedic Surgery, Keck School of Medicine of the University of Southern California, Los Angeles, United States

Introduction: Cervical degenerative disc disease may be managed with anterior, posterior, or circumferential fusion. Circumferential procedures can be performed during a single anesthetic episode or staged across separate operations. While both strategies are used clinically, evidence comparing complication rates between same-day and staged procedures is limited. This study compared perioperative complications between same-day and staged circumferential cervical fusions using a large national database. Materials and Methods: We queried the PearlDiver Mariner database (2010-Q3 2022) to identify adults undergoing anterior cervical discectomy and fusion (ACDF) combined with posterior cervical fusion. Patients with trauma, infection, malignancy, deformity, rheumatoid arthritis, or age < 18 were excluded. Patients were grouped into Same-Day Surgery (circumferential fusion during one anesthetic) and Staged Surgery (procedures within 1-14 days). Demographics, comorbidities, and perioperative complications were assessed. Outcomes included 30-day medical and surgical complications, readmissions, and revision rates. Odds ratios (OR) with 95% confidence intervals (CI) were calculated; p < 0.05 was considered significant. Results: A total of 8,578 patients were identified: 6,496 underwent same-day surgery and 2,082 underwent staged surgery. Staged patients were slightly older (60.7 vs 60.0 years, p < 0.01) with higher comorbidity indices (5.21 vs 4.75, p < 0.001). Diabetes was more common in the staged group (p < 0.001), while sex, obesity, and tobacco use were comparable. Staged procedures were associated with significantly higher 30-day complication rates, including infection (8.5% vs 5.6%, OR 1.57, p < 0.001), radiculopathy (34.8% vs 16.0%, OR 2.58, p < 0.001), anemia (18.0% vs 12.2%, OR 1.42, p < 0.001), pneumonia (0.9% vs 0.4%, OR 2.29, p = 0.03), urinary tract infection (7.6% vs 5.7%, OR 1.40, p < 0.001), ileus (1.1% vs 0.7%, OR 1.99, p = 0.04), stroke (3.8% vs 2.2%, OR 1.55, p = 0.02), neurologic complications (32.3% vs 17.9%, OR 1.89, p < 0.001), mechanical complications (5.7% vs 2.1%, OR 2.82, p < 0.001), and transfusion (1.5% vs 0.7%, OR 2.00, p = 0.02). Same-day surgery had lower overall complication rates despite comparable demographics aside from comorbidity burden. Conclusion: Staged circumferential cervical fusions were associated with significantly higher complication rates compared with same-day procedures, including infection, anemia, respiratory, thromboembolic, neurologic, and mechanical complications. These findings suggest that same-day circumferential fusion may provide a safer perioperative profile in appropriately selected patients, while staged procedures may carry higher risks due to prolonged hospitalization and recovery. Optimizing patient selection and perioperative management is essential to reduce morbidity in this population.

ID: AGS26/952

A437: Cervical sagittal alignment after multilevel ACDF: correction goes along with loss of compensation

Marco Manzetti 1,2,3 , Fabio Galbusera 4 , Andrea Cina 4 , Dezsö Jeszenszky 1 , Francois Porchet 1 , Jani Puhakka 1 , Raluca Reitmeir 1 , Mario Ropelato 1 , Maja Gocević 1 , Daniel Haschtmann 1 , Frank Kleinstuck 1 , Alberto Ruffilli 2,3 , Cesare Faldini 2,3 , Tamas Fekete 1 , Markus Loibl 1

1 Schulthess Klinik, Spine Center, Zürich, Switzerland

2 Istituto Ortopedico Rizzoli, Clinica Ortopedica e Traumatologica I, Bologna, Italy

3 Department of Biomedical and Neuromotor Science - DIBINEM, Bologna, Italy

4 Schulthess Klinik, Department of Teaching, Research and Development, Zürich, Switzerland

Introduction: Cervical spondylosis can lead to loss of lordosis or even kyphosis, altering sagittal alignment and triggering compensatory mechanisms to maintain horizontal gaze. Although multilevel anterior cervical discectomy and fusion (ACDF) is commonly used, its impact on sagittal alignment and compensatory changes remains unclear. This study investigates changes in cervical sagittal alignment before and after multilevel ACDF and the associated compensatory mechanisms. Material and Methods: We performed a retrospective analysis of 298 patients undergoing 2- or 3-level ACDF for cervical spondylosis. Preoperative and postoperative sagittal parameters including C0-C2 lordosis, C2-C7 lordosis, surgical (fused-level) lordosis, unfused-level lordosis, T1 slope (T1S), T1S minus cervical lordosis (T1S-CL), and cervical sagittal vertical axis (c-SVA) were measured using a previously described deep learning model. Demographic data were collected at the same time points. Changes from baseline to follow-up were calculated. Correlations between changes in surgical lordosis and sagittal parameters were assessed via linear regression, Pearson’s correlation, and multiple logistic regression analysis. Results: The mean preoperative C2-C7 lordosis was 11.3° ± 13.6°, increasing significantly to 20.7° ± 15.6° after surgery (p < 0.0001). Segmental correction at fused levels averaged 6.5° in two-level and 8.6° in three-level ACDF cases. C0-C2 lordosis decreased from 20.5° ± 13.0° to 12.6° ± 12.6° (p < 0.0001), while T1S-CL also decreased from 20.9° ± 11.5° to 13.1° ± 14.7° (p < 0.0001). Changes in surgical lordosis correlated negatively with changes in unfused-level lordosis (R = -0.26, p < 0.01) and with C0-C2 lordosis (R = -0.11, p = 0.061). A strong inverse correlation was also observed between changes in C2-C7 and C0-C2 lordosis (R = -0.60, p < 0.01). Notably, the strongest correlation overall was found between changes in T1S-CL and C0-C2 lordosis (R = 0.63, p < 0.0001), suggesting that greater sagittal mismatch is associated with increased cranio-cervical extension. A moderate positive correlation was also found between surgical lordosis and changes in T1 slope (R = 0.29, p < 0.01). Conclusion: Multilevel ACDF can increase surgical lordosis and reduce compensatory mechanisms in the upper cervical and subaxial spine, via restoration of C2-C7 lordosis and correction of cervical mismatch. Notably, the observed relationship between T1S-CL and C0-C2 lordosis highlights the biomechanical role of cranio-cervical hyperextension as a compensatory response to sagittal mismatch. These findings support a shift toward viewing cervical alignment as a dynamic, multi-segmental system in which segmental correction can influence adjacent segment behavior, as described by the unfused lordosis. While long-term outcomes remain multifactorial, reducing compensation may lessen mechanical stress on unfused levels and help prevent future deterioration. Routine preoperative assessment of sagittal alignment-including mismatch and upper cervical compensation-should be integrated into surgical planning to guide alignment goals and improve long-term stability.

ID: AGS26/3175

A438: Impact of preexisting heart failure on postoperative adverse events after cervical spine fusion: a matched cohort analysis

Muaz Wahid 1 , Elias Nasser 1 , Zuhair Zaidi 1 , Salah Aoun 1 , Andrew Simpson 1

1 UT Southwestern Medical Center, Dallas, United States

Introduction: Heart failure (HF) is a major perioperative risk factor, yet its impact after cervical spine fusion is under-characterized. As cervical fusions increase in an aging population, surgeons increasingly encounter patients with preexisting HF. We quantified 90-day and 1-year adverse events among adults undergoing cervical fusion, comparing those with versus without HF in a large, multicenter electronic health-record network to inform risk stratification and perioperative pathways. Material and Methods: We conducted a retrospective matched-cohort study in the TriNetX US Research Network (2010-2023). Adults (≥ 18 years) undergoing anterior or posterior cervical fusion below C2 were eligible. Patients with a documented HF diagnosis before index surgery were matched 1:1 to non-HF comparators using propensity scores from 38 covariates (demographics, comorbidities, medication use, baseline laboratories). Nearest-neighbor matching used a 0.1 SD caliper; balance was verified via standardized mean differences. Seventeen outcomes were assessed from postoperative day 1 through 90 days and 1 year, including mortality, myocardial infarction (MI), stroke, pulmonary embolism (PE), deep-vein thrombosis (DVT), pneumonia, sepsis, acute renal failure, emergency department (ED) visits, wound disruption, superficial infection, and selected neuropsychiatric/substance-use endpoints. Effects were estimated as odds ratios (ORs) with 95% confidence intervals (CIs) and p-values. Final matched cohorts included 3,716 HF and 3,716 non-HF patients. Results: At 90 days, HF was associated with substantially higher odds of mortality (OR 5.01, p < 0.0001), MI (9.43, p < 0.0001), PE (3.41, p < 0.0001), DVT (2.60, p < 0.0001), stroke (2.13, p < 0.0001), pneumonia (5.51, p < 0.0001), sepsis (4.20, p < 0.0001), acute renal failure (2.41, p < 0.0001), and ED visits (1.18, p = 0.0118); superficial infection (0.77, p = 0.14), wound disruption (0.76, p = 0.079), and opioid use disorder (1.00, p = 1.00) were not significantly different. Somatic symptoms were more frequent in HF, including chest pain (3.57, p < 0.0001), dizziness (1.67, p = 0.0002), and shortness of breath (4.47, p < 0.0001); alcohol use disorder was also higher (1.72, p = 0.0002). At 1 year, elevated risks persisted or intensified: mortality (3.78, p < 0.0001), MI (6.96, p < 0.0001), PE (3.08, p < 0.0001), DVT (3.30, p < 0.0001), stroke (1.84, p = 0.0002), pneumonia (3.79, p < 0.0001), sepsis (3.49, p < 0.0001), acute renal failure (2.35, p < 0.0001), and ED visits (1.45, p < 0.0001). Absolute rate differences illustrated clinical magnitude-for example, 90-day mortality was 6.76% in HF vs 1.43% in non-HF, and 1-year mortality was 11.62% vs 3.36%. Wound disruption and superficial infection remained nonsignificant at both horizons; opioid use disorder remained nonsignificant at 1 year (1.18, p = 0.27). Dementia (1.40, p = 0.027) and alcohol use disorder (1.49, p = 0.0011) were higher at 1 year in HF. Conclusion: Preexisting HF confers a broad, durable increase in postoperative risk after cervical fusion, spanning cardiopulmonary complications (MI, pneumonia), thromboembolism (DVT/PE), sepsis, acute renal failure, health-care utilization (ED visits), and mortality at 90 days and 1 year, without corresponding increases in superficial wound complications. These findings support incorporating HF status into preoperative counseling and risk models, optimizing guideline-directed medical therapy, and deploying enhanced perioperative bundles (e.g., pulmonary hygiene, VTE prophylaxis stratified for HF, vigilant hemodynamic/renal management, and early infection surveillance). Future work should stratify by HF subtype and severity (ejection fraction, NYHA class) to personalize mitigation strategies for this high-risk cervical fusion population.

ID: AGS26/671

A439: Predictors of return-to-work among patients undergoing cervical spine surgery for radiculopathy - A SpineCORe Study Group QOD analysis

Michael Tawil 1 , Mohamad Bydon 2 , Anthony Asher 3 , Domagoj Coric 3 , Paul Kim 3 , Matthew McGirt 3 , Eric Potts 4 , Kevin Foley 5 , Michael Wang 6 , Kai-Ming Fu 7 , Michael Virk 7 , Christopher Shaffrey 8 , Oren Gottfried 8 , John Knightly 9 , Scott Meyer 9 , Paul Park 5 , Cheerag Upadhyaya 10 , Chun-Po Yen 11 , Andrew Chan 12 , Luis Tumialan 13 , Dean Chou 12 , Juan Uribe 13 , Jay Turner 13 , Regis Haid 14 , Praveen Mummaneni 1

1 University of California, San Francisco, San Francisco, United States

2 University of Chicago, Chicago, United States

3 CNSA, Charlotte, United States

4 Goodman Campbell, Carmel, United States

5 Semmes Murphey, Memphis, United States

6 University of Miami, Miami, United States

7 Cornell, New York, United States

8 Duke, Durham, United States

9 Maxim Spine, Cedar Knolls, United States

10 UNC, Chapel Hill, United States

11 UVA, Charlottesville, United States

12 Columbia University Vagelos College of Physicians and Surgeons, New York, United States

13 Barrow, Phoenix, United States

14 Atlanta Brain and Spine, Atlanta, United States

Introduction: Returning to work is an essential indicator of functional recovery and surgical success. This study aimed to identify predictors of return to work at 1 year after cervical spine surgery for radiculopathy. Material and Methods: We selected patients from the 14-site SpineCORe study group in QOD who underwent surgery for cervical radiculopathy. Return-to-work status was assessed if patients returned to work at any time within 12 months postoperatively. Univariate and multivariate logistic regression analyses were performed to identify independent predictors of return-to-work. Results: Among 3,069 patients with return-to-work data, 2,394 (78%) returned to work by 1 year. Patients who returned to work were significantly younger, had private insurance, had lower comorbidity rates, and reported better baseline and postoperative patient-reported outcome measures (p < 0.05) . In multivariate analysis, the following factors were associated with lower odds of returning to work by 1 year: older age (OR 0.95, p < 0.001), Medicare (OR 0.14, p < 0.001) or Medicaid insurance (OR 0.32, p < 0.001), depression (OR 0.63, p = 0.04), current liability claim (OR 0.44, p = 0.006), longer length of stay (OR 0.85, p = 0.03), and higher NDI at 3 months (OR 0.90, p < 0.001). In multivariate analysis, the following were associated with higher odds of returning to work: advanced post-college education (OR 2.33, p = 0.001) and higher EQ-5D scores at 3 months (OR 6.57, p < 0.001). Conclusion: Approximately 78% of patients with cervical radiculopathy and available return-to-work data returned to work within 1 year of surgery. Socioeconomic and psychological factors, along with early postoperative functional outcomes, were key predictors. These findings may guide preoperative counseling and postoperative expectations.

ID: AGS26/746

A440: Substance-specific outcomes after cervical fusion: multicenter, propensity-matched analysis of tobacco, non-tobacco nicotine, and cannabis

Abdel Diab 1 , Reme Arhewoh 1 , Omar Diab 1 , Elinor Stern 1 , Ilyas Aleem 1 , Rakesh Patel 1 , Joshua Piche 1

1 University of Michigan, Ann Arbor, United States

Introduction: The negative impact of tobacco use on surgical outcomes, including cervical fusion procedures, is well-established, with evidence linking it to impaired healing, increased infections, and higher rates of complications. However, the effects of non-tobacco nicotine products (e.g., e-cigarettes) and cannabis use remain poorly defined, despite their growing prevalence. As these substances become more commonly used, understanding their specific associations with postoperative outcomes is crucial for risk stratification and perioperative management. This study evaluates the distinct effects of tobacco, non-tobacco nicotine dependence (NTND), and cannabis use on both short-term and long-term outcomes in cervical fusion procedures addressing a critical gap in perioperative risk stratification. Material and Methods: A retrospective cohort study was conducted using the TriNetX database, encompassing 104 healthcare organizations, and included only patients undergoing cervical fusion. Propensity matching was performed using age, sex, BMI, race, and comorbidities such as diabetes, hypertension, and CKD. Primary outcomes included 90-day complications (hospital readmission, sepsis, infection, myocardial infarction, mortality, and renal failure). Secondary outcomes included 2-year mechanical complications (pseudoarthrosis, proximal junctional kyphosis [PJK], reoperations, wound disruption, and spinal infections). Statistical analyses included risk differences (RD), odds ratios (OR), and Kaplan-Meier survival analysis for reoperations. Results: The study included three distinct substance-use cohorts: NTND (n = 414), cannabis (n = 659), and tobacco (n = 8,273), and respectively matched controls. After matching, baseline characteristics were well-balanced across groups. • NTND Cohort: NTND users exhibited significantly higher 90-day hospital readmissions compared to controls (11.1% vs. 5.6%, p = 0.004; OR 2.12, 95% CI 1.26-3.58). At 2 years, pseudoarthrosis rates were significantly lower in NTND users compared to controls (7.4% vs. 13.7%, p = 0.004; OR 0.50, 95% CI 0.31-0.81). Other mechanical complications could not be assessed due to limited sample sizes. • Cannabis Cohort: Cannabis users demonstrated increased 90-day hospital readmission rates compared to controls (11.8% vs. 7.0%; RD = 4.9%, p = 0.003; OR 1.789, 95% CI 1.221-2.620). No significant differences were observed in 2-year pseudoarthrosis or other mechanical complications. • Tobacco Cohort: Tobacco users experienced significantly higher rates of 90-day complications, including sepsis (1.2% vs. 0.6%, p < 0.001), infection (1.3% vs. 0.7%, p = 0.001), myocardial infarction (0.7% vs. 0.3%, p < 0.001), mortality (0.7% vs. 0.4%, p = 0.011), and hospital readmissions (11.1% vs. 7.0%, p < 0.001). At two years, tobacco use remained associated with increased risks of reoperation (p = 0.02), wound disruption (p = 0.017), spinal infection (p < 0.001), and proximal junctional kyphosis (PJK) (p = 0.004). Paradoxically, tobacco users demonstrated lower rates of pseudoarthrosis compared to controls (10.2% vs. 17.1%, p < 0.001; OR 0.55, 95% CI 0.50-0.60). Conclusion: NTND and cannabis use primarily contribute to increased hospital readmissions within 90 days. Tobacco use, serving as a positive control, is strongly associated with increased short-term complications, including infection, sepsis, and readmissions, as well as higher long-term risks of reoperation and wound-related complications following cervical fusion. Paradoxically, both tobacco and NTND users exhibited unexpectedly lower rates of pseudoarthrosis. These may reflect data artifacts, survivorship bias, difference in follow up patterns etc. Overall, these findings emphasize the comparatively reduced harm of other substances in comparison with tobacco but underscore the need for targeted preoperative counseling and tailored perioperative care for patients with substance use, with further research required to clarify underlying mechanisms and refine management strategies tailored to substance use patterns.

ID: AGS26/822

A441: Discordance between minimal clinically important difference and patient satisfaction after surgery for cervical spondylotic myelopathy: a 5-year analysis from the quality outcomes database

Abraham Dada 1 , Nima Alan 1 , Jay Kumar 1 , Andrew Chan 2 , Erica Bisson 3 , Mohamad Bydon 4 , Anthony Asher 5 , Kevin Foley 6 , Paul Kim 5 , Eric Potts 7 , Michael Wang 8 , Kai-Ming Fu 9 , Michael Virk 9 , Christopher Shaffrey 10 , Oren Gottfried 10 , John Knightly 11 , Paul Park 6 , Cheerag Upadhyaya 12 , Chun-Po Yen 13 , Luis Tumialan 14 , Dean Chou 2 , Juan Uribe 14 , Jay Turner 14 , Regis Haid 15 , Praveen Mummaneni 1

1 University of California, San Francisco, United States

2 Columbia, New York, United States

3 University of Utah, Salt Lake City, United States

4 University of Chicago, Chicago, United States

5 CNSA, Charlotte, United States

6 Semmes-Murphey, Indianapolis, United States

7 Goodman Campbell, Indianapolis, United States

8 University of Miami, Miami, United States

9 Cornell, New York, United States

10 Duke, Durham, United States

11 Maxim Spine, Morristown, United States

12 University of North Carolina, Chapel Hill, United States

13 University of Virginia, Charlottesville, United States

14 Barrow, Phoenix, United States

15 Atlanta Brain and Spine, Atlanta, United States

Introduction: Patient satisfaction with surgical outcomes may not always align with MCID achievement. This study identifies factors associated with 5-year dissatisfaction among patients who met MCID criteria following surgery for cervical spondylotic myelopathy (CSM). Material and Methods: Data was obtained from the Spine CORE QOD CSM, prospective dataset which includes patients from 14 sites. The cohort included 1,085 patients with a 5 year follow-up rate of 80%. Patients with complete baseline and 5-year PROs and satisfaction data were analyzed. MCID achievers (EQ-5D or NDI) were stratified by satisfaction using the NASS Patient Satisfaction Index (scores 1-2 = satisfied; 3-4 = dissatisfied). Univariate tests and multivariable logistic regression identified preoperative features associated with dissatisfaction. Results: 721 patients had complete baseline and 5-year postop PRO and satisfaction data. At 5-year follow up 527 patients (73.1% of 721) met MCID. Of those who met MCID, 40 (7.6%) unsatisfied patients were compared to 487 (92.4%) satisfied patients. At baseline, both cohorts were similar in terms of demographics and type of surgery. Dissatisfied patients were more likely to be smokers (35.0 % vs 15.4 %; p = 0.002), lack private insurance (65.0 %v s 45.8 %; p = 0.019), and be functionally dependent at baseline (28.2 % vs 15.2 %; p = 0.034). They also presented with worse baseline PROs: NDI (48.0 ± 17.1 vs 40.5 ± 20.4; p = 0.025), VAS arm pain (6.4 ± 3.1 vs 5.1 ± 3.4; p = 0.015), EQ-5D (0.50 ± 0.20 vs 0.60 ± 0.20; p = 0.024) and mJOA (11.2 ± 2.8 vs 12.1 ± 2.7; p = 0.048). On multivariable logistic regression, only preoperative smoking status correlates with dissatisfaction (odds ratio 2.42; 95%CI 1.13-4.97; p = 0.019). Conclusion: 73.1% of CSM patients achieved MCID thresholds, and 92.4% of these reported satisfaction at 5 years after surgery. The 7.6 % who remained dissatisfied despite meeting MCID were more likely to be smokers at baseline.

OP50: Tumor Prognostics

ID: AGS26/1066

A442: Does duration of symptoms in metastatic spine disease affect recovery and outcomes? A systematic review and meta-analysis

Harsh Jain 1 , Nick De Oliviera 2 , Advith Sarikonda 1 , Ilya Laufer 3 , Scott Lawrence Zuckerman 1

1 Vanderbilt University Medical Center, Nashville, United States

2 Vanderbilt University School of Medicine, Nashville, United States

3 NYU Langone Medical Center, New York, United States

Introduction: Despite widespread recognition that earlier intervention may improve outcomes in metastatic epidural spinal cord compression (MESCC), the specific impact of symptom duration on post-treatment recovery remains understudied. We sought to evaluate the association between the duration of pre-treatment symptoms, specifically motor function, and clinical outcomes, specifically motor recovery and survival, in patients undergoing radiotherapy (RT) or surgery for MESCC, through a systematic review and meta-analysis. This project was initiated by AO Spine Knowledge Forum Tumor. Material and Methods: A systematic literature review was performed in accordance with Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines using PubMed/Embase databases (01/01/1990-05/03/2025). Study inclusion criteria were: adults with MESCC with pre-treatment symptom duration and its association with outcomes after RT or surgery. Primary exposure was duration of pre-treatment symptoms, which was most often motor deficits. Covariates included age, sex, visceral metastases, and extent of vertebral involvement. Interventions included RT or surgery. Primary outcomes were motor recovery and ambulation; secondary outcomes included survival and local control. Pooled risks were estimated using random-effects models, stratified by intervention and outcome. Due to limited sample size, meta-analysis was performed for the following interventions/outcomes: RT with motor recovery and survival; surgery with survival. Results: Of 4639 articles reviewed, 37 met the inclusion criteria (26 RT cohorts, 11 surgical cohorts). RT: Symptom duration in all RT studies (26/26,100%) was defined as the time from the onset of motor weakness to RT initiation. Longer duration or slower development of motor symptoms was associated with improved motor recovery (Pooled-Risk = 2.08, 95%CI: 1.68-2.58, p < 0.001) and decreased risk of mortality (improved survival) (Pooled-Risk = 0.72, 95%CI: 0.69-0.76, p < 0.001). Although the literature consistently reported that longer symptom duration was associated with better ambulatory function and a reduced risk of local recurrence, the small number of studies precluded a meta-analysis and pooled-risk. Surgery: Symptom duration was defined as the time from onset of neurological deficit to surgery in 7/11 studies (64%); three studies (27%) used ambulatory status, and one study (9%) used both definitions. In the surgical cohort, longer duration of symptoms was associated with increased risk of death (worse survival) (Pooled-Risk = 1.28, 95%CI: 0.54-3.03, p = 0.575), although it was not statistically significant. Though a meta-analysis and pooled risk could not be conducted for motor recovery due to small numbers, most studies demonstrated that longer duration of symptoms was associated with worse motor recovery, although findings regarding ambulatory function were inconsistent. Covariate analysis revealed visceral metastases were associated with lower rates of positive outcomes in both surgical and RT cohorts. Conclusion: Symptom duration significantly influenced outcomes in MESCC, with opposing effects by treatment modality. Longer duration was associated with improved survival and motor recovery in the RT cohort, whereas it was associated with decreased survival in the surgical cohort, though this was not statistically significant. Most surgical studies demonstrated an inverse association between symptom duration and motor recovery. Visceral metastases consistently predicted lower rates of favorable outcomes across both treatment groups. These findings highlight the importance of tailoring intervention strategies not only to symptom trajectory but also to underlying disease burden, supporting an individualized, modality-specific approach to the management of MESCC.

ID: AGS26/1302

A443: Utility of intraoperative magnetic resonance imaging (iMRI) in spinal intramedullary tumor surgery

Wei-Ting Lee 1 , Yen-Heng Lin 2 , Yu Cheng Huang 2 , Keng-Chian Lin 2 , Dar-Ming Lai 1 , Fon-Yih Tsuang 1,3

1 Division of Neurosurgery, Department of Surgery, National Taiwan University Hospital, Taipei, Taiwan

2 Division of Neuroradiology, Department of Medical Imaging, National Taiwan University Hospital, Taipei, Taiwan

3 Spine Tumor Center, National Taiwan University Hospital, Taipei, Taiwan

Introduction: Spinal intramedullary tumors are uncommon and present significant surgical challenges due to their delicate anatomical location within the spinal cord and their frequently ill-defined margins. Preservation of neurological function while achieving maximal tumor resection remains the primary surgical goal. Intraoperative magnetic resonance imaging (iMRI), which has proven beneficial in brain tumor surgery, offers real-time imaging that may assist in assessing the extent of resection and guiding further surgical decisions. This study evaluates the clinical utility and feasibility of iMRI during spinal intramedullary tumor resection, with an emphasis on the extent of resection, tumor localization, and intraoperative guidance. Material and Methods: We conducted a retrospective case series at a single tertiary referral center, enrolling 22 patients who underwent spinal intramedullary tumor resection with iMRI guidance between April 2022 and August 2025. Preoperative assessments included conventional MRI and diffusion tensor imaging (DTI) in selected cases for tumor characterization and surgical planning. Intraoperatively, iMRI was performed after initial resection to evaluate residual tumor and to guide further resection where feasible. Postoperative MRI was conducted during follow-up to assess resection extent and early outcomes. Results: Among the 22 cases, tumors were located in the cervical (36.4%), thoracic (50.0%), and lumbar (13.6%) spine. Histopathological diagnoses included ependymomas, hemangioblastomas, astrocytomas, and other tumor types. iMRI detected residual tumors in 8 cases (36.3%) that were aiming for gross total resection, leading to additional resection in 5 cases. The mean operative time was 342.2 minutes (229-535 minutes), with an average iMRI examination duration of 53.5 minutes (30-70 minutes). The mean postoperative MRI follow-up period was 82.7 days. Among the patients who underwent postoperative MRI, no residual tumor was detected during follow-up, except for one case where recurrence occurred at a different spinal level. Case demonstrations are included to illustrate the clinical application of iMRI in guiding both additional resection and localization of subtle lesions. Notable imaging artifacts included air bubbles, hematomas, and contrast leakage. Conclusion: iMRI is a valuable tool in the resection of spinal intramedullary tumors, providing real-time intraoperative information that enhances surgical precision and may improves the gross total resection rate. Although not all residual tumors can be safely resected, the integration of iMRI into the surgical workflow can increase resection confidence and potentially improve surgical outcomes. Future research should focus on refining iMRI-integrated surgical protocols and evaluating their long-term impact on patient outcomes.

ID: AGS26/1367

A444: Incidence and predictors of local recurrence following surgical management of spinal metastases

Zainah Moukahhel 1,2 , Olivia Zhou 1,2 , Leonardo Olijnyk 1,2 , Jesse Shen 1,2 , Fidaa Al-Shakfa 1,2 , Zhi Wang 1,2

1 Université de Montréal, Centre Hospitalier de l'Université de Montréal, Montréal, Canada

2 Centre Hospitalier de l'Université de Montréal, Montréal, Canada

Introduction: Spinal metastases represent a common complication of advanced malignancies and remain a significant source of morbidity. Although recent advances in surgical technique and adjuvant therapies have improved the prognosis, local recurrence after operative management continues to pose a major clinical challenge. A better understanding of this population is essential to guide treatment strategies and offer better quality of life. In this context, we evaluated the incidence, associated risk factors, and outcomes of spinal metastases in patients treated surgically at the Centre Hospitalier de l’Université de Montréal (CHUM). Material and Methods: We retrospectively reviewed all patients who underwent surgery for histologically confirmed spinal metastases at CHUM and had the first operation between January 2015 and October 2024. In patients with surgeries at distinct, non-adjacent levels, each operated level was analyzed independently for recurrence at that site. Collected data included demographics, primary tumor type, spinal level, surgical modality, time to recurrence, and survival time. Recurrence was defined as symptomatic and radiological tumor regrowth at the same operated level. Statistical analysis was performed using SPSS v.25. Results: A total of 377 patients were included in the analysis. Mean overall survival of 541 days (95% CI 482-600). The average age at the time of surgery was 63.2 years. Among all patients, 70, 178 and 113 were smokers, non-smokers and former smokers respectively (18.6%, 47.3% and 30%). Symptomatic local recurrence occurred in 8.5% of patients (n = 32), including 5.1% (n = 19) within the first 6 months, 1.1% (n = 4) between 6-12 months, and 2.7% (n = 9) beyond 1 year postoperatively. Recurrence rates varied significantly by spinal level (p = 0.011), with similar rates observed in the cervical (8%, 15/175) and thoracic (7%, 12/155), but higher in the lumbar spine (30%, 3/7). Recurrence ranged from 0% to 50% according to cancer types. Although there was a tendency of association with primary tumor, it did not reach statistical significance (p = 0.054). The highest rates were seen in esophageal (50%, 2/4), pancreatobiliary (33%, 2/6), neuroendocrine (14%, 1/7), and prostate tumors (13%, 8/60). Lung cancer was the most common operated primary tumor (28%, n = 104), yet recurrences occurred in 7.7% of patients (8/104). Breast cancer was the second most frequent (19%, n = 72), with 4.2% of patients experiencing recurrence (3/72). Renal cancer represented 11% of cases (n = 40) and showed a recurrence of 10% (4/40). No demographic, operative (including surgical approach and type), or radiotherapy-related factors (pre-operative or post-operative) were associated with recurrence. Conclusion: Symptomatic local recurrence after surgery for spinal metastases was uncommon (<10%), but significantly higher in the lumbar spine. Primary tumor type showed a possible, though non-significant, association. Importantly, these findings derive from the largest single-center series to date, with a substantial cohort of 377 patients, enhancing the robustness and generalizability of the results. This study underscores that both compression severity and tumor location should be carefully considered in postoperative risk stratification and surveillance planning.

ID: AGS26/2097

A445: Evaluating the real-world predictive utility of Karnofsky and ECOG performance status for 90-day survival after oncologic surgery for metastatic spinal tumors

Rafael De la Garza Ramos 1 , Ali Haider Bangash 2 , Sertac Kirnaz 2 , Rose Fluss 2 , Victoria Cao 2 , Alexander Alexandrov 2 , Liza Belman 2 , Saikiran Murthy 2 , Yaroslav Gelfand 2 , Reza Yassari 2

1 Montefiore Medical Center, New York, United States

2 Montefiore Einstein, New York, United States

Introduction: Performance status is frequently cited in the literature as an independent predictor of survival following surgery in patients with metastatic spine disease. However, there is limited evidence regarding its standalone predictive performance for short-term postoperative outcomes. In this study, we aimed to directly evaluate how well Karnofsky (KPS) and Eastern Cooperative Oncology Group performance status scores (ECOG-PS) predict 90-day survival-a commonly used threshold for surgical candidacy-in patients undergoing surgery for metastatic spinal tumors. Material and Methods: We conducted a retrospective cohort study of 175 adult patients who underwent surgery for metastatic spine disease at a single institution between 2012 and 2025. All patients had documented preoperative KPS and ECOG-PS scores and at least 90 days of follow-up or death. Univariable logistic regression was used to assess the association between performance status and 90-day survival. Predictive performance was assessed across three domains: discrimination (AUC), calibration (Brier score and calibration plots), and clinical utility (decision curve analysis [DCA]). Results: The crude 90-day survival rate was 73%. Both KPS (OR 1.2, 95% CI 1.01-1.05; p = 0.001) and ECOG-PS (OR 0.51, 95% CI 0.36-0.73; p < 0.001) were statistically associated with survival on univariable analysis. However, discrimination was modest (AUC 0.65 for KPS, 0.68 for ECOG-PS), and calibration was fair (Brier scores 0.185 and 0.182, respectively). DCA revealed minimal net benefit for either score across most threshold probabilities, with ECOG-PS showing slightly better performance at intermediate thresholds (30-60%)-the zone of greatest clinical uncertainty. Conclusion: Despite being widely cited as an independent predictor of postoperative survival in patients with metastatic spinal tumors, performance status assessed via the KPS and ECOG-PS demonstrated only modest discriminatory ability, calibration, and clinical utility when used alone to predict 90-day survival. These findings highlight the important difference between statistical association and real-world clinical utility of a single metric to predict outcome in this patient population.

ID: AGS26/2300

A446: European retrospective multicenter study in patients with vertebral metastases (Train-METASTRA)

Giovanni Barbanti Bròdano 1 , Áron Lazáry 2 , Jorrit-Jan Verlaan 3 , Peter Vajkoczy 4 , Mariarita Intagliata 1 , Cristiana Griffoni 1 , Julia Szita 2 , Netanja Harlianto 3 , Hannah Miller 4 , Serena Moscato 5 , Ariele Santello 5 , Sabato Mellone 5 , Luca Cristofolini 6

1 Complex Spinal Surgery Unit, IRCCS Istituto Ortopedico Rizzoli, Bologna, Italy

2 National Center for Spinal Disorders, Buda Health Center, Budapest, Hungary

3 Department of Orthopedics, University Medical Center Utrecht, Utrecht, Netherlands

4 Department of Neurosurgery, Charité-Universitätsmedizin Berlin, Berlin, Germany

5 Department of Electrical, Electronic, and Information Engineering, University of Bologna, Bologna, Italy

6 Department of Industrial Engineering, University of Bologna, Bologna, Italy

Introduction: Train-METASTRA is a retrospective multicentre study conducted within the framework of the European project METASTRA: “Computer-aided effective fracture risk stratification of patients with vertebral metastases for personalized treatment through robust computational models validated in clinical settings.” Currently, fracture risk assessment in vertebral metastases is imprecise. Clinicians rely on the Spine Instability Neoplastic Score (SINS), which evaluates five radiographic criteria and mechanical pain to classify the spine as stable, potentially unstable, or unstable, with cases requiring referral to spine surgeons. However, SINS alone has low reliability, with sensitivity and specificity < 50%, due to limited parameters, dependence on expert opinion, and small validation cohorts. METASTRA aims to improve the prevention of pathological fractures in patients with vertebral metastases and to provide patient-tailored treatment by developing computational and biomechanical models that reliably stratify patients according to their fracture risk and can be incorporated into a decision support system (DSS). To develop these computational models, a large dataset of clinical and radiographic data from patients with vertebral metastases was collected through the retrospective multicentre study “Train-METASTRA”, carried out in four European clinical centres. Material and Methods: Patients included in the study were ≥ 18 years old, with vertebral metastases in the thoracic or lumbar spine not surgically treated, had a CT scan available at the time of lesion diagnosis and at least one follow-up visit at ≥ 3 months after the first evaluation. Clinical data were collected in pseudo-anonymized form using an electronic case report form (eCRF) based on RedCap, and imaging data were uploaded to a dedicated platform, in compliance with privacy regulations. Data quality was ensured by centralized monitoring across participating centres. SINS scores were collected for the metastatic lesions. For approximately 90% of lesions, SINS was calculated using a standardized RedCap form based on the established parameters: radiographic spinal alignment, vertebral body collapse, posterolateral element involvement, lesion location, bone lesion type, and mechanical/postural pain. For the remaining lesions, SINS values were retrieved from the clinical records, representing scores previously assigned by treating clinicians. Results: The total number of patients enrolled across the participating centres was 2,041, of whom 53.5% were male. The mean age was 65 years. The most common primary tumours metastasizing to the spine were breast cancer (35.05%) and prostate cancer (25.35%). A total of 3,467 lesions were analysed (1,838 thoracic, 1,629 lumbar), predominantly lytic (47.44%). Approximately 77.31% of the lesions were treated with surgery, systemic therapy, and/or radiotherapy: 86.53% remained stable, while 362 lesions (9.39%) progressed to pathological fracture, with 10.40% of these associated with grade 2 epidural spinal cord compression (Bilsky classification). SINS scores were available for 2,490 lesions, with a mean value of 5.18. Among the lesions that progressed to fracture, the mean SINS score was 8.81. Conclusions: Preliminary analysis suggests that while SINS score provides a general indication of instability, it does not fully predict fracture progression, highlighting the need for complementary computational models to improve risk stratification and enable more personalized, targeted interventions. Acknowledgement: METASTRA project funded by the European Union HEU topic HLTH-2022-12-01 grant 101080135.

ID: AGS26/985

A447: Effect of treatment center volume on clinical outcomes for primary malignant intramedullary spinal tumors

Ataollah Shahbandi 1 , Peter Palmer 1 , Kevin Wojcik 1 , Pegah Ghamasaee 1 , Saman Shabani 1

1 Department of Neurosurgery, Medical College of Wisconsin, Milwaukee, United States

Introduction: The impact of facility volume on patient outcomes for primary malignant intramedullary tumors remains unclear. This study aimed to evaluate survival differences based on facility volume, analyze variations in treatment approaches, and assess the independent association between facility caseload and survival as well as perioperative outcomes. Material and Methods: This retrospective cohort study utilized data from the National Cancer Database from 2004 to 2022. Patients diagnosed with primary malignant spinal intramedullary tumors were included in the present study. To determine the optimal caseload cutoff for survival prediction, maximally selected rank statistics were employed, revealing a threshold of 32 patients spanning the period from 2004 to 2022. Patients were categorized based on facility volume: high-volume facilities (> 32 cases over the study period) and low-volume facilities (≤32 cases). The primary outcome was overall survival (OS), while secondary outcomes included treatment utilization and perioperative outcomes in surgically treated patients. Results: A total of 3,667 patients were included, with 502 treated at high-volume facilities and 3,165 at low-volume facilities. Patients treated at high-volume facilities were significantly younger (46.8 ± 15.2 years versus 49.5 ± 15.8 years; p < 0.001) and had a significantly higher proportion of males (56.2% versus 50.4%; p = 0.018) compared to those at low-volume facilities. Patients treated at high-volume facilities were significantly more likely to have private insurance or managed care (73.6% versus 59.7%), whereas those at low-volume facilities had significantly higher rates of government-sponsored insurance (36.7% versus 25.6%) or were uninsured (3.6% versus 0.8%; p < 0.001). Additionally, a significantly greater proportion of patients at high-volume facilities resided in ZIP codes with a median household income of ≥ $63,000 (44.1% versus 34.4%; p < 0.001) and in metropolitan areas (78% versus 74.4%; p = 0.017). Treatment at high-volume facilities was associated with significantly improved OS (p < 0.001), with a 10-year survival rate of 81.4% compared to 69.8% at low-volume facilities. While surgical procedures, radiation therapy, and systemic therapy utilization remained stable at high-volume facilities, low-volume facilities exhibited a significant increase in surgical procedures (rate ratio [RR] 1.021, 95% confidence interval [CI] 1.011-1.032; p < 0.001) and a significant decline in radiation therapy use (RR 0.971, 95%CI 0.955-0.987; p < 0.001). After adjusting for confounding variables, treatment at high-volume facilities remained independently associated with lower mortality risk (hazard ratio 0.737, 95%CI 0.578-0.94; p = 0.014), a reduced likelihood of receiving radiation therapy (odds ratio [OR] 0.539, 95%CI 0.413-0.697; p < 0.001), and significantly lower 30-day readmission rates following surgical resection (2.1% versus 6.9%; OR 0.313, 95%CI 0.147-0.586; p = 0.001). Conclusion: Despite the rarity of primary malignant intramedullary tumors, treatment at high-volume facilities was associated with significantly improved survival and lower 30-day readmission rates following surgical resection, even after adjusting for potential confounders. These findings underscore the importance of specialized centers in optimizing patient outcomes and can inform referral patterns and treatment strategies for these complex tumors.

ID: AGS26/2917

A448: Proposal of a novel approach to minimally invasive stabilization and short-course neoadjuvant denosumab treatment before en bloc resection in enneking stage 3 giant cell tumors of the mobile spine

Chiara Cini 1 , Stefano Boriani 2 , riccardo ghermandi 1 , Valerio Pipola 1 , Stefano Bandiera 1 , Marco Girolami 1 , Giuseppe Tedesco 1 , Silvia Terzi 1 , Giovanni Barbanti Bròdano 1 , Alessandro Gasbarrini 1,2

1 Istituto Ortopedico Rizzoli, Bologna, Italy

2 ALMA Mater Studiorum, Bologna, Italy

Introduction: En bloc resection of giant cell tumor (GCT) is technically more demanding as this tumor is typically soft and friable. The risk of breakage of the tumor mass during the specimen manipulation is consistent with possibility of inadequate oncological margins. A few studies investigated the technical limits of en bloc resection and the role of Denosumab in less accessible GCTs of spine. The aim of our study is to compare a cohort of consecutive cases of GCTs of the mobile spine treated with a minimally invasive stabilization (MIS) followed by neoadjuvant Denosumab treatment (NDT) before en-bloc resection and patients treated with NDT followed by en-bloc resection. Material and Methods: This retrospective observational study analyzed 17 patients with Enneking stage 3 GCTs of mobile spine with intermediate spinal instability neoplastic score (SINS). 8 patients (Group 1) were treated with MIS accompanied by transpedicular biopsy at the onset and started NDT for at least two months. After NDT, en-bloc resection was performed in all cases. In 9 patients (Group 2) diagnosis was confirmed by CT guided trocar biopsy followed by NDT and subsequent en-bloc resection. Dosage was always 120 mg subcutaneous each month with loading doses in the first month on days 1, 8, 15, 22. The morphologic changes (linear dimensions, regional segmentations, tumor volume and total vertebral body volume) in GCTs following NDT were analyzed quantitatively using CT scan before en-bloc resection. The MRI Signal Intensity (SI) ratios of tumor/muscle was calculated. Results: The mean NDT duration was 4.25 ± 3.2 months in Group 1 and 7.4 ± 3.2 in Group 2. The CTscan analysis showed a significant mean reduction of posteromedial vertebral compartments after NDT in both groups [16.15 ± 1.46 mm to 11.87 ± 1.72 mm (p < 0.001)] suggesting localized densification in structural regions. Both groups achieved a significant decrease in tumor volume [10.94 ± 1.23 cm 3 to 7.32 ± 1.43 cm 3 (p = 0.0156)] and a reduction in total vertebral body volume [16.92 ± 2.11cm 3 to 15 ± 1.29 cm 3 (p = 0.0225)] aligning with the known shrinkage effect of Denosumab. No significant difference in transverse diameters reduction was found comparing the two groups (p = 0.9374), suggesting lateral stability. The values of Hounsfield units of vertebral bodies increased by > 50% in both groups. SI ratios of tumor/muscle in T2WI of plain MRI decreased by > 25% in all cases. Among the SINS categorical scores, radiographic alignment showed a similar distribution with most patients classified as baseline score 0 (p = 0.432). Body collapse score significantly increased in Group 2 with 2 preoperative pathological fractures associated with segmental kyphosis. The operative time was lower in Group 1 (495 ± 138 vs 575 ± 220 min). No evidence of mechanical complication was observed in Group 1 at the final follow-up (78 ± 2.3 months). A rod fracture occurred in one patient of Group 2 who underwent revision surgery at 11 months of follow-up. 2 (22.2%) patients of Group 2 experienced distal junctional failure and required revision surgery. No evidence of LR was found in our population. Conclusion: MIS associated with Denosumab as a planned neoadjuvant treatment offers advantages in preoperative surgical planning and downstage the surgical procedure preventing segmental kyphosis allowing a better reconstruction to reduce the risk of mechanical complication.

ID: AGS26/662

A449: Emergent vs. delayed decompression in cauda equina syndrome secondary to tumors: what does the evidence say?

Humaid Al Farii 1

1 Sultan Qaboos University, Muscat, Oman

Background: Cauda equina syndrome (CES) is a severe neurological condition resulting from compression of the cauda equina nerve roots. Its prevalence secondary to tumors is relatively low but presents significant diagnostic and treatment challenges. CES can lead to serious complications, including permanent neurological deficits, bladder and bowel dysfunction, and chronic pain. In the case of disc herniation, there is a well-established body of literature showing that delayed surgery worsens neurological recovery due to irreversible damage to these compressed nerves. Understanding the impact of surgical timing on patient outcomes in tumor-related CES is crucial, as it can guide clinical decision-making and optimize surgical planning. If evidence supports that delayed intervention does not compromise neurological outcomes, this allows the opportunity to conduct thorough preoperative evaluations, obtain tissue biopsies to guide treatment, and plan for complete tumor resection with appropriate instrumentation. Thus, the aim of this study is to examine the relationship between surgical timing and patient outcomes in tumor-related CES. Materials and Methods: PubMed and Embase databases were used to conduct searches for relevant articles evaluating the effect of surgical timing on patient outcomes in tumor-related CES. Included studies were those of CES with neoplastic involvement, wherein the outcomes were evaluated and timing (emergent vs. delayed) of the operations were mentioned. Excluded articles were non-surgical studies and those with no outcome data. A total of 36 patients were included in the study of whom satisfied the inclusion criteria. Results: There was a male predominance (21/36, 58.3%), and the patients’ mean (standard deviation [SD]) age was 46.75 (15.8) years. The most common tumor type were paragangliomas (14/36, 38.9%), and the most common presenting symptoms included pain (35/56, 97.2%), weakness (25/36, 69.4%), and incontinence (16/36, 44.4%). Of these patients, 29 of them underwent an emergent operation (80.5%), while 7 of them underwent an elective operation (19.4%). All elective procedures yielded persistent symptoms, while of the emergent operations, 3/29 (10.3%) passed away due to disseminated disease, 6/29 (20.7%) fully recovered, and 20/29 (69%) had residual symptoms. Mean (SD) follow-up time was 21.2 (33.4) months. Conclusions: Our review found that emergent operations reduce the risk of symptom persistence in tumor-related CES. However, it is worth noting that there is an underwhelming body of literature assessing the time interval between hospital admission and initiation of surgery in CES caused by tumors. Most studies included here only referenced whether the admission was emergent or elective, rather than precise quantitative measures of time. Thus, we advocate for more specific research evaluating how surgical timing specifically affects outcomes in tumor-related CES.

ID: AGS26/706

A450: Knowledge, attitudes and practices of intraoperative autologous blood cell salvage in tumour-related spine surgery: a survey of singapore spine society members

Hui Zhen Grace Chua 1 , Derek Chao Kuan Twan 1 , Yeong Huei Ng 1 , Naresh Kumar 2 , Jiong Hao Tan 2 , Youheng Ou Yang 1

1 Singapore General Hospital, Singapore, Singapore , 2 National University Hospital, Singapore, Singapore

Background: Intraoperative autologous blood cell salvage (IOCS) allows blood lost intraoperatively to be processed and re-transfused, reducing allogenic blood transfusion requirements. IOCS used in tumour-related spine surgery is limited due to concerns of tumour-cell reinfusion. Recent studies suggest that the risk of reintroduction of viable tumour cells is low and meta-analyses have not found increased tumour recurrence. This study aims to investigate the knowledge, attitudes and practices of members of the Singapore Spine Society (SSS) regarding usage of IOCS in metastatic spine tumour surgery (MSTS) and primary spinal tumour surgery (PSTS). Material and Methods: An online questionnaire was sent by electronic-mail to all 108 members of SSS in July 2025. The questionnaire included surgeon’s demographics, practice settings, case volumes, availability of IOCS and their support and usage of IOCS in both metastatic and primary spinal tumour surgeries respectively. Participants rated their support for IOCS on a five-point Likert scale, and listed their perceived benefits and concerns with usage of IOCS. Results: A total of 36 valid responses were received. Majority were Orthopaedic surgeons (83.3% vs 16.7% Neurosurgeons) working in public hospitals (69.4%), with more than 10 years of working experience as a spinal surgeon (55.6%). Most (91.7%) work in institutions that are able to provide IOCS, of which 72.7% had IOCS available for more than five years. The main perceived benefits of IOCS across both MSTS and PSTS are reduction in allogenic transfusion (86.1% MSTS vs 75.0% PSTS), decreased morbidity/complications (44.4% vs 50.0%), cost savings (38.9% vs 33.3%), and decreased mortality (27.8% vs 22.2%). Concerns regarding its use include potential systemic tumour seeding (63.9% MSTS vs 69.4% PSTS), coagulopathy (52.8% vs 38.9%), infection (36.1% vs 25.0%), and high costs (27.8% vs 27.8%). In MSTS, 77.8% of surgeons supported using IOCS but only 50.0% reported clinical usage. Mean support ratings were significantly higher in public-hospital surgeons (4.24 vs 2.55 private-hospital, p-value < 0.0001), surgeons performing more than ten MSTS cases a year (4.16 vs 3.24, p-value = 0.003), and those who use it in clinical practice (4.50 vs 2.94, p-value < 0.0001). This is compared to PSTS, where 58.3% of surgeons support its use, but only 22.2% use it. Seventeen surgeons had not performed surgery for PSTS before. Mean support ratings in PSTS were significantly higher in surgeons who use it (4.38 vs 2.61, p-value < 0.0001), but not significant based on practice settings (3.20 public vs 2.55 private, p-value=0.068). Overall mean support ratings were higher for MSTS (3.72) than PSTS (3.00, p-value = 0.034). Conclusion: The surveyed members of the SSS expressed support for IOCS use in both MSTS and PSTS, with MSTS receiving significantly higher support. Higher surgical volume and personal experience with IOCS significantly increased support for IOCS in both MSTS and PSTS. The greatest perceived benefit was a reduction in allogenic transfusion, while potential systemic tumour seeding was the most common concern. Research focused on defining the actual risk of potential systemic tumour seeding, and promoting education of the existing evidence base which suggests that this risk is low, may help spur IOCS adoption in clinical practice.