ABBREVIATIONS:
hemilaminectomy and microdiscectomy
length of stay.
Lumbar radiculopathy secondary to disk herniation or protrusion is one of the most commonly treated neurosurgical conditions. Conservative pain management and transforaminal epidural steroid injections remain the mainstay of first-line therapies, 1 with surgery as second line to refractory pain or for treatment of progressive neurological deficits. Initially treated through an open-only approach, introduction of the surgical microscope to the procedure led to the development of the lumbar hemilaminectomy and microdiscectomy (HLMD) in the 1970s. The widespread adoption of HLMD has reduced operative times and length of stay (LOS). 2 As surgical techniques were further refined in the 1980s, outcomes of microdiscectomy superseded those of open discectomy, with lengths of stay being halved and the average time for return to work significantly shortened. 3 In most current neurosurgical practices, lumbar HLMD has become an outpatient procedure, with demonstration of excellent cost-effectiveness and patient outcomes with this model. 4 Furthermore, with even greater focus in the past 10 years transitioning to societal cost-benefit surgery, there has been a push to transition all 1-level and 2-level lumbar decompression surgeries to outpatient, with evidence of noninferior patient outcomes and significantly reduced hospital costs. 5
However, incidental durotomy, a relatively common complication of lumbar decompression surgery, often precludes patients being sent home on the day of surgery. 5 Incidental durotomy is cited to occur in 3% to 27% of lumbar decompression surgeries, with a higher incidence in revision surgeries due to adhesive scar tissue. 6 - 9 Most studies and centers use a protocol of bedrest followed by gradual mobilization, requiring an inpatient stay after surgery. However, there is very little evidence to support this management protocol. In fact, recent investigations have demonstrated improvements in complication rates and LOS, as well as reduced health care costs without an increase in reoperation rates with early mobilization less than 24 hours after durotomy. 10 - 12
The paradigm of same-day discharge after primary repair of incidental durotomies has been proposed in minimally invasive lumbar surgery with success. 13 However, existing studies on treatment protocols for incidental durotomies are comprised mixed patient cohorts, including cervical, thoracic, and lumbar surgeries, and investigate a combination of multi-level and single-level surgeries as well as a combination of both decompression-only and decompression and fusion surgeries. Anticipated LOS, even within lumbar degenerative surgeries, differs significantly between the type of surgery being performed. Therefore, the aim of this study was to investigate all patients undergoing only single-level lumbar HLMD by a single surgeon over an extended period of time with a practice of sending patients home on postoperative day (POD) 0, even after incidental durotomy repair during index surgery. We hypothesized that patients with incidental durotomies could be safely sent home without any bedrest and would not have higher rates of readmissions or reoperations compared with the patients without durotomies.
METHODS
The medical records of patients undergoing single-level lumbar HLMD from 2013 to 2021 by a single surgeon at our academic institution were identified in our institutional complex spine database based on the review of the electronic medical record. Institutional review board approval was obtained (Approval Number 855679). The study was deemed to be minimal risk by the Institutional review board, and therefore, patient consent was not required. Patients were included if they were older than 18 years and undergoing single-level–only HLMD. Patients were excluded if they underwent multi-level surgeries, complete laminectomies, or placement or removal of instrumentation in addition to the discectomy. Preoperative variables collected included patient age, body-mass index, basic comorbidities, and whether they were undergoing revision surgery. Intraoperative variables included procedure duration, levels of procedure, and whether or not there was a durotomy. All durotomies in this series were repaired primarily with either 4-0 silk or 4-0 braided nylon suture (Nurolon, Ethicon Inc.). Postoperative variables collected included LOS, 30-day readmission, 90-day readmission, and reoperations within 1 year. Data were analyzed in Excel (Microsoft), and descriptive statistics were used.
The immediate postoperative management of a primarily repaired intraoperative spinal fluid leak at our institution is as follows: In the presence of a satisfactory primary repair, patients are returned to the recovery room after their procedure. In the immediate postoperative period, they are positioned supine. There head of the bed is gradually elevated, as tolerated, and the patient is monitored for headaches and neurological symptoms. If a patient develops headaches, they are again placed supine. They are gradually progressed to sitting followed by assisted ambulation. Patients are considered suitable for discharge home after a spinal fluid leak if a satisfactory primary repair has been achieved, and they are ambulatory within hours of the procedure without symptoms of an active leak. Typically, patients have a follow-up phone call from a nurse practitioner 1 to 2 days postoperative to monitor their progress. However, if they are unable to maintain an upright posture without headaches or dizziness, the patient is admitted for further expectant treatment. If the symptoms persist, an MRI scan would be performed, and if an active leak is present, a re-exploration is considered or the placement of a lumbar drain. The determination to re-explore is based on the judgment of the surgeon as to the difficulty in achieving the primary closure on the first attempt and whether it is thought that this could be optimized by second attempt to achieve closure.
RESULTS
From 2013 through 2021, 338 unique patients underwent 346 independent single-level lumbar HLMD with a single surgeon at our institution. Average age at surgery was 52 (SD = 15) years. Seventeen patients were undergoing revision surgeries. The mean body-mass index was 27 (SD = 4.8).
Among these surgeries, there were 17 durotomies (4.9%). For the 17 patients undergoing revision surgery for the index operative level, there were 3 durotomies (17.6%). The median LOS for patients who had a durotomy was 0 (IQR 1). Similarly, the median LOS for nondurotomy patients was 0 (IQR 0). For the entire patient population, 278 surgeries resulted in discharge on day 0 (80%), 31 surgeries resulted in discharge on day 1 (9%), with the remaining 37 patients remaining inpatient for >24 hours (10.7%). All but 5 patients with durotomy returned home on POD 0 (70.5%). All patients who remained inpatient after durotomy remained for reasons unrelated to the dural tear. One patient had significant preoperative comorbidities including active multiple myeloma with end-stage renal disease and remained inpatient for 17 days for postoperative delirium and chronic kidney disease management. Three patients remained 1, 2, and 5 days, respectively, for pain control issues arising in the postoperative setting. The fifth patient was kept inpatient until POD 1 for observation due to significant bleeding intraoperatively.
Within the durotomy group, there were two 90-day readmissions (11.8%) and 1 within 30 days (5.9%). One of these patients was readmitted for bacteremia secondary to chemotherapy infusions for myeloma and was the same patient who remained for 17 days after his index surgery. The other patient was readmitted for a superficial wound infection requiring a superficial washout. In the nondurotomy group, there were 5 readmissions within 30 days (1.4%) and 16 readmissions within 90 days (4.5%).
In addition, for surgeries complicated by durotomy, 2 required reoperation within 1 year (11.8%). One reoperation was for the previously mentioned superficial infection, while the other was for pseudoarthrosis and disk herniation at a different spinal level from the index surgery. Within the nondurotomy group, there were 27 reoperations within 1 year (7.8%). Given the small sample size, no statistically significant differences were found between the durotomy and nondurotomy groups regarding LOS, readmission, and reoperation.
DISCUSSION
Lumbar discectomy is usually a straightforward procedure with excellent results for treatment of radiating leg pain. As with any procedure, there are specific risks inherent to the intervention. Complications during HLMD include neurological injury to an isolated nerve root or cauda equina, a spinal fluid leak, or induction of spinal instability. Of these complications, spinal fluid leak is usually evident during the procedure and requires intraoperative judgment regarding management. The optimal treatment for an intraoperative spinal leak is primary repair with suture. The limitations to achieving a primary repair include the location of the tear on the nerve root and/or thecal sac, the quality and consistency of the dura, and access to the tear with a needle and suture within the confines of the laminotomy. Tamponade of the leak with packing or application of physiological or synthetic glues is other option. The ultimate effectiveness of tamponade or application of glue to secure the repair is less certain and able to be controlled and monitored postoperatively. A secondary technique to salvage an ineffective primary repair is a meticulous fascial closure to contain the spinal fluid to the subfascial space. A watertight fascial closure reduces the risk of incurring infection, and possibly meningitis, in addition to limiting the volume of potential space available in which an active leak can accumulate. The limited soft tissue dissected in HLMD further reduces the volume of potential space for the spinal fluid to accumulate and reduces the likelihood of postural symptomatology from an active leak. Although bedrest does provide short-term symptomatic relief for an active leak, it is unclear if bedrest improves the success rate of an adequate primary repair or if it can be a useful adjunct to promote healing of an active pseudomeningocele. Spinal fluid diversion with a lumbar drain can be used to facilitate healing/closure of a suboptimal primary repair or in the presence of an active leak.
The complications associated with an active spinal fluid leak include postural headaches, radicular pain from a herniated nerve root through the dural perforation, impaired wound healing, persistent drainage, and localized wound infection or meningitis. Wound-related complications and meningitis typically do not occur within the first few days after surgery. Headaches and radicular pain from a pseudomeningocele are not associated with catastrophic complications in the short term.
The shift in modern health care toward high value care has emphasized the importance of cost-effectiveness and decreasing hospital-associated costs while maintaining or improving patient outcomes. Lumbar HLMD is a common neurosurgical procedure associated with significant improvement in patient quality of life and rapid return to work. 3 With advancement in surgical techniques, anesthetic techniques, and perioperative patient optimization, lumbar HLMD has transitioned to a standard outpatient surgery. 4 , 5 , 14 , 15 However, incidental durotomy remains a common complication of lumbar decompression surgeries. 6 - 9 Growing recent evidence suggests that the traditional management of durotomies with bedrest and slow liberalization and mobilization is unnecessary and often associated with greater incidence of complications and unnecessary health care costs. 10 - 12 In this study, we propose that same-day discharge in patients undergoing single-level lumbar HLMD with primarily repaired durotomies is both safe and feasible.
Our patient cohort had an incidental durotomy rate of 4.9%, in keeping with rates in comparable published studies of about 3% to 27%. No patients in our population required reoperations for complications related to cerebrospinal fluid leak. All incidences of 90-day readmission and 1-year reoperation in the durotomy group were similarly unrelated to cerebrospinal fluid leak, as described in the Results section above. These findings provide a helpful evidence base for our continued institutional protocol of same-day discharge for patients undergoing HLMD, even for patients with primarily repaired durotomies.
In this study, all durotomies in the primary surgeon's cases were repaired primarily using 4-0 suture. This raises the question of whether or not the method of dural repair influences the safety of same-day discharge. Both direct and indirect dural repair techniques have been popularized. 16 Limited comparative studies have shown no significant differences in durotomy-related complications whether repaired with direct suturing techniques, indirect patching techniques, or a combination of the two. 17 - 19 This is particularly relevant with the growth of minimally invasive and endoscopic approaches for lumbar discectomies, where watertight suture closure of dural defects may not always be feasible. Although this study cannot address the influence of these differences in operative approach and durotomy repair, these data suggest that similar outcomes would likely be observed for patients undergoing indirect dural repairs as well. In addition, it will become important to address whether performing multi-level decompressions or full laminectomies affect the safety of same-day discharge after durotomy in lumbar decompression surgery. There is a theoretical concern that the larger potential space associated with bilateral, multi-level exposures with greater extent of bony decompression may lead to a higher incidence of cerebrospinal fluid hypotension symptoms and pseudomeningoceles requiring operative repair.
Limitations
This study was limited by its retrospective nature, which may lead to limited assessment of underlying confounders. In addition, only 17 incidences of durotomy were identified in our population, limiting further statistical analysis.
In an attempt to increase the number of patients w could be incorporated in the series, we reviewed management techniques and results of 2 of the senior author's colleagues. One of the surgeons had 8 dural perforations of 272 cases, 2 of which had no spinal fluid leak. All were repaired with suture and/or muscle patch. Two patients were observed overnight, and 6 patient were admitted (from 2 to 8 days). Another surgeon had 4 durotomies of 106 discectomies over the same interval of time. Instead of a primary suture repair, the leaks were treated with DuraSeal® (Codman Integra), with or without a muscle graft. Two patients were sent home the same day, 1 was admitted, and 1 had a subsequent readmission with symptoms of cerebrospinal fluid hypotension. The patient who was admitted had a protracted hospitalization for respiratory failure that pre-existed the surgery.
Because of the different surgeon-dependent techniques for intraoperative and postoperative management, these additional patients were not added directly to our results' data, to avoid introducing variables to the study which were otherwise controlled for with a uniform approach by a single surgeon. The repair techniques were different, and the decision on overnight observation varied between surgeons. These data confirm the very low incidence of this complication and reinforce the concept that there is significant variability regarding the management of intraoperative cerebrospinal fluid leaks among neurosurgeons. In contrast to the other management approaches, our data suggest that primary watertight suture closure is an effective technique for treating durotomies, which precludes overnight observation.
Based on the results of this study, we aim to establish a clear protocol for same-day discharge in these patients across our entire institution. We also aim to expand this protocol to patients undergoing minimally invasive lumbar discectomy, endoscopic lumbar discectomy, and those undergoing both unilateral and bilateral decompressions at 1 or 2 lumbar levels. Further studies of patients undergoing more extensive decompressions will help clarify the limits of such policies.
CONCLUSION
Overall, the results of this study indicate that same-day discharge remains safe in patients undergoing single-level HLMD who have an incidental durotomy that is primarily repaired. We consider it safe to discharge the patient home after an adequate primary repair if the patient remains asymptomatic. If they developed a recurrent leak, this would manifest as a positional headache or, much less likely, new onset radiculopathy from nerve root herniation or extrinsic pressure on the thecal sac or nerve from the pseudomeningocele. The other differential diagnosis for postoperative radicular pain would be a recurrent disk herniation. No catastrophic neurological problem should result from persistent or recurrent spinal leak in the immediate postoperative period. If the patient develops recurrent symptoms, they should be readmitted for symptomatic management and a follow-up MRI.
Acknowledgments
Author Contributions: JLG: Data Collection, Data Analysis, Manuscript Composition, Manuscript Review. MS: Conception, Data Collection, Manuscript Review. CW: Data Collection, Manuscript Composition, Manuscript Review. NR: Conception, Manuscript Review. WW: Conception, Manuscript Review. PM: Conception, Data Collection, Data Analysis, Manuscript Composition, Manuscript Review.
Funding
This study did not receive any funding or financial support.
Disclosures
The authors have no personal, financial, or institutional interest in any of the drugs, materials, or devices described in this article.
REFERENCES
1. Wilby MJ Best A Wood E Surgical microdiscectomy versus transforaminal epidural steroid injection in patients with sciatica secondary to herniated lumbar disc (NERVES): a phase 3, multicentre, open-label, randomised controlled trial and economic evaluation . Lancet Rheumatol. 2021 ; 3 ( 5 ): e347 - e356 . 33969319 10.1016/S2665-9913(21)00036-9 PMC8080892
2. Caspar W . A new surgical procedure for lumbar disc herniation causing less tissue damage through a microsurgical approach . Adv Neurosurg. 1977 : 74 - 80 . doi.: 10.1007/978-3-642-66578-3_15
3. Zappulla RA Hollis P Ryder J Noncortical origins of the spinal motor evoked potential in rats . Neurosurgery. 1988 ; 22 ( 5 ): 846 - 852 . 3380273
4. Linhares D Fonseca JA Ribeiro da Silva M Cost effectiveness of outpatient lumbar discectomy . Cost Eff Resour Alloc. 2021 ; 19 ( 1 ): 19 . 33771175 10.1186/s12962-021-00272-w PMC8004396
5. Mooney J Michalopoulos GD Zeitouni D Outpatient versus inpatient lumbar decompression surgery: a matched noninferiority study investigating clinical and patient-reported outcomes . J Neurosurg Spine. 2022 ; 37 ( 4 ): 485 - 497 . 35523251 10.3171/2022.3.SPINE211558
6. Khan MH Rihn J Steele G Postoperative management protocol for incidental dural tears during degenerative lumbar spine surgery: a review of 3,183 consecutive degenerative lumbar cases . Spine (Phila Pa 1976). 2006 ; 31 ( 22 ): 2609 - 2613 . 17047553 10.1097/01.brs.0000241066.55849.41
7. McMahon P Dididze M Levi AD . Incidental durotomy after spinal surgery: a prospective study in an academic institution . J Neurosurg Spine. 2012 ; 17 ( 1 ): 30 - 36 . 22540168 10.3171/2012.3.SPINE11939
8. Tafazal SI Sell PJ . Incidental durotomy in lumbar spine surgery: incidence and management . Eur Spine J. 2005 ; 14 ( 3 ): 287 - 290 . 15821921 10.1007/s00586-004-0821-2 PMC3476743
9. Baker GA Cizik AM Bransford RJ Risk factors for unintended durotomy during spine surgery: a multivariate analysis . Spine J. 2012 ; 12 ( 2 ): 121 - 126 . 22342249 10.1016/j.spinee.2012.01.012 PMC3299921
10. Najjar E Hassanin MA Komaitis S Karouni F Quraishi N . Complications after early versus late mobilization after an incidental durotomy: a systematic review and meta-analysis . Eur Spine J. 2023 ; 32 ( 3 ): 778 - 786 . 36609888 10.1007/s00586-023-07526-6
11. Verma K Freelin AH Atkinson KA Graham RS Broaddus WC . Early mobilization versus bed rest for incidental durotomy: an institutional cohort study . J Neurosurg Spine. 2022 ; 37 ( 3 ): 460 - 465 . 35303709 10.3171/2022.1.SPINE211208
12. Farshad M Aichmair A Wanivenhaus F Betz M Spirig J Bauer DE . No benefit of early versus late ambulation after incidental durotomy in lumbar spine surgery: a randomized controlled trial . Eur Spine J. 2020 ; 29 ( 1 ): 141 - 146 . 31552537 10.1007/s00586-019-06144-5
13. Francavilla TL Weiss MC Davis R . Safe discharge of patients from an ambulatory care center after incidental durotomy during minimally invasive spine surgery . Int J Spine Surg. 2019 ; 13 ( 4 ): 386 - 391 . 31531289 10.14444/6053 PMC6724752
14. Gruskay JA Fu M Bohl DD Webb ML Grauer JN . Factors affecting length of stay after elective posterior lumbar spine surgery: a multivariate analysis . Spine J. 2015 ; 15 ( 6 ): 1188 - 1195 . 24184639 10.1016/j.spinee.2013.10.022
15. Dodo Y Okano I Kelly NA Risk factors for ambulatory surgery conversion to extended stay among patients undergoing one-level or two-level posterior lumbar decompression . Spine (Phila Pa 1976). 2023 ; 48 ( 11 ): 748 - 757 . 36917719 10.1097/BRS.0000000000004621
16. Hassanzadeh H Bell J Bhatia M Puvanesarajah V . Incidental durotomy in lumbar spine surgery; risk factors, complications, and perioperative management . J Am Acad Orthop Surg. 2021 ; 29 ( 6 ): e279 - e286 . 33539059 10.5435/JAAOS-D-20-00210
17. Brazdzionis J Ogunlade J Elia C Wacker MR Menoni R Miulli DE . Effectiveness of method of repair of incidental thoracic and lumbar durotomies: a comparison of direct versus indirect repair . Cureus. 2019 ; 11 ( 7 ): e5224 . 10.7759/CUREUS.5224 31565626 PMC6758957
18. Kamenova M Leu S Mariani L Schaeren S Soleman J . Management of incidental dural tear during lumbar spine surgery. To suture or not to suture? World Neurosurg. 2016 ; 87 : 455 - 462 . 26700751 10.1016/j.wneu.2015.11.045
19. Jo DJ Kim KT Lee SH Cho MG Seo EM . The incidence and management of dural tears and cerebrospinal fluid leakage during corrective osteotomy for ankylosing spondylitis with kyphotic deformity . J Korean Neurosurg Soc. 2015 ; 58 ( 1 ): 60 - 64 . 26279815 10.3340/jkns.2015.58.1.60 PMC4534741