Humeral shaft fractures are often caused by low-energy trauma in elderly patients and mainly related to osteoporosis. 4 Moreover, pathologic fractures also occur more frequently with increasing age. 22 Almost 10% of humerus shaft fractures are metastatic pathologic fractures. 4 The number of patients reaching the metastatic stage is steadily increasing due to the aging population and advanced oncological therapies. 20 Although some cancers, such as renal cell, breast, or prostate cancer, preferentially metastasize to bone, virtually any malignancy has the potential to develop bone metatases. 3 , 12 Osseous metastases are most commonly spread through hematogenous dissemination. 15 , 18 Regardless of the primary tumor location, carcinomas frequently metastasize to the spine, the diaphyseal region of the femur, and the proximal third of the humerus. 9 , 18 These affected bony structures are prone to pathologic fractures. 8
Compound osteosynthesis is one treatment option in patients with impending or actual pathologic humeral fractures and refers to internal plate or nail fixation augmented with bone cement to create a more stable construct. 13 It provides immediate pain relief through reposition and stable fracture fixation ensuring rapid postoperative mobilization. 11 Thus, activities of daily living could be resumed as soon as possible. 11 This technique is mainly used for long tubular bones including the humerus. 13
In cancer patients, postoperative complications leading to reoperation should be avoided as far as possible. 7 The performance of compound osteosynthesis for pathologic fractures of the proximal humerus is not sufficiently covered in the literature. Therefore, the aim of this retrospective study was to review a cohort of patients with compound osteosynthesis for this specific indication. The primary goal was to assess all complications or reoperations and to determine the success of the fracture therapy until death (considering the competing risks of death and the failure of osteosynthesis).
Patients and methods
The survival and complications of compound osteosynthesis for pathologic humeral fractures performed at the Kantonsspital St. Gallen between September 2007 and October 2020 were studied in this retrospective monocentric cohort study. The most recent data on survival were collected on January 1, 2022, and the dataset completed. Inclusion criteria were impending or complete pathologic humeral fracture, primary surgical treatment with compound osteosynthesis, and complete radiological records. As shown in the CONSORT flowchart of patient selection in Fig. 1 , 1 patient who explicitly denied further use of his data was excluded. The local ethics committee granted the approval for use of patient data without their consent for deceased patients and for patients treated before 2015 who could not be contacted. Figure 1 The flowchart of patient selection. LCP , Locking Compression Plate; AP , anteroposterior; T1 , T1-weighted MRI scan.
Surgical techniques and implants
With the patient in a beach-chair position, the affected arm was draped so that it could be held by an assistant and moved freely in all directions. The standard deltopectoral approach (for more proximal metastases) and subsequent distal extension into an anterolateral approach (for more mid-shaft and distal metastases) were used. After deep dissection to the bone, the tumor mass was thoroughly curetted in the sense of an intralesional tumor resection and specimens collected for histology as well as bacteriology. The fractures were anatomically reduced and, if necessary, temporarily fixed with Kirschner wires. A locking plate of appropriate length (LCP or long Philos plate) was then applied, all screws tightened and the positioning controlled using an image intensifier. Bone cement was applied through the fenestration used earlier for curettage, and its intramedullary distribution was controlled with an image intensifier. Alternatively, the screws were partially retracted, the cement was applied, and then the screws were tightened again. Unintentional cement leakage through the screw holes or the defect zone of the pathological fracture were fluoroscopically controlled and removed if necessary. The wound was drained, sutured, and dressed as usual. The postoperative recommendation was passive, and active-assisted mobilization of the shoulder for 6 weeks at the maximum permitted abduction and flexion of 90° to protect the construct.
Fig. 2 illustrates an example of preoperative and postoperative imaging. Figure 2 Case of a 53-year-old male patient with a renal cell carcinoma diagnosed 11 years before the fracture occurred. ( A ) Preoperative anteroposterior view showing osteolytic osseous metastasis metaphyseal below the humeral head. ( B ) Magnetic resonance imaging: Gadolinium-enhanced fat-suppressed T1-weighted coronal view showing bone metastases in the proximal humeral metaphysis with medial minimally compressed pathologic fracture. ( C ) Postoperative anteroposterior view after compound osteosynthesis. The bright, patch-like areas in the bone are cement-filled regions.
Data collection
We created a case report form with all variables necessary for descriptive and survival analysis using competing risk analysis (CRA). Data were extracted from our institutional patient management system containing written reports and radiographs. When needed, missing reports from external follow-up examinations were requested from general practitioners or other hospitals. The data of the follow-up visits were collected and recorded as long as available or until the patient’s death. Final information was collected January 1, 2022.
Statistical analysis
All statistical analyses were performed using R (R: A language and environment for statistical computing; R Foundation for Statistical Computing, Vienna, Austria). Descriptive statistics included counts and proportions for categorical data and means and standard deviation for numeric data. By performing the CRA (R-package “cmprsk”, version 2.2-11), the 2 events “failure of osteosynthesis” and “death” were differentiated within the survival analysis.
Results
This study includes 36 patients [23 men (64%) and 13 women (36%)] treated operatively with compound osteosynthesis of the humerus. Complete fracture was found in 31 patients (86%), whereas 5 patients (14%) had an impending fracture. Demographics and baseline clinical data are in Table I . Table I Patient characteristics. Total N = 36 Complete fracture N = 31 Immediate fracture N = 5 Age, y Median (min, max) 71.6 (47.5, 93.9) 71.5 (47.5, 93.1) 71.7 (52.9, 83.0) Gender (%) M 23 (63.9%) 19 (61.3%) 4 (80%) F 13 (36.1%) 12 (38.7%) 1 (20%) Side (%) L 21 (58.3%) 19 (61.3%) 2 (40%) R 15 (41.7%) 12 (38.7%) 3 (60%) Primary tumor (%) Pulmonary 10 (27.8%) 7 (22.6%) 3 (60.0%) Breast 7 (19.4%) 7 (22.6%) 0 (0.0%) Hematologic 6 (16.7%) 5 (16.1%) 1 (20.0%) Prostatic 4 (11.1%) 4 (12.9%) 0 (0.0%) Other (kidney, gastrointestinal, bone) 5 (13.9%) 5 (16.1%) 0 (0.0%) Primary tumor of unknown origin 4 (11.1%) 3 (9.7%) 1 (20.0%)
Table I
Patient characteristics.
| Total N = 36 | Complete fracture N = 31 | Immediate fracture N = 5 | |
| Age, y Median (min, max) | 71.6 (47.5, 93.9) | 71.5 (47.5, 93.1) | 71.7 (52.9, 83.0) |
| Gender (%) | |||
| M | 23 (63.9%) | 19 (61.3%) | 4 (80%) |
| F | 13 (36.1%) | 12 (38.7%) | 1 (20%) |
| Side (%) | |||
| L | 21 (58.3%) | 19 (61.3%) | 2 (40%) |
| R | 15 (41.7%) | 12 (38.7%) | 3 (60%) |
| Primary tumor (%) | |||
| Pulmonary | 10 (27.8%) | 7 (22.6%) | 3 (60.0%) |
| Breast | 7 (19.4%) | 7 (22.6%) | 0 (0.0%) |
| Hematologic | 6 (16.7%) | 5 (16.1%) | 1 (20.0%) |
| Prostatic | 4 (11.1%) | 4 (12.9%) | 0 (0.0%) |
| Other (kidney, gastrointestinal, bone) | 5 (13.9%) | 5 (16.1%) | 0 (0.0%) |
| Primary tumor of unknown origin | 4 (11.1%) | 3 (9.7%) | 1 (20.0%) |
In our patient cohort, lung carcinoma (27.8%) was the most common primary tumor. In 21 cases (58%), the left humerus was fractured. Approximately ⅔ of the fractures were in the shaft region, ⅓ proximal, and only rarely was the fracture localized more distally. In 50% of the patients, the pathologic fracture occurred within 12 months of the tumor diagnosis. More than 65% of the patients included had already reached the metastatic stage at initial diagnosis. In 5 cases of impending but symptomatic fractures, compound osteosynthesis was performed prophylactically. In 78% of the cases, the fracture extended through the medial and lateral cortex in the conventional AP X-ray. The lateral cortex was affected in 89% of the cases in which only 1 cortex was involved.
Surgical treatment was mainly elective; however, in 8 cases (22%), it was performed as an emergency procedure. The average duration of surgery was 117 minutes. Two different types of plates were used for stabilization during the observed period: the Philos plate (DePuy Synthes, Raynham, MA, USA) and the LCP plate (DePuy Synthes, Raynham, MA, USA). Regarding the number of used plates, 2 plates were used in only 8 cases, mainly from 2018 onward. In the other cases, only 1 plate was used. The postoperative hospital stay ranged from 2 to 27 days (median = 7.5 days).
Complications and revisions
At hospital discharge, 81% of the patients were oligosymptomatic. Complications recorded were neurological deficits (radial paresis, n = 3) and postoperative hematoma (n = 2), none of these requiring revision. All of the 3 radial nerve palsies were not pre-existing but occurred postoperatively. One of these recovered completely over time, with persistent radial paresis in 5.6% of patients. Two patients died during hospitalization due to their underlying neoplastic disease. During follow-ups, screw loosening (n = 1) and refractures (n = 2) were registered as failures (8.3%); these were accompanied with limited mobility and pain during movement; 2 of these were treated conservatively, and 1 underwent revision surgery (2.8%). These complications occurred in 7 of 36 patients (19.4%) and are considered not specifically related to the index surgery. In 1 patient, 2 complications were recorded: postoperative hematoma and screw loosening. The mechanical failure occurred between 4 and 9 months postoperatively, with all using 1 plate in the initial surgery. These plates did not extend over the entire length of the bone. In the 3 investigated failures, bone metastases were located in the humeral shaft or at the proximal end, whereas the tumor entity differed between lung carcinoma, breast carcinoma, and prostate carcinoma.
The median survival after this surgical intervention was 26.6 weeks (186 days). However, 4 patients died within the first 30 days after compound osteosynthesis. Comparing impending and complete fractures, no statistical difference of the time between cancer diagnosis and death ( P = .576) or the time from compound osteosynthesis and death ( P = .238) was found.
The CRA showed that there was no censoring for reasons other than death. During the first 2 years, the competing event of death was frequent, and substantially reduced the cohort at risk ( Fig. 3 ). Failures of osteosynthesis occurred early—usually within the first year. Figure 3 Competing risk analysis plot. Gray: at risk; orange: failure; green: deceased; red dots: revision events (refracture or screw loosening).
Discussion
In this study, compound osteosynthesis of the humerus was a valid therapeutic option for patients with impending or complete pathologic humerus fractures. The failure rate (8.3%) was low and occurred usually during the first year after surgery. In the vast majority of cases, the fracture could be stabilized until the end of life with only 1 surgical reintervention in the whole series.
End of life occurred at a median of 26.6 weeks (186 days) after this primary surgical intervention though. Compared to other authors, this study showed a shorter survival after this surgery. Moura et al documented a mean survival time after the surgery of 309.29 ± 33.71 days in 86 fixations. Thirty-point five percent of their patients suffered from breast carcinoma. 14 In the study by Pretell et al, it was recorded that the mean postoperative survival time of 22.7 months in 23 cases with a pathological fracture of the humerus. Forty percent of their cases consisted of patients with multiple myeloma. 17 Toepfer et al reported a mean survival of 11.5 months after surgery. The most common tumor entities responsible for a pathological fracture were renal cell and breast carcinoma, accounting for 33% of all pathological fractures in their cohort. 20 Piccioli et al disclosed a mean survival of 8.3 months after surgery for pathological humeral fractures in 87 cases. In 39% of those cases, breast carcinoma or renal cell carcinoma was the primary tumor. 16 As previously stated by Wedin et al, the variations in survival rates may be explained, apart from other factors, by the differences in primary tumors, stage of the disease, number of impending fractures, and selection criteria. 23
The literature on pathologic humerus fractures and compound osteosynthesis is sparse, and thus comparison of the results is difficult. One of the reasons for this is that the choice of treatment method is influenced by factors such as the tumor entity, its staging, and the prognosis associated with it. In general, rapid postoperative recovery of activities of daily living can improve personal autonomy, especially considering that the concept of the end of life has changed in the last decades. Today, with advances in systemic therapies metastatic disease often can be considered a chronic condition rather than a palliative one. This has led modern medicine to focus on the long-term well-being and autonomy of the patient. Correspondingly, the expectations of both patients and their families have become more important for medical decision making. 2 Furthermore, surgical stabilization of the weakened or fractured humerus significantly reduces pain and usually leads to a rapid reduction of analgesic treatment. This counteracts polypharmacy in patients who are already seriously ill. 10
Cancer patients in the advanced metastatic stage have a reduced life expectancy. 21 Applying CRA we could show that from surgery and up to 2 years, the risk of death is clearly dominant over the risk of osteosynthesis failure. Therefore, it can be expected that patients will still have intact compound osteosynthesis until death caused by cancer. Probably due to the low number of impending fractures in our cohort, we could not confirm the findings of Groot et al that patients with impending pathological fractures of long bones have better secondary outcomes than patients with complete fractures. 6
Our study has some limitations. First, due to the retrospective study design on nonstandardized data records, the information extracted has been reduced to chart reviews without any validated clinical scores. Second, the sample size was small and heterogeneous in terms of type of the primary tumor and different localization of the metastatic lesions in the humerus, which make it difficult to generalize and compare the results. More cases of pathological fractures caused by malignant tumors with a longer life expectancy could have significantly altered our results. The aim of any composite osteosynthesis should be to ensure permanent stability of the affected limb and to avoid premature implant failure. The longer the affected patient lives, the higher the risk of failure of the surgical treatment, as often no osseous healing of a pathological fracture can be expected. 5 , 19 The median overall survival after surgery in our cohort was low (26.6 weeks).
Therefore, double plate osteosynthesis in combination with bone cement augmentation of the bone defect and adjacent medullary canal should be discussed for all patients with a longer prognosis to avoid implant failure and revision surgery. From a biomechanical standpoint, 90-90 double plating is stronger than single plating and requires less soft tissue dissection compared to spanning plate fixation. 1 Oligoostotic or singular osseous metastatic manifestation in chronic malignant diseases with a good prognosis and systemic treatment options may even warrant wide resection and alloarthroplastic reconstruction, similar to primary malignant bone tumors. 16
Additional statements on adjuvant therapy cannot be made because we do not have the complete data for this.
Conclusion
The present study shows that compound osteosynthesis of the humerus using locking plates is a suitable option for patients with pathological humeral fractures. Depending on the patient’s life expectancy, it is usually successful until the end of life and has a low rate of mechanical failure (or the need of subsequent revision) and an acceptable rate of complication.
Disclaimers:
Funding: No funding was disclosed by the authors.
Conflicts of interest: The authors, their immediate families, and any research foundation with which they are affiliated have not received any financial payments or other benefits from any commercial entity related to the subject of this article.
References
1 Anderson J. Erickson J. Thompson R. Jr. Chao E. Pathologic femoral shaft fractures comparing fixation techniques using cement Clin Orthop Relat Res 131 1978 273 278 657635
2 Carr D. Luth E.A. Well-being at the end of life Annu Rev Sociol 45 2019 515 534 10.1146/annurev-soc-073018-022524 33311838 PMC7731514
3 Coleman R.E. Croucher P.I. Padhani A.R. Clézardin P. Chow E. Fallon M. Bone metastases Nat Rev Dis Primers 6 2020 83 10.1038/S41572-020-00216-3 33060614
4 Ekholm R. Adami J. Tidermark J. Hansson K. Törnkvist H. Ponzer S. Fractures of the shaft of the humerus an epidemiological study of 401 fractures J Bone Joint Surg Br 88 2006 1469 1473 10.1302/0301-620X.88B11 17075092
5 Freyschmidt J. Die pathologische Fraktur aus Sicht des Radiologen Trauma Berufskrankh 10 2008 165 166 10.1007/S10039-008-1363-3/METRICS
6 Groot O.Q. Lans A. Twining P.K. Bongers M.E.R. Kapoor N.D. Verlaan J.J. Clinical outcome differences in the treatment of impending versus completed pathological long-bone fractures J Bone Joint Surg Am 104 2022 307 315 10.2106/JBJS.21.00711 34851323
7 Hoellwarth J.S. Weiss K. Goodman M. Heyl A. Hankins M.L. McGough R. Evaluating the reoperation rate and hardware durability of three stabilizing implants for 105 malignant pathologic humerus fractures Injury 51 2020 947 954 10.1016/j.injury.2020.02.124 32143857
8 Hong S. Youk T. Lee S.J. Kim K.M. Vajdic C.M. Bone metastasis and skeletal-related events in patients with solid cancer: a Korean nationwide health insurance database study PLoS One 15 2020 e0234927 10.1371/JOURNAL.PONE.0234927 PMC7367479 32678818
9 Koob S. Plöger M.M. Bornemann R. Lehmann R.P. Alex D. Placzek R. Intramedullary nailing versus compound plate osteosynthesis in pathologic diaphyseal humerus fractures: a retrospective cohort study Am J Clin Oncol 45 2022 379 380 10.1097/COC.0000000000000937 35983966
10 Kose E. Wakabayashi H. Yasuno N. Polypharmacy and malnutrition management of elderly perioperative patients with cancer: a systematic review Nutrients 13 2021 1961 10.3390/nu13061961 34200493 PMC8227653
11 Lindecken K. Kühr J. Schander K. Verbundosteosynthesen bei pathologischen Frakturen infolge eines Mammakarzinoms Geburtshilfe Frauenheilkd 43 1983 191 192 6553002 10.1055/s-2008-1037086
12 Migliorini F. Maffulli N. Trivellas A. Eschweiler J. Tingart M. Driessen A. Bone metastases: a comprehensive review of the literature Mol Biol Rep 47 2020 6337 6345 10.1007/S11033-020-05684-0 32749632
13 Modaressi K. Bode-Lesniewska B. Studer G. Hofer S. Fuchs B. Zürich S. Knochenmetastasen-Abklärung und Therapie Schweiz Med Forum 13 2013 571 579
14 Moura D.L. Alves F. Fonseca R. Freitas J. Casanova J. Treatment of pathological humerus-shaft tumoral fractures with rigid static interlocking intramedullary nail-22 years of experience Rev Bras Ortop (Sao Paulo) 54 2019 149 155 10.1016/j.rbo.2017.10.012 31363260 PMC6529327
15 Niethard F.U. Pfeil J. Biberthaler P. Duale Reihe Orthopädie und Unfallchirurgie. 9th ed. Stuttgart: Georg Thieme Verlag 2022
16 Piccioli A. MacCauro G. Rossi B. Scaramuzzo L. Frenos F. Capanna R. Surgical treatment of pathologic fractures of humerus Injury 41 2010 1112 1116 10.1016/J.INJURY.2010.08.015 20828692
17 Pretell J. Rodriguez J. Blanco D. Zafra A. Resines C. Treatment of pathological humeral shaft fractures with intramedullary nailing. A retrospective study Int Orthop 34 2010 559 563 10.1007/s00264-009-0833-y 19572132 PMC2903153
18 Rath B. Tingart M. Migliorini F. Eschweiler J. Zureik R. Hardes J. Differentiated treatment strategies for bone metastases of the extremities Orthopä 48 2019 752 759 10.1007/S00132-019-03791-W/FIGURES/3 31444515
19 Sckell A. Mittlmeier T. Grundkonzepte der operativen Frakturbehandlung, geeignete Implantatwahl, Erscheinungsbild in der Bildgebung und weiterführende Bildgebung Kompend Bildgeb Frakturdiagnostik 2023 97 129 10.1007/978-3-662-63602-2_6
20 Toepfer A. Lenze U. Pohlig F. Eisenhart-Rothe Rv Gerdesmeyer L. Kirchhoff C. [Pathological fractures of the humerus: experience with 76 cases in a musculoskeletal oncology centre] Z Orthop Unfall 154 2016 364 369 10.1055/S-0042-105160 27273274
21 Tulotta C. Ottewell P. Endocrine-related cancer C tulotta and P ottewell IL-1B in bone metastasis R421-R434 the role of IL-1B in breast cancer bone metastasis Endocr Relat Cancer 25 2018 R421 R434 10.1530/ERC-17-0309 29760166 PMC5987176
22 Tytherleigh-Strong G. Walls N. McQueen M.M. The epidemiology of humeral shaft fractures J Bone Joint Surg Br 80 1998 249 253 9546454 10.1302/0301-620x.80b2.8113
23 Wedin R. Hansen B.H. Laitinen M. Trovik C. Zaikova O. Bergh P. Complications and survival after surgical treatment of 214 metastatic lesions of the humerus J Shoulder Elbow Surg 21 2012 1049 1055 10.1016/J.JSE.2011.06.019 21982491
This study has been approved by the Ethical Committee of Eastern Switzerland, Switzerland (study ID: BASEC No. 2021-02019).