Abstract
Background
The femur is the most common site of metastasis in the appendicular skeleton, and metastatic bone disease negatively influences quality of life. Orthopaedic surgeons are often faced with deciding whether to prophylactically stabilize an impending fracture, and it is unclear if prophylactic fixation increases the likelihood of survival.
Questions/purposes
Is prophylactic femur stabilization in patients with metastatic disease associated with different overall survival than fixation of a complete pathologic fracture?
Methods
We performed a retrospective, comparative study using the national Veterans Administration database. All patient records from September 30, 2010 to October 1, 2015 were queried. Only nonarthroplasty procedures were included. The final study sample included 950 patients (94% males); 362 (38%) received prophylactic stabilization of a femoral lesion, and 588 patients (62%) underwent fixation of a pathologic femur fracture. Mean followup duration was 2 years (range, 0-7 years). We created prophylactic stabilization and pathologic fracture fixation groups for comparison using Common Procedural Terminology and ICD-9 codes. The primary endpoint of the analysis was overall survival. Univariate survival was estimated using the Kaplan-Meier method; between-group differences were compared using the log-rank test. Covariate data were used to create a multivariate Cox proportional hazards model for survival to adjust for confounders in the two groups, including Gagne comorbidity score and cancer type.
Results
After adjusting for comorbidities and cancer type, we found that patients treated with prophylactic stabilization had a lower risk of death than did patients treated for pathologic femur fracture (hazard ratio = 0.75, 95% CI, 0.62–0.89; p = 0.002).
Conclusions
In the national Veterans Administration database, we found greater overall survival between patients undergoing prophylactic stabilization of metastatic femoral lesions and those with fixation of complete pathologic fractures. We could not determine the cause of this association, and it is possible, if not likely, that patients treated for fracture had more aggressive disease causing the fracture than did those undergoing prophylactic stabilization. Currently, most orthopaedic surgeons who treat pathological fractures stabilize the fracture prophylactically when reasonable to do so. We may be improving survival in addition to preventing a pathological fracture; further study is needed to determine whether the association is cause-and-effect and whether additional efforts to identify and treat at-risk lesions improves patient outcomes.
Level of Evidence
Level III, therapeutic study.
Introduction
The skeletal system is one of the most common sites of metastatic disease [18, 30]. With advances in life expectancy and an aging population, the incidence of new cancer diagnoses has risen steadily. It is estimated that more than 1.6 million new cancers were diagnosed in 2016 in the United States, and more than 250,000 patients in the United States are living with metastatic bone disease [20]. Additionally, while metastatic disease is generally not curable, patients may live with their disease for a prolonged and increasing period; for example, the median survival times for patients with metastatic disease from thyroid, prostate, and breast cancers are 48, 40, and 24 months, respectively [1]. The annual estimated cost of treating metastatic bone disease in the United States is USD 12 billion [5, 16, 20]. Providing quality care for patients with metastatic bone disease is an increasing challenge for the US health care community, including orthopaedic surgeons.
The femur is the most common site of metastatic disease and pathologic fracture in the appendicular skeleton [18, 31, 37]. Predicting which patients will have a pathologic fracture has been a topic of study for decades [4, 12, 22]. The rationale for attempting to predict a lesion’s propensity to cause fracture is that prophylactic stabilization causes less morbidity than operative fixation after a fracture occurs. The main goal of prophylactic treatment is palliation, by improving quality of life and reducing pain. In several small nonrandomized series with sample sizes ranging from 22 to 105 patients, prophylactic fixation was suggested to have better clinical outcomes in terms of hospital length of stay, ability to ambulate postoperatively, quality of life, and even mortality rates, than prophylactic stabilization [3, 9, 21, 36, 39].
Questions remain about the magnitude of the benefit of prophylactic stabilization, limited in part by the small numbers of patients in prior studies. Another question is whether prophylactic stabilization may influence overall survival in patients with metastatic disease. The rationale for potential benefit may be similar to the clinical scenario for proximal femur fractures in geriatric patients, in whom early treatment was associated with improved outcomes and survival [23, 24, 32]. In those with metastatic bone disease, prophylactic stabilization may allow for ongoing mobilization and avoidance of treatment delays compared with treatment after a pathologic fracture.
Using a large, nationwide, clinically integrated database, we therefore asked: Is prophylactic femur stabilization in patients with metastatic disease associated with different overall survival than fixation of a complete pathologic fracture?
Patients and Methods
Study Design and Setting
This retrospective, comparative, large-database study included patients cared for in Veterans Health Administration facilities nationwide, with surgery occurring between September 30, 2010 and October 1, 2015. Patients’ information in the Veterans Health Administration database was then reviewed until death or the last followup examination. This work was supported by resources at and the use of facilities at the Portland VA Medical Center.
Participants
The Veterans Health Administration databases contain patient care data for the largest integrated health care system in the United States. The Veterans Administration Informatics and Computing Infrastructure Corporate Data Warehouse is a frequently updated relational database. Clinical data are entered into the Veterans Administration Computerized Patient Record System in the form of diagnoses, vital signs, laboratory values, and notes. These data populate specified tables that are connected by patient identifiers and billing codes including ICD and Current Procedural Terminology (CPT). This structure allows for effective and efficient data retrieval using structured query language. The Veterans Health Administration databases include data from all patient encounters, including inpatient, day patient, outpatient, clinic appointments, rehabilitation appointments, and rehabilitation and skilled nursing facility stays. It also includes information on veterans treated outside of the Veterans Health Administration but whose treatment was paid for by the Veterans Administration. Unlike most nationwide databases [6, 7], patient data at the Veterans Administration are available from first treatment through last followup or death. It is primarily a clinical database, rather than a claims-based administrative database [25, 28]. Therefore, it is particularly well-suited for answering questions about longer-term outcomes, such as survival, in a large nationwide cohort [17, 27].
After obtaining institutional review board approval for the study, we accessed the Informatics and Computing Infrastructure Corporate Data Warehouse data. All patient records in the selected 5-year period were queried. These dates allowed for a minimum of 2 years of followup and coincided with a time period when the ICD-9 was used. Patients with surgical records including the CPT codes 27187 (prophylactic treatment, femoral neck and proximal femur) or 27495 (prophylactic treatment, femoral shaft) were categorized into the prophylactic stabilization group. We classified patients with the ICD-9 codes 733.14 (pathologic fracture, neck of femur) and 733.15 (pathologic fracture, other part of femur), and ICD-9 code 733.10 (pathologic fracture, unspecified site) and a corresponding CPT code of 27245 (open treatment, femur/hip with nail), 27244 (open treatment, femur/hip with plate), 27507 (open treatment, femoral shaft fracture with plate), or 27506 (open treatment, femoral shaft with nail) into the fixation for pathologic fracture group. We included only nonarthroplasty procedures.
Description of Experiment, Treatment, or Surgery
The type of surgical treatment was defined by the CPT codes entered into the database. Individual operative reports were not reviewed for this study, and individual indications for surgery were not evaluated.
Variables, Outcome Measures, Data Sources, and Bias
For patients in each group, we extracted variables and covariates including age, gender, cancer type, date of death or last followup, and comorbidities from the database. For comorbidity comparisons, we used the Gagne comorbidity score. The Gagne comorbidity score combines the conditions in the Charlson and Elixhauser measures [13]. Developed using 120,679 Pennsylvania Medicare enrollees and validated in New Jersey Medicare enrollees, the Gagne score offers improvements in comorbidity summarization and can better predict 30-day, 90-day, 180-day, and 1-year mortality than these independent scoring systems [13, 26].
Our query yielded 950 patients; 362 (38%) underwent prophylactic stabilization of a femoral lesion, and 588 (62%) underwent fixation of a pathologic femur fracture (Table 1). During a mean followup period of 2 years (range, 0-7 years), 662 patients (70%) died. Patients with prophylactic stabilization were younger, on average, than those with fixation of a pathologic fracture (69 years versus 67 years; p = 0.011) and patients with prophylactic stabilization had a lower preoperative Gagne Score than did patients with fixation of a pathologic fracture (6.2 versus 6.8; p = 0.016). Sex (average, 94% males and followup time were similar between the groups. The primary cancer type was identifiable in 948 patients (Table 2). There was an overall difference in the distribution of cancer diagnoses between the two groups. Using pairwise testing, we found that there were more colon and liver cancers in the fixation of pathologic fractures group than in the prophylactic stabilization group (24 versus three, p = 0.006; and 19 versus three, p = 0.020), but these two diagnostic groups represented only 6% of the total study population. Patients in the prophylactic stabilization group had improved unadjusted survival at all measured time points from 3 months to 5 years after surgery (by log-rank test, p = 0.018; Table 3). The unadjusted hazard ratio (HR) for the risk of death was lower in the prophylactic stabilization group than that in patients who underwent internal fixation for a pathologic fracture (HR = 0.82, 95% CI 0.70–0.97; p = 0.018).
Table 1.
Patient characteristics
Table 2.
Cancer diagnoses
Table 3.
Unadjusted survival according to cohort
Statistical Analysis, Study Size
The primary endpoint of the analysis was overall survival, and all analyses were preplanned. We estimated univariate survival using the Kaplan-Meier method; between-group differences were compared using the log-rank test. An a priori power analysis estimated a greater than 80% chance to detect a 10% improvement in survival at 2 years, assuming 700 total patients; a baseline mortality rate of 80% at 2 years; 5 years of accrual and a minimum of 2 years of followup; and an alpha of 0.05. We used covariate data to create a Cox proportional hazards model for death to adjust for confounders in the two groups, and the final multivariate model was created using a backward procedure with the Akaiki information criterion [38] for variable selection. Statistical analyses were completed in R version 3.4.1 (R Core Team, Vienna, Austria) using packages cpower, survival, and ggplot2. This study was designed and reported using the Strengthening the Reporting of Observation Studies in Epidemiology criteria [35].
Results
Patients in the prophylactic stabilization group had improved survival compared with those who underwent fixation after a pathological femur fracture for metastatic disease (Fig. 1). In the multivariate model, which included the Gagne comorbidity score and primary cancer diagnoses, the risk of death was lower for patients with prophylactic stabilization than for patients who were treated for a pathological fracture with internal fixation (HR = 0.75, 95% CI 0.62–0.89; p = 0.002) (Table 4).
Fig. 1.
The graph shows a Kaplan-Meier survival curve for prophylactic femur stabilization and fixation of pathologic fractures (log-rank test, p = 0.018).
Table 4.
Association of patient age and comorbidities with the risk of death after prophylactic femur stabilization versus fixation of pathologic fractures
Discussion
An aging population combined with improvements in cancer treatments has increased the number of patients living with metastatic disease in the United States [18, 31, 37]. Mirels’ criteria [22], published in 1989, are frequently used to determine the likelihood that a metastatic lesion will progress to a pathologic fracture and thereby direct treatment. Subsequent studies have both supported and questioned the validity and reliability of Mirels’ criteria [8, 10, 19, 34], but the desire to predict which lesions lead to fracture stems from a supposition that patients would benefit from prophylactic stabilization when compared with patients who subsequently have a fracture. After adjusting for confounding variables of cancer type and comorbidities, we found that patients treated with prophylactic femoral stabilization had greater survival than did patients who underwent operative fixation after pathological fracture. However, because of database limitations, we could not control for all potentially relevant confounding variables, and so it is likely that some of the survival difference observed was caused by unrelated factors, such as metastatic disease burden or tempo.
Limitations
This study had a number of limitations. As a retrospective analysis using a clinical database with continuously collected data, it is susceptible to selection bias and may be limited by the quality and accuracy of data entered into the database [6, 7, 25, 28]. We did not analyze specific oncologic diagnoses, nor the functional outcomes of these patients. Even in this large sample, subgrouping by detailed entities would have limited the study’s power. Instead, the analysis covered broad categories of malignancy based on the high-level categorization of the ICD-Oncology (ICD-O). The association of prophylactic stabilization across these diagnoses groups supports internal validity, but we recognize that important distinctions such as those between small and non-small cell lung cancer were not made. Although we reported on an association between prophylactic stabilization and improved overall survival, clearly there are patient- and tumor-specific factors that influence outcome. We did not evaluate the burden of metastatic disease, nor the rate of progression or stage at presentation. Likewise, the influence of chemotherapy or targeted treatments across this heterogeneous population was not addressed. It is possible that unaccounted-for factors would explain the association identified here.
Another limitation stems from the specific demographics of the US veteran population; our population was overwhelmingly men and few patients had breast cancer. Our findings are not necessarily generalizable to women. Furthermore, the specific indications for prophylactic stabilization were not ascertained. One possible explanation for these findings is that patients undergoing prophylactic stabilization are healthier or have a smaller overall burden of metastatic disease than those undergoing fixation; the Gagne score was lower in the prophylactic stabilization group than in the fracture fixation group, but after adjusting for Gagne score in the multivariate analysis, we found that the association with survival remained. Unidentified cofounders may moderate the observed association further.
Determining differences in survival between patients who undergo prophylactic stabilization versus those with fixation of pathologic fractures within a large, nationwide population is challenging. Databases that can capture information on a large, geographically diverse population are generally administrative, lack the depth of data required to compare differences in population comorbidities, and cannot follow patients long enough to determine survival differences [7, 14, 15]. Asking the same question in different databases can also result in disparate conclusions [6]. However, the Veterans Administration database has been shown to be a powerful tool, when used appropriately, for performing similar population-based analyses and developing practice-changing guidelines [11, 17, 27]. With a long followup period within an integrated care system, we were able to follow patients for well beyond the 30- or 90-day duration frequently seen in database studies [7].
We specifically excluded patients treated with arthroplasty from the analysis so we could compare similar treatments. Inclusion of arthroplasty would also require a distinction between intralesional procedures and tumor resection for oligometastatic versus multifocal metastatic disease; these are important areas for future study but were not part of our focused question. We also did not report on the total population of veterans treated with metastatic disease in the femur. Defining this specific phenotype is not possible by ICD-9 codes. Lacking this denominator is a potential for selection bias because it limits an understanding of the patterns of care; this is an issue shared by other series [2-4, 9, 14, 29].
In all database studies in which patients are selected via methods such as CPT or ICD-9 billing codes, there may be coding errors. Alternative methods for future work may include natural language processing and machine learning to decrease reliance on these systems [26, 33]. However, the weaknesses of data entry errors and lack of standardized indications must be balanced by the advantage of external generalizability from a national study reflecting care as provided and documented [27].
Timing: Prophylactic Stabilization Versus Fixation of Pathological Fractures
We found that patients treated with prophylactic stabilization had improved survival compared with patients treated with internal fixation after a pathological femoral fracture. Although we have not proven cause and effect, our finding fits with those from other studies. Our results are supported by relatively small series showing improved clinical outcomes and lower treatment costs when metastatic lesions are treated with prophylactic stabilization; however, this has never been shown within such a large and geographically diverse sample [2, 3, 5, 9, 21, 36, 37, 39]. In a retrospective review of 65 patients [3], those treated with prophylactic stabilization of a femoral metastasis had lower immediate mortality, improved ambulation, and shorter hospital stay than patients treated for a pathologic femur fracture. In a similar single-site review of a series of 40 patients [5], the overall cost of treatment was lower for prophylactic stabilization than for pathologic fracture fixation, and at 1 year, there was no difference in mortality between the two groups. A large-database study [29] examined differences in mortality between similar groups in a single Canadian province (Ontario) between 1992 and 1998. Among that study’s sample of 558 patients, 37% underwent prophylactic stabilization, which was associated with improved survival at all postoperative timepoints. Similar to our study, this association with improved survival persisted after adjustment for comorbidity, age, and cancer type [29]. The distribution of prophylactic stabilization to pathologic fracture fixation was also similar to our results, with most operations occurring after a pathologic fracture. But the current study included patients from across the United States, and although the patients were 94% male, this national scope across disparate regions and surgeons supports the external validity of these findings. In addition, although previous database studies examining trends in the care of metastatic lesions of the femur have been published [14, 29], none have the length of followup or number of patients available in the Informatics and Computing Infrastructure Corporate Data Warehouse system that ours did [14]. This is a large and geographically diverse study examining differences in mortality between patients treated with prophylactic stabilization of femoral metastatic lesions and patients undergoing fixation of pathologic femoral fractures.
Conclusions
In a large, integrated health care system in the United States, prophylactic surgical stabilization of metastatic lesions of the femur in a predominately male population was associated with improved survival compared with fixation of pathologic femur fractures. This result persisted after adjustment for patient comorbidities and cancer type. However, most operations occurred after a completed pathologic fracture. Although we could not control for all potential confounding variables that could have contributed to the observed increase in survival in patients treated prophylactically—and in particular, we could not control for the burden or tempo of metastatic disease nor specific indications for prophylactic stabilization—we note that our results support and extend those from other, smaller studies on the same topic [3, 9, 21, 29, 36, 39]. Taken together, our findings and those other studies tend to support prophylactic stabilization for these patients, who often have substantial medical comorbidities, and whose lifespans generally are limited. Future efforts could investigate whether initiatives to more actively identify and treat at-risk lesions are able to increase the ratio of prophylactic stabilization to completed fracture fixation while also improving patient outcomes.
Acknowledgments
We thank James Hayden MD PhD, for his thoughts and advice on study design.
Footnotes
Each author certifies that he or she has no commercial associations (e.g., consultancies, stock ownership, equity interest, patent/licensing arrangements) that might pose a conflict of interest in connection with the submitted article.
All ICMJE Conflict of Interest Forms for authors and Clinical Orthopaedics and Related Research® editors and board members are on file with the publication and can be viewed on request.
Each author certifies that his or her institution approved the human protocol for this investigation and that all investigations were conducted in conformity with ethical principles of research.
This work was performed at the Portland VA Medical Center, Portland, OR, USA.
Disclaimer: The contents do not represent the views of the U.S. Department of Veterans Affairs or the United States Government.
References
- 1.American Cancer Society. Cancer Facts and Figures 2018. Atlanta: American Cancer Society; 2018. Available at: https://www.cancer.org/content/dam/cancer-org/research/cancer-facts-and-statistics/annual-cancer-facts-and-figures/2018/cancer-facts-and-figures-2018.pdf Accessed December 1, 2018. [Google Scholar]
- 2.Ampil FL, Sadasivan KK. Prophylactic and therapeutic fixation of weight-bearing long bones with metastatic cancer. South Med J. 2001;94:394-396. [PubMed] [Google Scholar]
- 3.Arvinius C, Parra JL, Mateo LS, Maroto RG, Borrego AF, Stern LL. Benefits of early intramedullary nailing in femoral metastases. Int Orthop. 2014;38:129-132. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Beals RK, Lawton GD, Snell WE. Prophylactic internal fixation of the femur in metastatic breast cancer. Cancer. 1971;28:1350-1354. [DOI] [PubMed] [Google Scholar]
- 5.Blank AT, Lerman DM, Patel NM, Rapp TB. Is prophylactic intervention more cost-effective than the treatment of pathologic fractures in metastatic bone disease? Clin Orthop Relat Res. 2016;474:1563-1570. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Bohl DD, Russo GS, Basques BA, Golinvaux NS, Fu MC, Long WD, 3rd, Grauer JN. Variations in data collection methods between national databases affect study results: a comparison of the nationwide inpatient sample and national surgical quality improvement program databases for lumbar spine fusion procedures. J Bone Joint Surg Am. 2014;96:e193. [DOI] [PubMed] [Google Scholar]
- 7.Bohl DD, Singh K, Grauer JN. Nationwide databases in orthopaedic surgery research. J Am Acad Orthop Surg. 2016;24:673-682. [DOI] [PubMed] [Google Scholar]
- 8.Damron TA, Morgan H, Prakash D, Grant W, Aronowitz J, Heiner J. Critical evaluation of Mirels' rating system for impending pathologic fractures. Clin Orthop Relat Res. 2003:S201-207. [DOI] [PubMed] [Google Scholar]
- 9.Edwards SA, Pandit HG, Clarke HJ. The treatment of impending and existing pathological femoral fractures using the long gamma nail. Injury. 2001;32:299-306. [DOI] [PubMed] [Google Scholar]
- 10.Evans AR, Bottros J, Grant W, Chen BY, Damron TA. Mirels' rating for humerus lesions is both reproducible and valid. Clin Orthop Relat Res. 2008;466:1279-1284. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Fehringer EV, Mikuls TR, Michaud KD, Henderson WG, O'Dell JR. Shoulder arthroplasties have fewer complications than hip or knee arthroplasties in US veterans. Clin Orthop Relat Res. 2010;468:717-722. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Fidler M. Incidence of fracture through metastases in long bones. Acta Orthop Scand. 1981;52:623-627. [DOI] [PubMed] [Google Scholar]
- 13.Gagne JJ, Glynn RJ, Avorn J, Levin R, Schneeweiss S. A combined comorbidity score predicted mortality in elderly patients better than existing scores. J Clin Epidemiol. 2011;64:749-759. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Gendi K, Hennessy D, Heiner J. The burden of metastatic disease of the femur on the Medicare system. Springerplus. 2016;5:1916. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Grauer JN, Leopold SS. Editorial: large database studies--what they can do, what they cannot do, and which ones we will publish. Clin Orthop Relat Res. 2015;473:1537-1539. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Guy GP, Jr., Ekwueme DU, Yabroff KR, Dowling EC, Li C, Rodriguez JL, de Moor JS, Virgo KS. Economic burden of cancer survivorship among adults in the United States. J Clin Oncol. 2013;31:3749-3757. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Hadlandsmyth K, Vander Weg MW, McCoy KD, Mosher HJ, Vaughan-Sarrazin MS, Lund BC. Risk for prolonged opioid use following total knee arthroplasty in veterans. J Arthroplasty. 2018;33:119-123. [DOI] [PubMed] [Google Scholar]
- 18.Hage WD, Aboulafia AJ, Aboulafia DM. Incidence, location, and diagnostic evaluation of metastatic bone disease. Orthop Clin North Am. 2000;31:515-528. [DOI] [PubMed] [Google Scholar]
- 19.Jawad MU, Scully SP. In brief: classifications in brief: Mirels' classification: metastatic disease in long bones and impending pathologic fracture. Clin Orthop Relat Res. 2010;468:2825-2827. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Li S, Peng Y, Weinhandl ED, Blaes AH, Cetin K, Chia VM, Stryker S, Pinzone JJ, Acquavella JF, Arneson TJ. Estimated number of prevalent cases of metastatic bone disease in the US adult population. Clin Epidemiol. 2012;4:87-93. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Mickelson MR, Bonfiglio M. Pathological fractures in the proximal part of the femur treated by Zickel-nail fixation. J Bone Joint Surg Am. 1976;58:1067-1070. [PubMed] [Google Scholar]
- 22.Mirels H. Metastatic disease in long bones: A proposed scoring system for diagnosing impending pathologic fractures. 1989. Clin Orthop Relat Res. 2003:(415 Suppl):S4-13. [DOI] [PubMed] [Google Scholar]
- 23.Moja L, Piatti A, Pecoraro V, Ricci C, Virgili G, Salanti G, Germagnoli L, Liberati A, Banfi G. Timing matters in hip fracture surgery: patients operated within 48 hours have better outcomes. A meta-analysis and meta-regression of over 190,000 patients. PLoS One. 2012;7:e46175. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Orosz GM, Magaziner J, Hannan EL, Morrison RS, Koval K, Gilbert M, McLaughlin M, Halm EA, Wang JJ, Litke A, Silberzweig SB, Siu AL. Association of timing of surgery for hip fracture and patient outcomes. JAMA. 2004;291:1738-1743. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Patel AA, Singh K, Nunley RM, Minhas SV. Administrative databases in orthopaedic research: pearls and pitfalls of big data. J Am Acad Orthop Surg. 2016;24:172-179. [DOI] [PubMed] [Google Scholar]
- 26.Quan H, Sundararajan V, Halfon P, Fong A, Burnand B, Luthi JC, Saunders LD, Beck CA, Feasby TE, Ghali WA. Coding algorithms for defining comorbidities in ICD-9-CM and ICD-10 administrative data. Med Care. 2005;43:1130-1139. [DOI] [PubMed] [Google Scholar]
- 27.Radcliff TA, Henderson WG, Stoner TJ, Khuri SF, Dohm M, Hutt E. Patient risk factors, operative care, and outcomes among older community-dwelling male veterans with hip fracture. J Bone Joint Surg Am. 2008;90:34-42. [DOI] [PubMed] [Google Scholar]
- 28.Ratliff JK, Balise R, Veeravagu A, Cole TS, Cheng I, Olshen RA, Tian L. Predicting occurrence of spine surgery complications using "big data" modeling of an administrative claims database. J Bone Joint Surg Am. 2016;98:824-834. [DOI] [PubMed] [Google Scholar]
- 29.Ristevski B, Jenkinson RJ, Stephen DJ, Finkelstein J, Schemitsch EH, McKee MD, Kreder HJ. Mortality and complications following stabilization of femoral metastatic lesions: a population-based study of regional variation and outcome. Can J Surg. 2009;52:302-308. [PMC free article] [PubMed] [Google Scholar]
- 30.Sarahrudi K, Hora K, Heinz T, Millington S, Vecsei V. Treatment results of pathological fractures of the long bones: a retrospective analysis of 88 patients. Int Orthop. 2006;30:519-524. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Sim FH. Metastatic bone disease of the pelvis and femur. Instr Course Lect. 1992;41:317-327. [PubMed] [Google Scholar]
- 32.Uzoigwe CE, Burnand HG, Cheesman CL, Aghedo DO, Faizi M, Middleton RG. Early and ultra-early surgery in hip fracture patients improves survival. Injury. 2013;44:726-729. [DOI] [PubMed] [Google Scholar]
- 33.Uzuner O, Stubbs A. Practical applications for natural language processing in clinical research: The 2014 i2b2/UTHealth shared tasks. J Biomed Inform. 2015;(58 Suppl):S1-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Van der Linden YM, Dijkstra PD, Kroon HM, Lok JJ, Noordijk EM, Leer JW, Marijnen CA. Comparative analysis of risk factors for pathological fracture with femoral metastases. J Bone Joint Surg Br. 2004;86:566-573. [PubMed] [Google Scholar]
- 35.Vandenbroucke JP, von Elm E, Altman DG, Gotzsche PC, Mulrow CD, Pocock SJ, Poole C, Schlesselman JJ, Egger M, Initiative S. Strengthening the Reporting of Observational Studies in Epidemiology (STROBE): explanation and elaboration. Int J Surg. 2014;12:1500-1524. [DOI] [PubMed] [Google Scholar]
- 36.Ward WG, Holsenbeck S, Dorey FJ, Spang J, Howe D. Metastatic disease of the femur: surgical treatment. Clin Orthop Relat Res. 2003;415:S230-244. [DOI] [PubMed] [Google Scholar]
- 37.Weikert DR, Schwartz HS. Intramedullary nailing for impending pathological subtrochanteric fractures. J Bone Joint Surg Br. 1991;73:668-670. [DOI] [PubMed] [Google Scholar]
- 38.Zhang Z. Variable selection with stepwise and best subset approaches. Ann Transl Med. 2016;4:136. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Zickel RE, Mouradian WH. Intramedullary fixation of pathological fractures and lesions of the subtrochanteric region of the femur. J Bone Joint Surg Am. 1976;58:1061-1066. [PubMed] [Google Scholar]