Abstract
Introduction
Bisphosphonate medications reduce the risk of osteoporotic fragility fractures; however, prolonged use increases the risk of atypical femur fractures (AFF). As the population ages and the burden of osteoporosis rises, the relative incidence of AFFs is expected to grow. Current expert guidelines recommend discontinuing bisphosphonates and initiating alternative anti-fracture therapies (AFT), such as denosumab, teriparatide, or raloxifene, following an AFF to support skeletal health and reduce future fracture risk. However, it is unclear which patients receive these medications after an AFF. This study aims to identify demographics associated with the initiation of alternate osteoporosis therapies following an AFF.
Methods
We conducted a retrospective cohort study using data from the TriNetX Registry to identify patients who were on bisphosphonate therapy that had an AFF. The primary outcome was initiation of teriparatide, denosumab, or raloxifene within one year after AFF. We compared patient demographics and baseline characteristics between those initiated on AFT after an AFF and those who were not. Pairwise comparisons of proportions were performed between the groups using two-proportion Z-tests with Bonferroni correction to account for multiple comparisons.
Results
We identified 784 patients with AFFs following bisphosphonate use. 71 patients (9.5%) were initiated on AFT following the index fracture. Teriparatide (83.1%) was the most commonly prescribed AFT, followed by denosumab (21.9%), then raloxifene (≤14.1%). Patients initiated on AFT were more likely to have a previous fragility fracture (66.2% vs 45.2%, P < 0.001). Other patient demographics assessed did not show any significant differences.
Discussion and Conclusion
Despite expert recommendations, AFT was initiated in only a small percentage of patients following AFFs. These findings highlight a significant care gap in secondary fracture prevention and underscore the need for a multidisciplinary approach, including coordination between endocrinology, orthopaedics, and primary care, to improve adherence to post-fracture osteoporosis management guidelines.
Introduction
The aging population in the United States is rapidly expanding with the proportion of adults over 65 years old tripling over the last century. 1 As the patient population continues to age, the rates of osteoporosis and osteoporosis related fragility fractures are increasing.2,3 There are currently over ten million Americans with osteoporosis affecting an estimated 11.5% percent of Americans age fifty or over.4,5 In addition, more than forty three million people in the United States have decreased bone density, putting them at risk of future development of osteoporosis. 6 Osteoporosis increases the risk of fracture and it is projected that by 2025, there will be more than three million osteoporosis related fractures each year in the United States.7-9 These fractures lead to decreased quality of life, disability, and increased mortality.10-13 Treatment of osteoporotic fractures also impose a high cost to the healthcare system with an estimated cost of 25.3 billion dollars annually in 2025. 14 Pharmacologic prevention of osteoporosis decreases rates of future fracture and is cost effective.15,16 Given the high burden of osteoporotic fragility fractures on both patients and the healthcare system, prevention is imperative.
Bisphosphonates are the most common class of pharmacologic therapy used for the treatment of osteoporosis and have been shown to decrease rates of osteoporotic fragility fracture.17,18 However, long-term use of bisphosphonates has been linked atypical femur fractures (AFFs).19,20 These fractures, while uncommon, are associated with high rates of mortality and disability similar to other femur fractures.21-23 Expert guidelines recommend discontinuing bisphosphonate therapy after an AFF, due to the possibility of delayed union or nonunion, increased surgical complications, or contributing to development of an AFF on the contralateral side.24-26 In their place, alternative anti-fracture therapies (AFT), such as teriparatide, denosumab, or raloxifene, are recommended to reduce future fracture risk in high risk patients.24,26
Despite these recommendations, it is unclear how often patients are transitioned to alternative therapies following an AFF and which clinical factors influence these prescribing decisions. The purpose of this study was to evaluate the use of AFT after bisphosphonate-associated AFFs in a large, multi-institutional cohort, and to identify demographic and clinical factors associated with their initiation.
Materials and Methods
The TriNetX Registry, a multicenter, multinational registry of deidentified medical records was queried and deidentified patient data from 108 health care organizations was obtained from the national database on September 15th, 2025. 27 This cohort was then evaluated using descriptive statistics and study specific cohorts were delineated. Institutional Review Board Approval was not required given the strict use of deidentified patient records with no individually identifiable data.
Cohort Selection
Patients who had an atypical femur fracture from the years 2005 to 2025 were identified using International Classification of Diseases, 10th Revision (ICD-10) code M84.75 (Atypical Femur Fracture). Atypical femur fractures can occur in patients without prior bisphosphonate exposure.28,29 Therefore, patients without documented bisphosphonate use were excluded to ensure accurate cohort classification. Patients were then excluded if they had a history of hypercalcemia or neoplasm. Alternative anti-fracture therapy (AFT) was defined as teriparatide, denosumab, or raloxifene. To be included in the AFT group, patients must have been prescribed any of the above drugs within one year following AFF. We chose one year as the time frame for AFT initiation because previous studies have evaluated osteoporosis medication initiation within one to two years following a fragility fracture.30-32 Given the mortality risk associated with femur fractures, we chose the shorter one year interval to more accurately capture treatment decisions after an AFF. We excluded patients who died within one year after their AFF. Patients without at least one year of follow-up after their AFF were also excluded. This process is summarized in Figure 1. Flowchart demonstrating cohort selection
Statistical Analysis
Alternative Anti-fracture Therapy Initiation Comparison
Pairwise comparisons of proportions for demographics and previous fragility fracture were performed between patients who had AFT prescribed after AFF and patients who were not prescribed AFT after AFF. Specific demographics assessed included age at time of AFF, gender, race, and ethnicity. We also compared previous medical conditions including prior osteoporotic fragility fracture and BMI. We used two-proportion Z-tests with Bonferroni correction to account for multiple comparisons. A P-value of less than 0.05 was set for statistical significance. A power analysis was not performed because this was a retrospective observational study that was intended to describe prescribing patterns following AFFs rather than test a predefined hypothesis. Figures were created using Microsoft Excel (Microsoft Corporation, Redmond, WA) and Microsoft Visio (Microsoft Corporation, Redmond, WA). All statistical analyses were performed using the built-in analysis tools available within the TriNetX platform (TriNetX, LLC, Cambridge, MA).
Results
Baseline Demographics of Patients Who are Prescribed AFT vs Not After AFF
There were a total of 1696 patients with AFFs identified. Of these patients, 930 (54.8%) were excluded from analysis due to no documented history of prior bisphosphonate usage. This left a total of 766 (45.2%) patients with bisphosphonate associated AFF and 748 patients were included in analysis after excluding patients who died within one year after their AFF. The recorded one year mortality rate was 2.3% overall (18/766), occurring in 4.2% of patients prescribed AFT (3/74) and 2.2% of those not prescribed AFT (15/692). There was no significant difference between mortality rates in patients who were or were not prescribed AFT after AFF (4.2% vs 2.2%, P = 0.309).
Difference in Baseline Characteristics Between Atypical Femur Fractures With and Without Alternative Antifracture Therapy
Types of AFT Prescribed
Teriparatide was the most commonly prescribed AFT after an AFF with 59 out of 71 (83.1%) patients having received a prescription. Denosumab was the second most common, prescribed to 17 out of 71 (23.9%) patients. Raloxifene was prescribed least often, prescribed to 10 or less patients out of 71 patients (14.1%).
Discussion
In our retrospective multicenter study, we found the rate of patients started on AFT after an AFF was low at 9.5%, despite current guideline recommendations. Our study showed a strong predominance of female patients, which aligns with the known epidemiology of osteoporosis, where older women make up the majority of those diagnosed and treated for osteoporosis. 5 This finding is consistent with prior literature showing that treatment initiation rates after fragility fractures are generally low. For example, Ross et al 30 reported that rates of osteoporosis management and secondary preventative treatment after primary fragility fractures remain suboptimal, with many patients not receiving pharmacologic therapy despite clear indications. While some analyses suggest that factors such as age, sex, and other demographics may influence fracture risk, results across studies vary. 6 In our cohort, we did not observe any clear differences in AFT initiation based on demographic factors. 6 This study highlights a disconnect between current guidelines and actual practice patterns.
In our cohort, 54.8% of patients with an AFF did not have a documented history of bisphosphonate use. This aligns with previously reported rates of AFF in patients not taking bisphosphonates of 22 to 60%.28,29 Patients who were initiated on AFT after an AFF were significantly more likely to have a previous diagnosis of a previous osteoporotic fragility fracture. A prior osteoporotic fragility fracture is one of the strongest predictors of subsequent future fractures, with the highest risk occurring in the first 24 months after the initial event. 33 Similarly, fragility fractures tend to cluster in time, highlighting a critical window to intervene after a first fracture to help prevent subsequent events. 34 In patients with previous osteoporotic fragility fractures, quick initiation of AFT after AFF may be especially important. Failing to treat patients during this period may increase future fracture risk. One limitation is we were unable to differentiate between the specific number or types of previous osteoporotic fragility fractures in our patient cohorts. Determining whether the number of previous fragility fractures or type of fragility fractures type are more strongly associated with AFT initiation requires further research.
In our study, the observed one-year mortality rate following an AFF was 2.3%, which is notably lower than the approximate 10% one-year mortality rate reported in a meta-analysis by Charoenngam et al 35 (95% CI, 5 - 16%). Like many other large databases, TriNetX is unable to record deaths which occur outside of the included healthcare systems, potentially showing an artificially lowered death rate.27,36 Patients who die after discharge or who seek care in a different healthcare system may not be recorded, leading to an underestimation of true mortality.
Most patients who were prescribed AFT after AFF were started on teriparatide. While this study is unable to comment on the effectiveness of this medication in preventing repeat fracture, there is limited evidence that teriparatide, a recombinant parathyroid hormone 1-34, may improve fracture healing and be beneficial after AFF. 24 Teriparatide’s anabolic mechanism, through intermittent PTH stimulation, increases osteoblast activity and promotes new bone formation. 24 This provides a biologically plausible rationale for supporting fracture healing in patients with AFF. However, randomized controlled trials specifically evaluating AFF are scarce, and current recommendations to prescribe teriparatide to improve bone healing after AFF remain cautious, usually individualized based on fracture pattern, comorbidities, and prior antiresorptive therapy. 24
Alternative osteoporosis treatments vary in their effectiveness and safety profiles. Bisphosphonates and denosumab both reduce common osteoporotic fractures in trials and clinical practice, but the evidence around AFF is more nuanced.3,18 Prolonged bisphosphonate use has the strongest association with AFF, though the overall risk remains low compared with their benefits for high-risk patients.18,19 Denosumab has also been linked to AFF in some reports, but the population-level evidence for a causal association is less consistent than for long-term bisphosphonates and remains debated. 20 Current guidelines emphasize individualized treatment decisions that take into account prior fractures, current fracture risk, previous therapies, and patient-specific goals to optimize outcomes.
Our study has several limitations. This includes the retrospective, observational nature of our study, which only allows inferences of association, and not causation. No power analysis was performed, limiting our ability to assess whether the study was adequately powered to detect meaningful differences. We were unable to assess for subsequent fracture risk following the index AFF, limiting our ability to evaluate the clinical impact of AFT initiation. It was also not possible for this study to account for individual medical decision making and possible contraindications for some patients to receive AFT. We did not include other osteoporosis therapies such as abaloparatide, romosozumab, calcium, or vitamin D supplementation. This may limit the completeness of our assessment of pharmacologic treatment patterns following AFF. Additionally, we were unable to evaluate exact timelines of when AFT was started after AFF, which may have important implications for treatment effectiveness. We were also unable to assess socioeconomic status or insurance type, which may influence both prescribing patterns and access to care.30,37 The true mortality rate was likely underreported, given the limitations of the ability of the database to record out of hospital deaths. If there are ten or fewer patients with a given set of criteria, TriNetX will round the number to ten to protect patient confidentiality. 27 This occurred in our analysis of patients who were prescribed raloxifene, likely overestimating the true rate of raloxifene prescription after AFF. While we attempted to include patients initiated on AFT if they had an active prescription within 1 year after their index AFF, we were unable to determine whether patients were compliant with their medication, the dose of medication prescribed, or medication adherence.
Conclusion
Despite guideline recommendations, initiation of AFT following bisphosphonate associated AFF remains low. Teriparatide is the most commonly prescribed alternative therapy, but overall use is limited, highlighting a significant care gap in secondary fracture prevention. Our study provides important insights into real-world patterns of AFT use following bisphosphonate associated AFFs. These findings emphasize the importance of a multidisciplinary approach, including coordination with endocrinology, orthopaedics, and primary care, to optimize post fracture management and ensure appropriate initiation of AFT. Future studies to assess the outcomes of AFT on secondary fracture risks are needed.
Footnotes
Ethical Considerations
This review was exempt from the University of Texas Health San Antonio Institutional Review Board approval given its retrospective nature and lack of patient-identifiable information.
Author Contributions
Andrew Ni, MD contributed to conceptualization, data curation, formal analysis, investigation, methodology, project administration, resources, software, visualization, writing – original draft, and writing - review & editing, Azophi Moffat contributed to formal analysis, writing – original draft, and writing – review and editing, Jordan Robbins MD contributed to methodology, data analysis, writing – original draft, writing – review & editing, Dean Slocum MD contributed to conceptualization, writing – review and editing, Abhi Rashiwala MD contributed to conceptualization, writing– review & editing, Katerina Papanikolaou MD contributed to methodology, writing– review & editing, Ravi, Karia MD contributed to methodology, supervision, project administration, and writing– review & editing.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.