Endocrine Feedback Loop – Episode 50

Topic: Bisphosphonates and Atypical Femoral Fractures
Date: June 20, 2024
Host: Dr. Chase Hendrickson (Vanderbilt University)
Featured Experts:

• Dr. Amal Shibli Rahal (University of Iowa)
• Dr. Ghada El Haj Fulahan (American University of Beirut)
  Fellows: David Berger (Columbia), Camilla Villavencencio (Mayo Clinic), Mazen Almagrabi (McGill), Lippy Marion (Mass General)

Overview & Purpose

This milestone 50th episode, recorded live at Endo 2024 in Boston, focuses on the risk of atypical femoral fractures (AFF) associated with bisphosphonate use in osteoporosis management. The discussion centers on a forthcoming JCEM Danish population case-cohort study and its implications for absolute risk, benefit-harm calculation, and patient counseling.

Key Discussion Points & Insights

1. Bisphosphonates – Benefit & Rare Side Effects

• Bisphosphonates remain the most widely used treatment for osteoporosis, with demonstrated fracture risk reduction:
  • Spine fractures: up to 70%
  • Hip fractures: up to 50%
  • Other non-vertebral sites: up to 25%
  • (02:59, Dr. Rahal)
• Increasing concern exists regarding long-term side effects, notably atypical femoral fractures (AFF)—rare, but highly discussed in patient care.
• The goal of the featured study: Quantify AFF risk and identify scenarios where risks might outweigh benefits.

2. AFF Diagnostic Criteria & Pathophysiology

• Diagnostic Criteria:

  • Four out of five major criteria per revised 2014 guidelines:
    • Location below lesser trochanter and above supracondylar flare
    • Minimal/no trauma
    • Transverse/oblique fracture starting in lateral cortex
    • Complete or incomplete through cortices
    • Minimal comminution (clean “sledgehammer” break)
  • Minor criteria (e.g., prodromal pain, delayed healing, etc.) are informative but not essential.
    (04:02, Dr. Fulahan)

• Mechanism:

  • Bisphosphonates suppress bone remodeling, impeding microcrack repair and potentially leading to AFF.
  • Not all AFFs are related to bisphosphonates; possible genetic, geometric, or other risk factors may play a role.

  “The skeleton is a very dynamic organ... What bisphosphonates do is suppress this remodeling... These [microcracks] may extend to the point where you get your AFF.”
  (05:00, Dr. Fulahan)

3. Study Design Strengths – Danish Case-Cohort Approach

• Case-cohort design: Hybrid between cohort (measuring outcome incidence) and case-control (identifying rare outcomes and exposures).
• Danish National Patient Registry (est. 1977): Robust, comprehensive linkage of inpatient, outpatient, prescription, and imaging data.
  • Image review for AFF was conducted by blinded radiologists, minimizing bias.
• The study's main exposure variable: Duration of bisphosphonate use (<1 year, 1-3 years, 3-5 years, >5 years).

4. Study Population & Methods

• Cases: 4,973 Danish adults (≥50 years) with subtrochanteric or femoral shaft fractures (2010-2014), narrowed to 189 confirmed AFFs by radiograph review.
• Comparison cohort: 37,021 age/sex-matched Danish adults; 699 traditional hip fractures identified.
• Covariates: Age, gender, prior fracture, co-morbidities, medication use (e.g., glucocorticoids, PPIs), but lacked DEXA, smoking status, SES, vitamin D, or femur geometry. (15:04-16:24, Panel Discussion)

5. Main Results & Findings

A. Bisphosphonate Use and AFF/Hip Fracture Incidence

• 68.8% of AFF cases had ever used bisphosphonates (vs. 8.8% of general cohort).

• Incidence of AFF by Duration (per 10,000 person-years):

  • <1 year: 0.9
  • 1–3 years: 1.4
  • 3–5 years: 2.0

  • 5 years: 4.9

• Incidence rapidly declines after discontinuation—up to 77% risk reduction after ≥3 years off drug.

• Hip fracture rate: Declined with longer bisphosphonate use (from 102.4 to 45 per 10,000 person-years with >5 years of therapy).

  “The rate of hip fracture remains much, much higher than that of an AFF... around 4.9 per 10,000 patient years with >5 years use.”
  (24:30, Dr. Rahal)

B. Risk Factors Identified

• AFF:

  • Increased with cumulative glucocorticoids, PPI use, hypertension (though biological plausibility is debated).
  • Dose-dependent increase with bisphosphonate duration—almost ninefold for >7 years of use.

• Hip fracture:

  • Increased with age, comorbidities, prior fracture; reduced with male gender and bisphosphonate use.

C. Number Needed to Treat/Harm (NNT/NNH):

• 3 years bisphosphonate:
  • NNT to prevent 1 hip fracture: 94
  • NNH to cause 1 AFF: 2,970
• Extrapolated 10 years: Curves converge—benefit/harm balance becomes less favorable the longer therapy is continued.
• No clinical scenario found in this dataset wherein AFF risk outweighed the anti-fracture benefit at 5 years.

6. Clinical Application & Counseling

Communicating Risk to Patients

• Patients commonly overestimate rare side effects due to media and social influences.

• Use absolute risk numbers, not just relative risk, in discussions.

  “If I were to follow 10,000 women over three years of bisphosphonate use, that risk would be in single digits—5 over 10,000. The absolute risk increase... is almost infinitesimal compared to the benefits.”
  (30:10, Dr. Fulahan)

• For every ~200 fractures (hip, spine, wrist) prevented over 5 years, only 1 AFF might be induced (from JBMR 2010 task force data).

Patient Adherence & Messaging

• Unified messaging among physicians, clear handouts, and consistent educational material are critical.

  “When you prevent mixed opinions or you engage physicians in a common approach... the patient is happier and... more compliant.”
  (44:24, Dr. Fulahan)

7. Other Notable Insights & Questions

• AFF without bisphosphonate exposure: ~30% of AFFs occurred in non-users; implied multifactorial etiology (mechanical, genetic, geometric factors) beyond the drug itself.

• Ethnicity and extrapolation: Population studied was largely North European; relative risks are likely generalizable, but absolute risks (NNT/NNH) must be recalculated for other (e.g., Asian) populations.

  “Relative risk may be applicable, but not the average absolute risk... The numbers for number needed to treat/harm only apply to the population where they were generated.”
  (37:49, Dr. Fulahan)

8. Strengths & Limitations of the Study

Strengths:

• Centralized, blinded radiographic adjudication
• Medication exposure from pharmacy (no recall bias)
• Longitudinal, population-based design
• Power calculation included for robustness

Limitations:

• No DEXA, smoking, SES, or detailed ethnic/geometry data
• Rare outcome: underpowered to distinguish risk by bisphosphonate type or sequential regimens
• Results most applicable to North European, vitamin D–replete populations

Notable Quotes & Memorable Moments

• Study design explanation & value:

  “The reason you want a cohort is to give you important information about how frequently something is occurring... but you also need the case-control aspect since we're talking about a rare outcome.”
  (07:22, Dr. Hendrickson)

• On balancing benefits and risks:

  “At five years, number needed to treat [is] 56, and number needed to harm 1,424—overwhelmingly favoring bisphosphonate treatment.”
  (28:46, Dr. Hendrickson quoting article)

• Patient anxiety perspective:

  “Patients are getting mixed messages from multiple sources... we have to make sure the clinic discussion is solid and well documented, with something for them to take home.”
  (29:44, Dr. Fulahan)

Important Timestamps

• Discussion of AFF criteria & mechanism: 04:02–06:38
• Danish registry & study methodology: 10:09–16:24
• Incidence findings (bisphosphonate duration & risk): 23:20–24:30
• Risk communication and counseling: 29:44–32:52
• Generalizability & population differences in risk: 37:49–39:44
• Panel assessments of study quality: 40:21–42:47
• Closing advice on patient messaging: 43:23–45:24

Conclusion & Practice Implications

The panel concludes that, for most patients—including those on long-term (>5 years) bisphosphonates—the risk of AFF is exceedingly low and is counterbalanced by much greater benefit in hip fracture risk reduction. Stopping drugs leads to a rapid risk reduction for AFF. However, patient counseling must be grounded in absolute numbers, with clear, consistent messaging, especially as patient fears about rare side effects are widespread and not always evidence-based.

Takeaway for Clinicians:
Continue bisphosphonates in appropriate patients with regular reassessment after 3–5 years; reassure patients regarding the rarity of AFF, especially relative to benefits; advocate for unified, clear education efforts; and be mindful in applying absolute risk data to more diverse populations.

End of Summary