Atypical Femoral Fractures: Causes, Risk Factors, and Clinical Management

What Exactly Is an Atypical Femoral Fracture?

An atypical femoral fracture is a low-energy or spontaneous fracture of the femoral diaphysis or subtrochanteric region that does not result from high-impact trauma. The fracture occurs in a highly specific pattern: transverse or oblique orientation at the lateral cortex, often with medial cortical spike formation, and a characteristic periosteal stress reaction on the lateral cortex that radiologists call the "dreaded black line." ASBMR diagnostic criteria require at least four of five major features to confirm the diagnosis.

The ASBMR 2014 Task Force defined the five major diagnostic criteria as: location in the femoral diaphysis or subtrochanteric region, association with minimal or no trauma, a transverse or short oblique fracture line, non-comminuted pattern, and a complete fracture that extends through both cortices or an incomplete fracture involving the lateral cortex only. Minor features include generalized periosteal thickening, increased cortical thickness, a prodromal aching or dull thigh pain, bilateral presentation, and delayed healing.

Typical hip fractures, by contrast, occur at the femoral neck or intertrochanteric region and arise from falls in osteoporotic bone. AFFs occupy a mechanistically different category: they resemble fatigue fractures in cortical bone that has been rendered excessively stiff by years of suppressed bone remodeling.

How Bisphosphonates Cause Atypical Femoral Fractures

Long-term bisphosphonate use is the dominant modifiable risk factor for AFFs, though the absolute risk remains low compared to the fracture prevention benefit in most patients. Bisphosphonates inhibit osteoclast-mediated resorption by binding hydroxyapatite and triggering osteoclast apoptosis. Over years of use, this suppression of bone turnover reduces the normal micro-damage repair cycle. Fatigue micro-cracks accumulate in cortical bone faster than they can be cleared, and the lateral femoral cortex, which bears tensile stress during normal gait, becomes the site of crack coalescence.

A 2011 population-based cohort study published in the New England Journal of Medicine (N=205 AFF cases) found that bisphosphonate use carried an adjusted odds ratio of 47.3 for subtrochanteric or diaphyseal fractures with the characteristic AFF pattern. The absolute risk was still low: 2.3 per 10,000 patient-years at 2 years of use, rising to 78 per 10,000 patient-years at 8 or more years of use. The data make duration the critical variable.

Alendronate (Fosamax), the most widely prescribed oral bisphosphonate, appears most frequently in AFF case series simply because it is the most commonly prescribed agent. Zoledronic acid (Reclast), given as an annual 5 mg intravenous infusion, carries a similar mechanism and a similar AFF signal in long-term registries. The FDA added an AFF warning to all bisphosphonate labels in 2010 after the pharmacovigilance data became sufficient to require labeling action.

Asian women, particularly those of East Asian descent, appear to have a two-to-four-fold higher AFF risk than white women at equivalent durations of therapy. Concomitant glucocorticoid use and low body weight also amplify risk. Hip geometry, specifically a more bowed femoral shaft, has been proposed as a biomechanical contributor.

The Prodrome: Recognizing Thigh Pain Before Complete Fracture

Roughly 70% of patients who sustain an AFF report a prodromal period of dull, aching, or sharp thigh or groin pain lasting weeks to months before the fracture completes. This window is clinically important because an incomplete AFF on imaging is a surgical emergency of a different urgency level than a completed fracture.

Plain radiographs sometimes miss early periosteal changes. When a patient on long-term antiresorptive therapy presents with new thigh pain and a normal X-ray, the next step is MRI of the full femur or a technetium-99m bone scan. MRI identifies the stress reaction and any cortical crack with greater sensitivity. A 2012 case series in JAMA Internal Medicine noted that prophylactic intramedullary nail fixation of incomplete AFFs prevented progression to complete fracture in 13 of 14 consecutive patients, compared to continued conservative management where 10 of 11 patients eventually fractured completely.

Any patient on bisphosphonate therapy for five or more years who reports new thigh or groin pain deserves bilateral femur imaging. Not just the symptomatic side. Bilateral involvement eventually develops in approximately 28 to 47% of patients with a confirmed unilateral AFF, and the contralateral fracture often follows within two years.

Osteonecrosis of the Jaw: A Separate but Related Antiresorptive Complication

Osteonecrosis of the jaw (ONJ) is a different complication from AFFs but shares the same underlying biology: impaired bone turnover from antiresorptive therapy leaves dead bone unable to remodel and repair after local trauma. The jaw is uniquely vulnerable because its rich vascular supply is paradoxically coupled with high local bone turnover demands, particularly after dental extraction or implant placement.

ONJ is defined clinically as exposed, necrotic bone in the maxillofacial region that persists for more than 8 weeks in a patient receiving or previously treated with antiresorptive or antiangiogenic agents, in the absence of head-and-neck radiation. The AAOMS 2022 position paper on medication-related osteonecrosis of the jaw stratified risk by route, dose, and indication. Risk with oral bisphosphonates for osteoporosis is approximately 0.01 to 0.1%, rising significantly with IV bisphosphonates used in oncology (0.7 to 6.7%) and reaching up to 1 to 15% in some high-dose cancer series.

Denosumab (Prolia), a RANK-L inhibitor administered as a 60 mg subcutaneous injection every 6 months for osteoporosis, carries an ONJ risk broadly comparable to oral bisphosphonates at osteoporosis doses. Risk factors include dental extraction, poor oral hygiene, dentures causing mucosal trauma, corticosteroid co-therapy, diabetes, and smoking.

Preventive dental assessment before initiating antiresorptive therapy is the single most effective risk-reduction strategy. Elective invasive dental procedures should ideally be completed and healed before starting therapy. In patients already on therapy, the decision to pause antiresorptives before a procedure requires individualized assessment, because the fracture prevention benefit of continued therapy may outweigh the modest ONJ risk reduction from a drug holiday. The American Dental Association's evidence-based clinical practice guideline published in JADA in 2022 concluded that the benefit of discontinuing oral bisphosphonates before dental procedures is not established for patients at low systemic risk.

Denosumab Rebound: The Fracture Surge After Discontinuation

Stopping denosumab without a bridging strategy produces one of the most clinically dangerous rebound phenomena in musculoskeletal pharmacology. Unlike bisphosphonates, which remain embedded in bone mineral for years after the last dose, denosumab's effect on bone resorption is fully reversible. Once the drug clears (roughly 6 months after the last injection), osteoclast activity surges well above baseline, bone turnover markers spike, and bone mineral density can fall by 5 to 15% within 12 months.

The clinical consequence is not merely a return to pre-treatment fracture risk. Multiple spontaneous vertebral fractures have been documented in patients who discontinued denosumab, even when they had no pre-existing vertebral fractures. A 2017 analysis in Osteoporosis International (N=1,001) found that 15% of patients who discontinued denosumab developed at least one new vertebral fracture within 24 months of stopping, with a mean of 2.7 new vertebral fractures per affected patient. That fracture clustering pattern is consistent with a rebound rather than simple disease recurrence.

The following bridging protocol reflects current ASBMR and Endocrine Society guidance, organized into a practical clinical decision framework for denosumab discontinuation:

Step 1. Assess duration of prior denosumab use. Patients who received fewer than 2 injections (less than 12 months) may have limited skeletal integration; transition to a bisphosphonate immediately after the last injection.

Step 2. For patients on denosumab for 2 or more years, administer a single dose of zoledronic acid 5 mg IV approximately 6 months after the last denosumab injection. This timing corresponds to the point when bone turnover markers (specifically serum CTX) begin rising, signaling the onset of rebound resorption.

Step 3. Monitor bone turnover markers at 6 and 12 months post-zoledronate. If CTX remains suppressed at 12 months, one zoledronate dose may suffice. If CTX has risen above the upper limit of normal, a second dose at 12 months is advisable.

Step 4. Do not leave a gap. Any interval exceeding 7 months between the last denosumab injection and the first bisphosphonate dose leaves the skeleton unprotected during the rebound window.

The Endocrine Society 2019 clinical practice guideline on osteoporosis in postmenopausal women states explicitly: "If denosumab is discontinued, antiresorptive therapy should be initiated to prevent the rapid loss of bone density and increased fracture risk."

Diagnosing and Staging Atypical Femoral Fractures

Once an AFF is suspected, the imaging and biochemical workup determines surgical urgency. A systematic approach prevents delayed diagnosis.

Plain radiography of the full femur (anteroposterior and lateral) is the first-line study. Look specifically for lateral cortical thickening, periosteal reaction at the lateral cortex, and any visible lucency (the "dreaded black line"). Complete fractures are usually visible. Incomplete fractures affecting only the lateral cortex may appear as a faint line or focal cortical thickening.

When X-ray is negative and clinical suspicion persists, MRI of the full femur is the preferred next study. MRI identifies bone marrow edema and cortical signal changes associated with stress reactions before a frank fracture line is visible. Sensitivity for incomplete AFF exceeds 90% in case series. CT of the femur can delineate fracture geometry for surgical planning but has lower sensitivity for early stress reactions.

Bilateral femur imaging is standard at initial diagnosis. The 2014 ASBMR Task Force Report in JBMR explicitly recommends evaluating the contralateral femur in every confirmed AFF patient because the bilateral risk compounds over time.

Bone turnover markers (serum P1NP for formation, serum CTX for resorption) and 25-hydroxyvitamin D levels should be measured. Prolonged bisphosphonate therapy may produce measurably suppressed CTX. If CTX is below 100 pg/mL, bone turnover is markedly suppressed, which supports the AFF diagnosis and informs the drug holiday decision.

Management: Drug Holidays, Surgery, and Anabolic Therapy

Bisphosphonate drug holiday. For patients who have completed 5 years of oral or 3 years of IV bisphosphonate therapy without AFF, a drug holiday of 1 to 2 years is generally recommended for those at moderate fracture risk. Patients with high hip fracture risk (T-score below -2.5 at the hip, prior hip fracture) should continue therapy or transition to an alternative. The drug holiday decision requires weighing the small cumulative AFF risk against the continued benefit of fracture prevention.

In a patient who has sustained a confirmed AFF, the bisphosphonate should be stopped immediately. The fracture itself is evidence of the accumulated cortical fatigue that antiresorptive therapy was perpetuating.

Surgical fixation. Complete AFFs require intramedullary nail fixation rather than plate fixation, because intramedullary nailing shares load along the entire diaphysis and addresses the mechanical environment that produced the fracture. Complete fractures that are not fixed are unlikely to heal with conservative management given the underlying biology.

Incomplete AFFs with cortical involvement of 50% or more are generally referred for prophylactic intramedullary nailing. Incomplete fractures with minimal cortical involvement may be managed conservatively with protected weight-bearing, cessation of the offending drug, and close imaging follow-up, but the threshold for surgical intervention should remain low.

Teriparatide. Teriparatide (Forteo), recombinant human PTH(1-34) at 20 mcg subcutaneously daily, is the preferred anabolic agent after AFF because it actively stimulates bone remodeling and new cortical bone formation. A prospective case series in JBMR (N=32) found that teriparatide therapy after AFF was associated with a mean time to radiographic healing of 4.9 months compared to 9.3 months without teriparatide. The FDA-approved duration is up to 24 months. Abaloparatide (Tymlos) at 80 mcg daily is an alternative with comparable mechanism and similar fracture-healing data in smaller series.

Calcium and vitamin D. All patients with AFF should receive calcium (total intake 1,000, 1 to 200 mg/day from diet and supplement combined) and vitamin D (sufficient to maintain 25-OH-D above 30 ng/mL). These are adjunctive, not sufficient as sole therapy.

Long-Term Monitoring and Return to Antiresorptive Therapy

After a drug holiday and fracture healing, the decision to restart antiresorptive therapy requires re-evaluation of the underlying osteoporosis risk. FRAX calculation, DXA with hip and spine BMD, and the clinical risk factor profile all contribute.

If the 10-year probability of major osteoporotic fracture on FRAX exceeds 20%, or the hip fracture probability exceeds 3%, the benefit of continued antiresorptive therapy likely outweighs the AFF risk for most patients. In this scenario, restarting with a non-bisphosphonate antiresorptive (denosumab or raloxifene) or continuing with a bisphosphonate at a lower dose or longer interval may be considered.

Romosozumab (Evenity), the anti-sclerostin antibody given as 210 mg (two subcutaneous injections) monthly for 12 months, offers an anabolic-antiresorptive dual mechanism and does not share the cortical suppression mechanism of bisphosphonates. The ARCH trial (N=4,093) in NEJM found that romosozumab followed by alendronate reduced clinical fracture risk by 27% compared to alendronate alone over 24 months. For patients who need antiresorptive therapy after AFF recovery, romosozumab may be a reasonable option if cardiovascular risk is acceptable, given its black-box warning for myocardial infarction and stroke.

Imaging follow-up after AFF should include repeat bilateral femur X-rays at 3, 6, and 12 months post-fixation, and annual DXA for at least 3 years after therapy changes.