In December 1996, a group of investigators reported the results of the Fracture Intervention Trial, a randomized controlled trial that compared the effect of alendronate plus calcium or calcium supplementation alone on the risk of fractures in women who already had evidence of vertebral fractures. The results showed that in patients managed with alendronate, there was a 51% decrease in the risk of hip fractures, a 46% decrease in the risk of vertebral fractures, and a 44% decrease in the risk of distal radial fractures when compared with patients managed with calcium alone. These results, as well as those from several other reports, supported the regulatory approval of alendronate (marketed under the trade name Fosamax) for the treatment of postmenopausal osteoporosis in the United States and many countries abroad. Alendronate thus became the first drug in a class of compounds known as the nitrogen-containing bisphosphonates to demonstrate these beneficial effects.

Approximately a decade later, and after millions of patients had undergone treatment, some disturbing reports suggested a potential suppression of bone turnover in association with long-term alendronate therapy. Bone biopsies from selected patients suggested diminished kinetic indices of bone formation. This was believed to lead to increased susceptibility to fracture and delayed healing of nonspinal fractures such as fractures of the femoral shaft. Additional reports suggested the occurrence of insufficiency or low-energy fractures in patients who used alendronate for several years. While epidemiological findings suggested that these fractures are very rare even among women who have been managed with bisphosphonates for as long as a decade, the American Society for Bone and Mineral Research convened a task force to understand the pathophysiology of these atypical fractures and to gain further information on the association of these fractures with bisphosphonates. The term “atypical femoral fracture” was adopted to distinguish this type of fracture as a unique entity in order to avoid a suggestion that it is exclusively associated with bisphosphonate use.

Atypical femoral fractures can occur anywhere along the shaft of the femur from just distal to the lesser trochanter to just proximal to the supracondylar flare of the distal femoral metaphysis. They may be transverse or short-oblique in configuration, are typically noncomminuted or minimally comminuted, are associated with minimal or no trauma, and may be associated with a medial spike. Incomplete fractures may involve only the lateral cortex. Because these fractures occur as a result of brittle failure while most osteoporotic patients show some ductility with deformation prior to failure, atypical femoral fractures most likely occur through bone that has undergone alterations in its mechanical and material properties. This further supports the notion that these fractures are unique and distinct from typical osteoporotic fractures of the femur.

While current evidence suggests a strong relationship between the use of bisphosphonates and the genesis of atypical femoral fractures, we now know that denosumab, a drug that inhibits osteoclastogenesis but is unrelated to the bisphosphonates, also may be associated with these fractures. Moreover, some patients who have never taken bisphosphonates or denosumab have presented with what appear to be atypical femoral fractures. Thus, atypical femoral fractures are not exclusive to patients who use osteoclast-inhibiting drugs, and this presents a more complicated picture regarding the etiology of this unique type of fracture.

In the March 2015 issue of JBJS Reviews, Blood et al. summarize current thinking regarding the evaluation and treatment of atypical femoral fractures. The authors note that these fractures can be treated successfully with intramedullary nailing and discontinuation of bisphosphonate therapy. However, there is a potential for a delay in healing. Prodromal thigh pain and radiographic evidence of a radiolucent line in patients with a history of atypical femoral fracture or chronic bisphosphonate use are strong indicators of impending fracture. In these patients, prophylactic fixation should be considered. In addition, patients with prodromal thigh pain who are receiving chronic bisphosphonate therapy but do not have radiographic evidence of incomplete fracture require further workup and may benefit from magnetic resonance imaging. For patients who have incomplete fractures and no pain, the authors recommend a trial of conservative therapy, including protected weight-bearing, discontinuation of bisphosphonate therapy, and supplementation with calcium and vitamin D (800 to 1000 IU) per day. While no recommendation currently exists regarding the duration of bisphosphonate therapy, most experts recommend discontinuation after five years. Moreover, as bisphosphonates are not the only class of compounds that may be associated with these fractures, further information is needed in order to make informed decisions regarding the use of specific drugs and the duration of their use. While treatment of atypical femoral fractures with an anabolic therapy such as parathyroid hormone has been proposed, there are no definitive data to support this suggestion at this time.

The use of bisphosphonates and denosumab to treat osteoporosis represents a major step forward. However, it is possible that there are specific subsets of patients who are more sensitive to pharmacological suppression of bone remodeling and who may not be candidates for this kind of therapy. Further investigation is required to understand the safety of prolonged use of osteoclast-inhibiting drugs.

Thomas A. Einhorn, MD, Editor

Click here for another OrthoBuzz post about this JBJS Reviews article.