Background: Recent case reports have identified an association between long-term bisphosphonate treatment and stress fractures of the femoral shaft. The risk of such fractures in bisphosphonate users has not been determined.
Methods: We identified women over 55 years of age with the specific fracture pattern by searching the operation registry of the hospitals in 2 healthcare districts. Prevalence of bisphosphonate treatment was provided by a Swedish national registry covering all drugs delivered to all individuals since 2005.
Results: The number of women on bisphosphonate treatment was 3,087. Of these, 5 had femoral stress fractures. They had been taking bisphosphonates for 3.5 to 8.5 years. The incidence density for a patient on bisphosphonate was 1/1,000 per year (95% CI: 0.3-2). In the remaining 88,869 women who were not taking bisphosphonates, there were 3 stress fractures. Thus, their risk (without correction for inhomogeneity in age distribution) was 46 times less (95% CI: 11-200).
Interpretation: These results are rough estimations based on a comparatively small material. Still, a treatment-associated incidence density of 1/1,000 is acceptable, considering that bisphosphonate treatment is likely to reduce the incidence density of any fracture by 15/1,000 according to a large randomized trial (Black et al. 1996).