Low-energy fractures of the humeral shaft and bisphosphonate use

Atypical fractures of the femur have been reported to occur in patients on long-term treatment with bisphosphonates; however, causality has not been proven, and it is not known whether similar fractures may occur in other long bones. We addressed this issue by examining the relationship between humeral shaft fractures and bisphosphonate use. We identified all patients aged ≥50 years consecutively admitted to a single center with a new fracture of the humerus. All individual radiographs were examined and fracture site was classified. A case-control study was undertaken in patients with humeral shaft fractures, and controls were sex- and age-matched patients with proximal humeral fractures in a 1:4 ratio. Patients with shaft fractures and radiographic characteristics similar to those of atypical femoral fractures were compared with those with ordinary shaft fractures. The association between "atypical" fractures and bisphosphonate or glucocorticoid use was examined. A total of 198 patients had a low-energy fracture of the humerus; 20 of these patients had a shaft fracture (10%). These 20 patients (cases) were matched with 80 patients with proximal fractures (controls). Bisphosphonates were used by 5% of cases and by 6.3% of controls (odds ratio [OR], 0.80; 95% confidence interval [CI], 0.09-6.85); glucocorticoids were used by 10% of cases and 8.8% of controls (OR, 1.15; 95% CI, 0.23-5.83). There was no difference in cortical thickness between cases and controls and bisphosphonate or glucocorticoid users and nonusers. Four of the 20 patients with shaft fractures had "atypical" radiographic features, with significantly increased cortical thickness, but none of these had ever been treated with bisphosphonates or glucocorticoids. Our results show that low-energy fractures of the humeral shaft with "atypical" radiographic characteristics are infrequent and are not associated with the use of bisphosphonates or glucocorticoids.