Effect of osteoporosis treatments on risk of non-vertebral fractures: review and meta-analysis of intention-to-treat studies.

Most osteoporosis treatments have proven efficacy in reducing the risk of vertebral fractures, whereas evidence is less straightforward for prevention of non-vertebral fractures. Conclusions as to the efficacy of a treatment should be based primarily on analyses of the intention to treat (ITT) population rather than on exploratory subgroup analyses; however, non-vertebral anti-fracture efficacy has been largely derived by post-hoc subgroup analyses. This review and meta-analysis was performed to assess non-vertebral anti-fracture efficacy of several osteoporosis therapies, including a more stringent assessment of the ITT populations. Data on non-vertebral anti-fracture efficacy, a defined endpoint of the ITT analyses and confirmed by radiographs, were obtained from randomized, placebo-controlled, phase III clinical trials of at least 3-year duration. Meta-analyses were performed for the two bisphosphonates, alendronate and risedronate. Relative risks (RR), 95% confidence intervals (CI) and statistical significance for active treatment compared with placebo were calculated. Eleven clinical trials met the criteria for review, three of which showed statistically significant ( P < or =0.05) non-vertebral anti-fracture efficacy in the ITT population: two trials with risedronate and one trial with strontium. A meta-analysis showed significant reductions in the relative risk of non-vertebral fracture for both alendronate (RR=0.86; 95% CI: 0.76-0.97, P =0.012) and risedronate (RR=0.81; 95% CI: 0.71-0.92, P =0.001). Risedronate and strontium ranelate were the only treatments to show non-vertebral anti-fracture efficacy in this robust assessment of anti-fracture efficacy of osteoporosis therapy using ITT populations in trials of 3 years or more in duration. Risedronate was the only agent shown to demonstrate efficacy in more than one trial. Meta-analysis showed that both alendronate and risedronate provide non-vertebral anti-fracture efficacy.