Network Meta-Analysis of Pharmacological Agents for Osteoporosis Treatment and Fracture Prevention

Background and Objective: Osteoporosis afflicts a large number of populations in the world and is featured by systemic impairment of bone mass and strength which may further trigger an increase in the risk of fragile fractures. This network meta-analysis (NMA) is designed to distinguish therapies more preferable than others with respect to efficacy and safety. Methods: We searched the medical literature for relevant studies systematically. Both direct and indirect evidence were synthesized to compare the efficacy, described by odds ratios (OR) and 95% credible intervals (CrI). Moreover, the surface under cumulative ranking curve was calculated to rank probabilities with respect to clinical outcomes. The new non-vertebral fractures, hip and wrist fractures, and adverse events were evaluated in this NMA. Results: Patients treated by alendronate, denosumab, teriparatide were associated with a reduced risk of new non-vertebral fractures compared to those treated by placebo. Alendronate, denosumab and zoledronic acid had better efficacy in preventing hip fractures. With respect to wrist fractures prevention, no significant difference was observed. Zoledronic acid exhibited significantly increased risk of adverse events than placebo, alendronate, denosumab, and raloxifene. According to SUCRA, teriparatide ranked highest in new non-vertebral fractures prevention, etidronate and denosumab balanced safety and efficacy well. Conclusion: In summary, teriparatide appeared to be the most efficacious drug for preventing new non-vertebral fractures, while etidronate and denosumab were preferable for balancing safety and efficacy well