Discussion
Postmenopausal women with osteoporosis have a high risk of fracture. Several drugs were used for the treatment of osteoporosis and the prevention of fractures. However, which drug is more effective than the others remains unknown. The objective of this study was to compare the efficacy of denosumab, teriparatide, and oral bisphosphonates for preventing vertebral and nonvertebral fractures in postmenopausal women with osteoporosis.
The included RCTs did not provide treatment estimates of these interventions for the prevention of fractures. Conducting studies with large sample sizes and long follow-up head-to-head trials with multiple treatment arms is difficult. Considering these reasons, MTC is appropriate for providing important comparative assessments. This technique could offer a similar value in the face of multiple competing treatments and lack of direct comparisons.
Several meta-analyses compared the efficacy of interventions by using MTC. Freemantle et al assessed the effects of several osteoporosis treatments on the risk of fractures in men and women. Ellis et al performed a meta-analysis comparing the efficacy of bazedoxifene with the efficacy of oral bisphosphonates for reducing fracture risk in postmenopausal women with osteoporosis. To our knowledge, our study was the first to compare the efficacy of teriparatide, denosumab, and oral bisphosphonates for preventing vertebral and nonvertebral fractures in postmenopausal women with osteoporosis. Results of our meta-analysis indicated that all of the therapies, except etidronate, demonstrated a statistically significant reduction of fractures in comparison with placebo. Teriparatide and denosumab were more effective than alendronate and risedronate for reducing vertebral fractures. Teriparatide, denosumab, alendronate, and risedronate also reduced the risk of nonvertebral fractures compared with placebo. Results of subgroup analysis showed that denosumab, alendronate, and risedronate reduced the risk of hip fracture and that risedronate also reduced the risk of upper-arm fracture.
However, there were some limitations in our study. First, a potential weakness of this meta-analysis was the fact that the included trials were probably different in study design. The follow-up period of each study varied. Second, the small sample size and the lack of head-to-head trials may increase the uncertainty of results. Third, we could not assess publication bias. Despite these limitations, we believe that our analysis could contribute to clinical decision-making in the treatment of postmenopausal women with osteoporosis.