Differing effects on hip and nonspine fracture risk reduction among N-containing bisphosphonates: Review and meta-analysis

Abstract

Objectives: N-containing bisphosphonates reduce the risk of vertebral fractures (VFx): however, results have not shown consistent effects on hip and other nonspine fractures. Systematic reviews are useful to summarize effects that often vary among individual trials, and meta-analysis provides a more precise estimate when results are consistent across pooled trials.(1) Methods: We surveyed earlier systematic reviews and meta-analyses ogether with subsequent reports (2-4) of randomized, placebo-controlled trials with data on nonspine and/or hip fractures, and used meta-analysis where appropriate to test for heterogeneity and derive pooled estimates. Results: For risedronate (RIS) (n=12958 subjects in trials), the relative risk (RR) reduction was 27% for nonspine fracture (RR=0.73; 95% CI= 0.61, 0.87)1 and 26% for hip fracture (RR=0.74; 0.58, 0.94); there was no significant interaction of treatment and age >80 and there was no heterogeneity by dose (2.5 or 5 mg/day). For alendronate (ALN), heterogeneity existed; doses >10 mg/day were significantly more effective for reduction of nonspine and hip fracture. The RR reduction with ALN >10 mg (n=3723) was 49% for nonspine fractures (RR=0.51; 0.38, 0.69).1 The RR reduction for hip fracture was 55% (RR=0.45; 0.28, 0.71) when osteoporotic women in FIT (ALN 5 mg/day in years 1–2 then 10 mg/day for remainder of trial) were included (n=6804), and 55% (RR=0.45; 0.18, 1.13) for doses >10 mg/day (n=3723).1 One post-hoc analysis suggested an effect of ibandronate (IBN) on non-spine fractures in patients with severe osteoporosis, but the limited fracture data and differences in dosing precluded pooling studies for meta-analysis. No data are available on the effect of IBN on hip fracture. Consequently, there is no consistent evidence of hip or nonspine fracture risk reduction available for meta-analysis. No data are available on the effect of IBN on hip fracture. Consequently, there is no consistent evidence of hip or nonspine fracture risk reduction available for meta-analysis with IBN at a dose of 2.5 mg/day.(4) Conclusions: The magnitude of effect on hip fracture appears to differ among drugs. Based on current data, ALN reduces the risk of hip and nonspine fracture by 49–55% and RIS by 26–27%. There is insufficient and/or inconsistent evidence of an effect on these fractures for ibandronate.(4) 1. Cranney et al. Endocr Rev 2002;23:570–8. 2. Papapoulos Osteoporos Int 2005;16:468–74. 3. McClung et al. NEJM 2001;344:333–40. 4. Chesnut et al. JBMR 2004;19:1241–9.