Effects of statins on bone mineral density: A meta-analysis of clinical studies

Context: Statins inhibit HMG-CoA reductase, preventing synthesis of mevalonate but also of isoprenoids, which affect osteoclast activity. Aminobisphosphonates share this effect. In vitro and in vivo, statins show convincing anabolic and anti-resorptive bone effects. However, in a clinical meta-analysis (MA), they did not prevent hip fractures. Objective and design: Our meta-analysis studied the impact of statins on bone mineral density (BMD) at various sites and compared the effects of lipophilic and more hydrophilic statins. Data sources: Our PubMed and Ernbase queries using two keywords (statins, BMD) were updated to October 2006. Data collection: Two readers independently collected BMDs from studies. Data synthesis: Twenty-one studies, mostly observational (three randomized controlled trials and one pseudo-randomized study), were assessed. Two studies were excluded (no control groups). Three studies could not be analyzed. The sixteen studies analyzed mainly included postmenopausal osteopenic women (2971 patients under statins). Statins significantly increased BMD at total hip (TH) and femoral neck (FN). Effect sizes (ESs) were modest: 0.21 at TH (95% confidence interval [CI]: 0.16-0.25) and 0.20 at FN (CI: 0.08-0.28). Among women, statins acted similarly (ES: 0.20 for TH and 0.18 for FN; CI: 0.14-0.25 and 0.06-0.31 respectively); lipophilic statins (simvastatin, lovastatin) almost entirely caused this effect, at both TH (ES: 0.20; CI: 0.15-0.26) and FN (ES: 0.22; CI: 0.06-0.37). Conclusion: Our findings of modest but statistically significant beneficial effects of statins on hip BMD should promote large double-blind randomized controlled trials on their bone effects, in view of their major beneficial cardiovascular effects with excellent safety profile. (c) 2007 Elsevier Inc. All rights reserved.