PURPOSE

Our objective was to determine the association of clinically relevant adverse events from a systematic review and meta-analysis of statin randomized controlled trials (RCT).

METHODS

We performed the meta-analysis in the manner of a Cochrane Collaboration systematic review. Outcomes were discontinuances of therapy or muscle-related symptoms due to adverse events. We searched for articles from 1982 through June 2006 in MEDLINE and other databases. The main inclusion criteria were double blind, placebo controlled RCTs with a monotherapy intervention of any marketed statin and active surveillance of adverse events. We excluded studies of drug interactions, organ transplants, or exercise, or those not meeting all of the study quality criteria. The primary analysis was a statin formulation stratified fixed-effect model using Peto odds-ratios (POR). Secondary analyses explored the stability of the primary results.

RESULTS

Over 86,000 study participants from 119 studies were included. Available statins were associated with a lower POR of discontinuance (overall: 0.88 [0.84, 0.93], largest effect with pravastatin: 0.79 [0.74, 0.84]), an elevated POR of rhabdomyolysis (1.59 [0.54, 4.70]) and myositis (2.56 [1.12, 5.85]), and null odds of myalgia (1.09 [0.97, 1.23]). Cerivastatin by comparison demonstrated larger PORs for discontinuances and muscle-related adverse events. Secondary analyses demonstrated the stability of the results.

CONCLUSIONS

Overall, discontinuation of statin therapy due to adverse events was no worse than placebo. The risks of muscle-related adverse events were in general agreement with the known risks of statins.