Level 1: Likely reliable evidence
A systematic review of 39 English-language, randomized, double-blind, placebo-controlled trials, each involving at least 100 adults with hyperlipidemia, evaluated adverse effects associated with statin monotherapy (Circulation. 2006;114:2788-2797). Myalgias were defined as any musculoskeletal pain or symptoms without documented creatine kinase (CK) elevation; rhabdomyolysis was defined as CK elevation at least 10 times the normal upper limit; hepatotoxicity was defined as any notation of elevated alanine aminotransferase (ALT) or aspartate aminotransferase (AST).
Currently approved statins were evaluated in 35 randomized trials with 74,102 patients: atorvastatin (Lipitor) in five trials with 12,148 patients, fluvastatin (Lescol) in four trials with 2,295 patients, lovastatin (Altoprev, Mevacor) in nine trials with 17,178 patients, pravastatin (Pravachol) in eight trials with 15,262 patients, rosuvastatin (Crestor) in four trials with 835 patients, and simvastatin (Zocor) in seven trials with 26,649 patients (two trials evaluated two different statins). Cerivastatin (Baycol) was analyzed separately (four trials with 2,282 patients). Average follow-up time was 17 months (range 1.5-64.8 months).
Overall, comparing statins (except cerivastatin) vs. placebo, there were no significant differences in rates of myalgias (15.4% vs. 18.7% in 21 trials with 48,138 patients), CK elevation (0.9% vs. 0.4% in 16 trials with 41,457 patients), rhabdomyolysis (0.2% vs. 0.1% in 20 trials with 68,110 patients), or drug discontinuation due to any adverse effects (5.6% vs. 6.1% in 26 trials with 45,268 patients). Rates of aminotransferase elevations were 1.4% with statins compared with 1.1% with placebo (P <.01, NNH 333) in 28 trials with 62,184 patients. Comparing individual statins vs. placebo, the only significant or nearly significant differences were rates of myalgias with atorvastatin (5.1% vs. 1.6%, P =.04, NNH 29), rhabdomyolysis with rosuvastatin (0.5% vs. 0%, P =.08), and aminotransferase elevations with fluvastatin (1.6% vs. 0.3%, P <.01, NNH 77) and with lovastatin (1.2% vs. 0.6%, P =.05, NNH 166).
Compared with placebo, cerivastatin was associated with increased rates of CK elevation (3.7% vs. 2.9%, P =.06), rhabdomyolysis (1.1% vs. 0%, P <.01, NNH 91), and aminotransferase elevations (1% vs. 0%, P <.01, NNH 100).
There were no significant differences in myalgias (1.9% vs. 2.9%) or drug discontinuation due to any adverse effects (3% vs. 1.1%).