Forty-two RCTs were included in the meta-analyses: 13 short-term RCTs (n=4,414 patients, follow-up ranged from 0.02 to 0.42 years) where the risk of bias was low in five trials, high in four trials and unclear in four trials; and 29 longer-term RCTs (n=134,755, follow-up ranged from one to 5.2 years) where the risk of bias was low in all trials except one.
Short-term RCTs: Statin treatment reduced the incidence of atrial fibrillation (OR 0.61, 95% CI 0.51 to 0.74; 13 RCTs). There was significant statistical heterogeneity (p<0.001).
Longer-term RCTs: Statin treatment was not associated with a significant reduction in atrial fibrillation when compared with control (0.95, 95% CI 0.88 to 1.03; 22 RCTs). There was no evidence of significant heterogeneity. The difference between the pooled results for shorter- and longer-term trials was significant (p<0.001).
Trials of more intensive versus standard statin regimens (28,964 randomised patients and 1,419 events) yielded similar results (OR 1.00, 95% CI 0.90 to 1.12; seven RCTs). There was an indication of statistical heterogeneity (p=0.05).
Sensitivity analyses did not significantly alter the results. Subgroup analyses suggested neither evidence that statins prevented first diagnosed atrial fibrillation events nor that the effect of statin treatment differed according to previous disease status.