Ninety-seven RCTs (n=276,116) were included. Of these RCTs, 35 assessed statins, 17 assessed fibrates, 8 assessed resins, 2 assessed niacin, 14 assessed n-3 fatty acids and 17 assessed dietary interventions.
Overall mortality.
Overall mortality was significantly reduced with statins (RR 0.87, 95% CI: 0.81, 0.94; heterogeneity P=0.05, I2=30%) and n-3 fatty acids (RR 0.77, 95% CI: 0.63, 0.94; heterogeneity P=0.01, I2=53%; heterogeneity appeared to be due to one lower quality RCT). There was no statistically significant difference in mortality for the other interventions examined.
Statins significantly reduced overall mortality in primary and secondary prevention studies. Heterogeneity was reduced in both subgroup analyses (P=0.50 for primary prevention and P=0.42 for secondary prevention).
For n-3 fatty acids, there was insufficient evidence to assess the effects on mortality in primary prevention studies. The sensitivity analysis found that lower quality studies of statins and fibrates tended to report higher risk reductions, although these differences were not statistically significant.
Cardiac mortality.
Cardiac mortality was significantly reduced with statins (RR 0.78, 95% CI: 0.72, 0.84; heterogeneity P=0.42, I2=3%), resins (RR 0.70, 95% CI: 0.50, 0.99; heterogeneity P=0.83, I2=0%) and n-3 fatty acids (RR 0.68, 95% CI: 0.52, 0.90; heterogeneity P=0.001, I2=66%; heterogeneity appeared to be due to one lower quality RCT). Noncardiovascular mortality.
Only fibrates significantly increased noncardiovascular mortality compared with the control (RR 1.13, 95% CI: 1.01, 1.27; heterogeneity P=0.80, I2=0%). A post hoc analysis found no increase in deaths from neoplasia, but insufficient data precluded further examination.
A univariate meta-regression found that the percentage of patients with established CHD and trial duration explained a statistically significant amount of variability in overall mortality among studies.