Twenty-seven relevant RCTs were identified (n=175,634). Detsky quality scores of RCTs ranged from 16 to 20. Sixteen RCTs were double-blind. Mean duration of follow-up was 3.4 years (range 0.3 to 5.5 years). A placebo was used in 10 RCTs.
Risk of all-cause death was reduced by dihydropyridine CCBs (OR 0.96, 95% CI 0.93 to 0.99). Although risk of all-cause death was significantly reduced by all CCBs, it was not significantly affected by non-dihydropyridine CCBs.
Risk of heart failure was increased by CCBs compared with active treatment (OR 1.17, 95% CI 1.11 to 1.24, p=0.0001 for heterogeneity) and decreased compared to placebo (OR 0.72, 95% CI 0.59 to 0.87) and in the subgroup of coronary heart disease patients (OR 0.76, 95% CI 0.61 to 0.95), but not in hypertensive patients (for these patients there was significant heterogeneity, p=0.001). ACEs decreased risk of heart failure compared to CCBs (OR 1.19, 95% CI 1.08 to 1.31).
CBCs did not increase the risk of: myocardial infarction (significant heterogeneity, p=0.004); cardiovascular death; and major cardiovascular events (significant heterogeneity, p=0.0001). Compared to placebo alone, CCBs significantly reduced risk of major cardiovascular events (OR 0.76, 95% CI 0.62 to 0.93, p=0.002 for heterogeneity).
CBCs decreased risk of fatal or nonfatal stroke (OR 0.86, 95% CI 0.82 to 0.90) and when compared to ACEs (OR 0.87, 95% CI 0.78 to 0.97). This result was only significant for dihydropyridine CCBs (OR 0.85, 95% CI 0.80 to 0.90) and not for non-dihydropyridine CCBs.
Additional sensitivity analysis: For all-cause death, the protective effect of CBCs was not present in trials with less than 20% of diabetic patients, more than 20% coronary heart disease patients, more than 10% heart failure patients and more than 10% previous myocardial infarction patients. Results of other sensitivity analyses were reported.
Meta-regression analysis for potential effect modifiers: A significant relationship was identified between blood pressure reduction and a favourable effect of CCBs on cardiovascular death, major cardiovascular events, heart failure and stroke. There was an association between major cardiovascular events and smoking or diabetes and an inverse relationship between baseline presence of heart failure and heart failure events.
Publication bias was identified only for myocardial infarction.