Seven cholesterol-lowering RCTs (2,603 patients), 6 blood- pressure-lowering RCTs (7,572 patients) and 5 glucose-lowering RCTs (5,159 patients) were included in the review, but not all were combined in the meta-analyses.
No significant statistical heterogeneity was found. The control-arm event rates were highest in the lipid-lowering trials and lowest in the glucose-lowering trials.
Lipid lowering (5 RCTs).
The cardiac event rates were significantly lower for lipid-lowering drugs than the control. The aggregated cardiac events RR was 0.75 (95% CI: 0.61, 0.93). The person-years needed-to-treat were 106 (95% CI: 62, 366). The cardiac event rates were significantly lower in secondary prevention studies (3 RCTs; RR 0.77, 95% CI: 0.62, 0.96) than in primary prevention studies (2 RCTs; RR 0.44, 95% CI: 0.17, 1.20).
Blood-pressure lowering (3 RCTs).
The cardiac event rates were significantly lower for blood-pressure- lowering drugs than the control. The aggregated cardiac events RR was 0.73 (95% CI: 0.57, 0.94). The person-years needed-to-treat were 157 (95% CI: 88, 726).
Glucose lowering (2 RCTs, both primary prevention).
There was no statistically-significant difference in the cardiac event rates for glucose-lowering regimens versus control. The aggregated cardiac events RR was 0.87 (95% CI: 0.74, 1.01). The person-years needed-to-treat was 419 (95% CI: 197, not meaningful).
Similar patterns of results were seen across individual cardiovascular outcomes; the results were reported in the review.
The exclusion of one earlier RCT did not affect the results substantially.