Three RCTs (n=29,375) were included, all published between 2002 and 2004. Data on all-cause mortality were included from two further trials in a sensitivity analysis.
A meta-analysis of two studies using the same primary end point (heterogeneity, p=0.69) showed a statistically significant reduction with ARBs in the composite measure of cardiovascular mortality, MI and stroke (RRR 0.87, 95% CI: 0.80, 0.96, p=0.004). The primary end point in the other trial was a composite measure of sudden cardiac death, fatal MI, death during or after percutaneous coronary intervention or coronary artery bypass graft, death due to heart failure, death associated with recent MI on autopsy, heart failure requiring hospital management, nonfatal MI, or emergency procedures to prevent MI, but excluded stroke. Once this trial was added in the meta-analysis there was no statistically significant difference in the primary end points between ARBs and the control groups (RRR 0.93, 95% CI: 0.82, 1.06, p=0.26; heterogeneity, p=0.059).
There was no statistically significant difference between ARBs and the control groups for all-cause mortality (RRR 0.96, 95% CI: 0.88, 1.06, p=0.45; heterogeneity, p=0.22).
ARBs were associated with a higher risk of MI than the control groups (RRR 1.12, 95% CI: 1.01, 1.26, p=0.041; heterogeneity, p=0.68), although these findings were heavily influenced by one trial.
For stroke, a meta-analysis of two homogeneous studies showed a statistically significant reduction with ARBs (RRR 0.75, 95% CI: 0.65, 0.87, p=0.0001; heterogeneity, p=0.87). The other trial showed an increased risk of stroke with ARBs, but the difference was not statistically significant (RRR 1.14, 95% CI: 0.97, 1.33, p=0.11).
A meta-analysis of two studies reporting data on heart failure showed no statistically significant difference between ARBs and the control groups (RRR 0.90, 95% CI: 0.80, 1.01, p=0.073).
ARBs were associated with a statistically significant reduction in new-onset of diabetes compared with the control groups (RRR 0.80, 95% CI: 0.74, 0.86, p<0.001; heterogeneity, p=0.74). The number-needed-to-treat per year was 255 (95% CI: 189, 392).
A sensitivity analysis was performed by including two further studies carried out in renal patients, with most of the participants being hypertensive. When including these two trials in the meta-analysis there was no statistically significant difference in all-cause mortality between ARBs and the control groups (RRR 0.98, 95% CI: 0.92, 1.05, p=0.57; heterogeneity, p=0.48). The RRR was not pooled for other outcomes of interest because of statistically significant heterogeneity.