Abstract 2752: Prevention of Stroke by Antihypertensive Drugs: An Updated Network Meta-analysis
Lowering blood pressure with drug therapy prevents stroke, but whether a specific type of antihypertensive drug is better for this purpose than others has remained controversial. Traditional meta-analyses that include the most recent clinical trial data have shown significantly better stroke prevention with an initial calcium channel blocker (CCB), compared to an initial diuretic/beta-blocker, which in turn is significantly better than an initial angiotensin converting-enzyme (ACE-) inhibitor. Although these meta-analyses show no significant inhomogeneity for stroke, we performed a network meta-analysis (which accounts for both direct and indirect comparisons in clinical trials; Stat Med., 2002;21:2313) to compare the relative potency of five classes of initial antihypertensive drugs with placebo. In 64 clinical trials involving 309,657 subjects, initial drug therapy (and the number of patients with new strokes/number of subjects at risk) included: Placebo: 3061 of 67,410 (4.54%); ACE-inhibitor: 1800 of 43,908 (4.10%); angiotensin II receptor blocker (ARB): 817 of 17,124 (4.77%); Beta-blocker: 1916 of 55,810 (3.43%); CCB: 2220 of 74,286 (3.00%); and Diuretic: 1543 of 51,119 (3.02%). Using Placebo as the standard of comparison (39 arms), the incoherence was very small (ω= 0.000009047), and the odds ratios, 95% confidence intervals, and P-values were: Unlike the previous meta-analyses, but similar to the network meta-analysis ( JAMA, 2003;289:2534–44) that included data from only 42 trials and 192,478 patients, these results suggest that all antihypertensive drugs significantly prevented stroke in clinical trials, and that an initial calcium antagonist is only slightly (and not significantly) better than an initial diuretic.