A systematic review and network meta-analysis 1 included 22 studies with a total of 143 153 subjects without diabetes at randomisation. In the network meta-analysis an initial diuretic was used as the standard of comparison and the odds ratios for different drugs were: angiotensin-receptor blocker, ARB 0.62 (95% CI 0.51 to 0.77, p<0.0001); ACE inhibitor 0.67 (0.57 to 0.79, p<0.0001); placebo 0.75 (0.63 to 0.89, p=0.001); calcium-channel blocker, CCB 0.79 (0.67 to 0.92, p=0.004); β blocker 0.93 (0.78 to 1.11, p=0.43). The association of antihypertensive drugs with incident diabetes is therefore lowest for ARB and ACE inhibitors followed by placebo and CCB, β blockers and diuretics in rank order. However, the ORs for the agents differed when placebo was used as the referent agent: only the initial diuretic (OR 1.34, 95% CI 1.12 to 1.60, p=0.001) and the β blocker (1.25, 1.05 to 1.48, p=0.01) retain significance. The individual ORs for an ARB (0.84, 0.70 to 1.00, p=0.055), ACE inhibitor (0.90, 0.78 to1.04, p=0.16), and CCB (1.05, 0.90 to 1.24, p=0.53) were all non-significant, perhaps due to the lower statistical power for these comparisons than for those involving the diuretic. Odds ratios did not differ between an ARB and an ACE inhibitor (0.93, 0.77 to1.12, p=0.44), or between a diuretic and a β blocker (0.93, 0.78 to 1.11, p=0.43). Also a traditional meta-analysis was performed; thiazide diuretic and β blocker increased significantly the incidence of diabetes compared with placebo, and ACE inhibitor and ARB decreased significantly incident diabetes compared with placebo.
Another meta-analyses 2 included 12 studies with a total of 94492 subjects. Beta-blocker therapy resulted in an increased risk for new-onset diabetes mellitus (RR 1.22, 95% CI 1.12 to 1.33) compared with nondiuretic antihypertensive agents. For the secondary end points, beta blockers resulted in a 15% increased risk for stroke, with no difference for the end point of death or myocardial infarction.
A network meta-analysis 3 included 38 articles with a total of 224 140 subjects. The pairwise meta-analysis and the network meta-analysis showed that compared with placebo, ACE ihibitors and ARBs showed a significant advantage, CCBs showed a mild or no impact, and β-blockers showed an opposite effect on the risk of new-onset diabetes mellitus. ACE inhibitors and ARBs had high ranking probabilities (72.77% and 79.81%, respectively) in preventing new-onset diabetes mellitus. Meanwhile, diuretics showed the lowest ranking probabilities of 4.44%, even less than placebo, which had a ranking probability of 45.19%.
Comment: The quality of evidence is downgraded by inconsistency (heterogeneity in interventions).
Primary/Secondary Keywords