A5862 Does diabetes mellitus modify the effect of antihypertensive treatment across blood pressure levels? Systematic review and metaregression analysis

Abstract

Objectives: Previous meta-analyses have found different treatment effects in people with diabetes compared to non-diabetics at low blood pressure levels. These analyses are susceptible to ecological bias, however, because studies in people with diabetes differ from those in non-diabetics. The aim of this systematic review was to assess within-trial differences in treatment effect between diabetics and non-diabetics, to minimize the risk of ecological bias. Methods: We performed a systematic review according to the recommendations from the Cochrane Collaboration. We included randomized controlled trials of antihypertensive treatment versus placebo or different treatment goals, including both people with and without diabetes. Relative risks in treated versus controls were calculated for each patient group in each study. Relative treatment effects in people with diabetes compared to non-diabetics were calculated for each study. Relative treatment effects were analyzed in relation to baseline systolic blood pressure using metaregression. Results: 19 trials, including 123 211 participants, were analyzed. Diabetes modified treatment effect across blood pressure levels for myocardial infarction (p = 0.04) and heart failure (p = 0.01). For high baseline blood pressure values, treatment was more beneficial in people with diabetes compared to non-diabetics, whereas at levels below 152 mm Hg for myocardial infarction and 141 mm Hg for heart failure, treatment was less beneficial in people with diabetes. Interactions were non-significant for mortality outcomes and stroke. Conclusion: The effect of antihypertensive treatment depends on an interaction between systolic blood pressure and diabetes. These results further support that intensive blood pressure lowering may be harmful in the diabetic population.