To assess the ability of olmesartan (OLM) to prevent or delay left ventricular remodeling and hypertrophy in patients with type 2 diabetes.
This prespecified ECG substudy of Randomised OlmesArtan and Diabetes MicroAlbuminuria Prevention (ROADMAP), which compared OLM with placebo, assessed the signs of left ventricular remodeling in patients with a 12-lead ECG at baseline and after at least 2 years. Cornell voltage QRS duration product (primary objective), Cornell voltage index and Sokolow–Lyon index were assessed.
In total, 9418 ECG recordings and 1513 patients from ROADMAP were analyzed (placebo, n = 736; OLM, n = 777). Quartiles defined by baseline Cornell voltage QRS duration product were assessed and the proportion of patients in the highest quartile (≥200 mV∗ms) increased from 24.0 to 26.5% in the placebo group and decreased from 25.5 to 22.3% in the OLM group [odds ratio (OR) 0.598 (95% confidence interval [CI] 0.440–0.813); P = 0.0011]. The OR did not change after adjustment for baseline parameters. By the end of study, 38.7% of patients in the placebo group and 34.7% in the OLM group shifted from a lower to a higher quartile or remained in the highest quartile of Cornell voltage QRS duration product [OR 0.797 (95% CI 0.637–0.996); P = 0.0465]. This translated into a 20.3% risk reduction with OLM and suggested OLM attenuated the progression of left ventricular remodeling versus placebo.
OLM substantially delayed the development of left ventricular remodeling in type 2 diabetes. This effect was not explained by the differences in blood pressure control. Thus, OLM delayed the onset of microalbuminuria, as well as the ECG signs of cardiac structural adaptation in type 2 diabetes.