Left ventricular hypertrophy (LVH) is highly prevalent in patients with coronary artery disease (CAD), even in the absence of hypertension and is an independent predictor of cardiovascular mortality. Metformin has been shown to regress LV mass (LVM) in animal models of LVH. We hypothesise that metformin may regress LVH in non-diabetic and normotensive CAD patients with pre-diabetes and/or insulin resistance.Methods
In this randomised double-blind placebo controlled trial, 68 patients (mean age 65±8 y, 25% females) with prediabetes (defined using American Diabetes Association criteria of HbA1c ≥39 mmol/mol and less than 48 mmol/mol) and/or insulin resistance (defined by fasting insulin resistance index ≥2.7) were assigned to receive either metformin (2 g daily dose) or placebo for 12 months. An intention-to-treat (ITT) and per-protocol analysis was designed to determine the effect of metformin on the following study endpoints: Primary endpoint was change in left ventricular mass indexed to height 1.7 (LVMI), assessed by magnetic resonance (MRI) imaging; other endpoints were changes in LVM, changes in body weight, office blood pressure (BP) and biomarkers.Results
In the ITT analysis (n=61), metformin treatment significantly reduced: LVMI (metformin −2.7±2.3 g/m1.7 vs placebo −1.4±2.7 g/m1.7; p=0.05), LVM (metformin −6.5±5.6 g vs placebo −3.45±6.5 g; p=0.05), body weight (lowered by 3.6 kgs, p=0.002), office systolic BP (metformin −4.8±15.6 mmHg vs placebo 4.6±15.7 mmHg; p=0.02) and reduced concentration of thiobarbituric acid reactive substances (TBARs), a biomarker for oxidative stress (p=0.04). In the on-per protocol analysis (n=56), metformin resulted in a greater reduction of LVMI (metformin −3.1±1.9 g/m 1.7 vs placebo −1.2±2.7 g/m 1.7; p=0.005), LVM (metformin −7.5±4.6 g vs placebo −3.1±6.3 g; p=0.005) and greater weight reduction of 4.2 kgs (p=0.001).Conclusions
Metformin treatment significantly reduced LVMI, LVM, office SBP, body weight and oxidative stress. These results reveal a novel mechanism for the cardioprotective effect of metformin and raise the possibility of using metformin in patients without type 2 diabetes with CAD.