Weight change and risk of cardiovascular disease among adults with type 2 diabetes: more than 14 years of follow-up in the Tehran Lipid and Glucose Study

Abstract

Background: To examine the impact of weight change on incident cardiovascular disease and coronary heart disease (CVD/CHD) among an Iranian population with type 2 diabetes mellitus (T2DM). Methods: The study population included 763 participants with T2DM aged ≥ 30 years without a history of CVD and cancer at baseline. Two weight measurements done at baseline and about 3 years later. Based on their weight change, they categorized into: > 5% loss, 3–5% loss, stable (± < 3%), 3–5% gain, > 5% gain. Participants were then followed for incident CVD/CHD annually up to 20 March 2018. Multivariable Cox proportional hazard models, adjusted for age, sex, body mass index, educational level, current smoking, glucose-lowering drug use, family history of CVD, hypertension, hypercholesterolemia, chronic kidney disease, and fasting plasma glucose (FPG) were applied to estimate the hazard ratios (HRs) and 95% confidence intervals (CIs) of weight change categories for incident CVD/CHD, considering stable weight as reference. Results: After the weight change measurement, during a median follow-up of 14.4 years, 258 CVD and 214 CHD occurred. Over 5% weight gain was associated with reduced risks of CVD and CHD development by the HRs of 0.70 [95% CI 0.48–1.01; P-value: 0.058] and 0.61 [0.40–0.93], respectively, in multivariable analysis. After further adjustment for FPG change, the HRs of weight gain > 5% were attenuated to 0.75 [0.51–1.10; P-value: 0.138] and 0.66 [043–1.01; P-value: 0.053] for incident CVD and CHD, respectively. The effect of weight loss > 5% was in opposite direction among those older versus younger than 60 years; with suggestive increased risk (not statistically significant) of incident CHD/CVD for the older group. Moreover, weight gain > 5% significantly reduced the risk of CHD only among those older than 60 years (P-value for interaction < 0.2). Furthermore, weight gain > 5% had an association with lower risk of CVD and CHD among sulfonylurea users (0.56 [0.32–0.98] for CVD and 0.54 [0.29–0.99] for CHD). Conclusions: Our results with a long-term follow-up showed that weight gain > 5% was associated with better CVD/CHD outcomes among Iranian participants with T2DM, especially older ones. Moreover, we did not find an unfavorable impact on incident CVD/CHD for sulfonylurea-induced weight gain.