Effects of diabetes and obesity in patients undergoing coronary artery bypass graft surgery: ten-year results from the Arterial Revascularisation Trial

Abstract

Abstract Background Coronary artery bypass graft (CABG) surgery is an effective way to manage advanced coronary heart disease. Diabetes is associated with higher risk of mortality in patients undergoing CABG but the effects of obesity, and the interactions between diabetes and obesity in the context of CABG, have not been fully explored. Methods The Arterial Revascularisation Trial (ART) compared ten-year mortality in patients undergoing CABG randomised to single or bilateral internal thoracic arterial grafts. Effects of diabetes and obesity on mortality and risk (sternal wound complications) were explored. Diabetes was defined according to clinical history at baseline and combined insulin and non-insulin dependent patients, and obesity was defined as body mass index (BMI) ≥30kg/m2 at baseline. Results Information on diabetes, obesity and ten-year mortality were documented for 3094 patients who were included in the analysis. Mean age was 64 years, 86% male, mean weight 82kg and BMI 28. Diabetes was documented in 24% and obesity in 30% and as expected obese patients were more likely to be diabetic. The hazard ratios (95% confidence intervals) for ten-year mortality using the no diabetes, non-obese group as a reference were 1.33 (1.08, 1.64) (diabetes, non-obese), 0.93 (0.75, 1.16) (no diabetes, obese) and 1.18 (0.91, 1.54) (diabetes, obese) groups (Figure 1). Similar patterns of risk were observed whether patients were randomised to receive single or bilateral internal thoracic arteries and diabetes appeared to be the main driver of elevated risk of early sternal wound complications following CABG. Conclusions Obesity and diabetes are closely associated but our analysis shows that patterns of risk of death after CABG are not additive for these characteristics, and obesity may actually be associated with lower death rates in patients with and without diabetes. These findings will be explored further to understand mechanisms and interactions with other factors. Figure 1. Hazard ratios at 10 years Funding Acknowledgement Type of funding source: None