Association of estimated glucose disposal rate with atrial fibrillation, heart failure and cardiovascular mortality in patients with diabetes: a prospective cohort study from the UK Biobank.

Background: Estimated glucose disposal rate (eGDR) was a novel non-insulin-based marker of insulin resistance (IR), which had been used in many studies to evaluate the clinical prognosis of diabetes. However, the association of eGDR with atrial fibrillation (AF), heart failure (HF) and cardiovascular mortality in patients with diabetes remains unclear. Methods: The study utilized UK Biobank data from 31,733 participants. Kaplan-Meier curves and Log-rank tests assessed AF, HF, and cardiovascular mortality incidence. Multivariate Cox models and restricted cubic splines analyzed the associations of eGDR with these outcomes. Polygenic Risk Score (PRS) analysis evaluated the joint effects of eGDR and PRS. Boruta algorithm filtered key predictive variables. Subgroup analysis was performed using cardiovascular high-risk factors, and mediation analysis explored the relationships of eGDR with the outcomes. Results: Subjects with higher eGDR were more likely to be female, younger, more physically active, non-smoker, and non-drinker. The cumulative incidence of AF, HF, and cardiovascular mortality in the higher quartiles of GDR were significantly lower than those in the lowest quartile (log-rank Conclusions: This study demonstrated that higher eGDR levels were associated with a decreased risk of AF, HF, and cardiovascular mortality. Therefore, eGDR may serve as a valuable tool for predicting the risk of AF, HF, and cardiovascular mortality in patients with diabetes.