RESUMO
BACKGROUND: To estimate glucose disposal rate (eGDR) as a newly validated surrogate marker of insulin resistance. Few studies have explored the association between changes in eGDR levels and stroke incidence. This study aims to explore the effect of the level of eGDR control on stroke and events. METHODS: Data were obtained from the China Longitudinal Study on Health and Retirement (CHARLS). The eGDR control level was classified using K-means cluster analysis. Logistic regression analysis was used to explore the association between different eGDR control levels and incident stroke. Restrictive cubic spline regression was used to test the potential nonlinear association between cumulative eGDR and stroke incidence. RESULTS: Of the 4790 participants, 304 (6.3%) had a stroke within 3 years. The odds ratio (OR) was 2.34 (95% confidence interval [CI], 1.42-3.86) for the poorly controlled class 4 and 2.56 (95% CI, 1.53-4.30) for the worst controlled class 5 compared with class 1 with the best controlled eGDR. The OR for well-controlled class 2 was 1.28 (95% CI, 0.79-2.05), and the OR for moderately controlled class 3 was 1.95 (95% CI, 1.14-3.32). In restrictive cubic spline regression analysis, eGDR changes are linearly correlated with stroke occurrence. Weighted quartile and regression analysis identified waist circumference and hypertension as key variables of eGDR for predicting incident stroke. CONCLUSIONS: Poorly controlled eGDR level is associated with an increased risk of stroke in middle-aged and elderly people. Monitoring changes in eGDR may help identify individuals at high risk of stroke early.