The association between insulin resistance assessed by estimated glucose disposal rate and stroke prevalence and mortality in non-diabetic people: evidence from two prospective cohorts

Abstract Background The estimated glucose disposal rate (eGDR), serving as a measure of insulin resistance (IR), provides a simpler and more accessible method for assessing insulin sensitivity. However, its association with stroke and mortality in non-diabetic patients remains to be fully clarified. Methods Data from the National Health and Nutrition Examination Survey (NHANES) Study (2003–2014, n = 11,063, age ≥ 45) were examined. Participants with diabetes, coronary heart disease (CHD), or missing key data were excluded. eGDR was calculated based on waist circumference, hypertension status, and glycated hemoglobin (HbA1c). The primary outcomes were stroke prevalence and all-cause, cardiovascular, and cerebrovascular disease mortality. For stroke outcomes, a cross-sectional analysis was conducted and multivariate logistic regression was employed for assessment; whereas for all mortality outcomes, longitudinal analysis was performed using multivariate Cox proportional hazards models. The association between eGDR and these outcomes was investigated using Kaplan-Meier survival analysis, Cox proportional hazards regression models, restricted cubic splines (RCS), and mediation analysis, adjusted for demographic and clinical variables. An additional cohort of 6,873 participants from the China Longitudinal Health and Aging Study (CHARLS) was applied to evaluate the association further. Results Between 2003 and 2014, the NHANES study documented 380 stroke cases. A higher eGDR showed a significant linked to lower stroke prevalence (OR 0.44, 95% CI: 0.31–0.62, P < 0.001), as well as reduced all-cause mortality (HR 0.78, 95% CI: 0.68–0.89), CVD mortality (HR 0.70, 95% CI: 0.53–0.94) and cerebrovascular diseases mortality (HR 0.54, 95% CI: 0.31–0.94) over a median 117.6 months. Each one-unit increase in eGDR was associated with a 15% decrease in stroke prevalence (OR 0.85, 95% CI 0.79–0.9). The protective impact of a higher eGDR against stroke was consistently observed across most subgroups, but significantly stronger in individuals < 60 years. CHARLS demonstrated the research findings. Conclusions Elevated eGDR levels showed an independent correlation with a significantly reduced prevalence of stroke and decreased risks of mortality in non-diabetic adults. Graphical abstract