LAUSR

Objective To investigate sex and race differences in the association between fasting blood glucose (FBG) and risk of ischemic stroke (IS). Methods This prospective longitudinal cohort study included adults age ≥45 years at baseline in the Reasons for Geographic And Racial Differences in Stroke Study, followed for a median of 11.4 years. The exposure was baseline FBG (mg/dL); suspected IS events were ascertained by phone every 6 months and were physician-adjudicated. Cox proportional hazards were used to assess the adjusted sex/race-specific associations between FBG (by category and as a restricted cubic spline) and incident IS. Results Of 20,338 participants, mean age was 64.5 (SD 9.3) years, 38.7% were Black, 55.4% were women, 16.2% were using diabetes medications, and 954 IS events occurred. Compared to FBG <100, FBG ≥150 was associated with 59% higher hazards of IS (95% confidence interval [CI] 1.21–2.08) and 61% higher hazards of IS among those on diabetes medications (95% CI 1.12–2.31). The association between FBG and IS varied by race/sex (hazard ratio, FBG ≥150 vs FBG <100: White women 2.05 [95% CI 1.23–3.42], Black women 1.71 [95% CI 1.10–2.66], Black men 1.24 [95% CI 0.75–2.06], White men 1.46 [95% CI 0.93–2.28], pFBG×race/sex = 0.004). Analyses using FBG splines suggest that sex was the major contributor to differences by race/sex subgroups. Conclusions Sex differences in the strength and shape of the association between FBG and IS are likely driving the significant differences in the association between FBG and IS across race/sex subgroups. These findings should be explored further and may inform tailored stroke prevention guidelines.