As the association between insulin resistance and ischaemic stroke is conflicting, our study aimed to examine the association between triglyceride–glucose (TyG), a surrogate marker of insulin resistance, and incident ischaemic stroke, and also to further assess the potential effect of modification by several known risk factors of stroke.

The Rural Chinese Cohort Study enrolled 11,777 participants, aged ≥ 40 years, who were free of stroke and cardiovascular disease at baseline during 2007–2008, and who were then followed during 2013–2014. TyG was determined using the following formula: Ln[fasting triglycerides (mg/dL) × fasting glucose (mg/dL)/2]. The relative risk (RR) and 95% confidence interval (CI) of incident ischaemic stroke associated with TyG were estimated using modified Poisson regression models.

During a median follow-up duration of 6 years, 677 new ischaemic stroke cases were identified. After multivariate adjustment, RR (95% CI) values for ischaemic stroke were 1.33 (1.01–1.75), 1.57 (1.17–2.10) and 1.95 (1.34–2.82) in TyG quartile (Q) 2, 3 and 4 groups, respectively, compared with Q1. A significant interaction between TyG index and age for risk of ischaemic stroke (P_interaction < 0.001) was also observed. However, no significant interaction was found between TyG index and other potential risk factors of risk for ischaemic stroke, although there were significant positive associations with female, non-smoker, non-drinker, low or moderate physical activity, non-obese and non-type 2 diabetes mellitus groups.

Elevated TyG index is an independent predictor of ischaemic stroke in the general population, and insulin resistance may be positively associated with future stroke risk.