Composite indices combining the TyG index with non-traditional lipid parameters were consistently associated with incident stroke in middle-aged and older Chinese adults, with TyG-LCI showing the strongest association (HR 1.84; 95% CI 1.37-2.47), suggesting these measures may serve as adjuncts for primary stroke risk assessment.
During a median 7-year follow-up of 7,578 stroke-free participants, 481 incident strokes occurred.
TyG-LCI showed the strongest association with incident stroke risk among all composite indices after full adjustment.
Restricted cubic spline analyses revealed non-linear exposure-response relationships between composite indices and incident stroke risk.
Composite indices yielded higher AUC values than conventional lipid parameters for predicting incident stroke.
The TyG index partially mediated the associations of four non-traditional lipid parameters (CRI-II, AC, Non-HDL-C, and CRI-I) with incident stroke.
Seven non-traditional lipid parameters were combined with the TyG index to create composite indices for stroke risk assessment, all derivable from routine glucose and lipid tests.
This research suggests that combining a measure called the triglyceride-glucose (TyG) index — which reflects how well the body handles insulin — with other non-standard cholesterol-related measurements creates composite scores that can better predict stroke risk than traditional lipid tests alone. The study followed nearly 7,600 Chinese adults aged 45 and older for about 7 years, during which 481 strokes occurred. Among the combinations tested, the TyG-LCI (TyG index multiplied by the lipid combination index) was the strongest predictor, with people in the highest quarter of scores having an 84% higher stroke risk compared to those in the lowest quarter. The study also found that the TyG index itself plays an important role in explaining why some other lipid measurements are linked to stroke. For example, more than 60% of the relationship between non-HDL cholesterol and stroke risk appeared to operate through the TyG index pathway, suggesting insulin resistance is a key driver connecting lipid abnormalities to stroke. Three specific composite scores — TyG-AIP, TyG-CRI-I, and TyG-LCI — not only outperformed conventional lipid measurements in discriminating who would or would not have a stroke, but also meaningfully improved the ability to correctly reclassify individuals into higher or lower risk categories. This research suggests that these composite scores, all calculable from routine blood glucose and lipid panel tests already commonly ordered in clinical practice, could serve as useful additional tools for identifying middle-aged and older adults at higher risk of stroke. Because these measures capture both insulin resistance and abnormal fat metabolism simultaneously, they may provide a more complete picture of cardiovascular risk than any single marker alone. This is particularly relevant for populations with high rates of metabolic disorders and where stroke remains a leading cause of death and disability.
Xiao D, Zhou L, Xiao Y, Yu C, Chen W, Fan J, et al.. (2026). Joint assessment of triglyceride-glucose index and non-traditional lipid parameters for incident stroke risk in middle-aged and older chinese adults: A prospective cohort study based on CHARLS.. Journal of stroke and cerebrovascular diseases : the official journal of National Stroke Association. https://doi.org/10.1016/j.jstrokecerebrovasdis.2026.108740