Studies have shown that non-traditional lipid profiles have a better association with stroke than traditional blood lipids in clinical applications, other studies have drawn different conclusions.

This study was a large-scale study with a median follow-up of 8.4 years. The hazard ratio (HR) and 95% Confidence interval (CI) of lipid variables for risk of incident stroke were analyzed by multivariable Cox proportional hazard models.

During the follow-up, 502 new strokes (310 ischemic, 187 hemorrhagic, and 5 unclassified strokes) occurred among the 5099 hypertensive patients. Comparing with the lowest quarter, the HR of future ischemic stroke (IS) in the highest were 1.41(95%CI, 1.03–1.92) for TC, 1.60 (95%CI, 1.15–2.22) for TG, 1.03 (95%CI, 0.75–1.42) for HDL-C, 1.77 (95%CI, 1.29–2.44) for LDL-C, 1.42 (95%CI, 1.03–1.94) for non-HDL, 2.09 (95%CI, 1.45–3.00) for TC/HDL, 2.08 (95%CI, 1.46–2.96) for LDL/HDL, 1.86 (95%CI 1.33–2.60) for TG/HDL, respectively. No significant association was observed between lipid-related indicators and hemorrhagic stroke. The results of statistical differences showed that the correlation between LDL/HDL and the risk of ischemic stroke in non-traditional lipids was higher than that of other traditional lipids (P < 0.001), except for LDL (P = 0.056).

We didn’t find that HDL was associated with the risk of stroke and all the lipid parameters were not associated with the risk of hemorrhagic stroke. LDL/HDL was associated with a higher risk of ischemic stroke than other lipids and should be considered for clinical diagnosis and future disease prevention.