Abstract
BACKGROUND: Lipoprotein(a) [Lp(a)] is a risk factor of coronary heart disease, however, its effects on stroke are less well-defined. METHODS: We performed a single-center, retrospective case-control study in 1,953 and 196 ischemic stroke and hemorrhagic stroke in-hospital patients, respectively. Controls were healthy individuals that were matched for sex and age (±5 years) for the ischemic (1:1 ratio) and hemorrhagic (1:2 ratio) stroke. Lp(a) concentration was measured using the latex agglutination turbidimetric method. Logarithmic transformation and quartile categorization were applied to adjust for the skewed distribution of Lp(a). Conditional logistic regression models were used to assess the association between Lp(a) and stroke risk. RESULTS: Median Lp(a) concentration was higher in stroke patients when compared with controls (12.2 vs. 8.60 mg/dL) and hemorrhagic strokes (14.40 vs. 13.40 mg/dL). The conditional multivariate analysis revealed a positive association between Lp(a) and ischemic stroke (OR =2.03, 2.36, and 2.03 for quartiles 2, 3 and 4, respectively, vs. quartile 1; P<0.001). In addition, elevated Lp(a) was also significantly associated with increased hemorrhagic stroke risk, after adjusted for potential covariates (OR =1.93, 3.24, and 2.19 for quartile 2, 3 and 4 respectively vs. quartile 1, P<0.05). The stratified analyses for ischemic and hemorrhagic stroke revealed significant association between elevated log-transformed Lp(a) and ischemic stroke in men. Furthermore, there was a trend towards a higher stroke risk for younger patients compared with older patients. CONCLUSIONS: Elevated serum Lp(a) is significantly positively correlated with ischemic and hemorrhagic stroke risk in the Chinese Han population, especially among men and younger patients.