Regular ArticleInverse association between serum lipoprotein(a) and cerebral hemorrhage in the Japanese population

IntroductionAlthough lipoprotein(a) (Lp(a)) is involved in cardiometabolic disease processes, the association between serum Lp(a) and stroke and/or its subtypes has not yet been elucidated among Japanese people. This study investigated the association between Lp(a) and the incidence of stroke and/or its subtypes in the general Japanese population.Materials and MethodsThis population-based prospective cohort study included 10,494 community-dwelling participants (4,030 males/6,464 females). The incidence of stroke and its subtypes was the primary outcome. The subjects were divided into tertiles based on their Lp(a) levels, and the risk of all stroke and stroke subtypes was examined using Cox's proportional hazard model.ResultsA total of 393 subjects (199 males and 194 females) with stroke were identified during a follow-up duration of 10.7 years. The multivariate-adjusted hazard ratios for all stroke events were 0.55 (95% confidence interval: 0.38-0.81) and 0.69 (0.49-0.99) in the 2nd (9-19 mg/dl) and 3rd tertiles (≥ 20 mg/dl) of Lp(a) in reference to the 1st tertile (< 9 mg/dl) in males, and 0.85 (0.59-1.24) and 0.76 (0.52-1.11) in 2nd (10-22 mg/dl) and 3rd tertiles (≥ 23 mg/dl) of Lp(a) in reference to the 1st tertile (< 10 mg/dl) in females. The multivariate-adjusted hazard ratios for cerebral hemorrhage were 0.26 (0.10-0.67) and 0.34 (0.15-0.76) in the 2nd and 3rd tertiles of Lp(a) in reference to the 1st tertile in males, and were 0.48 (0.23-1.04) and 0.44 (0.21-0.96) in the 2nd and 3rd tertiles of Lp(a) in females.ConclusionsLp(a) was associated with the incidence of cerebral hemorrhage in the general Japanese population, particularly among males, while a similar trend was seen among females. A low Lp(a) level may be a marker of the risk of cerebral hemorrhage in this population.