Abstract P404: Sex-specific Associations of Plasma Endothelin-1 With Incidence of Stroke in African Americans: the Jackson Heart Study

Abstract

Background: Endothelin-1 (ET-1) is a vasoconstrictive peptide with cerebral action, also associated with the degree of atherosclerosis. Previous research found that estrogen has inhibitory effect on ET-1 level, which points to putative differences of ET-1 by sex. As stroke is the fifth leading cause of death in the U.S., killing approximately 133,000 people per year, more among African Americans (AA), we aimed to identify the sex differences of the association between ET-1 and incident stroke among AA. Methods: Our AA study participants (N= 4995), with ages 21-90 years and no prevalent stroke, were enrolled in the Jackson Heart Study in Jackson, Mississippi. Stroke events were ascertained through 2011 by reviewing medical records for hospitalizations of JHS participants. Plasma ET levels were measured at the baseline study visit (2000-2004) and expressed in picograms/ml by QuantiGlo Human Endothelin-1 Immunoassay (R&D Systems Inc). Hazard ratios (HR) and 95% confidence intervals (95% CI) were estimated using Cox proportional hazard regression models. Several interactions were assessed such as by sex, smoking and hypertension. Results: Baseline plasma ET-1 level (mean=1.35, median= 1.30) was significantly lower in women (mean =1.32) than in men (mean =1.40; P-value <0.0001). Sex was not an effect modifier of the association between ET-1 and time to stroke event (p for interaction =0.18). After adjusted for age, ET-1 was associated with reduced risk of stroke [HR (95% CI) for all gender: 0.61 (0.40, 0.94), P trend =0.015, among men: 0.32 (0.14, 0.72), P trend =0.003, but not among women: 0.96 (0.54, 1.68), P trend =0.68, for highest vs. lowest quartile, respectively]. After additional adjustment for hypertension and smoking status, the association was still significant among men [0.35 (0.15, 0.79), P trend =0.007, for highest vs. lowest quartile, respectively], but not among women. No statistically significant associations were found among both men and women after additional adjustment for BMI, cigarette smoking, anti-hypercholesterolemic medication status, diabetes and hypertension status, cholesterol levels and homocysteine. No interactions by smoking or hypertension were detected. Conclusion: Our study indicates a non-statistical significant lower risk of stroke with higher levels of endothelin-1, with variation by sex. Our findings warrant confirmation in other epidemiologic studies.