Association of Plasma Sphingomyelin Levels and Incident Coronary Heart Disease Events in an Adult Population: Multi-Ethnic Study of Atherosclerosis.
Joseph Yeboah,Coleen McNamara,Xian-Cheng Jiang,Ira Tabas,David M. Herrington,Gregory L. Burke,Steven Shea
DOI: https://doi.org/10.1161/atvbaha.109.199281
2010-01-01
Arteriosclerosis Thrombosis and Vascular Biology
Abstract:Objective— A high plasma sphingomyelin level has been associated with subclinical atherosclerosis, coronary artery disease, and worse prognosis in subjects with acute coronary syndromes. We wanted to assess the predictive value of plasma sphingomyelin levels for incident coronary heart disease (CHD) events in the Multi-Ethnic Study of Atherosclerosis. Methods and Results— The plasma sphingomyelin level was measured in 6809 of 6814 subjects (age, mean±SD, 62.2±10.2 years) participating in the Multi-Ethnic Study of Atherosclerosis, a population-based cohort study of adults free of clinical cardiovascular disease at baseline recruited at 6 clinic sites in the United States. The subjects consisted of 52.8% females, 38.5% whites, 11.8% Chinese, 27.8% blacks, and 21.9% Hispanics. Cox proportional hazard analysis was used to examine the association between plasma sphingomyelin level and 5 years of adjudicated incident CHD events, including myocardial infarction, resuscitated cardiac arrest, angina, CHD-related death, and revascularization (coronary artery bypass grafting or percutaneous transluminal coronary angioplasty). The mean±SD plasma sphingomyelin level was 48.0±16.0 mg/dL. A total of 189 subjects had an adjudicated CHD event during the 5 years of follow-up. In the Kaplan-Meier analysis, subjects with a plasma sphingomyelin level higher than the sex-specific median had a similar event-free survival rate compared with subjects with a plasma sphingomyelin level at or less than the sex-specific median (97.16% versus 97.00%; log rank P =0.71). In the univariate Cox proportional hazard analysis, the plasma sphingomyelin level was not a predictor of an incident CHD event (hazard ratio, 0.992 [0.982–1.004]; P =0.09). In our multistage multivariate Cox proportional hazard models, a higher plasma sphingomyelin level had a modest negative association with incident CHD events when total cholesterol, high-density lipoprotein, and triglycerides were included in the model (hazard ratio, 0.985 [0.973–0.996]; P =0.008) and also in our full model after adjusting for age, sex, total cholesterol, high-density lipoprotein, triglycerides, diabetes mellitus, cigarette smoking, systolic and diastolic blood pressure, medication use for blood pressure, and 3-hydroxy-3-methylglutaryl–coenzyme A reductase inhibitor use (hazard ratio, 0.984 [0.973–0.996]; P =0.002). In other models, the plasma sphingomyelin level was not associated with incident CHD events. Conclusion— A high plasma sphingomyelin level is not associated with an increased risk of incident CHD in population-based adults free of clinical cardiovascular disease at baseline.