Abstract P368: Systemic Inflammation Does Not Mediate the Inverse Association between Usual Alcohol Consumption and a 39-Year Death Risk from Incident Coronary Heart Disease Independent of Genes and Common Environment among Men

Abstract

INTRODUCTION: Alcohol consumption is associated with coronary heart disease (CHD), and systemic inflammation also contributes to CHD risk. It is unknown to what extent systemic inflammation plays a role in the association between alcohol consumption and CHD. We examined the hypothesis that systemic inflammation mediated the relationship between usual alcohol consumption and long-term CHD death risk in a temporal order: alcohol, inflammation, and death, independent of genes and common environment. METHODS: We included 535 white, male, middle-aged (mean 48 years; range 42-55), veteran twins (106 monozygotic and 97 dizygotic twin pairs; 60 monozygotic and 69 dizygotic unpaired twins), who were free of CHD at baseline (1969-1973) and had plasma collected at least 8 years after the baseline examination in the follow-up exams from the National Heart, Lung, and Blood Institute Twin Study. Usual alcohol consumption data were collected at baseline through a dietary history interview. Inflammatory biomarkers, interleukin-6 and high-sensitivity C reactive protein, were measured by ELISA. Vital data and death causes were collected through death certificates until Dec 31, 2009. A frailty survival model, accounting for clustering within co-twins, was used to partition the overall association into within-pair and between-pair associations. We controlled for total caloric intake and parsimonious known CHD risk factors, including body mass index and modified Framingham Risk Score. RESULTS: There were 54 incident CHD deaths during a 39-year follow-up (median follow-up of 34 years). By treating twins as individuals, after controlling for caloric intake and known CHD risk factors, the overall association showed that daily alcohol consumption was negatively associated with CHD death risk [partial coefficient β (95% confidence interval (CI)): -0.063 (-0.14, 0.02), converted to hazard ratio (HR) (95% CI) as 0.94 (0.77, 1.02)]. Interaction between the within-pair effect and zygosity was not significant ( P =0.58). Using the pooled data by zygosity, in within-pair comparison between co-twins that additionally controlled for genes and common environment, the association was dramatically strengthened by 220% [β (95% CI): -0.14 (-0.26, -0.02); HR (95% CI): 0.87 (0.77, 0.98)]. Further adjustment for inflammatory biomarkers did not alter the within-pair association magnitude: one-serving increment in daily alcohol consumption was associated with a 13% (95% CI: 2%, 23%) lower 39-year CHD death risk after controlling for genes, common environment, caloric intake, and parsimonious known CHD risk factors. CONCLUSIONS: Systemic inflammation does not mediate alcohol consumption-related risk of death from incident coronary heart disease independent of genes and common environment. Funding(This research has received full or partial funding support from the American Heart Association, National Center)