B-PO02-004 TOTAL AND BEVERAGE-SPECIFIC ALCOHOL CONSUMPTION AND INCIDENT ATRIAL FIBRILLATION

Abstract

Although binge drinking and moderate-to-high alcohol consumption are risk factors for developing atrial fibrillation (AF), the impact of low alcohol consumption remains controversial. Using prospective cohort data from the UK Biobank, we characterised the associations of alcohol consumption with incident AF. Self-reported alcohol consumption from a baseline questionnaire was calculated as standard drinks (8g alcohol) per week. Non-drinkers and those with a history of AF were excluded. Incident AF was assessed through hospitalisation and death records. Consumption was modelled as restricted cubic splines in multivariate Cox regression models and corrected for regression dilution bias. The cohort consisted of 381,236 participants with 11,538 incident AF events over a median (IQR) follow-up duration of 7.9 (7.3-8.6) years. Participants consuming greater amounts of alcohol were younger, more likely male, of white ethnicity, and largely more comorbid. Associations for total and beverage-specific alcohol consumption with risk of AF are shown in the Figure. When included in the analysis, past drinkers but not never drinkers appeared to have a greater risk of AF compared to current drinkers (HR 1.15, 95% CI 1.05-1.26 and HR 1.03, 95% CI 0.94-1.13 respectively). Spirit consumption in females was associated with greater AF risk, though no other significant gender interactions were observed. Amongst current drinkers, low alcohol consumption up to 7 standard drinks/week was not associated with increased AF risk. Red and white wine intake was associated with lower AF risk in comparison to beer, cider and spirits.