Background: Alcohol is considered to be a major cause of fatty liver (FL). In contrast, however, recent investigations have suggested that moderate alcohol consumption is protective against FL. To clarify the role of alcohol consumption in FL development, we examined the association between drinking patterns and FL prevalence.
Methods: We enrolled 9,886 male participants at regular medical health checks. Each subject's history of alcohol consumption was determined by questionnaire. The subjects were classified according to alcohol consumption as non-, light, moderate, and heavy drinkers (0, <20, 20-59, and ≥60 g/day, respectively). FL was defined by ultrasonography. Independent predictors of FL were determined by logistic regression analysis.
Results: The prevalence of FL displayed a "U-shaped curve" across the categories of daily alcohol consumption (non-, 44.7%; light, 39.3%; moderate, 35.9%; heavy drinkers, 40.1%; P < 0.001). The prevalence of FL was associated positively with body mass index and other obesity-related diseases and inversely with alcohol consumption (light, odds ratio [OR] 0.71, 95% confidence interval [CI] 0.59-0.86; moderate, OR 0.55, CI 0.45-0.67; heavy, OR 0.44, CI 0.32-0.62) as determined by multivariate analysis after adjusting for potential confounding variables. In addition, examination of drinking patterns (frequency and volume) revealed that the prevalence of FL was inversely associated with the frequency of alcohol consumption (≥21 days/month) (OR 0.62, CI 0.53-0.71) but not with the volume of alcohol consumed.
Conclusions: Our observations suggest that alcohol consumption plays a protective role against FL in men, and consistent alcohol consumption may contribute to this favorable effect.