Mendelian randomisation suggests no beneficial effect of moderate alcohol consumption on the severity of nonalcoholic fatty liver disease

BackgroundPrevious epidemiological studies suggest that patients diagnosed with nonalcoholic fatty liver disease (NAFLD) who drink light to moderate amounts of alcohol (up to similar to 30 g per day) have less severe histological lesions compared with nondrinkers. However, while the cross-sectional nature of current evidence precludes assessment of causality, cumulative lifetime-exposure of moderate alcohol consumption on histological outcomes has never been evaluated. AimTo overcome these limitations, a Mendelian randomisation study was performed using a validated genetic variant (rs1229984 A;G) in the alcohol dehydrogenase (ADH1B) gene as a proxy of long-term alcohol exposure. MethodsWe first assessed whether the instrumental variant (rs1229984) was associated with the amount of alcohol consumption in our cohort. We further explored the association between the variant and histological outcomes; a sample of 466 individuals, including 266 patients with NAFLD confirmed by liver biopsy, was studied. ResultsWe found that carriers of the A-allele consumed significantly lower amounts of alcohol compared with noncarriers (2.3 5.3 vs. 8.18 +/- 21 g per day, mean +/- s.d., P = 0.03). The analysis of association with the disease severity showed that carriers of the A-allele had lower degree of histological steatosis (1.76 +/- 0.83 vs. 2.19 +/- 0.78, P = 0.03) and lower scores of lobular inflammation (0.54 +/- 0.65 vs. 0.95 +/- 0.92, P = 0.02) and NAFLD-Activity Score (2.9 +/- 1.4 vs. 3.7 +/- 1.4, P = 0.015) compared with noncarriers. ConclusionMendelian randomisation analysis suggests no beneficial effect of moderate alcohol consumption on NAFLD disease severity.