Body Fat Distribution and Risk of Incident and Regressed Nonalcoholic Fatty Liver Disease

BACKGROUND & AIMS: Some studies have examined correlations between visceral adipose tissue (VAT) and subcutaneous adipose tissue (SAT) with nonalcoholic fatty liver disease (NAFLD) or between VAT and NAFLD. We investigated the longitudinal association between body fat distribution (VAT vs SAT) and incidence and regression of NAFLD, adjusting for risk factors, in a large population-based cohort. METHODS: We collected data from adults who underwent abdominal ultrasonography (to identify liver fat), abdominal fat computed tomography scan, and blood tests from March 2007 through December 2008. Each patient underwent an anthropometric assessment and completed a questionnaire about their medical history, physical activity, and diet. Our final analysis involved 2017 subjects from the initial cohort who participated in a voluntary follow-up health screen performed in 2011 and 2013. The median follow-up time was 4.43 years. RESULTS: We found 288 incident cases of NAFLD; 159 patients had NAFLD regression during the follow-up period. An increasing area of VAT was associated with higher incidence of NAFLD in the multivariable analysis (highest quintile vs lowest quintile of VAT hazard ratio [HR], 2.23; 95% confidence interval [CI], 1.28-3.89; P for trend = .002; HR, 1.36 [per 1 standard deviation]; 95% CI, 1.16-1.59). An increased area of SAT was significantly associated with regression of NAFLD (highest quintile vs lowest quintile of SATHR, 2.30; 95% CI, 1.28- 4.12; P for trend = .002; HR, 1.36 [per 1 standard deviation]; 95% CI, 1.08-1.72). CONCLUSIONS: In a large cohort study, larger areas of VAT were longitudinally associated with higher risk of incident NAFLD (during a period of approximately 4 years). In contrast, larger areas of SAT were longitudinally associated with regression of NAFLD. These data indicate that certain types of body fat are risk factors for NAFLD, whereas other types could reduce risk for NAFLD.