[Association between regional fat mass and risk of nonalcoholic fatty liver disease in overweight/obese adults].

OBJECTIVE

To explore the relationship between fat distribution and non-alcoholic fatty liver(NAFLD) in overweight/obese adults.

METHODS

This cross-sectional study included 736(190 men and 546 women) 19-56 years old overweight/obese people in Beijing were selected by convenient sampling. Their age and body mass index(BMI) distribution were 36(31-46) years old and 28.0(26.2-30.7), respectively. The body fat mass and regional fat mass were measured by dual energy X-ray absorptiometry(DXA), and Logistic regression model was used to analyze the association between regional fat mass and the risk of NAFLD.

RESULTS

The prevalence of NAFLD was 70.0%(515/736) in overweight/obese population. In the multivariate Logistic model, after adjusting for age, gender, BMI, hypertension and body fat mass, waist circumference(WC), thigh fat mass and android fat mass were significantly association with NAFLD risk(P<0.05), but no association was found between arms, trunk and gynoid fat mass and NAFLD risk. There were interactions between thigh fat mass and age(P_(interaction)<0.001) and BMI group(P_(interaction)=0.001). Subgroup analysis showed that thigh fat mass and NAFLD risk were significantly associated in ≤36-year-old(OR=0.62, 95%CI 0.48-0.81), male(OR=0.32, 95%CI 0.16-0.64) and overweight(OR=0.48, 95%CI 0.36-0.64) groups, but in the >36-year-old, female and the obesity group this association was not statistically significant. There was an interaction between trunk fat mass and age group(P_(interaction)=0.009). There was a positive correlation between trunk fat mass and NAFLD risk in >36-year-old group(OR=1.63, 95%CI 1.35-1.97), but no association was found in ≤36-year-old group. In addition, we also found that a significant interaction between gynoid fat mass and BMI group on NAFLD(P_(interaction)<0.001). In overweight, gynoid fat mass was negatively correlated with the risk of NAFLD(OR=0.12, 95% CI 0.06-0.25), but in the obesity group, the association was not statistically significant. There were no statistically significant interactions between WC, arms fat mass and android mass and age, sex and BMI groups.

CONCLUSION

WC, android fat mass and thigh fat mass are associated with the risk of NAFLD. Thigh fat mass has a significant interaction with age and BMI group on the risk of NAFLD(only in ≤36-year-old group, male and overweight group a significant protective effect of thigh fat on NAFLD was found, but not in >36-year-old group, female and obesity group). Trunk fat mass had an interaction with age(the association between trunk fat mass and NAFLD was significant in >36-year-old group). Gynoid fat mass and BMI group also have a significant interaction on NAFLD(the detrimental effect of gynoid fat on NAFLD is much more profound in the obesity group).