Abstract
OBJECTIVE: To investigate the association between adipose distribution and hepatic steatosis in American adults and to assess whether this association varies among different blood glucose states. METHODS: Data from the American National Health and Nutrition Examination Survey (NHANES) 2017-2018 were analyzed. The subjects were divided into three groups: diabetes, prediabetes and normal glucose tolerance (NGT). Hepatic steatosis was quantified by median controlled attenuation parameter (CAP), which was measured by ultrasound transient elastography. Total abdominal fat volume, visceral adipose tissue (VAT) volume, total percent fat, trunk percent fat, android percent fat and android to gynoid ratio (AGR) was measured by dual-energy X-ray absorptiometry (DXA). RESULTS: Data pertaining to 2986 participants (1581 with hepatic steatosis) were included in the analysis. In the NGT group, the proportion of S0 (<5% of the hepatocytes with fatty infiltration) was 58.9%, and 25.2% for S3 (≥66% of the hepatocytes with fatty infiltration). In contrast, the proportion of S0 was 11.1%, while S3 accounts for as high as 68.7% in the diabetes group. In the NGT group, all parameters of fat distribution revealed a positive relation with the occurrence of hepatic steatosis (p<0.05) except total percent fat (p=0.872) after adjusting for confounding factors. In the prediabetes group, VAT volume, trunk percent fat, android percent fat and AGR had significant influence on hepatic steatosis (p<0.05). As for diabetes, only AGR remained significantly correlated with hepatic steatosis (p=0.004). CONCLUSION: For NGT individuals, high level of total abdominal fat volume, VAT volume, trunk percent fat, android percent fat and AGR all can be used to predict hepatic steatosis. For diabetes, only AGR can predict hepatic steatosis among the surveyed parameters of adipose distribution.