Abdominal obesity phenotypes are associated with the risk of developing non-alcoholic fatty liver disease: insights from the general population

BACKGROUND: The diversity of obesity-related metabolic characteristics generates different obesity phenotypes and corresponding metabolic diseases. This study aims to explore the correlation of different abdominal obesity phenotypes with non-alcoholic fatty liver disease (NAFLD). METHODS: The current study included 14,251 subjects, 7411 males and 6840 females. Abdominal obesity was defined as waist circumference ≥ 85 cm in males and ≥ 80 cm in females; according to the diagnostic criteria for metabolic syndrome recommended by the National Cholesterol Education Program Adult Treatment Panel III, having more than one metabolic abnormality (except waist circumference criteria) was defined as metabolically unhealthy. All subjects were divided into 4 abdominal obesity phenotypes based on the presence ( +) or absence (− ) of metabolically healthy/unhealthy (MH) and abdominal obesity (AO) at baseline: metabolically healthy + non-abdominal obesity (MH(−)AO(−)); metabolically healthy + abdominal obesity (MH(−)AO(+)); metabolically unhealthy + non-abdominal obesity (MH(+)AO(−)); metabolically unhealthy + abdominal obesity (MH(+)AO(+)). The relationship between each phenotype and NAFLD was analyzed using multivariate logistic regression. RESULTS: A total of 2507 (17.59%) subjects in this study were diagnosed with NAFLD. The prevalence rates of NAFLD in female subjects with MH(−)AO(−), MH(−)AO(+), MH(+)AO(−), and MH(+)AO(+) phenotypes were 1.73%, 24.42%, 7.60%, and 59.35%, respectively. Among male subjects with MH(−)AO(−), MH(−)AO(+), MH(+)AO(−), and MH(+)AO(+) phenotypes, the prevalence rates were 9.93%, 50.54%, 25.49%, and 73.22%, respectively. After fully adjusting for confounding factors, with the MH(−)AO(−) phenotype as the reference phenotype, male MH(−)AO(+) and MH(+)AO(+) phenotypes increased the risk of NAFLD by 42% and 47%, respectively (MH(−)AO(+): OR 1.42, 95%CI 1.13,1.78; MH(+)AO(+): OR 1.47, 95%CI 1.08,2.01); the corresponding risks of MH(−)AO(+) and MH(+)AO(+) in females increased by 113% and 134%, respectively (MH(−)AO(+): OR 2.13, 95%CI 1.47,3.09; MH(+)AO(+): OR 2.34, 95%CI 1.32,4.17); by contrast, there was no significant increase in the risk of NAFLD in the MH(+)AO(−) phenotype in both sexes. CONCLUSIONS: This first report on the relationship of abdominal obesity phenotypes with NAFLD showed that both MH(−)AO(+) and MH(+)AO(+) phenotypes were associated with a higher risk of NAFLD, especially in the female population. These data provided a new reference for the screening and prevention of NAFLD. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1186/s12876-022-02393-9.