BMI–hepatic steatosis phenotypes and prevalent carotid plaque: a cross-sectional health examination study

Abstract Hepatic steatosis is frequently detected during health examinations, but its vascular relevance among individuals with normal BMI remains insufficiently characterized. This study examined the cross-sectional association of ultrasound-detected hepatic steatosis and BMI–hepatic steatosis phenotypes with prevalent carotid plaque, and further explored metabolic heterogeneity among normal-BMI individuals with hepatic steatosis. This cross-sectional retrospective study included 95,157 eligible health examination records from 68,761 adults. Ultrasound-detected hepatic steatosis was used as an operational imaging-based exposure and was not considered equivalent to formal metabolic dysfunction-associated steatotic liver disease (MASLD). Examination records were classified into four BMI–hepatic steatosis phenotypes: normal BMI without hepatic steatosis, normal BMI with hepatic steatosis, BMI ≥ 24 kg/m² without hepatic steatosis, and BMI ≥ 24 kg/m² with hepatic steatosis. Multivariable logistic regression with participant-clustered robust variance estimates was used to estimate adjusted odds ratios and 95% confidence intervals for prevalent carotid plaque. Among normal-BMI examination records with hepatic steatosis, associations of non-high-density lipoprotein cholesterol (non-HDL-C), triglyceride-glucose index, uric acid-to-high-density lipoprotein cholesterol (HDL-C) ratio, and atherogenic index of plasma with prevalent carotid plaque were further evaluated. Ultrasound-detected hepatic steatosis was associated with higher odds of prevalent carotid plaque after multivariable adjustment (OR 1.167, 95% CI 1.111–1.226). Compared with normal-BMI examination records without hepatic steatosis, those with hepatic steatosis had higher odds of prevalent carotid plaque (OR 1.229, 95% CI 1.147–1.317), as did examination records with BMI ≥ 24 kg/m² and hepatic steatosis (OR 1.151, 95% CI 1.087–1.219). In the primary logistic regression model, BMI ≥ 24 kg/m² without hepatic steatosis was not significantly associated with prevalent carotid plaque. Among normal-BMI examination records with hepatic steatosis, marker-specific linear analyses identified associations of non-HDL-C, triglyceride-glucose index, uric acid-to-HDL-C ratio, and atherogenic index of plasma with prevalent carotid plaque. Sensitivity analyses using a stricter normal-BMI definition, modified Poisson regression, and restriction to the first eligible examination per participant yielded broadly similar association patterns; no statistically significant interaction by sex or age group was observed. Among adults undergoing health examinations, ultrasound-detected hepatic steatosis within the normal-BMI range was associated with higher odds of prevalent carotid plaque, suggesting that normal BMI alone may not exclude plaque-related cardiometabolic burden. Associations of metabolic markers with prevalent carotid plaque further indicated metabolic heterogeneity within this phenotype. Prospective studies are needed to determine whether these markers improve prediction of plaque progression and cardiovascular events.