Severity of nonalcoholic fatty liver disease on sonography and risk of coronary heart disease

PURPOSE: To evaluate the relationship between nonalcoholic fatty liver disease (NAFLD) on sonography (US) and the risk of coronary heart disease (CHD) as well as the predictive value of US-diagnosed NAFLD to determine intermediate/high CHD risk compared with as well as in combination on with NAFLD fibrosis score (NFS). METHODS: This retrospective study included 1,276 patients examined between November 2015 and August 2016. NAFLD was categorized as absent, mild, moderate, or severe based on liver-to-kidney echogenicity, visibility of intrahepatic vessel walls, and the diaphragm. The Framingham risk score (FRS) and NFS were used to predict CHD risk and hepatic fibrosis severity. Spearman correlation test, multivariate-adjusted logistic regression analysis, and receiver operating characteristic curves were used for statistical evaluation. RESULTS: FRS increased as NAFLD severity increased, and US-determined NAFLD severity and FRS were highly positively correlated (r = 0.683, p < 0.001). The odds ratios for intermediate/high CHD risk increased with increasing NAFLD severity. The predictive performance of US-determined NAFLD severity for determining intermediate/high CHD risk in NAFLD patients was 0.738. There was no significant difference between US-determined NAFLD severity and NFS in terms of identifying intermediate/high CHD risk (p = 0.88). However, the combination of US-determined NAFLD severity and NFS significantly improved the ability to distinguish intermediate/high CHD risk compared with that of US-determined NAFLD severity or NFS alone (p < 0.001 for both). CONCLUSIONS: US-determined NAFLD severity was well correlated with FRS and associated with the prevalence of intermediate/high CHD risk. The combination of US-determined NAFLD severity and NFS may be useful for predicting CHD risk.