Performance and limitations of steatosis biomarkers in patients with nonalcoholic fatty liver disease.

Abstract

Several steatosis biomarkers are available with limited independent validation. To determine diagnostic value and limitations of several steatosis biomarkers using liver biopsy as reference standard in a large cohort of patients with suspected NAFLD. Three hundred and twenty-four consecutive liver biopsies were included. Histological steatosis was categorised as none (<5%), mild (5-33%), moderate (33-66%) and severe (>66%). Five steatosis biomarkers were measured: fatty liver index (FLI), NAFLD liver fat score (NAFLD-LFS), hepatic steatosis index (HSI), visceral adiposity index (VAI) and triglyceride × glucose (TyG) index. Steatosis grades prevalence was: none 5%, mild 39%, moderate 30% and severe 27%. Except for VAI, the steatosis biomarkers showed a linear trend across the steatosis grades. However, their correlation with the histological amount of steatosis was only weak-moderate. All steatosis biomarkers had an adequate diagnostic accuracy for the presence of steatosis: AUROCs for FLI, LFS, HSI, VAI and TyG were 0.83, 0.80, 0.81, 0.92 and 0.90. However, their ability to quantify steatosis was poor: none of them distinguished between moderate and severe steatosis and the AUROCs for predicting steatosis >33% were 0.65, 0.72, 0.65, 0.59 and 0.59 for FLI, LFS, HSI, VAI and TyG. Both fibrosis and inflammation significantly confounded the association between steatosis biomarkers and steatosis. The steatosis biomarkers were all correlated with HOMA-IR, independent from histological steatosis. All five steatosis biomarkers can diagnose steatosis and are correlated with insulin resistance. They are confounded by fibrosis and inflammation, and do not accurately quantify steatosis; this may limit their clinical utility. More research is needed to identify truly independent and quantitative markers of steatosis.