Ultrasound Fatty Liver Indicator: A Simple Tool for Differentiating Steatosis From Nonalcoholic Steatohepatitis: Validity in the Average Obese Population

Objectives: Steatosis, nonalcoholic steatohepatitis (NASH), and fibrosis/cirrhosis represent a spectrum of fatty liver disease. The ultrasound fatty liver indicator (US-FLI) evaluates ultrasound (US) features to identify stages of fatty liver disease. We hypothesized that US features could be independent predictors of NASH and that the US-FLI differentiates steatosis from NASH in the average obese population. Methods: A retrospective analysis of 208 patients with normal (n = 14), steatotic (n = 89), and NASH (n = 105) livers was performed. Liver/biliary disease and a history of alcohol intake were excluded. Ultrasound metrics included liver-kidney contrast, posterior attenuation, vessel blurring, difficulty visualizing the gallbladder wall, difficulty visualizing the diaphragm, and areas of focal fatty sparing. A statistical comparison of the 3 groups as well as fibrosis stage I and II/III NASH groups was performed. Logistic regression identified independent predictors of NASH. Results: Gallbladder wall visualization and vessel blurring were different between the steatosis and NASH groups (P ≤.01). Gallbladder wall visualization was specific for NASH (89%), and vessel blurring was sensitive for NASH (93%). A US-FLI score of 4 or lower suggested the absence of NASH (negative predictive value, 88%; sensitivity, 91%). Logistic regression revealed vessel blurring as the only US predictor of NASH (P ≤.01). However, the area under the curve (0.649) showed poor performance in differentiating steatosis from NASH when the US-FLI score was 5 or higher. Conclusions: Our data suggest that the US-FLI may differentiate steatosis from NASH in the average obese population. Vessel blurring and poor gallbladder wall visualization were the most important metrics. Identification of NASH was enhanced by including the US-FLI score with vessel blurring.