Steatosis Grading Comparison Between Qualitative Ultrasonography and Magnetic Resonance Spectroscopy in Patients With Nonalcoholic Fatty Liver Disease

Abstract

Objective: Assess the correlation between the qualitative sonographic score for detecting hepatic steatosis (HS) and liver fat quantification, using proton magnetic resonance spectroscopy (MRS). Materials and Methods: Sixty-six patients with known or suspected nonalcoholic fatty liver disease (NAFLD) underwent ultrasonography (US) and magnetic resonance imaging (MRI). The qualitative sonographic score and fat quantification, measured by MRS, were the techniques used. A Kappa coefficient was used for agreement calculation, and a Fisher test was used to assess the normality of the variables. The MRS results were the gold standard for US quality assessment. Results: The agreement between MRS and US was 50% (Kappa 0.35). Ultrasonography results were more severe in 42.4% of the cases, and 66.7% of the patients had a body mass index greater than 30 ( P = .017). For diagnosis of HS, US presented 100% sensitivity, 30.8% to 60% specificity, 61.7% to 72.7% accuracy, 40% to 69.2% false-positive rate, 0% false-negative rate, 53.8% positive predictive value, and 100% negative predictive value. Conclusion: Ultrasonography is a reliable exam for detecting HS, although not for grading purposes. Therefore, US is not a good predictor of HS severity for the management and follow-up of NAFLD.