Abstract: Objective The aim of this study was to explore the diagnostic efficacy of magnetic resonance imaging (MRI) iterative decomposition of water and fat with echo asymmetry and the least squares estimation quantification sequence (IDEAL-IQ) in the evaluation of hepatic steatosis in patients with non-alcoholic fatty liver diseases(NAFLD). Methods A total of 120 patients with NAFLD and 65 healthy individuals were enrolled in our hospital between February 2018 and February 2020, and all underwent MRI scan. The magnetic resonance spectroscopy (MRS) of hydrogen proton and IDEAL-IQ quantitative indexes, e.g. relative lipid content (RLC) and fat fraction (FF) were obtained. The patients with NAFLD received liver biopsies. The diagnostic efficacy of IDEAL-IQ for hepatic steatosis was analyzed by ROC. Results Serum ALT, GGT, TC, TG, and the RLC and FF in patients with NAFLD were (54.8±10.6)U/L, (103.1±5.6)U/L, (6.2±0.7)mmol/L, (2.7±0.6)mmol/L, (17.4±9.2)% and (18.0±6.5)%, all significantly higher than [(25.7±11.4)U/L,(59.7±2.7)U/L,(4.6±0.4)mmol/L, (1.6±0.2)mmol/L, (3.5±1.3)% and (6.2±2.1)%, respectively, P＜0.05] in healthy persons; the RLC and FF in 15 patients with severe hepatic steatosis were(32.5±4.8)% and (28.6±4.0)%, both significantly higher than [(23.1±6.2)% and (20.8±5.1)%, P＜0.05] in 39 NAFLD patients with moderate liver steatosis or [(10.6±3.0)% and (13.9±4.2)%, P＜0.05] in 66 patients with mild liver steatosis; the AUC was 0.894, with the sensitivity (Se) of 88.7%, the specificity (Sp) of 70.8% as the RLC＞23.7 % was set as the cut-off-value, and the AUC was 0.870, with the Se of 97.0%, the Sp of 90.0% when the FF＞20.4 % was set as the cut-off-value in diagnosing moderate to severe liver steatosis, the latter superior to the former (Z=3.025,P＜0.05). Conclusion The qualitative and quantitative diagnosis of hepatic steatosis as an non-invasive approach in patients with NAFLD by MRI IDEAL-IQ is promising, and warrants further investigation.

Key words: Non-alcoholic fatty liver disease, Hepatic steatosis, Magnetic resonance imaging, IDEAL-IQ, Diagnosis