Nonalcoholic fatty liver disease: Quantitative assessment of liver fat content by computed tomography, magnetic resonance imaging and proton magnetic resonance spectroscopy

Abstract

The purpose of this study was to evaluate the clinical application of imaging technology in the quantitative assessment of fatty liver with magnetic resonance imaging (MRI) and proton MR spectroscopy. Overall 36 patients with diffuse fatty liver who had undertaken the computed tomography (CT) scan, MRI and proton MR spectroscopy (1H MRS) were analyzed. Their body mass index (BMI) was measured and their liver to spleen CT ratio (L/S) calculated on the plain CT scan. MR T1-weighted imaging (T1WI) was obtained with in-phase (IP) and out-of-phase (OP) images. T2-weighted imaging (T2WI) was acquired with or without the fat-suppression technique. The liver fat content (LFC) was quantified as the percentage of relative signal intensity loss on T1WI or T2WI images. The intrahepatic content of lipid (IHCL) was expressed as the percentage of peak value ratio of lipid to water by 1H MRS. The results of BMI measurement, CT L/S ratio, LFC calculated from MR T1WI and T2WI images, as well as IHCL measured by 1H MRS were 27.26 +/- 3.01 kg/m2, 0.88 +/- 0.26, 13.80 +/- 9.92%, 40.67 +/- 16.04% and 30.98 +/- 20.43%, respectively. The LFC calculated from MR T1WI, T2WI images and IHCL measured by 1H MRS correlated significantly with the CT L/S ratio (r = -0.830, P = 0.000; r = -0.736, P = 0.000; r = -0.461, P = 0.005, respectively). BMI correlated significantly only with the liver fat contents measured by T1WI IP/OP and 1H MRS (r = -0.347, P = 0.038; r = -0.374, P = 0.025, respectively). CT, MR imaging and 1H MRS were effective methods for the quantitative assessment of LFC. The MR imaging, especially 1H MRS, would be used more frequently in the clinical evaluation of fatty liver and (1)H MRS could more accurately reflect the severity of fatty liver.