Diffusion-weighted imaging (DWI) of hepatocellular carcinomas: a retrospective analysis of the correlation between qualitative and quantitative DWI and tumour grade

Abstract

To evaluate the application of qualitative and quantitative diffusion-weighted imaging (DWI) in predicting the histological grade of hepatocellular carcinoma (HCC). Two hundred and fifty-four patients with pathologically confirmed HCC who underwent hepatic DWI on a 1.5-T platform (b = 0, 600s/mm2) were evaluated retrospectively. HCCs were divided into well-, moderately, and poorly differentiated groups. The relationships between naked-eye signal intensity (SI), SI values, apparent diffusion coefficient (ADC) values on DWI, and the histopathological differentiation of HCC were analysed. Receiver operating characteristic (ROC) curves were drawn to determine the optimal operating points (OOPs) of the SI and ADC values to predict the tumour grade. A weak negative correlation (r=-0.350, p<0.05) was obtained between naked-eye SI and histological grade. There was a significant difference in mean SI values between well- (68.32±31.71) and moderately (102.39±45.55)/poorly (114.55±32.15) differentiated HCC but not between moderately and poorly differentiated HCC. The OOP of the SI value by ROC curve analysis was 66.5 to predict well-differentiated HCC. The mean ADC values of well-, moderately, and poorly differentiated HCC were 1.67±0.13×10-3, 1.31±0.16×10-3, and 1.08±0.11×10-3mm2/s, respectively, with significant differences between any two combinations of groups. The OOPs of ADC to diagnose well- and poorly differentiated HCC were 1.5×10-3 and 1.24×10-3mm2/s, respectively. Qualitative and quantitative SI and ADC values at DWI may be useful to estimate the histological grade of HCC preoperatively and non-invasively.