Yong-Sheng Xu1,2, Hai-Feng Liu1,2, Da-Li Xi2,3, Jin-Kui Li1, Zhao Liu1, Rui-Feng Yan1, Jun-Qiang Lei4. 1. Department of Radiology, First Hospital of LanZhou University, No. 1, Donggang West Road, Chengguan District, Lanzhou, Gansu, 730000, People's Republic of China. 2. First Clinical Medical College of LanZhou University, Lanzhou, Gansu, People's Republic of China. 3. Department of Pathology, First Hospital of LanZhou University, Lanzhou, Gansu, People's Republic of China. 4. Department of Radiology, First Hospital of LanZhou University, No. 1, Donggang West Road, Chengguan District, Lanzhou, Gansu, 730000, People's Republic of China. leijq2011@126.com.
Abstract
PURPOSE: The study evaluated the relationship between the histological grade of hepatocellular carcinoma (HCC) and the histogram-derived parameters of apparent diffusion coefficient (ADC) obtained from the whole-lesion assessment of diffusion-weighted magnetic resonance (MR) imaging in the liver. METHODS: A total of 51 patients were included. The parameters were correlated with the Edmondson-Steiner grades by using the Spearman correlation coefficient (ρ). The differences of ADC parameters between different tumor histological grades were compared using the Mann-Whitney U test. The extent to which each parameter aided in differentiating tumors with poor performance (III, IV) and fair performance (I, II) was assessed by using the area under the receiver operating characteristic curve (Az). RESULTS: The 25th percentile ADC exhibits the most negative correlation with histological grade (ρ = - 0.397), followed by the 30th percentile ADC (ρ = - 0.395), the minimum ADC value (ρ = - 0.390) and the 20th percentile ADC (ρ = - 0.385), whereas the minimum ADC value yielded the highest Az (0.763) in the discrimination of tumor foci with poor differentiation from fairly differentiated HCCs. The minimum ADC of 4.15 × 10-3 mm2/s or lower was considered to indicate poorly differentiated performance, and the corresponding sensitivity and specificity were 66.7 and 90.9%, respectively. CONCLUSION: The 25th percentile ADC showed a stronger correlation with the histological grade of HCC than other ADC parameters, and the minimum ADC value might be an optimal metric for determining poor and fair differentiations of HCC in DWI.
PURPOSE: The study evaluated the relationship between the histological grade of hepatocellular carcinoma (HCC) and the histogram-derived parameters of apparent diffusion coefficient (ADC) obtained from the whole-lesion assessment of diffusion-weighted magnetic resonance (MR) imaging in the liver. METHODS: A total of 51 patients were included. The parameters were correlated with the Edmondson-Steiner grades by using the Spearman correlation coefficient (ρ). The differences of ADC parameters between different tumor histological grades were compared using the Mann-Whitney U test. The extent to which each parameter aided in differentiating tumors with poor performance (III, IV) and fair performance (I, II) was assessed by using the area under the receiver operating characteristic curve (Az). RESULTS: The 25th percentile ADC exhibits the most negative correlation with histological grade (ρ = - 0.397), followed by the 30th percentile ADC (ρ = - 0.395), the minimum ADC value (ρ = - 0.390) and the 20th percentile ADC (ρ = - 0.385), whereas the minimum ADC value yielded the highest Az (0.763) in the discrimination of tumor foci with poor differentiation from fairly differentiated HCCs. The minimum ADC of 4.15 × 10-3 mm2/s or lower was considered to indicate poorly differentiated performance, and the corresponding sensitivity and specificity were 66.7 and 90.9%, respectively. CONCLUSION: The 25th percentile ADC showed a stronger correlation with the histological grade of HCC than other ADC parameters, and the minimum ADC value might be an optimal metric for determining poor and fair differentiations of HCC in DWI.