HUANG Yao 1,2,3 , YIN Peiyuan 4 , YIN Longlin 1,2,3,4,5 , LI Xiaoyan 2,3,5 , SUN Ju 2,3
  • 1. School of Medicine, University of Electronic Science and Technology of China, Chengdu 610056, P. R. China;
  • 2. Department of Radiology, Sichuan Academy of Medical Sciences & Sichuan Provincial People’s Hospital, Chengdu 610072, P. R. China;
  • 3. Institute of Radiological Medicine, Sichuan Academy of Medical Sciences & Sichuan Provincial People’s Hospital, Chengdu 610072, P. R. China;
  • 4. Clinic Medical College, Southwest Medical University, Luzhou, Sichuan 646099, P. R. China;
  • 5. School of Medical Imaging, North Sichuan Medical College, Nanchong, Sichuan 637000, P. R. China;
YIN Longlin, Email: yinlonglin@163.com
Export PDF Favorites Scan Get Citation

Objective To explore the value of magnetic resonance diffusion weighted imaging (DWI) in preoperative Bismuth-Corlette classification of hilar cholangiocarcinoma (HCCA). Methods A total of 53 HCCA patients confirmed by postoperative pathology were retrospectively included. The accuracy of two sequence combinations, namely dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) + magnetic resonance cholangiopancreatography (MRCP) and DCE-MRI + MRCP + DWI, in evaluating the longitudinally involved bile duct segments and Bismuth-Corlette classification of HCCA was compared. Additionally, the correlation between apparent diffusion coefficient (ADC) values and tumor Bismuth-Corlette classification as well as degree of differentiation was analyzed. Results There were 318 bile duct segments in 53 HCCA patients. The accuracy rate of DCE-MRI + MRCP was 93.7% (298/318), the sensitivity was 91.5% (161/176), and the specificity was 96.5% (137/142); The accuracy rate of DCE-MRI + MRCP + DWI was 96.5% (307/318), the sensitivity was 96.0% (169/176), and the specificity was 97.2% (138/142). The accuracy of DCE-MRI + MRCP + DWI was higher than that of DCE-MRI + MRCP, with a statistically significant difference (χ2=275.902, P<0.001). Receiver operating characteristic (ROC) curve analysis showed that the area under the ROC curve (AUC) of DCE-MRI + MRCP + DWI was 0.966 [95%CI (0.940, 0.983), P<0.000 1], and its diagnostic efficacy was superior to that of DCE-MRI + MRCP [AUC=0.940, 95%CI (0.908, 0.963), P<0.000 1]. The DeLong test indicated a statistically significant difference in AUC between the two sequences (Z=2.633, P=0.008 5). The accuracy rates of preoperative Bismuth-Corlette classification of HCCA evaluated by DCE-MRI + MRCP and DCE-MRI + MRCP + DWI were 86.8% (46/53) and 94.3% (50/53), respectively. After adding the DWI sequence, the consistency between Bismuth-Corlette classification results and surgical pathological classification results (Kappa=0.922, P<0.001) was higher than that of DCE-MRI + MRCP sequence (Kappa=0.820, P<0.001), with a statistically significant difference (χ2=160.370, P<0.001). In addition, the ADC value of HCCA was negatively correlated with tumordegree of differentiation (rs=–0.524, P<0.001), but had no significant correlation with its Bismuth-Corlette classification (rs=–0.058, P=0.682). Conclusions DCE-MRI + MRCP + DWI sequence can effectively improve the accuracy in preoperative evaluation of the involvement of bile duct segments and Bismuth-Corlette classification of HCCA, which provides guidance for precise preoperative surgical planning in clinical practice. In addition, the ADC value can provide additional information required for non-invasive preoperative prediction of the prognosis of HCCA patients.

Copyright © the editorial department of CHINESE JOURNAL OF BASES AND CLINICS IN GENERAL SURGERY of West China Medical Publisher. All rights reserved