Related Experiment Video
Updated: Aug 13, 2026
Portal Vein Injection of Colorectal Cancer Organoids to Study the Liver Metastasis Stroma
Published on: September 3, 2021
Predictors of unresectability after portal vein embolization for centrally located cholangiocarcinoma
Ahmed Abdelrafee1, Julio Nunez2, Luis Cano2
1Department of Surgery, Paul-Brousse Hospital, Assistance Publique Hôpitaux de Paris, Centre Hépato-Biliaire, Villejuif, France; Gastrointestinal Surgery Center, Department of General Surgery, Faculty of Medicine, Mansoura University, Mansoura, Egypt.
Background:
The curative treatment of perihilar cholangiocarcinomas and centrally located intrahepatic cholangiocarcinomas often requires major hepatectomy preceded by portal vein embolization. This strategy, however, is associated with a high rate of dropouts before operation or failure of resection at the time of operative exploration. We aimed to identify predictors of unresectability (dropout or failure of resection) after portal vein embolization for centrally located cholangiocarcinoma, including perihilar cholangiocarcinomas and intrahepatic cholangiocarcinomas.
Method:
All patients undergoing portal vein embolization for a planned resection of a centrally located cholangiocarcinoma between 2000 and 2018 in our center were evaluated retrospectively. Predictors of unresectability were determined under intention-to-treat conditions, based on clinical, biologic, and radiologic data collected before portal vein embolization.
Results:
Eighty-eight consecutive patients scheduled for portal vein embolization before operative exploration were included, 56 of whom (64%) underwent curative resection and 32 (36%) of whom were not resected, including those who did not undergo exploration (n = 11) and those operated on but not resected (n = 21). The most common cause of unresectability was tumor progression (62%). A psoas muscle index <500 mm2/m2 (P = .04), high body mass index (P = .023), and low serum albumin level (P = .007) were associated with unresectabilty on multivariate analysis. A composite score including these variables (cutoffs determined after receiver operating characteristic curve analysis) was proposed and achieved accurate discrimination regarding unresectability (area under the curve = 0.82, P < .001).
Conclusion:
Predictors of unresectability after portal vein embolization for centrally located cholangiocarcinoma were identified, with sarcopenic overweight patients having a greater risk of unresectability. This preoperative score enables a fairly accurate prediction of unresectability in a given patient. These simple, objective, and inexpensive parameters should be considered in all patients with centrally located cholangiocarcinoma scheduled to undergo portal vein embolization.
