177 - Combined PVE/HVE versus PVE: Increased Biliary Adverse Events and Surgical Dropout Unique to Perihilar Cholangiocarcinoma

Monday, April 13, 2026

4:12 PM - 4:21 PM ET

Presenting Author(s)

Aakash N. Gupta, MD

Resident Physician
Stanford University, United States

Author/Co-author(s)

GP

George A. Poultsides, MD, MS

Professor of Surgery (Surgical Oncology)
Stanford University, United States
BV

Brendan C. Visser, MD

Professor of Surgery (Hepatobiliary and Pancreatic Surgery)
Stanford University, United States
Lynne Martin, MD

Resident
Stanford, United States
AP

Andrew Picel, MD

Clinical Associate Professor
Stanford Health Care, United States
David S. Wang, MD, FSIR

Clinical Associate Professor
Interventional Radiology, Stanford University, United States
JL

John D. Louie, MD

Professor of Radiology
Stanford University, United States
Daniel Y. Sze, MD, PhD

Professor of Radiology
Stanford University, United States

Purpose: To compare surgical resection rate of primary and metastatic liver tumors following combined portal and hepatic vein embolization (PVE/HVE) versus portal vein embolization (PVE) alone. To assess biliary adverse events (AEs) and surgical dropout after embolization in perihilar cholangiocarcinoma (pCCA).

Materials and Methods: A single-center retrospective study was performed in 174 patients undergoing PVE/HVE (N=49) or PVE alone (N=125) between 2008 to 2025 for colorectal liver metastases (CRLM, N=58), other liver metastases (OLM, N=30), hepatocellular carcinoma (HCC, N=33), pCCA (N=30), intrahepatic cholangiocarcinoma (iCCA, N=20), and benign disease (N=3). PVE was performed using nBCA; HVE was performed using plugs with or without nBCA. Completion of surgical resection was recorded. For pCCA, baseline characteristics were comparable in PVE/HVE (N=12) and PVE (N=18) subgroups: presence of endoscopic biliary drain (92% vs. 78%, p=0.62), percutaneous biliary drain (42% vs. 61%, p=0.46), prior cholangitis (58% vs. 39%, p=0.46), total bilirubin (1.8 vs. 3.8 mg/dL, p=0.30), and cancer antigen 19-9 (148 vs. 191 U/mL, p=0.55). Biliary AEs within 2 months after embolization were recorded. Univariate and multivariate logistic regressions were performed to identify factors affecting resection. Categorical and continuous variables were compared by Fisher’s exact and Wilcoxon rank-sum tests.

Results: For pCCA, surgical resection was lower following PVE/HVE compared to PVE (2/12 [17%] vs. 11/18 [61%], p=0.019). For all other malignancies, resection was comparable after PVE/HVE versus PVE: CRLM (15/17 [88%] vs. 31/41 [76%], p=0.48), OLM (8/8 [100%] vs. 16/22 [73%], p=0.16), HCC (3/4 [75%] vs. 20/28 [71%], p=0.58), and iCCA (4/6 [67%] vs. 12/14 [86%], p=0.34). For pCCA, choice of PVE/HVE or PVE was the sole factor associated with resection in multivariate analysis (OR=8.0, 95% CI: 1.1-58.5, p=0.041). Incidence of biliary AEs was similar in PVE/HVE and PVE (8/12 [67%] vs. 7/18 [39%], p=0.26). However, severity of AEs after PVE/HVE included cholangitis (N=3, 25%), cholangitis with intrahepatic abscesses (N=4, 33%), and death from biliary sepsis (N=1, 13%). Only cholangitis (N=7, 38%) occurred after PVE. Zero of 8 patients (0%) with biliary AEs after PVE/HVE and 4 of 7 patients (57%) with biliary AEs after PVE completed hepatectomy (p=0.026).

Conclusion: Surgical resection of pCCA was lower after PVE/HVE than PVE alone but similar in all other malignancies. PVE/HVE should be performed with caution in pCCA as patients may be more prone to severe biliary AEs which jeopardize surgical candidacy.