Propensity Score-Matched Analysis of Liver Venous Deprivation and Portal Vein Embolization Before Planned Hepatectomy in Patients with Extensive Colorectal Liver Metastases and High-Risk Factors for Inadequate Regeneration

Abstract

Background Liver venous deprivation (LVD) is known to induce better future liver remnant (FLR) hypertrophy than portal vein embolization (PVE). The role of LVD, compared with PVE, in inducing FLR hypertrophy and allowing safe hepatectomy for patients with extensive colorectal liver metastases (CLM) and high-risk factors for inadequate hypertrophy remains unclear. Methods Patients undergoing LVD (n = 22) were matched to patients undergoing PVE (n = 279) in a 1:3 ratio based on propensity scores, prior to planned hepatectomy for CLM at a single center (1998-2023). The propensity scores accounted for high-risk factors for inadequate hypertrophy, namely pre-procedure standardized FLR (sFLR), body mass index, number of systemic therapy cycles, an extension of PVE to segment IV portal vein branches, prior resection, and chemotherapy-associated liver injury. Results The matched cohort included 78 patients (LVD, n = 22; PVE, n = 56). Baseline characteristics were comparable. The number of tumors in the whole liver was similar but more LVD patients had five or more tumors in the left liver (32% vs. 11%; p = 0.024). Post-procedure sFLR was similar but LVD patients had a significantly higher degree of hypertrophy (16% vs. 11%; p = 0.017) and kinetic growth rate (3.9 vs. 2.4% per week; p = 0.006). More LVD patients underwent extended right hepatectomy (93% vs. 55%; p = 0.008). Only one patient had postoperative hepatic insufficiency after PVE, and no patients died within 90 days of hepat…