Efficacy of Preoperative Portal Vein Embolization Among Patients with Hepatocellular Carcinoma, Biliary Tract Cancer, and Colorectal Liver Metastases: A Comparative Study Based on Single-Center Experience of 319 Cases.

BACKGROUND: Efficacy of preoperative portal vein embolization (PVE) has been established; however, differences of outcomes among diseases, including hepatocellular carcinoma (HCC), biliary tract cancer (BTC), and colorectal liver metastases (CLM), are unclear. METHODS: Subjects included patients in a prospectively collected database undergoing PVE (from 1995 to 2013). A future liver remnant (FLR) volume ≥40% is the minimal requirement for patients with an indocyanine green retention rate at 15 min (ICGR15) <10%, and stricter criteria (FLR volume ≥50%) have been applied for patients with 20% > ICGR15 ≥ 10%. Patient characteristics and survivals were compared among those three diseases, and predictors of dropout and better FLR hypertrophy were determined. RESULTS: In 319 consecutive patients undergoing PVE for HCC (n = 70), BTC (n = 172), and CLM (n = 77), the degree of hypertrophy did not significantly differ by cancer types (median 10, 9.6, and 10%, respectively). Eighty percent (256 of 319) of patients completed subsequent hepatectomy after a median waiting interval of 24 days (range 5-90), while dropout after PVE was more common in BTC or CLM (odds ratio 2.75, p = 0.018), mainly because of disease progression. Ninety-day liver-related mortality after hepatectomy was 0% in the entire cohort, and 5-year overall survival of patients with HCC, BTC, and CLM was 56, 50, and 51%, respectively (p = 0.948). No patients who dropped out survived more than 2.5 years after PVE. CONCLUSION: PVE produced equivalent FLR hypertrophy among the three diseases as long as liver function was fulfilling the preset criteria; however, the completion rate of subsequent hepatectomy was highest in HCC. PVE followed by hepatectomy was a safe and feasible strategy for otherwise unresectable disease irrespective of cancer types.