BACKGROUND/AIMS: The prognosis ofhepatocellular carcinoma with macroscopic portal vein tumor thrombosis is extremely poor. The risk factors may differ at different postoperative intervals. This study was undertaken to clarify the surgical outcome and time dependency of factors influencing survival in these patients. METHODOLOGY: We analyzed clinicopathological variables of 381 hepatocellular carcinoma patients with macroscopic portal vein tumor thrombosis who underwent hepatic resection. Survival rates were calculated using Kaplan-Meier method. The stratified Cox models were used to identify factors independently influencing short- and long-term survival, respectively. RESULTS: The cumulative 1-, 2-, 3-, 5-, and 10-year survival rates in 381 patients were 47%, 23%, 16%, 12%, 6%, respectively. The 1-, 3-, and 5-year survival rates calculated from time of re-resection were 36%, 14% and 0% in patients undergoing re-resection for intrahepatic recurrence within 2 years after first operation, and 85%, 53% and 32% in those more than 2 years after first operation (P<0.05). Multivariate analysis showed that portal vein infusion chemotherapy, serum alpha-fetoprotein > 20 mg/L and positive surgical margin were significant prognostic factors within 2 years after operation. In contrast, alanine aminotransferase > 80 U/L was the only significant factor beyond 2 years after operation. CONCLUSIONS: The survival of hepatocellular carcinoma patients with macroscopic portal vein tumor thrombosis was poor, but the prognosis of patients who had tumor recurrence more than 2 years after operation was much better than those with tumor recurrence within 2 years. Evaluation of time-dependency of risk factors may have important clinical implication in determining the therapeutic strategy.
BACKGROUND/AIMS: The prognosis ofhepatocellular carcinoma with macroscopic portal vein tumor thrombosis is extremely poor. The risk factors may differ at different postoperative intervals. This study was undertaken to clarify the surgical outcome and time dependency of factors influencing survival in these patients. METHODOLOGY: We analyzed clinicopathological variables of 381 hepatocellular carcinomapatients with macroscopic portal vein tumor thrombosis who underwent hepatic resection. Survival rates were calculated using Kaplan-Meier method. The stratified Cox models were used to identify factors independently influencing short- and long-term survival, respectively. RESULTS: The cumulative 1-, 2-, 3-, 5-, and 10-year survival rates in 381 patients were 47%, 23%, 16%, 12%, 6%, respectively. The 1-, 3-, and 5-year survival rates calculated from time of re-resection were 36%, 14% and 0% in patients undergoing re-resection for intrahepatic recurrence within 2 years after first operation, and 85%, 53% and 32% in those more than 2 years after first operation (P<0.05). Multivariate analysis showed that portal vein infusion chemotherapy, serum alpha-fetoprotein > 20 mg/L and positive surgical margin were significant prognostic factors within 2 years after operation. In contrast, alanine aminotransferase > 80 U/L was the only significant factor beyond 2 years after operation. CONCLUSIONS: The survival of hepatocellular carcinomapatients with macroscopic portal vein tumor thrombosis was poor, but the prognosis of patients who had tumor recurrence more than 2 years after operation was much better than those with tumor recurrence within 2 years. Evaluation of time-dependency of risk factors may have important clinical implication in determining the therapeutic strategy.