RESUMEN
From January 1986 to December 1988, 85 patients (55 men and 30 women, mean age 59 years) with metastatic liver tumors were treated with hepatic artery embolization (TAE) or infusion (HAI). Sixty-eight patients with successful catheterization were treated with TAE using iodized oil (Lipiodol) mixed with anticancer agent (ACA). In 12 of 68 patients with hypervascular tumors gelatin sponge was added. Patients with unsuccessful catheterization were treated with hepatic artery infusion of ACA. Forty-three patients received oral chemotherapy following TAE or HAI. Overall, the 6-month, and 1- and 2-year survival rates were 69.5, 31.8 and 4.1 per cent, respectively (mean 233 days). A univariate analysis of prognostic factors showed that number of metastases, stage, treatment times and oral chemotherapy were all significant factors (p less than 0.05). Ascites, jaundice, percentage of hepatic replacement and treatment protocol also had some influence (p less than 0.1). Sex, age, primary site, elevation of tumor markers, other metastatic lesions, portal vein involvement and difference in anticancer agent had no prognostic significance. A multivariate analysis using Cox's proportional hazard model revealed that the number of treatments had the most important prognostic significance, followed by oral chemotherapy, stage and percentage of hepatic replacement.