PURPOSE: To prospectively evaluate the technical feasibility, effectiveness, and complications of percutaneous radiofrequency (RF) ablation for hepatic malignancies during temporary percutaneous balloon occlusion (PBO) of a large hepatic or portal vein. MATERIALS AND METHODS: During a 4-year period, RF ablation was performed in 201 patients (106 men, 95 women; age range, 41-88 years) with 233 liver tumors. Institutional review board approval was obtained to attempt RF ablation during PBO for 18 tumors that were larger than 35 mm (mean, 43 mm +/- 7.6 [standard deviation]; range, 36-60 mm) and did not abut a portal or hepatic vein 4 mm in diameter or larger (group 1), 58 tumors 35 mm or smaller (mean, 23 mm +/- 7.3; range, 12-35 mm) that abutted a large vessel (group 2), and 20 tumors that were both larger than 35 mm (mean, 42 mm +/- 5.7; range, 38-50 mm) and abutted a large vessel (group 3). RF ablation without PBO was performed for 137 tumors 35 mm or smaller (mean, 22 mm +/- 6.8; range, 9-35 mm) and remote from large vessels (group 4). Rate of local tumor progression was estimated with the Kaplan-Meier method, and tumor progression-free rates were compared between the four groups with the log-rank test. Complications were compared by using the Fisher exact test between the four groups and between the two RF devices used. RESULTS: PBO was achieved in 94 of 96 attempts (98%), including 64 of 64 hepatic veins and 30 of 32 portal branches. After a mean follow-up of 18 months +/- 9, 10 tumors in eight patients were lost to follow-up. Local tumor progression was observed in six (40%) of 15 tumors in group 1, in six (11%) of 56 tumors in group 2, in eight (40%) of 20 tumors in group 3, and in 12 (9%) of 130 tumors in group 4. Combined analysis of tumor size and the use of PBO showed that size was the only prognostic factor for tumor progression, with a hazard ratio of 4.9 (95% confidence interval: 2.4, 9.9) (P < .001). There were no differences between groups 2 and 4. Asymptomatic, transient postprocedure venous thrombosis was seen in nine of 94 RF ablations with PBO, while occlusion of one permanent portal branch induced segmental liver atrophy. There were no differences in rates of complications (5% and 6% for RF ablation with and that without PBO, respectively). CONCLUSION: RF ablation with PBO provides tumor control for tumors smaller than 35 mm in diameter that abut vessels 4 mm or larger, equivalent to tumor control of the same-size tumors away from vessels. PBO does not seem to affect the results of RF ablation for tumors 35 mm or larger.