ABSTRACT
BACKGROUND: Radiofrequency ablation (RFA) is gaining increased acceptance for the local control of liver tumors. Essential for achieving local tumor control are reproducible volumes of ablation that encompass the tumor and a margin of normal liver parenchyma. The technical algorithm for performing ablations was arrived at in an animal model using normal liver. Limited amounts of data exist as to whether this translates to the human tumor model. METHODS: We analyzed 531 ablated lesions in 154 patients undergoing laparoscopic RFA using RITA Medical Systems Starburst XL catheter deployed to a final diameter of 2-5 cm. The first 54 patients (algorithm 1) were treated with a larger initial deployment to 3 cm and incremental advancement of the catheter to the final diameter with a 20-min ablation time for a 5-cm lesion. The subsequent 100 patients (algorithm 2) were treated with a smaller initial deployment of 2 cm, incremental advancement to the final diameter, and 14-min total ablation time for a 5-cm lesion. Lesion size was measured on 1 week postablation CT scans. Analysis was performed using the two-tailed t-test. RESULTS: Ablation zones tended to be larger with the second method. On 1 week postablation CT scans, mean +/- SEM lesion sizes created using the first and second algorithms were 3.7 +/- 0.1 cm vs 4.0 +/- 0.1 cm at 3 cm deployment ( p < 0.05); 4.3 +/- 0.1 cm vs 4.8 +/- 0.1 cm at 4 cm deployment ( p < 0.05), and 5.5 +/- 0.1 cm vs 5.6 +/- 0.2 cm at 5 cm deployment ( p > 0.05), respectively. The mean +/- SEM total ablation times for the first and second algorithms were 7.9 +/- 0.3 min vs 7.0 +/- 0.2 min at 3 cm deployment ( p < 0.05); 13.3 +/- 0.3 min vs 11.1 +/- 0.02 min at 4 cm deployment ( p < 0.05); and 27.8 +/- 1.2 min vs 21.4 +/- 1.2 min at 5 cm deployment ( p < 0.05), respectively. The small SEM values indicate little variation in lesion size. CONCLUSIONS: These results show that both algorithms create dependable and reproducible zones of ablation, essential for reliable tumor destruction. Algorithm 2 demonstrates that creating an initial small core of ablation with rapid coagulation of the center of the lesion allows for equivalent, if not larger, final volumes to be performed in less time.