Marco Montorsi, Roberto Santambrogio, Mara Costa, Paolo Bianchi, Angela Palmisano, Matteo Donadon, Eliana Moroni, Hepatobiliary Surgical Unit, Milano, Italy

Hepatic resection is still considered the treatment of choice for HCC on liver cirrhosis in patients not candidates to liver transplantation. Radiofrequency ablation is a new emerging modality aimed to get a complete ablation of the tumor. At present very few prospective studies are available comparing these two modalities. Aim of our study was to compare the results of HR and RF in the treatment of HCC on liver cirrhosis. From 2/97 to 2/03, a total of 98 patients were enrolled in this prospective study. Inclusion criteria were a single HCC nodule < 5 cm, Child A-B class of liver function and no previous treatment. HR was conducted according to the systematic segmentectomy technique proposed by Makuuchi; RF was performed trough a laparoscopic approach under general anesthesia. Intraoperative ultrasound (IOUS) was performed immediately before the treatment in both groups: 40 pts were included in the HR group and 58 pts in the RF group. The two groups were homogeneous as far as age, sex, class of liver function and tumor characteristics are concerned. IOUS diagnosed new malignant intrahepatic nodules in both groups (10% after HR vs. 18% after RF; p=0.241), which were treated in the same session. Operative mortality was nil and there was the same 19% rate of operative morbidity within the two groups. All patients were submitted to a follow-up protocol including biochemical examination, liver ultrasound and spiral CT at 1, 3, 6 and 12 months and twice a year thereafter. Mean follow-up was 21 + 15 months after RF and 24 + 15 months after HR. 95% of patients in the RF group showed a complete HCC necrosis at 1 month. Actuarial survival at 4 yrs was similar (66% after HR and 50% after RF; p=0.645). There was a significant higher incidence of intrahepatic recurrences after RF than after HR (53% vs. 35%; p= 0,017). A multivariate analysis (Cox model) showed the only significant factor predictive of an intrahepatic recurrence was the type of the treatment (beta: 0,234; hazard ratio: 2.339; 95% C.I: 2.14-2.54). On the other hand, only the level of alpha-fetoprotein (beta: 0.612; hazard ratio: 2.947; C.I: 2.54-3.36) was independent predictor of survival rate. With the limitations of a non-randomized study and the small sample size, the results of our study showed a significant lower incidence of intrahepatic recurrences after HR compared to RF. This seems not to have influence on the overall survival, probably because of a prompt treatment of the recurrence.