Between June 2002 and September 2007, 116 consecutive patients with incompletely treated HCCs after TACE were referred to the department of radiology for RF ablation. Inclusion criteria for this study were as follows: 1) a single HCC smaller than or equal to 5 cm in diameter; 2) lesions visible on ultrasonography (US), with an acceptable and safe path between the lesion and the skin as observed on the US scan; 3) no extrahepatic metastases present; 4) no imaging evidence of tumor invasion into the major portal or hepatic vein branches; 5) liver cirrhosis classified as Child-Pugh class A or B; 6) no history of encephalopathy, ascites refractory to diuretics, or variceal bleeding, and 7) no previous treatment for HCC except TACE. Consequently, 39 patients with a single HCC (33 men and 6 women; age range 48-84 years; mean age 66 years) were enrolled in this study. Tumor size ranged from 1.1 to 5.0 cm in diameter (mean, 2.4 cm).

The diagnosis of incompletely treated HCC was made when there were areas within HCC that had not undergone complete necrosis based on the follow-up CT or MR imaging performed within 1 month after TACE. These areas were considered to be present by observing incomplete accumulation of lipiodol within HCC or arterially enhancing portion within HCC on follow up CT or MR imaging (19). Of the 39 patients, 22 patients were treated with one cycle of TACE, 13 patients with 2 cycles and 4 patients with 3 cycles.

All RF ablation was performed percutaneously under real-time sonographic guidance. A 20 or 15 cm long, 17 G, Cool-tip electrode with a 2 or 3 cm long exposed metallic tip (Valleylab, Boulder, CO, USA) was used to deliver radiofrequency energy. A 200 W, 480 kHz monopolar radiofrequency generator (CC-1, Valleylab, Boulder, CO, USA) was used as the energy source. A standard grounding pad (Valleylab, Boulder, CO, USA) was placed on each of the patient's thighs.

RF ablation procedures were performed with the patient under local anesthesia in the ultrasonography suite. The most appropriate approach was determined to avoid damaging large vessels near the targeted HCC, and then a single electrode was directly inserted through the skin and positioned at the center of the HCC under US guidance using a 3.75 MHz convex probe. The tip of the electrode was then further advanced to the deepest margin of the tumor. RF energy was applied for 8-12 min in each treatment session. During the procedure, a thermocouple embedded within the electrode tip continuously measured the temperature of the tip; tissue impedance was monitored by circuitry incorporated within the generator. When the desired current could not be applied without observing an elevation in impedance suggestive of tissue boiling, the generator automatically switched to the pulsed-RF technique. At the end of treatment, the RF ablation application and cooling circuits were simultaneously interrupted, and the heated tissues were allowed to heat the electrode by diffusion until the maximum temperature was recorded. To prevent bleeding, bile leakage, and tumor seeding, the intrahepatic needle track was treated with thermocoagulation and the electrode was removed. Tumors larger than 3.5 cm needed multiple overlapping ablations (20).

One hour after the initial treatment, contrast-enhanced dynamic CT was performed. When the thickness of the ablative margin was at least 0.5 cm for ablation of index tumor, the treatment was completed. When a residual enhanced lesion was seen on dynamic CT, an additional RF ablation was performed for the residual enhanced lesion. The effectiveness of the RF ablation technique was evaluated with dynamic CT performed 1 month after the first RF ablation session was performed. When no enhancing lesion was seen on CT, the technique effectiveness was defined as complete. When residual enhancing lesion was still seen on CT, the technique effectiveness was defined as incomplete. The HCCs showing incomplete technique effectiveness were not treated with additional RF ablation but rather with TACE.

Thereafter, the patients were evaluated on scheduled follow-ups every 2-4 months with contrast-enhanced dynamic CT to detect local recurrence. Local recurrence was defined as the development of tumoral enhancement in or around the ablation zone as observed on follow-up CT. Additionally, overall survival was evaluated and defined as the interval between the first RF ablation and either death or last follow-up visit. At each follow-up visit, blood tests, including those for liver function and serum alpha-fetoprotein levels, were performed.

Ten possible prognostic factors were postulated and each was divided into two categories for analyses. The possible prognostic factors were tumor size (≤ 2 cm or > 2 cm maximum dimension), number of previous TACE treatments (1 or ≥ 2), patient age, patient gender, antibody against hepatitis C virus (anti-HCV) (positive or negative), hepatitis B surface antigen (HBs Ag) (positive or negative), Child-Pugh class (A or B), history of alcoholic liver disease (yes or no), level of alpha-fetoprotein (≤ 100 ng/mL or > 100 ng/mL), and interval between TACE and RF ablation (≤ 2 months or > 2 months).

The complete ablation rate and major complication rate were calculated. The local recurrence free rate and overall survival rate were evaluated using the Kaplan-Meier method. Possible prognostic factors influencing local recurrence and overall survival were analyzed using a Cox proportional hazards regression model. The level of significance (p value) was set at 0.05 for all tests. SPSS statistical analysis software (SPSS Inc.) was used.

In 36 (92.3%) of 39 patients, complete tumor ablation was depicted on dynamic CT performed 1 month after treatment (Fig. 1). The 3 patients with a single HCC showing an incomplete response were not treated with a second RF ablation treatment because of the heat sink effect from the adjacent portal vein in one patient and large diameter (5 cm and 3.8 cm, respectively) in other 2 patients (Fig. 2). Afterward, the 3 patients in whom RF ablation treatment failed underwent subsequent TACE.

No procedure-related death was observed. Complications were observed in two (5.1%) of 39 patients. Hepatic abscess, which was a major complication, occurred in one patient. The patient had developed a high fever 7 days after RF ablation and the CT findings in the ablated zone was similar to those of a hepatic abscess. Subsequently, US-guided percutaneous drainage was performed. Portal venous thrombosis was observed in the other patient which was a minor complication.

The follow-up periods ranged from 90 to 1950 days (mean 767 days). The local recurrence free rates at 1, 2, 3, and 5 years were 81.7%, 63.1%, 53.6%, and 35.7%, respectively (Fig. 3). During the follow-up period, local recurrence developed in 13 (33.3%) of 39 patients. At the time of local tumor recurrence, distant recurrence was observed in 3 (7.7%) of 39 patients. Tumor diameter was found to be a significant prognostic factor affecting local recurrence (p = 0.047) (Table 1).

The overall survival rates at 1, 2, 3, and 5 years, were 96.9%, 82.9%, 67.8%, and 48.4%, respectively (Fig. 4). No significant prognostic factors affecting overall survival rate were found (Table 2).