Learning curve for radiofrequency ablation of liver tumors - Prospective analysis of initial 100 patients in a tertiary institution

Objective: This study aims to evaluate the effect of operator experience on the treatment outcomes of radiofrequency ablation (RFA) for malignant liver tumors. Summary Background Data: RFA is gaining popularity as the ablative therapy of choice for liver tumors. It is generally considered a simple and safe technique, and little attention has been paid to the importance of operator experience in this treatment. A learning curve in this treatment modality has not been documented before. Patients and Methods: The clinical data and treatment outcomes of the initial 100 patients undergoing RFA for liver tumors (hepato-cellular carcinoma, n = 84; metastasis, n = 15; cholangiocarcinoma n = 1) were collected prospectively. All patients were managed by a single team of surgeons and interventional radiologists. The data of the first 50 patients (group I) and the second 50 patients (group II) were compared. Results: RFA was performed by percutaneous (group I, n = 22; group II, n = 19), open (group I, n = 26; group II, n = 30) or laparoscopic (group I, n = 2; group II, n = 1) approach. In group I, 30 patients (60%) had a solitary tumor and 20 (40%) had multiple tumors; in group II, 35 patients (70%) had a solitary tumor and 15 (30%) had multiple tumors (P = 0.295). The size of the largest tumor was comparable between groups I and II (median, 2.8 cm in both groups; P = 0.508). Group II had significantly shorter hospital stay (median, 4.0 versus 5.5 days; P = 0.048), lower morbidity rate (4% versus 16%; P = 0.046) and higher complete ablation rate (100% versus 85.7%; P = 0.006) than group I. There was 1 hospital death (2%) in group I and 0 in group II. By multivariate analysis, treatment period (group I versus group II) was an independent significant factor affecting the morbidity rate and complete ablation rate. Conclusions: A low complication rate and a high complete ablation rate could be achieved with the accumulated experience from the first 50 cases of RFA for liver tumors by a specialized team. This study demonstrates that there is a significant learning curve in RFA for liver tumors.