射頻消融術(RFA)乃治療肝細胞癌眾多非手術治療方法中最有效者，近年來已被廣泛採用，尤其對於三公分以下肝腫瘤療效尤佳。本研究分析台灣中部某醫學中心收治病患，評估射頻消融術治療肝細胞癌的長期療效及併發症，並利用統計分析及分類法中的M5’模式樹，嘗試找出最佳的療效時間與消融次數，以供初學者參考，並期望能提升台灣治療肝細胞癌的治療成果。本研究收集個案醫院2003年6月至2011年5月，共210位病患於8年中接受324次RFA治療358顆肝細胞癌，腫瘤大小0.8公分~10公分，平均2.69公分。所有病患皆接受電腦斷層導引射頻消融治療，並於術後立即追蹤雙相對比劑電腦斷層攝影，然後於術後2個月及每隔3個月定期追蹤其AFP、肝功能及CT。結果顯示，我們整體的手術併發症是1.2%(4/324)，一年、三年及五年的存活率為91.9%、53.2%及35.7%。若以術後2個月CT評估腫瘤壞死率，則小於或等於3公分的腫瘤有95%達到完全壞死，若腫瘤小於或等於5公分則有89%腫瘤完全壞死。腫瘤的局部復發率分別是一年11.7%及三年17.6%。進一步利用決策樹及M5’模式樹來分類與分析，我們發現若腫瘤靠近超過3mm以上的肝內血管，較容易治療不完全及產生局部復發。若以3公分水冷式消融探針為例，2.5公分以下肝腫瘤，單針一次消融即可治癒肝腫瘤，但若腫瘤大於2.5公分或靠近3mm以上大血管，則需二次以上的消融治療方能克竟全功。利用電腦斷層導引的射頻消融術是一種既安全又有效的非手術治療肝細胞癌的主要方法，尤其是小於3公分的肝癌，RFA的治療效果與手術切除無異，值得推廣採用以降低病患的手術風險及健保與病患的財務負擔。為了降低局部復發率，其消融灶應至少比腫瘤邊緣多5毫米以上，尤其是靠近3mm以上肝內血管的肝細胞癌。 Radiofrequency ablation (RFA) has been accepted as the most effective non-surgical method for the treatment of hepatocellular carcinoma (HCC) less than 3cm in size. In this study, we retrospectively review the clinical data and follow-up images of the patients with HCC received RFA in 8 years. The therapeutic effectiveness and complication rate of RFA for managing HCC in this series were evaluated. Further more, we utilize OLAP, CART and M5’ for classification of our patients’ group and try to find the best way for obtaining the better results in the future. From June 2003 to May 2011, totally 210 patients with 358 HCC lesions were referred to a medical center at Mid-Taiwan for 324 sessions of RFA treatment. The tumor size ranged from 0.8 to 10cm with average of 2.69cm. All patients received CT-guided RFA under local anesthesia with PCA pain control. The dual phases CT scan was performed as well as liver function test and alpha-fetoprotein for follow-up tumor response at immediately, 2 months, and then every 3 months after the procedure. The overall complication rate was 1.2% (4/324 procedures) and the 1, 3 and 5 years of survival rate were 91.9%, 53.2% and 35.7% respectively. The complete tumor necrosis rate of tumors less than 3cm or 5cm were 95% or 89% and the local recurrence rate at 1 and 3 years were 11.7% and 17.6%.We further classified the patients with OLAP, CART and M5’ to analyze the risk factors of local recurrence and revealed that tumors nearby the vessels larger than 3mm had high risk of local recurrence. If the tumor size is less than 2.5cm without large vessel surrounded, one ablation with 3cm exposed cool-tip needle is enough, otherwise, more than two ablations may be indicated for tumors larger than 2.5cm and/or nearby the large vessels. In conclusion, CT-guide RFA is a safe and cost-effective non-surgical modality for the treatment of hepatocellular carcinoma, especially for the tumor mass less than 3cm in size. For the purpose of decrease of local tumor recurrence, the safe margin must be 5mm beyond the tumor at least.
