射频消融治疗早期肝癌的荟萃分析

A meta-analysis of radiofrequency ablation for early hepatocelluar carcinoma

目的根据现有的临床研究资料评价射频消融（RFA）治疗早期肝癌的疗效与安全性。方法检索Cochrane图书馆、EMBASE、PubMed、OVID等数据库和中国期刊网中的随机对照试验（RCT）文献，将Jadad评分≥3分的高质量论文纳入研究，并提取纳入研究的特征信息。采用STATA9．0软件进行数据分析，首先进行异质性检验，根据异质性检验结果选择相应的效应模型。然后进行敏感性分析，并以漏斗图和Egger回归方程评定有无发表偏倚。结果共6篇RCT文献纳入本研究，共包含862例早期肝癌患者。研究表明与其他治疗方法相比，RFA治疗可明显增加早期肝癌患者的3年生存率，并减少了肝内局部复发率，OR值分别为2．06（95％CI为1．54～2．77，P=0．000）和0．40（95％CI为0．28～0．57，P=0．000）；而在肝内转移、远处转移、严重副作用等方面差异无统计学意义。漏斗图基本呈现下宽上窄左右对称的图形，Egger线性回归显示3年生存率、肝内局部复发率、肝内转移、远处转移、严重副作用等方面的P值分别为0．670、0．160、0．884、0．087、0．317，提示无发表偏倚。结论早期肝癌经RFA治疗可获得优于其他疗法的局部疗效和3年生存率。RFA具有微创、简便和经济的优点，有可能作为早期肝癌的首选治疗手段之一。

Objective To compare the effectiveness and safety of radiofrenquency ablation （RFA） with other therapeutic methods for patients with early hepatocelhiar carcinoma （HCC）. Methods Randomized clinical trials （ RCTs ） which compared the efficacy or safety of RFA with other therapeutic methods for primary hepatocellular carcinoma in Cochrane library, EMBASE, PubMed, OVID and CBM were searched. Trials were considered of high quality if methodological quality score was 3 or more according to Jadad standard. Statistical heterogeneity between trials was evaluated by STATA 9. 0 and considered to exist when P 〈 0. 1. Heterogeneity of the included articles was tested and used to select proper effective model for calculation. Sensitivity analysis was performed and publication bias was investigated through visual inspection of funnel plots and Egger regression model. Results Six RCTs including 862 cases were analyzed. As compared with other therapeutic approaches, RFA significantly increased 3-year overall survival rate and reduced local recurrence rate of early hepatocelluar carcinoma; the total OR were 2. 06 （ 95 % CI being 1.54 - 2. 77, P = 0. 000 ） and 0. 40 （ 95 % CI being 0. 28 - 0. 57, P = 0. 000 ） respectively. As compared with other therapeutic approaches, the total OR of new HCC recurrence rates, extrabepatic metastasis rate and major complications in patients with HCC treated with RFA were 0. 92 （ 95% CI being 0.68 - 1.24）, 0. 98 （ 95% CI being 0. 30 - 3.22 ）, 1.35 （ 95% CI being 0. 49 - 3.77 ） respectively, showing no significant differences （ P 〉 0. 05 ）. Inspection of the funnel plots for all outcome measures did not reveal evidence of publication bias （P = 0. 670,0. 160,0. 884,0.087,0. 317, respectively, by Egger regression model ）. Conclusions RFA is superior to other treatment methods with respect to local recurrence and 3 year overall survival in early HCC and is the preferred therapeutic method for small HCC because it is minimally invasive, simple and convenient.