含左氧氟沙星方案治疗复治肺结核临床疗效的Meta分析

Meta-analysis on clinical efficacy of combination regimens with levofloxacin in retreatment of pulmonary tuberculosis

目的评价含左氧氟沙星方案治疗复治肺结核的临床疗效。方法计算机检索中国生物医学文献光盘数据库、中国期刊全文数据库、万方数字化期刊、维普中文科技期刊等数据库,检索年限:1994年—2008年。按纳入、排除标准纳入合格的随机对照研究并对其进行质量评价,Meta分析采用RevMan4.2.7软件进行。结果共纳入8篇研究,包括791例患者,但这些研究的质量普遍不高。Meta分析结果表明,一线抗结核药加左氧氟沙星在提高痰菌阴转率、促进病灶吸收方面,优于一线抗结核药治疗复治肺结核。其OR和95%CI分别为3.70[2.59,5.27]、3.10[2.25,4.28]。左氧氟沙星方案与单纯一线抗结核药治疗比较,可能促进空洞闭合,其OR和95%CI为2.41[1.50,3.89],但RR随机(95%CI)为1.63[0.99,2.68]。不良反应方面无统计学差异,其OR和95%CI为0.94[0.58,1.53]。结论尽管有证据支持左氧氟沙星与一线抗结核药联用在提高痰菌阴转率、促进病灶吸收方面,优于一线抗结核药治疗复治肺结核。由于纳入研究的数量少、潜在的发表偏倚和质量不高,尚不能对其疗效得出肯定性的结论。因此,继续开展高质量、大样本的随机对照试验非常必要。

Objective To assess the clinical efficacy of combination regimens with Levofloxacin （LVFX） in retreatment of pulmonary tuberculosis. Methods We searched CBM, CNKI, Wan Fang Digitalization Periodical, and Wei Pu Chinese technology periodical from 1994 to 2006 for randomized controlled trials（RCTs）, which were enrolled according to the criteria for inclusion and exclusion, the quality were evaluated and meta-analysis performed by RevMan4.2. 7 software. Results We found 8 trials including 791 patients with poor qualities. The meta-analysis showed that there were statistically significant differences on the rates of sputum negative and radiographic improvement between containing Levofloxacin group and control group （OR 3.70, 95% CI 2.59 to 5.27, and OR3.10, 95% CI 2.25 to 4.28）. The rate of cavities closure in the study group maybe higher than that of the control（OR 2.41, 95% CI 1.50 to 3.89, RR 1.63, 95% CI 0.99 to 2.68） There was no difference on side effects（OR 0.94,95% CI 0.58 to 1.53）. Conclusions So far, there are a few evidences to support the result, which the clinical efficacy on the rates of sputum negative and the radiographic improvement in the study group is better than that of the control. Due to few studies, potential publication bias, and low quality of trails, we could not draw reliable conclusions from the present data. High quality RCTs should be performed further.