利奈唑胺和万古霉素对革兰阳性球菌肺炎治疗效果的荟萃分析

Linezolid versus vancomycin in the treatment of pneumonia caused by Gram-positive cocci：meta-analysis of randomised controlled trials

目的 采用荟萃分析方法对现已发表的利奈唑胺和万古霉素治疗革兰阳性球菌肺炎的文献进行综合分析,评价利奈唑胺的疗效及其安全性是否优于万古霉素.方法 检索Medline数据库、Embase数据库、Ovid数据库、Cochrane图书馆及中文生物医学期刊数据库中的相关文献.检索年限均从建库到2009年2月,并查阅所有纳入文献的参考文献.外文检索词包括linezolid、glycopeptides、vancomycin、pneumonia、gram-positive cocci、saureus、MRSA、enterococcus及streptococci.限定语言为中文或英文,限定对象为＂人＂.中文检索词为相应的主题词.纳入用英文或中文发表的比较利奈唑胺和万古霉素治疗革兰阳性球菌肺炎疗效的随机对照试验,由2名评价员独立筛查文献、评价质量和提取资料.采用Jadad量表及随机分配方案隐藏方法评估纳入研究的方法学质量;采用x2检验鉴定研究间的异质性,使用随机效应或固定效应模型合并研究;采用敏感性分析方法探讨研究结果的影响因素.结果共纳入7个随机对照研究,包括1425例革兰阳性球菌肺炎患者.荟萃分析结果显示,利奈唑胺治疗结束后在临床可评估患者的临床治愈率优于万古霉素（OR=2.16,95%CI为1.13～4.16,P〈0.05）;随访结束后,临床可评估患者（OR=1.11,95%CI为0.81～1.53,P〉0.05）及意向性治疗患者（OR=1.01,95%CI为0.78～1.31,P〉0.05）利奈唑胺的临床治愈率、微生物学总治愈率（OR=1.31,95%CI为0.85～2.04,P〉0.05）、金黄色葡萄球菌清除率（OR=1.45,95%CI为0.84～2.51,P〉0.05）、耐甲氧西林金黄色葡萄球菌清除率（OR=1.36,95%CI为0.51～3.61,P〉0.05）、链球菌清除率（OR=4.27,95%CI为0.01～1365.87,P〉0.05）及肠球菌清除率（OR=0.75,95%CI为0.03～17.51,P〉0.05）与万古霉素相同.利条唑胺治疗组与万古霉素治疗组的病死率（OR=0.80,95%CI为0.59～1.07,P〉0.05）及不良反应总体发生率（OR=1.06,95%CI为0.68～1.64,P〉0.05）比较差异无统计学意义.结论 对革兰阳性球菌肺炎患者,利奈唑胺虽在治疗刚结束时的临床疗效优于万古霉素,但1～4周随访结束后,两组临床疗效无明显差异.

Objective To study the randomised controlled trials on the efficacy and safety in patients with pneumonia caused by Gram-positive cocci treated with linezolid versus with vancomycin.Methods The data were collected from the MEDLINE database,EMBASE,OVID,the Cochrane library,and Chinese Biomedical Database,and the references of eligible studies were manually screened.Randomized controlled trials published in the English and Chinese literatures comparing linezolid with vancomycin in patients with pneumonia caused by Gram-positive cocci were eligible for inclusion.Results Seven randomized controlled studies comparing linezolid with vancomycin were analyzed,focusing on 1425 patients with pneumonia caused by Gram-positive cocci.It was found by meta-analysis that,with respect to clinical treatment success,linezolid was more effective than vancomycin in clinically evaluation patients at the end-of-treatment visit （OR=2.16,95% CI 1.13-4.16,P〈.05）,however,there was no difference in clinical treatment success for clinically evaluation patients （OR=1.11,95% CI0.81-1.53,P〉.05）and intention-to-treat patients （OR=1.01,95% CI0.78-1.31,P〉.05）at the test-of-cure visit.With respect to microbiological treatment success,linezolid was as effective as vancomycin in microbiologically evaluation patients at the test-of-cure visit （OR=1.31,95% CI 0.85-2.04,P〉.05）.Additionally,empirical treatment with linezolid was not associated with increased eradication rates for saureus strains （OR=1.45,95% CI0.84-2.51,P〉.05）and methicillin resistant Staphylococcus aureus strains （OR=1.36,95% CI 0.51-3.61,P〉.05）in comparison with vancomycin in microbiologically evaluation patients at the test-of-cure visit,and,there was no difference in eradication rates for streptococcal species （OR =4.27,95% CI 0.01-1365.87,P〉.05）and enterococcal species （OR=0.75,95% CI 0.03-17.51,P〉.05）.Mortality was similar between the groups （OR=0.80,95% CI 0.59-1.07,P〉0.05）.Treatment with linezolid was not associated with more adverse effects in general in comparison with vancomycin （OR=1.06,95% CI 0.68-1.64,P〉.05）.Conclusions Linezolid is at least as effective as vancomycin for treatment of patients with pneumonia caused by gram-positive cocci although rigorously designed large sample sized,randomized double blind clinical trials are required to further demonstrate and support the conclusion.