ICU患者中心静脉导管血流感染危险因素的Meta分析

目的通过Meta分析明确重症监护病房(ICU)患者中心静脉导管血流感染(CRBSI)的危险因素。方法计算机检索The Cochrane Library、PubMed、Embase、CBM、CNKI、WanFang Data数据库中关于ICU患者CRBSI危险因素的研究文献,应用RevMan 5.3软件对纳入文献进行Meta分析。结果符合纳入标准的文献有12篇,总样本量为14 422例,5篇队列研究,7篇病例对照研究,NOS评分为4~6分。Meta分析结果显示:导管留置时间(WMD=12.25,95%CI=5.55~18.94)、多腔导管(OR=3.52,95%CI=1.46~8.52)、股静脉置管(OR=2.44,95%CI=1.34~4.46)、肠外营养(OR=2.47,95%CI=1.18~5.21)、住ICU的时间(WMD=10.01,95%CI=4.17~15.85)、APACHEⅡ评分(WMD=4.46,95%CI=1.25~7.66)、合并糖尿病(OR=1.83,95%CI=1.08~3.09),各组差异有统计学意义(均P0.05)。结论导管留置时间、多腔导管、股静脉置管、肠外营养、住ICU的时间、APACHEⅡ评分、合并糖尿病是ICU患者CRBSI危险因素。因纳入研究总体方法学质量一般,尚需设计严谨、大样本的前瞻性队列研究对结果进行进一步验证。

Meta analysis on risk factors for central venous catheter related bloodstream infection in intensive care unit patients

ObjectiveTo evaluate risk factors for central venous catheter related bloodstream infection(CRBSI) in intensive care unit(ICU) patients by Meta analysis. MethodsLiteratures about risk factors of CRBSI in ICU patients were retrieved from databases of Cochrane Library， PubMed， Embase， CBM， CNKI， and WanFang Data, RevMan 5.3 software was used for Meta analysis. ResultsThere are 12 literatures in accordance with the inclusion criteria, with a total sample size of 14 422 cases, 5 cohort studies and 7 case control studies, the Newcastle Ottawa Scale(NOS) scores were 4-6 points. According to Meta analysis, duration of indwelling catheter（WMD, 12.25 ［95%CI, 5.55-18.94］）, multi cavity catheter（OR, 3.52 ［95%CI , 1.46-8.52］）, femoral vein catheterization（OR, 2.44 ［95%CI, 1.34-4.46］）, parenteral nutrition（OR, 2.47 ［95%CI, 1.18-5.21］）, length of stay in ICU（ WMD, 10.01 ［95%CI, 4.17-15.85］）, APACHE II score（WMD, 4.46 ［95%CI, 1.25-7.66］）, and diabetes mellitus（OR, 1.83 ［95%CI, 1.08-3.09］）were significantly different in each group (all P<0.05). ConclusionRisk factors for CRBSI in ICU patients are duration of indwelling catheter, multi cavity catheter, femoral vein catheterization, parenteral nutrition, length of stay in ICU, APACHE II score, and diabetes mellitus. However, due to the limitation of methodological quality of included studies, more strictly designed and large sample prospective studies are needed to verify the result.