不同无针输液接头对中心静脉导管血栓性堵管影响的系统评价

Effects of needleless connectors on central venous catheter-related thrombotic occlusions： a systematic review

目的通过系统评价临床上不同种类输液接头，从而选择将导管血栓性堵管降至最低的类型用于临床。方法计算机检索万方、知网、生物医学文献数据库、Cochrane、PubMed、Embase、CINAHL，收集所有讨论不同输液接头对中心静脉导管血栓性堵管影响的研究。由2名研究者独立进行文献检索、文献筛选、资料提取和质量评价，并对合格研究进行描述或Meta分析。结果共有9篇文献符合纳入标准，3篇随机对照试验，3篇非随机试验性研究，3篇回顾性观察研究，对其中5个研究进行了Meta分析，结果显示：分隔膜输液接头的导管发生血栓性堵管（不完全＋完全堵管）较正压接头高[RR=0．59，95％CI 0．39-0．91，P=0．02]。标准输液接头、负压接头的导管堵塞发生率均较正压接头高，但合并效应无统计学意义（P〉0．05）。结论鉴于有限的证据，无法确定将中心静脉导管血栓性堵管降至最低的接头类型，仍需开展大样本、高质量的随机对照试验。

Objective To evaluate the influence of different types of needleless connector on central venous catheter-related thrombotic occlusions. Methods A detailed search was performed to identify literature about the impact of needleless connector on central venous catheter-related thrombotic occlusions, using the Cochrane Library and the databases of Pubmed, EMbase, CINAHL, CBM, CNKI and WanFang. The meta-analysis or descriptive review were performed after two authors independently searching databases, extracting data and assessing quality of included studies. Results Nine published studies were included for systematic review, which contained three randomized controlled trials, three non-randomized experimental studies and three retrospective observational studies. The results of meta-analysis for five studies showed that the split septum connector was associated with a significantly higher incidence of catheter-related thrombotic occlusions（complete occlusions and partial occlusions） than positive- displacement mechanical valve（RR=0.59, 95% CI 0.39-0.91, P=0.02）. Standard connector and negative-displacement mechanical valve had a higher occlusions rates than positive-displacement mechanical valve ,but the difference did not reach statistical significance after merging the studies（P〉0.05）. Conclusion It is difficult to decide which needleless connector has the lowest rate of central venous catheter-related thrombotic occlusions because of the insufficient evidence. More large-scale and high-quality randomized controlled trials are required for further confirmation.