中心静脉导管堵塞风险预测模型研究质量的系统评价

Systematic review of risk prediction models for central venous catheter occlusion

目的系统评价中心静脉导管堵塞风险预测模型。方法检索Cochrane Library、CINAHL、Embase、PubMed、Web of Science、万方数据库、维普数据库、中国生物医学文献数据库和中国知网关于中心静脉导管堵塞风险预测模型的研究,检索时限为建库至2023年6月9日。2名研究员独立筛选文献和提取数据,并进行质量评价。结果共纳入7项中心静脉导管堵塞风险预测模型研究,受试者工作特征曲线下面积为0.660-0.820。出现频次较高的预测因子包括血栓史、体重指数、D-二聚体水平、血流速度、导管尖端位置。偏倚风险评估显示纳入研究均存在较高偏倚风险,6项研究的适用性较好,其余1项研究的适用性较差,主要原因是样本量不足、未报告盲法、未恰当处理缺失数据等。结论现中心静脉导管堵塞风险预测模型的偏倚风险较高,未来应重点关注研究设计的方法学细节和报告的规范性,并开展多中心、大样本的研究,进一步优化与验证模型,提高模型性能。

Objective To systematically evaluate the risk prediction models of central venous catheter blockage.Methods The Cochrane Library,CINAHL,Embase,PubMed,Web of Science,Wanfang,VIP,Chinese Biomedical Literature Database(CBM)and CNKI were searched for the researches on the risk prediction models of central venous catheter blockage.The retrieval period was from the establishment of the database to June 9,2023.Two investigators independently screened the literature and extracted the data,and evaluated the quality.Results A total of seven studies on the risk prediction model for central venous catheter occlusion were included,and the area under the ROC curve ranged from 0.660 to 0.820.The predictors with the highest frequency were thrombus history,body massindex,D-dimer level,blood flow velocity,and catheter tip position.The risk of bias assessment showed that the included studies had high risk of bias,six studies had good applicability and the remaining one study had poor applicability,mainly due to insufficient sample size,no blinding method reported and no proper missing data handling.Conclusion The existing risk prediction models of central venous catheter blockage had high risk of bias.In the future,we should focus on the methodological details of the study design and the standardization of reporting,and multi-center and large sample studies should be carried out to further optimize and validate the model and improve the model performance.