术后谵妄风险预测模型的系统评价

Systematic evaluation of the risk prediction models for post-operative delirium

目的 系统评价术后谵妄风险预测模型。方法 在中国知网、万方数据库、维普、中国生物医学文献数据库、pubmed、embase(OVID)、EBSCO (ASP)、Cochrane图书馆等数据库全面检索国内外有关术后谵妄预测模型的研究,检索时限均为建库至2021年6月。由2名研究者独立筛选文献、提取资料后,采用PROBAST评估预测模型的质量,然后对预测模型进行系统评价。结果 共纳入21个风险预测模型,其中15个为前瞻性研究,15个模型的受试工作者曲线下面积超过0.7,整体偏倚风险高,适用性风险低;预测模型中包含最多的谵妄易感因素是认知功能损害和年龄,最常见的诱发因素是内环境紊乱(酸碱失衡、电解质紊乱)。结论 纳入研究的整体预测性能较好,适用性风险较低,但整体偏倚风险较高,医务人员应该结合自身实际,将已有模型进行验证用于指导临床,开展针对性的预防措施降低术后谵妄的发生。

Objective To systematically evaluate the risk prediction model for postoperative delirium in adults. Methods Domestic and foreign studies on prediction models of postoperative delirium were searched comprehensivelyin CNKI,Wanfang Database, VIP,Chinese Biomedical Literature Database, pubmed, embase(OVID),EBSCO(ASP),Cochrane Library and other databases, and the retrieval period was established from the database establishment to June 2021. After two researchers independently screened the literature and extracted data, PROBAST was used to evaluate the quality of the prediction model, and then the prediction model was systematically evaluated.Results A total of 21 risk prediction models were included, 15 of which were prospective studies. The area under the receiver operating characteristic curvesin 15 models were more than 0.7, and the overall risk of bias was high and the risk of applicability was low. The cognitive impairment and age were the most commonly reported predisposing factors of delirium among all prediction model, and the most common predisposing factors were disorders of the internal environment(acid-base imbalance, electrolytedisturbance). Conclusion The overall prediction performance of the included studies is good, and the applicability risk is low, but the overall risk of bias is high. Medical staff shouldcombine their own actual conditions to verify the existing model for clinicalguidance, and carry out targeted preventive measures to reduce the occurrence of postoperative delirium.