目的 系统评价急性缺血性脑卒中（acute ischemic stroke，AIS）出血转化风险预测模型，为临床决策工具的选择及护理风险评估工具的开发提供参考和借鉴。 方法 检索中国知网、万方、PubMed、Embase等数据库中AIS出血转化预测模型的相关研究，检索时间为建库至2022年12月。对纳入文献进行质量评价，并依据数据提取清单提取资料。 结果 共纳入27篇文献，包括30项出血转化风险预测模型，C统计量0.682~0.956，25项进行了内部验证，18项进行了外部验证。美国国立卫生院卒中量表(NIH Stroke Scale，NIHSS)、发病到开始治疗时间、年龄、血糖、收缩压是多变量风险预测模型中占比较高的独立预测因子。 结论 已开发的风险预测模型具有良好的区分度，但模型偏倚风险较高。NIHSS评分、发病到治疗时间、血糖、年龄、血压是AIS出血转化的独立危险因素，有助于高危患者识别。

Objective To systematically evaluate the risk prediction models for hemorrhagic transformation(HT) in acute ischemic stroke(AIS)，and to provide reference for the selection of clinical decision-making tools and the development of nursing risk assessment tools. Methods Databases including CNKI，Wanfang，PubMed，Embase and other databases were searched for studies on the prediction model of AIS HT from the inception to December 2022.The quality of the included literature was evaluated，and the data were extracted according to the data extraction checklist. Results A total of 27 articles were included，including 30 risk prediction models of HT.The C statistics were 0.682-0.956.25 models were internally validated，and 18 models were externally validated.The National Institutes of Health Stroke Scale (NIHSS)，time from onset to treatment，age，blood glucose and systolic blood pressure were the independent predictors with high proportion in the multivariate risk prediction model. Conclusion The developed risk prediction models have good discrimination，but the risk of model bias is high.NIHSS score，time from onset to treatment，blood glucose，age and blood pressure are independent risk factors for HT of AIS，which are helpful to identify high-risk patients.

R47；R473.54

江苏省科技计划（资金）项目（BE2022668）；中华护理学会科研课题专项资助（ZHKYQ202108）