⽬的：系統評估冠⼼病患者非計劃再入院風險預測模型，為相關預測模型的構建、應⽤和優化提供參考依據。⽅法：計算機檢索 Cochrane Library、PubMed、Embase、Web of Science、中國知網、維普、萬⽅和中國⽣物醫學⽂獻數據庫中發表的關於冠⼼病患者非計劃再入院風險預測模型的相關研究，檢索時限為建庫至 2025 年 4 ⽉ 10 ⽇。由 2 位研究者獨⽴篩選⽂獻、提取資料，使⽤預測模型構建研究數據提取和質量評價清單對納入⽂獻的質量進⾏評價。結果：共納入 8 項研究，3 篇為模型的開發研究，5 篇為模型的開發與驗證研究。8 項⽂獻的模型區分度良好（AUC: 0.719~0.9773），N 末端 B 型利鈉肽前體、年齡、⾼⾎壓、住院時間、NYHA ⼼功能分級是冠⼼病患者非計劃再入院風險預測模型最常⾒的預測因數。所納入⽂獻的總體偏倚風險相對較⾼，但 8 個模型的適⽤性均較好。結論：冠⼼病患者非計劃再入院風險預測模型尚存在⼀些不⾜，未來應進⼀步提⾼相關模型的研究質量。

Objective: To systematically evaluate risk prediction models for unplanned readmission in patients with coronary heart disease, and to provide references for the construction, application, and optimization of relevant prediction models. Methods: Computerized searches were conducted in Cochrane Library, PubMed, Embase, Web of Science, CNKI, VIP, Wanfang, and CBM databases for studies related to risk prediction models for unplanned readmission in patients with coronary heart disease, with the search period spanning from database inception to April 10, 2025. Two researchers independently screened the literature, extracted data, and evaluated the quality of the included studies using a checklist for data extraction and quality assessment of prediction model development studies. Results: A total of 8 studies were included, comprising 3 model development studies and 5 model development and validation studies. The models in the 8 studies demonstrated good discriminative performance (AUC: 0.719–0.977). NT-proBNP, age, hypertension, length of hospital stay, and NYHA functional classification were the most common predictors in the risk prediction models for unplanned readmission in patients with coronary heart disease. The overall risk of bias in the included studies was relatively high, but the applicability of all 8 models was satisfactory. Conclusion: Current risk prediction models for unplanned readmission in patients with coronary heart disease have certain limitations, and future efforts should focus on improving the quality of research on such models.