ObjectiveTo conduct a comprehensive analysis of risk prediction models for acute kidney injury (AKI) following Stanford type A aortic dissection surgery through a systematic review. MethodsA systematic search was performed in English and Chinese databases such as PubMed, EMbase, ProQuest, Web of Science, China National Knowledge Infrastructure (CNKI), VIP, Wanfang, and SinoMed to collect relevant literature published up to January 2025. Two researchers completed the literature screening and data extraction. The methodological quality of the prediction models was assessed using bias risk assessment tools, and a meta-analysis was performed using R version 4.3.1, with a focus on evaluating the predictive factors of the models. Results A total of 15 studies were included (13 retrospective cohort studies, 1 prospective cohort study, and 1 case-control study), involving 22 risk prediction models and a cumulative sample size of 4 498 patients. The overall applicability of the included studies was good, but all 15 studies exhibited a high risk of bias. The meta-analysis revealed that the area under the curve (AUC) for the predictive performance of the models was 0.834 [95%CI (0.798, 0.869)]. Further subgroup analysis indicated that the number of predictive factors was a source of heterogeneity. Additionally, hypertension [OR=2.35, 95%CI (1.55, 3.54)], serum creatinine [OR=1.01, 95%CI (1.00, 1.01)], age [OR=1.05, 95%CI (1.02, 1.09)], and white blood cell count [OR=1.14, 95%CI (1.06, 1.22)] were identified as predictors of AKI following type A aortic dissection surgery. Conclusion Currently, the predictive models for AKI after type A aortic dissection surgery demonstrate good performance. However, all included models carry a high risk of bias. It is recommended to strengthen multicenter prospective studies and external validation of the models to enhance their clinical applicability.
