连续性肾脏替代治疗滤器凝血风险预测模型的构建及验证  

Construction and Verification of a Filter Coagulation Risk Prediction Model for Continuous Renal Replacement Therapy

作　　者：贾燕 李娜[1] 刘考[1] 武海环 JIA Yan;LI Nai;LIU Kao;WU Haihuan(Department of Organ Transplantation,Qilu Hospital,Shandong University,Jinan,Shandong 250063;Qilu Hospital Health Management Center,Shandong University,Jinan,Shandong 250063)

机构地区：[1]山东大学齐鲁医院器官移植科,山东济南250063 [2]山东大学齐鲁医院健康管理中心,山东济南250063

出　　处：《智慧健康》2024年第25期68-72,共5页Smart Healthcare

摘　　要：目的 构建连续性肾脏替代治疗(continuous renal replacement therapy,CRRT)滤器凝血风险预测模型,并验证其临床预测效果。方法 选择2021年7月—2022年7月在山东大学齐鲁医院接受CRRT的257例患者作为建模组,根据是否发生滤器凝血分为发生组和未发生组,将两组的指标应用二元Logistic回归构建预测模型。于2022年8月—2023年1月纳入112例患者作为验证组对模型进行验证。结果 本研究中建模组CRRT滤器凝血发生率为19.46%(50/257),验证组为15.18%(17/112)。最终纳入模型的预测变量为是否应用血管活性药物、血流量、超滤量。结论 是否应用血管活性药物、血流量、超滤量时发生CRRT滤器凝血的独立危险因素。本研究构建的模型预测效能良好,可为临床治疗提供参考。Objective To construct a filter coagulation risk prediction model for continuous renal replacement therapy(CRRT)and verify its clinical predictive effect.Methods The paper chose 257 patients treated with CRRT in Qilu Hospital of Shandong University from July 2021 to July 2022,and divided them into occurrence group and non occurrence group based on whether filter coagulation occurred,constructed a prediction model based on indicators between two groups with binary logistic regression.112 cases were enrolled from August 2022 to January 2023 to verify the model.Results Filter coagulation incidence rate of the CRRT in modeling group was 19.46%(50/257),while that in verification group was 15.18%(17/112).Final predictive variables included in the model included whether vasoactive drugs were used,blood flow and ultrafiltration volume.Conclusion Vasoactive drugs,blood flow and ultrafiltration volume are independent risk factors for occurrence of CRRT filter coagulation.The model constructed in the study has good predictive performance and can provide reference for clinical treatment.

关 键 词：连续性肾脏替代治疗 滤器凝血 风险预测模型 

分 类 号：R459.5[医药卫生—治疗学]