异基因造血干细胞移植术后并发肛周感染预测模型的构建与验证OA
Construction and validation of a risk prediction model for patients complicated with perianal infection after allogeneic hematopoietic stem cell transplantation
目的:探讨异基因造血干细胞移植术后并发肛周感染的危险因素,构建风险预测模型并验证.方法:选取2019年1月—2023年12月在重庆市某三级甲等医院进行异基因造血干细胞移植术的259例病人作为建模组,通过单因素和多因素分析筛选相关危险因素,利用R语言构建风险预测模型.选取2024年1月—2025年3月该院进行异基因造血干细胞移植术的75例病人作为验证组.采用受试者工作特征曲线下面积(AUC)、Hosmer-Lemeshow拟合优度检验及决策曲线评估及验证预测模型.结果:建模组肛周感染发生率为18.53%,验证组肛周感染发生率为24.00%.异基因造血干细胞移植术后并发肛周感染的危险因素包括肛周疾病史、腹泻、粒细胞缺乏期持续时间及术前预后营养指数.根据危险因素构建异基因造血干细胞移植术后并发肛周感染预测模型.建模组的AUC值为0.773,灵敏度为0.604,特异度为0.853,最大约登指数为0.457;验证组的AUC值为0.726,灵敏度为0.745,特异度为0.897,最大约登指数为0.442.决策曲线分析显示该模型在较宽风险阈概率内具有良好的临床获益.结论:构建的异基因造血干细胞移植术后并发肛周感染预测模型具有良好的预测性能,可为早期评估异基因造血干细胞移植术后并发肛周感染提供科学依据.
Objective:To explore the risk factors for patients complicated with perianal infection after allogeneic hematopoietic stem cell transplantation,and to construct and verify the risk prediction model.Methods:A total of 259 patients who underwent allogeneic hematopoietic stem cell transplantation at a tertiary grade A hospital in Chongqing from January 2019 to December 2023 were selected as the modeling group.Relevant risk factors were screened through univariate and multivariate analyses,and a risk prediction model was constructed using R language.A total of 75 patients who underwent allogeneic hematopoietic stem cell transplantation in hospital from January 2024 to March 2025 were selected as the validation group.The model was evaluated and validated using the receiver operating characteristic curve area(AUC),Hosmer-Lemeshow goodness-of-fit test,decision curve analysis,and validation of the prediction model adopted.Results:The incidence of perianal infection in the modeling group was 18.53%,while that in the validation group was 24.00%.The risk factors of perianal infection after allogeneic hematopoietic stem cell transplantation included the history of perianal diseases,diarrhea,duration of neutropenia,and preoperative prognostic nutritional index.A predictive model for perianal infection after allogeneic hematopoietic stem cell transplantation was constructed based on these risk factors.The AUC value of the modeling group was 0.773,with a sensitivity of 0.604,a specificity of 0.853,and a maximum Youden index of 0.457.The AUC value of the validation group was 0.726,with a sensitivity of 0.745,a specificity of 0.897,and a maximum Youden index of 0.442.The decision curve analysis indicated that this model had good clinical benefits within a wide range of risk threshold probabilities.Conclusions:The predictive model for patients complicated with perianal infection after allogeneic hematopoietic stem cell transplantation has excellent predictive performance.It could provide a scientific basis for the early assessment of perianal infection after allogeneic hematopoietic stem cell transplantation.
吴霞;叶洪敏;马海霞;胡盼;邓建川;甘秀妮
重庆医科大学附属第二医院,重庆 400000重庆医科大学附属第二医院,重庆 400000重庆医科大学附属第二医院,重庆 400000重庆医科大学附属第二医院,重庆 400000重庆医科大学附属第二医院,重庆 400000重庆医科大学附属第二医院,重庆 400000
造血干细胞移植肛周感染危险因素预测模型
hematopoietic stem cell transplantationperianal infectionrisk factorspredictive model
《护理研究》 2026 (16)
2815-2823,9
重庆市自然科学基金创新发展联合基金资助项目,编号:CSTB2023NSCQ-LZX0052
评论