肠源性脓毒症患者28 d死亡风险预测模型构建  

The construction of a 28 d death risk prediction model in patients with gut-derived sepsis

作　　者：萧意颖 廖吕钊 沈超 陈雅婷[1] 夏国莲[2] 江荣林[2] Xiao Yiying

机构地区：[1]浙江中医药大学,310053 [2]浙江中医药大学附属第一医院,310006 [3]浙江医院,310030

出　　处：《浙江临床医学》2025年第2期162-165,共4页Zhejiang Clinical Medical Journal

基　　金：浙江省中医药科技计划项目(2023ZL425);浙江省卫生健康委员会科技项目(2020KY663);浙江中医药大学自然科学青年探索项目(2021JKZKTS046B)。

摘　　要：目的构建肠源性脓毒症患者28 d死亡风险的预测模型,并评估该模型的预测价值。方法选取MIMIC IV数据库中221例肠源性脓毒症患者数据,根据28 d生存情况分为存活组(169例)和死亡组(52例),进行单因素和多因素Cox回归分析,确定与28 d死亡相关的危险因素,并基于多因素Cox回归结果构建预测模型。采用ROC曲线评估模型的预测能力。采用Kaplan-Meier进行生存分析以验证模型的预测价值。采用列线图对模型进行可视化。采用K折交叉验证对模型进行内部验证。采用Harrell's C-Index、校准曲线和决策曲线分析(DCA)分别对模型的区分度、校准度以及临床实用性进行评估。结果221例患者28 d死亡风险的关键预测因子包括年龄、白细胞、平均动脉压、活化部分凝血酶原时间、碱性磷酸酶、乳酸。采用上述指标构建肠源性脓毒症患者28 d死亡风险预测模型。模型的ROC曲线下面积为0.832(95%CI:0.766~0.899)。Kaplan-Meier生存分析结果提示高危组28 d死亡率大于低危组(P<0.0001)。K折交叉验证(K=10)平均AUC值为0.833(95%CI:0.778~0.888)。Harrell's C-Index为0.813(95%CI:0.734~0.892)。校准曲线接近理想曲线。DCA结果提示该模型在ICU环境中使用具有较高的临床价值。结论该预测模型有助于早期识别肠源性脓毒症高危患者,及时提供干预治疗,从而可能改善患者预后。Objective To develop a predictive model for 28-day mortality risk in patients with gut-derived sepsis and to evaluate the predictive value of this model.Methods A retrospective analysis was performed on 221 patients with gut-derived sepsis from the MIMIC IV database,who were divided into a survival group(169 cases)and a death group(52 cases)based on their 28-day survival status.Univariate and multivariate Cox regression analyses were performed to identify risk factors associated with 28-day mortality,and a predictive model was constructed based on the results of the multivariate Cox regression.The predictive ability of the model was assessed using the receiver operating characteristic(ROC)curve.Kaplan-Meier survival analysis was used to validate the predictive value of the model.A nomogram was used to visualize the model.The model was internally validated using K-fold crossvalidation.Harrell's C-Index,calibration curves,and decision curve analysis(DCA)were used to evaluate the discriminative ability,calibration,and clinical utility of the model,respectively.Results The key predictive factors for 28-day mortality risk in 221 patients included age,white blood cell count,mean arterial pressure,activated partial thromboplastin time,alkaline phosphatase,and lactate.A predictive model for 28-day mortality risk in patients with gut-derived sepsis was constructed using these indicators.The area under the ROC curve of the model was 0.832(95%CI:0.766~0.899).Kaplan-Meier survival analysis indicated that the high-risk group had a higher 28-day mortality rate than the low-risk group(P<0.0001).The average area under the curve(AUC)value of K-fold cross-validation(K=10)was 0.833(95%CI:0.778~0.888).Harrell's C-Index was 0.813(95%CI:0.734~0.892).The calibration curve was close to the ideal curve.DCA results suggested that the model had a high clinical benefit when used in the ICU environment.Conclusion The predictive model can help clinicians identify high-risk patients with gut-derived sepsis at an early stage,provide timely inte

关 键 词：肠源性脓毒症 危险因素 预测模型 感染 

分 类 号：R73[医药卫生—肿瘤]