发热伴血小板减少综合征预后模型构建与验证

Construction and validation of a prognostic model for severe fever with thrombocytopenia syndrome

目的探究发热伴血小板减少综合征(severe fever with thrombocytopenia syndrome,SFTS)患者预后的影响因素,构建列线图预测模型并验证评价。方法回顾性分析2019年4月至2024年5月于安徽医科大学第二附属医院住院治疗SFTS成人患者的临床资料,根据院内生存情况分为生存组和死亡组。采用二元Logistic回归分析确定SFTS患者预后不良的独立危险因素并构建列线图预测模型,使用Bootstrap法重复抽样1000次进行验证,采用受试者工作特征(receiver operating characteristic,ROC)曲线及其曲线下面积(area under curve,AUC)、校准曲线和决策曲线评估列线图模型的区分度、校准度和临床适用性。结果共纳入271例SFTS患者,其中生存组215例,死亡组56例。二元Logistic回归分析显示年龄[OR=1.089,95%CI(1.018,1.166)]、病毒载量[OR=2.047,95%CI(1.192,3.514)]、消化道出血[OR=5.368,95%CI(1.635,17.625)]、肺部真菌感染[OR=6.446,95%CI(2.115,19.645)]、血清铁蛋白[OR=8.198,95%CI(1.850,36.340)]是SFTS患者院内死亡的影响因素。列线图预测模型AUC值为0.936[95%CI(0.906,0.965)],Bootstrap法重复抽样1000次后的AUC为0.928[95%CI(0.898,0.960)],校准曲线、决策曲线显示模型具有较好的一致性和净收益。结论年龄、病毒载量、消化道出血、肺部真菌感染、血清铁蛋白是SFTS院内死亡的影响因素,本研究构建的SFTS患者院内死亡风险的列线图预测模型可辅助临床识别高危患者,具有一定的临床使用价值。

Objective To explore the risk factors for the prognosis of patients with severe fever with thrombocytopenia syndrome(SFTS),and establish and validate a nomogram prediction model.Methods A retrospective analysis was conducted on the clinical data of adult patients with SFTS admitted to the Second Hospital of Anhui Medical University from April 2019 to May 2024.Patients were divided into the survival group and death group according to their survival status in the hospital.Binary Logistic regression analysis was used to determine independent influencing factors for poor prognosis in SFTS patients,and then a nomogram prediction model was constructed.The differentiation,calibration and clinical applicability of the nomogram prediction model were evaluated using the receiver operating characteristic(ROC)curve,area under curve(AUC),calibration curve,and decision curve.Results A total of 271 SFTS patients were included,of which 215 were in the survival group and 56 in the death group.Binary Logistic regression analysis showed that age[OR=1.089,95%CI(1.018,1.166)],viral load[OR=2.047,95%CI(1.192,3.514)],gastrointestinal bleeding[OR=5.368,95%CI(1.635,17.625)],pulmonary fungal infection[OR=6.446,95%CI(2.115,19.645)],and serum ferritin[OR=8.198,95%CI(1.850,36.340)]were independent influencing factors for the poor prognosis of SFTS patients.The AUC value of the nomogram prediction model was 0.936[95%CI(0.906,0.965)],and the AUC of the Bootstrap method after repeated sampling 1,000 times was 0.928[95%CI(0.898,0.960)].The calibration curve and decision curve showed that the model had good consistency and net returns.Conclusion Age,viral load,gastrointestinal bleeding,pulmonary fungal infection,and serum ferritin are the influencing factors for the poor prognosis of SFTS.The nomogram prediction model for the prognosis of SFTS patients constructed in this study can assist in clinical identification of high-risk patients,showing certain clinical value.