发热伴血小板减少综合征死亡风险预测模型构建研究  

Construction of a mortality prediction model for severe fever with thrombocytopenia syndrome

作　　者：喻才正 阿依努尔·吐尔孙 迪里努尔·吾不力 雷清 刘伟[1] Yu Caizheng;Ayinuer·Tuersun;Dilinuer·Wubuli;Lei Qing;Liu Wei(Department of Public Health,Tongji Hospital,Tongji Medical College,Huazhong University of Science and Technology,Wuhan 430030,China;Department of Nephrology,Tongji Hospital,Tongji Medical College,Huazhong University of Science and Technology,Wuhan 430030,China)

机构地区：[1]华中科技大学同济医学院附属同济医院公共卫生科,武汉430030 [2]华中科技大学同济医学院附属同济医院肾病内科,武汉430030

出　　处：《中华临床感染病杂志》2023年第5期354-359,共6页Chinese Journal of Clinical Infectious Diseases

基　　金：国家自然科学基金(82302065)。

摘　　要：目的:建立发热伴血小板减少综合征(SFTS)患者死亡风险预测模型,为临床评估SFTS不良预后提供科学依据。方法:回顾性收集华中科技大学同济医学院附属同济医院2017年1月至2023年6月120例SFTS住院患者的临床资料,根据临床预后分为存活组(n=89)和死亡组(n=31)。采用多因素Logistic回归分析影响SFTS临床预后的危险因素,并建立死亡风险预测模型。通过受试者工作特征曲线(ROC)和曲线下面积(AUC)分析模型的预测价值。采用SPSS 23.0软件对数据进行处理和分析。结果:Logistic回归分析发现,皮肤出血点(OR=5.171,95%CI 1.617~16.530,P=0.006)、神志改变(OR=5.481,95%CI 1.540~19.512,P=0.009)、乳酸脱氢酶(LDH,OR=1.002,95%CI 1.001~1.004,P<0.001)和肌酐(OR=1.018,95%CI 1.007~1.029,P=0.002)是SFTS患者临床预后的独立危险因素。基于回归分析结果建立死亡风险预测模型:Logit(P)=-6.623+皮肤出血点×1.643+神志改变×1.701+LDH(U/L)×0.002+肌酐(μmol/L)×0.018。该预测模型的AUC为0.91(95%CI 0.86~0.96,P<0.001),其预测能力高于皮肤出血点(Z=3.788,P<0.001)、神志改变(Z=5.728,P<0.001)、LDH(Z=2.309,P=0.021)和肌酐(Z=2.064,P=0.039)。结论:基于皮肤出血点、神志改变、LDH和肌酐构建的SFTS患者死亡风险预测模型,对于SFTS患者预后具有较好的预测价值。Objective To construct a mortality prediction model for severe fever with thrombocytopenia syndrome(SFTS)and to evaluate its prediction ability.Methods Clinical data of 120 hospitalized patients with SFTS at Tongji Hospital,Tongji Medical College,Huazhong University of Science and Technology from January 2017 to June 2023 were retrospective analyzed.Based on clinical prognosis,patients were divided into survival group(n=89)and fatal group(n=31).The risk factors of SFTS mortality were analyzed with multivariate Logistic regression,based on which a mortality risk prediction model was constructed.The predictive value of the model was examined with receiver operator characteristic(ROC)curve.SPSS 23.0 sofware was used to process and analyze the data.Results Logistic regression analysis indicated that skin petechiae(0R=5.171,95%CI 1.617-16.530,P=0.006),mental disturbance(0R=5.481,95%CI 1.540-19.512,P=0.009),increased serum lactate dehydrogenase level(OR=1.002,95%CI 1.001-1.004,P<0.001),and increased serum creatinine level(0R=1.018,95%Cl:1.007-1.029,P=0.002)were independent risk factors for SFTS mortality.A mortality risk prediction model was established based on the regression coefficient of risk factors:Logit(P)=-6.623+skin petechiae×1.643+mental disturbance×1.701+lactate dehydrogenase level(U/L)×0.002+creatinine level(μmol/L)×0.018.The area under ROC curve(AUC)of the prediction model was 0.91(95%CI 0.86-0.96,P<0.001),and its predictive ability was higher than that of skin petechiae(Z=3.788,P<0.001),mind change(Z=5.728,P<0.001),lactate dehydrogenase(Z=2.309,P=0.021),and creatinine(Z=2.064,P=0.039).Conclusion The mortality prediction model constructed based on skin petechiae,mental disturbance,lactate dehydrogenase,and creatinine has good predictive value for the prognosis of SFTS patients.

关 键 词：发热伴血小板减少综合征 预后 预测模型 

分 类 号：R511[医药卫生—内科学] R181.3[医药卫生—临床医学]