维持性血液透析患者合并2019冠状病毒病死亡的危险因素及生存预测模型构建

Risk factors of death and construction of a survival prediction model in maintenance hemodialysis patients with corona virus disease 2019

该研究为回顾性队列研究。纳入2022年12月15日至2023年1月15日苏州两家三级医院因2019冠状病毒病(corona virus disease 2019,COVID-19)住院的80例维持性血液透析(maintenance hemodialysis,MHD)患者,比较存活组(48例)和死亡组(32例)临床资料的差异。采用Cox回归模型分析MHD患者合并COVID-19死亡的危险因素,并构建生存预测模型。结果显示,死亡组年龄、男性比例、合并糖尿病比例、低氧血症比例、病灶累积肺段数量、C反应蛋白、降钙素原、血清铁蛋白、白细胞介素6、D-二聚体及血磷均高于存活组(均P<0.05)。年龄增加(HR=1.039,95%CI 1.007~1.072,P=0.017)、合并糖尿病(HR=2.668,95%CI 1.018~6.991,P=0.046)、C反应蛋白升高(HR=1.006,95%CI 1.001~1.011,P=0.012)及血磷升高(HR=1.573,95%CI 1.015~2.438,P=0.043)为MHD患者合并COVID-19死亡的独立影响因素。以年龄、糖尿病、C反应蛋白及血磷构建患者生存预测模型,其7、14、21 d生存时间的联合模型的受试者工作特征曲线下面积分别为0.751(95%CI 0.690~0.811)、0.768(95%CI 0.712~0.824)和0.780(95%CI 0.729~0.831)。交叉验证作为内部验证获得的一致性指数(C-index)为0.797(95%CI 0.757~0.837)。该研究显示,年龄增加、合并糖尿病、C反应蛋白升高及血磷升高是MHD患者合并COVID-19死亡的独立危险因素,以其构建的生存预测模型效能良好。

It was a retrospective cohort study.Eighty maintenance hemodialysis(MHD)patients with corona virus disease 2019(COVID-19)were enrolled,among whom 48 patients survived and 32 died.The clinical data between the survival and death groups were compared.The Cox regression model was used to analyze the risk factors of death in MHD patients with COVID-19,and a survival prediction model was constructed.The results showed that age,lesion-cumulative number of lung segments,C-reactive protein,procalcitonin,serum ferritin,interleukin-6,D-dimer,serum phosphorus,and proportions of males,diabetes and hypoxemia in the death group were higher than those in the survival group(all P<0.05).Increased age(HR=1.039,95%CI 1.007-1.072,P=0.017),diabetes(HR=2.688,95%CI 1.018-6.991,P=0.046),increased C-reactive protein(HR=1.006,95%CI 1.001-1.011,P=0.012),and increased serum phosphorus(HR=1.573,95%CI 1.015-2.438,P=0.043)were independent influencing factors of death in MHD patients with COVID-19.The survival prediction model was established based on age,diabetes,C-reactive protein and blood phosphorus.The area under the receiver operating characteristic curve of the combined model for survival time at 7-day,14-day,and 21-day were 0.751(95%CI 0.690-0.811),0.768(95%CI 0.712-0.824),and 0.780(95%CI 0.729-0.831),respectively.The concordance index of cross-validation as internal validation was 0.797(95%CI 0.757-0.837).Increased age,diabetes,elevated C-reactive protein and elevated blood phosphorus are independent risk factors of COVID-19 death in MHD patients,and the survival prediction model built by those factors has good efficacy.