摘要
目的新型冠状病毒感染(COVID-19,简称新冠感染)已成为全球大流行疾病,严重者可导致多器官损伤综合征,死亡率高,因此迫切需要可靠的生物标志物,用来早期筛选出可能进展为重症感染者,这对于改善临床预后具有重要意义。方法回顾性分析2022年12月8日—2023年1月31日因首次感染新冠病毒就诊于中国人民解放军联勤保障部队第九〇三医院呼吸与危重症医学科的94例患者的临床资料,依据新型冠状病毒感染诊疗方案(试行第十版)的分型,将其分为2组,其中轻型和中型为非重症组(57例),重型和危重型为重症组(37例)。通过单因素分析筛选出2组差异有统计学意义的指标,再通过多因素logistic回归分析筛选出影响早期新冠感染严重程度的独立危险因素。采用ROC曲线分析对应指标对早期新冠感染严重程度分层的预测价值,最后用列线图评估预测模型的检验效能。结果C反应蛋白(CRP)/白蛋白(ALB)、肺实变、冠心病可作为新冠感染严重程度的预测因素,其中CRP/ALB的ROC曲线下面积为0.762(95%CI:0.663~0.862,P<0.001),肺实变曲线下面积为0.682(95%CI:0.573~0.791,P=0.003),冠心病曲线下面积为0.638(95%CI:0.520~0.756,P=0.024)。CRP/ALB联合肺实变预测指标AUC为0.801(95%CI:0.708~0.894)。绘制列线图预测模型及其校正曲线,结果表明该预测模型具有较高的准确性。结论CRP/ALB联合肺实变能较好地预测新型冠状病毒感染严重程度。
Objective Corona Virus Disease 2019(COVID-19)has become a global pandemic disease,which can lead to multiple organ damage syndrome and high mortality.Therefore,there is an urgent need for reliable biomarkers that can be used for early screening for possible progression to severe infection,which is important for improving clinical prognosis.Methods For this retrospective study,basic information and clinical data from December 8,2022 to January 31,2023 were extracted from the medical records of patients with COVID-19 admitted to the 903rd Hospital of the People's Liberation Army.On the basis of a new type of coronavirus infection diagnosis and treatment plan 10(trial version)of parting,after into and exclusion criteria,the patients were divided into two groups,non-severe group(n=57)and severe group(n=37).Univariate analysis was used to screen out statistically significant indicators between the two groups,then multivariate analysis logistic regression was used to screen out independent risk factors affecting the severity of early COVID-19.The receiver operating characteristic curve was used to evaluate and analyze the predictive value of the corresponding indicators for the severity stratification of COVID-19 in the early stage.Finally,a nomogram was used to evaluate the test efficacy of the prediction model.Results The C-reactive protein/albumin ratio(CRP/ALB),lung consolidation and coronary heart disease were predictors of severity prediction,the area under the ROC of CRP/ALB was 0.762(95%CI:0.663-0.862,P<0.001),the area under the ROC of lung consolidation was 0.682(95%CI:0.573-0.791,P=0.003),and the area under the ROC of coronary heart disease was 0.638(95%CI:0.520-0.756,P=0.024).The AUC of CRP/ALB and lung consolidation was 0.801(95%CI:0.708-0.894).The nomogram prediction model and its calibration curve showed that the prediction model had high accuracy with the actual situation.Conclusion CRP/ALB combined with lung consolidation can predict the severity of COVID-19.
作者
洪佳慧
刘永洋
房宇坤
仇奕尹
方凯
陈清勇
HONG Jiahui;LIU Yongyang;FANG Yukun;QIU Yiyin;FANG Kai;CHEN Qingyong(The Second Clinical College of Zhejiang Chinese Medical University,Hangzhou,Zhejiang 310053,China;不详)
出处
《中华全科医学》
2024年第7期1116-1120,共5页
Chinese Journal of General Practice
基金
浙江省医药卫生科技计划项目(2020KY1025)。
