Objective: To investigate the application of severity classification according to the protocol on the Diagnosis and Treatment of coronavirus disease 2019(COVID-19)by the National Health Commission of China, pneumonia severity index(PSI) and CURB-65 in risk stratification and prognostic assessment of COVID-19. Methods: Clinical data of 234 in-hospital patients with COVID-19 were collected and retrospectively reviewed in Wuhan Tongji Hospital. Patients were divided into 3 groups (common, severe, and critical type) at admission according to the sixth version of the protocol issued by the National Health Commission of China on Diagnosis and Treatment of COVID-19. At the same time, the severity of pneumonia was calculated by PSI and CURB-65, and the patients were stratified into 3 risk groups, namely mild, moderate, and severe groups. The hospital mortality rate was evaluated in each group. Sensitivity, specificity, positive predictive values, negative predictive values, and the area under the receiver operating characteristic(ROC) curve(AUC) for predicting hospital mortality in each rule were assessed. Results: According to the severity classification of Chinese protocol, the proportion of patients with common type, severe type, and the critical type was 15.8%, 75.6%, and 8.5%, respectively. No in-hospital death occurred in the common type. As for PSI and CURB-65, greater proportions of patients were classified as low risk(79.1% and 75.6%, respectively), while smaller proportions of patients were classified as moderate and high risk(16.2%, 15.0%; 4.7%, 9.4%, respectively). In-hospital death occurred in low and moderate risk patients identified by these 2 scoring systems. The mortality of the critical group of the Chinese protocol was 65%, and the sensitivity and specificity of predicting in-hospital mortality were 36.4% and 97.0%, respectively. The mortality in the high risk group of PSI and CURB-65 was 100% and 77.3%. The risk class V of PSI and CURB-65 score 3-5 had high specificity(100% and 97.4%, respectively)but low sensitivity(33.3% and 51.5%, respectively)in predicting in-hospital mortality. The AUC of the Chinese protocol severity classification, PSI, and CURB-65 was 0.735, 0.951, and 0.912. The optimal cut-off point of PSI was risk class Ⅳ, and the sensitivity and specificity for predicting mortality were 90.9% and 90.5%. The optimal cut-off point of CURB-65 was score 2, and the corresponding sensitivity and specificity were 84.8% and 85.6%. Conclusions: PSI and CURB-65 can be used for risk stratification and prognostic assessment in patients with COVID-19.
目的： 探讨新型冠状病毒肺炎（COVID-19）诊疗方案临床分型、肺炎严重度指数（PSI）和CURB-65在COVID-19患者危险度分层及预后评估中的应用。 方法： 回顾性分析234例武汉同济医院中法新城院区收治的COVID-19患者的临床资料，入院时根据新型冠状病毒肺炎诊疗方案将患者分为普通型、重型和危重型，同时使用PSI、CURB-65进行危险度分层，将患者分为低危、中危和高危组。比较各组患者病死率，评价各评估方法预测死亡的敏感度、特异度、阳性预测值、阴性预测值、受试者工作特征（ROC）曲线下面积（AUC）。 结果： 依据诊疗方案临床分型，普通型、重型和危重型患者的比例分别为15.8%、75.6%和8.5%，普通型患者未出现院内死亡；而PSI、CURB-65分别将患者中79.1%、75.6%划分为低危组，16.2%、15.0%划分为中危组，4.7%、9.4%划分为高危组，低危组和中危组均存在死亡病例。入院时诊疗方案临床分型为危重型的患者住院期间病死率为65%，预测死亡的敏感度和特异度分别为36.4%、97.0%，而PSI和CURB-65高危组病死率分别为100%和77.3%，其预测死亡的敏感度分别为33.3%、51.5%，特异度分别为100%和97.5%。诊疗方案临床分型、PSI和CURB-65预测死亡的AUC分别为0.735、0.951、0.912。PSI预测死亡的最佳截点值为Ⅳ级，相对应的敏感度为90.9%，特异度为90.5%；CURB-65的最佳截点值为2分，预计死亡的敏感度为84.8%，特异度为85.6%。 结论： PSI和CURB-65可用于COVID-19患者危险度分层和预后评估。.
Keywords: Coronavirus disease 2019; Prognosis; Severity of illness.