新型冠状病毒肺炎首次临床和CT特征价值分析

目的：分析2019新型冠状病毒肺炎（COVID-19）首次临床和CT特征价值。方法：回顾性分析90例临床确诊COVID-19患者的首诊症状、一般炎症指标和胸部CT，包括13例重型/危重型和77例轻型/普通型，并由两名经验丰富的影像诊断医师进行分析和比较。结果：相对于轻型/普通型COVID-19患者，重型/危重型最重要的临床特征为年龄较大，一般炎症指标上升，胸部CT显示肺部炎症指数（PII）更高。受试者工作特征曲线（ROC）显示C-反应蛋白（CRP）、PII区分重型/危重型和轻型/普通型COVID-19患者的灵敏度分别为69.2%、69.2%，特异度分别为83.1%、87.0%，曲面下面积（AUC）分别为0.775[95% CI （0.649，0.901）；P=0.002]、0.770[95% CI （0.621，0.920）；P=0.002]，其联合的AUC为0.785[95% CI （0.644，0.927）；P=0.001]。结论：重型/危重型和轻型/普通型COVID-19患者的临床症状、炎症指标和胸部CT特征具有显著性区别，结合年龄、CRP和PII的检查，有助于早期识别重型/危重型COVID-19和评估其预后。

The First Clinical and CT Value Analysis of COVID-19

Objective:To analyze the value in the first clinical and chest CT features of Corona Virus Disease 2019(COVID-19). Methods:The first clinical symptoms、general inflammatory indicators and chest CT features of 90 patients with clinically diagnosed COVID-19 were retrospectively analyzed, which included 13 cases of severe/critical type and 77 cases of light/ordinary type. The results were analyzed and compared by two experienced radiologists. Results:Compared with the light/ordinary COVID-19 patients, the most important clinical characteristics of the severe/critical type were older age, general inflammatory indicators increased, and chest CT showed higher pulmonary inflammatory index(PII). The receiver operating characteristic curve(ROC) showed that the sensitivity of C-reactive protein(CRP) and PII to distinguish between severe/critical COVID-19 and light/ordinary COVID-19 was respectively 69.2% and 69.2%; The specificity was 83.1% and 87.0%, respectively; The Area Under Curve(AUC) were 0.775[95% CI(0.649, 0.901); P=0.002] and 0.770[95% CI(0.621, 0.920); P=0.002], respectively. The AUC of their association was 0.785[95% CI(0.644, 0.927); P=0.001]. Conclusion:Clinical symptoms, inflammatory indicators and chest CT features were significantly different between severe/critical COVID-19 and light/ordinary COVID-19 patients, and it is helpful for early identification of severe/critical COVID-19 and assessment of its prognosis when combined with age, PII and CRP.