Multicenter Development and Validation of a Novel Risk Nomogram for Early Prediction of Severe 2019-Novel Coronavirus Pneumonia

Background: Severe cases of coronavirus disease 2019 (COVID-19) rapidly develop acute respiratory distress leading to respiratory failure, with remarkably high short-term mortality rates. At present, there is no reliable risk stratification tool for COVID-19 patients. We aimed to construct and validate a model for early identification of severe cases of COVID-19. Methods: SARS-CoV-2 infected patients from two centers in Guangzhou and one center in Wuhan were included retrospectively, and divided into the train and external validation cohorts. All patients with non-severe COVID-19 during hospitalization were followed for more than 15 days following admission and patients who deteriorated to severe COVID-19 were assigned to the severe group. Least absolute shrinkage and selection operator (LASSO) algorithm and logistic regression model were used to construct a nomogram for risk prediction in the train cohort. The predictive accuracy and discriminative ability of nomogram were evaluated by area under the curve (AUC) and calibration curve. Decision curve analysis (DCA) and clinical impact curve analysis (CICA) were conducted to evaluate the clinical applicability of our nomogram. Findings: The train cohort consisted of 189 patients, while the two independent validation cohorts consisted of 165 and 18 patients. Among all cases, 72 (19.35%) patients developed severe COVID-19. We generated the nomogram containing one clinical and six serological indicators (age, serum lactate dehydrogenase, C-reactive protein, the coefficient of variation of red blood cell distribution width, blood urea nitrogen, albumin, direct bilirubin) that could early identify severe COVID-19 patients. The nomogram showed remarkably high diagnostic accuracy in distinguishing individuals with severe COVID-19 from non-severe COVID-19 (AUC 0.914 [95% CI 0.852–0.976] in the train cohort; 0.856 [0.795-0.916] in validation cohort 1. The calibration curve for probability of severe COVID-19 showed optimal agreement between prediction by nomogram and actual observation. DCA and CICA further indicated that our nomogram conferred significantly high clinical net benefit. Interpretation: Our nomogram is a potentially useful prediction tool for risk assessment of COVID-19 patients and early identification of severe COVID-19 patients. Risk stratification will enable better management and optimal use of medical resources via patient prioritization and thus significantly reduce mortality rates.Funding Statement: Science and Technology Program of Guangzhou, China (201804010474)Declaration of Interests: The author(s) declare(s) that there is no conflict of interest regarding the publication of this paper.Ethics Approval Statement: The study was approved by the Ethics Committee of the Eighth People's Hospital of Guangzhou (20200547). Written informed consent was waived by the Ethics Commission of the Third Affiliated Hospital of Sun Yat-sen University for emerging infectious diseases.