A Simple Multiparametric Score System to Predict In-Hospital Mortality of COVID-19 Patients

Abstract

Background: Several clinical, laboratory and instrumental prognostic indicators for coronavirus disease 2019 (COVID-19) have been found. Combining all the different predictors in a score would make easier and more accurate the risk assessment of COVID-19 patients. To this purpose, we examined a large number of COVID-19 patients. First, we identified the best predictors of in-hospital mortality at admission. Then, we calculated a score system to capture the contribution of the various prognostic indicators.Methods: Prospective multicenter study (ELCOVID) referring to central-northern Italy. This project is registered on ClinicalTrials.gov (identifier: NCT04367129). COVID-19 patients admitted to the hospital in the period May-September 2020 were enrolled. Clinical, laboratory and electrocardiographic (ECG) records were collected at admission. Patients were followed-up and in-hospital mortality constituted the primary endpoint. A risk scoring system to predict prognosis was derived by independent predictors of in-hospital mortality.Findings: A total of 1014 patients fulfilled inclusion criteria. Demographic, clinical, laboratories and ECG characteristics were collected. Median age was 74 (IQR 64-82) years, and most patients were male (61%). During a median follow-up of 12 (IQR 7-22) days, 359 (35%) patients died. Age (HR 2.25, 95%CIs 1.72-2.94, p Interpretation: This score system stratifies prognosis and may be important for the management of COVID-19 patients admitted to the hospital.Trial Registration: ClinicalTrials.gov (identifier: NCT04367129).Funding Statement: None.Declaration of Interests: None declared.Ethics Approval Statement: ELCOVID is a prospective observational study approved by the local Ethics Committee and involves 15 hospitals in the Emilia Romagna and Lazio, two regions in northern and central Italy heavily affected by the pandemic.