Simple Demographics and Laboratory Findings on Admission May Predict f In-hospital Mortality in Patients with SARS-CoV-2 Infection

Background: Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) infection could result in different laboratory abnormalities. The prediction of the outcome based on simple demographics and laboratory parameters could be useful for clinical purposes. The objective of the study is to develop and validate a score (Covid19-score) based on demographics and laboratory findings, performed at hospital admission in patients with a SARS-CoV-2 infection confirmed on a reverse transcriptase-polymerase chain reaction of the nasopharyngeal swab, to predict in-hospital mortality. Methods: Three cohorts of patients from three different hospitals were studied consecutively. The studied data came from patients’ electronic records. On the basis of the retrospective analysis of the mortality in the developing cohort from the first hospital the cut-off points predicting in-hospital mortality for gender, age, hemoglobin, mean corpuscular volume, platelet count, leukocyte count, sodium, potassium, creatinine level, C-reactive protein level were found and Covid19-score as a sum of points was calculated for each patient. The area under the receiver operating characteristic curve (AUC) of the Covid19-score for predicting survival to hospital discharge was counted. The Covid19-score was validated using data of patients from a second hospital. The significance of Covid19-score was confirmed on the prospective cohort of patients collected from a third hospital,Results: AUC of the Covid19-score for predicting survival to hospital discharge was 0.89 (0.84-0.95) p<0.001 in developing cohort, 0.850 (0.75-0.88) p<0.001 in validation cohort and 0.773 (0.731-0.816) p<0.001 in the prospective cohort. Conclusion: The Covid19-score is useful in predicting the clinical outcome for hospitalized patients with SARS-CoV-2 infection