392. Predictors of Mortality in Hospitalized Patients with COVID-19; A Single Centered Retrospective Analysis. (31st December 2020)

• Record Type: Journal Article
• Authors: Harmouch, Farah N; Shah, Kashyap; Goel, Harsh
• Is Part Of: Open forum infectious diseases. Volume 7:Number 1(2020) Supplement
• Page Start: S265
• Page End: S266
• Publication Date: 2020-12-31
• Subjects: Communicable diseases -- Periodicals; Medical microbiology -- Periodicals; Infection -- Periodicals; 616.9
• DOI: 10.1093/ofid/ofaa439.587
• Languages: English
• ISSNs: 2328-8957

Abstract

Abstract: Background: The Coronavirus disease-2019 (COVID-19) has been responsible for the death of over 400, 000 people with a continuous rise in prevalence and mortality globally. Identifying hospitalized patients at high mortality risk is critical for triage and health-care resource management regionally, nationally, and globally. We present a retrospective analysis of predictors of mortality in hospitalized COVID-19 patients. Methods: Electronic health records (EHR) of patients admitted between March 1 and April 18, 2020 to St. Luke's University Hospital with a primary diagnosis of COVID-19 were reviewed for medical co-morbidities and initial biochemical/inflammatory markers. Survivors vs non-survivors were compared using χ 2 test, Student's t-test, and Mann-Whitney U-test as appropriate. Univariate logistic regression was used to identify candidate variables for multivariate analysis, which were then included in stepwise backward logistic regression. Statistical analyses were done on SPSS v26 software (IBM, Armonk, NY). Results: Clinical characteristics, biochemical abnormalities and results of univariate regression in our cohort of 560 patients are noted in table 1. Multivariate regression revealed age, congestive heart failure (CHF), and creatinine≥ 1.5 mg/dl as significant predictors of mortality while race (Caucasian), vascular disease, lymphopenia, and elevated ferritin approached significance (Table 2). Table 1: Baseline clinical characteristics, overall and by mortality. Continuous variables are presented as median (25th-75th percentile), and categorical variables as n (%) Significance of difference between subgroups (survivors versus non-survivors) *p≤0.05, **p≤0.01, ***p≤0.001 Table 2: Results of stepwise backward conditional logistic regression for predicting mortality among hospitalized COVID-19 patients. (n=334, 287 survivors and 47 non-survivors). ALC – Absolute lymphocyte count, S.E. – Standard error of B. Conclusion: We present one of the largest cohorts to date of hospitalized COVID-19 patients. Age, CHF, and renal disease were significant independent predictors of mortality. Though several inflammatory markers (d-dimer, CRP, procalcitonin) initially predicted mortality, they failed in multivariate analysis, questioning their role in risk-stratifying COVID-19 hospitalized patients. Interestingly, IL-6 used in those severely ill patients to assess candidacy for IL-6 inhibitor therapy (Tocilizumab) failed to predict mortality in our study. Our analysis was limited due to its retrospective nature and unfortunately large amounts of data were missing for some variables (ESR, BNP, IL-6 levels). The missing data was due to rapidly evolving institutional protocols early during the pandemic, leading to non-uniform assessment of these markers. Disclosures: All Authors : No reported disclosures