Predictors of mortality in minority patients admitted to the ICU with COVID-19 infection

ABSTRACT

INTRODUCTION:Coronavirus-2 disease 2019 (COVID-19) is a novelty virus that caused a worldwide pandemic. It can cause mild to critical illness requiring intensive care unit (ICU) admission. In the United States, Black and Hispanic individuals comprise a disproportionately high number of infections and deaths due to COVID-19, likely related to underlying social and healthcare disparities.1,2 There are limited studies identifying predictors of outcome among COVID-19,3 in minority patients. The aim of this study was to identify the predictors of mortality among laboratory confirmed COVID-19 minority patients with severe clinical disease admitted to the ICU. METHODS:Clinical data at the time of ICU admission was extracted from electronic records for a total of 95 sequentially admitted patients to the medical ICU with confirmed COVID-19 diagnoses. Demographics, comorbidities, laboratory values that included inflammatory markers, ICU course, mortality and discharge status data were collected. The primary outcome was ICU mortality treated as a binary outcome. Summary characteristics were described based on survival status with a test of significance using ANOVA, kwallis and chisquare as appropriate. A univariate logistic regression was used to identify mortality predictor variables of statistical significance which were then included in a final multivariate regression model. Inflammatory markers were added individually to this finalized model to avoid collinearity. Findings were summarized using odds ratios and confidence intervals. RESULTS:The mean (SD) age was 61.54(14) years, 34(36%) were men, 67(71%) were African Americans and 20 (16%) were Hispanic. Most common comorbidities were hypertension 55 (58%) and diabetes 46 (48%). Fifty-three (56%) were intubated, 23 (25%) required pressor support, and 15 (16%) patients had their initial blood culture positive. Inflammatory markers were elevated in most all patients which was associated with mortality. ICU mortality was 48% (45 patients). Univariate analysis identified age ≥ 65yrs (odds ratio [OR]=1.25;95% CI,1.02-1.52;p= 0.032), higher SOFA scores of 2 and 3{ (OR=1.74, 95% CI ,1.05-2.89,p=0.035) and (OR=1.90,95%CI,1.1-3.29;p=0.024 respectively)}, vasopressor use ( OR=1.77;95%CI,1.44-2.18;p<0.001), severe ARDS (OR=;1.45;95%CI,1.05-2.01;p=0.027), mechanical ventilation use (OR=1.46;95%CI,1.22-1.79;p<0.001), procalcitonin>2.5ng/ml (OR=1.84;95% CI, 95%CI,1.03-3.29;p=0.042), ferritin>2000ng/ml (OR=1.45;95% CI,1.12-1.89;p=0.007), CRP>20mg/dl (OR=1.67 OR=;95CI,1.3-2.13;p<0.001) and LDH>400 (OR=1.68;95%C,1.26-2.23;p<0.001) as predictors of ICU morality. Of these, only age ≥ 65yrs, mechanical ventilation and vasopressor use remained statistically significant independent predictors of mortality in multivariable regression model. CONCLUSIONAmong predominantly minority patients with severe COVID-19 admitted to the ICU, older patients who become intubated, requiring vasopressor support and/or had elevated biomarkers of inflammation had a significantly higher ICU mortality.