514. Clinical features and outcomes of 112 patients with SARS-CoV-2 infections requiring intensive care in a public healthcare system in South Florida

BACKGROUND: SARS-CoV-2 infections may result in intensive care unit (ICU) admission and mortality. We characterized patients with SARS-CoV-2 infections that required intensive care in a public healthcare system in South Florida. METHODS: Consecutive patients with SARS-CoV-2 infections, admitted to the ICUs of Memorial Healthcare System (MHS) in South Florida, from March 7 to May 14, 2020 were retrospectively reviewed. The primary outcome was the rate of in-ICU death. A Cox proportional hazards regression model was used to determine the independent risk factors for death during ICU stay. RESULTS: Of 702 patients with SARS-CoV-2 infections admitted to MHS from March 7 to May 14, 2020, 112 (16%) patients required ICU admission. The median age was 67 years (IQR 58–75) and 68 (61%) were men. The most common comorbidities were hypertension (88; 79%) and diabetes (43; 38%). All patients received hydroxychloroquine therapy, 42 (38%) received convalescent plasma transfusion, 39 (35%) received Tociluzimab, and 14 (13%) received Remdesivir. 90 (80%) patients required invasive mechanical ventilation for an average of 18 days (median = 14 days). 47 (42%) received vasopressors, and 34 (30%) received renal replacement therapy. As of May 14, 2020, 40 (36%) patients had died, 64 (57%) discharged alive, and 8 (7%) remained in the ICU. The ICU length of stay was comparable between discharged patients and those that died. An increased duration of mechanical ventilation was observed in discharged patients (Figure 1). Of 64 discharged patients, 9 (14%) were readmitted to ICU, with a median time to readmission of 4 days (IQR, 1.5–6.5). By multivariable Cox regression analysis, chronic pulmonary disease was one independent risk factor associated with in-ICU mortality, whereas renal replacement therapy, convalescent plasma transfusion, and tocilizumab administration were negatively associated with in-ICU mortality (Figure 2). Figure 1. Clinical outcomes of critically ill COVID-19 patients required intensive care. [Image: see text] Figure 2. Forest plots for adjusted hazard ratio of in-ICU mortality in critically ill COVID-19 patients required intensive care. [Image: see text] CONCLUSION: In a public healthcare system in South Florida, patients with SARS-CoV-2 infection that required intensive care had 36% mortality. Mortality was associated with chronic pulmonary disease and inversely associated with renal replacement therapy, convalescent plasma transfusion, and tocilizumab administration. DISCLOSURES: All Authors: No reported disclosures