Predicting intensive care unit admissions for COVID-19 patients in the emergency department

Objective: Determining the parameters that can predict the requirement of intensive care unit (ICU) admissions among the coronavirus disease 2019 (COVID-19) patients presented to the emergency departments (EDs). Methods: In adult consecutive patients admitted (March 15 - April 15, 2020) to the ED of a state hospital for COVID-19, we retrospectively analyzed demographic data, symptoms, laboratory tests, and chest computed tomography (CT) on arrival. Results: We included 458 patients [213 (46.5%) females, median age 48 y]. Body temperature, respiration rate, C-reactive protein (CRP), D-dimer, ferritin values, and the number of comorbidities were significantly higher in patients admitted to the ICU than others. Also, diffuse infiltration in chest CT is more common in patients who need ICU follow-up. As a result of the binary regression analysis, a statistically significant correlation was found between the presence of dyspnea (odds ratio [OR]: 12.55), tachypnea (relative risk [RR] >= 18) (OR: 14.54), multiple comorbidities (>= 2) (OR: 23.39), diffuse infiltration in CT (OR: 14.52), and CRP (>= 45 mg/L) (OR: 4.71); and the need for ICU admission. Conclusion: It has been concluded that the presence of dyspnea and tachypnea, elevated CRP, presence of multiple comorbidities, and diffuse infiltration in CT may predict the need for ICU admissions of the patients, who presented to the EDs.