Risk Stratification of Patients with COVID-19 Presenting at the Emergency Department

Abstract

Background: Since the outbreak of the coronavirus disease 2019 (COVID-19) clinical characteristics and risk factors for unfavorable outcome of patients with COVID-19 have been published, mainly from Wuhan, China. However, a risk score including patient characteristics and specific cut-off values of laboratory parameters for better risk stratification and prediction for requiring mechanical ventilation (MV) of these patients is lacking.

Methods: All patients with confirmed COVID-19 from a single center in Munich, Germany, who presented at the emergency department were retrospectively included into the present study. Demographic data, clinical symptoms, radiological results, laboratory parameters, treatment, and outcome data were obtained from electronic medical records and paper based patient charts. Follow-up of all patients was at least 14 days except they were discharged before.

Findings: A total of 60 patients were included into the present study. Of them, 26·6% (n=16) patients required MV in the ICU. Acute respiratory distress syndrome based on the WHO definition was observed in 41·6% (n=25). Patients who required MV had more often cardiovascular and pulmonary co-morbidities compared to those who did not require MV. On admission, patients with need for MV had lower blood oxygen saturation and increased indices of inflammation, renal failure and liver failure as well as higher lactate dehydrogenase and high-sensitive (hs) troponin I. A risk score consisting of arterial hypertension, need for oxygen therapy, and ROC derived cut-offs for procalcitonin (0.09 µg/l), lactate dehydrogenase (348 U/l), and hs-troponin I (15 ng/l) predicted the need for MV or ARDS with high sensitivity and specificity. A score of ≥3 out of 5 points showed sensitivity of 94% and specificity of 91% for the need of MV.

Interpretation: A simple risk score consisting of clinical and laboratory parameters may facilitate and improve risk stratification at initial presentation of patients with COVID-19 at the emergency department.

Funding: No funding was available for the present study.

Declaration of Interests: The authors do not report any conflict of interest regarding this study.

Ethics Approval Statement: Demographic data, clinical symptoms, radiological results, laboratory parameters, treatment, and outcome data were obtained from electronic medical records and paper based patient charts. All patient data were anonymized directly after data entry into the database. Therefore, written informed consent was waived. This is approved by the local ethics committee.