Search - Critical Care Science (CCS)

ABSTRACT

Objective:

To compare, within a cohort of patients with acute respiratory failure, the phenotypes of patients with and without COVID-19 in the context of the pandemic and evaluate whether COVID-19 is an independent predictor of intensive care unit mortality.

Methods:

This historical cohort study evaluated 1001 acute respiratory failure patients with suspected COVID-19 admitted to the intensive care unit of 8 hospitals. Patients were classified as COVID-19 cases and non-COVID-19 cases according to real-time polymerase chain reaction results. Data on clinical and demographic characteristics were collected on intensive care unit admission, as well as daily clinical and laboratory data and intensive care unit outcomes.

Results:

Although the groups did not differ in terms of APACHE II or SOFA scores at admission, the COVID-19 group had more initial symptoms of fever, myalgia and diarrhea, had a longer duration of symptoms, and had a higher prevalence of obesity. They also had a lower PaO2/FiO2 ratio, lower platelet levels than non-COVID-19 patients, and more metabolic changes, such as higher levels of blood glucose, C-reactive protein, and lactic dehydrogenase. Patients with non-COVID-19 acute respiratory failure had a higher prevalence of chronic obstructive pulmonary disease/asthma and cardiopathy. Patients with COVID-19 stayed in the hospital longer and had more complications, such as acute kidney failure, severe acute respiratory distress syndrome and severe infection. The all-cause mortality rate was also higher in this group (43.7% in the COVID-19 group versus 27.4% in the non-COVID-19 group). The diagnosis of COVID-19 was a predictor of intensive care unit mortality (odds ratio, 2.77; 95%CI, 1.89 – 4.07; p < 0.001), regardless of age or Charlson Comorbidity Index score.

Conclusion:

In a prospective cohort of patients admitted with acute respiratory failure, patients with COVID-19 had a clearly different phenotype and a higher mortality than non-COVID-19 patients. This may help to outline more accurate screening and appropriate and timely treatment for these patients.