Predictive Factors for Intubation in Coronavirus Disease Patients Admitted in the ICU

Abstract

Object: Since it began in Wuhan in December 2019, the Coronavirus Disease pandemic has affected more than 500 million people and caused more than 6 million deaths. Identifying risk factors for severe cases has become a major issue. We evaluated whether patient characteristics upon intensive care unit admission could predict later intubation. We also compared outcomes for patients undergoing early versus delayed intubation. Methods: This is a retrospective, monocentric study carried out in a medical university intensive care unit between August 2020 and January 2021. Demographic, clinical, biological and imaging data were collected (on arrival and on day 2). We examined intubation timing (before or after 48h hours after intensive care unit admission), ventilatory features and outcomes for intubated patients. Results: SAPS2, high steroid dosages, pulmonary superinfection, extensive CT pulmonary lesions, polypnea and elevated oxygen requirements were associated with a higher need of intubation. Biological features on admission were non-discriminatory. Delayed intubation seemed to be associated with more severe acute respiratory distress syndrome, but mortality did not vary. Discussion and conclusion: Intubation can be predicted using a multimodal approach including clinical and imaging features. Early clinical evaluation plays a key role in identifying patients likely to be intubated. Delaying intubation could lead to respiratory worsening.