ABSTRACT
BACKGROUND: The coronavirus disease 2019 (COVID-19) pandemic has put significant pressure on hospitals and in particular on intensive care units (ICU). Some patients develop acute hypoxemic respiratory failure with profound hypoxia, which likely requires invasive mechanical ventilation during prolonged periods. Corticosteroids have become a cornerstone therapy for patients with severe COVID-19, though only little data are available regarding their potential harms and benefits, especially concerning the risk of a ventilator-associated lower respiratory tract infection (VA-LRTI). METHODS: This retrospective multicenter study included patients admitted in four ICUs from Belgium and France for severe COVID-19, who required invasive mechanical ventilation (MV). We compared clinical and demographic variables between patients that received corticosteroids or not, using univariate, multivariate, and Fine and Gray analyses to identify factors influencing VA-LRTI occurrence. RESULTS: From March 2020 to January 2021, 341 patients required MV for acute respiratory failure related to COVID-19, 322 of whom were included in the analysis, with 60.6% of them receiving corticosteroids. The proportion of VA-LRTI was significantly higher in the early corticosteroid group (63.1% vs. 48.8%, p = 0.011). Multivariable Fine and Gray modeling considering death and extubation as competing events revealed that the factors independently associated with VA-LRTI occurrence were male gender (adjusted sHR:1.7, p = 0.0022) and corticosteroids (adjusted sHR: 1.44, p = 0.022). CONCLUSIONS: in our multicenter retrospective cohort of COVID-19 patients undergoing MV, early corticosteroid therapy was independently associated with VA-LRTI.