Mortality rate, intensive care unit length of stay and time to orotracheal intubation of COVID-19 patients under different non-invasive ventilatory therapies: retrospective cohort study

ABSTRACT

Introduction: COVID-19 may lead to heterogeneous needs for ventilator therapy, whether oxygen therapy (OT), noninvasive ventilation (NIV), high-flow nasal catheter (HFNC) or their combination (NIV + HFNC). The purpose of the study was to describe, retrospectively, the mortality rate, intensive care unit length of stay (ICU-LOS) and time to orotracheal intubation of COVID-19 patients under OT, NIV, HFNC or combined (NIV + HFNC). A retrospective cohort study was done analyzing official medical data from March 2020 up to July 2021. (CAAE 52534221.5.0000.5249). Method(s) The inclusion criteria were age > 18 years-old, and positive swab test for COVID-19 or computed tomography consistent of COVID-19. The exclusion criteria were hospital LOS less than 3 days, patients whose therapy (OT, NIV, HFNC or NIV + HFNC) lasted less than 48 h, and missing data about the outcome variables. The primary outcome was mortality rate, while secondary outcomes were ICU-LOS and time to orotracheal intubation. Chi-Square test was used to assess mortality rate. The Mann-Whitney U test was applied to assess differences in ICU-LOS and time to orotracheal intubation (p < 0.05). Result(s) Overall, 1371 patients were enrolled. 880, 120, 35, and 148 patients were submitted to OT, NIV, HFNC or NIV + HFNC, respectively. The mortality rates were 8.4%, 29.6%, 22.2%, and 33.2% for OT, NIV, HFNC or NIV + HFNC, respectively (p < 0.001). The ICU-LOS was higher in NIV + HFNC (median [IQR] 15 days [16]) than NIV (9 days [10]) and OT (4 days [5], p < 0.001). The time to orotracheal intubation was higher in NIV (6 days [6]), HFNC (6 days [4.5]), and NIV + HFNC (6 days [6]) than OT (2 days [4]), p < 0.001. Mortality rate and ICU-LOS were higher in those patients requiring the combination of NIV and HFNC. Conclusion(s) Although the type of ventilator therapy may be associated to increased mortality rate and ICU-LOS, we cannot assure causality due to exploratory nature of the retrospective study, but a marker of severity.