Background: This study aimed to explore the clinical practice of noninvasive respiratory support (NIRS) in COVID-19 patients, and assess its clinical outcomes, safety, and prognostic factors.
Methods: A multicenter, retrospective observational study was conducted with consecutive adult COVID-19 patients who underwent NIRS at nine Japanese hospitals between January 1 and September 30, 2021. NIRS modalities included high-flow nasal cannula oxygen therapy (HFNC), noninvasive ventilation (NIV), and continuous positive airway pressure (CPAP). NIRS failure, defined as intubation or in-hospital death, intubation rate, in-hospital mortality, and the incidence of SARS-CoV-2 infection in health care workers (HCWs) were evaluated. The respiratory rate-oxygenation (ROX) index was examined as a predictor of NIRS failure.
Results: We included 694 patients with a median age of 61 years, of whom 516 (74.4 %) were men. HFNC was the first-line respiratory support for 622 (89.6 %) patients, followed by CPAP in 54 (7.8 %) and NIV in 18 (2.6 %). The NIRS failure rate was 41.5 %, with an intubation rate of 32.0 % and in-hospital mortality of 18.4 %. The ROX index at baseline and 1 and 24 h after NIRS initiation were independent predictors of NIRS failure (odds ratio [95 % CI] 0.89 [0.85-0.94], p < 0.001; 0.84 [0.78-0.89], p < 0.001; 0.84 [0.78-0.90], p < 0.001, respectively). Nine possible NIRS-related HCW infections were reported.
Conclusion: The majority of COVID-19 patients in this study were treated with HFNC, with a NIRS failure rate of about 40 %. The ROX index within 24 h effectively predicted NIRS failure. HCW infections were infrequent, indicating the relative safety of NIRS.
Keywords: COVID-19; Continuous positive airway pressure; High-flow nasal cannula oxygen therapy; Japan; Noninvasive ventilation.
Copyright © 2025. Published by Elsevier B.V.