Abstract
Non-invasive respiratory support (NIRS) has increasingly been used in the management of COVID-19-associated acute respiratory failure, but questions remain about the utility, safety, and outcome benefit of NIRS strategies. We identified two randomised controlled trials and 83 observational studies, compromising 13 931 patients, that examined the effects of NIRS modalities-high-flow nasal oxygen, continuous positive airway pressure, and bilevel positive airway pressure-on patients with COVID-19. Of 5120 patients who were candidates for full treatment escalation, 1880 (37%) progressed to invasive mechanical ventilation and 3658 of 4669 (78%) survived to study end. Survival was 30% among the 1050 patients for whom NIRS was the stated ceiling of treatment. The two randomised controlled trials indicate superiority of non-invasive ventilation over high-flow nasal oxygen in reducing the need for intubation. Reported complication rates were low. Overall, the studies indicate that NIRS in patients with COVID-19 is safe, improves resource utilisation, and might be associated with better outcomes. To guide clinical decision making, prospective, randomised studies are needed to address timing of intervention, optimal use of NIRS modalities-alone or in combination-and validation of tools such as oxygenation indices, response to a trial of NIRS, and inflammatory markers as predictors of treatment success.