Trial ISRCTN 16912075
Publication Perkins G, RECOVERY-RS, 2021 (preprint)
Dates: 4/6/2020 to 5/3/2021
Funding: Public/non profit (UK National Institute for Health Research)
Conflict of interest: Yes

Methods
RCT
Blinding: Unblinded
Location : Multicenter / UK
Follow-up duration (days): 30
Inclusion criteriaAdult (≥18-years); hospitalized patients with known or suspected COVID-19; acute respiratory failure, defined as peripheral oxygen saturations (SpO2) of 94% or below despite receiving a fraction of inspired oxygen (FiO2) of at least 0.4, and were deemed suitable for tracheal intubation if treatment escalation was required
Exclusion criteriaImmediate (<1-hour) need for invasive ventilation; known pregnancy; planned withdrawal of treatment; a contraindication to an intervention, based on the judgement of the treating clinician, precluded randomization to that trial arm
Interventions
Treatment 1
Continuous positive airway pressure
Control
Conventional oxygen therapy
Treatment 3
High-flow nasal oxygenation
Participants
Randomized
1272 participants n1=380/ n2=475/ n3=417
Characteristics of participants
N=1272
Mean age : 57.3
844 males
Severity : Mild: n=0 / Moderate: n=0/ Severe: n=1272 Critical: n=0
Primary outcome
In the register
Composite outcome comprising tracheal intubation or mortality within 30 days.
In the report
NR
Documents
available

Protocol

Yes. In English

Statistical plan

Yes

Data-sharing stated

NR
Risk of bias
Overall
The overall risk of bias reported in the table corresponds to the highest risk of bias for the outcomes assessed for the systematic review
Some concerns
General comment In addition to the pre-print article, the study registry and protocol were used in data extraction and risk of bias assessment. There were no substantive differences between the pre-print article and the registry and protocol in population, procedures, interventions and outcomes. This trial was a part of the RECOVERY platform trial and had an adaptive design that allowed the study to stop early if one or both interventions were more effective than the control group. The study did not achieve the planned sample size and stopped recruitment due to a rapid decline in UK COVID-19 case numbers, and the need to share accumulated data to inform international treatment of COVID-19 patients. Clinical outcomes were reported in pairwise comparisons in populations defined by intervention availability at time of randomization (CPAP vs standard randomized when only CPAP alone or both CPAP + high-flow were available as experimental arm interventions; HFNO versus standard when only HFNO or both HFNO + CPAP were available as experimental arm interventions). Clinical outcomes were not extracted pending author contact for 3-arm data.