This site is optimised for modern browsers. For the best experience, please use Google Chrome, Mozilla Firefox, or Microsoft Edge.

West Midlands-led RECOVERY-RS trial finds continuous positive airway pressure (CPAP) reduces need for invasive ventilation in hospitalised COVID-19 patients

Landmark UK trial compared three commonly used respiratory interventions to establish which works best for COVID-19 patients with acute respiratory failure

Participants who received continuous positive airway pressure (CPAP) were less likely to require invasive mechanical ventilation from COVID-19

Researchers found no benefit from high flow nasal oxygenation (HFNO) over standard oxygen therapy

Based on this evidence, the authors say CPAP should be considered for hospitalised patients with COVID-19 needing increasing oxygen - reducing the need for invasive ventilation and relieving pressure on intensive care services

The Respiratory Strategies in COVID-19; CPAP, High-flow, and Standard Care (RECOVERY-RS) trial has demonstrated that treating hospitalised COVID-19 patients who have acute respiratory failure with continuous positive airway pressure (CPAP) reduces the need for invasive mechanical ventilation.

Preliminary data from the trial also suggest that the routine use of high flow nasal oxygenation (HFNO), which can consume large amounts of oxygen, should be reconsidered as it did not improve outcomes for COVID-19 patients compared with conventional oxygen therapy.

RECOVERY-RS, led by the University of Warwick and Queen’s University Belfast, is the world’s largest non-invasive respiratory support trial for COVID-19 - with over 1200 participants taking part across 48 UK hospitals. The multi-centre, adaptive, randomised controlled trial compared the use of CPAP (oxygen and positive pressure delivered via a tightly fitting mask), with HFNO (high pressure oxygen delivered up the nose), against standard care (standard oxygen therapy).

All three interventions are commonly used to treat COVID-19 patients before they are moved onto invasive ventilation in a critical care bed, but it was not known which, if any, resulted in better outcomes.


Over 13 months, between April 2020 and May 2021, a total of 1,272 hospitalised COVID-19 patients with acute respiratory failure, aged over the age of 18, were recruited to the study and randomly allocated to receive one of three respiratory support interventions as part of their hospital care.

380 (29.9%) participants received CPAP; 417 (32.8%) participants received HFNO; and 475 (37.3%) received conventional oxygen therapy.

The primary outcomes assessed through the trial were whether the patient went on to require tracheal intubation (invasive mechanical ventilation) or died within 30-days of beginning treatment through the trial.

In the comparison of CPAP and conventional oxygen therapy, the likelihood of patients going on to require invasive mechanical ventilation or die within 30-days of treatment was significantly lower in those who were treated with CPAP, than those who received standard care. In the CPAP group, 137 of 377 participants (36.3%) either needed mechanical ventilation or died within 30 days, compared with 158 of 356 participants (44.4%) in the conventional oxygen therapy group.

There was no difference in primary outcomes between patients in the HFNO and conventional oxygen therapy groups. In the HFNO group, 184 of 414 participants (44.4%) went on to require mechanical ventilation or die, compared with 166 of 368 participants (45.1%) in the conventional oxygen therapy group.

Based on these results, one person would avoid needing invasive ventilation within intensive care units (ICU) for every 12 people treated with CPAP instead of standard oxygen therapy.

Professor Gavin Perkins, Chief Investigator and Professor in Critical Care Medicine at Warwick Medical School at the University of Warwick said: “The RECOVERY-RS trial showed that CPAP was effective at reducing the need for invasive ventilation, thus reducing pressures on critical care beds. The routine use of high flow nasal oxygenation, which can consume large amounts of oxygen, should be reconsidered as it did not improve outcomes. By giving patients the most effective treatment to begin with, we can help prevent resource shortages in our NHS and make sure the right type of ventilation is available to patients when it is required.

“This is the first large trial of different types of ventilation in COVID-19. While it is encouraging that these results can help reduce the number of people who require invasive ventilation, it is important to stress that, where it is needed, invasive ventilation can be lifesaving.”

Professor Danny McAuley, Chief Investigator and Professor and Consultant in Intensive Care Medicine at the Royal Victoria Hospital and Queen's University Belfast said: "Over the COVID pandemic, we've seen a large number of patients requiring high levels of oxygen and admission to ICU for invasive ventilation, causing a huge strain on staff and beds.

"The results of this trial are really encouraging as they have shown that by using CPAP, invasive ventilation may not be needed for many patients with COVID-19 requiring high oxygen levels. Avoiding invasive ventilation is not only better for the patients, but it also has important resource implications as it frees up ICU capacity.

“This research should help healthcare professionals in the UK and beyond manage patients with COVID-19, to improve patient outcomes while helping to lessen the burden on resources."

Patient case studies

Patient case study 1 - Lisa’s story

Lisa Broadhurst, 42, from Northfield, Birmingham, took part in the RECOVERY-RS trial in January after contracting Covid-19.

She was rushed to hospital by ambulance on 13 January with breathing problems and dropping oxygen levels, 11 days after first experiencing symptoms of aches and pains, loss of taste and smell and a severe headache.

Diagnosed with severe Covid-19 and pneumonitis, Lisa, who is also asthmatic, said: “I’ve never been so scared in my life. I had no control of my deteriorating health. I remember Face Timing my family and letting them know how poorly I was. All I could think about was whether I’d go home or whether covid would kill me like so many other patients. The nurses would squeeze my hand and reassure me to stay strong. It was a heartbreaking time and I wouldn't wish that feeling on anyone.”

Lisa, who received CPAP as part of the trial, remembers being connected to the machine: “The pressure of the oxygen took a little while to get used to. It was a scary experience and took my breath away at first. All I could think about was making it home to my family. I fought as hard as I could to stay alive, I cried a lot and I was scared and overwhelmed.

“But throughout the time on the machine I could see I was getting better. It was the best feeling ever leaving the hospital and walking outside into the fresh air knowing I was going home to my loved ones.”

Covid-19 still has a lasting impact on her health and her lungs are much more sensitive.

But she added: “I’m extremely grateful I took part in the trial - the level of care I received from everyone was amazing. The nurses and doctors showed so much support. I couldn’t have asked for better and taking part in the trial saved my life. I will always forever be in debt to the NHS - they helped me to go back home to my family. Research really does help benefit others and it’s the reason I’m alive today.”

Patient case study 2 - Phil’s story

Phil Cawkwell, age 38, from Erdington took part in the RECOVERY-RS Covid-19 trial in 2020. He knew only too well what some patients go through when they are hospitalised with Covid-19 because he is a member of the Anaesthetics Team at Good Hope Hospital in Sutton Coldfield.

Following three weeks spent intubating* patients with the disease during March 2020, he started to feel unwell and tested positive for COVID-19. A week after, he woke up at home, where he lives alone, unable to breathe. He said: “I felt dizzy and scared and I was gasping for air’. An ambulance was called and he was given oxygen by the paramedics.

“I was petrified on my way to hospital and convinced I was going to die. The paramedics kept calling out my observations and as I am medically trained I knew just what they meant. I had a terrible feeling of doom and kept thinking about my three kids.

“It was truly traumatic - of course I knew I would be a candidate for intubation when I got to hospital and I really didn’t want to be taking up resource in the Intensive Care Unit, so I was only too pleased to agree to take part in the RECOVERY-RS trial, as I knew I would be randomised to one of three different treatments.”

“I am incredibly grateful I was offered the chance to take part in the trial. It’s a fantastic opportunity and gives you hope at a dark time to think you might be helping other people. I have always been interested in research and I would say to anyone who gets the chance - just do it. We have to go forward with medical interventions and I would definitely think about taking part in further trials in future.”

Phil has made a very slow recovery and is now back at work after 11 months off sick, having suffered from significant pneumonia and problems with his heart.

“Even in October, I had fatigue and breathlessness but in the last three months I am back to riding my bike and can walk a mile without having to stop. I can’t emphasise enough how grateful I am to have had the opportunity to take part in RECOVERY-RS. Research has been so valuable in the fight against Covid - and of course I have had my vaccine. There’s no reason not to.”

Professor Jonathan Van-Tam, Deputy Chief Medical Officer said: “COVID-19 has placed huge pressure on our hospitals and intensive care units, and our doctors, nurses and all NHS staff have stepped up to meet that challenge. A key part of tackling COVID has been the improvements that staff have identified and then implemented in terms of how to best care for COVID patients.

“This study, funded by the NIHR, provides valuable evidence around how non-invasive respiratory support can be used to improve patient outcomes.
“Reducing invasive mechanical ventilation is better for patients and reduces pressures on mechanical ventilator capacity across the NHS.

“I want to thank the team of doctors, researchers and patient volunteers involved in today’s excellent results – hospitals across the country can now use these data to further improve care for patients and reduce the demand for mechanical ventilation as we get closer to what might still be a challenging winter period.”

Professor Lucy Chappell, Chief Scientific Adviser (CSA) for the DHSC and the National Institute for Health Research (NIHR) Chief Executive Officer, said: “Research such as this has been a huge asset to the COVID-19 response, allowing us to fine-tune our approach and improve care for patients in hospital.

“I am hugely grateful to the teams at the University of Warwick and Queen’s University Belfast for their contribution to our understanding of the virus through this NIHR-funded study, and particularly how to treat it.

“This data will help ensure hospitalised patients with COVID-19 get the best possible care, making a difference to patients and intensive care units across the country.”

Professor Nick Lemoine, Medical Director at the National Institute for Health Research (NIHR) Clinical Research Network said: “Preliminary results from this NIHR-supported trial provide important evidence which will help shape clinical practice worldwide around respiratory support interventions for hospitalised COVID-19 patients. The study will undoubtedly help improve outcomes for patients - while potentially alleviating pressure on hospital beds and critical care services.

“We sincerely want to thank everybody involved - the patients who took part in their darkest hour, and the NHS doctors and nurses who helped deliver the study right across the UK.”

Professor Simon Ball, Executive Medical Officer at University Hospitals Birmingham said: "This is an important study that will significantly influence treatment decisions. It is an example of how well NHS hospitals can deliver studies to improve clinical practice. This includes the definition of treatments that are beneficial, in this case CPAP, but just as importantly those with no apparent benefit, in this case high flow nasal oxygen. The best possible care we deliver is that focused by evidence.”


Both CPAP and HFNO have been widely used worldwide in the management of COVID-19 throughout the pandemic for patients who need high levels of additional oxygen. If these treatments are not successful, patients need to be sedated and treated with a ventilator in intensive care.
Although both CPAP and HFNO are commonly used in other lung conditions, prior to the RECOVERY-RS study, it was unknown how safe and effective they were for people with breathing difficulties arising from COVID-19.

The trial is led by Joint Chief Investigators Professor Gavin Perkins at the University of Warwick, and Professor Danny McAuley at Queen's University Belfast. It was funded and supported by the National Institute for Health Research (NIHR) as a prioritised urgent public health COVID-19 study.

RECOVERY-RS was one of the first COVID-19 studies to be classed as urgent public health research by the UK’s Chief Medical Officers in order to urgently identify strategies to reduce the need for invasive mechanical ventilation. Launched in April 2020 as COVID-19 hospitalisation began to soar, the NIHR Clinical Research Network provided prioritised research delivery support to rapidly set the study up at hospital sites across the UK and enroll participants. The NIHR’s research infrastructure, expertise and delivery support has been critical to the trial’s success.

The preliminary results of this evaluation of the data will be available as a pre-print on medRxiv on 5 August 2021 and will be submitted to a leading peer-reviewed medical journal. The results will be presented in detail at a free virtual Critical Care Reviews conference session ( on Thursday 5 August at 7.30pm.


Notes to editors

For further information or interviews with the chief investigators or trial participants, please contact Ben Lingard at the National Institute for Health Research: or or call 020 3328 6730.

Full details of the study protocol and related materials.

Available for interview:
Co-Chief Investigator: Prof Gavin Perkins, University of Warwick
Co-Chief Investigator: Prof Danny McAuley, Queen’s University Belfast
Trial participant: Lisa Broadhurst (via video call / telephone)
Trial participant: Phil Cawkwell (via video call / telephone)

Photographs of the Chief Investigators and participant case studies are available.

RECOVERY-RS trial: Background and supporting information

COVID-19 causes people to have breathing difficulties and many patients in hospital need respiratory support - with 76% of hospitalised patients requiring supplemental oxygen and 9% requiring tracheal intubation and invasive mechanical ventilation. Early in the pandemic, international experiences highlighted the potential risk that intensive care units (ICUs) might become overwhelmed by rising case numbers and the high mortality in patients that required invasive mechanical ventilation. This drove an urgent public health need to identify strategies to reduce the need for invasive mechanical ventilation.

The RECOVERY-RS trial compares three different ways of providing respiratory support to patients suspected or confirmed of having COVID-19 to establish which work best. All three methods are currently in use in clinical practice and patients received one of these, chosen at random. Primary outcomes assessed which treatment is more effective in relation to survival of patients and the need for intubation (tube inserted into a patient's throat to help them breathe). Secondary outcomes assessed how long it takes to intubate a patient, time to death, survival rate in critical care, and the length of stay, both in critical care and overall over a 30 day period, or until the patient leaves hospital.

Primary outcomes

Participant characteristics were similar at baseline (tables 1 and 2). The mean age was 57.4 (95% CI, 56.7 to 58.1) years, 66.4% were male, and 65.3% of white ethnicity.

In the comparison of CPAP and conventional oxygen therapy, the need for tracheal intubation or mortality at 30-days was significantly lower in the CPAP group (CPAP 137 of 377 participants (36.3%) vs conventional oxygen therapy 158 of 356 participants (44.4%); unadjusted odds ratio 0.72; 95% CI 0.53 to 0.96, P=0.03).

There was no difference in primary outcomes between the HFNO and conventional oxygen therapy groups (HFNO 184 of 414 participants (44.4%) vs conventional oxygen therapy 166 of 368 participants (45.1%); unadjusted odds ratio 0.97; 95% CI 0.73 to 1.29, P=0.85).

In conclusion, CPAP, compared with conventional oxygen therapy, reduced the composite outcome of intubation or death within 30 days of randomisation in hospitalised adults with acute respiratory failure due to COVID-19. There was no beneficial effect observed, compared with conventional oxygen therapy, with the use of HFNO.

NIHR Clinical Research Network support

Throughout the set up and delivery of the RECOVERY-RS study, the NIHR Clinical Research Network (CRN) supported the study team to identify new potential sites, while maintaining strong communication with study Principal Investigators through its 15 Local Clinical Research Networks (LCRNs) to provide support, feedback, and advice on study delivery.

This invaluable support ensured the research could be quickly embedded in hospital sites across the NIHR’s network, while the development and dissemination of promotional materials boosted the profile of the study nationally.