Glucocorticoids in Adults With Acute Respiratory Distress Syndrome: Randomised Clinical Trial (GuARDS Trial)

Research question: What is the clinical and cost effectiveness of dexamethasone given early (within 72 hours of diagnosis) in acute respiratory distress syndrome (ARDS)?

Design: Multi-centre, parallel group, allocation concealed, open label, pragmatic, group sequential design randomised controlled trial, with internal pilot Setting: Approximately 65 UK intensive care units (ICUs) Target population: Adult ICU patients with moderate-to-severe ARDS (a) Provision of informed consent

(b) Aged 16-years or older (c) Admitted to intensive care unit or high dependency unit (ICU)

(d) Receiving respiratory support via invasive mechanical ventilation or non-invasive ventilatory support (non-invasive ventilatory support includes mask or helmet) or high flow nasal cannula (HFNC) >30L/min (e) Within 72 hours of diagnosis of ARDS with moderate to severe hypoxaemia defined as

i) a known acute clinical insult or new or worsening respiratory dysfunction (Note: this includes new deterioration at any time-point during the ICU stay), and ii) opacities on chest imaging not fully explained by effusions, lobar/lung collapse/atelectasis, or nodules, and iii) respiratory failure not fully explained by cardiac failure or fluid overload, and

iv) assessment of hypoxaemia done with either PaO2/FiO2 ratio <26.7 kPa from arterial blood gases, or SpO2/FiO2 <235 with SaO2<97% Exclusion criteria: (a) ARDS due to microbiologically confirmed SARS-Co-V2 infection (COVID-19 ARDS)

(b) Major upper gastrointestinal bleeding during current hospital admission, defined as requiring endoscopy and transfusion for two or more units of packed red blood cells. This exclusion criterion will exclude patients with contraindications to the glucocorticoids on the grounds of safety.

(c) High dose glucocorticoids are required for a separate proven clinical indication at the time of randomisation as withholding treatments that have been deemed clinically effective, would be unethical. (d) Known hypersensitivity to dexamethasone (e) Infections that are not being effectively treated as determined by the treating medical team.

(f) Planned intensive care treatment withdrawal within next 24 hours as determined by the treating medical team (g) Patients who are known to be pregnant (h) Previous enrolment in the GuARDS trial

Health technologies being assessed: Intravenous dexamethasone 20mg OD day 1 to day 5, reduced to 10 mg OD day 6 to day 10. Measurement of costs and outcomes Primary: All-cause mortality 60 days post-randomisation Secondary:

-In hospital: Successful extubation; Duration of mechanical ventilation; ICU, and hospital length of stay; and adverse events. -At 60-days: HRQoL; Health service use -At 90-days: HRQoL; Health service use; Mortality -At 180-days: Reintubation; Mortality; HRQoL; Health service use

Health economic evaluation: Cost-effectiveness from NHS and personal social services perspective as per NICE reference case specifications. Follow up: Up to 180 days post-randomisation. Sample size: 854 per group (1708 in total).

We estimate duration of recruitment to achieve a sample size of 1708 will be 42 months (at 0.8 participants/site/month). We will undertake an internal pilot for the first 9 months of recruitment, to set-up all 60 sites and recruit 268 patients. Participants will be followed-up for 6 months and further 6-months for close-down, data-cleaning, analysis and reporting. The total study duration is 60 months.

Expertise: Our multi-professional team includes clinical trialists, leaders in critical care, pharmacists, methodologists, statistician, health economist, and family representatives, to optimise all aspects of trial delivery.

Anticipated impact and dissemination: The results of this definitive trial will establish best clinical care where there is uncertainty, and will determine best value for services. Dissemination will take multiple routes including publication in international, high impact factor, journals, presentation at relevant national and international conferences, and communication with involved participant and family networks including social media and lay press.