Résumé
Introduction: Hypoxaemic acute respiratory failure (ARF) in intensive care unit (ICU) patients is associated with high mortality. Three main devices are used to provide oxygen to hypoxaemic ARF patients: non-invasive ventilation (NIV), high-flow nasal cannula oxygen (HFNO) and standard oxygen (first-attempt device in usual care). To date, no multicentre randomised controlled study has compared NIV and HFNO to standard oxygen with day 28 mortality rate as primary outcome in hypoxaemic ARF in non-selected patients. Our hypothesis is that NIV and/or HFNO is superior to standard oxygen to reduce day 28 mortality rate in hypoxaemic ARF. Methods and analysis: The Key oxygenation Interventions in Surgical and non-Surgical patients (KISS) trial is an adaptive investigator-initiated, multicentre, stratified, parallel-group unblinded trial with an electronic system-based randomisation. Patients with hypoxaemic ARF were randomly assigned to one of three groups: the 'NIV-group' to receive curative NIV combined with HFNO delivered between NIV trials, or the 'HFNO' group to receive HFNO alone, or the 'standard oxygen-group' to receive oxygen therapy alone.The primary endpoint is day 28 all-cause mortality. The main secondary endpoint is intubation rate at day 28. The exploratory endpoints are intubation rates at day 3 and day 7; oxygenation up to day 7; need for other rescue oxygen therapy up to day 7; ICU and hospital length of stay; and mortality rates in ICU, hospital and at day 90.The main objective is to assess if NIV and/or HFNO is superior to standard oxygen to reduce day 28 mortality rate in hypoxaemic ARF. Additional comparisons between predefined stratum following randomisation will be performed: (1) medical versus postoperative admissions, (2) among medical (immunocompromised vs non-immunocompromised) and (3) among postoperative (abdominal vs cardio-thoracic).An adaptive design will be used. Two interim analyses will be performed after 700 and 1400 included patients among the 2100 planned. Ethics and dissemination: The study project has been approved by the appropriate ethics committee 'Comité-de-Protection-des-Personnes Sud-Est V-23-CHUM-01 Cat2 2022-A02761-42/1'. Informed consent is required. The results will be submitted for publication in a peer-reviewed journal and presented at one or more scientific conferences. If NIV and/or HFNO reduce the mortality at day 28, NIV and/or HFNO could be proposed to become one of the first-line therapies in hypoxaemic ARF patients. Trial registration number: NCT05812911.