Decision Support System to Evaluate VENTilation in ARDS

The study is a randomised control trial, comparing the effects of mechanical ventilation in ARDS provided as standard care, with mechanical ventilation set according to the advice of the Beacon Care system. The study will compare if the use of the system results in a better application of PEEP and driving pressure across all severities and phases of ARDS.

The broad timeframe for the study is as follows. The study will be conducted at three sites. Two of these sites are outside the UK, one in France and one in Italy.

The systems will be delivered and installed starting September/October 2019 in the UK site, followed by France and Austria in November 2019. A one month period of training will then follow where all doctors and nurses involved in using the system will be trained. During this period none of the systems advice will be applied. This training will be supported by study partners, and nurses trained as super users of the system, by Mermaid Care A/S, the company producing the system, and by project investigators who have previously conducted studies using the system.

Patient inclusion and data collection will be performed at in the UK sites over a 15 month period from start October 2019 to January 2020. In both European sites, inclusion and data collection will be performed over a 12 month period from start January 2020 to December 2020. Final data analysis and report writing will be performed by March 2021.

For all study sites, all adult patients diagnosed with ARDS in the cardio-thoracic intensive care unit (ICU) undergoing mechanical ventilation will be considered to see if they are eligible for the study.

Patients will only be considered eligible if they are Within 7 days of an acute and potentially reversible episode of acute respiratory failure, their chest radiograph with bilateral infiltrates is consistent with evidence of pulmonary oedema but not fully explained by cardiac failure and they present with hypoxaemia. In addition, patients are not considered eligible due to being aged under 18, not having an arterial catheter for blood sampling at study start, consent declined, over 7 days of mechanical ventilation, treatment withdrawal imminent within 24 hours, Do Not Attempt Resuscitation order in place and for having a severe chronic respiratory disease requiring domiciliary ventilatioin. For each patient, if identified as eligible, advice to participate in the study will be sought.

Critically ill patients are and often unconscious and may not be able to grant consent. Therefore, the Patient Informed Sheet (PIS) and Consent Form will be requested from a third party acting as a consultee; in most cases this person will be a Personal Consultee, who is someone who knows the person lacking capacity and is able to advise the researcher about that person's wishes and feelings in relation to the project and whether they should join the research. This person must be interested in the welfare of the patient in a personal capacity, not in a professional capacity or for remuneration and will mostly likely be family member, carer or friend, etc. Where the personal consultee is not available on site, the researcher may contact the personal consultee by telephone and seek verbal advice. The verbal agreement will be recorded in the telephone consultee declaration form. The telephone consultee declaration form will be signed by a second member of staff who has witnessed the telephone advice. A copy of the PIS will be emailed to the personal consultee.

Where no Personal Consultee is available, the researcher will nominate a professional person to assist in determining the participation of a person who lacks capacity. A Nominated Professional Consultee is someone who will be appointed by the researcher to advise the researcher about the person's (who lacks capacity) wishes and feelings in relation to the project and whether they should join the research. An independent clinician not treating the patient will be asked to be the nominated consultee. A patient information sheet will be distributed immediately following the patient being identified as eligible for the study.

The study investigators will ask the patient if they can use their existing hospital records. Without their consent, no additional information about the patient will be collected for the purposes of the study. However, to maintain integrity of the randomised trial, all information collected up to that time will still be used and analysed as part of the study.

If consent is obtained and the patient is still eligible, then they will be randomly allocated to one of two groups. This random allocation will be performed according to the randomisation template with random block allocation.

If the patient is allocated to the standard care group then all care will be according to usual practice. If the patient is allocated to the Beacon group, mechanical ventilation will be set with advice from the system. In either case, the study ends for the individual patient either with successful transfer of the patient to another department, or patient death. All other care will be according to usual practice.

110 patients will be included in the study, with 55 patients in each of the groups. 55 patients per group will allow to detect a difference of 3cmH2O in driving pressure between the groups with 90% power and a two sided alpha of 0.05 assuming a control group driving pressure of 15 cmH2O with a standard deviation of 2.5 cmH2O and including a 40% dropout. The investigators have used data from the MIMIC dataset (as published in Serpo Neto et al) for the estimation of the driving pressure. In view of the longitudinal analysis, loss to follow up has taken account of an average mortality of 34% and a 6% drop out.