BackgroundMechanical chest compression devices consistently deliver high-quality chest compressions. Small very low-quality studies suggest mechanical devices may be effective as an alternative to manual chest compressions in the treatment of adult in-hospital cardiac arrest patients. The aim of this feasibility trial is to assess the feasibility of conducting an effectiveness trial in this patient population.MethodsCOMPRESS-RCT is a multi-centre parallel group feasibility randomised controlled trial, designed to assess the feasibility of undertaking an effectiveness to compare the effect of mechanical chest compressions with manual chest compressions on 30-day survival following in-hospital cardiac arrest.Over approximately two years, 330 adult patients who sustain an in-hospital cardiac arrest and are in a non-shockable rhythm will be randomised in a 3:1 ratio to receive ongoing treatment with a mechanical chest compression device (LUCAS 2/3, Jolife AB/Stryker, Lund, Sweden) or continued manual chest compressions. It is intended that recruitment will occur on a 24/7 basis by the clinical cardiac arrest team. The primary study outcome is the proportion of eligible participants randomised in the study during site operational recruitment hours. Participants will be enrolled using a model of deferred consent, with consent for follow-up sought from patients or their consultee in those that survive the cardiac arrest event.The trial will have an embedded qualitative study, in which we will conduct semi-structured interviews with hospital staff to explore facilitators and barriers to study recruitment.DiscussionThe findings of COMPRESS-RCT will provide important information about the deliverability of an effectiveness trial to evaluate the effect on 30-day mortality of routine use of mechanical chest compression devices in adult in-hospital cardiac arrest patients.Trial registrationISRCTN38139840, date of registration 9th January 2017.Electronic supplementary materialThe online version of this article (10.1186/s13049-018-0538-6) contains supplementary material, which is available to authorized users.