Can real-time feedback improve the simulated infant cardiopulmonary resuscitation performance of basic life support and lay rescuers?

Background Performing high-quality chest compressions during cardiopulmonary resuscitation (CPR) requires achieving of a target depth, release force, rate and duty cycle. Objective This study evaluates whether 'real time' feedback could improve infant CPR performance in basic life support-trained (BLS) and lay rescuers. It also investigates whether delivering rescue breaths hinders performing high-quality chest compressions. Also, this study reports raw data from the two methods used to calculate duty cycle performance. Methodology BLS (n=28) and lay (n=38) rescuers were randomly allocated to respective 'feedback' or 'no-feedback' groups, to perform two-thumb chest compressions on an instrumented infant manikin. Chest compression performance was then investigated across three compression algorithms (compression only; five rescue breaths then compression only; five rescue breaths then 15: 2 compressions). Two different routes to calculate duty cycle were also investigated, due to conflicting instruction in the literature. Results No-feedback BLS and lay groups demonstrated <3% compliance against each performance target. The feedback rescuers produced 20-fold and 10-fold increases in BLS and lay cohorts, respectively, achieving all targets concurrently in >60% and >25% of all chest compressions, across all three algorithms. Performing rescue breaths did not impede chest compression quality. Conclusions A feedback system has great potential to improve infant CPR performance, especially in cohorts that have an underlying understanding of the technique. The addition of rescue breaths-a potential distraction-did not negatively influence chest compression quality. Duty cycle performance depended on the calculation method, meaning there is an urgent requirement to agree a single measure.