Validation of Physiologic CPR Quality Using NOn-inVasive Waveform Analytics (CPR-NOVA)

点击翻译按钮获取中文摘要

Project Abstract Pediatric cardiac arrest affects thousands of children each year. Progressive heart and lung failure is a predisposing cause in many of these events. Despite improvements in survival outcomes over the past two decades, more than half of these children still do not survive. As new brain injury complicates care among survivors, the burden to these children and the public's health is substantial. Cardiopulmonary resuscitation (CPR) – the medical procedure of providing chest compressions and ventilations during cardiac arrest – saves lives, and higher quality CPR is more effective at doing so. One method to improve CPR quality is through the use of CPR quality monitoring defibrillators. By providing real- time feedback on CPR mechanics targets such as chest compression depth and rate, they represent the best patient care option currently available to improve CPR performance. Unfortunately, most of these devices are either not approved for children or use pads that are too large for many pediatric patients. Thus, current technology limits the benefit of meaningful CPR quality monitoring to a small percentage of the children who suffer a cardiac arrest. Given the strong association between pediatric CPR quality and outcomes, new methods to monitor CPR quality are urgently needed to improve the care of this vulnerable population. Physiologic-directed CPR is a promising technique that uses the hemodynamic response of the patient to guide the ongoing resuscitation effort. This approach overcomes the technological limitations of existing CPR quality monitoring technology by using data from patient monitors. Unfortunately, because many patients do not have intra-arterial lines in place at the time of arrest to guide CPR, its clinical impact has been limited. To overcome this limitation, the objective of this ancillary application is to leverage the existing infrastructure of the National Institute of Child Health and Human Development-funded Collaborative Pediatric Critical Care Research Network (CPCCRN) and the unique hemodynamic waveform database of the National Heart, Lung, and Blood Institute-funded parent R01 – the ICU-Resuscitation (ICU-RESUS) Project – to validate two noninvasive physiologic CPR monitors applicable to nearly every pediatric cardiac arrest: 1) end-tidal carbon dioxide (ETCO2); and 2) PhotoPlethysmoGraphy (PPG) obtained via pulse oximetry. Using sophisticated machine learning methods, a prospective observational analytic investigation is proposed with the following Aims: 1) Evaluate ETCO2 as a noninvasive CPR quality monitor among children receiving at least 1 minute of CPR in a CPCCRN intensive care unit; and 2) Using novel machine learning classification algorithms, evaluate PPG and other candidate physiologic waveforms as noninvasive CPR quality monitors. By leveraging the substantial infrastructure of the CPCCRN, the novel hemodynamic waveform database of ICU-RESUS, and advanced machine learning analytics, this submission represents a unique opportunity to validate two noninvasive physiologic pediatric CPR quality monitors to improve clinical care and save lives. !