Department of Communications Engineering, University of the Basque Country, UPV/EHU, 48013 Bilbao, Spain.
Department of Communications Engineering, University of the Basque Country, UPV/EHU, 48013 Bilbao, Spain.
Resuscitation. 2018 Mar;124:63-68. doi: 10.1016/j.resuscitation.2017.12.013. Epub 2017 Dec 12.
Capnography has been proposed as a method for monitoring the ventilation rate during cardiopulmonary resuscitation (CPR). A high incidence (above 70%) of capnograms distorted by chest compression induced oscillations has been previously reported in out-of-hospital (OOH) CPR. The aim of the study was to better characterize the chest compression artefact and to evaluate its influence on the performance of a capnogram-based ventilation detector during OOH CPR.
Data from the MRx monitor-defibrillator were extracted from OOH cardiac arrest episodes. For each episode, presence of chest compression artefact was annotated in the capnogram. Concurrent compression depth and transthoracic impedance signals were used to identify chest compressions and to annotate ventilations, respectively. We designed a capnogram-based ventilation detection algorithm and tested its performance with clean and distorted episodes.
Data were collected from 232 episodes comprising 52 654 ventilations, with a mean (±SD) of 227 (±118) per episode. Overall, 42% of the capnograms were distorted. Presence of chest compression artefact degraded algorithm performance in terms of ventilation detection, estimation of ventilation rate, and the ability to detect hyperventilation.
Capnogram-based ventilation detection during CPR using our algorithm was compromised by the presence of chest compression artefact. In particular, artefact spanning from the plateau to the baseline strongly degraded ventilation detection, and caused a high number of false hyperventilation alarms. Further research is needed to reduce the impact of chest compression artefact on capnographic ventilation monitoring.
呼气末二氧化碳监测（capnography）已被提议作为监测心肺复苏（CPR）期间通气率的一种方法。先前在院外（OOH）CPR 中已经报道过，由于胸部按压引起的振荡而导致的呼末二氧化碳描记图失真的发生率很高（高于 70%）。本研究的目的是更好地描述胸部按压伪影，并评估其对基于呼末二氧化碳描记图的通气探测器在 OOH CPR 中的性能的影响。
从 MRx 监护除颤器中提取 OOH 心搏骤停事件的数据。对于每个事件，在呼末二氧化碳描记图中注释了胸部按压伪影的存在。使用同步的压缩深度和经胸阻抗信号分别识别胸部按压和通气。我们设计了一种基于呼末二氧化碳描记图的通气检测算法，并使用干净和失真的事件对其性能进行了测试。
数据来自 232 个事件，共包含 52654 次通气，每个事件的平均（±SD）为 227（±118）次。总体而言，42%的呼末二氧化碳描记图存在失真。胸部按压伪影的存在降低了算法在通气检测、通气率估计和检测过度通气方面的性能。
使用我们的算法在 CPR 期间基于呼末二氧化碳描记图的通气检测受到胸部按压伪影的影响。特别是，从平台到基线的伪影强烈降低了通气检测的性能，并导致大量的假过度通气警报。需要进一步的研究来减少胸部按压伪影对呼末二氧化碳监测的影响。
