Abstract Body: Introduction: During cardiopulmonary resuscitation (CPR), continuous chest compressions cause retrograde movement of gases during the inspiratory and expiratory phases of ventilation, termed reversed airflow (RF). Quantifying the effect of RF on ventilation quality is challenging due to the lack of data in prehospital settings.
Aim: Analyze the incidence of RF and its effect on ventilation volume and waveform capnography during manual prehospital CPR.
Methods: We retrospectively analyzed prehospital monitor-defibrillator recordings with additional mainstream waveform capnography and spirometry-based airflow. For each episode, ventilations were automatically annotated using the waveform capnography signal. The onset and offset of each ventilation were manually identified. We then applied a custom-made algorithm to estimate the volume from the airflow reliably. Ventilations during chest compressions were identified using the transthoracic impedance. RF events were manually classified as either negative or positive airflow, corresponding to the inspiratory or expiratory phases of ventilation, respectively. For each RF event, associated volume was computed and waveform capnography change was annotated.
Results: Of the twenty-five patients examined, airflow was recorded during chest compressions in seven of them (28%) with 719 ventilations affected by chest compressions. Of these, 471 (65%) had RF events with a median (IQR) of 2 (1-3) events per ventilation. Out of a total of 1116 RF events, 875 (78%) occurred during the inspiratory phase and 241 (22%) during the expiratory phase, with a median volume change of 8.1 (3.1-11.5) ml and 1.8 (0.9-2.8) ml, respectively (p<0.001). RF events with impact on the waveform capnography were 124 (11%) with a median CO₂ change of 3.3 (0.9-10.5) mmHg and 117 of them (94%) during the inspiratory phase.
Conclusion: Reversed airflow frequently occurs in two-thirds of breaths when ventilations and compressions are administered at the same time and more likely during inspiration. The associated volumes were small but also greater during inspiration. Occasionally, RF was observed in the waveform capnography as peaks in the baseline. Further research is warranted to better understand RF and waveform capnography interactions.