Introduction:
In 40-70% of out-of-hospital cardiac arrest (OHCA) cases, chest compressions (CCs) during CPR induce measurable oscillations in capnography (E
T
CO
2
). Recent studies suggest the magnitude and frequency of oscillations are due to intrathoracic airflow dependent on airway patency. These oscillations can be quantified by the Airway Opening Index (AOI), ranging from 0-100%. We sought to develop, automate, and evaluate multiple methods of computing AOI throughout CPR.
Methods:
We conducted a retrospective study of all OHCA cases in Seattle, WA during 2019. E
T
CO
2
and impedance waveforms from LifePak 15 defibrillators were annotated for the presence of intubation and CPR, and imported into MATLAB for analysis. Four proposed methods for computing AOI were developed (Fig. 1) using peak E
T
CO
2
in conjunction with ΔE
T
CO
2
(oscillations in E
T
CO
2
from CCs). We examined the feasibility of automating ΔE
T
CO
2
and AOI calculation during CCs throughout OHCA resuscitation and evaluated differences in mean AOI using each method. Statistical significance was assessed with ANOVA (alpha = 0.05).
Results:
AOI was measurable in 312 of 465 cases. Mean [95% confidence interval] AOI across all cases was 34.3% [32.0-36.5%] for method 1, 27.6% [25.5-29.7%] for method 2, 22.7% [21.1-24.3%] for method 3, and 28.8% [26.6-31.0%] for method 4. Mean AOI was significantly different across the four methods (p<0.001), with the greatest difference between method 1 and 3 (11.6%, p<0.001), but no significant difference between methods 2 and 4 (p=0.44). Mean ΔE
T
CO
2
was 7.76 [7.08-8.44] mmHg.
Conclusion:
We implemented four proposed methods of automatically calculating AOI during OHCA. Each method produced a different average AOI. Consistent, automated methods to measure AOI provide the foundation to evaluate if, and how, AOI may change with treatment or predict outcomes. These four approaches require additional investigation to understand which may be best suited to improve OHCA care.
Letter by Rottenberg Regarding Article “Improved Neurological Outcome With Continuous Chest Compressions Compared With 30:2 Compressions-to-Ventilations Cardiopulmonary Resuscitation in a Realistic Swine Model of Out-of-Hospital Cardiac Arrest”
