Ventricular Fibrillation Waveform Analysis During Chest Compressions to Predict Survival From Cardiac Arrest

Abstract

Background: Quantitative measures of the ventricular fibrillation (VF) electrocardiogram (ECG) waveform can assess myocardial physiology and predict cardiac arrest outcomes, making these measures a candidate to help guide resuscitation. Chest compressions are typically paused for waveform measure calculation, as compressions cause ECG artifact. However, such pauses contradict resuscitation guideline recommendations to minimize CPR interruptions. We evaluated a comprehensive group of VF measures with and without ongoing compressions to determine their performance under both conditions for predicting functionally-intact survival, the study’s primary outcome. Methods: Five-second VF ECG segments were collected with and without chest compressions prior to 2755 defibrillation shocks from 1151 out-of-hospital cardiac arrest patients. Twenty-four individual measures and three combination measures were implemented. Measures were optimized to predict functionally-intact survival (Cerebral Performance Category score ≤ 2) using 460 training cases, and their performance evaluated using 691 independent test cases. Results: Measures predicted functionally-intact survival on test data with an area under the receiver operating characteristic curve (AUC) ranging from 0.56-0.75 (median=0.73) without chest compressions and from 0.53-0.75 (median=0.69) with compressions (p<0.001 for difference). Of all measures evaluated, the support vector machine model ranked highest both without chest compressions (AUC=0.75, 95% CI 0.73-0.78) and with compressions (AUC=0.75, 95% CI 0.72-0.78) (p=0.75 for difference). Conclusions: VF waveform measures predict functionally-intact survival when calculated during chest compressions, but prognostic performance is generally reduced compared to compression-free analysis. However, support vector machine models exhibited similar performance with and without compressions while also achieving the highest AUC. Such machine learning models may therefore offer means to guide resuscitation during uninterrupted CPR.

Adult Advanced Life Support: 2020 International Consensus on Cardiopulmonary Resuscitation and Emergency Cardiovascular Care Science with Treatment Recommendations

“…The evidence updates restricted the search to January 2008 to January 2020 and identified one large RCT conducted in 2013 45 and 20 observational studies. 46 , 47 , 48 , 49 , 50 , 51 , 52 , 53 , 54 , 55 , 56 , 57 , 58 , 59 , 60 , 61 , 62 , 63 , 64 , 65 In addition, there is an ongoing multicentre RCT of real-time amplitude spectrum area to guide defibrillation (NCT03237910). Although the VF waveform analyses and outcomes studied were highly heterogeneous, given the amount of data available, an updated SysRev was suggested.…”

Discovering hidden information in biosignals from patients using artificial intelligence

“…Another example is the use of ECG rhythms recorded by automated external defibrillators during chest compressions in 1151 cases of cardiac arrest [22]. In the study, the AI model could predict a functionally intact survival rate after acute cardiac arrest.…”

Estimating the amplitude spectrum area of ventricular fibrillation during cardiopulmonary resuscitation using only ECG waveform

“…Both animal and clinical studies have demonstrated that quantitative VF signal analysis can be used as a non-invasive tool to optimize the timing of defibrillation, and allow CPR to be tailored to each individual heart (8,9). Amplitude spectrum area (AMSA), as one of the most efficient predictors for successful defibrillation (9)(10)(11), has been shown to be correlated with coronary perfusion pressure, and can reflect the energy state of the myocardium (12,13). Typically, the highest value of AMSA is frequently observed at the onset of VF, and declines as time elapses without treatment.…”

“…However, reliable AMSA can only be obtained during the interval of CC pause because CC-produced artifacts may lead to overestimation of AMSA and reduction of predictive performance (11,19). Suppressing the artifact using additional reference recordings, such as transthoracic impedance or acceleration signals, may be a possible solution (20,21).…”