Ventricular Fibrillation Waveform Changes during Controlled Coronary Perfusion Using Extracorporeal Circulation in a Swine Model

Abstract

Background: Several characteristics of the ventricular fibrillation (VF) waveform have been found predictive of successful defibrillation and hypothesized to reflect the myocardial energy state. In an open-chest swine model of VF, we modeled "average CPR" using extracorporeal circulation (ECC) and assessed the time course of coronary blood flow, myocardial metabolism, and myocardial structure in relation to the amplitude spectral area (AMSA) of the VF waveform without artifacts related to chest compression.

Methods: VF was induced and left untreated for 8 minutes in 16 swine. ECC was then started adjusting its flow to maintain a coronary perfusion pressure of 10 mmHg for 10 minutes. AMSA was calculated in the frequency domain and analyzed continuously with a 2.1 s timeframe and a Tukey window that moved ahead every 0.5 s.

Results: AMSA progressively declined during untreated VF. With ECC, AMSA increased from 7.0 ± 1.9 mV·Hz (at minute 8) to 12.8 ± 3.3 mV·Hz (at minute 14) (p < 0.05) without subsequent increase and showing a modest correlation with coronary blood flow of borderline statistical significance (r = 0.489, p = 0.0547). Myocardial energy measurements showed marked reduction in phosphocreatine and moderate reduction in ATP with increases in ADP, AMP, and adenosine along with myocardial lactate, all indicative of ischemia. Yet, ischemia did not resolve during ECC despite a coronary blood flow of ~ 30% of baseline.

Conclusion: AMSA increased upon return of coronary blood flow during ECC. However, the maximal level was reached after ~ 6 minutes without further change. The significance of the findings for determining the optimal timing for delivering an electrical shock during resuscitation from VF remains to be further explored.

Fig 1. Effects of ventricular fibrillation and simulated resuscitation using extracorporeal circulation (EC) on the ventricular fibrillation amplitude-spectral area (AMSA) and left ventricular (LV) volume and wall thickness.

The EC flow was first titrated to maintain a coronary perfusion pressure of 10 mmHg (not shown) from minute 9 to 18 (Low-Flow EC) and then increased during the defibrillation effort and the post-resuscitation interval to generate a higher (H) mean aortic pressure of 40 mmHg (H-EC). Arrows signal the time of the electrical shocks.

Fig 2. Myocardial measurements at baseline (BL) and during ventricular fibrillation (VF), at 10 and 16 minutes while on low-flow extracorporeal circulation simulating the hemodynamic conditions of closed-chest CPR.

DO₂, oxygen delivery; VO₂, oxygen consumption; GCV, great cardiac vein. Mean ± SEM of 16 experiments. Data was analyzed by one-way repeated measures ANOVA and differences shown after Holm-Sidak method for multiple pairwise comparisons. a vs BL, b vs VF10; *p ≤ 0.05, †p ≤ 0.001.

Fig 3. Myocardial measurements at baseline (BL) and during ventricular fibrillation (VF), at 10 and 16 minutes while on low-flow extracorporeal circulation simulating the hemodynamic conditions of closed-chest CPR.

pCr, phosphocreatine; ATP, adenosine triphosphate; ADP, adenosine diphosphate; AMP, adenosine monophosphate; Ado, adenosine. Data was analyzed by one-way repeated measures ANOVA and differences shown after Holm-Sidak method for multiple pairwise comparisons. a vs BL, b vs VF10; *p ≤ 0.05, †p ≤ 0.001.

Fig 4. Correlations between AMSA measured 1 minute after induction of VF (VF-1 min) and measurements after 8 minutes of untreated VF (VF-8 min) and at 11, 14, and 17 minutes while ECC was on.