Optimizing timing of ventricular defibrillation

Objective: Our intent was to evolve a prognosticator that would predict the likelihood that an electrical shock would restore a perfusing rhythm. Such a prognosticator was to be based on conventional electrocardiographic signals but without constraints caused by artifacts resulting from precordial compression. The adverse effects of "hands off" intervals for rhythm analyses would therefore be minimized. Such a prognosticator was further intended to reduce the number of electrical shocks and the total energy delivered and thereby minimize postresuscitation myocardial dysfunction. Design: Observational study. Subjects. Medical research laboratory of a university-affiliated research and educational institute. Subjects: Domestic pigs. Interventions. Ventricular fibrillation was induced in an established porcine model of cardiac arrest. Recordings of scalar lead 2 over the frequency range of 4-48 Hz were utilized. The area under the curve representing the amplitude and frequency was defined as the amplitude spectrum area (AMSA). Measurements and Main Results. A derivation group of 55 animals yielded a threshold value of AMSA that uniformly predicted successful resuscitation. A separate group of 10 animals, a validation group, confirmed that an AMSA value of 21 mV-Hz predicted restoration of perfusing rhythm after 7 of 8 electrical shocks and failure of electrical conversion in 21 of 23 electrical shocks, yielding sensitivity and specificity of about 90%. The negative predictive value of AMSA was 95% and statistically equivalent to that of coronary perfusion pressure, mean amplitude, and median frequency. The positive predictive value that would prompt continuation of cardiopulmonary resuscitation without interruption for an unsuccessful defibrillation attempt was greatly improved with AMSA (78%) as compared with coronary perfusion pressure (42%), mean amplitude (32%), and median frequency (29%). Conclusion: AMSA has the potential for guiding more optimal timing of defibrillation without adverse interruption of cardiopulmonary resuscitation or the delivery of unsuccessful high energy electrical shocks that contribute to postresuscitation myocardial injury.