Novel CPR System That Predicts Return of Spontaneous Circulation from Amplitude Spectral Area Before Electric Shock in Ventricular Fibrillation

• Published in: Resuscitation 2017-04, Vol.113, p.8-12
• Main Authors: Nakagawa, Yoshihide, Amino, Mari, Inokuchi, Sadaki, Hayashi, Satoshi, Wakabayashi, Tsutomu, Noda, Tatsuya
• Format: Article
• Language: English
• Subjects: Adult; Aged; Amplitude spectral area; Area Under Curve; Blood Circulation; Cardiopulmonary Resuscitation - instrumentation; Cardiopulmonary Resuscitation - methods; Cardiopulmonary Resuscitation - standards; Defibrillation; Electric Countershock - adverse effects; Electric Countershock - methods; Electric Countershock - statistics & numerical data; Electrocardiography - methods; Emergency; Emergency Medical Services - methods; Emergency Medical Services - statistics & numerical data; Female; Humans; Japan - epidemiology; Male; Middle Aged; Out-of-hospital cardiac arrest; Predictive Value of Tests; Prognosis; Reproducibility of Results; ROC Curve; Time Factors; Ventricular fibrillation; Ventricular Fibrillation - diagnosis; Ventricular Fibrillation - epidemiology; Ventricular Fibrillation - physiopathology; Ventricular Fibrillation - therapy; Waveform analysis
• ISSN: 0300-9572; 1873-1570
• DOI: 10.1016/j.resuscitation.2016.12.025

Abstract Aim Amplitude spectral area (AMSA), an index for analysing ventricular fibrillation (VF) waveforms, is thought to predict the return of spontaneous circulation (ROSC) after electric shocks, but its validity is unconfirmed. We developed an equation to predict ROSC, where the change in AMSA (ΔAMSA) is added to AMSA measured immediately before the first shock (AMSA1). We examine the validity of this equation by comparing it with the conventional AMSA1-only equation. Method We retrospectively investigated 285 VF patients given prehospital electric shocks by emergency medical services. ΔAMSA was calculated by subtracting AMSA1 from last AMSA immediately before the last prehospital electric shock. Multivariate logistic regression analysis was performed using post-shock ROSC as a dependent variable. Results Analysis data were subjected to receiver operating characteristic curve analysis, goodness-of-fit testing using a likelihood ratio test, and the bootstrap method. AMSA1 (odds ratio (OR) 1.151, 95% confidence interval (CI) 1.086–1.220) and ΔAMSA (OR 1.289, 95% CI 1.156–1.438) were independent factors influencing ROSC induction by electric shock. Area under the curve (AUC) for predicting ROSC was 0.851 for AMSA1-only and 0.891 for AMSA1 + ΔAMSA. Compared with the AMSA1-only equation, the AMSA1 + ΔAMSA equation had significantly better goodness-of-fit (likelihood ratio test P < .001) and showed good fit in the bootstrap method. Conclusions Post-shock ROSC was accurately predicted by adding ΔAMSA to AMSA1. AMSA-based ROSC prediction enables application of electric shock to only those patients with high probability of ROSC, instead of interrupting chest compressions and delivering unnecessary shocks to patients with low probability of ROSC.