Comparative performance assessment of commercially available automatic external defibrillators: A simulation and real-life measurement study of hands-off time. (January 2017)

• Record Type: Journal Article
• Authors: Savastano, Simone; Vanni, Vincenzo; Burkart, Roman; Raimondi, Maurizio; Canevari, Fabrizio; Molinari, Simone; Baldi, Enrico; Danza, Aurora I.; Caputo, Maria Luce; Mauri, Romano; Regoli, Francois; Conte, Giulio; Benvenuti, Claudio; Auricchio, Angelo
• Is Part Of: Resuscitation. Volume 110(2017)
• Page Start: 12
• Page End: 17
• Publication Date: 2017-01
• Subjects: AED -- CPR -- Resuscitation -- Hands-off time; Resuscitation -- Periodicals; Resuscitation -- Periodicals; Réanimation -- Périodiques; Electronic journals; 616.025
• DOI: 10.1016/j.resuscitation.2016.10.006
• Languages: English
• ISSNs: 0300-9572

Abstract

Abstract: Purpose: Early and good quality cardiopulmonary resuscitation (CPR) and the use of automated external defibrillators (AEDs) improve cardiac arrest patients' survival. However, AED peri- and post-shock/analysis pauses may reduce CPR effectiveness. Methods: The time performance of 12 different commercially available AEDs was tested in a manikin based scenario; then the AEDs recordings from the same tested models following the clinical use both in Pavia and Ticino were analyzed to evaluate the post-shock and post-analysis time. Results: None of the AEDs was able to complete the analysis and to charge the capacitors in less than 10 s and the mean post-shock pause was 6.7 ± 2.4 s. For non-shockable rhythms, the mean analysis time was 10.3 ± 2 s and the mean post-analysis time was 6.2 ± 2.2 s. We analyzed 154 AED records [104 by Emergency Medical Service (EMS) rescuers; 50 by lay rescuers]. EMS rescuers were faster in resuming CPR than lay rescuers [5.3 s (95%CI 5–5.7) vs 8.6 s (95%CI 7.3–10). Conclusions: AEDs showed different performances that may reduce CPR quality mostly for those rescuers following AED instructions. Both technological improvements and better lay rescuers training might be needed.