Experience with the use of automated external defibrillators in out of hospital cardiac arrest

Aim: To describe the sequences of arrhythmias, number of shocks delivered and the number of failures in a consecutive series of patients with out-of-hospital cardiac arrest attended by our emergency medical service (EMS) and in whom cardio-pulmonary resuscitation (CPR) was initiated and in whom automated external defibrillators (AEDs) were used. Patients: All patients with out-of-hospital cardiac arrest attended by the EMS and in whom AEDs were used. Time for inclusion in the study: January Ist, 1987 to December 31st, 1992. Results: In all there were 1781 out of hospital cardiac arrests during the study period. Among them AEDs were used in 383 cases (22%). The total number of interpreted rhythms delivered in these patients was 2719. Among all rhythm sequences coarse ventricular fibrillation (VF) was found on 375 occasions (14%); fine VF on 107 occasions (4%) and ventricular tachycardia (VT) on 12 occasions (0.4%). In ten cases with coarse VF (nine patients) the AED did not advise a shock (2.7%). In five of those nine patients a human error was interpreted as the explanation and in four there was a possible technical error. In these four patients defibrillation was delayed by 33-43 s, respectively. Among the 2225 rhythm sequences not judged as VF/VT the AED advised a shock on one occasion (0.04%). Conclusion: Among patients with coarse VF AED gave inaccurate instructions in 2.7%. However, the majority of the failures were judged to be caused by human errors. (C) 1998 Elsevier Science Ireland Ltd. All rights reserved.