Integrating the existing emergency medical system with automated external defibrillators for out-of-hospital cardiac arrest

We investigated whether a program based on diffuse deployment of automated external defibrillators operated by trained volunteers and laypersons across the largest county in Italy would safely and effectively improve the current survival rate among patients with out-of-hospital cardiac arrest. Compared with historical control subjects, the new strategy resulted in a 3-fold increase in 1-year survival free of neurologic impairment.

Several years ago, automated external defibrillators (AEDs) were made available to law enforcement staff, firefighters, and emergency medical technicians in Italy's largest county, Brescia, which comprises 1.6% of the country's total area and has a population of 1,112, 628. Initial studies on defibrillation in public access areas have shown promising results.1-4 There are limited data, however, on outcomes when the entire defibrillation capability of urban and rural areas relies on laypersons and volunteers.

Subjects and methods

We performed the Brescia Early Defibrillation Study (BEDS) to determine whether survival among patients with out-of-hospital cardiac arrest was improved with the widespread use of AEDs operated by laypersons and trained volunteers in Brescia, Italy. The emergency medical service (EMS), which is coordinated by a single dispatch center, has served the region since 1992. For 9 years before the start of the study, defibrillation was performed solely in the 5 medically equipped ambulances and with manual defibrillators in the emergency departments of the 10 county hospitals. In January 2000, the use of AEDs by laypersons and trained volunteers was established. Automated external defibrillator instruction consisted of 5 hours of training in theory, practice, and basic life support, and was performed by qualified teachers. The director of the emergency medical department provided certification for use of AEDs to all qualified individuals.

Consecutive patients who experienced out-of-hospital cardiac arrest were included in the study. We documented patient data obtained at the location of the cardiac arrest, throughout hospitalization, and during follow-up on standardized forms and recorded it in a central database.


Between July 2000 and June 2002, 702 subjects were enrolled in the study prospectively. As shown in Table 1*, witnessed events, sex, percentage of outdoor events, and events in which cardiopulmonary resuscitation was performed by a witness were similar between the 692 subjects enrolled in the historical group (recruited between June 1997 and May 1999) and the prospectively studied group. No significant differences between the 2 groups were shown in the partial response time, prevalence of causes of cardiac arrest, and therapies delivered.

After implementing the AED program, the rate for 1-year survival free of neurologic impairment increased significantly, from 0.9% to 3.0% overall, from 0.5% to 2.5% in the rural areas, and from 1.4% to 4.0% in the urban areas. Among dispatched cardiac arrests, the larger percentage of survivors in the urban areas was correlated with a larger population per square kilometer covered per deployed device, a shorter response time, and a larger percentage of witnessed events, compared with the rural areas. There were 3 times as many additional survivors in the urban areas (1.5 per 100,000) as in the rural areas (0.5 per 100,000). Table 2* shows a comparison between the historical and prospective groups categorized by territory.

For patients who received defibrillation within 8 minutes of collapse, the 1-year survival rate was 12.5%, compared with those who received defibrillation 8 or more minutes after collapse (1-year survival rate, 1.7%). However, at the 1-year follow-up, those who received defibrillation 8 or more minutes after collapse were responsible for 30.6% of all survivors free of neurologic impairment.

Implementation of the AED program cost €1,017 ,514 initially and €681,766 for maintenance of the system. For each quality-adjusted life year saved, further costs included €39,388 during start-up and €23,661 during maintenance of the system.


The implementation of emergency layperson-integrated service can be employed chiefly in areas such as Italy or in less-developed regions with budget, logistic, or legislative restrictions regarding the use of AEDs. It may also be applied to more organized systems as well, including in areas where results of present EMS systems are unsatisfactory. Survival rates may even increase in areas that already have better outcomes than those shown in the present study.

Although the survival rate of 4.1% shown in the BEDS is not exceptional, it is similar to the rates shown in some studies in which more-structured EMS systems supported by paramedics were evaluated.5-24 One of the main reasons for this low survival rate is probably because the response time was > 17 minutes after collapse, which is not typical of most structured environments. But this rate signifies a helpful reference for less-structured EMS systems and provides hope for further improvement in response time in the areas evaluated.

In our study, 77% of subjects with witnessed cardiac arrest received defibrillation > 8 minutes after collapse. Optimal standards for early defibrillation programs require that at least 90% of patients receive defibrillation within 8 minutes of collapse. In our study, however, about one third of subjects who received defibrillation later than 8 minutes after collapse had 1-year survival free of neurologic impairment. This result highlights the need for aggressive resuscitation in patients who undergo late defibrillation and the need to reassess the optimal standards for early defibrillation programs.


In the BEDS, we evaluated whether an AED program operated by trained volunteers and laypersons would safely and effectively improve current survival rates among subjects with out-of-hospital cardiac arrest. Results showed that there was a 3-fold increase in 1-year survival free of neurologic impairment compared with historical control subjects.

*Readers wishing to see the tables for this paper should consult the print version.

Related Videos
Muthiah Vaduganathan, MD, MPH | Credit: Brigham and Women's Hospital
Viet Le, PA-C
Clyde Yancy, MD
Steve Nissen, MD
Related Content
© 2023 MJH Life Sciences

All rights reserved.