Study in New England Journal of Medicine Shows Community Training With AEDs Doubles Survival

08/11/04    Portland, Ore.

OHSU emergency medicine team led this study locally

Emergency medicine researchers at Oregon Health & Science University led Oregon's study of public access to defibrillation, which is featured in today's New England Journal of Medicine. The Public Access to Defibrillation (PAD) trial found that twice as many lives can be saved when properly trained volunteers perform cardiopulmonary resuscitation (CPR) and use automated external defibrillators (AEDs) after someone has suffered cardiac arrest. The study showed 30 people survived in the CPR-plus-AED group and 15 people survived in the CPR-only group. The study's results are based on 239 attempted resuscitations across the study sites.

OHSU was one of 24 community-based research centers in the United States and Canada that coordinated significant community participation and support for the PAD trial. More than 19,000 volunteer responders from 993 community locations participated.

"This study demonstrates that lay persons, without a duty to respond, can use AEDs successfully and effectively to improve outcomes from out-of-hospital cardiac arrests in public places," said Mohamud Daya, M.D., OHSU's principal investigator in the study and associate professor of emergency medicine, OHSU School of Medicine. "Given the dismal outcomes from out-of-hospital cardiac arrests nationally, it will be important to build on the results of this study and create more volunteer responder-based AED programs in workplaces and other public locations."

Numerous cases of AEDs being used to save lives in airports and other public places have been reported. However, this is the first randomized, controlled study to scientifically determine whether nonmedical volunteers can be trained to perform CPR and use AEDs effectively and if they can save more lives than volunteers who just perform CPR. In both circumstances, early activation of the local emergency medical services (EMS) was a part of the study design. AEDs in this study were only approved for use in adults with cardiac arrest.

Twice as many people who suffered cardiac arrest survived when the volunteer called 911, performed CPR and used an AED. However, this does not decrease the value of performing CPR when there is rapid local EMS response because a significant number of people survived with this treatment alone. Indeed, the two treatments are complementary since CPR provides blood flow to the heart and the brain until the heart rhythm can be restored by a defibrillator.

The PAD trial study coordinators provided two to four hours of initial training to lay volunteers, then retrained them one or more times during the study. An important aspect of the study was the presence of nonmedical volunteers who came forward to be trained and respond to a medical emergency at their work sites. As part of the study, Daya and his team coordinated the training of more than 1,500 community volunteers at 80 sites in the Portland-metropolitan area, Salem and Vancouver. Half of the sites received training in CPR, while the other half received training in AED use, as well as training in CPR. At the end of the study, all sites were offered AEDs donated by Philips Medical Systems and training in their use.

Community participation and EMS support in Oregon and southwest Washington were key to the success of the study. "Our site volunteers and private and public EMS agencies made the study possible," Daya said.

The study also was unique because it required the community's notification and consultation to conduct the study before any research and AED use could begin. This was required because patients who are in cardiac arrest cannot consent before entering a study.

Researchers used the PAD data to determine two other key findings. The study found that volunteers at CPR-plus-AED sites were more likely to perform CPR than volunteers at sites without AEDs. This suggests that either people are more motivated to help others when they have the support of an AED or the voice commands given by the AED help guide the resuscitation. Researchers also found that trained volunteers could perform CPR and use an AED without hurting themselves or the patient.

"Knowing that AEDs can be used by properly trained volunteers in a variety of workplace environments is an important first step to introducing these devices in additional public settings. Indeed, the FDA is considering whether to approve the sale of these devices to the public," said Jerris Hedges, M.D., professor and chairman of emergency medicine, OHSU School of Medicine, co-principal investigator and an author of safety and CPR-use sub-studies.

The two-year study was funded by the National Heart, Lung and Blood Institute, the American Heart Association, Cardiac Science/Survivalink, Medtronic Physio-Control, Philips Medical Systems, Laerdal Medical Corp. and Guidant Corp.

More than 250,000 deaths from "out-of-hospital" cardiac arrests occur each year in the United States, and half of them occur suddenly. Most sudden cardiac arrests are caused when the heart stops beating effectively due to an abnormal heart rhythm. AEDs are able to quickly jump-start the heart back into a normal rhythm with an electrical shock. The time it takes from the cardiac arrest until the patient receives a rhythm-stabilizing electrical shock can be the difference between life and death. Nationally, the survival rate from sudden cardiac arrest is 5 percent, and only 20 percent of arrests occur in public locations. The NHLBI is funding a multi-center, 7,000-patient study designed to evaluate whether providing the devices to families of heart attack patients will improve survival if a cardiac arrest occurs in that person's home. This is an important next step because 80 percent of cardiac arrests occur in the home.

Other members of OHSU's PAD study team include: Jonathan Jui, M.D.; Terri Schmidt, M.D.; Lynn Wittwer, M.D. (Southwest Washington Medical Center); Maggie Gunnels, Ph.D.; Heather Brooks; Denise Griffith; and Chris Burke.