Report on AED Failures: A Reminder that Maintenance is Vital

WASHINGTON-- More than 1,000 cardiac arrest deaths over 15 years are connected to the failure of automated external defibrillators (AEDs); battery failure accounted for almost one-quarter of the failures. The study was published online last week in Annals of Emergency Medicine ("Analysis of Automated External Defibrillator Device Failures Reported to the Food and Drug Administration" http://bit.ly/ox6YYr).

"Survival from cardiac arrest depends on the reliable operation of AEDs," said lead study author Lawrence DeLuca, MD, EdD, of the University of Arizona Department of Emergency Medicine in Tucson. "AEDs can truly be lifesavers but only if they are in good working order and people are willing to use them."

Researchers analyzed reports to the Food and Drug Administration (FDA) about all adverse events connected to use of an AED between January 1993 and October 2008. Of the 40,787 AED-related events reported to the FDA, 1,150 adverse events connected to fatalities were reported (3%). Almost half (45 percent) of failures occurred during the attempt to charge and deliver a recommended shock to the person in cardiac arrest. Problems with pads and connectors accounted for 23.7 percent of the failures and battery power problems accounted for 23.2 percent of the failures.

Sudden cardiac arrest is a leading cause of death in North America and Europe. Odds of survival decline by 7 to 10 percent per minute of delay in defibrillation. Even as AEDs have proliferated in public places such as airports and offices, bystanders are reluctant to use them. An Annals of Emergency Medicine study published earlier this year found that less than half of people in public places reported being willing to use an AED and more than half were unable to recognize one.

"AEDs are like any other piece of medical equipment: They can experience unexpected failures," said Dr. Deluca. "I would recommend that people maintain AEDs as recommended by the manufacturer. If an unexpected device failure occurs it is vitally important to promptly contact both the manufacturer and the FDA. Then be sure to return the unit (and accessories such as pads or batteries) to the manufacturer immediately so that it can be analyzed and a cause for the failure identified and fixed."

Read the abstract here.

SOURCE: American College of Emergency Physicians/ PR Newswire

What You Need to Know About “AED Failures"

A Commentary by the Sudden Cardiac Arrest Foundation

AEDs save lives when they are properly maintained and used quickly by bystanders

A new study, “Analysis of Automated External Defibrillator (AED) Device Failures Reported to the Food and Drug Administration (FDA),” published online in the Annals of Emergency Medicine, requires attention and careful analysis.[i] It should also serve as a reminder to program coordinators that device maintenance is fundamental to the success of every AED program.

Researchers at the Arizona Emergency Medicine Research Center, the University of Arizona Department of Medicine, and the Christus Spohn-Texas A&M Emergency Medicine Residency conducted a retrospective analysis of the FDA’s Manufacturer and User Device Experience (MAUDE) database, a post-market surveillance system that collects voluntary “reports on adverse events involving medical devices.”[ii] Their report is summarized here.

The DeLuca, et al. study captured a total of 40,787 “AED related events” reported to the FDA over a 15-year period (June 1993-October 2008). Their analysis indicates there were 1,150 “AED related event” reports associated with an unsuccessful resuscitation. Although the term “failure” is used in the title of the study and elsewhere, 39,563 of “AED related events” apparently involved successful resuscitations.

It is important to note that the study does not indicate that the 1,150 unsuccessful resuscitations were caused by the “AED related event.” Association does not imply causation.

There are several other key considerations:

• Based on other studies, about 20 percent of sudden cardiac arrest (SCA) victims cannot be treated effectively with AEDs, even if the devices are working perfectly.[iii] It is unclear whether any of the 1,150 “AED related events” associated with unsuccessful resuscitations involved SCA patients who were unlikely to have benefited from treatment with an AED.
• Whether the devices documented in this study were properly maintained is also unclear. Only 1.5% of adverse event reports included records of a maintenance program or schedule. So it is impossible to know whether “AED failures” occurred in well-maintained devices—or devices that were inadequately maintained.
• Low battery messages were considered device “failures,” though in reality, such messages indicate the devices worked successfully by alerting users to the need for device maintenance. (Note to self: Low battery messages should be addressed as soon as they occur—not when the devices are needed for an emergency.)
• The study does not address the total number of AED uses, since the MAUDE database includes only voluntary, self-reported information. Thus, it is impossible to know the true incidence of adverse events.

Interestingly, there was a dramatic downward trend in the number of reports of adverse events during the study period.[iv] This occurred as device sales were increasing dramatically. More than 2 million AEDs have been sold in North America since their debut in 1995, including automated devices used by professional rescuers. [v] More than a half million units were sold during the last three years of the study (2006-2008) alone.[vi]

“Over the past five years, we have seen the expansion of the automated external defibrillator segment for ‘Public Access’,” according to Frost & Sullivan, a global market research and consulting firm.

“This study actually suggests that AEDs are safe—and getting safer all the time,” according Nadine Levick, MD, MPH, Director of the EMS Safety Foundation in New York City, and principal investigator of the iRescU mobile device application.[vii]

When interpreting the data, it’s important to consider study period itself (1993-2008). While AEDs were developed in the 1980s, they were deployed by Emergency Medical Services (EMS), fire departments, and first responder organizations. These early units were large, heavy, and complicated to use. The FDA did not clear small, lightweight AEDs intended for public access until 1995 and their deployment did not begin until 1996. The American Heart Association first advocated bystander use of AEDs in 1995.[viii]

Another consideration is the research design. According to the FDA, “MAUDE data is not intended to be used to evaluate rates of adverse events.” The researchers acknowledge this fact in the limitations section of their report. “The FDA cautions investigators and the public not to use MAUDE to estimate the incidence of device failure,” the authors write. “MAUDE information is often incomplete and frequently no corroborating data are available.” Further, they note, since there is no central AED registry, MAUDE represents (at best) a surrogate database.

“We do not know the source of the reports of device failure from the database,” said Roger D. White, MD, of the Mayo Clinic in Rochester, MN, a member of the Sudden Cardiac Arrest Foundation Advisory Council, in an interview. “If reports originate with professional rescuer organizations, this is more serious concern than if they originate with laypersons who may not understand that failure to restore the pulse is not necessarily a device failure.”

In any case, every effort must be made to reduce AED failures. According to Dr. White: “AED failures should not happen with any frequency. This study provides an opportunity to recognize the serious need for strong medical oversight of AED programs and well-planned and executed maintenance plans. It also reminds us of the need for systematic reporting and analysis of AED uses. Analysis of every incident, including the data from the AED, is strongly recommended. AED data analysis provides critical information on interventions with AEDs.”

This view is part of a rising clamor for an improved system that moves on beyond MAUDE. The Office of the National Coordinator for Health Information Technology, for example, is calling for a new system that would:

• Increase the reporting rate of adverse events
• Improve the quality of the reported data
• Collect and analyze more information that makes it possible to better identify trends
• Make the data more dynamic with continuous updating to identify trends in real-time.[ix]

Putting the Study in Context

When considering the DeLuca, et al. study, it’s important to consider the big picture. The annual incidence of out-of-hospital cardiac arrest is about 295,000 and the mean rate of survival to hospital discharge is 8%.[x] If these numbers are extrapolated to calculate incidence and survival rates during the 15-year study period, approximately 4,425,000 people would have suffered SCA and 354,000 people would have survived. If this is true, “AED failures” would have occurred in a fraction of one percent of cases.

Interpreting the Study for Lay Audiences

Interpreting the meaning of this study may be complex for the general public. It’s important for the public to understand first and foremost that properly used AEDs cannot cause deaths. They are, by definition, used to treat people who are already clinically dead (i.e., people whose hearts are no longer effectively pumping blood). Whether or not those treated with AEDs can be successfully resuscitated (i.e., an effective pulse is restored), depends on multiple factors, including the time interval between sudden death and AED use. Further, even in ideal circumstances, successful AED function will not always result in successful resuscitation.

Moreover, technology has advanced dramatically in recent years. “Technology is helping shape and drive the future of the external defibrillator market,” according to Frost & Sullivan. “(AEDs) are coming to market with features that help improve patient care. Users are now able to receive direct feedback on the quality of care they are providing. In addition to immediate feedback, devices are now programmed with cutting-edge treatment algorithms to improve survivability rates, having a significant impact on the effectiveness of the resuscitation.”

Reading between the lines, we believe this study provides some encouraging insights:

• AED failures are infrequent.  
• Most “AED failures” can be prevented through planning and implementation of effective maintenance programs developed under medical direction. (In our view, therefore, there should be no need to deploy “backup AEDs,” as recommended by the authors. The important thing is to maintain the ones already in place.)  
• AEDs save lives—as long as they are kept in working order and used promptly in cases of SCA.

The last point cannot be stressed enough. Studies have shown that the use of AEDs by bystanders greatly improves the chances of survival from SCA. In fact, while the overall rate of SCA survival in the U.S. is 8%, this rate increases to 34% when bystanders use AEDs in public settings.[xi]

The Real Challenge

Public awareness about AEDs and the need for bystander intervention is very low. Most people are unable to recognize an AED and would be unwilling to use them.[xii] The incidence of AED use by laypersons in cases of out-of-hospital cardiac arrest is also very low (2%[xiii]-4%[xiv]).

Perhaps the “AED failure” we need to be most concerned about is the failure to widely deploy AEDs wherever they are needed—and to engender their use by bystanders.

“AEDs are proven lifesaving devices,” according to Norman S. Abramson, MD, FACEP, FCCM, chairman of the Board of Directors of the Sudden Cardiac Arrest Foundation. “However, it is crucial to check them regularly and maintain them properly. And, it’s vitally important that the public becomes comfortable using them.”

-Mary Newman, MS, President, Sudden Cardiac Arrest Foundation

Note: It’s essential to keep AEDs well-maintained, to train anticipated users, to know how to easily locate devices, and to have a well-planned and executed response. For more information, click here.