AED Reclassification: When Near Perfect is Not Enough for the FDA

The FDA is proposing to reclassify AEDs as Class III medical devices. This change will require manufacturers to go through a much more time-consuming and costly regulatory process before their AEDs can be sold. Why? According to the agency's proposed rule, the primary purpose of reclassification is to reduce the risk of illness or injury resulting from AED failures. Yet, a study relied upon by the agency finds that AEDs work as intended during rescue attempts greater than 99 percent of the time. This is a clear example of the perfect as the enemy of the good (or even great) driving a regulatory action that will cost lives and money with very little public health benefit.

The FDA's main stated concerns are high AED recall rates and the associated risk of device failure. The agency relies heavily on a 2006 study¹ co-authored by Jignesh Shah, M.D., and William Maisel, M.D., MPH, to support its reclassification rule. Interestingly, Dr. Maisel is now deputy director of the FDA's Center for Devices and Radiological Health.

From my perspective, the fundamental issue the agency should be examining is whether AED failures occurred during rescue attempts and resulted in sudden cardiac arrest (SCA) patient deaths. AED failures alone, while they may need to be addressed, are insufficient to justify this rule if the failures are identified when the devices are in standby mode and corrected before they are used on patients.

With this question in mind, let's look at the data. The American Heart Association calculates there are about 360,000 out-of-hospital SCA events in the United States every year.² AEDs are used by bystanders trying to save SCA victims anywhere from 2.1 percent³ to 3.7 percent⁴ of the time. This tells us AEDs are used before EMS arrives between 7,560 and 13,320 times per year. However, national SCA survival rates are less than 8 percent, largely because a defibrillator does not arrive in time. Without defibrillation, SCA victims have almost no chance of surviving.

The Maisel study says there were 370 adverse event reports involving a death over a 10-year period but acknowledges "it is not possible to predict whether a given device malfunction directly led to a patient's death." Let's accept for the sake of discussion that they did. (Interestingly, the FDA's Executive Summary distributed at the January 25, 2011 Circulatory System Panel meeting convened to examine AED reclassification says only that "some" AED failures occurred during a rescue attempt but does not attempt to quantify how many.)

75,600 AED uses over 10 years (a 2.1 percent AED use rate), with 370 adverse events, translates into an AED failure rate of 0.49 percent, meaning AEDs worked as intended 99.51 percent of the time. 133,200 AED uses over 10 years (a 3.7 percent AED use rate), with 370 adverse events, translates into an AED failure rate of 0.28 percent, meaning AEDs worked as intended 99.72 percent of the time. The Maisel study suggests AEDs were used hundreds of thousands of times during the study period (I find no data supporting this statement), which would mean an even lower AED failure rate (e.g., 200,000 AED uses over 10 years would translate into an AED failure rate of just 0.19 percent). And, despite the identified AED recalls, the study notes that thousands of lives were saved. Seems to me it is hard to justify AED reclassification based on this data. (Some additional perspective: Implantable Cardioverter-Defibrillators (ICDs) have a failure rate as high as 1-2 percent⁵).

Another important issue relates to how many of these 370 adverse events could have been prevented by proper human intervention. It is likely the vast majority of issues would have been detected and corrected prior to attempted use if the AEDs were properly inspected and maintained. Researchers at the University of Alabama at Birmingham recently evaluated five years of data from Alabama communities involved in the National Institutes of Health Public Access Defibrillation (PAD) trial.⁶ They found over 50 percent of AEDs were not inspected in the prior 12 months, over 50 percent had expired electrodes, 25 percent had expired batteries, and 33 percent had unresolved maintenance alerts. In another study, 32 PAD sites were evaluated for compliance with AHA PAD site recommendations.⁷ None did.

These are people problems, not device problems that can be addressed through more FDA regulation. Clearly, efforts aimed at improving AED program performance over long periods of time will save more lives than AED reclassification.

One final point: Bystander AED use occurs during fewer than 4 percent of out-of-hospital SCA events. One reason for this is the lack of a sufficient number of AEDs available for use. Between 1.5 and 2.4 million AEDs are now found in public settings. Because AEDs have a limited coverage radius of no more than 300 feet, over 30 million AEDs are needed to meaningfully cover the urbanized United States. Most AEDs currently in public settings were purchased and placed voluntarily by their owners. There are few mandates and no insurance reimbursement. Unlike other public safety initiatives (police, fire, EMS), early defibrillation is not supported by tax dollars (other than government deployed AEDs), though this activity benefits the community at large. Thus, every effort should be made to encourage rather than discourage private sector involvement.

The FDA's reclassification rule will result in a longer time to market for new devices (138 days today versus 540 days under the more onerous regulatory pathway) and much higher regulatory compliance costs ($5,000 today versus more than $260,000 under the new regulatory regime).⁸ This approach will slow innovation, drive up the costs of these life-saving devices, and inhibit the rate of growth for AED deployment. This, in turn, will cost lives because the problem of having an insufficient number of public access AEDs will persist longer than if the rule is not adopted. All for a device that works as intended more than 99 percent of the time.

The FDA certainly has a role to play in ensuring AED safety and quality. An alternative regulatory path can achieve this result while preserving the benefits of rapid public access defibrillation. Basically, continuing to regulate AEDs as they are today but with additional regulatory oversight. That is the better approach and the approach that will save more lives. After all, that is supposed to be the FDA's primary mission.

References
1. Recalls and Safety Alerts Affecting Automated External Defibrillators
2. AHA Heart Disease and Stroke Statistics – 2013
3. Survival After Application of Automatic External Defibrillators Before Arrival of Emergency Medical Services
4. Out-of-Hospital Cardiac Arrest Surveillance — Cardiac Arrest Registry to Enhance Survival (CARES), United States, October 1, 2005–December 31, 2010
5. Patient Management: When ICD Leads Go Awry
6. Public Access to Defibrillation: a Five-Year Follow-Up; Shannon Stephens, Grant Cobb, Henry Wang (not yet published)
7. Community public access sites: Compliance with American Heart Association recommendations
8. AED manufacturers get regulatory jolt

Posted by Richard A. Lazar on May 30, 2013