Medical Device Daily Washington Editor

WASHINGTON – While the media spotlight has shone brightly on implantable cardioverter defibrillators (ICDs) of late, their external cousins – automated external defibrillators (AEDs) – have grown in numbers and in technological complexity.

AEDs have saved thousands of lives, but their increased prevalence has not been without increased problems, and the authors of an article appearing Aug. 9 in the Journal of the American Medical Association (JAMA) make the case that the FDA should push manufacturers to more closely track their equipment.

The agency wasted no time in responding to the recommendation, posting a statement Thursday on its web site stating that while it agrees with the bulk of the article's findings, it noted “a few points on which the agency differs.”

Authored by Jignesh Shah, MD, and William Maisel, MD, both of Beth Israel Deaconess Medical Center (Boston), the study based its findings on a review of FDA enforcement reports of recalls and safety alerts “to identify trends in these rates and to identify the types of malfunctions” leading to those actions.

Shah and Maisel reviewed recall reports (available at the FDA web site) between January 1996 and December 2005, covering “2.78 million AED device-years.” According to the authors, “21.2% of AEDs distributed during the study period were recalled, most often because of electrical or software problems.”

Defibrillators that lack automatic functions and recalls that did not affect critical AED components were not included in the numbers.

One point of confusion in the article is that the authors define the term “advisory” as including recalls and safety alerts, the latter of which are also known as field alerts. However, after reviewing the data that he and Shah used for the article, Maisel told Medical Device Daily that all the advisories in the authors' database were recalls.

The article notes that “confirmed fatal AED-related device malfunctions occurred in 370 patients,” but it does not specify the total number of times that the AEDs were actually used during the period studied.

Maisel and Shah also do not specify how many of the 370 were directly attributable to operation of the defibrillators – or some other circumstance that might be counted as an adverse event – because such data are not often available for device adverse event reports due to difficulties in attribution of cause.

Of the 52 recalls, which affected almost 386,000 AEDs and AED accessories, 15 were for accessories, such as cables and pads, and the balance of 37 were for the defibrillator unit itself. Among the unit failures were failure to detect and failure to shock, both of which could include software failures.

Software also is listed in its own category, however.

Of the 37 recalls for defibrillator units, the largest number in any one category was for failure to shock, the reason for five recalls that covered more than 51,000 defibrillators. Eight recalls were for electrical problems, pulling in almost 32,000 defibrillators.

Recalls for “failure to detect” were in a near tie for the second spot with electrical problems, also at just under 32,000. Class I recalls accounted for four of the 37 unit recalls, and 31 were Class II recalls.

As for accessory recalls, seven were for cables accounting for a total of somewhat more than 148,000 defibrillators, and four recalls for pad problems caught almost 51,000 defibrillators. All but one of the accessory recalls were class IIs.

The article reminds readers that Class I recalls are those that involve a “reasonable probability that use of the product will cause serious adverse health consequences” and class II recalls are for devices that “may cause temporary or medically reversible adverse health consequences or the probability of adverse consequences is remote.”

Shah and Maisel point out that FDA advisories deal with “the potential for a device to malfunction, not [necessarily with] an actual device malfunction.” They assert that “as such, they are a surrogate marker of device reliability.”

Qualifying this assertion, they acknowledge that “some advisories are issued even when the risk of device failure is less than 1%.”

The article also claims that “while a patient's ICD is routinely registered with the manufacturer . . . no such process occurs with AEDs” and that “the inability to track devices and end users makes it impossible to know how many AED units were actually fixed or taken out of service.”

This is one of the points that the agency disagrees with, insisting that it can document that “more than 95% of the AEDs affected by Class I recalls in 2005 were returned to the manufacturers or taken out of service. Fewer than 3% were lost or stolen.”

As to the JAMA article's argument that the number of AEDs affected by recalls has gone up, the FDA rebuts that too.

It says that “improvements in the devices' ability to self-diagnose hardware and software problems may contribute to this trend,” which often brings units out of distribution “before a device is ever used on a patient.”

Anne Devine, a communications specialist for Medtronic (Minneapolis) said, “Overall, Medtronic thinks it was a balanced article” but insisted that “we do track [AEDs] very carefully.” She told MDD that the company does not favor a different regulatory approach.