Risks Evaluated For Heart Implants

Manufacturers Urged to Boost Reliability, Publicize More Performance Data

Like a lot of cardiologists who specialize in arrhythmia, William Maisel receives regular notices about problems in some of the life-saving devices implanted in his patients’ chests. These “Dear Doctor” letters confront affected patients and their doctors with hard decisions about whether or not to replace a potentially flawed device with another one of unknown reliability.

Photo by Graham Ramsay

William Maisel found that the reliability of pacemakers is up, while that of implantable defibrillators is down. In his practice, Maisel implants a diverse array of devices so that fewer patients are subject to any given recall.

“Suddenly, I’m sitting with a list of patients that might have defective devices,” said Maisel, HMS assistant professor of medicine and director of the Pacemaker and Device Service at Beth Israel Deaconess Medical Center. “Even if a letter says there is a one in 1,000 chance of failure from this specific mechanism, we aren’t told about the expected failure rate of the replacement device. One of the missing pieces of information is the risk of replacing a device.”

Several years ago, Maisel set out to fill in the missing information and find out if the increase in advisories meant that the devices were becoming less reliable or that federal oversight on safety issues was becoming better.

Part of the answer is published in two studies in the April 26 Journal of the American Medical Association. From 1990 to 2002, the popularity and reliability of pacemaker generators has improved, and the malfunction replacement rate has decreased, Maisel reports. Generators for implantable cardioverter defibrillators (ICDs), on the other hand, had been mostly reliable until the last few years of the study, when their malfunction and replacement rates spiked.

Maisel presented his results at a meeting of device makers, electrophysiologists, and FDA officials in September. The meeting was prompted, in part, by the publicity triggered by the death of a 21-year-old college student with hypertrophic cardiomyopathy, who died when his defibrillator short-circuited. As required by law, Guidant, the manufacturer (which was recently acquired by Boston Scientific Corp.), alerted the FDA that 25 similar devices of the same model had failed in patients. But the company alerted physicians last May, only the day before The New York Times reported these events.

The device-maker maintained that the observed failure rate was so low that an alert could have unnecessarily exposed patients to the risk of device-replacement surgery, including infection. The statistical argument angered the student’s doctors. “Patients have a fundamental right to be fully informed when they are exposed to the risk of death no matter how low that risk may be perceived,” they wrote in the Sept. 27, 2005 Circulation.

Since the fall meeting, manufacturers have added important details to their latest public annual reports on device performance, Maisel said, though no single company includes all the desired information for informed decision-making.

On the same day Maisel’s papers came out in JAMA, the Heart Rhythm Society released its draft recommendations urging that manufacturers establish higher performance standards for its devices, publish more detailed annual data on malfunctions, and make the information publicly available, among other changes. The statement also urges physicians to report all problems directly to the FDA’s online MedWatch program at www.fda.gov/medwatch. Maisel and his colleagues at the society presented the recommendations on May 17 in Boston at the society’s annual meeting.

Net Benefits Nonetheless
For the vast majority of people who have received or are considering one of these implants, the benefits still trump any risks by a wide margin, Maisel said. The study results only apply to the small group of patients with devices that are the target of a recall or advisory. “This allows them to have a better understanding of the risks and benefits of replacing a potentially defective device,” he said.

“Many trials have demonstrated the clinical benefit of these devices in certain patients,” Maisel said. “Nothing in these studies overrides that benefit in that well-outlined group of patients who have been advised to have a device implanted. For those patients who already have implanted devices, nothing here requires anything different from their routine regular checkups with physicians to have the devices checked.”

Sudden cardiac death, usually from a sudden fast heart rhythm, is the leading cause of death in this country, edging out fatal heart attacks caused by inflamed and clogged arteries and killing more people than all cancers combined.

ICDs can sense and shock a malfunctioning heart back into its normal beat. Several clinical trials have shown that ICDs reduce the relative risk of death in selected high-risk patients by a third and lower the absolute risk of death from 19 percent to 14 percent in one year, which adds up to a lot of people avoiding the most common cause of death. Pacemakers are simpler devices that continuously or occasionally pulse to ensure normal beating of a dangerously slow heart. The devices typically last five to seven years and normally are replaced due to aging batteries.

Implant Replacement Rates
Malfunctioning batteries and electrical issues accounted for most of the premature device replacements in Maisel’s studies. In one paper, he teamed up with co-authors in the FDA’s Center for Devices and Radiologic Health. (In the course of proposing the study, Maisel became an FDA consultant; he chairs the FDA Circulatory System Medical Devices Advisory Panel.) The study analyzed the mandatory annual reports sent from manufacturers to the FDA for the numbers of device implants and the numbers, types, and causes of malfunctions. These details are not available on the publicly searchable FDA website.

From 1990 to 2002, U.S. doctors implanted 2.25 million pacemakers and almost 416,000 ICDs. During the same time, they removed 8,834 pacemakers and 8,489 ICDs due to a confirmed device malfunction. To identify trends, the researchers divided confirmed malfunctions by the annual number of new implants for an overall annual replacement rate of 4.6 per 1,000 pacemakers and 20.7 per 1,000 ICDs. Malfunctions were confirmed by the device-makers.

The manufacturers do not actively seek out device problems and instead assume patients are alive and devices working unless they hear otherwise. To address the possible underreporting of the problem, Maisel conducted a meta-analysis of three voluntary device registries in the United States, the United Kingdom, and Denmark, which actively follow long-term patient outcomes. He found similar malfunction and replacement rates.

In a third paper examining a different question, Canadian researchers reported that replacing flawed devices may be as dangerous as leaving them in. A retrospective analysis of one year at 17 centers in Canada showed a major complication rate of 5.8 percent, similar to complication rates of new implants reported in large randomized clinical trials.

The bottom line is that physicians and their patients can anticipate a 1 to 2 percent failure rate of these devices. “The big picture is that this is a bad disease and a very good therapy, but not a perfect therapy,” said Bruce Wilkoff, an electrophysiologist at the Cleveland Clinic Foundation, who wrote an editorial accompanying the study. “There is solid evidence that this reduces absolute mortality by 5 to 10 percent. The biggest tragedy would be if people [died after they] chose not to have a defibrillator implanted to avoid a complication.”

—Carol Cruzan Morton

top