Washington, DC - "There's a dark side to the therapy, but that's always been true. The problem is that people pretended it wasn't true."

Dr Bruce L Wilkoff (Source: Cleveland Clinic)

That's the reason for a wave of reform engulfing cardiology, heart-rhythm-device companies, and federal regulators, as offered by Dr Bruce L Wilkoff (Cleveland Clinic Foundation, OH), a coauthor of implantable-cardioverter-defibrillator (ICD) and pacemaker safety-oversight recommendations recently announced by the Heart Rhythm Society (HRS) [1]. He and other experts reached by heartwire hold those three parties, and perhaps the news media as well, accountable for failing to grasp and publicize a more realistic picture of ICDs that balances their immense contribution with insight that they aren't flawless.

The devices' limitations became apparent to everyone last year when the deaths of two patients with different models of Guidant defibrillators led to several class I recalls involving more than 70 000 patients around the world [2].

The fact that Merck has been on the front pages with Vioxx litigation, the fact that Guidant was on the front pages with litigation, I think has created a public distrust for pharma and device companies in general.

As widely described in medical journals, newspaper and TV-news accounts, and heartwire, Guidant had learned that design flaws in some of its Prizm 2 DR and Contak Renewal models could prevent delivery of needed therapy. The company met its regulatory responsibility to disclose the information to the US Food and Drug Administration (FDA) but, in what seemed like hubris to some, didn't report the problems to clinicians in practice and continued to allow some remaining unused devices to be implanted in patients.

The affair first came to public light in May 2005, when the New York Times reported on the witnessed sudden death of a young patient with one of the involved ICDs and the apparently secretive company response to it. A second patient with a flawed Guidant device died only days after the newspaper story. At fault was a different device model and design problem, but again it was one the company knew of. Only then were the recalls issued.

The events brought criticism to the industry and the FDA, led to reform-minded committees, and ultimately derailed Johnson & Johnson's deal to purchase Guidant—which was later sold to Boston Scientific. The committees produced an onslaught of recommendations for revamping the heart-rhythm-device safety oversight process [1,2,3,4,5].

Dr Douglas Zipes (Source: Indiana University)

Another coauthor of the HRS recommendations, Dr Douglas Zipes (Indiana University School of Medicine, Indianapolis), appears to speak for many when lauding the devices. "When you consider the complexity of an implantable cardioverter defibrillator the size of a small pager, when you think of what this device can do, it's astounding," he said to heartwire. "We've taken groups of patients with a 30% to 50% annual mortality—annual mortality—and reduced it to 1% or 2%. . . . There's nothing precipitous that has changed that over the past year, it's simply the realization that they don't work 100% of the time."

The devices are more likely to fail because of a lead-related problem or a programming choice the physician makes.

From about 1986 to 2000 or so, ICD therapy went through "infancy, childhood, and adolescence," according to Wilkoff. By 2004, it had matured enough to convince the Centers for Medicare and Medicaid Services to extend Medicare reimbursement from secondary-prevention use only to include a range of primary-prevention uses. Although device-related rates of malfunctions and clinical complications during the 1990s "were the same as they are today," he noted, so many devices are implanted in expanding legions of lower-risk patients now that "you can see every blemish as well as the good."

Wilkoff said that two recent analyses led by Dr William H Maisel (Beth Israel Deaconess Medical Center, Boston, MA), recently reported by heartwire, showed that ICD malfunctions "have rarely been associated with mortality" [6,7]. Moreover, he observed, the ICD malfunctions tracked in the two analyses occurred during the same era in which clinical trials were demonstrating the devices' overwhelming survival value. They are more likely to fail "because of a lead-related problem or a programming choice the physician makes" than they are from a malfunction.

Zipes observed that it wasn't malfunctions per se that put Guidant in hot water—in fact, its rates appeared to be low. It was the company's failure to be open about it, or at least to ensure that clinicians promptly received the news. "I think they realize they made the wrong decision, and they did betray a trust to the patients and to physicians by not going public with it." Patients also seem to feel that way, and the company has apparently responded. "Their product announcements are now far more thorough, and I think they'll become even more transparent."

The whole process has to be much more transparent at all levels.

Dr Robert G Hauser (Source: Minneapolis Heart Institute at Abbott Northwestern)

In an interview with heartwire, Dr Robert G Hauser (Minneapolis Heart Institute Foundation, MN) contrasted Guidant's behavior with that of Medtronic in February 2005, when it issued an advisory about possible premature battery depletion in three of its defibrillator models. Although "a higher proportion of Medtronic devices were replaced than Guidant devices," he said, the reaction from doctors and patients was far more accommodating. "You know why? Because Medtronic came out with the information right away." Hauser, also a coauthor of the recent HRS safety oversight recommendations, was one of two cardiologists caring for the 21-year-old patient whose death ushered in the Guidant recalls.

In September 2005, the physicians—Hauser and Dr Barry J Maron (Minneapolis Heart Institute Foundation)—published their take on the Guidant saga and what went wrong [3]. According to their report, Guidant had argued privately that its devices' potential for malfunction was too small to justify the risk of replacement, so it wasn't going to inform physicians or patients. "Guidant's statistical argument ignored the basic tenet that 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. "Furthermore, by withholding vital information, Guidant had in effect assumed the primary role of managing high-risk patients, a responsibility that belongs to physicians."

You can't have medical decisions being made by a group of engineers.

And the FDA, they continue, "was in possession of important information about the safety of the Prizm 2 DR many months before the death of our patient and, to the best of our knowledge, took no action." The authors speculate that "the agency was not prepared for the upsurge in ICD technology and the extraordinary growth in the number of ICD implantations that has occurred during the past five years."

"I think we need to give the FDA a chance to address the concerns that we raised in the report," Hauser said. "I know the people, they've got good intentions, I think they're focused on this issue, and I'm sensing that there's going to be progress made."

Zipes apparently also sees the current regulatory process as weak and the FDA's oversight capability as limited. "I think the FDA has been understaffed. I think they get swamped with data. They have proprietary issues whereby they feel they indeed know things about a particular company that are private, and they're in a difficult situation as to whether they should go public or not with something like that," he said. But it's now clear that "the whole process has to be much more transparent at all levels."

Currently, industry and hospitals are required to report all device problems to the FDA, according to Zipes, who said he's sure they comply. But, he added, "the ones for whom it's not mandated are us, the physicians. And I think in many respects we've dropped the ball because we have not been conscientious enough in reporting device failures."

Wilkoff contends that physicians and the public were overconfident about ICDs once the devices had established themselves. "I think people had an unrealistic picture and didn't want to hear the whole story. I think largely that people would rather deny the full story and not hear that there are potential problems. But we can't afford to do that anymore."

One of the things I'd like to see . . . is for patients to get the information before they see it on TV or in the papers.

"One of the things I'd like to see, and I know patients would like to see, is for patients to get the information before they see it on TV or in the papers—a system that will allow manufacturers to inform physicians and patients simultaneously," said Hauser. As recently reported by heartwire, the HRS preliminarily proposals include, among others, expanded and intensified monitoring of devices by industry and, for the first time in a serious way, by independent physicians; guidelines for when implicated devices should be left in patients or replaced; and prompt notification of patients, clinicians, and the FDA using language clearer than the current norm.

In response to physician and public reaction to its handling of the ICD problems, Guidant invited Dr Robert J Myerburg (University of Miami, FL) to lead a panel of experts in developing ways to strengthen and open up corporate device safety oversight and reporting practices. The panel's recommendations, released in March 2006 and reported by heartwire at the time, were aimed at improving physician involvement in the process, focus on safety issues in postmarket surveillance, communication within the company, transparency in product performance evaluation, and retrieval of unused devices involved in advisories or recalls that are already in the field.

They have to worry about share price, shareholders, competition, the cost of their devices. . . . So I think it's inappropriate for industry to guard itself.

The Myerburg report explicitly addresses Guidant's role in its device-malfunction and recall story and was welcome, observed Hauser. "You can't have medical decisions being made by a group of engineers," he said. What the HRS produced is a "call to action," a "framework around which a lot of additional work needs to be done by industry, the FDA, and the medical community."

According to Zipes, he and others have recommended that device postmarket surveillance be taken out of the companies' hands and given to independent physician-led oversight groups that resemble clinical-trial data and safety monitoring boards. "Industry—and this is not pejorative—has an inherent conflict of interest. They have to worry about share price, shareholders, competition, the cost of their devices. . . . So I think it's inappropriate for industry to guard itself."

"I think the public today is very distrustful of medical companies," Zipes said. "The fact that Merck has been on the front pages with Vioxx litigation, the fact that Guidant was on the front pages with litigation, I think has created a public distrust for pharma and device companies in general."

Will it slow the growth of defibrillators? Sure it will. That's not as much of a problem as if we lose trust.

Moreover, he noted, it's possible the public will start to believe that ICD performance is sinking as recommendations from the HRS, the Myerburg report, and other sources start to kick in with greater scrutiny and more careful reporting of events. That means industry will have to work that much harder to win back public trust. He proposes that industry or the HRS wage a public education campaign "to inform the public just what the devices are all about in terms of reliability."

Said Wilkoff, "In the short run, it will no doubt be painful. In the long run, I think what's happening is absolutely necessary and will make things much better. Will it slow the growth of defibrillators? Sure it will. That's not as much of a problem as if we lose trust.

"Over the past year, I think the most damage, unfortunately, was done by the way this information was revealed to the public and some of the ways it was covered in the press," Wilkoff said. Some of the coverage was overly sensational and "terrified the patients." His "plea" is that media representatives strive to understand that every device risk and benefit must be placed in the context of each patient's clinical situation and to realize that "we have a terrible disease and an excellent therapy that's not perfect."