Wednesday, January 7, 2009

Is Medtronic CEO Abrogating Responsibility Of His Company To Determine Device Safety?

It would appear so. If this is the case, this CEO and his board of directors should be dismissed as a danger to the public. The CEO, William A. Hawkins III, being a biomedical engineer, would take special blame for this distorted attitude.

In "Pre-emption' Cited as Major Case Is Tossed", Wall Street Journal, Jan. 7, 2009 we learn that a federal judge, U.S. District Judge Richard H. Kyle in Minneapolis, threw out lawsuits on behalf of thousands of patients with heart-defibrillator wires that have been shown to fracture and fail to function or dispatch potentially lethal shocks, concluding that a recent Supreme Court opinion made the dismissals inevitable.

Judge Kyle did allow that "the Court recognizes that at least some plaintiffs have suffered injuries from using Sprint Fidelis leads, and the Court is not unsympathetic to their plight," he wrote. Nevertheless, he said, the plaintiffs' remedy "lies with Congress."

Such a dismissal at his level is understandable on procedural grounds, but reflects problems in the laws and procedures rather than in the ability of rigorous, honest, uncorrupted medical science to help prevent such patient harm (and the need for resultant suits) in the first place.

In the article we also learn that:

At issue is the Sprint Fidelis defibrillator "lead," a wire that was the top such product world-wide when Medtronic Inc. pulled it from the market in October 2007. Defibrillators are electrical devices placed under the skin in a patient's shoulder. Connected to the heart, they have saved thousands of lives by sending out powerful jolts of electricity, restarting the normal pattern of an abnormally beating heart.

But defibrillators depend on the leads to carry that electricity to the heart, and the exceptionally thin wires have over decades tended to fail at a higher rate than the devices themselves.

The Sprint Fidelis had been surgically implanted in 268,000 patients world-wide when the Minneapolis company concluded it was fracturing at a higher rate than that of its more robust predecessor, the Sprint Quattro.When the wires fractured, some patients died, either allegedly because the defibrillator failed to dispatch electrical therapy to the heart or because the massive and multiple shocks may have caused patients' hearts to stop.

... Medtronic reports that its 42-month failure rate from all causes is 6.3%, far more than the 1.8% failure rate for its predecessor, the Sprint Quattro model 6947, at 48 months.

The obvious issue that exceptionally thin wires might fail at a higher rate in a living, moving body and heart is a key to understanding the major scientific issue here. It is one of medical and engineering common sense, both of which appear to have been lacking regarding this product (or perhaps were overruled by management fiat).

New and experimental drugs are known to pose risks. This is why pharmaceutical companies are required to do preclinical testing in animals, and, in fact, in good conscience would never presume to release a drug without years of extensive animal testing, even before the first human ever tries a new compound. It is realized that animal testing is a straightforward, first-tier way to eliminate the possibility a new drug is going to have 'hidden surprises' that could kill people.

In the WSJ we learn, however, that:

By early 2007, Sprint Fidelis was the lead used in 90% of new Medtronic patients, garnering about $1 billion in revenue since its introduction in late 2004.

The lead was marketed based not on any multiyear study in hundreds of human patients, but rather on "bench testing" of the product in a laboratory to determine how easily it might fail by putting multiple stresses on it.

Excuse me? Bench testing of a wire lead to be implanted in a beating human heart?

The key to this entire issue and how to help avoid it in the future lies in the last sentence.

That is a stunning admission if accurate.

As a physician and biomedical informaticist who did a clerkship in biomedical engineering in medical school, as an expert in building invasive cardiology information systems to in part evaluate cardiac device safety and efficacy, and as an NIH SBIR grant reviewer in the area of public health, had this been an NIH proposal for funding of development of the new wire I would have scored an application for such development of a defibrillator lead that lacked animal testing at the very bottom of the pile (lowest score).

I would also have argued vehemently at the roundtable discussion that the proposal needed revision to include such testing in a rigorous manner before it could even be considered seriously.

Apparently, private R&D in the device industry, funded by stockholders and insurance companies and patients themselves, does not undergo such rigorous review as proposals seeking public money.

If the WSJ account is accurate, and no animal testing was done, then real, live patients, not animals, became the front line for clinical testing of a new experimental device. Some paid with their lives.

If so, this is unacceptable.

The belief that bench testing (and likely computer modeling, per my post here on overconfidence in computing) of a device made with complex materials, themselves subject to the vagaries and uncertainties of materials science let alone living systems, is adequate for immediate in vivo use in humans is perhaps the biggest worldview gap between linear-thinking engineers and computer personnel on the one hand, and clinicians and life scientists on the other used to complexity and uncertainty in living systems. The linear, grade school level thinking about biomedical issues must stop.

If the WSJ account is accurate, the following questions must be asked:

  • Who made the decision to not do animal testing? Why?
  • Was input from clinicians, scientists or others that such testing should be done raised? If not, why not?
  • Was such input if it occurred overridden? If it was overridden, did the managers who overrode it have appropriate biomedical backgrounds to have overridden it?
  • Were profit issues a cause of lack of in vivo testing in animals of the new lead?
  • Does biomedical engineering as a discipline forego animal testing of devices meant for human use? If so, why? If not, why was it overridden here?
  • Did anyone in Medtronic object to release of this wire lead on the grounds of inadequate testing? If not, why not?
  • If anyone did object to release, were such objections overridden, and who did the overriding based on what criteria?


We perhaps get to the root of the problem in this admission in the WSJ article - the familiar (to HC Renewal readers) issue of leadership:

Medtronic Chief Executive Bill Hawkins lauded the decision [to dismiss the lawsuits] as supporting "the principle that the U.S. Food and Drug Administration is the appropriate body to determine the safety and efficacy of innovative technologies."

Mr. Hawkins seems to be implying that Medtronic is not an appropriate body to determine the safety and efficacy of its products. It would appear he is abrogating primary responsibility for the safety of his company's product, putting that responsibility in the hands of a government agency. I can presume his Board of Directors agrees with this assessment.

This is wrongheaded, cavalier, and has been the basis of recent major individual and class action lawsuits against the pharma industry. The medical device industry should not get a pass.

It is the primary responsibility of a medical R&D and manufacturing company to do all it can to reasonably determine the safety of its experimental devices and their iterative "improvements" before seeking approval for sale and use in humans.

FDA's role is to simply assure that the determination meets sound scientific, biomedical and legal grounds. They are not, should not be, and cannot be a "safety determination subcontractor" to industry.

It seems both Medtronic and FDA failed on this case, in fact. I addressed why making FDA the ultimate arbiter of safety was, in fact, itself risible and cavalier, in my post "Staring Idiocy In The Face: The FDA As The Gold Standard, Or Tin Plated Emperor?" regarding musician Diana Levine's unfortunate loss of her arm from a misapplied drug injection.

In any case, anyone with medical training who has had responsibility for direct patient care would have little trouble understanding the principle of not passing the buck on safety. Mr. Hawkin's lack of this experience, I add, seems to show in his stunning passing of the buck for safety to government.

If I were on the board of this company, a management house cleaning would be in order.

-- SS