Barriers to evidence based use of thrombolytics in ischemic stroke

I have written several posts on the low uptake of best evidence in clinical practice and have commented on some of the barriers to evidence adherence. The use of thrombolytic therapy, specifically TPA, in patients with ischemic stroke is a special case. It’s been 13 years since publication of the NINDS study, the trial on which FDA approval for TPA in ischemic stroke was based. Its findings were validated on reanalysis by independent observers and additional evidence has accumulated in support of the NINDS conclusions. Nevertheless TPA for ischemic stroke, 13 years post NINDS, remains one of the most hotly debated issues in medicine. Adherence to treatment guidelines is low. Only a small fraction of patients presenting to the ER with ischemic stroke are candidates, but of those it’s a safe bet that fewer than 50% are treated. Why? I will make a case here that a major barrier to guideline based thrombolytic therapy of ischemic stroke may have been erected, however good their intentions, by the pharmascolds.

A review in Emergency Medicine Clinics of North America chronicles the history of the controversy and notes:

Numerous professional and community organizations have endorsed tPA use in stroke, including: the American Academy of Neurology, American College of Chest Physicians, American Heart Association/American Stroke Association, Canadian Stroke Consortium, National Stroke Association, and multiple others. Additionally, the National Institutes of Health (NIH) sponsored national symposia on promoting treatment of acute stroke in 1997 and again in 2002.

Notably absent from the list of endorsing professional organizations was emergency medicine representation. In 1996 the American College of Emergency Physicians (ACEP) agreed ‘‘with reservations’’ to the new stroke guidelines. In 2002, ACEP issued a policy statement indicating ‘‘intravenous tPA may be an efficacious therapy’’ but that ‘‘there is insufficient evidence to endorse the use of tPA.when systems are not in place to ensure that.NINDS guidelines.are followed.’’ The decision to use tPA ‘‘should begin at the institutional level.’’ The American Academy of Emergency Medicine issued a position statement in 2002 to address issues of medical-legal liability. Their position paper stated, ‘‘Debate on the safety, efficacy and applicability of tPA has limited its widespread use. Nonetheless, an increasing number of liability suits are emerging against physicians for not administering tPA.’’ and stated that there was insufficient evidence to classify tPA use in stroke as a standard of care.

If emergency medicine’s official position on TPA has been at best lukewarm, in the trenches of emergency medicine it’s much more negative. Surveys of emergency physicians suggest that many have never used it or are strongly disinclined to do so. If you lurk on some emergency medicine blogs and listservs it’s even worse. It’s almost as if some providers are convinced it’s like giving cyanide. Clearly the field of emergency medicine, among medical specialties, is uniquely negative about TPA for ischemic stroke. I’ll get to the reasons in a moment, but think of what it means for patients. The ER is the patient’s best chance to get TPA in a timely manner, and, with neurologists joining all the other subspecialists in the flight from the hospital EPs are often the only available docs to treat patients. Emergency Medicine should be taking the lead in evidence based stroke care, not erecting barriers.

Why the resistance from emergency medicine all these years? Well, as it happens, some of EM’s most influential leaders are also some of the world’s leading pharmascolds who are critical of the American Heart Association’s ties with Genentech, the maker of TPA. Notable among them is Jerome R. Hoffman MD, MA, professor of Emergency Medicine at UCLA. He was skeptical of recommendations in favor of TPA soon after publication of the NINDS study and, starting with this commentary in the Western Journal of Medicine, has, through the years, published article after article critical of widespread adoption of the stroke guidelines concerning TPA treatment. Asked in an interview about his concern for the Pharma ties:

"I don't have reason to believe that there is a quid pro quo with anyone in the American Heart Association," he said. "On the other hand, many of the volunteers on the panel have worked for drug companies, and while people who do research for drug companies often deny that this has any affect on their science, studies show it does have an effect -- results tend to be better for proprietary research than for non-proprietary research."

That statement is reasonable enough but, having participated with Dr. Hoffman on the No Free Lunch listserv for several years I can assure you that his rants against Big Pharma and other corporate interests are not always so circumspect. He’s not one of Big Pharma’s admirers. To say the least.

So now after ECASS-3 and SITS-MOST have provided additional reassurance about the safety and efficacy of TPA in ischemic stroke what is Dr. Hoffman up to? He’s still questioning the findings of the NINDS trial. The latest paper, in the Annals of Emergency Medicine was posted on line ahead of print. The conclusion reads:

Our graphical method of presenting the NINDS trial results provides more detail than was conveyed in the original report and empowers readers to reach their own conclusions about the trial's meaning. Outcomes for placebo and treatment limbs
are sufficiently similar that larger trials, conducted under the same conditions as the NINDS trial, are needed to determine which patients benefit from this therapy.

Thirteen years after NINDS, and right on the heels of ECASS-3, just as doctors are becoming more comfortable with TPA these authors decide to publish a reanalysis of the NINDS findings from 13 years ago. Evidently Hoffman and his co-author David L. Schriger MD, MPH feel that after 13 years we still don’t have the research we need to determine whether some patients with ischemic stroke will benefit from TPA. What has their novel reanalysis of the NINDS data revealed that the rest of us don’t know? I’ll reserve judgment about the paper until I have access to the full text.

Meanwhile as emergency medicine’s key opinion leaders continue to discourage widespread adoption of published guidelines for thrombolysis in ischemic stroke doctors in the trenches of emergency medicine are faced with questions. How should they interpret 13 years of perseveration against the guidelines by their leaders? Is it healthy skepticism or extreme cynicism driven by contempt for the pharmaceutical industry? What do you think?

Pending my review of the full text of Hoffman’s paper here’s my take on TPA for ischemic stroke:

Some proponents, I concede, have been excessive in their zeal to promote the treatment (see this ridiculously inaccurate New York Times piece).

Properly patients stand a better chance of having a good stroke outcome when treated.

Some patients will die as a direct result of treatment with TPA while others who would have died from their ischemic stroke will survive as a result of improvement after TPA, making the mortality effect neutral in the aggregate.

Bad outcomes directly attributable to TPA are dramatic and pose special difficulties for families and treating physicians.

The treating physician needs to be armed with some numbers to get informed consent based on sufficiently detailed information. Some guidance can be found here.

I do not believe it should be considered “standard of care” but, then again, I don’t like that term for any treatment. It’s too dogmatic and only gives fuel to the trial lawyers.