The article was published several months ago in the British Medical Journal (BMJ), but stroke neurologists are still talking about it (2002; 324: 723-726). In a three-page feature entitled “Alteplase for Stroke: Money and Optimistic Claims Buttress the ‘Brain Attack’ Campaign,” medical journalist Jeanne Lenzer contended that financial interests might have influenced the American Heart Association's (AHA) 2000 practice guidelines that recommended recombinant tissue plasminogen activator (rt-PA) – or Alteplase, its trade name – for treatment of acute ischemic stroke, despite concerns about its safety and efficacy (Circulation 2000; 102 (suppl I): I204-I216).
WHAT THE ‘BMJ’ ARTICLE CLAIMS
In her article, Ms. Lenzer charged that the AHA had received $11 million from Genentech, the US manufacturer of rt-PA, in the decade before the recommendations were made; six of the eight panelists who supported the use of rt-PA in ischemic stroke had ties to Genentech; and no statements regarding conflict of interest were published with the guidelines.
Ms. Lenzer cited studies and position papers by several groups of emergency room physicians who questioned the safety of tPA, citing its risks for hemorrhage. They also questioned the AHA recommendation, which they claim was based on only one trial by the National Institute of Neurological Diseases and Stroke (NINDS) rt-PA Study Group. The NINDS rt-PA Study Group concluded that early stroke treatment with rt-PA was associated with better outcomes (New England Journal of Medicine 1995; 333: 1581-1587); (Neurology 2000; 55: 1649-1655).
REACTIONS TO STORY
Do the author's claims have merit? Should the AHA guidelines regarding rt-PA for stroke be revisited? Leading stroke experts and emergency medicine physicians, who are often the first to see people with strokes, reacted to the story with mixed response. In published letters to BMJ and interviews with Neurology Today, there were two threads of debate – one, which questioned – or affirmed – the science behind the trials and guidelines supporting the use of rt-PA for stroke, and the second, reacting to claims of financial conflicts of interest in developing the recommendations for stroke treatment.
In an editorial accompanying the BMJ article, Jeffrey L. Saver, MD, Associate Professor of Neurology at the University of California-Los Angeles (UCLA) Stroke Center, and others responded to both concerns. They wrote that data from six trials enrolling patients within three hours showed that rt-PA is a proven treatment of substantial benefit (Neurology 1999; 53: S3-14). They added that the “real scandal in acute stroke care is not that thrombolytic therapy is being used, but that it is not being used often or wisely enough.” They cited studies showing that only one to two percent of acute stroke patients in the US are receiving thrombolytic therapy (Journal of the American Medical Association 2000; 28:1151-1158).
The authors of the editorial acknowledged that conflicts of interest must be managed, but they noted that the writers of treatment guidelines in all fields of medicine have some form of interaction with the pharmaceutical industry, and that it is therefore unrealistic to “avoid all appearance of potential bias,” as is recommended by Ms. Lenzer.
The AHA responded to the BMJ article with a letter by AHA President David S. Faxon, MD. In that letter, Dr. Faxon affirmed that the AHA had “implemented a thorough, multileveled process to prevent any individual or special interest from unduly influencing its ‘Guidelines 2000 for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care.’”
He noted that the guideline development took more than 18 months and involved more than 300 of the world's leading resuscitation experts, and that the final published recommendations resulted from countless hours spent by panel committees, international scientific councils, an international editorial board, and members of the American Heart Association's Emergency Cardiovascular Care committee and writing groups.
“All those who developed the guidelines adhered to the American Heart Association's stringent conflict of interest policies and procedures,” Dr. Faxon wrote. “Participants submitted disclosure forms reporting all relevant relationships with external organizations, groups, and companies. Panel and subcommittee chairs reminded participants to report these relationships and if necessary, asked them to refrain from discussion or voting.”
NO INTRUSION OF FINANCIAL INTERESTS
In a phone interview with Neurology Today, another prominent stroke expert, James C. Grotta, MD, questioned the BMJ article's allegations that financial interests influenced the guidelines on rt-PA.
“The NINDS study, when compared to many pharmaceutically funded trials, tried to avoid any conflict of interest,” he said. “There was much less sponsor involvement than in an industry-funded trial.
“I take concerns about conflict of interest very seriously,” Dr. Grotta continued. “To complain that Genentech tried to influence people's behavior regarding rt-PA for stroke is exaggerated when compared with efforts of pharmaceutical companies for other conditions, where marketing efforts are far more vigorous. You can look through any journal, and you will never see an advertisement for rt-PA for stroke.”
Dr. Grotta, Director of the stroke program and Professor of Neurology at the University of Texas Medical School in Houston, also responded to the article in a lengthy letter to the BMJ in July on behalf of the NINDS rt-PA Study Group, in which he participated.
“We cannot speak to the actions of the AHA and the financial relationship of that organization to Genentech, but the original guidelines recommending tPA were formulated not by the AHA, but by independent ‘clinician-interpreters’ empanelled by the AHA; eight of 13 of whom had nothing to do with the NINDS trial, and none of whom personally were receiving funds from Genentech at the time the guidelines were formulated.”
NO MERIT TO CLAIMS
In a phone interview with Neurology Today, stroke expert Louis R. Caplan, MD, Chief of Cerebrovascular Disease at Beth Israel Deaconess Medical Center in Boston and Professor of Neurology at Harvard Medical School, noted that virtually all stroke neurologists are aware of the BMJ article. “Other neurologists and those who treat acute stroke patients should know that exists,” he said. “I do not think the issues regarding bias and financial considerations have any merit.”
GUIDELINES NEED REVISING
Dr. Caplan did express concerns, however, about the AHA guidelines regarding the use of rt-PA for stroke. “I do feel that the present guidelines badly need revising,” he said, “but the problems with the guidelines do not, in my opinion, relate to bias. Rather, they merely reiterate the inclusions and exclusions of the NINDS trial. No prior guidelines do this, since caring for individual patients is very different from a trial.”
Dr. Caplan noted that the present guidelines do not address the need for judgment and more extensive imaging when treatment is uncertain, and they lack specification regarding the type of physician and setting appropriate for the administration of rt-PA.
CAREFUL SELECTION OF PANELISTS
Gary Franklin, MD, who is Chair of the AAN Quality Standards Subcommittee (QSS) and Research Professor in the University of Washington Occupational Epidemiology and Health Outcomes Program, agreed with Dr. Caplan on the question of inclusion or exclusion criteria for the trial.
As Chair of the AAN QSS, Dr. Franklin was involved in the development of the 1996 guidelines on anticoagulants and antiplatelet agents in acute ischemic stroke, which were jointly developed by the AAN subcommittee and the American Stroke Association, a division of AHA. He noted that the QSS is extremely careful in establishing its expert panels, ensuring that they are balanced. None of the funding of the QSS efforts comes directly from pharmaceutical companies, he said.
PROBLEMS WITH IMPLEMENTATION
It is true, Dr. Franklin said, that the recommendation for rt-PA needs more study and specificity in terms of who should receive it and when it should be used. It is also true, he said, that the principal benefit of rt-PA occurs within the first 90 minutes after stroke onset. And therein lies the problem. Most patients do not arrive at the hospital within that time frame and do not receive timely evaluation within this window. At the same time, many hospitals – particularly community hospitals in rural settings – may not have the staff to respond with the treatment within the recommended treatment window.
Dr. Franklin noted that emergency medicine physicians, whom Ms. Lenzer cited in the BMJ article as having concerns about the data for rt-PA, may be concerned as much with implementation issues – how to respond in the most timely and effective manner – as with which patients should receive the drug.
EMERGENCY MEDICINE'S ROLE
In another letter in response to the BMJ article, Phillip B. Gorelick, Professor of Neurology at Rush Medical College in Chicago, IL, acknowledged that there are obstacles to be faced when stroke is accepted as an emergency. “Emergency room physicians must become experts in stroke diagnosis or must arrange that experts are available 24 hours a day in their ERs,” he wrote.
“They must have access to immediate and expert CT examination and interpretation. Without the requisite staffing and appropriate organization and skill, rt-PA treatment can be dangerous and is contraindicated.”
But, Dr. Gorelick added, “a gap in prerequisite organization is not an argument against the value of rt-PA in reducing stroke disability.”
Dr. Gorelick also noted that staffing levels may be a primary issue in recommendations against rt-PA. He cited a report from the Association of British Neurologists, for example, that found that the UK has only one neurologist per 177,000 persons.
He wrote: “The British report concludes that ‘it would be inappropriate for neurologists to participate in acute general medical intake in emergency rooms.’ By contrast, in Canada there are almost half as many neurologists with special interest in stroke (150) as the total number of all neurologists in all the various special subdivisions of neurology in the UK. The US figures are at least of the same proportion as the Canadian.
“Because the care of acute stroke patients is labor-intensive and demands the attendance of stroke experts, it may be some time before this new treatment strategy will be accepted, let alone implemented in the UK. We wonder if a measure of the skepticism about the value of rt-PA in acute stroke is not prejudiced by the probable inability of the British health-care system as presently constituted to implement the treatment safely.”
For its part, the American Academy of Emergency Medicine (AAEM) in the US has, in a position statement, stated that the objective data were insufficient to justify using rt-PA as a first-line therapy for acute ischemic stroke (AAEM Work Group on Thrombolytic Therapy in Stroke, 2002). Concerns were also raised by the Canadian Association of Emergency Physicians and by members of the American College of Emergency Physicians.
Jerome R. Hoffman, MD, Professor of Medicine/Emergency Medicine at the UCLA School of Medicine, who was quoted in the BMJ article by Ms. Lenzer, was one of the original AHA panel members for its 2000 stroke guidelines. But Dr. Hoffman requested that his name be removed from the AHA list of guideline authors when he dissented from the eight panelists who supported the use of rt-PA.
The BMJ article addresses two important issues, Dr. Hoffman told Neurology Today. “First, there are the clinical issues – whether or not thrombolytic therapy should have been made a class I therapy [from evidence provided by a prospective, randomized, controlled clinical trial with blinded outcome assessment, in a representative population] for acute ischemic stroke,” he said. “People like me believe it's not proven overall, and it's not clear if there is a subgroup of patients from among those who present with ischemic stroke for whom there is more benefit than harm. I find the decision to categorize this as a class I therapy rather astonishing.”
“The second issue is the question of whether [financial] conflicts of interest may influence scientific judgment,” he continued. “The author reported several facts about the AHA and proprietary organizations, and about panel members and proprietary organizations. These facts speak for themselves. I personally believe financial conflicts of interest are an important and powerful motivator, and there is a lot of evidence in the medical literature to support that. I do not think they are a conscious motivator in the large majority of cases … I believe that such relationships impair our judgment, whether we think they do or not.”
Deepi Goyal, MD, another emergency medicine physician who was a co-author of the AAEM position statement on thrombolytic therapy in acute ischemic stroke, agrees that the BMJ article has merit.
Dr. Goyal, Assistant Program Director for the Emergency Medicine Residency Program at Mayo Medical School in Rochester, MN, said: “My feeling is that the data to support rt-PA for stroke are controversial enough that we can't consider it to be the standard of care in stroke, given the studies that have been published thus far. I find it hard to implement the class I recommendation by the AHA.”
Dr. Goyal also pointed out the three-hour window within which rt-PA is considered beneficial applies to the few stroke patients who are able to arrive at the hospital within that time. “A very small minority of stroke patients get to the hospital within that window,” he said. “Whether all patients who meet that criteria get the same degree of benefit is controversial. The patients treated within less than 90 minutes of the onset of stroke got more benefit than the group who received treatment 90 to 180 minutes after onset. There may be a question regarding any benefit to thrombolytic therapy for the latter group. Until the data are public, it's difficult to generalize the recommendation of thrombolytic therapy to the entire group.”
STANDARD OF CARE?
How valid are concerns about the standard of care? Dr. Franklin pointed out that the term – standard of care – is used primarily in a legal context, but is not a concept used in medical training. The principle is applied in malpractice suits, he said, to determine whether a physician has practiced outside of a standard of care.
But, he continued, standard of care could vary by community, by specialty, and by common practice. “Just because the AHA might state that a treatment should be used wouldn't necessarily define it as a standard of care for all specialties.”
The AAEM articles clearly imply that rt-PA treatment for acute stroke is not a standard of care for ER doctors, Dr. Franklin noted. “Even for neurologists, if the stated recommendation by the AHA is only being implemented with caution and not as frequently as some stroke experts believe it should be, that would indicate that a standard of care for neurologists has not clearly been established.”
Given the controversy over rt-PA, was the AHA being unnecessarily scrutinized? In an interview, Ms. Lenzer told Neurology Today that the AHA and Genentech were not being singled out, but, she said, they were representative of a problem permeating medical research and nonprofit advocacy groups. “The relationship between drug companies and nonprofits is a widespread problem. The AHA is in no way worse than any other organization,” she said.
Dr. Franklin agreed, adding that the issue of potential financial conflicts is a serious problem in general, with the emergence of drugs and devices. “We (Washington state) recently reviewed a controversial technology for the treatment of lumbar disc disease. Most of the available literature had been produced by clinicians with substantial interests in the company producing the device. We only discovered this by reviewing SEC [Securities Exchange Commission] records. It was not apparent from the flimsy reporting in the articles themselves.”
This said, however, Dr. Franklin and other neurologists interviewed for this article maintain that the AHA guidelines – and the NINDS rt-PA trial data that are cited – were not necessarily influenced by drug company money.
“The guidelines are meant to be helpful, but they don't set stringent rules for what neurologists should or should not do. The guidelines should leave physicians room to exercise their judgment – to practice the art of medicine,” Dr. Franklin said.
FOR FURTHER READING
- “The AHA 2000 Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care.” Circulation 2000; 102 (suppl I): 1204-1216;
- The complete collection of letters in response to the BMJ article are posted online at http://bmj.com/cgi/eletters/324/7339/723#21631;
- Published data by the NINDS rt-PA Stroke Group – New England Journal of Medicine 1995; 333: 1581-1587; Neurology 2000; 55: 1649-1655;
- The AAEM Work Group on Thrombolytic Therapy in Stroke position statement of the American Academy of Emergency Medicine is online at www.aaem.org; click on “position statement;”
- “Guidelines for Thrombolytic Therapy for Acute Stroke: A Supplement to the Guidelines for Management of Patients with Acute Ischemic Stroke (from the Special Writing Group of the Stroke Council of the AHA).” Circulation 1996; 94: 1167-1174;
- “Thrombolytic therapy for acute ischemic stroke.” Canadian Journal of Emergency Medicine 2001; 3: 183-185;
- “Anticoagulants and Antiplatelet Agents in Acute Ischemic Stroke: Report of the Joint Stroke Guideline Development Committee of the American Academy of Neurology and the American Stroke Association.” Neurology 2002; 59: 13-22; Stroke 2002; 33: 1934.
HOW STRENGTH OF EVIDENCE WAS RANKED INGUIDELINES PROCESS
- Class I: Evidence is provided by one or more well-designed, randomized, controlled clinical trials.
- Class II: Evidence is provided by one or more well-designed, observational clinical studies with concurrent controls (for example, case-control and cohort studies).
- Class III: Evidence is provided by expert opinion, case series, case reports, and studies with historical controls.