The American College of Emergency Physicians and the American Academy of Neurology jointly issued a new clinical policy on thrombolytics for stroke, but the new statement has done little to resolve one of the most troubling rifts in emergency medicine.
The recommendations are nothing new in thrombolytics-for-stroke circles, first drawing attention after the original NINDS trial was published in 1995. (N Engl J Med 1995;333:1581.) The new guidelines, published in the February issue of Annals of Emergency Medicine, stamp an A-level recommendation on treating stroke patients who meet the NINDS criteria with tPA within three hours of symptom onset. They also place a B recommendation on tPA for stroke patients who meet the European Cooperative Acute Stroke Study (ECASS) criteria for treatment between three and 4.5 hours after symptom onset. (JAMA 1995;274:1017.) An introduction to the new policy understates the issue: “The use of IV tPA for stroke has been one of the most contentious medical treatments.”
That's a mild description of the bitter division within the emergency medicine community and between some emergency physicians and their neurologist counterparts. Or as David Newman, MD, the director of clinical research and an associate professor of emergency medicine at Mount Sinai School of Medicine in New York City and a critic of thrombolytics for stroke, put it in a podcast: “It is the biggest, baddest controversy in our field.”
More than 17 years after the first FDA recommendation, many in the emergency medicine community still adamantly say the treatment is too risky and has too little potential for benefit to implement in their emergency departments. None of the studies since the first NINDS (National Institute of Neurological Disease and Stroke) trial has persuaded them the treatment has benefit.
Those who back the treatment point out that it is the only approved treatment for acute ischemic stroke and insist that it is beneficial, at least in patients who get to the emergency department quickly, who receive diagnostic CT scans, and who meet the criteria for treatment. Thrombolysis for stroke is not without risk, of course, but the promise of reducing disability after stroke outweighs that benefit, they insist.
Peter Sandercock, DM, a professor of medical neurology and the director of Edinburgh Neuroscience at the University of Edinburgh in Scotland and the director of the most recently reported stroke-thrombolytic trial IST-3 (the third international stroke trial; Lancet 2012;379:2352), said the results from his study, which expanded the treatment window to six hours, showed that the treatment “does not increase mortality.” But critics said that was a subjective finding, especially when no statistical difference was found in survival between the two groups and when 104 patients in the treatment group experienced intracranial hemorrhage compared with 16 in the control group.
IST-3 studied 3,035 patients at 156 hospitals in 12 countries, and randomized 1,515 patients to receive tPA and 1,520 to the control group; 1,617 patients were over age 80. Mortality in the first seven days of treatment was higher in the treatment group (163 vs. 107) but lower in the treatment group between seven days and six months. A total of 408 in the treatment group (27%) and 407 in the control group (27%) had died at six months.
Dr. Sandercock said physicians and health organizations around the world are generally accepting the evidence, despite some opposition in Europe and the United States, and are reconfiguring the way services are delivered to those with stroke. They are making sure “people who might benefit are made aware it is there, and [that] their local hospital will have a system to make sure if they get there in time their chances will be maximized,” he said.
Perhaps the Cochrane Review of thrombolysis for acute ischemic stroke best describes the dilemma in its summary: “Overall, thrombolytic therapy appears to result in a significant net reduction in the proportion of patients dead or dependent in activities of daily living. This overall benefit was apparent despite an increase both in deaths [evident at seven to 10 days and at final follow-up] and in symptomatic intracranial haemorrhages. Further trials are needed to identify which patients are most likely to benefit from treatment and the environment in which thrombolysis may best be given in routine practice.” The Cochrane analysis evaluated 26 trials of 7,152 patients who received a variety of thrombolytics — streptokinase, recombinant tissue plasminogen activator, recombinant pro-urokinase, and desmoteplase. (Cochrane Database Syst Rev 2009 Oct 7;:CD000213.)
In contrast, experts at the www.thennt.com, a website that independently evaluates therapies and accepts no outside funding or advertisements, concluded no consistent or reliable benefit could be found in using thrombolytics for stroke after reviewing all of the 12 high-quality trials. They pointed out that five of those studies found no benefit of tPA for stroke, four found it harmful and stopped the trials early, and two found benefit.
The NNT says one of those two trials, NINDS-2 (N Engl J Med 1995;333:1581), should have compared the improvement in functional and stroke scores between groups (the outcome used in the closely related NINDS-1 study that found no benefit), but that the researchers instead inexplicably reported a less meaningful, dichotomous favorable vs. unfavorable outcome. This reported benefit, the website reported, was an NNT of 8 for a “favorable” outcome. The site also said the other study that found benefit, ECASS III, also reported more favorable outcomes with tPA but no difference in mortality, with an NNT of 15 for a “favorable” outcome. (See FastLinks.)
Dr. Newman, himself an expert in clinical science and one of the driving forces behind NNT, said “emergency physicians feel threatened, put on the spot. We're between a rock and a hard place. There is some chance we may kill our patients. The Cochrane Review found that thrombolytics increases mortality by one percent, and you could end up causing a severe bleed in the brain. No one wants to be the doctor who gives something harmful unless, in the aggregate, it ends up being helpful.”
It is important to be able to replicate findings, he said, because doctors rely on that confirmation. “It's how we have found out about many major frauds in science. We tend not to believe one study in isolation. NINDS, the one study that suggests a benefit in the zero to three-hour window, has never been replicated,” he said. “Those who believe in the tenets of science have a real problem with thrombolytics. Everyone holds out hope that somewhere there are folks who might benefit. But it is also true that by most of the scientific standards we hold ourselves to, we haven't found those people yet. Thrombolytics have not met the test.”
Arthur Pancioli, MD, a professor and the chair of emergency medicine at the University of Cincinnati College of Medicine, is a proponent of tPA for stroke, and has led the field in evaluating its use. He said he understood why some emergency physicians resist using the treatment, often because they are under pressure to use the thrombolytic without full backup. “One element required is a system in place for rapid diagnostics, rapid reading of those scans, and an expert consultant to ask the hard questions. As an emergency physician, I would expect that if my system wanted me to do this, I would have the backup and resources necessary.”
The backlash after the release of NINDS was expected because emergency physicians were told to use the new findings without those systems in place, said Dr. Pancioli, also a director of the University of Cincinnati stroke team. “Some of the most vociferous emergency physicians out there came out with potent arguments and eloquently questioned the science. Some of the questions were very good; some were just reactionary. It's time to be done arguing and ensure that the systems are in place and the candidates have the potential to receive the therapy,” he said.
Those resources, however, are not available everywhere. “As an emergency physician, I would really aggressively be against an expectation of doing this if I don't have the help available that I need,” Dr. Pancioli said, noting that he reads reports daily of patients who were not treated because resources were not available.
“Every hospital in suburbia cannot be a stroke center. If they cannot be one, then they have to know the facility to which they need to defer. How do I make sure there are no failed links in the chain?” he asked. “To treat a patient has risk. We know that. To not treat a patient has equal or greater risk. What you have done is raise their opportunity to have a better outcome.”
Dr. Pancioli admitted that many questions remain about treating patient within the 4.5-hour limit. “The FDA didn't buy the data for change, but we have randomized data to show that you can still offer benefit to patients.” The criteria will have to go beyond a simple time point to subselect the patients for whom the treatment works best, he said.
Edward C. Jauch, MD, a professor and the division director of emergency medicine and a professor of neurosciences at the Medical University of South Carolina, was the lead author of the American Heart Association-American Stroke Association “Guidelines for the Early Management of Patients with Acute Ischemic Stroke,” just issued in the journal Stroke (2013 Jan 31. [Epub ahead of print]; see FastLinks.) He said that new document is not a major change but evolutionary, calling for a cohesive team of health providers including emergency medical services personnel, emergency physicians, and stroke experts. “None should work in isolation. Having just one is not sufficient to treat patients well,” he said.
After the European Cooperative Acute Stroke Study (ECASS III) expanded the time window to 4.5 hours, the drug was approved for the longer time period in Europe. (N Engl J Med 2008;359:1317) A 2009 AHA practice advisory said those new data led heart association experts to feel it was reasonable to treat patients using those criteria, being careful about which patients were selected.
The FDA refused to approve the drug for the expanded time period, but Dr. Jauch and his colleagues talked to Genentech, the drug's manufacturer, and decided the issues were not based on safety. “We felt it was appropriate to leave the recommendation intact as we had in 2009,” he said.
The obvious next step for them, Drs. Pancioli and Jauch said, was to institute systems to speed stroke treatment. Phillip Scott, MD, an associate professor of emergency medicine at the University of Michigan School of Medicine in Ann Arbor, attempted to determine how to do that with the INSTINCT trial, which was just published in Lancet Neurology this year. (2013;12:139.) Their attempt to increase use of tPA for stroke in community hospitals did not result in a statistically significant change. Twenty-four hospitals took part in the study; 12 in the study group went through a process to identify barriers to treatment and find ways to overcome those problems while the other 12 served as controls. The proportion of patients admitted and treated with tPA increased more between the pre-intervention and post-intervention periods in the intervention hospitals than in the control hospitals, but the difference was not significant. The study was disappointing, Dr. Scott said, but he added that it was one way to start moving toward more widespread use of the treatment in the community.
Dr. Newman of thennt.com pointed out that that the federal Centers for Medicare and Medicaid Services has also created incentives for treatment in a stroke center, stoking interest in those centers. “It's all been generated to support the acute care of the stroke patients,” he said. “Once you get excited about it, it's often hard to remember or care about all of this ‘science' stuff.’
He acknowledged that stroke centers have had successes, with “great rehabilitation, early stroke care, investigation into diagnosis, and the prevention of future strokes.” They have also enhanced prevention of aspiration, a major factor in morbidity in post-stroke periods.
But one longtime critic of using tPA for acute ischemic stroke said medicine should not move toward widespread use when studies do not support that treatment. Jerome Hoffman, MD, a professor of emergency medicine at the David Geffen School of Medicine at the University of California Los Angeles, said he is particularly critical of the new guidelines. “Who got to write these, and how were they chosen?” he asked. “Why is there no one on the list who is an emergency physician skeptical of this?”
Despite how tPA proponents present the data, Dr. Hoffman said all studies of tPA for stroke have yielded negative results, and none proves that it should be given to patients, in or out of any time window. Some support tPA because it is in their interest to do so, said Dr. Hoffman, often because they receive compensation for serving as drug company consultants. “There are true believers who have a semi-religious zeal about this, particularly among neurologists. When you want to believe something, it's hard to believe the opposite. It takes a lot of proof before you say, ‘Maybe it's wrong.’”
Dr. Hoffman said tPA is very likely harmful, at least in community hands, and the chance that it is beneficial is remotely small. “It's not beneficial to a degree that could possibly matter,” he said.
This is a perfect example of the ways in which society does things that “are really bad that we shouldn't be doing, allowing ourselves to be led because of other motives that have nothing to do with health care. I think it's very sad, antiscientific, and disingenuous,” Dr. Hoffman said.
Dr. Newman said one way to dispel questions about the treatment would be to replicate NINDS in multiple centers with larger numbers. “That, by itself, would be revolutionary and wonderful,” he said. The trial should be powered to look at subgroups and examine outcomes at the granular level. “It should be funded by the federal government like NINDS was. It is fair to ask for replication.”
Another study should look at the outcomes from stroke centers, evaluating all comers, including those who do and do not receive thrombolytics, he said. “There is a standard our public holds us to. That standard is science. When we offer them a therapy that has the potential to do great harm, it must have been proven with great certainty to be beneficial overall. Few people would say we have met that standard in the science for this intervention,” said Dr. Newman.
Click and Connect! Access the links in EMN by reading this issue on our website or in our iPad app, both available on www.EM-News.com.
- Read the joint ACEP-AAN clinical policy, “Use of Intravenous tPA for the Management of Acute Ischemic Stroke in the Emergency Department,” at http://bit.ly/157tihK.
- The American Heart Association-American Stroke Association policy, “Guidelines for the Early Management of Patients with Acute Ischemic Stroke,” is available at http://bit.ly/ASAstatement.
- Review the NNT's review of thrombolytics for strokes at http://bit.ly/NNTthrombolytics.
- Dr. David Newman's blog post about the IST-3 study is available at http://bit.ly/NNTdelusions.
- Listen to David Newman, MD, and Ashley Shreves, MD, dissect the tPA-for-stroke issue in a SMART EM podcast at http://bit.ly/SMARTtpa.
AAEM, SAEM on tPA for Stroke
The policies from the American Academy of Emergency Medicine and the Society for Academic Emergency Medicine vary from the one just released by ACEP and the AAN. SAEM, in fact, now has no policy; it withdrew a previous one urging caution when using thrombolytics for stroke. (Read EMN's 2010 article: http://bit.ly/YbQ9pp.)
AAEM, on the other hand, found that the “objective evidence regarding the efficacy, safety, and applicability of tPA for acute ischemic stroke is insufficient to warrant its classification as standard of care.” The academy advises emergency physicians to use their discretion in its use until additional evidence clarifies the controversies. Read AAEM's policy at http://bit.ly/Wlwm9N.