Journal Logo



Reanalysis Fires Up Controversy about Alteplase for Stroke

SoRelle, Ruth MPH

doi: 10.1097/01.EEM.0000722376.68472.fc
    tpa, alteplase, stroke

    A reanalysis of the ECASS III trial found alteplase does not benefit stroke patients 3-4.5 hours after onset, drawing parallels with a previous reexamination that concluded the NINDS trial failed to establish benefit. Those two trials, long considered support for giving alteplase, leave no major support for the drug in stroke.

    Or as Salim Rezaie, MD, wrote on his blog, REBEM EM, “This means there are now ZERO RCTs supporting the practice of thrombolysis in acute ischemic stroke.” (

    Since the study sponsored by the National Institute of Neurological Disorders and Stroke (NINDS) appeared in 1995, some have questioned whether tissue plasminogen activator, now called alteplase, really benefits patients who suffer an acute ischemic stroke. (N Engl J Med. 1995;333[24]:1581;

    The study's authors came down on the positive side, concluding that the patients who received the drug were less likely to have negative effects from a stroke three months later if they were treated within three hours of the onset of symptoms compared with those who received placebo.

    The ECASS III trial by the European Cooperative Acute Stroke Study investigators pushed the window for treatment to 4.5 hours after the onset of symptoms. (New Engl J Med. 2008;359[13]:1317; In this second study, as in the first, the incidence of bleeding in the brain was higher in those who received the drug.

    The authors of ECASS III concluded: “This reanalysis confirms reports of concern that the baseline imbalances introduce such a risk of bias that conclusions of efficacy based on ECASS III data cannot be considered reliable. Clinicians, patients, policymakers, systematic review authors, clinical practice guideline developers, and drug regulators should reconsider interpretations and decisions regarding management of acute ischaemic stroke that were based on ECASS III results.”

    Baseline Imbalance

    The objections to the first study did not go unnoticed. The National Institute of Neurological Disorders and Stroke charged an independent committee with evaluating it, and found “a statistically significant and clinically important benefit of t-PA treatment, despite an increased incidence of symptomatic intracerebral hemorrhage in t-PA treated patients and subgroup imbalances in baseline stroke severity.” (Stroke. 2004;35[10]:2418; This imbalance was the major factor in prompting another evaluation of the study, the researchers said.

    The reanalysis of the ECASS III trial looked at similar issues. (BMJ Evidence-Based Medicine doi:10.1136/bmjebm-2020-111386), but came to a different conclusion. Researchers led by Brian Scott Alper, MD, now the chief medical knowledge officer at EBSCO, evaluated ECASS III because no one was looking at the statistical methodology despite Dr. Alper's efforts to point out the study's flaws. “My job is nothing to do with tPA or stroke,” he said. ECASS III was the only study of many that had a positive result, meaning patients who went to the hospital as many as 4.5 hours after a stroke showed a small benefit.

    The researchers found baseline imbalances in the two arms of patients, and those in the placebo group were twice as likely to have had a prior stroke, said Daniel Fatovich, MBBS, PhD, an emergency physician at Royal Perth Hospital in Western Australia. The stroke scores differed as well.

    “There was a failure of randomization,” he said. “One group had more severe disease than the other.” The analysis by the group led by Dr. Alper took all the imbalances into account, and used a variety of different statistical approaches to evaluate the study and its findings. Only the first approach produced the original result, Dr. Fatovich said.

    Sicker Placebo Group

    Dr. Alper said his group found in that case they could only achieve the positive finding if they used a modified National Institutes of Health Stroke Score. A major problem was seen at baseline—the placebo group was sicker. The original authors said they had done an adjusted analysis, and there was no need for concern, he said, adding that the adjustment was not well explained.

    “ECASS III was flawed because of the imbalance at baseline,” said Dr. Fatovich, who was not involved in either study or their analyses. The risk of brain bleeds remained the same in the reanalysis.

    The bottom line is that any benefit of tPA between 3-4.5 hours is much less than it is at 0-3 hours, said James C. Grotta, MD, the director of stroke research at the Clinical Institute for Research and Innovation at Memorial Hermann-Texas Medical Center, and the director of the Mobile Stroke Unit Consortium in Houston. Dr. Grotta was the leader of the NINDS trial at the University of Texas Health Science Center at Houston. “And the difference isn't just linear, it is really exponential,” he said. “In fact, the benefit is even greater at zero to one or two hours. And what we are seeing with our mobile stroke unit is that when we are treating patients in the first hour or so after onset, which is possible in that setting, the results show an even greater benefit.

    “I think the authors are saying that at 3-4.5 hours, the benefit becomes marginal,” Dr. Grotta said. “But now there are multiple studies that have enrolled patients out to six hours with tPA and you pool the data together, the signal becomes much clearer and there is still benefit to four-and-a-half hours and maybe even at five hours, but it becomes much less between four-and-a-half to five hours.”

    He said trials will always have imbalances at baseline. “The initial authors did the best they could to balance them, and these guys want to adjust the analysis,” Dr. Grotta said. “Rather than focus on that, it's better to get more numbers and then you can become more confident. I would point to all those papers that show that there is a decay of benefit as you get out to four-and-a-half hours, but there is still benefit.”

    Dr. Fatovich agreed. “No study is perfect, and every study has its biases and problems. You look at the results and come to a reasonable set of conclusions.”

    But Dr. Alper expressed concern that institutions and groups such as the American Heart Association and the American Stroke Association have made thrombolysis with tPA an important part of their guidelines. “These people are making guidelines on the basis of data that were not given a proper analysis,” he said. “Will they now say, ‘We had it wrong for a decade?’”

    Selective Analysis

    The American Heart Association and the American Stroke Association guidelines emphasize the administration of alteplase quickly to ensure patients receive the most benefit based on NINDS and ECASS III. The American College of Emergency Physicians gives it a level B classification, meaning its use is buttressed by moderate clinical certainty.

    Dr. Alper described in a blog post the lengths to which he and his colleagues had to go even to conduct their reanalysis after the inclusion of the treatment in the AHA/ASA guidelines. When they asked the lead ECASS III investigators if they wanted to collaborate on a multivariable analysis, they demurred but said the data were available from the sponsor. Obtaining that data and the subsequent clarifications took years, which is why the reanalysis and the findings didn't appear until this year.

    Reproducing the authors' reported adjusted analysis “took a combination of three distinct deviations from what would have been an optimal protocol if a protocol for adjusted analysis had existed,” Dr. Alper wrote in the BMJ Evidence-Based Medicine Spotlight post, “The Dangers of Selective Analysis: Has Stroke Treatment Been Misguided for a Decade?” (May 28, 2020; Clearly, he said, further studies and bigger numbers are needed.

    Jerome Hoffman, MD, a professor of medicine and emergency medicine at the University of California, Los Angeles, pondered the question of whether to treat ischemic stroke with tPA in 2006. At that time, only the NINDS trial had shown benefit, and Dr. Hoffman pointed out the questionable data behind the therapy and noted that a “handful of other RCT has confirmed such benefit, and several of them clearly found overall harm. (Emerg Med Australas. 2006;18[3]:215;

    “I am also often asked what would I do if I, or a member of my family, had a stroke and ‘qualified’ for thrombolysis, according to the NINDS guidelines,” Dr. Hoffman wrote. “Here, too, I do not have an easy answer, because much though I believe that such drugs may well ultimately be shown to do more harm than good, I also know it would be tragic to fail to use a therapy that could produce dramatic benefit in an individual patient, for such a devastating disease. I therefore typically answer along the lines of ‘I wish I knew.’”

    Share this article on Twitter and Facebook.

    Access the links in EMN by reading this on our website,

    Comments? Write to us at [email protected].

    Ms. SoRellehas been a medical and science writer for more than 40 years, previously at the University of Texas MD Anderson Cancer Center, the Houston Chronicle, and Baylor College of Medicine. She has received more than 60 awards, including the Texas Human Rights Foundation Award. She has been a contributor to EMN for more than 20 years.

    Copyright © 2020 Wolters Kluwer Health, Inc. All rights reserved.
    • TheSGEM2:51:05 PMHow many years will it take to get the right answer?