Share this article on:

Viewpoint: SAEM Retires Policy Statement on Lytic Therapy in Stroke

Chan, Yu-Feng Yvonne MD; Levine, Steven R. MD; Silbergleit, Robert MD; Jagoda, Andy S. MD

doi: 10.1097/01.EEM.0000370754.94295.c9


In January 2009, at the request of the Society for Academic Emergency Medicine (SAEM) Neurological Emergencies Interest Group, the Society's Board of Directors reviewed and then rescinded the policy statement “Clinical Policy Statement on Lytic Therapy in Stroke.” Although there had been no precedent for SAEM making statements on specific clinical practice issues, this policy was adopted in 2003 to support members who felt they were being put at medicolegal risk, and were being pressured on this issue by professional organizations outside of emergency medicine. The policy statement questioned the benefit and safety of thrombolysis for acute ischemic stroke, said the therapy was not the “standard of care,” and called for more research.

The Neurological Emergencies Interest Group asked for the review because clinical research since the policy statement was published has continued to provide additional corroborative clinical data on the efficacy of thrombolysis in acute ischemic stroke. Specifically, the European Cooperative Acute Stroke Study (ECASS) 3 confirmed an improved rate of neurological recovery after patients with stroke were treated with tPA. (N Engl J Med 2008;359[13]: 1317.) ECASS 3 was a randomized, double-blind, placebo-controlled clinical trial of more than 800 patients with acute stroke treated with intravenous rtPA or placebo between three and 4.5 hours after the time last seen normal. Published in the New England Journal of Medicine on Sept. 25, 2008, the trial showed an absolute benefit of seven percent in good neurological outcomes at three months post-stroke in patients treated with rtPA without any increase in mortality. (N Engl J Med 2008; 359[13]:1317.)

ECASS 3 was not only consistent with the prior positive NINDS tPA Stroke Trial, it was consistent with the prior negative ECASS and ATLANTIS randomized controlled trials. (N Engl J Med 1995;333[24]:1581; JAMA 1999; 282[21]:2019; JAMA 1995;274[13]:1017; Lancet 1998;352[9136]:1245.) Those trials had all shown consistent trends toward absolute improvements with tPA, but were negative on their primary outcomes because they had only been powered to detect larger benefits like those demonstrated in the NINDS trial, in which treatments were given much earlier after stroke onset. In 2004, Hacke published a pooled analysis of more than 2,700 patients from six randomized controlled trials of intravenous rtPA given up to six hours post-stroke that predicted a benefit for treatment with tPA that diminished with time from onset of stroke. (Lancet 2004;363[9411]: 768.) ECASS 3, the largest stroke thrombolysis trial ever performed, was designed to have the power to detect the more modest benefit predicted by the pooled analysis in the three- to 4.5-hour treatment window, and exactly confirmed the effect proposed by that model and the prior randomized controlled trial data.

It is important to note that stroke thrombolysis trials are all about finding and testing a specific protocol and way of identifying those patients in whom the therapy will be more beneficial than harmful. As in the previous ECASS trials, ECASS 3 excluded certain patients at highest risk for hemorrhagic complications including those over age 80, those with very high stroke severity scores (NIHSS >25), and those with early signs of massive infarct on initial CT. ECASS 3 was mandated by the EMEA, the European equivalent of the FDA, as a condition of approval of 0–3 hour rtPA for stroke in the European Union, and it was funded by the manufacturer. (N Engl J Med 2008;359[13]:1317.)

The current data, therefore, show that when appropriately selected and treated, one in three patients will benefit from IV rtPA given within three hours of symptom onset. Approximately one in six will benefit when treated between three and 4.5 hours. Approximately one in 30 will be harmed when treated within three hours, and approximately one in 37 will be harmed when treated three to 4.5 hours after symptom onset. (Stroke 2009;40[7]:2433.)

What about other phase IV data? Nonrandomized observational data on community implementation of this therapy predominantly demonstrate outcomes and safety consistent with that seen in the clinical trials, but included some reports where efficacy and safety were markedly worse than in the clinical trials. An NIH-sponsored cluster randomized trial of rtPA stroke protocol and education implementation in the community is ongoing (INSTINCT, Identifier: NCT00349479).

Two independent appropriately powered, randomized, placebo-controlled trials of intravenous thrombolysis in carefully selected patients with acute ischemic stroke now demonstrate clinically and statistically significant improvements in the proportion with favorable neurological outcomes with treatment without any increase in mortality. Pooled analyses of data from these and other randomized controlled trials, many of which were underpowered for the effect size ultimately demonstrated, are consistent with the data from the positive trials, and also confirm that earlier treatment significantly increases the size of this effect on outcome. High-quality U.S. community implementation data are forthcoming.

It is important that SAEM as an academic society maintain positions consistent with the scientific literature and best practice policies and procedures. The SAEM board rescinded the standing policy statement, and this time, the society no longer has a policy statement for or against the use of thrombolytics or stroke centers.

According to Jill M. Baren, MD, SAEM president, and Dr. Katherine Heilpern, the immediate past president, the board has returned to its traditional view that it is not the role of the society to take official positions on specific clinical care policies. The mission of the society is to advance patient care through research and education. The SAEM Neurological Emergencies Interest Group hopes the new data on thrombolysis in stroke will do just that.

Dr. Chan

Dr. Chan

© 2010 Lippincott Williams & Wilkins, Inc.