Correspondence
Should Thrombolytic Therapy Be the First-Line Treatment for Acute Ischemic Stroke?
N Engl J Med 1998; 338:761-763March 12, 1998
- Article
To the Editor:
Treatment of acute stroke seems to be controversial, as the Clinical Debate between Caplan et al. and Grotta (Oct. 30 issue)1,2 shows. Caplan and colleagues, on the con side, and Grotta, on the pro side, all stand firmly on the same side of the main argument — namely, that patients with acute stroke should receive aggressive treatment. In the United States today, the real alternative to therapy with tissue plasminogen activator (t-PA) is no therapy at all. Readers should not lose sight of this point in wading through the pros and cons of a specific treatment such as t-PA.
As for the specific choice of therapy, there is more clinical evidence in support of Grotta's advocacy of t-PA. First, the t-PA study by the National Institute of Neurological Disorders and Stroke (NINDS)3 provides much clearer statistical evidence in favor of t-PA than do the more anecdotal reports cited in favor of intraarterial urokinase. Second, the subgroup analysis in the t-PA study showed a benefit even in patients with small infarcts likely to be related to disease in small, penetrating arteries, which the opponents of t-PA therapy argue would not respond to t-PA. Such patients would not be found to have occlusion on magnetic resonance or conventional angiography and would hence not be candidates for intraarterial therapy. For this group, t-PA is a treatment offered as an alternative to no treatment. Third, even patients with large-vessel or embolic strokes had better responses to t-PA than to placebo. Those who recommend a more selective approach, with some patients receiving intraarterial therapy, should pursue randomized trials comparing this approach with intravenous t-PA.
Editor's note: Dr. Kirshner receives research funding from Genentech, the manufacturer of t-PA.
3 ReferencesHoward S. Kirshner, M.D.
Vanderbilt University School of Medicine, Nashville, TN 372121
Caplan LR ,Mohr JP ,Kistler JP ,Koroshetz W . Should thrombolytic therapy be the first-line treatment for acute ischemic stroke? Thrombolysis -- not a panacea for ischemic stroke. N Engl J Med 1997;337:1309-10, 1313
Full Text | Web of Science | Medline2
Grotta J. Should thrombolytic therapy be the first-line treatment for acute ischemic stroke? t-PA — the best current option for most patients. N Engl J Med 1997;1310-2, 1313.
3
The National Institute of Neurological Disorders and Stroke rt-PA Stroke Study Group
Full Text | Web of Science | Medline
To the Editor:
The Clinical Debate over intravenous recombinant t-PA for acute ischemic stroke reflects continued concern about an approach in which the outcome may appear to violate the precept, “First, do no harm.” Thrombolytic therapy does significantly increase the risk of intracranial hemorrhage in patients with stroke. One of the critical lessons of the European trials of thrombolytic therapy for stroke is the importance of careful selection of patients, and in fact, the failure of the European Cooperative Acute Stroke Study (ECASS)1 was due in part to the number of misrandomized patients. Recombinant t-PA was of benefit in an efficacy analysis of the ECASS target population, but when the data from all the patients (including those who were misrandomized) were analyzed, there were unacceptably high rates of intracranial hemorrhage.
Caplan and his coauthors are correct in reminding us that information from clinical trials must be adapted to individual patients. In that regard, continued accrual of data from phase 4 or postmarketing studies of recombinant t-PA should allow us to define eligibility criteria better in order to individualize further the treatment of acute stroke. Nevertheless, given the limited diagnostic information available at most institutions in our current framework (which consists of limited therapeutic options and a short time for treatment), we should make thrombolytic therapy available to as many eligible patients as possible with the proviso that informed consent with regard to the risks and benefits of treatment be obtained.
Intravenous recombinant t-PA is indeed not a magic bullet. Some patients will fare better with t-PA, some will have no benefit, and unfortunately some will fare worse. In an ideal world, we would be able to individualize our treatment options for patients with stroke rapidly, using both expert clinical acumen and appropriate ancillary studies. We should therefore strive for an environment in which patients with stroke can be treated on an emergency basis. In this regard, I support the calls to consider, as a matter of national public policy, creating a stroke system modeled after the levels of care for patients with trauma endorsed by the American College of Surgeons.2 In this model, with triage, patients can be moved through the emergency medical systems to clinical centers of excellence within a very short time. Given the complexities of care for stroke, such a system could only optimize the care of patients. In this regard alone, the approval of recombinant t-PA has provided us with a powerful stimulus to further the development of an infrastructure for the care of patients with acute stroke.
2 ReferencesMichael J. Schneck, M.D.
Rush Medical College, Chicago, IL 606121
Hacke W ,Kaste M ,Fieschi C , et al. Intravenous thrombolysis with recombinant tissue plasminogen activator for acute hemispheric stroke. JAMA 1995;274:1017-1025
CrossRef | Web of Science | Medline2
Gordon DL. “From the Editor” Stroke Council Newsletter. No. 1, 1997.
To the Editor:
As a practicing emergency care physician for 17 years, I am concerned about the clinical decision making that will be passed on to the emergency care doctor if no in-house neurologist is readily available. This is the situation in many nonteaching hospitals, especially after regular hours. Dr. Grotta's approach is especially worrisome. As Caplan et al. correctly state, there are many “conditions that can masquerade as stroke.” Many of us who are not formally trained in neurology might have difficulty differentiating, say, a hemiplegic migraine from stroke or a patient with a transient ischemic attack that may clear in a few hours from a possible candidate for t-PA therapy. In addition, there are metabolic and functional causes of weakness, postictal states such as Todd's paralysis, exposure to toxins, hypertensive disorders, and many other conditions that mimic stroke. In one study that involved 411 patients, members of a stroke team misdiagnosed stroke nearly 20 percent of the time before a computed tomographic (CT) scan was obtained.1 Caplan et al. rightly emphasize that expediency should not compromise the specificity of the diagnosis.
1 ReferencesJoseph P. D'Addesio, M.D.
6103 Rapid Creek Ct., Houston, TX 773451
Libman RB ,Wirkowski E ,Alvir J ,Rao TH . Conditions that mimic stroke in the emergency department: implications for acute stroke trials. Arch Neurol 1995;52:1119-1122
Web of Science | Medline
Author/Editor Response
The authors reply:
To the Editor: I agree with the comments of Drs. D'Addesio and Schneck. Dr. Kirshner states that the only treatment for acute stroke is t-PA. Control of risk factors, optimal management of blood pressure and blood volume, surgery, angioplasty, and use of platelet antiaggregants and standard anticoagulants are alternative treatments. The causes of stroke are heterogeneous; treatment must be tailored to the causative problems in individual patients. Fortunately, the nature of the brain and vascular lesions can now be quickly and safely determined by experienced clinicians using modern technology. Clinicians have a much greater chance of doing good and not doing harm if they know what is wrong with the patients they are treating. This means identifying the cause of brain ischemia. No active treatment is preferable to treatment that has as much chance of doing harm as of being beneficial.
The data from the NINDS trial that support the benefit of recombinant t-PA in patients with penetrating artery disease1 are extremely weak. It was impossible within the time constraints of that study (treatment within 90 or 180 minutes after the onset of symptoms and frequent delays in arrival at the hospital) to perform the thorough history taking and neurologic examinations needed to diagnose penetrating artery disease accurately and still obtain CT scans, obtain informed consent, and begin treatment. The findings on initial CT scans are negative in such patients. No vascular studies were performed before treatment, and if any were performed after treatment, they were not reported. The outcome in the placebo-treated patients with a diagnosis of small-vessel disease was far worse than in any other published study and worse than in any other vascular group in this study. This finding speaks strongly to the inaccuracy of the diagnosis. The number of patients (81) was too small for statistical significance; the treatment assignment was unbalanced (51 patients in the t-PA group vs. 30 in the placebo group), and the analysis was post hoc without prior stratification.
Furthermore, there are two strong reasons not to use recombinant t-PA: the prognosis for recovery in patients with penetrating artery disease is excellent without treatment, and autopsy studies have not shown thrombi within penetrating arteries. Why give a potentially risky treatment if the outlook is benign? If I had penetrating artery disease, I certainly would not want recombinant t-PA, and I believe in the golden rule as a guide to medical treatment.
Additional therapeutic trials are needed in patients in whom the underlying cause of brain ischemia has been identified. The administration of thrombolytic agents must be compared with other strategies in patients with defined vascular lesions.
1 ReferencesLouis R. Caplan, M.D.
Beth Israel Deaconess Medical Center, Boston, MA 022151
The National Institute of Neurological Disorders and Stroke rt-PA Stroke Study Group
Full Text | Web of Science | Medline
Author/Editor Response
Drs. Kirshner, Schneck, and D'Addesio raise important points — namely, the need to monitor our experience with intravenous recombinant t-PA as it is used in daily practice, and whether we need to establish regional centers for the treatment of acute stroke to carry out this therapy. The issue is not to rehash the data on the efficacy of recombinant t-PA, but to figure out how the largest number of eligible patients might benefit.
In talking with colleagues at medical centers or in community practice throughout the country and from my own experience in Houston, clinical results have so far largely reflected those reported in the NINDS trial. Papers presented at meetings in the next year should be carefully scrutinized, since they will undoubtedly report such phase 4 data, particularly the success of this therapy outside major stroke centers. Recent articles in the emergency medicine literature1,2 should clarify some of the myths and misunderstandings about intravenous thrombolysis that are prevalent among neurologists and nonneurologists alike.
A consensus conference recently recommended guidelines for the development of stroke-treatment centers.3 The development of such centers should obviate the concern expressed by Dr. D'Addesio. I agree that neurologic backup is a critical component of any comprehensive stroke-treatment center, whether or not patients receive recombinant t-PA. As we establish this therapy in our emergency departments, our emergency medicine colleagues deserve the best that neurologists can offer. Such centers already exist in Dr. D'Addesio's hometown, Houston. My colleagues and I have also established a mobile stroke team, which in collaboration with the emergency and neurologic staffs of various community hospitals in the city, will go to the emergency departments, assist with the treatment of patients, and train emergency department personnel.
However, one of the advantages of intravenous recombinant t-PA therapy for stroke as currently recommended and approved (as opposed to the unproved approach recommended by Caplan et al., which would necessitate extensive pretreatment testing) is that it can be carried out in any hospital with a CT scanner and, most important, with the medical expertise to select patients properly. Obtaining this expertise is absolutely essential but is not difficult. In his letter, Dr. D'Addesio indicates that he himself already knows the subtleties of selecting appropriate patients. One can hide behind the admonition, “First, do no harm,” but the result is, “First, do no good.”
To develop a network of such primary stroke centers able to carry out intravenous thrombolysis, a number of steps must be taken, as outlined in the consensus conference.3 Neurologic backup is necessary, but it is also critical that the essentials of stroke management and the specifics of selecting patients for thrombolytic therapy be integrated more prominently into the education of all primary care and emergency physicians, nurses, and prehospital health care workers.
Cost-effective therapy is available for many patients with acute stroke. The challenge to the health care system is to reorganize the way stroke is managed in terms of teaching, triage, and treatment.
3 ReferencesJames Grotta, M.D.
University of Texas Medical School–Houston, Houston, TX 770301
Haley EC Jr ,Lewandowski C ,Tilley BC . Myths regarding the NINDS rt-PA Stroke Trial: setting the record straight. Ann Emerg Med 1997;30:676-682
CrossRef | Web of Science | Medline2
Kasner SE ,Grotta JC . Emergency identification and treatment of acute ischemic stroke. Ann Emerg Med 1997;30:642-653
CrossRef | Web of Science | Medline3
Marler JR, Jones PW, Emr M, eds. Proceedings of the National Symposium on Rapid Identification and Treatment of Acute Stroke. Bethesda, Md.: National Institutes of Health, 1997.
- Citing Articles (1)
Citing Articles
1
Joanna M Wardlaw, Veronica Murray, Eivind Berge, Gregory J del Zoppo, Joanna M Wardlaw. 2009. Thrombolysis for acute ischaemic stroke. .
CrossRef
