to the editor: Dr. Selman and colleagues recently presented thrombolytic agents as the new standard of care in the early management of ischemic stroke.¹ I believe the significant risk of early intracerebral hemorrhage makes the establishment of community-wide protocols premature.

In the Cochrane Database of Systemic Reviews,² Wardlow and colleagues searched the world literature and critically reviewed the 12 randomized trials that studied thrombolytic therapy administered within 14 days of ischemic stroke (see figure). This review presents a number of concerns. Three of the four studies using streptokinase were interrupted because of unacceptably high rates of early intracerebral hemorrhage. Because of the heterogeneity of studies, it is not clear whether the increased risk of early bleeding complication was specific to streptokinase or was caused by other variables. Only one of the 12 studies looked at concomitant use of aspirin or heparin in a randomized fashion, showing significant increase in early mortality with the group using streptokinase and aspirin. Three studies published data on mortality during the first two weeks following stroke, showing a 21.8 percent death rate in the treatment group versus 12.3 percent in the control group. Two of the 12 studies randomized patients to start thrombolytic therapy within three hours. In these studies, the five-fold increase in early fatal intracranial hemorrhage (6 percent in the treatment group versus 1 percent in the control group) was significant. This early treatment group was strongly influenced by the large number of patients contributed by the National Institute of Neurological Disorders and Stroke (NINDS) trial,³ which showed favorable outcomes in death and disability at follow-up in the treatment group.

Many unanswered and important questions remain. What is the appropriate dose, timing, and agent of choice in thrombolytic therapy to ensure improved outcomes while minimizing the risk of intracerebral hemorrhage? Is the use of aspirin a risk factor for intracranial bleeding with thrombolytic therapy? Many patients at risk for stroke take aspirin or warfarin (Coumadin) prophylactically, and the question of whether antithrombotic agents in the preceding 24 hours should preclude thrombolytic therapy has not been answered. Can the promising results of the NINDS trial be reproduced by other studies? Can thrombolytic therapy be successfully used outside of tertiary care settings?⁴

Clinicians and patients are eager for a therapy that improves the tragic consequences of stroke, and thrombolytic agents appear promising. However, more information is needed before we commit the enormous community resources required to enact protocols currently proposed by specialty organizations. Education for individual patients and the public should include a balanced review of the risks and uncertainties of treatment in the discussion of “clot buster” therapy for stroke. Patients given tissue plasminogen activator (t-PA) or other thrombolytic agents as therapy for acute stroke should be enrolled in a registry or randomized trial to hasten the development of a sound scientific footing for the use of these agents in stroke.

The use of evidence-based health care recommendations is crucial if we are to use medical resources wisely and limit unsuccessful interventions for our patients. The Cochrane Database of Systematic Reviews is a valuable resource in this effort.

in reply: Although we would like to thank Dr. Killeen for her comments on our article, we take exception to her suggestions and conclusions. We believe that, according to the guidelines established by the American Academy of Neurology and the American Heart Association, which were developed on the basis of the NINDS recombinant tissue plasminogen activator (rt-PA) stroke study, ¹ if the clinical presentation and the computed tomographic scan are consistent with the diagnosis of acute stroke within three hours of onset, then the patient must be considered a candidate for thrombolytic therapy.

Dr. Killeen implies that meta-analysis of thrombolysis data indicates that the significant risk of early fatal intracerebral hemorrhage makes the establishment of community-wide protocols premature. We believe that conclusions based on meta-analysis of 12 randomized thrombolytic trials will be misleading because the trials vary in patient population, agents used, concomitant therapies, and in the critical issue of time. As noted in a recent editorial, “It is impossible to say, on the basis of present evidence alone, whether the results of a large randomized, controlled trial or those of a meta-analysis of many small studies are more likely to be close to the truth.”² It must be remembered that the FDA was persuaded to approve rt-PA for treatment of acute stroke on the basis of the results of the NINDS stroke study¹ Fewer than 4 percent of the patients in the NINDS stroke study were eligible for thrombolytic therapy due to time delays^1,3 Only with the establishment of community-wide education programs and protocols will there be a sufficient number of patients available to answer the questions that Dr. Killeen poses.

The bioethics of thrombolysis have been discussed by Furlan and Kanoti, who note, “There can be no completely satisfying answer to the question, ‘Under what current circumstances can we justify exposing ischemic stroke patients to an increased acute risk of hemorrhage and death from thrombolysis?’ as long as the data are incomplete.”⁴ We do not wish to minimize the risk of hemorrhage. The incidence of 6.4 percent symptomatic hemorrhage in the NINDS stroke study must not be overlooked, nor must it be considered a deterrent to prudent use of an appropriate therapeutic tool for selected patients. As in many other therapies, the physician's responsibility is to present patients or their representatives with the choices, and then to provide the preferred therapy as well as possible.

We would argue that physicians and patients must learn about intravenous thrombolysis and the need to regard stroke as a medical emergency, so that recognition, transportation and treatment can be accomplished within the existing time constraints for safe administration of this therapy. ³ We do not see justification for delaying the mobilization of resources that are required for implementation of thrombolytic therapy. Dr. Killeen's claim that it would not be a wise use of our medical resources to do otherwise is not supported by our analysis of the cost-effectiveness of thrombolytic therapy⁵ Whether the 11 percent to 13 percent absolute benefit justifies the use of thrombolytic therapy in the face of the 6.4 percent symptomatic hemorrhage rate¹ must be individualized for each patient. Without citywide protocols for evaluation and transfer, however, most patients will never be permitted this opportunity.