Treating Ischemic Stroke As an Emergency

NEUROLOGICAL REVIEW Treating Ischemic Stroke as an Emergency

Harold P. Adams, Jr, MD

he success of treatment with tissue plasminogen activator serves as an impetus to approach stroke as a medical emergency; diagnosis and treatment must be accurate and prompt. The initial evaluation should be straightforward and aimed at confirming ischemic stroke as the cause of the patient’s acute neurologic impairments. Until the Tusefulness of diagnostic tests to demonstrate an arterial occlusion is established in emergent management, their application before treatment should not be mandated. Most individuals with acute ischemic stroke will receive their initial, key treatment in a community setting. Thus, strategies for emergent treatment should aim at management of patients whose strokes are diagnosed and first treated by emergency treatment and primary care physicians with the collaboration of neurologists. Arch Neurol. 1998;55:457-461 Ischemic stroke is the most common acute ment also is influenced by the patient’s neurologic illness and a leading cause of condition and his/her wishes. Most pa- death, disability, and human suffering. Fi- tients with acute ischemic stroke will not nancially, it also is an expensive disease. be treated initially by neurologists or stroke Stroke costs the US economy more than care specialists. Because of limited avail- $40 billion per year.1 Thus, successful ability of acute stroke care units, most pa- management of ischemic stroke has vast tients are admitted to community hospi- public health implications. In the past, is- tals that do not have special expertise in chemic stroke was approached with ni- the treatment of cerebrovascular disease. hilism by both the public and health care The primary goal of modern treat- providers.2 This negative attitude is be- ment of acute ischemic stroke is to limit ing abandoned in response to a revolu- or reverse the brain injury so that the pa- tion in the management of stroke. Stroke tient can recover as much as possible. A now is being treated as the life-threat- sustained improvement in neurologic out- ening disease it is. Several factors under- come is the measure to judge responses to pin modern management. The interval treatment.3 Ischemic stroke is a complex from onset until treatment is critical. Is- vascular and metabolic process that evolves chemic stroke usually is caused by throm- over minutes to hours. The concept of the boembolic occlusion of an artery supply- ischemic penumbra emphasizes that al- ing a portion of the brain, and improving though a core of infarcted tissue might not perfusion will be fundamental for suc- be salvageable with any intervention, an cess. The spectrums of neurologic impair- area of dysfunctional, ischemic brain might ments and causes of stroke are broad; as be rescued if a therapy is prescribed a result, the prognosis of patients varies promptly. Experimental studies4-6 amply considerably. Besides treating the stroke demonstrate that the interval from onset itself, management includes measures to of ischemia until initiation of treatment is prevent or control medical or neurologic critical for success for any therapy. The complications, rehabilitation, and thera- window for effective treatment may be only pies to prevent recurrent stroke. Manage- a few hours. The maximum interval is not known but an intervention prescribed From the Division of Cerebrovascular Diseases, Department of Neurology, University more than 24 hours after onset of symp- of Iowa College of Medicine, Iowa City. toms is not likely to be of much benefit.

ARCH NEUROL / VOL 55, APR 1998 457

©1998 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/27/2021 Starting treatment as soon as possible is and will con- chemistry studies, including serum electrolytes and blood tinue to be a driving force in emergent stroke care. glucose level; arterial blood gases, if hypoxia is sus- The push to start treatment quickly is the basis of ef- pected clinically; and cervical spine x-ray films, if the pa- forts to increase public and professional awareness of stroke, tient is unconscious and no medical history is available which is summarized in the term brain attack.7-15 Delays about the onset of neurologic impairments. In addition, in recognition, transport, evaluation, and treatment must the protocol should include components for general emer- be overcome to meet the short time windows required for gent management and the initiation of treatments of the current emergent care. The public’s knowledge about the stroke, including deciding whether a patient might be presentations of stroke and the best responses to their ap- treated with tissue plasminogen activator (Table). The pearance is limited.16-18 A concerted public education cam- published guidelines7-12 for care of acute stroke can be paign is needed; in particular, high-risk patients and their used as the templates for writing these protocols. family members, neighbors, friends, and coworkers should Intravenous administration of tissue plasminogen be instructed on the common symptoms of stroke. The mes- activator has engendered controversy.20-24 Intravenous sage to the public also should emphasize that stroke is treat- thrombolysis is not a cure-all but, at present, it is the only able but that time is critical. The educational program for therapy of proven efficacy for management of individu- acute ischemic stroke (brain attack) can be modeled on that als with acute ischemic stroke. While intra-arterial ad- used for acute myocardial ischemia (heart attack). Neu- ministration of thrombolytic drugs has hypothetical ad- rologists should lead these educational efforts in their com- vantages, its superiority over intravenous thrombolysis munities. is not proved.11,12,25 Although the results of some stud- Emergency medical services (EMS) should give acute ies26-30 are promising, neither antithrombotic drugs nor stroke the same priority as acute heart disease. Educa- neuroprotective agents are established as effective in im- tional programs on the importance of stroke care are proving outcomes after stroke. Until other therapies or needed for all components of EMS systems, including dis- treatment strategies are of proven value in improving neu- patchers and rescue squads. Evaluation in the field by rologic outcome after ischemic stroke, the goal of emer- EMS personnel should be prompt and transportation to gent management should be aimed at treatment with in- a hospital should be speedy. The first assessments should travenous tissue plasminogen activator. include relevant medical history, measurement of vital The short time window for effective and safe treat- signs, and a brief neurologic examination. If a para- ment with tissue plasminogen activator has implica- medic is in attendance, blood work and an electrocar- tions for all physicians, including neurologists. Initial man- diogram can be obtained and an intravenous line can be agement likely will occur in an emergency department. placed. A protocol that coordinates care between the EMS Most emergency departments are staffed by emergency and hospital can expedite transfer. The rescue squad medicine specialists or primary care physicians and in should inform the hospital that a patient with a possible many hospitals neurologists are not readily available. Many stroke is being transported to the emergency depart- patients cannot reach large tertiary medical centers with ment. The patient should go to a hospital that has the stroke specialists within 3 hours; in particular, patients ability to emergently perform computed tomography on in rural areas likely cannot be transported to a major hos- a 24-hour per day and a 7-day per week basis. Such a re- pital quickly enough to be treated with tissue plasmino- quirement might mean bypassing the closest hospital be- gen activator. Thus, a strategy to provide emergent care, cause it does not have imaging facilities. including the use of tissue plasminogen activator, must In turn, the hospital should have an acute stroke care be developed for a community setting. Hachinski31 ad- protocol to meet ambitious but feasible goals to rapidly vised that 3 criteria must be in place for the successful assess and treat patients.15 The availability of a treat- use of thrombolytic drugs: (1) availability of a physician ment of proven efficacy for improving outcomes of in- with expertise in the diagnosis and management of stroke, dividuals with ischemic stroke (tissue plasminogen ac- (2) accessibility to modern brain imaging studies, and tivator) but which must be given within 3 hours of onset (3) capability to handle potential complications of treat- of stroke highlights the importance of these ef- ment, especially brain hemorrhage. One could argue that forts.11,12,19 The protocol should include a list of physi- these requirements are needed for treatment of individu- cians, nurses, laboratory and radiology personnel, and als with stroke, regardless of the use of tissue plasmino- pharmacists who will be members of an acute stroke care gen activator. To expedite care of individuals with stroke, team. In institutions that do not have continuous cov- neurologists should be readily available for consulta- erage by stroke specialists or neurologists, primary care tions in emergency departments. Pending increased in- or emergency medicine physicians likely will be the lead- volvement of neurologists in emergency management, ers of this team. The protocol should outline steps in emer- guidelines for acute ischemic stroke should be written gent evaluation that focus on confirming ischemic stroke with the primary audience being physicians who are not as the likely cause of the neurologic impairments and de- neurologists. Strategies that permit rapid access to a neu- tecting acute medical or neurologic complications. The rologist for individuals located in remote places should emergent diagnostic studies for assessment of a patient be explored; potential approaches include telemedicine with a suspected acute ischemic stroke should include or air evacuation. A possible scenario would be for the the following: computed tomography of the brain with- patient to be treated with tissue plasminogen activator out contrast; electrocardiogram; chest x-ray film; com- in a community hospital with the advice of a neurolo- plete blood cell count and platelet count; prothrombin gist via telemedicine followed by air evacuation to a ter- time and activated partial thromboplastin time; blood tiary medical center for subsequent treatment.

ARCH NEUROL / VOL 55, APR 1998 458

©1998 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/27/2021 Besides being a life-threatening neurologic disease, ischemic stroke is a symptom of an underlying vascular Algorithm for Decisions About Administration disease that leads to thromboembolic occlusion of an ar- of Tissue Plasminogen Activator for Treatment of Acute Ischemic Stroke tery supplying a portion of the brain. The cause of stroke affects the patient’s prognosis; people with strokes sec- What is the interval from the onset of neurologic symptoms? ondary to occlusion of a penetrating brain artery (lacu- If symptoms Ͼ3 h, not eligible for treatment nes) generally have a better prognosis than those with If stroke occurred during sleep (Ͼ3 h) or if onset is uncertain, not occlusions of a major extracranial or intracranial ves- eligible for treatment sel.32,33 Determination of the most likely cause of stroke Can the patient’s symptoms be attributed to an illness other than has implications on plans to prevent recurrent stroke; for ischemic stroke? In particular, screen for Hypoglycemia example, anticoagulants are prescribed to most individu- Seizures with a postictal paralysis als with cardioembolic stroke while antiplatelet aggre- Craniocerebral trauma gating drugs are given to most patients with stroke sec- Any recent events that could increase the risk of bleeding ondary to extracranial or intracranial atherosclerosis.34 complications and that could contraindicate treatment with tissue Still, the diagnosis of cause of stroke can be difficult; phy- plasminogen activator? In particular, ask about Myocardial infarction sicians, including neurologists who have special exper- Ischemic stroke 35 tise in stroke, often disagree. Even diagnosis of a lacu- Craniocerebral trauma nar stroke can be difficult. A lacunar syndrome may not Bleeding be due to a lacunar infarction and the misdiagnosis could Surgery hamper the use of potentially effective treatment.36 In ad- Is the patient being treated with warfarin or heparin? dition, etiologic diagnoses made in an emergency de- The use of these medications leading to a prolongation of the 37 prothrombin time or activated partial thromboplastin time partment often change in subsequent days. Unfortu- precludes treatment with tissue plasminogen activator nately, an accurate diagnosis of subtype of ischemic stroke Is the patient’s blood pressure Ͼ185 mm Hg systolic or Ͼ110 mm Hg often requires the use of ancillary tests (ie, transesoph- diastolic? ageal echocardiography, carotid duplex, or arteriogra- An elevated blood pressure precludes treatment with tissue phy) that may not be available at a community hospital. plasminogen activator for most cases because the blood pressure cannot be stabilized within the 3-hour time window In addition, the reliability, sensitivity, and specificity of How severe are the neurologic impairments using the National these tests when performed in an emergency setting are Institutes of Health Stroke Scale? not known. Such information is critical because of the Patients with minimal neurologic impairments may not need to be potential for incorrect treatment if an ancillary test gives treated with tissue plasminogen activator because their either false-positive or false-negative information. Even prognosis usually is good. Conversely, caution should be exercised in giving tissue plasminogen activator to patients if these tests are available in a community hospital, the with severe strokes because of the risk of intracranial bleeding time required to mobilize personnel and other re- Does computed tomography demonstrate findings of a major sources to perform the test endangers the opportunity infarction? for treatment with tissue plasminogen activator. Con- Early computed tomographic findings of a multilobar infarction siderable research is needed to test the assumption that preclude treatment with tissue plasminogen activator the diagnosis of subtype of stroke is needed before starting emergent treatment. Data from clinical trials give limited support for the urgency of making an “accurate” quired personnel and sophisticated interventional capa- stroke subtype diagnosis.19,30 In particular, the Ameri- bilities. In addition, withholding treatment with a therapy can trial19 of tissue plasminogen activator did not note of proven value (intravenous tissue plasminogen activa- differences in responses to thrombolytic therapy among tor) while marshaling the resources to give intra- people with disparate subtypes. Until available evi- arterial thrombolytic therapy is problematic, particu- dence shows that the diagnosis of stroke subtype criti- larly when intra-arterial therapy is not established as cally affects acute management, ancillary diagnostic tests superior to intravenous therapy. Considerable research to determine the likely cause of stroke should not be man- is needed to test the hypothesis that knowledge of the dated before emergent treatment. At present, data do not vascular anatomy underlying stroke is necessary before support, in general, a recommendation to tailor acute man- starting emergent treatment. Until such evidence is avail- agement of stroke based on the presumed cause. able, examination of the vascular anatomy before treat- Caplan22,23 has articulated the importance of dem- ment should not be mandated. onstrating an arterial occlusion before making deci- Physicians know that the clinical severity of stroke sions about acute treatment. To test this assumption, tri- strongly predicts outcomes and affects decisions about als are testing the value of intra-arterial thrombolytic treatment. The National Institutes of Health Stroke Scale therapy in treatment of patients with arteriographically (NIHSS) is a widely used and accepted system to nu- confirmed occlusions of extracranial or intracranial ar- merically rate the severity of a patient’s stroke.37-40 The teries.11,25 While results are promising, intra-arterial throm- aggregate score of the NIHSS strongly correlates with out- bolytic therapy has not been shown convincingly to be comes.41,42 Patients with very mild deficits (low NIHSS either safer or more effective than intravenous therapy. score) generally have a favorable prognosis regardless of The requirement for visualization of an arterial occlu- treatment; these patients might not need to be treated with sion by arteriography before treatment greatly limits the tissue plasminogen activator or any other acute inter- use of the therapy; a minority of patients can be treated vention. Conversely, patients with multilobar infarc- because only a limited number of hospitals have the re- tions have high NIHSS scores; these patients usually have

ARCH NEUROL / VOL 55, APR 1998 459

©1998 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/27/2021 poor outcomes regardless of treatment with tissue plas- success of tissue plasminogen activator means that atti- minogen activator, and the risk of hemorrhagic compli- tudes toward stroke must be changed—patients can be cations is higher in this group.43,44 Because the NIHSS is treated and their outcomes improved. Neurologists should a useful way to quantify the patient’s neurologic signs and be in the forefront of efforts to treat stroke as a medical because the NIHSS score influences decisions about acute emergency and to develop specialized stroke care facili- treatment, neurologists should lead efforts to teach phy- ties in their communities. However, the critical part of sicians about the nuances of performing and scoring this care of most individuals with stroke will not be in spe- scale. Goldstein and Samsa40 showed that non- cialized stroke centers, and neurologists should not de- neurologists can effectively use the NIHSS in assess- mand that all patients with acute stroke be evaluated and ment of patients with stroke. treated only at tertiary-level, specialized centers. Such a Decisions about acute treatment also are affected by requirement could undermine the public’s and medical the results of brain imaging studies. Current guidelines community’s support for emergent stroke care. Rather, for the use of tissue plasminogen activator recommend neurologists should strive to collaborate with primary care not treating patients who have computed tomographic and emergency medicine physicians in increasing the evidence of a multilobar stroke.11,12 Fortunately, the ag- availability of stroke therapies. gregate NIHSS score correlates with the size of stroke found by brain imaging.37 Thus, the computed tomo- Accepted for publication January 14, 1998. graphic findings will be most critical among people with This article was funded in part by grant RO1- major deficits (high NIHSS score). Other brain imaging NS27863 from the US Public Health Service, National In- studies, such as perfusion and diffusion magnetic reso- stitute of Neurological Disorders and Stroke, National nance imaging, might improve selection of patients for Institutes of Health, Bethesda, Md. treatment.3,45-48 However, some clinicians have cau- Reprints: Harold P. Adams, Jr, MD, Division of Cere- tioned against the overdependence on such ancillary di- brovascular Diseases, Department of Neurology, Univer- agnostic tests.49 The value of these studies in a clinical sity of Iowa, 200 Hawkins Dr, Iowa City, IA 52242 (e-mail: setting has not been ascertained and the superiority of [email protected]). these tests over the clinical findings, such as the NIHSS score, or the results of computed tomography need to be REFERENCES determined. Until data are available that prove the use- fulness of ancillary brain imaging studies in the selection of patients to treat, these tests should not be re- 1. Taylor TN, Davis PH, Torner JC, et al. Lifetime costs of stroke in the United States. quired before starting emergent treatment of ischemic Stroke. 1996;27:1459-1466. 2. Biller J, Love BB. Nihilism and stroke therapy. Stroke. 1991;22:1105-1107. stroke. 3. Zivin JA. Diffusion-weighted MRI for diagnosis and treatment of ischemic stroke. The patient’s general health has a major impact on Ann Neurol. 1997;41:567-568. management of acute ischemic stroke. Severe comorbid 4. Astrup J, Siesjo B, Symon L. Thresholds in cerebral ischemia: the ischemic pen- medical or neurologic diseases affect prognosis and de- umbra. Stroke. 1981;12:723-725. 5. Pulsinelli W. Pathophysiology of acute ischaemic stroke. Lancet. 1992;339: cisions about treatment. For example, a patient who is 533-536. disabled because of dementia, heart disease, or arthritis 6. Grotta J. Rodent models of stroke limitations. Arch Neurol. 1996;53:1067- likely will not be nondisabled after management of stroke. 1070. The wishes of the patient and his/her family also should 7. Emergency Cardiac Care Committee and Subcommittees, American Heart As- be considered. Stroke affects a large number of elderly sociation. Guidelines for cardiopulmonary resuscitation and emergency cardiac care, IV: special resuscitation situations—stroke. JAMA. 1992;268:2242-2250. individuals who fear the disease more because of the po- 8. McDowell FH, Brott TG, Goldstein M, et al. Stroke: the first six hours. J Stroke tential for the loss of independence than for the poten- Cerebrovasc Dis. 1993;3:133-144. tial for death. They may not wish aggressive therapies that 9. Futrell N, Millikan CH. Stroke is an emergency. Dis Mon. 1996;42:199-264. might prolong their illness or result in chronic, severe 10. Adams HP Jr, Brott TG, Crowell RM, et al. Guidelines for the management of disability. Similarly, families might not want therapies patients with acute ischemic stroke. Stroke. 1994;25:1901-1914. 11. Adams HP Jr, Brott TG, Furlan AJ, et al. Guidelines for thrombolytic therapy for that they consider to be excessive. acute stroke: a supplement to the guidelines for the management of persons with At present, intravenous use of tissue plasminogen acute ischemic stroke. Stroke. 1996;27:1711-1718. activator is the only treatment that is known to be effec- 12. Report of the Quality Standards Subcommittee of the American Academy of Neu- tive in emergent management of patients with acute is- rology. Practice advisory: thrombolytic therapy for acute ischemic stroke. Neu- chemic stroke. The strategy used by the National Insti- rology. 1996;47:835-839. 13. Pan European Consensus Meeting on Stroke Management. J Intern Med. 1996; tute of Neurological Disorders and Stroke tissue 240:173-180. plasminogen activator investigators proved to be suc- 14. Broderick JP. Logistics in acute stroke management. Drugs. 1997;54(suppl 3): cessful; patients within 3 hours of stroke can benefit from 109-117. treatment if a similar approach is used. Intravenous throm- 15. Marler JR, Jones PW, Emr M. Proceedings of a National Symposium on Rapid Identification and Treatment of Acute Stroke. Bethesda, Md: Office of Scientific bolysis has limitations. It cannot be given with impu- and Health Reports, National Institute of Neurological Disorders and Stroke, Na- nity; bleeding complications occur, especially among in- tional Institutes of Health; 1997. NIH publication 97-4239. dividuals with severe strokes. Patients must receive the 16. Alberts M, Bertels C, Dawson DV. An analysis of time of presentation after stroke. medication within 3 hours and, thus, a minority of pa- JAMA. 1990;263:65-68. tients can be treated. 17. William LS, Bruno A, Rouch D, Marriott DJ. Stroke patients’ knowledge of stroke: influence on time to presentation. Stroke. 1997;28:912-915. Intravenous thrombolysis is the first, not the final, 18. Kothari R, Sauerbeck L, Jauch E, et al. Patients’ awareness of stroke signs, symp- treatment that will be shown to be effective in improv- toms and risk factors. Stroke. 1997;28:1871-1875. ing neurologic outcome after ischemic stroke. Still, the 19. The National Institute of Neurological Disorders and Stroke rt-PA Stroke Study

ARCH NEUROL / VOL 55, APR 1998 460

©1998 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/27/2021 Group. Tissue plasminogen activator for acute ischemic stroke. N Engl J Med. 35. Gordon DL, Bendixen BH, Adams HP Jr, Clarke W, Kappelle LJ, Woolson RF, 1995;333:1581-1587. and the TOAST Investigators. Interphysician agreement in the diagnosis of sub- 20. Caplan LR, Mohr JP, Kistler JP, Koroshetz W. Thrombolysis: not a panacea for types of acute ischemic stroke: implications for clinical trials. Neurology. 1993; ischemic stroke. N Engl J Med. 1997;337:1309-1310. 43:1021-1027. 21. Grotta JC. t-PA: the best current option for most patients. N Engl J Med. 1997; 36. Toni D, Fiorelli M, DeMichele M, et al. Clinical and prognostic correlates of stroke 337:1310-1313. subtype misdiagnoses within 12 hours from onset. Stroke. 1995;26:1837- 22. Caplan LR. New therapies for stroke. Arch Neurol. 1997;54:1222-1224. 1840. 23. Caplan LR. Stroke thrombolysis: growing pains. Mayo Clin Proc. 1997;72: 37. Brott T, Adams HP Jr, Olinger CP, et al. Measurements of acute cerebral infarc- 1090-1092. tion: a clinical examination scale. Stroke. 1989;20:864-870. 24. Haley EC Jr, Lewandowski C, Tilley BC, and the NINDS rt-PA Stroke Study Group. 38. Goldstein LB, Bertels C, Davis JN. Interrater reliability of the NIH Stroke Scale. Myths regarding the NINDS rt-PA Stroke Trial: setting the record straight. Ann Arch Neurol. 1989;46:660-662. Emerg Med. 1997;30:676-682. 39. Wityk RJ, Pessin MS, Kaplan RF, Caplan LR. Serial assessment of acute stroke 25. del Zoppo GJ, Higashida RT, Furlan AJ. The case for a phase III trial of cerebral using the NIH Stroke Scale. Stroke. 1994;25:362-365. intraarterial fibrinolysis. AJNR Am J Neuroradiol. 1994;15:1217-1222. 40. Goldstein LB, Samsa GP. Reliability of the National Institutes of Health Stroke 26. Clark WM, Warach SJ, Pettegrew LC, Gammans RE, Sabounjian LA, for the Citi- Scale: extension to non-neurologists in the context of a clinical trial. Stroke. 1997; coline Stroke Study Group. A randomized dose-response trial of citicoline in acute 28:307-310. ischemic stroke patients. Neurology. 1997;49:671-678. 41. Pullicino P, Snyder W, Granger C. The NIH Stroke Scale and the FIM in stroke 27. Grotta J, Hantson L, Wessel T, for the LUB-INT-9 Lubeluzole Study. The efficacy and rehabilitation. Stroke. 1992;23:99. safety of lubeluzole in patients with acute ischemic stroke. Stroke. 1997;28:271. 42. Muir KW, Weir CJ, Murray GD, Povey C, Lees KR. Comparison of neurological 28. International Stroke Trial Collaborative Group. The International Stroke Trial (IST): scales and scoring systems for acute stroke prognosis. Stroke. 1996;27:1817- a randomised trial of aspirin, subcutaneous heparin, both, or neither among 19,435 1820. patients with acute ischaemic stroke. Lancet. 1997;349:1569-1581. 43. The NINDS t-PA Stroke Study Group. Intracerebral hemorrhage after intrave- 29. CAST (Chinese Acute Stroke Trial) Collaborative Group. CAST: randomised placebo- nous t-PA therapy for ischemic stroke. Stroke. 1997;28:2109-2118. controlled study of early aspirin in 20,000 patients with acute ischaemic stroke. 44. The NINDS t-PA Stroke Study Group. Generalized efficacy of t-PA for acute stroke: Lancet. 1997;349:1641-1649. subgroup analysis of the NINDS t-PA Stroke Trial. Stroke. 1997;28:2119-2125. 30. Publications Committee for the Trial of Org 10172 in Acute Stroke Treatment 45. Warach SJ, Dashe JF, Edelman RR. Clinical outcome in ischemic stroke pre- (TOAST) Investigators. Low molecular weight heparinoid Org 10172 and out- dicted by early diffusion-weighted and perfusion magnetic resonance imaging: come after ischemic stroke: a randomized controlled trial. JAMA. In press. a preliminary analysis. J Cereb Blood Flow Metab. 1996;16:53-59. 31. Hachinski V. Thrombolysis in stroke: between the promise and peril. JAMA. 1996; 46. Koroshetz WJ, Gonzalez G. Diffusion-weighted MRI: an ECG for “brain attack?” 276:995-996. Ann Neurol. 1997;41:565-566. 32. Landi G, Cella E, Boccardi E, Musicco M. Lacunar versus non-lacunar infarcts: 47. Lutsep HL, Albers GW, DeCrespigny A, Kamar GN, Marks MP, Moseley ME. Clini- pathogenetic and prognostic differences. J Neurol Neurosurg Psychiatry. 1992; cal utility of diffusion-weighted magnetic resonance imaging in the assessment 55:441-445. of ischemic stroke. Ann Neurol. 1997;41:574-580. 33. Arboix A, Garcia-Eroles L, Massons J, Oliveres M. Predictive factors of in- 48. Lovblad KO, Baird AE, Schalug G, et al. Ischemic lesion volumes in acute stroke hospital mortality in 986 consecutive patients with first-ever stroke. Cerebro- by diffusion-weighted magnetic resonance imaging correlate with clinical out- vasc Dis. 1996;6:161-165. come. Ann Neurol. 1997;42:164-170. 34. Feinberg WM, Albers GW, Barnett HJM, et al. Guidelines for the management of 49. Chimowitz MI, Logigian EL, Caplan LR. The accuracy of bedside neurological di- transient ischemic attacks. Stroke. 1994;25:1320-1335. agnoses. Ann Neurol. 1990;28:78-85.

ARCH NEUROL / VOL 55, APR 1998 461