Advertisement
Mayo Clinic Proceedings Home
MCP Digital Health Home

Of Stroke Treatment, Algorithms, Trials, and Such

  • Louis R. Caplan
    Correspondence
    Address reprint requests to Dr. L. R. Caplan, Department of Neurology, New England Medical Center, 750 Washington Street, #314, Boston, MA 02111
    Affiliations
    Department of Neurology New England Medical Center, Boston, Massachusetts
    Search for articles by this author
      In this issue of the Mayo Clinic Proceedings (pages 1027 to 1039), Brown and colleagues from the Mayo Clinic stroke service present a detailed, carefully considered, timely algorithm that incorporates their current advice for assessment and management of patients with minor strokes and transient ischemic attacks (TIAs). The report includes the rationale behind the practical advice and provides support for their positions from studies in the literature.
      My reaction is unreserved admiration. Bravo. I believe in algorithms—they are the best approach for now and the future. The Mayo stroke service is among the most experienced and distinguished in the United States. Brown and his current associates were preceded by Siekert, Millikan, Whisnant, Sundt, Karnes, and Sandok—all well-known leaders in the field of cerebrovascular disease. Furthermore, the algorithm is scholarly, relatively clear, and practical. I am aware of only four previous published attempts at presenting guidelines for management of stroke,
      • Feinberg WM
      • Albers GW
      • Barnett HTM
      • Biller J
      • Caplan LR
      • Carter LP
      • et al.
      Guidelines for the management of transient ischemic attacks: from the ad hoc committee on guidelines for the management of transient ischemic attacks of the Stroke Council of the American Heart Association.
      • Adams Jr, HP
      • Brott TG
      • Crowell RM
      • Furlan AJ
      • Gomez CR
      • Grotta J
      • et al.
      Guidelines for the management of patients with acute ischemic stroke: a statement for healthcare professionals from a special writing group of the Stroke Council, American Heart Association.
      • Brott TG
      • Goldstein M
      • Grotta JG
      • Heros RC
      • Latchaw RE
      • McDowell FH
      • et al.
      Stroke: the first six hours; emergency evaluation and treatment.
      • Frey JL
      Decision-making in transient ischemic attack and stroke.
      all of which differed considerably and only one of which provided an algorithm.
      • Frey JL
      Decision-making in transient ischemic attack and stroke.
      Why should algorithms be used? Hearing or reading the most erudite didactic presentation often does not tell a physician how to approach and manage an actual patient. The clinical approach to patients involves a series of questions; decisions are sequential and depend on answers to these queries.
      • Caplan LR
      • Caplan LR
      The “what-if” approach mimics the bedside experience better than does a didactic lesson. Algorithms and decision trees graphically illustrate the what-if approach and define key decision points.
      Do algorithms have negative aspects? Algorithms concern disease. Physicians treat unique patients. The preferences, biases, and knowledge of patients and their families, the presence of other illnesses and conditions, and individual needs, wants, and demands are difficult to incorporate into an algorithm. The nuances of complex diseases such as stroke are also difficult to include in an algorithm, as are unusual disorders, findings, and exceptions to the rules. Algorithms are not meant to be law. They are only guides or maps to help direct the way in caring for the “average” patient.
      What should algorithms include? Of foremost importance is identification of the key questions. Physicians who are not specialists in a particular field often do not know what questions are critical and what data are essential. If they knew the right questions, they could better refer to appropriate resources to help interpret answers. The next step is determining, at each decision point, the key data items that influence which arrow or path in the algorithm to follow. For example, in an assessment of whether cardiac-origin embolism is likely, the key questions are as follows: Does the patient have ischemia in one or multiple vascular territories? hemorrhagic infarcts? heart symptoms, signs, or known cardiac disease? severe and sudden onset of neurologic signs?
      • Timsit SG
      • Sacco RL
      • Mohr JP
      • Foulkes MA
      • Tatemichi TK
      • Wolf PA
      • et al.
      Brain infarction severity differs according to cardiac or arterial embolic source.
      rapid clearing of a severe neurologic deficit?
      • Minematsu K
      • Yamaguchi T
      • Omae T
      ‘Spectacular shrinking deficit’: rapid recovery from a major hemispheric syndrome by migration of an embolus.
      systemic embolism? If cardiac-origin embolism is likely, the action response is cardiac evaluation and possibly interim treatment—such as with anticoagulants, as mentioned in the algorithm. What-if questions are essential. For example, if a patient has an attack of unilateral cerebral hemisphere dysfunction, what is the first step? If duplex ultrasonography shows carotid artery disease, then what? If duplex scans show normal findings, should some treatment be administered during evaluation and, if so, when and what type?
      What preceded algorithms? Assessment and treatment of patients with cerebrovascular disease have a short history. Miller Fisher described TIAs and carotid artery disease in the early 1950s. From 1950 until now, the main approach was to prescribe treatment depending on the time course of ischemia—that is, TIA, progressing stroke, reversible ischemic neurologic deficit, or “completed stroke.” Time courses, however, do not predict the presence or absence of brain infarction, the cause of ischemia, or the prognosis.
      • Caplan LR
      TIAs: we need to return to the question, ‘What is wrong with Mr. Jones?’[editorial].
      • Caplan LR
      Terms describing brain ischemia by tempo are no longer useful: a polemic (with apologies to Shakespeare).
      Any of the common vascular pathologic conditions, such as carotid artery disease, cardiac-origin embolism, hypercoagulability, or penetrating artery occlusive disease, can cause TIAs, prolonged but reversible deficits, or persistent signs. Treatment should depend mostly on what is wrong with the patient,
      • Caplan LR
      TIAs: we need to return to the question, ‘What is wrong with Mr. Jones?’[editorial].
      including assessment of the following factors: (1) the site, nature, and severity of the causative vascular (cardiac or hematologic) lesion; (2) the pathophysiologic features of the symptoms (such as hypoperfusion, embolism, or vasoconstriction); and (3) the presence and severity of the brain lesion—is the brain normal, infarcted, or reversibly ischemic (“stunned”)? The cookbook approach to treatment based on time course has no theoretical basis and has failed miserably during the past nearly half-century. Treatment based on time course should become a relic of the past, of historical interest only.
      Another familiar approach, unfortunately still espoused by some clinicians, is what I call the “skin cream” method. My grandfather once had an itchy rash. His dermatologist gave him seven creams (labeled A through G) and the following instructions: if A does not work, use B; if B does not work, use C; and so forth. For patients with stroke, some physicians recommend use of aspirin; if that fails, use ticlopidine; if that fails, use warfarin. If all else fails, call a surgeon (or a priest). Wise patients ask, “What do you mean, if the treatment fails?” Truthful physicians might answer, “Well, for example, if you go blind, become paralyzed, cannot talk or walk,….” My response would be “no, thanks.” I want the treatment most likely to be effective used first, not after a disaster.
      Unequivocally, the tempo and skin cream approaches should be quickly abandoned and placed in museums to be pondered by future generations as the folly of their ancestors. Instead, use algorithms, decision points, and what-if advice.
      What do I like and not like about the currently presented algorithm? I agree with recommendations about the use of anticoagulants. Some clinicians have dismissed these drugs as having no proven use. I agree but respond that they have never been appropriately evaluated. Strong theoretic factors suggest their utility in certain vascular lesions,
      • Caplan LR
      as indicated by the authors of this algorithm.
      I agree with the inclusion of tests and treatments not yet proved by randomized trials (but identified as such). Trials are not feasible in many situations because of insufficient numbers of patients. Time and considerable amounts of money are needed to complete trials. If only proven treatments were prescribed, physicians would be a decade or so behind current practice. Would you choose a 1986 tested and certified Ford over an untested 1993 Rolls-Royce? Progress would be halted if only proven tests, treatments, and strategies were used.
      Whether or not to hospitalize is overemphasized. Patients with brain ischemia should be assessed by experts with access to modern technology who can perform rapid evaluation and testing. The brain is the Rolls-Royce of the body—it defines humanity and individuality. Would anyone trust a Rolls-Royce to an inexperienced mechanic at an ill-equipped local gas station? For assessment, by whom and with what are more important considerations than whether to hospitalize.
      Race, sex, and risk factors are not sufficiently emphasized as important factors in evaluation. In general, white men most frequently have extracranial carotid and vertebral artery disease, whereas blacks and persons of Chinese, Japanese, and Thai origin and women more often have intracranial disease.
      • Caplan LR
      • Gorelick PB
      • Hier DB
      Race, sex and occlusive cerebrovascular disease: a review.
      Demographic features should be considered in planning the assessment of the patient.
      Patients with involvement of the anterior circulation and those with posterior circulation disease should be assessed in the same fashion and to the same extent.
      • Caplan LR
      Vertebrobasilar disease: should we continue the double standard of managing patients with brain ischemia? [editorial].
      This algorithm suggests different approaches. The two circulations share the same blood, blood pressure, heart, risk factors, and pathologic conditions. Modern neuroimaging and ultrasound studies offer effective noninvasive evaluation of each area. The only differences are anatomic—where to insonate and image.
      Lacunar infarcts are not considered. I believe, as do many others,
      • Mohr JP
      Lacunes.
      that the causes of lacunae and the management of patients with penetrating artery disease differ from those in patients with other stroke mechanisms.
      In the algorithm, too much emphasis is placed on the number of ischemic spells. Many patients with severe strokes have no or only a single TIA.
      What do I advise? The following guidelines for management of patients are important.
      • 1.
        Determine what is wrong with the patient. A precise diagnosis has considerable intrinsic value. I do not believe in just testing for treatable causes. Knowing what is wrong with the patient also indicates what is not wrong and helps in the ascertainment of prognosis and omission of tests and treatments not likely to work. Logical although unproven treatments can be tried if the situation warrants. Knowledge facilitates frank discussions with the patient and family members about the diagnosis, prognosis, and treatment options.
      • 2.
        Assessment should be done immediately and rapidly. The brain is vulnerable and must be protected at all costs.
      • 3.
        The responsible physician should have extensive knowledge and experience in brain disease and stroke and needs access to modern technology and experienced colleagues (such as cardiologists, hematologists, neuroradiologists, and surgeons).
      • 4.
        Use the algorithm as a guide, but consider the individual patient and family and their preferences and exceptions to the rules.
      • 5.
        Be open and responsive but skeptical about new tests and treatments.
      • 6.
        Weigh benefits versus risks in the individual patient in choosing tests and treatments. Unlike established therapeutic trials, no physician is obligated to use the same treatment for all patients. Tailoring should depend on the details of the individual problem and patient.
      I applaud the work of Brown and colleagues. It is an excellent start in the right direction. Their algorithm needs application and testing in the field and may be modified by experience. Nothing is exactly right the first time. The Mayo Clinic stroke service merits our appreciation and thanks for a brave and thoughtful undertaking.

      REFERENCES

        • Feinberg WM
        • Albers GW
        • Barnett HTM
        • Biller J
        • Caplan LR
        • Carter LP
        • et al.
        Guidelines for the management of transient ischemic attacks: from the ad hoc committee on guidelines for the management of transient ischemic attacks of the Stroke Council of the American Heart Association.
        Stroke. 1994; 25: 1320-1335
        • Adams Jr, HP
        • Brott TG
        • Crowell RM
        • Furlan AJ
        • Gomez CR
        • Grotta J
        • et al.
        Guidelines for the management of patients with acute ischemic stroke: a statement for healthcare professionals from a special writing group of the Stroke Council, American Heart Association.
        Stroke. 1994; 25: 901-1914
        • Brott TG
        • Goldstein M
        • Grotta JG
        • Heros RC
        • Latchaw RE
        • McDowell FH
        • et al.
        Stroke: the first six hours; emergency evaluation and treatment.
        Stroke Clin Updates. 1993; 4: 3-12
        • Frey JL
        Decision-making in transient ischemic attack and stroke.
        BNI Q. 1994; 10: 22-29
        • Caplan LR
        The Effective Clinical Neurologist. Blackwell Scientific Publications, Cambridge (MA)1990
        • Caplan LR
        Stroke: A Clinical Approach. 2nd ed. Butterworth-Heinemann, Boston1993
        • Timsit SG
        • Sacco RL
        • Mohr JP
        • Foulkes MA
        • Tatemichi TK
        • Wolf PA
        • et al.
        Brain infarction severity differs according to cardiac or arterial embolic source.
        Neurology. 1993; 43: 728-733
        • Minematsu K
        • Yamaguchi T
        • Omae T
        ‘Spectacular shrinking deficit’: rapid recovery from a major hemispheric syndrome by migration of an embolus.
        Neurology. 1992; 42: 157-162
        • Caplan LR
        TIAs: we need to return to the question, ‘What is wrong with Mr. Jones?’[editorial].
        Neurology. 1988; 38: 791-793
        • Caplan LR
        Terms describing brain ischemia by tempo are no longer useful: a polemic (with apologies to Shakespeare).
        Surg Neurol. 1993; 40: 91-95
        • Caplan LR
        • Gorelick PB
        • Hier DB
        Race, sex and occlusive cerebrovascular disease: a review.
        Stroke. 1986; 17: 648-655
        • Caplan LR
        Vertebrobasilar disease: should we continue the double standard of managing patients with brain ischemia? [editorial].
        Heart Dis Stroke. 1993; 2: 377-381
        • Mohr JP
        Lacunes.
        Stroke. 1982; 13: 3-11
      1. Adv Neurol. 1993; 62: 1-241