Editorials

Treating acute ischaemic stroke

BMJ 1995; 311 doi: https://doi.org/10.1136/bmj.311.6998.139 (Published 15 July 1995) Cite this as: BMJ 1995;311:139
  1. Philip M W Bath, Wolfson senior lecturer in stroke medicine
  1. Department of Medicine, King's College School of Medicine and Dentistry, London SE5 9PJ

    Still no effective drug treatment available

    Now that drug treatment for acute myocardial infarction has been largely sorted out, the world of vascular medicine has moved on to acute stroke. Given that stroke is the third commonest cause of death in the West and the most important cause of adult disability,1 it is surprising that no treatment exists that has been conclusively shown to reduce the risk of death or disability. Despite a massive worldwide effort to rectify this, early results are disappointing.

    In the West most (85%) strokes result from cerebral infarction after arterial occlusion. Ischaemia induces activation of the glutamate-calcium cascade and cytodestructive enzymes, the release of free radicals, and ultimately cell death.2 Surrounding the core of dead neural tissue lies a penumbra of neurones that may survive or die, which has become the focus of pharmacological activity.

    The main classes of agents include thrombolytics (to accelerate reperfusion), anticoagulants and antithrombotics (to prevent further occlusion), vasodilators (to increase cerebral blood flow), antileucocytic drugs (to prevent inflammation), glutamate antagonists, calcium antagonists, and free radical scavengers. Anticoagulants, antithrombotics, and antileucocytic drugs may also reduce the early risk of deep vein thrombosis and pulmonary embolism, which are important complications of stroke.

    Two sets of events have occurred recently that considerably advance our knowledge of treatment of stroke. Firstly, the results of several trials of thrombolysis have recently been presented or published (although their interpretation is complex). One study, the multicentre acute stroke trial of Europe, stopped prematurely because of increased early death in patients treated with streptokinase.3 The Australian streptokinase trial also finished early because patients given thrombolysis between three and four hours after the onset of stroke fared worse than those treated within three hours.4 The European cooperative acute stroke study of 608 patients reported a non-significant increase in death and severe disability at three months in patients receiving recombinant tissue plasminogen activator.5 Finally, the multicentre acute stroke trial in Italy has suspended recruitment at 616 patients (of a planned 1200) to await further analysis of the existing data on thrombolysis. A common finding in these trials is that the outcome after thrombolysis seems worse in the first four weeks (with more deaths) but better after three to six months. This is analogous with the results of carotid endarterectomy for severe symptomatic carotid stenosis, which carries a definite early risk but survivors are subsequently relatively protected from ipsilateral stroke.

    The second important event is the publication of the Cochrane Database of Systematic Reviews on disk, which includes the first set of overviews from the Cochrane Stroke Review Group6; six reviews relate to the treatment of stroke, three to secondary prevention in atrial fibrillation, and one each to care in a stroke unit and carotid endarterectomy patch angioplasty.

    In contrast to the recent trials of thrombolysis, meta-analysis of six older and smaller studies (899 patients in total) suggests that early thrombolysis results in a non-significant fall in early death (odds ratio 0.62 (95% confidence interval 0.31 to 1.23)) and symptomatic bleeding into cerebral infarction (0.60 (0.21 to 1.67)).7 An update of this analysis including the recent thrombolysis trials is urgently needed to determine whether further trials of thrombolysis are warranted and, if so, how they should be designed.

    An analysis of trials of anticoagulant treatment in which patients were enrolled within two weeks of their stroke suggests that such treatment reduces the risk of deep vein thrombosis (eight trials, 750 subjects; odds ratio 0.18 (0.13 to 0.25)). There was no significant effect, however, on mortality, and a non-significant increase in symptomatic intracranial haemorrhage was observed.8 Four trials of antiplatelet agents (ticlopidine or aspirin-dipyridamole) have also been reviewed; no significant changes in deaths, deep vein thrombosis, or symptomatic intracerebral haemorrhage were found.9

    Randomised trials of intravenous epoprostenal showed no effect on deaths at one month.10 In contrast, intravenous glycerol reduced early deaths (six trials, n=454; odds ratio 0.57 (0.36 to 0.90) but not late deaths.11 Future reviews by the Cochrane Collaboration will include studies of calcium antagonists, corticosteroids, gangliosides, haemodilution, oxpentifylline, and vinpocetine in acute stroke. It is evident from these systematic overviews that too few patients receiving each potential treatment have been studied and that much larger trials are required. The international stroke trial, the first “mega trial” in stroke, is assessing the effects of aspirin or heparin, or both, on 20000 patients enrolled within two days of the onset of stroke and will report within 18 months (details on recruitment from Peter Sandercock 0131 343 6639).

    Developing drug treatments for cerebral ischaemia will be far more difficult than for myocardial infarction; 20 years of trials have yet to provide an effective drug treatment for stroke. We will probably have to wait until the next century before treatments other than antithrombotic and anticoagulant treatment (for example, neuroprotection) have been tried in sufficient numbers of patients to assess their efficacy and safety adequately. Sadly, little regard has being given to developing treatments for primary intracerebral haemorrhage, although newer trials of neuroprotective agents (including the planned international stroke trial 2) are starting to include patients with this condition.

    To accelerate the development of short term treatments for stroke six actions are necessary. Firstly, hospitals capable of participating in multicentre trials should join them to speed recruitment. Secondly, hospitals will need to open acute stroke units to facilitate such trials12 13 (much as coronary care units made treatment of myocardial infarction trials possible). Thirdly, most ongoing trials require the involvement of patients within three to 12 hours of the onset of stroke. Patients, general practitioners, and hospital staff need to appreciate that early presentation to hospital will increase the population of patients eligible for, and hence who may benefit from, such trials; the National Institutes of Health's trial of tissue plasminogen activator managed to enrol 300 patients within 90 minutes of the onset of stroke so it is possible to greatly reduce the time that elapses before the initiation of treatment. Fourthly, some trials of drugs that alter haemostasis (especially thrombolytics, anticoagulants, and antithrombotics) require prior computed tomography to exclude primary haemorrhage; the routine availability of 24 hour tomography is now required in all acute hospitals. Fifthly, short term drug treatments for primary intracerebral haemorrhage need to be developed. Lastly, systematic reviews spanning the whole range of the management of stroke are required to help define future lines of research.

    References

    View Abstract

    Sign in

    Log in through your institution

    Subscribe