Article Text

Download PDFPDF

Radiology of stroke
  1. O M P Jolobe1
  1. 1Department of Adult Medicine, Tameside General Hospital, Fountain Street, Ashton-under-Lyne OL6 9RW, UK
    1. J M Wardlaw2
    1. 2Department of Clinical Nurosciences, Western General Hospital, Crewe Road, Edinburgh EH4 2XU, UK

      Statistics from

      Request Permissions

      If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

      The optimum time frame for imaging embolic infarcts for stigmata of haemorrhagic transformation should have merited discussion under the heading “special clinical circumstances”,1 not least because of conflicting evidence about the benefits versus risks of early anticoagulation in the context of unpredictable evolution of embolic infarcts with or without anticoagulant treatment. In a study comprising 30 patients with cardiogenic cerebral embolism, three patients with an initially non-haemorrhagic cerebral infarct, visualised by computed tomography within 12 hours of stroke onset, showed asymptomatic haemorrhagic transformation in the absence of anticoagulant treatment 2–8 days later. One other patient in this subgroup did, however, develop sudden worsening of hemiparesis despite having initially presented with a small infarct.2 Among 1457 patients anticoagulated with unfractionated heparin in the presence of embolic cerebral infarct associated with atrial fibrillation, haemorrhagic transformation (within 14 days) was significantly commoner (p < 0.0001) than in their non-heparinised counterparts.3 Ischaemic stroke recurred with a 4.9% frequency in the latter subgroup (comprising 1612 patients) during that time frame, and this complication was significantly less common (p = 0.001) in their anticoagulated counterparts.3 Anti-coagulation in the presence of haemorrhagic transformation has been advocated as being without risk on the basis of the outcome in 12 patients so treated,4 but the caveat is that, in another study also involving patients with embolic stroke, the subsequent development of haemorrhagic transformation in 5 of 231 patients anticoagulated with heparin was associated with significant clinical deterioration.5 It is this unpredictability in the consequences and tempo of haemorrhagic transformation and in the impact of early anticoagulation on this phenomenon that causes anxiety among clinicians at all levels of experience.


      Author's reply

      The strongest risk factor for haemorrhagic transformation of cerebral infarction in patients not treated with any antithrombotic or thrombolytic drug is simply having an extensive large infarct. It so happens that cardiogenic cerebral embolism often results in large cerebral infarcts (because thrombi arising from the heart are often large and block a largish vessel). Thus, there is an association between cardiogenic cerebral embolism and haemorrhagic transformation. This association may be exaggerated by administering antithrombotic or anticoagulant or thrombolytic therapy to these patients. Unfortunately, it is difficult to draw conclusions on the risks and benefits of anticoagulant treatment from non-randomised studies. The complete data from the international stroke trial (20 000 patients) clearly show that heparin started within the first 48 hours of stroke and continued for the first 14 days may reduce the risk of recurrent ischaemic stroke but at the risk of increasing haemorrhagic stroke and death, and as a result there is no net benefit.1 When to start or continue anticoagulant treatment after stroke in the small proportion of patients with a clear cardiac source of embolism is a decision that needs to be made in light of each patient's risk factors and continues to be a thorny problem. However, the benefits of aspirin (while less effective) are still worthwhile with less risk of haemorrhagic transformation. There is no easy answer to this problem.