Background: Retrospective studies suggested that cerebral microbleeds (MB) on magnetic resonance images (MRI) increase risk of intracerebral haemorrhage (ICH).
Objective: To compare the benefit of anti-thrombotic agents in stroke prevention (absolute risk reduction 2.49 -6 %) versus risk of ICH in ischaemic stroke patients with MB.
Materials and methods: We prospectively studied patients admitted consecutively for acute ischaemic stroke between 1999 and 2004. MB on MRI were documented. Primary end points were subsequent ICH, recurrent cerebral infarct (CI) and mortality.
Results: A total of 908 patients were recruited. MB were identified in 252 (27.8 %) patients. Mean follow-up period was 26.6 +/- 15.4 months. Risk of subsequent ICH increased significantly with quantity of MB: 0.6 % (no MB), 1.9 % (1 MB), 4.6 % (2-4 MB) and 7.6 % (>or= 5 MB) (p < 0.001). There was also a significant increase in mortality from ICH: 0.6 %, 0.9 %, 1.5 % and 3.8 % respectively (p = 0.054). Rate of recurrent CI was 9.6 %, 5.6 %, 21.5 % and 15.2 % respectively (p = 0.226). Mortality from CI and myocardial infarction did not increased with quantity of MB. Survival analyses showed that age, presence of MB, mixed cortical-subcortical distribution of MB were independent predictors of subsequent ICH.
Conclusion: Risk and mortality of ICH increased with quantity of MB. As tendency to recurrent CI exceed that of ICH, anti-thrombotic agents are still warranted. However, in patients with >or= 5 MB, the high risk and mortality of ICH seem to outweigh the modest benefit of antithrombotic agents. Extra precautions should be taken to minimize risk of ICH. Further studies in patients on Coumadin and assessment of functional outcome are warranted to support these preliminary findings.