To the Editor:—Should we test for CYP2C9 before initiating anticoagulant therapy in patients with atrial fibrillation? The answer to the question1 is “yes”, and we should also test for other risk factors predisposing to anticoagulant-related intracerebral haemorrhage (ICH), including the ApoE genotype, which is a marker for cerebral amyloid angiopathy, and, hence, increased risk of warfarin-related ICH.2 Other modalities for risk profiling include magnetic resonance imaging (MR), given the association between cerebral microbleeds and recurrent haemorrhagic stroke in patients treated with warfarin (odds ratio 7.383, 95% confidence interval 1.052 to 51.830) following ischaemic stroke.3 Computerised tomography, on the other hand, can be utilised for identifying leukoairosis, an independent risk factor for warfarin-related ICH (odds ratio 8.4, 95% CI 1.4–51.5) in patients with previous ischaemic stroke.4 Therapeutic options that can be utilised for patients at high risk of warfarin-related intracranial haemorrhage include not only aspirin,1 but also antiarrhythmic therapy for atrial fibrillation (AF), using either catheter ablation5 with the attendant 0.98 per 1,000 mortality risk6 or amiodarone,7 the former being by far the more successful modality for maintenance of sinus rhythm.5 Amiodarone, however, is associated with a lower rate of treatment-related fatality, although requiring great vigilance for drug-related toxicity.7 Therapeutic options to reduce haemorrhagic risk also include low-intensity anticoagulation with target INR in the range of 1.5–2.7 or 1.4–2.8.8