Mead et al very justifiably draw our attention to the value of clinical assessment of patients with stroke. The cohort they describe is heterogenous—that is, patients with acute stroke and non-acute patients attending an outpatients' clinic.1 Clinical features of stroke syndrome seem to predict the anatomical site and size of the ischaemic lesion. We presented an abstract at the 7^th European Stroke Conference2 describing the correlation of CT features and clinical stroke syndromes on Oxfordshire Community Stroke Project classification in a study of 202 patients with acute stroke, the clinical examination being performed within 3 days of admission to the hospital and the CT performed at a mean duration of 3.5 (SD 2.6) days. Whereas there was a very good correlation between these two measurements (p=0.01), the group with lacunar infarcts had a heterogenous presentation, only 20% of patients with a visible lacunar infarct on CT had a lacunar infarct syndrome on clinical examination, the rest having some form of cortical clinical feature—that is, dysphasia, neglect, or hemianopia in addition to motor, sensory, or sensorimotor deficit. Similarly a good proportion of patients with large (12%) and small cortical infarcts (24%) had …

Dr G E Mead