@article {van Latum132, author = {J van Latum and P J Koudstaal and L J Kappelle and F van Kooten and A Algra and J van Gijn}, title = {Comparison of CT in patients with cerebral ischaemia with or without non-rheumatic atrial fibrillation. European Atrial Fibrillation Trial and Dutch T I A Trial Study Groups.}, volume = {59}, number = {2}, pages = {132--137}, year = {1995}, doi = {10.1136/jnnp.59.2.132}, publisher = {BMJ Publishing Group Ltd}, abstract = {In an attempt to distinguish between the CT characteristics of strokes of presumed cardioembolic origin and strokes caused by arterial disease, a comparison was made between the baseline CT of two prospective cohorts of patients with transient ischaemic attack or minor ischaemic stroke, with (n = 985) or without (n = 2987) non-rheumatic atrial fibrillation (NRAF). Of the patients with NRAF 54\% had evidence of cerebral infarction v 41\% of the controls (patients with sinus rhythm (SR); odds ratio (OR) 1.7; 95\% confidence interval (95\% CI) 1.4-1.9). Patients with NRAF more often had multiple infarcts (OR 1.4; 95\% CI 1.1-1.8), and more often infarcts that were not related to current neurological symptoms (OR 1.5; 95\% CI 1.2-1.8). For symptomatic infarcts, patients with NRAF more often had cortical end zone infarcts (OR 3.1; 95\% CI 2.6-3.8) and cortical border zone infarcts (OR 1.9; 95\% CI 1.3-2.9) than patients with SR. Conversely, symptomatic small deep infarcts (lacunae) were more often seen in patients with SR (OR 3.9; 95\% CI 2.8-5.4). Multivariate analyses showed that all these findings were independent of differences in baseline characteristics between the two study groups. The CT characteristics overlapped and did not allow a reliable distinction between cardioembolic and atherosclerotic causes of stroke in patients with NRAF.}, issn = {0022-3050}, URL = {https://jnnp.bmj.com/content/59/2/132}, eprint = {https://jnnp.bmj.com/content/59/2/132.full.pdf}, journal = {Journal of Neurology, Neurosurgery \& Psychiatry} }