Abstract

Background: Incidence of ischaemic stroke subtypes, classified by cause, seems to vary between communities. We aimed to prospectively ascertain the incidence of first-ever ischaemic stroke in a predominantly Hispanic-Mestizo population in the northern desertic region of Chile. Methods: Between July, 2000, and June, 2002, all patients with possible stroke or transient ischaemic attacks were identified from multiple overlapping sources and were rapidly assessed by two field neurologists. All identified patients were diagnosed by at least two stroke neurologists according to Trial of Org 10172 in Acute Stroke Treatment (TOAST) definitions and were followed up at 6 months. Annual incidence rates were age adjusted to WHO, European, and US populations by the direct method to allow comparisons. Findings: A total of 239 ischaemic strokes were identified, of which 185 (77%) were first-ever cases. 151 (82%) patients were hospitalised, of whom only 70 (38%) were assessed within 6 h of symptom onset. The mean age of patients was 66·4 years (SD 14·9) and 56% were men. The crude annual incidence rates (per 100 000) according to stroke subtype were: cardioembolic, 9·3; large-artery disease, 2·0; small-vessel disease, 15·8; other determined cause, 0·2; and undetermined cause, 17·4. Hypertension was the most common cardiovascular risk factor in all subtypes and atrial fibrillation was the most common cause of cardioembolic stroke. Case fatality at 30 days was highest in cardioembolic strokes (28%) and lowest in small-vessel disease (0%). Dependency or death at 6 months was also highest in cardioembolic strokes (62%) and lowest in small-vessel disease (21%). Interpretation: Incidence and prognosis of small vessel and cardioembolic infarction was similar to that in other populations and incidence of large-artery atherothrombotic infarction was lower than in most previous reports. Hypertension and atrial fibrillation were the most common risk factor and cause, respectively, of ischemic stroke in this population. These findings should help the national stroke programme in the prevention of cardioembolic stroke, increase access to specialists and acute brain imaging and vascular studies, and improve stroke care. © 2007 Elsevier Ltd. All rights reserved.