Search

Background Data on stroke risk factor profiles and atrial fibrillation (AF) for the Chinese population are sparse. This study identified risk factors for stroke among native Chinese with nonrheumatic AF. Methods In this retrospective investigation, patients diagnosed with nonrheumatic AF were identified from 18 hospitals in representative areas of the country, from January 2000 to April 2002, based on the medical records. All parameters relevant to AF were compared between AF patients with stroke and those without. The independent risk factors for stroke were assessed with a logistic regression analysis. Results Patients numbering 3,425 with AF were included, among whom 827 subjects were discharged on account of stroke. The prevalence of stroke in nonrheumatic AF patients was 24.15%. AF patients with stroke were significantly older than controls (73.31 ± 9.18 versus 68.22 ± 12.29 y, p < 0.001) and more likely to have a history of hypertension (71.0 versus 51.6%, p < 0.001) and diabetes (17.9 versus 11.1%, p = 0.001). Both, systolic and diastolic blood pressure, are significantly higher in patients with stroke. Of all the parameters of echocardiography, there was strong evidence that left atrial (LA) thrombi significantly increased risk of stroke. Patients with persistent AF were more likely to have stroke than paroxysmal AF patients, while lone AF is less in patients with stroke than in those without. The rate of anticoagulation treatment is only 9.27%, but there were no significant differences between the 2 groups. In multivariate analysis, age ≥ 75 y (odds ratio [OR] 1.76; 95% confidence interval [CI] 1.08-2.98), history of hypertension (OR 1.52; 95% CI 1.28-1.80), diabetes (OR 1.39; 95% CI 1.11-1.76), high systolic blood pressure (OR 1.71; 95% CI 1.21-2.28), LA thrombi (OR 2.77; 95% CI 1.25-6.13) were independent predictors for stroke. Conclusions The prevalence of stroke in hospitalized nonrheumatic AF patients was high. The population-specific risk factors for stroke were age ≥ 75 y, diabetes, history of hypertension, high systolic blood pressure and LA thrombi by transesophageal echocardiography (TEE). These merit further evaluation. Copyright © 2009 Wiley Periodicals, Inc.