A simple score for predicting paroxysmal atrial fibrillation in acute ischemic stroke

Abstract Background and purpose Our aim in this study was to investigate factors associated with paroxysmal atrial fibrillation (PAF) in acute stroke patients and to develop a risk score to predict the presence of PAF. Methods We retrospectively enrolled patients with acute ischemic stroke within 24 h of onset between June 2006 and April 2008. Patients with sustained AF were excluded. Patients were divided into two groups according to the presence of PAF: the PAF group or the non-PAF group. The clinical factors associated with PAF were investigated. Furthermore, we devised a new risk score to predict the presence of PAF. Results There were 215 patients enrolled. The PAF group had 32 (14.9%) patients. Multivariate logistic regression analysis demonstrated that NIHSS score ≥ 8 (OR, 4.2; 1.38–12.88), left atrial size ≥ 3.8 cm (OR, 4.8; 1.65–13.66), mitral valvular disease (OR, 7.5; 2.17–25.90), and plasma BNP level ≥ 144 pg/ml (OR, 12.8; 4.12–40.00) were independent factors associated with PAF. We developed a risk score from these variables (total score 0 to 5): NIHSS score ≥ 8 (1 point); left atrial size ≥ 3.8 cm (1 point); mitral valvular disease (1 point); and BNP level ≥ 144 pg/ml (2 points). The frequency of PAF was 0% with a score of 0, 4% with a score of 1, 14% with a score of 2, 26% with a score of 3, 50% with a score of 4 and 100% with a score of 5 Conclusion Our simple score can predict the presence of PAF during hospitalization in acute ischemic stroke.