Background: Patients with mitral stenosis, especially those with atrial fibrillation, are at increased risk for thromboembolic complications. Size of the left atrium, left atrial appendage dysfunction and severity of mitral stenosis are known risk factors for thromboembolism in patients with mitral stenosis. It has been postulated that F-wave amplitude on surface ECG is correlated with left atrial size, left atrial appendage function, and risk of thromboembolism in patients with nonrheumatic atrial fibrillation. The aims of this study were as follows: (1) to examine the relationship between surface ECG F-wave amplitude and left atrial appendage function, and (2) to assess the clinical significance of F-wave amplitude as it relates to risk of thromboembolism in a group of patients with rheumatic mitral stenosis. Material and Methods: One hundred patients with rheumatic mitral stenosis and permanent atrial fibrillation but without moderate to severe mitral insufficiency were investigated by transthoracic and transesophageal echocardiography. Two groups were formed according to the presence of a coarse (Group 1; n=60; F-wave in lead V-1 greater than or equal to 1 mm) or fine (Group 2; n=40; F-wave in lead V-1 < 1 mm) F-wave on surface ECG. Results: Comparison showed that Group I had significantly smaller mitral valve area (1.44+/-0.6 versus 2 1.7+/-0.74 cm(2), respectively; P<0.05), lower peak left atrial appendage flow velocity (18.8+/-2.1 versus 25.6+/-1.9 cm/s, respectively; P<0.005), higher-grade spontaneous echo contrast (2.05+/-1.44 versus 0.98+/-1.14, respectively; P<0.0001), and higher prevalence of thromboembolism (35% versus 12.5%, respectively; P<0.012). Multiple logistic regression analysis revealed that mitral valve area, left atrial appendage peak velocity, and coarse F-wave were independent clinical risk factors for thromboembolism in this patient group. Conclusion: The data suggest that presence of a coarse F-wave on surface ECG is associated with left atrial appendage dysfunction, and indicates higher thromboembolic risk in patients with predominant rheumatic mitral stenosis. (C) 2003 Elsevier Ireland Ltd. All rights reserved.