A total of 168 consecutive patients with predominant rheumatic mitral
stenosis were evaluated by transthoracic (TTE) and transesophageal ech
ocardiography (TEE). Of the 168 patients, 35 had previous embolic even
ts (group I) and 133 had no emboli (group II). A total of 77 (45.8%) p
atients had atrial fibrillation. The frequency of atrial fibrillation
was higher in group I than group II (68.6% vs 39.8%, < 0.001). The inc
idence of left atrial enlargement was greater in group I (p < 0.001).
Mitral valve area was found to be smaller in group I compared to group
II (p < 0.001). In group I 83.3% and 29.2% of the patients with atria
l fibrillation had left atrial spontaneous echo contrast (SEC) and lef
t atrial thrombus, respectively, and 72.7% of the patients with sinus
rhythm had left atrial SEC. In group II 79.2% and 20.8% of the patient
s with atrial fibrillation had left atrial SEC and left atrial thrombu
s whereas 28.6% and 2.6% of the patients with sinus rhythm had left at
rial SEC and left atrial thrombus, respectively. The incidence of left
atrial thrombus was significantly different in those patients with co
mpared to those without embolic events (20% vs 9.7%, p < 0.01). In gro
ups I and II, 28 of 35 (80%) and 64 of 133 (48.1%) patients had left a
trial SEC (p < 0.01). Patients with left atrial SEC had a greater left
atrial size (p < 0.01) and smaller mitral valve area (p < 0.01). Left
atrial size was normal in 2 patients with left atrial SEC and SEC was
not found in 55 patients with enlarged left atrium. Multiple logistic
regression analysis showed that atrial fibrillation, mitral valve are
a and left atrial enlargement were independent predictors of the SEC (
p < 0.001) and left atrial SEC was the principal determinant of thromb
oembolism. These data suggest that regardless of rhythm and atrial siz
e, left atrial SEC is a principal determinant of thromboembolic risk i
n mitral stenosis. TEE may be able to detect those patients with mitra
l stenosis at risk for emboli and guide appropriate therapy.