To assess right atrial appendage (RAA) flow and its possible relationship t
o left atrial appendage (LAA) flow in chronic nonvalvular atrial fibrillati
on (AF), transesophageal echocardiography (TEE) was performed in 26 patient
s with chronic nonvalvular AF (group I). For the purpose of comparison, an
additional group of 27 patients with chronic valvular AF due to mitral sten
osis (group II) was analyzed. The clinically estimated duration of AF in gr
oup I was significantly longer than that of group II (8.7 +/- 3.4 versus 2.
7 +/- 1.1 years). Although right atrial size and RAA maximal area were larg
er in group I than those in group II, left atrial size was larger in group
II than that in group I. Group II had larger LAA maximal areas than group I
, but this difference did not reach statistical significance. The two group
s were not different with respect to the RAA or LAA emptying velocities. Si
gnificant correlations were observed between echocardiographic parameters o
f the two atria in patients with nonvalvular AF (r range, 0.4 to 0.7). In c
ontrast, in patients with valvular AF, no correlation was observed between
the echocardiographic parameters of the two atria (appendage emptying veloc
ity, r = 0.38, p = 0.051; atrial size, r = - 0.03, p = 0.89; maximal append
age area, r = 0.07, p = 0.75, respectively). There were no significant diff
erences in the presence of right and left atrial spontaneous echo contrast
and thrombus between the groups. All of the right and left atrial thrombi w
ere confined to their respective appendages and were found in the atria wit
h spentaneous echo contrast. Both RAA and LAA thrombi were present in one p
atient. In conclusion, our findings suggest that AF could affect both atria
equally in nonvalvular AF, in contrast to valvular AF. Therefore, the asse
ssment of RAA function as well as LAA may be important in patients with chr
onic nonvalvular AF.