Results:
12 patients unable to attend follow-up visits and 10 patients having
poor echogenicity were excluded from the study. In total, 137 patients
with hypertrophic cardiomyopathy (70 % male, mean age 49.6 ± 14.2
years) were included in the study. Echocardiographic and
clinical characterizations of study populations are summarized in table
1.
During a follow-up period of 5 years (mean follow-up duration, 48.9±11.4
months) 37 patients (16.9%) developed AF on standard 12-lead ECG or
during their 24-hour Holter recordings.
At follow-up, the patients with occurrence of AF were older than in
patients without AF (56.7±13.1vs. 47.6±13.9, p= 0.001) (Table 1).
Compared to the patients who did not develop AF, those who developed AF
showed statistically lower global PALS (%) (18.8±7.6 vs. 22.2±6.6, p=
0.008) and global conduit strain (%) (12±5.6 vs. 14.6±5.6, p=0.023),
while there was no statistically significant difference in PACS %
(6.7±2.7 vs. 7.5±2.7 p= 0.14) (Table 2) (Figure 1).
The atrial peak time of mid septum (msn) (418.5 ± 70.1 vs 390.1 ± 73.9,
p= 0.04) and of the apical septum (msn) (441.7 ± 73.4 vs 405.2 ± 77.9,
p= 0.016), mid lateral (msn) (501.4 ± 83.8 vs 461.7 ± 77.6, p= 0.012)
and apical lateral (msn) (507.3± 77.9 vs. 470.7.5±78.8, p= 0.02)
segments were longer in the group of patients, in whom AF developed. The
peak atrial time was not statistically significantly different in the
other segments between the two groups. Atrial dispersion obtained from
the standard deviation of the 12 segments was observed to be higher in
the AF developing group (61.4 ± 23.2 vs 43.1±15.8, p=<0.001)
(Table 2) (Figure 2,3). LA diameter (4.2 ± 0.6 vs 4.1 ± 0.5, p = 0.1)
and LAVi (43.9±10.6 vs. 44.4±11, p= 0.8) were similar in the AF
developing group and in the group with no AF (Table 1).
In correlation analysis, there was a moderate inverse correlation
between artial dispersion(msn) and PALS (r: -0.29, p=0.001) and
There were no correlations between atrial dispersion (msn) and age
(years), LA diamater (cm), LAVi (mL/m2), GLPS (%),
respectively (Table 3).
The multivariate Cox regression analysis, including atrial dispersion,
PALS, age, LA diamater was used to determine independent predictors of
AF occurrence during the follow-up. Atrial dispersion (msn) (HR 1.017,
95% CI: 1.001-1.03, p= 0.035) and age (years) (HR 1.03, 95% CI:
0.986-1.047, p= 0.045) were found to be independent predictors of AF
occurrence (Table 4).
In the ROC analysis atrial dispersion > 44.7 msn predicted
occurrence of AF with 82.4% sensitivity and 64 % specificity (AUC:
0.75, p= <0.001 CI: 0.66-0.84). (Figure 4)