Results:
Baseline Characteristics:
A total of 101 patients with AF were included in the study. The mean age was 60.1 ± 11.1 years, 69% were male, the average BMI was 29.22 ± 5.08 and the average body surface area (BSA) was 1.83 ± 0.68 m2. Of all patients, 23.8% had heart failure with reduced ejection fraction (HFrEF), 6.9% had heart failure with preserved ejection fraction (HFpEF), 46.5% had hypertension (HTN), and 7.9% had a history of stroke or other systemic thromboembolism (SSE). 33.7% of patients underwent some form of cardioversion and 38.6% had failed a trial of antiarrhythmic therapy. The average left ventricular ejection fraction (LVEF) was 48.85%.
In terms of baseline left atrial characteristics, the mean LAVI was 38.66 ± 10.2 mL/m2 , the mean LA sphericity index was 73.6 ± 18.9, the mean LAA volume was 6.76 ± 4.47 mL, and the mean fibrosis level was 15.51 ± 8.02%. Left atrial appendage shapes were categorized into the popular “windsock” (6.93%), “cactus”(5.94%), “chicken wing”(77.2%) and “cauliflower” (9.90%) categories (20). The baseline characteristics of the LA, LAA, and LAA ostia in our study cohort are given in Table 1.
LAA Ostium Dimensions
The average LAA ostium area was 3.32 ± 1.17 cm2. When accounting for BSA, the indexed mean LAA ostium area was 1.67 ± 0.56 cm2/m2. Since LAA ostia can be approximated with a centroid with a major and minor axis, the average major axis length was 2.44 ± 0.48 cm, the average minor axis length was 1.72 ± 0.34 cm, with the average perimeter at 6.35 ± 1.17 cm.
When exploring the relationship between the LAA ostia characteristics and various risk factors and comorbidities, patients with HFrEF had a significantly larger LAA ostial area than patients without HFrEF (3.88 vs. 3.20 cm2, respectively, p=0.0148). This result remained significant when comparing the LAA ostial area indexed by BSA (1.89 vs. 1.60 cm2/m2, respectively, p=0.0268). No statistically significant associations were found between LAA ostial area and other risk factors or comorbidities. The analysis results are presented in Table 2.
Left Atrial Appendage Ostium Dimensions and its Association with Left Atrial Morphological and Functional Parameters
The LAA ostial parameters were further examined for their correlation with LA and LAA morphological parameters. The LAA ostial area was shown to have a moderate positive correlation with LA volume (r=0.31, p=0.0017) and LAA volume (r=0.42, p<0.0001). LAA ostial area was largest in patients with a “cauliflower” LAA shape morphology (4.53 cm2 vs 3.30 cm2 [average of other morphology classes], p=0.011). No association was found between LAA ostial area and LA sphericity, LAVI or LA fibrosis level.
LAA ostial parameters were also correlated with LA functional metrics. Total LAEF correlated negatively with LAA ostial area (r=-0.289, p=0.0057). Indexed LAA ostial area was also negatively correlated with total LA strain (r=-0.248, p=0.0185), and passive LA strain (r=-0.208, p=0.049). There was a trend for negative correlation between LAA ostial and active LA strain, but this was not statistically significant (p=0.064).
Analysis results correlating LAA ostial parameters with LA morphological and functional markers are presented in Tables 3A/3B and 4, respectively. A selection of correlation analysis results from this analysis are also shown in Figure 2.
Changes in Left Atrial Appendage Ostium Following Catheter Ablation
A subgroup analysis on 37 patients who underwent pre- and post-ablation CMR studies was performed. On average, the post-ablation CMR was performed at 49.6 ± 26.7 weeks after procedure day. The LAA ostial area reduced significantly from 3.84 ± 1.15 cm2 before ablation to 3.42 ± 0.96 cm2 after ablation (p=0.0004). 11 of the 37 patients (29.7%) experienced a reduction in ostial size to below 3.5 cm2while the minor axis length of the LAA ostia decreased from 1.92 cm to 1.77 cm after ablation (p=0.0004), the major axis length did not demonstrate a significant reduction (p=0.1139). Other LA and LAA structural and functional markers were also examined in these patients and compared pre- and post-ablation. Other parameters such as LA volume, LAA volume, LAEF and various components of LA strain did not change significantly after ablation. This subgroup analysis on pre- and post-ablation changes is given in Table 5.
Additional analysis was performed to examine LVEF change pre-/post-ablation and its association with LAA ostium area. First, LVEF increased from a pre-ablation average of 48.26% to a post-ablation average of 53.62% (p=0.015). Correlation of pre-ablation LVEF and pre-ablation LAA ostium area showed a near-significant negative trend (r=-0.21, p=0.083). There was no correlation between the post-ablation change in LVEF and the change in LAA ostium area (p=0.671). There was also no correlation between the post-ablation change in LVEF and the change in LAA ostium area (p=0.513). Results of this correlation analysis are presented in Figure 3.