Left atrial appendage orifice area and morphology is closely associated with flow velocity in patients with nonvalvular atrial fibrillation

BACKGROUND: Thromboembolic events are the most serious complication of atrial fibrillation (AF), and the left atrial appendage (LAA) is the most important site of thrombosis in patients with AF. During the period of COVID-19, a non-invasive left atrial appendage detection method is particularly important in order to reduce the exposure of the virus. This study used CT three-dimensional reconstruction methods to explore the relationship between LAA morphology, LAA orifice area and its mechanical function in patients with non-valvular atrial fibrillation (NVAF). METHODS: A total of 81 consecutive patients with NVAF (36 cases of paroxysmal atrial fibrillation and 45 cases of persistent atrial fibrillation) who were planned to undergo catheter radiofrequency ablation were enrolled. All patients were examined by transthoracic echocardiography (TTE), TEE, and computed tomography angiography (CTA) before surgery. The LAA orifice area was obtained according to the images of CTA. According to the left atrial appendage morphology, it was divided into chicken wing type and non-chicken wing type. At the same time, TEE was performed to determine left atrial appendage flow velocity (LAAFV), and the relationship between the left atrial appendage orifice area and LAAFV was analyzed. RESULTS: The LAAFV in Non-chicken wing group was lower than that in Chicken wing group (36.2 ± 15.0 cm/s vs. 49.1 ± 22.0 cm/s, p-value < 0.05). In the subgroup analysis, the LAAFV in Non-chicken wing group was lower than that in Chicken wing group in the paroxysmal AF (44.0 ± 14.3 cm/s vs. 60.2 ± 22.8 cm/s, p-value < 0.05). In the persistent AF, similar results were observed (29.7 ± 12.4 cm/s vs. 40.8 ± 17.7 cm/s, p-value < 0.05). The LAAFV in persistent AF group was lower than that in paroxysmal AF group (34.6 ± 15.8 cm/s vs. 49.9 ± 20.0 cm/s, p-value < 0.001). The LAAFV was negatively correlated with left atrial dimension (R = − 0.451, p-value < 0.001), LAA orifice area (R= − 0.438, p-value < 0.001) and left ventricular mass index (LVMI) (R= − 0.624, p-value < 0.001), while it was positively correlated with LVEF (R = 0.271, p-value = 0.014). Multiple linear regression analysis showed that LAA morphology (β = − 0.335, p-value < 0.001), LAA orifice area (β = − 0.185, p-value = 0.033), AF type (β = − 0.167, p-value = 0.043) and LVMI (β = − 0.465, p-value < 0.001) were independent factors of LAAFV. CONCLUSIONS: The LAA orifice area is closely related to the mechanical function of the LAA in patients with NVAF. The larger LAA orifice area and LVMI, Non-chicken wing LAA and persistent AF are independent predictors of decreased mechanical function of LAA, and these parameters might be helpful for better management of LA thrombosis.