Usefulness of three-dimensional transthoracic echocardiography for the classification of congenital bicuspid aortic valve in children.

AIMS Because the classification of congenital bicuspid aortic valve (BAV) is of importance to predict a possible valvular dysfunction, we sought to assess the feasibility, the reproducibility, and the accuracy of three-dimensional transthoracic echocardiography (3D-TTE) to accurately depict the morphology of the leaflets in a BAV. METHODS AND RESULTS Seventy-two consecutive children, who were suspected of having a BAV on two-dimensional transthoracic echocardiography (2D-TTE), were included in this prospective study. 2D-TTE and 3D-TTE views of a BAV were recorded by the same investigator, and then were analysed separately by two confirmed paediatric cardiologists. For each of these two imaging techniques, the spatial position of cusps and raphes was noted for each patient. Intra-observer concordance and inter-observer concordance were evaluated to assess the reproducibility of the techniques. Feasibility of 3D-TTE was 100%. Median acquisition time of 3D-TTE was 117 (98.5-176.8) s. Image quality seemed to be better with 3D-TTE compared with 2D-TTE. When using 3D-TTE, the diagnosis was reconsidered for 12 patients (17%). Only 44.4% of uncertain BAV cases identified by 2D-TTE were confirmed by 3D-TTE. Furthermore, 3D-TTE seems to provide a better visualization of the leaflet morphology, leading to reclassification for 34.4% (95% CI 22.9-47.3) of the patients. Agreement for the BAV classification between 2D-TTE and 3D-TTE was therefore moderate (κ = 0.46). Both inter-observer concordance and intra-observer concordance were good (κ = 0.91 and κ = 0.93, respectively) for 3D-TTE. CONCLUSION 3D-TTE is feasible and provides accurate description of a BAV in children. Compared with 2D-TTE, 3D-TTE seems to enable a better visualization of the structural geometry of the leaflets.