Level of agreement between three-dimensional transthoracic and transesophageal echocardiography for mitral annulus...

Mitral annulus assessment is of utmost importance for the management of patients with mitral valve (MV) abnormalities, as it helps to determine the decision for surgical or transcatheter treatment. Three-dimensional (3D) transesophageal echocardiography (TOE) has been the only reliable echocardiographic method for the evaluation of the mitral annulus by now. However, newer transthoracic echocardiography (TTE) 3D probes have enabled to provide accurate measurements as well and become a valuable tool when TOE is contraindicated. The aim of this study is to assess the feasibility of 3D TTE analysis of mitral annulus and the level of agreement with 3D TOE measurements.

A total of 121 consecutive patients were assessed with 3D TTE and TOE. All mitral annulus parameters were retrospectively analyzed with the dedicated 4D autoMVQ application. Bland-Altman analysis and intraclass correlation coefficient were used for the comparison and agreement between the two methods. Half of our patients had normal mitral valves and served as control group, while the other half had various mitral valve pathologies.

AutoMVQ analysis was not feasible in 11 out of 121 TTE examinations (91% feasibility) and in 4 out of 121 TOE examinations (96% feasibility). Mitral annular area and perimeter were slightly larger in TTE than those measured by TOE (12.7 ± 3.6 vs. 12.4 ± 3.2 cm² for area and 12.7 ± 1.7 vs. 12.5 ± 1.6 cm for perimeter), however still showing strong correlation (r = .942 and r = .922, respectively). The majority of mitral valve measurements (anterior-posterior, medial-lateral and commissural diameter, aorto-mitral angle and anterior leaflet length) were similar among the two methods with strong correlation (r > .80). Inter-trigonal distance, posterior leaflet length and tenting height showed weaker agreement between TTE and TOE (r = .687, r = .687, r = .634, respectively). Mitral annular dimensions (by 3D area) were found to be significantly larger in patients with MV pathology (13.5 ± 3.5 vs. 11 ± 2.3 cm² ), atrial fibrillation (14.4 ± 3 vs. 11.4 ± 2.8 cm² ), left ventricular (13.8 ± 3.1 vs. 11.7 ± 3.1cm² ) and left atrial dilatation (13 ± 3.3 vs. 10.6 ± 2.3cm² ) compared to the individuals in the control group (p < .001 for all comparisons).

Assessment of the MV with 3D TTE with dedicated MVQ software is feasible and accurate, showing strong correlation and agreement with TOE measurements.