Agreement between 3D TTE and TOE measurements
The Bland-Altman analysis showed a good agreement between 3D TTE and TOE measurements of mitral annular area and perimeter for the whole study cohort (Figures 3a and 3b) as well as in the subgroups with MV pathology (Figures 3c and 3d) and those with normal MV (Figures 3e and 3f). More specifically, for mitral annular area measurements, the Bland–Altman analysis indicated a mean difference between 3D TOE and TTE of a) -0.34 cm2 (CI: 0.56–0.11cm2) with limits of agreement -2.6–1.8cm2, for all subjects, b) of -0.57 cm2(CI: 0.89–0.25cm2) with limits of agreement -2.8–1.6cm2, for those with MV pathology c) of -0.08 cm2(CI: 0.40–0.22cm2) with limits of agreement -2.2–2.0cm2, for those with normal MV. For mitral annular perimeter measurements similarly, the Bland–Altman method indicated a mean difference between 3D TOE and TTE of a) -0.19 cm (CI: 0.33–0.06cm) with limits of agreement -1.5-1.1cm, for all subjects b) of 0.34 cm (CI: 0.17–0.51cm) with limits of agreement -0.9–1.5cm, for those with MV pathology c) of 0.04 cm (CI: 0.16–0.24cm) with limits of agreement -1.3–1.4cm, for those with normal MV. Therefore, the Bland–Altman analysis demonstrated acceptable limits of agreement between 3D TTE and TOE MV measurements.
We also calculated the intraclass correlation coefficient (ICC) for every parameter among the 3D TTE and TOE (with the same operator in every measurement for all patients). The ICC for MV annular area was 0.94 (CI: 0.91–0.95) whereas it was 0.92 for MV annular perimeter (CI: 0.88–0.95) between 3D TTE and TOE, in all subjects. For tenting area and posterior leaflet length, the ICC was 0.68 (CI: 0.56-0.77) and 0.69 (CI: 0.57-0.78) respectively.
Similar values of ICC were obtained for 3D MV parameters both in those with MV pathology and those with normal MV.