Discrepancy between mitral valve areas measured by two-dimensional planimetry and three-dimensional transoesophageal echocardiography in patients with mitral stenosis

Objective To compare mitral valve area (MVA) measurements obtained by 2D transthoracic planimetry and 3D transoesophageal echocardiography (TOE) in patients with mitral stenosis (MS), and to determine the causes of discrepancies between the two techniques. Design Reliability and agreement study. Setting Tertiary referral centre. Patients Eighty-seven patients with MS. Methods MVA was determined by transthoracic 2D planimetry and 3D TOE. Clinical and echocardiographic variables were evaluated. The angle (M alpha) between the lines of the true mitral valve (MV) tip and the echo beam-to-MV tip was measured at early diastole from the parasternal long-axis view obtained from 2D echocardiography. Results Although MVA measurements using 2D planimetry and 3D TOE showed good agreement (intraclass correlation coefficient, 0.853; p<0.001), 2D planimetry overestimated MVA by 0.19 +/- 0.2 cm(2) compared with 3D TOE (p<0.001). Left atrial (LA) dimension obtained from the parasternal long-axis view at end-systole (p=0.012), M alpha (p<0.001), and left ventricular ejection fraction (p=0.022) were independent determinants of the MVA difference (MVA by 2D-MVA by 3D TOE; MVA2D-3D) according to multiple linear regression analysis. The LA dimensions correlated with M alpha (r=0.352, p=0.001). The best cut-off values for predicting significant overestimation by 2D planimetry (MVA2D-3D>0.2 cm(2)) were LA dimension >= 49 mm (78% sensitivity, 72% specificity) and M alpha >= 9.5 degrees (56% sensitivity, 89% specificity). Conclusions Because 2D planimetry tends to overestimate MVA, 3D TOE should be considered for accurate MVA assessment, especially in patients with a large LA and large M alpha.