Assessment of left atrial functional parameters using a novel dedicated analysis tool for real-time three-dimensional echocardiography: validation in comparison to magnetic resonance imaging

Buechel, Ronny ; Sommer, Gregor ; Leibundgut, Gregor ; Rohner, Andreas ; Riede, Florian ; Kessel-Schaefer, Arnheid ; Kaufmann, Beat ; Zellweger, Michael ; Bremerich, Jens ; Handke, Michael

In: The International Journal of Cardiovascular Imaging, 2013, vol. 29, no. 3, p. 601-608

Ajouter à la liste personnelle
    Summary
    Real-time three-dimensional echocardiography (RT3DE) is superior to two-dimensional echocardiography in assessing left atrial (LA) parameters, but to date algorithms developed for the left ventricle were applied due to a lack of dedicated LA software. In addition, no data are available on RT3DE assessment of active atrial contraction. The aim of this study was to validate a novel RT3DE analysis tool specifically dedicated to evaluate the LA. Cardiac magnetic resonance imaging (MRI) served as standard of reference. Fifty-five patients scheduled for pulmonary vein isolation underwent cardiac MRI and RT3DE. On ultrasound image datasets, a dynamic polyhedron model of the LA was generated from which LA maximum and minimum volumes (LAmax and LAmin), passive atrial emptying fraction (LAEF), and active atrial ejection fraction (LAEFtrue) were derived and compared to values obtained from cardiac MRI. High intraclass correlations between RT3DE and MRI were found for LAmax (r=0.94, p<0.001), LAmin (r=0.95, p<0.001), LAEF (r=0.92, p<0.001), and LAEFtrue (r=0.87, p<0.001). Similarly, Bland-Altman analysis revealed narrow limits of agreement for LAmax (−28.6 to 14.1ml), LAmin (−26.8 to 12.4ml), LAEF (−11.2 to 14.9%), and LAEFtrue (−10.6 to 6.8%). LAmax, LAmin and LAEFtrue were measured significantly (p<0.05) lower by RT3DE (111±38ml vs. 118±39ml, 73±38ml vs. 80±41ml, and 23±14% vs. 27±14%, respectively). Interobserver and intraobserver RT3DE measurements correlated closely. RT3DE using a novel dedicated software tool is valid, accurate and reproducible for assessing LA dimensional and functional parameters. This study corroborates previous reports and extends its validity to the assessment of active LA contraction