Determinants of discrepancies between two-dimensional echocardiographic methods for assessment of maximal left atrial volume

AIMS: The determinants of discrepancies among two-dimensional echocardiographic (2D-E) methods for left atrial volume (LAV) assessment are poorly investigated. METHODS AND RESULTS: Maximal LAV was measured in 613 individuals (282 healthy subjects,180 athletes, and 151 hypertensives; age 45 ± 20 years, 62% male) using the ellipsoid model (LAVEllips), the area-length method (LAVAL), and the Simpson's rule (LAVSimps). On the basis of a mathematical model, two left atrial (LA) geometry indexes were tested as predictors of discrepancies between methods: the ratio between LA medial-lateral diameter (MLD) and LA anteroposterior diameter (APD); and the ratio between LA area in the four-chamber view and that of an ellipse with the same diameters [deviation from ellipse (DE)-coefficient]. Discrepancies among methods were consistently present in the overall population and across all study groups. MLD/APD and the DE-coefficient together predicted 76 and 68% of differences between biplane LAVAL and LAVEllips, and between biplane LAVSimps and LAVEllips, respectively. The DE-coefficient was the only determinant of LAVAL/LAVSimps difference (β = 0.167, P < 0.0001). Body mass index was the strongest predictor of discrepancies between single-plane and biplane approaches of LAVAL (β = 0.427, P < 0.0001) and LAVSimps (β = 0.424, P < 0.0001). In additional analyses, biplane LAVAL showed the best agreement with LAV obtained by three-dimensional echocardiography and the best reproducibility and repeatability. CONCLUSION: LA geometry is the main determinant of inconsistencies between 2D-E methods for measuring maximal LAV. Body mass index is the strongest determinant of differences between single-plane and biplane approaches. Different 2D-E methods cannot be used interchangeably for diagnosis and follow-up. The biplane area-length method should be preferred, particularly in overweight-obese subjects.