Relation Between Number of Component Views and Accuracy of Left Ventricular Mass Determined by Three-Dimensional Echocardiography

• Published in: The American journal of cardiology Vol. 99; no. 9; pp. 1321 - 1324
• Main Authors: Chuang, Michael L., MD, Salton, Carol J., BA, Hibberd, Mark G., MD, PhD, Manning, Warren J., MD, Douglas, Pamela S., MD
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 01.05.2007; Elsevier Limited
• Subjects: Adult; Aged; Cardiology; Cardiomyopathy, Dilated - diagnostic imaging; Cardiomyopathy, Dilated - pathology; Cardiovascular; Cardiovascular system; Comparative studies; Echocardiography, Three-Dimensional - methods; Ethnicity; Female; Heart; Heart Ventricles - diagnostic imaging; Heart Ventricles - pathology; Humans; Image Processing, Computer-Assisted; Male; Middle Aged; Organ Size; Reproducibility of Results; Time Factors; Women
• ISSN: 0002-9149; 1879-1913
• DOI: 10.1016/j.amjcard.2006.12.045

Three-dimensional echocardiography (3DE) allows the accurate determination of left ventricular (LV) mass, but the optimal number of component or extracted 2-dimensional (2D) image planes that should be used to calculate LV mass is not known. This study was performed to determine the relation between the number of 2D image planes used for 3DE and the accuracy of LV mass, using cardiovascular magnetic resonance (CMR) imaging as the reference standard. Three-dimensional echocardiography data sets were analyzed using 4, 6, 8, 10 and 20 component 2D planes as well as biplane 2D echocardiography and CMR in 25 subjects with a variety of LV pathologies. Repeated-measures analysis of variance and the Bland-Altman method were used to compare measures of LV mass. To further assess the potential clinical impact of reducing the number of component image planes used for 3DE, the number of discrepancies between CMR and each of the 3DE estimates of LV mass at prespecified levels (i.e., ≥5%, ≥10%, and ≥20% difference from CMR LV mass) was tabulated. The mean LV mass by magnetic resonance imaging was 177 ± 56 g (range 91 to 316). Biplane 2-dimensional echocardiography significantly underestimated CMR LV mass (p <0.05), but LV mass by 3DE was not statistically different from that by CMR regardless of the number of planes used. However, error variability and Bland-Altman 95% confidence intervals decreased with the use of additional image planes. In conclusion, transthoracic 3DE measures LV mass more accurately than biplane 2-dimensional echocardiography when ≥6 component 2D image planes are used. The use of >6 planes further increases the accuracy of 3DE, but at the cost of greater analysis time and potentially increased scanning times.