Evaluation of automated measurement of left ventricular volume by novel real-time 3-dimensional echocardiographic system: Validation with cardiac magnetic resonance imaging and 2-dimensional echocardiography

• Published in: Journal of cardiology Vol. 61; no. 4; pp. 281 - 288
• Main Authors: Shibayama, Kentaro, MD, Watanabe, Hiroyuki, MD, FJCC, Iguchi, Nobuo, MD, Sasaki, Shunsuke, MD, Mahara, Keitaro, MD, Umemura, Jun, MD, Sumiyoshi, Tetsuya, MD, FJCC
• Format: Journal Article
• Language: English
• Published: Netherlands Elsevier Ltd 01-04-2013
• Subjects: Aged; Algorithms; Cardiac Volume; Cardiovascular; Computer Systems; Echocardiography; Echocardiography, Three-Dimensional - methods; Ejection fraction; Feasibility Studies; Female; Heart Diseases - diagnostic imaging; Heart Diseases - physiopathology; Heart Ventricles - pathology; Heart Ventricles - physiopathology; Heart Ventricles - ultrastructure; Humans; Magnetic Resonance Imaging; Male; Middle Aged; Sensitivity and Specificity; Software; Transthoracic; Ventricular function; Ventricular Function, Left; Ventricular function; Echocardiography; Magnetic resonance imaging; Transthoracic; Ejection fraction
• ISSN: 0914-5087; 1876-4738
• DOI: 10.1016/j.jjcc.2012.11.005

Abstract Background Traditional 3-dimensional echocardiography (3DE) with volumetric scanning technique requires several heart cycles for full-volume acquisition and complicated manual contouring of left ventricular (LV) endocardium. The new real-time 3DE (RT3DE) system allows acquisition of an instantaneous full-volume dataset in a single heart cycle and automated measurement of LV volume by the algorithm software. However, it has not been evaluated adequately whether automated measurement by RT3DE has better agreement with cardiac magnetic resonance imaging (CMR) than 2-dimensional echocardiography (2DE) with CMR. Purpose This study aimed to evaluate the accuracy of automated measurement of LV volume using RT3DE compared with 2DE and CMR. Methods and results Forty-four consecutive patients who underwent RT3DE, 2DE, and CMR were evaluated in this study. The feasibility of automated measurement by RT3DE was 93.2% and the mean operation time was 6 min. LV volume and ejection fraction (EF) from semi-automated measurement [end-diastolic volume: r = 0.96, limits of agreement (LOA) −30.5 to 39.3 ml; end-systolic volume: r = 0.97, LOA −22.6 to 32.7 ml; EF: r = 0.90, LOA −16.1 to 14.2%, respectively] had better agreement with CMR than those from 2DE ( r = 0.87, LOA −50.5 to 72.2 ml; r = 0.93, LOA −34.1 to 65.2 ml; r = 0.89, LOA −20.9 to 10.0%, respectively). Conclusion Semi-automated measurement by RT3DE has better agreement with CMR than 2DE in LV volume and EF. In addition, it is simple to operate and acceptable in feasibility for the clinical setting although there may be room for further learning required to incorporate small hypertrophic LV into the automated algorithm software.