Assessment of right ventricular volumes in hypoplastic left heart syndrome by real-time three-dimensional echocardiography: comparison with cardiac magnetic resonance imaging

Abstract

Accurate assessment of right ventricular (RV) volumes and function is important in patients with hypoplastic left heart syndrome (HLHS). We prospectively sought to determine the reproducibility of three-dimensional (3D) echocardiography and its agreement with cardiac magnetic resonance imaging (CMR) in HLHS. Twenty-eight patients underwent CMR followed immediately by transthoracic 3D echocardiography under general anaesthesia. Semi-automated border detection software was used to determine echocardiographic RV volumes. Inter- and intra-observer variability, correlation and levels of agreement between techniques were determined. The median age was 0.37 years (0.18-9.28 years) and weight 6.24 kg (3.42-32.50 kg). Intra- and inter-observer variability was excellent for both techniques. Median (range) measurements for 3D echocardiography and CMR were; end-diastolic volume (EDV) 23.6 mL (6.5-63.2) and 30.6 mL (11.8-87.9), end-systolic volume (ESV) 12.6 mL (3.7-37.0) and 14.9 mL (5.8-33.9), stroke volume (SV) 11.2 mL (2.8-33.0) and 17.1 mL (6.0-54.1), ejection fraction (EF) 48.2% (31.2-64.9), and 56.5% (42.7-72.2). Correlation coefficients were r = 0.85, 0.84, 0.83, and 0.74, respectively (P < 0.01 for all). Volumetric data were expressed as a percentage of the echocardiographic volume to CMR volume. When compared with CMR, 3D echocardiography underestimated EDV, ESV and SV by 26.7% (SD ± 20.2), 10.6% (±28.1), and 37.5% (±20.1), respectively. The difference in volume appeared largest at low ventricular volumes. EF was 8.3% (±7.3) lower by 3D echocardiography compared with CMR. Both 3D echocardiography and CMR volumes appear highly reproducible. Measurements obtained by 3D echocardiography are significantly lower than those obtained by CMR, with wide limits of agreement such that these two methods cannot be used interchangeably.