Comparison of two-dimensional and three-dimensional echocardiographic strain in children with CHD

Background: In CHD, three-dimensional strain analysis may overcome limitations of Doppler and two-dimensional strain of the left ventricle. The aims of this study were to evaluate feasibility and reproducibility of three-dimensional longitudinal, circumferential, and radial systolic strain by three-dimensional speckle-tracking echocardiography compared with two-dimensional echocardiography. Methods: Patients with CHD, biventricular circulation with a systemic left ventricle, and who had two-and three-dimensional imaging performed on the same day from 2010 to 2014 were included. Quantitative two-and three-dimensional strain analyses were performed (two-dimensional cardiac performance analysis version 1.2 and four-dimensional left ventricular analysis version 3.1). Intra-and inter-observer variabilities were calculated on 25 studies. Results: A total of 30 patients, including 19 (61%) males, with a median age of 3.6 years (0.1-22 years) were included. The mean fractional shortening was 34.6 +/- 5.3%, and the mean ejection fraction was 62.0 +/- 6.4%. Measurement of two-and three-dimensional strain was feasible in >95% of segments. Good correlation was observed between longitudinal and circumferential strain (r = 0.92, p <= 0.001 and r = 0.87, p <= 0.001), but not radial strain (r = 0.29, p = 0.2). Intra-and inter-observer agreements were better for three-dimensional compared with two-dimensional strain, and better for both two-and three-dimensional longitudinal and circumferential strains compared with radial strain. Conclusion: Left ventricular three-dimensional strain analysis is feasible in children with CHD. The reproducibility of longitudinal and circumferential strain by three-dimensional analyses is better. Further longitudinal studies are warranted for the potential clinical application of this new technology.