Simultaneous strain–volume analysis by three-dimensional echocardiography: validation in normal subjects with tagging cardiac magnetic resonance

Objective The aim of this study is to compare threedimensional echocardiography strain-volume analysis with tagging cardiac magnetic resonance (cMR) measurements. Background and methods Strain-volume analysis represents a noninvasive method to assess myocardial function and volumes simultaneously. It can be derived from echocardiography and speckle-tracking; however, it shows some variability that can limit clinical utilization. A threedimensional approach partially overcomes these limitations since full-volume acquisition avoids images being foreshortened and geometrical reconstruction. In the study presented here, 23 healthy subjects were studied by threedimensional echocardiography and cMR during the same session. Images were stored and the better cardiac cycle was chosen for simultaneous analysis of volumes and longitudinal (Long) and circumferential (Circ) strain. By means of full-volume acquisition all parameters can be calculated for each frame of the cardiac cycle using the speckle-tracking method. With cMR, left ventricle volumes were calculated as recommended; myocardial strains were computed in short-axis and long-axis views using the tagging technique. For each patient, volumes and strain values were plotted in a Cartesian system for strain-volume analysis. Data were compared between the two methods using Bland-Altman analysis based on mean difference and 95% limits of agreement (LoA). Results The volume as measured by three-dimensional echocardiography and cMR was comparable with the slightly higher end-diastolic volumes measured by cMR (mean difference 15.24 ml; LoA S53.6 to 26.5 ml, endsystolic volume 0.3 ml; LoA S19.9 to 20.5 ml). Long shortening was very similar in the two methods (1.5%; LoA S3.9 to 7%), whereas Circ strain was systematically lower with cMR (S8.5%; LoA S15.5 to S1.5%). Very similar values between three-dimensional echo and cMR both for Slope of strain-volume curves (S0.015; LoA S0.08 to 0.05) and ratio (S0.001; LoA S0.04 to 0.04) were observed in the longitudinal plane. Analysis of strain-volume per patient showed a significant correlation coefficient between techniques for both Long Slope (r = 0.65; P = 0.001) and Long Ratio (r = 0.70; P = 0.001). Conclusion Longitudinal strain-volume analysis performed with three-dimensional speckle-tracking echocardiography is closely comparable with cMR, which is usually considered the gold standard for volume and function assessment.