Contemporary Left Ventricular Mass Assessment via 2D Echocardiography–a Comparison with Cardiac Magnetic Resonance (CMR) Imaging

Abstract

Background: Left ventricular (LV) wall thickness/LV hypertrophy (LVH) has been assessed by echocardiography (echo) using LV m-mode in parasternal long axis view (PLAx). LV mass (LVM) calculations can be derived from m-mode measurements or 2D in the PLAx and more recently by the area/length method from the parasternal short axis view (PSSAx). The aim was to compare LV mass derived from both the 2D PSLAx and PSSAx methods with MRI calculated LV mass as the gold standard. Method: 103 patients referred for echocardiography and CMR within a 3-month period were included. Indexed measurements for LVM from the PLAx and PSSAx views were calculated using ASE recommendations. Three-dimensional, indexed mass was calculated from the CMR images by a separate experienced CMR observer, blinded to the echo data. Results: There was good correlation between both echo derived mass calculations and MRI using Pearson's correlation coefficient with the PSLAx r = 0.81 (p < 0.0001) and PSSAx r = 0.93 (p < 0.001). Using Bland-Altman analysis the bias (and limits of agreement) of the PSLAx was -117.74 g/m2 (-258.59 to 23.11 g/m2) and PSSAx was -1.18 g/m2 (-22.4 to 20.05 g/m2). Conclusion: Despite good correlation between echo and CMR mass, PSLAx calculated mass demonstrated a large bias. The PSSAx method is more robust but with broad limits of agreement suggesting that echo and CMR calculations of LVM are not interchangeable with echo overestimating LV mass. Given this, serial assessment of LVM should be performed by the same modality for follow-up and when using 2D echo, PSSAx is a more reliable and accurate method and hence preferable in clinical practice.