Echocardiographic Strain Assessment
Strain assessment was performed off-line using vector velocity imaging (Siemens Acuson SC 2000, version 5). All images analyzed were obtained at 50-80 frames per second at end expiration. The region of interest was traced with a point-and-click approach on the endocardium of the RV free wall at end-diastole in the RV-focused apical four-chamber view. A broader region of interest was subsequently generated and manually adjusted if necessary. The Program automatically divided the RV free wall into three segments and performed the analysis of the deformation frame by frame. This process allowed an automated confirmation of the contour and generated deformation values. The peak strain values from the three free wall segments were averaged, and the mean value was taken as the RV free wall strain (RVFWS) (5,7,11, 12, 13).
For the right atrium (RA), the endocardial border was traced in the apical 4-chamber view, excluding the appendage and the Eustachian valve from the RA cavity. RA longitudinal strain curves were generated throughout the cardiac cycle with R-R gating. Accuracy of the automated border tracking was verified and manually adjusted if needed. Tracking was repeated three times, and averages were used for analysis as reported in guidelines (5,8, 11). The peak RA reservoir strain in ventricular systole, conduit strain in early diastole and peak contractile phase strain during atrial systole/ late diastole were measured and expressed as percentage. RA total reservoir phase and RA contractile phase were assessed by measuring the corresponding peak strains. The conduit strain was calculated as the difference between RA total reservoir strain and RA contractile strain (Figure 1 ).
To assess intra- and inter-observer reproducibility of RA reservoir, conduit, and contractile strain, thirteen randomly sampled analyses were repeated twice by the same observer and by a second observer without knowledge of previous findings, respectively.