Geert Kleinnibbelink

Chapter 4 94 were exported and a cubic spline was applied to normalize for heart rate. This allowed the presentation of temporal strain and rotation across the cardiac cycle. Strain-volume/area loops. The longitudinal strain-volume/area relationship (for methodology of derivation, see Supplemental 1 and Oxborough et al. 14 ) was assessed using the following parameters: (I) the linear strain-area slope (Sslope) and early strain- area slope during first 5% of volume ejection in systole (ESslope); (II) end-systolic peak longitudinal strain (peak strain); (III) the early linear strain-area slope during first 5% (EDslope) and late linear strain-area slope (LDslope) during last 5% of volume increase in diastole; and (IV) diastolic uncoupling (i.e. difference in strain between systole and diastole at any given area), divided into uncoupling during early (Uncoupling ED) and late diastole (Uncoupling LD). 14, 15 Statistical analysis Statistical analyses were performed using SPSS Statistics 24 (SPSS, Inc., Chicago, Illinois). All parameters were visually inspected for normality and tested with Shapiro- Wilk normality tests. Continuous variables were reported as mean±SD and categorical variables were presented as proportions. Paired-Samples T-tests were used to compare echocardiographic continuous variables between the baseline and follow-up evaluation. Comparison of sex differences was performed using repeatedmeasurements ANOVA with Bonferroni post hoc correction for multiple comparisons. Consistency of intra-observer measurements of selected measurements were verified through the intra-class correlation coefficient (ICC). Therefore, both echocardiographs of 15 randomly chosen subjects were analysed by the same operator blinded from earlier results. ICC coefficients were as follows: RVOT-PLAX 0.96 (0.91-0.98), LVISd 0.91 (0.81-0.96), LVPWd 0.90 (0.80-0.95), LVIDd 0.97 (0.93-0.98), RV basal diameter 0.98 (0.96-0.99), RVEDA 0.98 (0.95-0.99), RA area 0.99 (0.97-0.99), LA volume 0.98 (0.95-0.99). In previous studies, we showed that strain measurements and both right and left ventricular loops have a good to excellent inter-observer variability. 14-17