Geert Kleinnibbelink

Cardiac Remodelling in Olympic Athletes 4 111 APPENDICES Supplemental 1. Strain-Area Loop – methods of derivation To calculate right ventricular (RV) strain-area loops the following steps have been taken. Temporal longitudinal strain values were exported to a spreadsheet (Excel; Microsoft Corp., Redmond, WA, USA). Using cubic spline interpolation, the global temporal longitudinal strain values were divided into 300 points for systole and 300 points for diastole in order to correct for variable heart rates. For both systole and diastole, the 300 strain values were then split into 5% increments of the cardiac cycle, providing longitudinal strain values at 10 time points in systole and 10 time points in diastole. Concomitant time points for the strain values were used in the same image and cardiac cycle to trace RV monoplane areas to provide simultaneous strain and area values. Using the individual strain–area loop, a linear regression line and a polynomial of two orders was applied to both diastolic and systolic parts of the loop. This derived polynomial equation allowed the derivationof strain at percentage increments of RV end-diastolic area (RVEDA). The longitudinal strain–area loop was assessed using the following parameters ( Figure 1 ): (a) early linear slope during first 5% of volume ejection in systole (ESslope), (b) the overall linear slope during systole (Sslope) and (c) end-systolic peak longitudinal strain (peak strain). In addition (un)coupling was termed to describe the relationship between systolic and diastolic strain for any given area/volume and was assessed during (d) early (Uncoupling ED), (e) late diastole (Uncoupling LD) and (f ) overall. Furthermore, (g) the early linear slope during first 5% (EDslope) and (h) late linear slope (LDslope) during last 5% of volume increase in diastole. The Sslope was derived as the gradient of the linear regression line over the systolic phase of the strain–area loop. Longitudinal peak strain was derived as the raw peak strain value from the longitudinal strain data. The Uncoupling ED and Uncoupling LD were calculated across the area between the systolic and diastolic polynomial curves. Using the equations of the polynomial regression lines, strain at % increments of RVEDA were calculated. By subtracting diastolic from systolic strain, the difference at each point was calculated. Based on individual RV fractional area change (RVFAC), the working range of the heart