P787 Interstudy reproducibility in the real world of speckle tracking echocardiography in direct comparison with cardiac magnetic resonance

Abstract

Abstract Funding Acknowledgements Fond national de la recherche scientifique (FNRS) Background New research methods to assess cardiac function such as 2D speckle tracking echocardiography (STE) are emerging in clinical practice after showing incremental prognostic information beyond ejection fraction in a variety of settings. However, comprehensive data regarding the day to day reproducibility is still lacking. Aim To assess the inter and especially the day to day intravariability of RV and LV GLS STE in direct comparison with cMR in asymptomatic volunteers and in patients with heart failure. Methods for the volunteers:30 asymptomatic volunteers ( 34 ± 9years, 33% were women) underwent 2 cMR and 2 echocardiography studies with a maximal interval between each study of 6 days and 20 minutes between each technique. RV strain was performed on a RV focus view and LV strain on 4, 3 and 2 chamber views using the segment software. cMR- RVEF and cMR-LVEF was performed in short axis view. Preliminary Results: Volunteers Average values for the the STE measurements was -24.2%±2.3 and -27.4%±2.1 for LV and RV GLS respectively. Average values for the 2 cMR measurements was 61.9% ±4.5 and 56.3%± 6.2% for cMR-LVEF and RVEF respectively. The interstudy varability coefficient of variability was lower in GLS-STE parameters (RV-GLS = 7.7% and LV-GLS = 6.3%) than for the cMR parameters (cMR-LVEF= 8.2% and cMR-RVEF= 12.9%). The superior interstudy reproducibility resulted in lower calculated sample sizes require by GLS compared with cMR to show clinically relevant changes in LV and RV function. (Figure 1) Conclusion LV and RV -GLS- STE has excellent interstudy reproducibility in normal hearts and is as good as cMR- LVEF and RVEF. Echocardiography cMR LV-GLS-STE RV-GLS-STE LVEF RVEF Mean difference ±SD 0.6 ± 1.88 0.5± 1.73 1.14 ± 5.22 0.98 ± 7.23 Coefficient of variation 7.7% 6.3% 8.2% 12.9% Sample Size for 2% absolute change in GLS or 5% absolute change in cMR- EF 18 16 22 44 Sample size required to detect a clinically significant change in global longidutdinal strain speckle tracking (GLS-STE) echocardiographie and cMR - ejection fraction (EF). RV and LV indicate Right ventricular and left ventricular respectively.