Reproducibility and repeatability of biventricular function/volume and strain parameters by 2D and 4D stress echocardiography in adult patients with repaired TOF

Abstract

Abstract Funding Acknowledgements Type of funding sources: None. Background The use of 2D and 4D during stress echocardiography to undertake complex measures in complex patients like patients with repaired TOF is challenging and the validity of these measures is not known yet. Methods For test-retest variability, 20 patients with repaired TOF with no or mild pulmonary regurgitation were selected randomly and underwent a cardiopulmonary exercise test (CPET) with echocardiography. Intra-observer variability study was performed for all 20 patients by the same observer. Interobserver variability study was performed for 5 patients by different experienced observer. Intraclass correlation coefficient (ICC), and coefficients of variation (COV) were used to quantify reproducibility and variability. Results For 2D measures, better reproducibility was observed for semiautomated 2D strain measures than 2D functional measures for biventricular systolic function at baseline and during the stress (ICC>.90 vs >.70, p<0.001), with least COV was observed (COV<10%). 4D semiautomated volumetric measures demonstrated less reproducibility during stress with highest COV was observed for 4D RV volume parameters (COV, 35%), followed by 4D LV volume parameters (COV, 27%). CPET had an excellent agreement of all measures (ICC >.90). with very low COV (<10%). Conclusions Semiautomated echo measures outperformed manual measures during stress echocardiography and can be performed with acceptable reproducibility. Variability is at highest for 4D semiautomated measures despite good reproducibility, while lowest variability was observed for 2D semiautomated measures of myocardial deformation.