Intra- and inter-observer reproducibility of global and regional magnetic resonance feature tracking derived strain parameters of the left and right ventricle.

OBJECTIVES: To investigate the reproducibility of regional and global strain and strain rate (SR) parameters of both ventricles and to determine sample sizes for all investigated strain and SR parameters in order to generate a practical reference for future studies.

MATERIALS AND METHODS: The study population consisted of 20 healthy individuals and 20 patients with acute myocarditis. Cine sequences in three horizontal long axis views and a stack of short axis views covering the entire left and right ventricle (LV, RV) were retrospectively analysed using a dedicated feature tracking (FT) software algorithm (TOMTEC). For intra-observer analysis, one observer analysed CMR images of all patients and volunteers twice. For inter-observer analysis, three additional blinded observers analysed the same datasets once. Intra- and inter-observer reproducibility were tested in all patients and controls using Bland-Altman analyses, intra-class correlation coefficients (ICCs) and coefficients of variation.

RESULTS: Intra-observer reproducibility of global LV strain and SR parameters was excellent (range of ICCs: 0.81-1.00), the only exception being global radial SR with a poor reproducibility (ICC 0.23). On a regional level, basal and midventricular strain and SR parameters were more reproducible when compared to apical parameters. Inter-observer reproducibility of all LV parameters was slightly lower than intra-observer reproducibility, yet still good to excellent for all global and regional longitudinal and circumferential strain and SR parameters (range of ICCs: 0.66-0.93). Similar to the LV, all global RV longitudinal and circumferential strain and SR parameters showed an excellent reproducibility, (range of ICCs: 0.75-0.97). Radial strain and SR parameters were less reproducible in the LV as well as in the RV when compared to circumferential or longitudinal parameters.

CONCLUSION: CMR FT using the TOMTEC algorithm is highly reproducible in health and disease in both, the LV and RV on a global and regional level. The only exceptions are radial strain and SR parameters, which should be used with caution within clinical studies. The sample sizes estimated on the basis of the present study might serve as a reference during the planning of future FT studies.