Long-axis myocardial strain assessment from cardiovascular magnetic resonance feature tracking in the German National Cohort (NAKO)

Abstract Background/Introduction While feature-tracking (FT) strain assessment has an increasing impact in cardiovascular magnetic resonance (CMR), its widespread application is limited by the lack of standardization and universal reference values. Although initiatives driven by the CMR community provided cut-offs, there is still a lack of population-wide reference values as well as data on the effect of sex and age. The German National Cohort (NAKO) is a large population-based study with randomly selected participants across Germany which includes a dedicated magnetic resonance imaging protocol. Purpose To assess the effect of sex and age on FT-based strain parameters derived from long-axis (LAX) CMR cine imaging in the NAKO. Methods Myocardial strain was assessed from LAX balanced steady state free precession-based cine images of 29,070 NAKO participants in three orientations (2-,3- and 4-chamber views) with 25 cardiac phases using dedicated software (CVI42 version 5.17.3). In a previous step, 1.359 cases with multiple or missing acquisitions were manually inspected for series selection and quality control. Segmentation was performed automatically in end-diastole and end-systole for the left and right ventricles (LV and RV respectively). FT-based strain values were derived for LV and RV global longitudinal strain (GLS) and radial strain (GRS). Implausible values (for example positive GLS) were excluded. GLS values were compared by sex (male, female) and age across the following age groups: 19-29, 30-39, 40-49, 50-59 and 60-74 years. A two-factor analysis of variance (ANOVA) was performed to investigate the effects of age group and gender on myocardial strain. Results Of 29,070 data sets, 28,256 (97.2 %) were available for the final analysis (mean age: 48.2 ± 12.3 years; 55.5% men). With increasing age, LV GLS increased, and GRS decreased, indicating a decline in LV function. In contrast, RV GLS decreased, and GRS increased with age in both sexes, suggesting an improvement in RV function (Figure 1 and 2). The ANOVA consistently revealed highly significant main effects of age group (all F 217, p 0.001) and gender (all F 396, p 0.001) on myocardial strain. Conclusion This study demonstrates that strain assessment in CMR needs to take sex and age into account, especially as the LV and RV show different trends of functional decline and incline, respectively. The age-dependent increase in RV strain values was unexpected and may indicate adaptive or compensatory mechanisms in the context of age-related changes in cardiopulmonary interaction.Sex and age impact on biventricular stra Strain values according to sex and age g