P1782 Left ventricular deformation proprieties in aortic coarctation: insights from multilayer strain imaging

Abstract

Abstract Background Patients with coarctation of the aorta (CoA) have proven impaired left ventricular (LV) myocardial deformation that will persist even after successful transcatheter intervention compared to normal controls. However, little is known about layer specific contraction in this clinical setting. Purpose To investigate multi-layer longitudinal strain in patients with repaired CoA, compared with a control group of healthy subjects. Methods In a case-control study, 13 CoA patients (F/M = 9/4, age = 15.1 ± 4.4 years) and 13 healthy age- and sex-matched controls, underwent a complete echocardiogram, including speckle tracking assessment for quantification of LV transmural global longitudinal strain (GLS), subendocardial longitudinal strain (LSsubendo), subepicardial longitudinal strain (LSsubepi), and myocardial longitudinal strain gradient (LSsubendo - LSsubpepi). Results CoA patients had similar blood pressure, heart rate, and body surface area in comparison with healthy controls. The two groups did not differ for ejection fraction and LV diastolic indices. Increased maximal gradient (>20 mmHg) in descending aorta was found in 9 (69.2%) CoA patients. In CoA population, 5 (38.5%) had LV concentric remodeling (RWT > 0.42) and 2 (15.4%) LV hypertrophy (Z score >2.0). LV mass index (92.4 ± 31.1 vs. 60.9 ± 12.5 g/m^2.7, p = 0.002), septal wall thickness (8.7 ± 2.0 vs 6.6 ± 0.9 mm p = 0.002), posterior wall thickness (8.2 ± 1.7 vs. 6.6 ± 1.4 mm, p = 0.02) and RWT (0.38 ± 0.06 vs. 0.27 ± 0.08, p < 0.001) were greater in CoA patients. By multi-layer deformation analysis, GLS (21.2 ± 1.9 vs. 22.9 ± 1.4%, p < 0.01), LSsubepi (19.3 ± 1.6 vs. 21.0 ± 1.6%, p < 0.01), LSsubendo (22.9 ± 1.8 vs. 25.4 ± 1.9%, p = 0.003) and longitudinal strain gradient (3.6 ± 1.1 vs. 4.7 ± 1.1, p < 0.02) were lower in CoA patients than in healthy controls. Separate sub-analyses in patients without LV hypertrophy, substantially confirmed the same results: GLS 20.9 ± 1.6 vs. 22.8 ± 0.9%, p = 0.004; LSsubepi 19.0 ± 1.4 vs. 21.1 ± 1.3%, p = 0.004; and LSsubendo 22.4 ± 1.2 vs. 25.3 ± 1.4%, p < 0.0001; strain gradient 3.4 ± 1.0 vs. 4.6 ± 1.1, p = 0.02. Notably, LSsubepi (18.6 ± 1.4 vs. 20.6 ± 1.0%, p = 0.03) and, with greater significance, LSsubendo (22.1 ± 1.1 vs. 24.7 ± 1.8%, p < 0.0001) were lower in CoA patients with increased aortic gradient versus those without significant gradient increase. Conclusions In CoA patients, layer specific strain imaging highlights an impairment of LV longitudinal deformation, which mainly involves LSsubendo and causes reduction of longitudinal strain gradient. This preferential impairment of subendocardium is particularly evident in patients with residual aortic gradient and is independent of LV hypertrophy. Abstract P1782 Figure. Multilayer strain in CoA vs. controls