Abnormal cardiac magnetic resonance-derived ascending aortic area strain demonstrates altered ventriculo-vascular function in Marfan syndrome

Abstract

Abstract Funding Acknowledgements Type of funding sources: None. Background Surveillance strategies for aortic disease based on the current guidelines support the utilization of maximum aortic diameters to evaluate the risk of aortic dissection, however there remains a need for improved imaging markers for this risk stratification. Purpose This study aimed to compare cardiac magnetic resonance (CMR)-derived ventriculo-vascular properties in patients with Marfan syndrome against normal controls, and to determine the relationship between aortic area strain and other clinical variables in this population. Methods Ascending (AAAS) and descending aortic area (DAAS) strain and left ventricular (LV) longitudinal (LS) and circumferential (CS) strain were retrospectively obtained from 20 CMR studies from 17 patients with Marfan syndrome. Aortic area strain was estimated using maximal and minimal luminal areas obtained from cross-sectional aortic images at the level of the pulmonary artery and was calculated as follows: (100 × (Areamax – Areamin)/Areamin). Additionally, LV mass (LVM), end-diastolic volumes (LVEDV) and ejection fraction (LVEF) were measured. Ventriculo-vascular properties were compared to a cohort of CMR studies from 25 healthy controls. Continuous variables were expressed as mean ± standard deviation (SD) or median (interquartile range). Relationships between aortic area strain and clinical variables were investigated using Spearman’s correlation coefficient. Results Mean Marfan patient age at CMR was 16.7 [10.1, 32.9] years. Twelve patients underwent genetic testing and had a Fibrillin-1 (FBN-1) mutation, other patients met the diagnostic criteria for Marfan syndrome. Four patients (4 studies) had undergone surgical aortic root replacement prior to CMR examination. When compared with the healthy controls, LVLS and LVCS were lower in Marfan (−16.0 ± 4.9 vs −19.7 ± 4.8, p = 0.015 and −18.5 ± 3.8 vs −27.0 ± 4.1, p <0.001; respectively) while AAAS was significantly decreased (19.4 [14.3, 25.1] vs 46.1 ± 11.3, p <0.001). LV EDV indexed to body surface area was significantly greater in Marfan (126.0 ± 45.1 vs 78.4 ± 11.9, p <0.001). No significant differences where observed in LVEF (59.3 [55.4, 56.9] vs. 60.1 ± 3.5, p = 0.440) or DAAS (30.3 [14.6, 44.8] vs 41.7 ± 6.4, p = 0.400). Both AAAS and DAAS did not show any significant correlations to LVLS, LVM, LVEF or LVEDV. Correlation analyses demonstrated that increasing age was associated with lower aortic strain (AAAS, R= −0.57, p = 0.010 and DAAS, R = -0.58, p = 0.008; respectively). Conclusions Ascending aortic area strain was reduced in Marfan syndrome compared to normal healthy controls, indicating altered vascular function. Both LVLS and LVCS were lower in Marfan compared to controls. AAAS shows promise as an aortic imaging marker in Marfan, and larger studies examining it in combination with standard aortic dimension measurements are needed to fully ascertain clinical value.