Segmental analysis of microvascular dysfunction and tissue characterization in hypertrophic cardiomyopathy by magnetic resonance imaging

Abstract Funding Acknowledgements Type of funding sources: None. Background While left ventricular hypertrophy (LVH) and myocardial fibrosis are frequently evaluated in hypertrophic cardiomyopathy (HCM), microvascular dysfunction is often overlooked. The aim of this study was to assess the association between wall thickness, microvascular dysfunction and tissue characteristics in HCM patients (P), comparing individual myocardial segments. Methods Prospective evaluation of adult P with HCM (P with "end-stage" HCM, prior septal reduction therapy or epicardial coronary artery disease were excluded). All underwent a cardiac magnetic resonance (CMR) protocol (1.5-T), from which the following parameters were analysed: maximal LV wall thickness (MLVWT), T1 and T2 mapping, extracellular volume (ECV), late gadolinium enhancement (LGE) and stress perfusion. Results were stratified according to the 16 American Heart Association segments. Multivariate regression analyses for perfusion defects and tissue characteristics were performed. Results 75 P (total of 1200 myocardial segments analysed), 63% male, mean age 55 ± 15 years, MLVWT of 20 ± 4.5mm (61% asymmetric septal LVH, 29% apical LVH and 8% concentric LVH). 28% presented LV outflow tract obstruction. MLVWT was greater in the basal and mid septum (table 1). Among the 424 segments (35.3%) with a perfusion defect, 286 (23.8%) had a defect only in the endocardial layer and 138 (11.5%) in both endocardial and epicardial layers with defects more often detected in hypertrophied segments. This association was verified in segments with MLVWT 12-14mm and MLVWT ≥15mm (OR 7.83, 95% CI 5.75-10.67, p < 0.001) (table 2). Among the 660 segments with normal MLVWT (≤11mm), 123 (19%) presented perfusion defects. A perfusion defect was more frequent in segments of obstructive HCM P (OR 1.48, 95% CI 1.13-1.92, p = 0.004). Microvascular dysfunction was associated with changes in tissue characteristics. For the same thickness, segments with perfusion defects had a higher T1 mapping (β-estimate 20.91, 95% CI 16.87-24.96, p < 0.001) and T2 mapping mean values than those without. Furthermore, regardless of MLVWT, segments with perfusion defects had LGE more often (OR 4.16, 95% CI 3.19-5.41, p < 0.001) and a higher ECV. On the other hand, among the 424 segments with a perfusion defect, 115 (27%) did not present LGE. MLVWT was associated with tissue characteristics. Comparing to non-hypertrophied segments, T1 mapping mean value was higher in segments with MLVWT 12-14mm and with ≥15mm (β-estimate 37.71, 95% CI 31.2-44.21, p < 0.001). Segments with MLVWT ≥15mm showed a significantly superior ECV mean value comparing to non-hypertrophied segments. LGE was more frequent in the more hypertrophied segments: MLVWT 12-14mm and ≥15mm (OR 9.02, 95% CI 6.42-12.67, p < 0.001) (Table 2). Conclusion Microvascular dysfunction is more prevalent in obstructive HCM, particularly in the more hypertrophied segments. The presence of microvascular dysfunction is associated with diffuse tissue abnormalities and replacement fibrosis. Abstract Table 1: Characteristics of AHA segments Abstract Table 2: Multivariable analyses