Mitral annular disjunction in Marfan syndrome: a multicenter cardiovascular magnetic resonance study.

backgroundData on the prevalence of mitral annular disjunction (MAD) in Marfan syndrome (MFS) based on cardiovascular magnetic resonance (CMR) is sparse. The purpose of this study was to assess prevalence, extent, and distribution of MAD in MFS using CMR and to examine its association with left heart parameters, aortic dimensions, and cardiovascular events.

methodsThis retrospective multicenter study included CMR studies of patients treated for MFS at four tertiary care medical centers with a (likely) pathogenic fibrillin-1 gene variant. Two radiologists (5 and 8 years of experience in CMR) evaluated datasets for MAD (at 4 points around the annulus, including measurement of extent) and mitral valve prolapse (MVP). Further assessment comprised volumetric and functional analysis of the left ventricle (LV), left atrial size, and aortic root diameters. Cardiovascular events included aortic (aortic surgery or aortic dissection), arrhythmic (sustained ventricular tachycardia or sudden cardiac death), and mitral events (mitral valve surgery, MVS).

resultsAmong 91 patients [(28.9±14.0 years, 47.3% female (n = 43/91)]81.3% (n = 74/91) had MAD (extent: 6.1±2.6 mm). MAD was mostly found at the inferior insertion (72.5% of patients, n = 66/91) and usually affected all sites (39.6% of patients, n = 36/91). Left heart parameters and aortic dimensions did not differ between MAD and no MAD groups (P>0.05). MAD extent and localizations showed significant correlations with LV dilatation (e.g., inferior MAD: r=0.62 for end-diastolic volume index), decreased LV ejection fraction (e.g., anterolateral MAD: r=-0.46), and MVP (e.g., MAD distance: r=0.83), which was found in 44.6% of patients (n = 33/74) with MAD while only affecting 11.8% (n = 2/17) without MAD (P=0.017). Based on receiver operating characteristic analysis for the prediction of MVP prevalence, a threshold of 7.1 mm MAD extent was identified as the optimal cut-off value (sensitivity: 77.1%, specificity: 89.3%). Additionally, subgroup analysis applying different thresholds of MAD extent revealed a significantly larger displacement of MVP and LV volumes as well as higher aortic root z scores for a threshold of ≥8 mm. After a mean follow-up of 4.0±3.0years, cardiovascular events (aortic: n=13/91 [14.3%], arrhythmic: n= 2/91 [2.2%], and mitral: n=2/91 [2.2%] of patients) did not differ significantly (all P>0.05) between no MAD and MAD groups regardless of applied thresholds although MVS was observed exclusively in patients with MAD.

conclusionThe high prevalence, large extent, and predominantly pan-annular distribution of MAD suggest a systemic annular pathology in MFS. Overall presence of MAD was not associated with changes to left heart parameters, aortic dimensions, and cardiovascular events. However, MAD, taking into account its extent and affected insertion sites, could serve as a potential marker of disease progression given the shown association of localizations and distance with LV dysfunction and remodeling as well as aortic enlargement and the formation of MVP.