Association of Mitral Annular Disjunction on Magnetic Resonance Imaging With Cardiovascular Outcomes

Mitral annular disjunction (MAD) is increasingly recognised on cardiac magnetic resonance (CMR) imaging, yet its clinical significance remains under discussion. We aimed to determine the prevalence, prognostic impact, and association with native T1 relaxation time of MAD assessed in both end-systole and end-diastole in an all-comer CMR cohort.

We analysed data from 1969 patients enrolled in a prospective CMR registry at the Vienna General Hospital. MAD was defined as ≥ 1 mm separation between the left atrial wall at the mitral valve hinge point and the adjacent left ventricular myocardium. Native T1 relaxation times were measured at septal and lateral positions of the mitral annulus in the 4-chamber view. All-cause mortality was analysed by Cox proportional hazards regression.

End-systolic MAD was present in 509 patients (25.9%), end-diastolic in 366 (18.6%). MAD patients were younger (59.6 ± 18.5 vs. 65.2 ± 17.9 years, p < 0.001), had fewer comorbidities, and had higher left ventricular ejection fractions (58.8% ± 10.4% vs. 56.8% ± 14.4%, p = 0.009). MAD distance correlated with native annular T1 relaxation time (r = 0.24, p < 0.001), and patients with MAD ≥ 5 mm showed higher T1 relaxation time than those < 5 mm (1041 ± 148 vs. 1010 ± 72 ms, p = 0.028), while mid-septal T1 relaxation time showed no difference (r = 0.06, p = 0.252). During 63.7 months median follow-up, 485 deaths occurred. After multivariable adjustment, end-diastolic MAD was not associated with mortality (HR 0.95, 95%-CI 0.74-1.23, p = 0.712).

MAD is common and prognostically neutral after adjustment. However, larger disjunction distances are associated with higher native T1 relaxation time values at the annulus, suggesting local subclinical myocardial remodelling.