CMR-derived mitral annular relaxation velocity refines characterisation of diastolic functional phenotypes in patients with severe aortic stenosis
J Eckelt, A Schulz, S Gersch, J Gronwald, B Beuthner, M Puls, C Schmidt, K Toischer, A Schuster, T LangeAbstract
Background
Aortic stenosis (AS) is associated with progressive left ventricular (LV) remodeling and diastolic dysfunction (DD). While cardiovascular magnetic resonance (CMR) enables assessment of myocardial structure, function and tissue characteristics, accurate assessment of DD has remained limited. Recently, CMR-derived mitral annular relaxation velocity (E´) has been introduced as a novel imaging parameter of longitudinal LV relaxation dynamics. However, its role in characterising DD in severe AS remains insufficiently studied.
Purpose
To characterise DD in patients with severe AS and preserved LV ejection fraction (EF) using CMR-derived E´ and to explore its relationship with left atrial (LA) function and atrio–ventricular functional coupling.
Methods
Patients with severe AS scheduled for transcatheter aortic valve replacement underwent pre-procedural CMR. CMR-derived E´ was calculated by measuring the distance from LV apex to mitral annulus in end-systole and end-diastole in long-axis views divided by the corresponding time interval, representing a surrogate marker of longitudinal mitral annular relaxation velocity (Figure 1). For exploratory diastolic characterisation, patients were grouped according to median splits of CMR E´ and LA reservoir strain (Es). An atrio-ventricular coupling index (AVCI) integrating E′ and LA Es using Z-Score normalisation was calculated to assess the relative contribution of atrial and ventricular functional components to overall diastolic impairment.
Results
A total of 95 patients were included (median age 80 [75-83] years; 44.3% women). CMR E´ correlated with markers of diastolic dysfunction, including NT-pro-BNP, LV mass index and LA-Es (all p<0.05), independently of AS subtype. No significant correlations were observed with CMR-based tissue characteristics (late gadolinium enhancement [p=0.09]; extracellular volume fraction [p = 0.74]). Combined assessment of CMR E´ (median 2.88 cm/s) and LA Es (median 15.0%) classified four distinct patterns of atrio–ventricular functional interaction. While three suggest a continuum of progressive LV DD with secondary LA impairment (1: E´↑/LA Es↑; 2: E´↓/LA Es↑; 3: E´↓/ LA Es↓), a fourth group demonstrated disproportionately high LA dysfunction (4: E´↑/LA Es↓) despite preserved CMR E´. Although LA Es was similarly reduced in Groups 3 and 4, Group 4 exhibited significantly lower NT-proBNP, higher LV concentricity index and elevated AVCI compared with Group 3 (all p<0.05).
Conclusion
CMR-derived E´ provides additional functional information on DD in patients with severe AS and preserved LVEF. Integration of CMR E´ with LA strain enables more refined characterisation of atrio–ventricular functional patterns and identifies a subgroup with disproportionate atrial dysfunction relative to longitudinal LV relaxation. This integrative approach may refine evaluation of DD and atrio-ventricular functional interplay in severe AS.Graphik 1Graphik 2