Causality analysis of the determinants of atrial functional mitral regurgitation

Abstract Background Atrial functional mitral regurgitation (AFMR) results from a complex interaction of atrial, ventricular, and valvular factors, which often overlap. However, the predominant mechanism and the dynamic relationships among these contributing pathways are not yet fully understood. Clarifying the relative impact of each mechanism is crucial for the development of effective, mechanism-specific treatments. Purpose The purpose of this study was to explore the directional influence of established etiological factors on AFMR using Granger causality analysis, a time series-based method that assesses potential causal relationships. Methods We analysed 20 patients diagnosed with AFMR, all presenting with atrial fibrillation or flutter and preserved left ventricular ejection fraction. The cohort had a mean age of 78.7 ± 7.2 years, with 70% (14 patients) being female. Each patient underwent transthoracic echocardiography, from which up to ten consecutive cardiac cycles in apical 4 chambers view were recorded (ranging from 25 to 92 frames per patient), yielding a total of 637 time points. For each frame, we measured mitral regurgitant jet area (MR area), left atrial and left ventricular volumes (LA and LV volumes), mitral annulus diameter, and tenting height to capture AFMR variability in relation to its potential determinants. Multivariate autoregressive modeling and Granger causality testing were employed to assess the directionality and significance of these relationships. Results All patients exhibited relatively enlarged LA volumes (median 57.8 [45.8–81.3] mL/m²), borderline MA diameters (median 34 [32.5–37] mm), and small LV volumes (median 53.0 [45.0–59.0] mL/m²). Granger causality analysis identified LA volume as the only variable with a significant directional influence on MR area (χ² = 40.643, p < 0.001). Additional significant causal relationships (ordered by effect size) included: MR area → LA volume (χ² = 45.258, p < 0.001), MR area → LV volume (χ² = 18.3, p < 0.001), LV volume → LA volume (χ² = 16.47, p < 0.001), MR area → tenting height (χ² = 14.39, p = 0.001), LV volume → tenting height (χ² = 10.771, p = 0.005), LA volume → tenting height (χ² = 8.19, p = 0.017), and MA diameter → LV volume (χ² = 7.937, p = 0.019) (Figure 1). Conclusions Left atrial remodeling appears to be the primary driver of AFMR, while other established determinants interact in complex ways that shape the broader pathophysiological landscape of the condition. These findings highlight the importance of targeting atrial remodeling in therapeutic strategies and suggest that treatment decisions in AFMR should be guided by the dominant underlying mechanism.Figure 1.AFMR Granger causality