Causality analysis of the determinants of ventricular functional mitral regurgitation

Abstract Background The severity of ventricular functional mitral regurgitation (VFMR) is influenced by multiple interrelated factors (Figure 1A), making it challenging to disentangle cause-and-effect relationships in its pathogenesis. To address this complexity, our study aimed to identify the key determinants of VFMR by analyzing beat-to-beat variability (Figure 1B) and applying Granger causality analysis. Granger causality is a statistical method that evaluates whether one time-series variable can predict future changes in another, thereby suggesting a directional causal influence (Figure 1C). Purpose To determine the primary physiological contributors to VFMR severity and variability by using high-resolution beat-to-beat data and Granger causality analysis, thereby distinguishing causal factors from mere associations. Methods We included 20 patients (mean age 77 ± 9 years) undergoing echocardiographic evaluation for VFMR, including 8 with atrial fibrillation (AF). In each patient, 10 consecutive cardiac cycles were analyzed, yielding a total of 685 time-stamped data points (Figure 1D). The severity of VFMR was assessed by measuring the mitral regurgitant color Doppler flow area (MRa), along with its potential determinants, including left ventricular (LV) volume, left atrial (LA) volume, mitral annulus (MA) area, mitral valve (MV) tenting height, and papillary muscle length. Results At baseline, MRa correlated significantly with LV volume and LA volume (r = 0.57, p = 0.001; r = 0.51, p = 0.007, respectively). Beat-to-beat variability in MRa was associated with variations in LV volume (Figure 1F), LA volume, mitral valve tenting height, and papillary muscle length, with strong autocorrelation observed as well (Figure 1E). Granger causality analysis (Figure 2) demonstrated significant directional relationships: LV volume → MRa, LA volume → MRa, Papillary muscle length → MRa. A trend toward significance was observed for: Tenting height → MRa. In a multivariate vector autoregressive model, LV volume, LA volume, MV tenting height, and papillary muscle length were independent predictors of MRa across the pooled VFMR cohort. Conclusions Left ventricular volume, left atrial volume, and papillary muscle length exert a causal influence on mitral regurgitant orifice dynamics in patients with ventricular functional mitral regurgitation. Understanding these directional relationships may support more targeted diagnostic and therapeutic approaches.