The impact of transcatheter edge‐to‐edge repair on right ventricle‐pulmonary artery coupling in patients with functional mitral regurgitation

Transcatheter edge-to-edge repair (TEER) with the MitraClip system (Abbott; Menlo Park, California, USA) represents a therapeutic option for functional mitral regurgitation (FMR) by reducing symptoms and improving clinical outcomes in selected patients.1, 2 In chronic MR, long-standing volume overload in the left-sided chambers leads to the development of pulmonary hypertension (PH).3 The right ventricle (RV) initially responds to the increased afterload with remodelling and enlargement, but eventually right ventricular dysfunction (RVD) occurs, which is recognized as an independent predictor of mortality in patients with heart failure (HF) and after mitral valve surgery.4-7 A negative prognostic role of RVD evaluated by tricuspid annular plane systolic excursion (TAPSE) at echocardiography has been suggested in patients undergoing TEER as well.8 The non-invasive surrogate of RV to pulmonary arterial (RV-PA) coupling TAPSE/pulmonary artery systolic pressure (PASP) ratio has demonstrated to improve prognostic stratification in patients with HF,9 being also able to detect impending RVD.10 Recent data suggested that TEER can be followed by RV reverse remodelling in patients with severe MR.11, 12 This study aimed to assess the short-term impact of TEER with MitraClip on TAPSE/PASP ratio in patients affected by heart failure with reduced ejection fraction (HFrEF) and FMR. Patients affected by HFrEF and moderate-to-severe or severe (3+ or 4+/4+) FMR undergoing TEER with MitraClip at two Italian centres (San Matteo Hospital in Pavia and Niguarda Hospital in Milan) between January 2013 and December 2020 were retrospectively included in the study. The investigation conforms to the principles outlined in the Declaration of Helsinki and was approved by the Local Ethics Committee. FMR was defined as MR caused by non-complete leaflet coaptation due to LV remodelling. Patients were considered for TEER by the local Heart Team if suffering from persisting HF symptoms despite optimal medical therapy. MitraClip procedure was performed as previously described13; procedural success was defined as residual MR ≤2, in absence of failure (abortion of the procedure and conversion to open surgery) at the end of the procedure.14 Patients underwent trans-thoracic echocardiography at baseline and at discharge. TAPSE and TAPSE/PASP ratio were collected as indexes of RV function and RV-PA coupling, respectively. TAPSE was measured as the peak excursion of the tricuspid annulus from end-diastole to end-systole in the apical 4-chamber view with the M-mode.15 PASP was calculated by adding the right atrial pressure (estimated by the diameter and collapsibility of the inferior vena cava) to the systolic pressure gradient of the tricuspid regurgitation.16 For patients in atrial fibrillation, the measurements were repeated at least three times and the average value was calculated. ΔTAPSE, ΔPASP and Δ(TAPSE/PASP) were calculated by subtracting post-procedural values from pre-procedural values. We stratified our cohort according to the distribution of the Δ(TAPSE/PASP) parameter, setting the median value to define patients as ‘responders’ and ‘non-responders’, respectively. Categorical variables were expressed as count (percentage) and compared with Fisher's exact test. Continuous variables were shown as mean ± standard deviation or median (interquartile range [IQR]). Normality distribution was assessed by visual inspection of the quantile-quantile plot. Student's paired t test was used for within-group comparison. For non-normally distributed variables, Mann–Whitney and Wilcoxon tests were used for between-groups and within-group comparison, respectively. Longitudinal effects on TAPSE and TAPSE/PASP ratio were analysed by fitting a mixed effect model for repeated measures (time measurement as fixed and individual subject as random-effects). The Holm method was used to adjust for post-hoc comparisons. The correlations between echocardiographic parameters were assessed with a nonparametric Kendall correlation coefficient. The Kaplan–Meier method was used to explore event-free survival. Statistical significance was set at a level of p < .05. All statistical analyses were performed using R software (http://www.R-project.org/version3.4.4). Fifty-two patients were included. The mean age was 63 (±10) years, 17 patients (29%) were affected by ischaemic cardiomyopathy. Baseline clinical and echocardiographic characteristics are shown in Table 1. Mean left ventricular (LV) end-diastolic diameter was 69 (63.2–73.7) mm, mean mitral effective regurgitant orifice area 0.31 (0.24–0.40) mm2. All patients underwent implantation of at least one clip and procedural success was achieved in 49 patients (94%). The mean time between pre-procedural and post-procedural echocardiograms was 7 (±3) days. Compared to baseline, pre-discharge echocardiography showed an improvement in TAPSE/PASP (0.49 ± 0.16 vs 0.42 ± 0.27 mm/mmHg, +0.11 [+0.09;+0.24], p < .001) (Figure 1). Both TAPSE and PASP significantly improved after the procedure (respectively, p = .009 and p < .001) (Table 2). ΔTAPSE was positively correlated with Δ(TAPSE/PASP) (R = 0.24, p = .017) and a negative correlation was found between ΔPASP and Δ(TAPSE/PASP) (R = −0.64, p < .001) (Figure 2). A small but significant reduction in LVEF was reported (27% vs 25%, −1.7 [−0.6;-2.7], p = .006). At 1 -year follow-up, death occurred in only one patient while hospitalization for HF in 11 patients. Median Δ(TAPSE/PASP) was 0.14 mm/mmHg and 22 patients were classified as responders. Differences between responders and non-responders with respect to clinical outcomes are presented in Table 3 and the Kaplan–Meier survival plot for cardiovascular death OR hospitalization for HF in both group is shown in Figure 3. Non-responders showed a borderline significant increase in the composite endpoint compared with responders at 6 months (5 [15.2%] vs 0 [0%], p = .067) and at 12 months (9 [28.1%] vs 2 [8.7%], p = .097) (Table 3). Our study on patients with HFrEF and FMR showed an improvement in TAPSE/PASP ratio after TEER with MitraClip. To the best of our knowledge, this analysis is the first that investigates the modification of TAPSE/PASP in patients with HFrEF and FMR treated with TEER. Previous studies have examined the impact of TEER with MitraClip on RV function, with inconsistent results.8, 12, 17 Sugiura et al. demonstrated an overall early improvement in the RV function (median time of 3 days after the procedure, IQR 2–5) expressed as fractional area change, while there was significant change neither in TAPSE nor in PASP among the groups.17 Conversely, Ledwoch et al. showed a significant increase in TAPSE after TEER (19 ± 5 vs 17 ± 5 mm, p = .006) at a mean follow-up of 4.9 ± 2.5 months.12 Interestingly, in a study including only patients with FMR, Godino et al. found an improvement in TAPSE at 6-months echo-matched analysis in patients with baseline RVD (19 ± 4.5 vs 15 ± 3.0 mm, p = .007).8 However, the majority of these studies used TAPSE to assess RV function, which represents only a rough indicator. Indeed, despite being the most immediate echocardiographic parameter for assessing RV function, TAPSE has several limitations, including not taking into account the RV interdependence from its afterload. Thus, the TAPSE/PASP ratio appears as a promising echocardiographic surrogate of the more precise RV-PA coupling measured during right heart catheterization and has also been shown to improve prognostic stratification in patients with HFrEF.10, 18-20 In this study including only FMR patients, we found that TAPSE/PASP significantly increases at discharge after TEER, compared with pre-procedural values (0.49 ± 0.16 vs 0.42 ± 0.27 mm/mmHg, +0.11 [+0.09;+0.24], p < .001). This finding is in line with previous observations reporting a positive effect on the haemodynamic of the right sections after MitraClip, with a possible RV positive remodelling.11, 21 Indeed, the reduction of MR following TEER has been shown to be associated with a reduction in pulmonary pressures as a consequence of the decreased volume overload in the left atrium and pulmonary vascular bed.11, 21 Since the RV performance is extremely pressure-dependent, it can be assumed that the reduction of pulmonary pressures could be one of the earliest mechanisms of improvement of RV-PA coupling and RV function itself. Interestingly, by analysing the correlation between the variation of TAPSE, PASP and their ratio before and after TEER, we found that the magnitude of correlation between ΔPASP and Δ(TAPSE/PASP) ratio was remarkably greater than those between ΔTAPSE and Δ(TAPSE/PASP) (Figure 2). These findings may confirm the hypothesis of an initial more pronounced influence of PASP on the improvement in RV–PA coupling rather than TAPSE after TEER. Consequently, it could be speculated that improvement in RV-PA coupling after TEER may occur in a stepwise fashion, eventually explaining the previous inconsistent results when assessing RV function using TAPSE, whose improvement may require time. However, further studies are needed to confirm these data, particularly the eventual persistent increase in TAPSE/PASP ratio over time. Recent studies have also suggested a possible prognostic role of TAPSE/PASP at baseline in patients undergoing MitraClip.20, 22 However, no data exists on the prognostic impact of TAPSE/PASP changes after TEER, especially in patients with FMR. Because there is no certainty regarding the best cut-off to use for TAPSE/PASP and Δ(TAPSE/PASP), we used the median Δ(TAPSE/PASP) of our population to distinguish responders from non-responders and evaluate their differences in terms of outcome. The incidence of clinical endpoints, particularly death, was low, considering the type of patients. This could be related to the baseline characteristics of the study population (e.g. not very high age, good renal function). Although the results of this study do not reach full statistical significance, the difference between the two groups in terms of percentage of the composite endpoint, as well as the pathophysiological plausibility, probably encourage testing this hypothesis in larger prospective studies. This study is limited by its observational nature and relatively small sample size. Moreover, TAPSE/PASP ratio is only a surrogate of RV-PA coupling. Indeed, right heart catheterization and cardiac magnetic resonance represent the gold standard for evaluating RV-PA coupling and RV size and function, respectively. However, cardiac magnetic resonance has limited availability and may be contraindicated in patients with HFrEF with pacemaker or implantable cardioverter defibrillator. Right heart catheterization was not regularly repeated before and after TEER in both centres. However, TAPSE/PASP ratio has demonstrated to be an independent predictor of survival in HFrEF patients and in this study more than one hundred repeated measurements were rigorously made, before and after the procedure. In our study, the percentage of patients with ischaemic dilated cardiomyopathy appears to be lower than in previous studies performed in patients with FMR; therefore, our results will need to be confirmed in populations with a greater presence of post-ischaemic patients. Nevertheless, there remains the pathophysiological plausibility of improved RV-PA coupling even in subjects with ischaemic aetiology dilated cardiomyopathy. The sample size and number of endpoints are relatively small to reliably assess a possible prognostic role of Δ(TAPSE/PASP). Despite these limitations, there are clear percentage differences which reached a trend towards statistical significance. In a cohort of patients with HFrEF and FMR, TEER resulted in acute improvement of TAPSE/PASP ratio. Despite this finding needs to be confirmed in larger studies with a longer follow-up, these results further expand the current knowledge on the impact of TEER on RV-PA coupling and RV function. Open access funding provided by BIBLIOSAN. Open access funding provided by BIBLIOSAN. The authors declare that there is no conflict of interest.