Chronic secondary mitral regurgitation (SMR) is a frequent finding in the setting of heart failure with reduced ejection fraction (HFrEF), and is associated with adverse prognosis. Since SMR is only one component of the left heart disease, mitral valve treatment may not be by itself curative, and the best therapy for chronic SMR is unclear. The benefit of mitral surgery in this setting has not been clearly demonstrated, and the majority of these patients have multiple complex comorbidities and are denied surgery in clinical practice [1, 2].

Edge-to-edge transcatheter mitral valve repair (TMVR) has been shown to be an effective and safe therapeutic option in patients with primary MR at high surgical risk [3]. In SMR, results from real-life registries and three retrospective studies have shown promising initial results, including a potential mortality benefit over isolated optimal medical treatment (OMT) [4‐8]. Recently, two prospective randomized trials assessed the benefit of TMVR on top of OMT in patients with severe SMR: the French Percutaneous Repair with the MitraClip Device for Severe Functional/Secondary Mitral Regurgitation (MITRA‐FR) trial, which showed no benefit from TMVR on top of OMT [9], and the American Cardiovascular Outcomes Assessment of the MitraClip Percutaneous Therapy for Heart Failure Patients with Functional Mitral Regurgitation (COAPT) trial, which demonstrated significant reductions in mortality and hospitalisations for heart failure as well as improvements in quality of life [10, 10]. One of the potential explanations that were discussed for the observed discrepancies in outcomes is the difference in effective regurgitant orifice (ERO) in each study: mean baseline ERO was 0.31 cm² in MITRA‐FR, vs. 0.41 cm² in COAPT cm², which led to a questioning of the yield of TMVR in the setting of low ERO (mainly < 0.3 cm²), even though guidelines advocate for an integrative approach to assess the severity of MR.

Accordingly, the aim of this study was two-fold: (1) to compare the characteristics and outcomes of SMR patients undergoing edge-to-edge TMVR in the retrospective EuroSMR (European Registry of Transcatheter Repair for Secondary Mitral Regurgitation) registry, according to the ERO threshold, and (2) to assess the yield of ERO in selecting candidates for TMVR, and eventually identify an optimal ERO threshold associated with mortality in SMR patients undergoing TMVR.

A total of 1237 patients with severe SMR underwent TMVR in the participating centres and were included in EuroSMR. In 1062 patients, ERO quantification before TMVR was available, which defined the study population for the current analysis. Baseline characteristics of the global population are presented in Table 1. Median age of the patients was 76.0 (IQR: 69.0–80.1) years, with 387 women (36.5%). Overall, 943 patients (89.4%) were in NYHA class III or IV at the time of the procedure. They presented with atrial fibrillation in 686 patients (64.7%), a prior stroke in 98 patients (9.2%), and a chronic pulmonary disease in 174 patients (16.4%). Regarding surgical risk, median EuroSCORE II was 6.7% (IQR: 3.9–11.9%). The aetiology of SMR was ischemic in 515 patients (51.6%).

Baseline echocardiographic parameters are described in Table 2. Median ERO was 0.28 cm² (IQR: 0.20–0.39). Median LVEF was 34.0% (IQR: 25.0–44.0%) with a median LV end-diastolic volume of 170 ml (IQR: 123–225 ml) and end-systolic volume of 110 ml (IQR: 70.0–160.0 ml). The median vena contracta diameter was 6.8 mm (IQR: 4.8–8.0 mm).

There has been a long debate on the impact of TMVR on outcomes in patients with chronic SMR and HFrEF. While four studies, three retrospective reports and the prospective randomized COAPT trial, demonstrated a survival benefit with TMVR [6, 7, 8, 10], the prospective randomized MITRA-FR study did not [9]. These contradictory outcomes have raised a host of new questions:

The EuroSMR registry was initiated to provide further insights and answers for the raised questions. Eight European high volume centres with a long experience in the treatment of SMR by TMVR included their patients, who were treated in clinical routine, into this retrospective real-world registry. Of note, a multidisciplinary heart team assessed each individual patient before the procedure to establish the severity of MR using a multiparametric echocardiographic approach, and the appropriateness of TMVR. When compared to the COAPT-TMVR patients group, EuroSMR patients with an ERO ≥ 0.3 cm² had several similarities in their baseline characteristics, including ERO (0.4 cm² vs. 0.41 cm², even though not all patients in COAPT did have an ERO > 0.3 cm²), LVEDV (174 ml vs. 194 ml), and LV-EF (33.2% vs. 31.3%) for EuroSMR vs. COAPT-TMVR, respectively. Furthermore, the two-year all-cause mortality rate was comparable with 30.0% vs. 29.1% for EuroSMR vs. COAPT-TMVR, respectively [10]. Due to the lack of a control group in EuroSMR, the true benefit of TMVR in this particular European patient population remains unclear, but the similarities with the COAPT-TMVR group might indicate a comparable outcome.

The surprising finding of EuroSMR is the comparable all-cause mortality rate in patients treated with TMVR and with an ERO < 0.3 cm², without any survival difference at 1 year and 2 years. Besides, spline curves showed a very weak relation between ERO and mortality, supporting the notion that ERO may not be an appropriate tool to predict outcome after TMVR, and that differences in ERO cannot solely explain the differences in outcomes between COAPT and MITRA-FR. Accordingly, a more integrative approach, assessing among others MR severity with additional qualitative and quantitative parameters, right heart function and clinical parameters, is needed to predict which patients with HFrEF will actually benefit from TMVR, or more importantly, in which patients TMVR is ineffective.

Quantification of SMR severity is challenging, and the interpretation of the MR severity assessment should be made with caution. LV volume and stroke volume, as determined by two-dimensional echocardiography with the use of Simpson’s rule, are often underestimated, especially in patients with dilated ventricles [15, 16]. Regarding ERO, the cut-off of > 0.2 cm² has traditionally been used in Europe to determine severe MR since data suggested that, adverse outcomes are associated with a smaller ERO in SMR than in primary MR. This observation is explained by a smaller regurgitant volume, which may still represent a relevant regurgitant fraction in the presence of compromised LV systolic function (and low total stroke volume) adding to the effects of elevated ventricular filling pressures. However, ERO calculated by PISA is affected by valve morphology and colour flow jet characteristics [17]. PISA measurement requires accurate visualisation of the aliasing line of the isovelocity surface and the regurgitant orifice, and suboptimal images can alter the reliability of the measurement. Besides, in SMR, Doppler methods for ERO area calculations by the flow convergence method are less reproducible and may underestimate severity because of the crescent shape of the regurgitant orifice. Finally, the reliability of ERO measurement in the setting of SMR is questionable given the variability of MR throughout systole, typically in a biphasic pattern, leading to a variation in measured ERO according to the timing of assessment, with an early systolic peak, followed by a rapid decrease to a midsystolic lowest value, then a rapid increase toward a late systolic peak [18]. The current EuroSMR findings demonstrate that selected patients with ERO above but also below 0.3 cm² benefit from TMVR, and have similar mortalities, and that it might not be reasonable to restrict TMVR to patients with ERO > 0.3 cm² [1, 2, 19].

Regarding the symptomatic improvement, MITRA-FR and COAPT showed a benefit of TMVR in terms of NYHA functional class. In COAPT, 72.2% of TMVR treated patients were in NYHA functional class I or II at 1-year follow-up, which compares well with the observed 68% rate of EuroSMR patients with an ERO ≥ 0.3 cm². As for patients with ERO ≤ 0.3 cm², the COAPT investigators recently presented the improvements in the physical exercise capacity and quality of life in patients with an ERO ≤ 0.3 cm² and large left ventricles vs. the remainder of patients (either ERO > 0.3 cm² or large left ventricles) [20]. In this analysis, both patient groups exhibited very similar improvements for both outcome parameters, indicating that also patients with smaller ERO experience a significant symptomatic benefit. This observation is also paralleled in the current EuroSMR registry with 61.7% of patients with an ERO < 0.3 cm² demonstrating an improvement of at least one NYHA functional class at follow-up. Given this recent COAPT subgroup analysis, as well as the EuroSMR results, which both demonstrate that a large proportion of patients with severe SMR and an ERO < 0.3 cm² will obtain a clinically relevant improvement from TMVR, TMVR should not be withheld from patients with an ERO < 0.3 cm². However, these results should not lead to excessive TMVR in patients with ERO < 0.3cm². Instead, experienced multidisciplinary heart teams have to assess TMVR appropriateness using an integrative approach of clinical aspects including the patient history of heart failure and current presentation, as well as qualitative, semi-quantitative, and quantitative echocardiographic parameters for the assessment of SMR severity [21].

Several limitations have to be acknowledged. This is an observational study without a control group, and there was no central adjudication of clinical status and echocardiographic parameters. Severity of MR was assessed using integrative approach including several echocardiographic parameters. Some of those variables, such as E wave velocity, pulmonary vein flow reversal, vena contracta area by 3D imaging or another measure of non-circular orifice were not consistently available in the database due to the absence of corelab and the retrospective multicentric nature of the registry that prevented the extensive collection of all echocardiographic parameters. All patients included were considered as under OMT, based on the local heart teams’ assessment; however, there was no clinical core lab to monitor this information, and the subsequent therapeutic adjustments were not documented. Several patients were lost to follow-up, as it is often the case in multicentric retrospective registries, and we do not have the specific cause of mortality, in particular, we cannot distinguish whether mortality was due to a cardiovascular cause. Finally, a selection bias might explain the observed differences in NYHA class before TMVR, e.g., the increased severity of the symptoms might explain why patients with lower ERO underwent TMVR despite a less severe MR according to ERO. However, as discussed above, it appears important to realize that the symptomatic improvement is comparable in patients with an ERO < or ≥ 0.3 cm².

The large EuroSMR registry reveals that ERO is not an appropriate tool to select SMR patients likely to benefit from TMVR. Besides, all-cause mortality after edge-to-edge TMVR does not differ between patients with a baseline ERO < or ≥ 0.3 cm², as long as the severity of MR was established using an integrative approach, and both groups exhibited a relevant clinical improvement, so that TMVR should not be withheld from such patients who remain symptomatic despite OMT, if the appropriateness for TMVR has been determined by an experienced heart team in dedicated heart valve centres.