Rodrigo Bagur; Angela McInerney

doi:10.4244/EIJ-E-24-00002

Transcatheter mitral valve (MV) implantation (TMVI) for the treatment of severe mitral regurgitation (MR) has attracted increased interest in recent years. Several dedicated devices are being developed and trialled despite facing technical challenges due to the anatomical complexity of the MV and its proximity to other cardiac structures. Screening failure has been reported to be close to 60% in a study assessing 11 different TMVI devices1. Furthermore, the appropriate use of TMVI is yet to be established, and many available devices have a paucity of data to support their use (Supplementary Table 1).

The transapical, tetherâbased Tendyne Mitral Valve System (Abbott) has emerged as a forerunner in the TMVI field. The 30-day, 1- and 2-year outcomes of 100 high surgical risk patients have been reported in a single-arm global feasibility study, showing a technical success rate (<1+ MR at 30 days) of 96% with sustained results at 1 and 2 years, while also reducing hospitalisations for heart failure23. These results in a high-risk population prompted the approval of the Tendyne device for commercial use in Europe (European conformity [CE] mark, 2020) for individuals with clinically significant MR and who were deemed ineligible for surgical mitral valve replacement (SMVR). This approval led to the establishment of the TENDyne European expeRience registry (TENDER)4, whose data for 30-day outcomes align with the initial global study23.

In this edition of EuroIntervention, Ziegelmueller and colleagues5 aim to compare the 30-day outcomes of elderly patients at intermediate surgical risk undergoing either TMVI with Tendyne or SMVR. Among 1,278 SMVR performed between 2000 and 2022, a total of 454 SMVR patients were eligible for a 2:1 propensity score-matched (PSM) study of 80 SMVR and 40 Tendyne individuals. The median age of the population was 78 years, and 60% were female, with median Society of Thoracic Surgeons Predicted Risk of Mortality (STS-PROM) scores of 4.05 and 5.20 for SMVR and Tendyne, respectively (p=0.08). Technical success, where MR was essentially eliminated, was achieved in 97.5% of patients and was similar in both groups. However, device success (82.5% vs 57.5%; p=0.04), and procedural success (75.0% vs 52.5%; p=0.07) “favoured” TMVI compared to SMVR. No differences in 30-day mortality, major bleeding, stroke, or requirement for dialysis were noted.

The authors should be commended for attempting this complex and courageous comparison; however, further data are needed to truly assess the role of TMVI versus SMVR rather than Tendyne on its own. Several observations must be highlighted to help interpret the current study. First, STS-PROM score calculations were partially conducted after matching and for the matched cohorts only. Second, the populations under comparison were heterogeneous, even after matching. Among others, there were differences regarding the aetiology of MR (primary MR: 69.0% and 45.0%, secondary MR: 2.5% and 17.5%, mixed aetiology: 22.5% and 27.5%; p=0.01), the prevalence of coronary artery disease (32.5% and 67.5%; p<0.001) and previous PCI (25.0% and 47.5%; p=0.023) for SMVR and Tendyne, respectively. Furthermore, 30% of SMVR patients underwent concomitant tricuspid valve repair, and two Tendyne patients received concomitant transcatheter aortic valve implantation (TAVI). These differences may be expected and are indeed difficult to match given that the current paradigm would suggest TMVI use only in those individuals who are unsuitable for SMVR or even transcatheter edge-to-edge repair (TEER). Moreover, while the TMVI patients represent a contemporary population, to obtain the actual PSM populations, the SMVR patients are from a 22-year period in the large German Heart Center Munich registry experience; this long period, during which surgical techniques and postoperative care certainly progressed, again potentially affected outcomes.

The main results of this study are not unexpected when comparing SMVR to minimally invasive catheter-based procedures, given the nature of the TMVI prosthesis. Greater device success in the TMVI arm was driven mainly because SMVR patients experienced a greater proportion of unplanned surgical re-intervention. One may wonder if this need for re-intervention was secondary to bleeding, which is often encountered in surgical patients and particularly on long pump runs versus the very unlikely scenario of SMVR “device-related” issues. In addition, higher residual transvalvular gradients were observed in the SMVR group; again, comparing valve haemodynamics between stented and stentless bioprotheses is like comparing apples to oranges, even more so considering the slender self-expanding engineering of the Tendyne TMVI compared with standard SMVR bioprostheses. Along these lines, greater procedural success in the TMVI group was also predominantly driven by a larger number of periprocedural events occurring in the SMVR group, such as postoperative hypotension requiring vasopressors, prolonged intubation, etc. It should be noted that the authors reported outcomes using the Mitral Valve Academic Research Consortium (MVARC) criteria6, which may have been the only sensible option considering TMVI as a comparator. Nonetheless, as stated by the authors, MVARC is meant to standardise definitions for outcome reporting of TMVI/TEER studies rather than for SMVR; hence, its applicability remains grossly uncertain, and perhaps this paper has correctly exposed this important matter.

The results of this study are therefore relevant and interesting in many aspects; however, they are insufficient to challenge current clinical practice. Indeed, comparing TMVI and SMVR may perhaps be a little too premature, particularly in an intermediate surgical risk group. Many unanswered questions remain, particularly regarding valve durability with TMVI devices − a question that has not yet been answered by the current literature which only extends to 2 years of reporting on the trajectory of MR and gradients in 44 and 32 patients, respectively3.

Furthermore, a sizeable rate of major adverse events was reported in the initial studies, including bleeding complications and device-specific events, such as valve thrombosis, endocarditis, migration or dislodgement, paravalvular leaks leading to haemolysis, and valve reintervention, most of which occurred within the first months after TMVI23478910. In fact, valve thrombosis was a major concern and has led to some of the TMVI devices being put on hold9. Paravalvular leaks during follow-up are also a major concern and relate to the mechanism of fixation and anchoring of the Tendyne. Therefore, one would expect that upon left ventricular remodelling or apical fatigue, leakage and valve dislodgement may become clinically meaningful during follow-up, thereby making it necessary to retension the tether or upsize the apical pad of the Tendyne device10.

In summary, the trajectory of catheter-based MV technologies is promising; however, a head-to-head comparison with SMVR in a randomised controlled trial setting would be desirable. Nonetheless, based upon the difficulties mentioned above in the screening of these patients, mainly because of anatomical suitability and multiple devices with different intrinsic technical features, such a trial may be difficult to undertake. Paraphrasing Morpheus in The Matrix, “There is a difference between knowing the path and walking the path.” The present study adds to the growing evidence on the use of TMVI with Tendyne but also puts the “blue pill” into context – we do not want the story to end, since there are a substantial proportion of individuals who are ineligible for TMVI, so SMVR should remain the preferred option in this instance. Rather, we face the reality of a complex subject matter exposed by the “red pill” and our willingness to change. A long walk on the path of catheter-based treatment of MV disease has started with the “red pill”, and only time will tell if this is the right path, though this must be through critical thinking and science.