Mitral valve-in-valve, valve-in-ring, and valve-in-MAC: the Good, the Bad, and the Ugly

This editorial refers to ‘Outcomes of transcatheter mitral valve replacement for degenerated bioprostheses, failed annuloplasty rings, and mitral annular calcification’†, by S.-H. Yoon et al., on page 441. In the present issue of the European Heart Journal, Sung-Han et al. report the largest series of mitral valve-in-valve (ViV), valve-in-ring (ViR), and valve-in-MAC (ViMAC) procedures from several centres from different countries, including the USA, Canada, Europe and Brazil.1 The main value of the study is to provide an overall real-world snapshot of which type of patients are treated today with these three methods and to report the results in terms of real clinical use, in the absence of a randomized trial in the field. The first important message is that patients treated worldwide are really high-risk candidates, with an average STS score of 9% (>10% for ViMAC) and a high prevalence of co-pathologies, suggesting that most of the time these three methods are offered to patients without surgical alternatives (Figure 1). The picture shows the three different procedures, all performed through transseptal access and using an Edwards Sapien 3, 29 mm prosthesis. Line 1: valve-in-valve. (A1) Positioning of the prosthesis inside the surgical valve, before the inflation of the balloon (B1). (C1) Post-procedural echocardiography showing no paravalvular leak and patent LVOT. Line 2: valve-in-ring; the surgical ring is a complete semi-rigid device (A2), which is deformed after the inflation of the balloon (B2); final echo shows minimal paravalvilar leak (C2). Line 3: valve-in-MAC; after the positioning of the prostheses within the calcified mitral annulus (A3), the balloon is inflated (B3). Final echo shows the implanted valve protruding in the LVOT, resulting in significant LVOT obstruction (C3). The second important message is that the results of the three methods are very different. Although overall 30-day mortality is relatively high (∼10%), the difference in terms of results are quite impressive, mainly due to the different incidence of peri-procedural complications and residual mitral regurgitation, which are much higher in ViR and ViMAC procedures compared with ViV. The high incidence of valve thrombosis in all three groups suggests that long-term anticoagulation is preferable after the procedure. In the present registry, mitral ViV (‘the Good’) has been associated with excellent results, with a technical success of 94.4% and procedural success of 74%. Thirty-day mortality is 6.2%, which is extremely good if we consider that in-hospital mortality for redo mitral replacement is between 9% and 12.6% and that the prevalence of functional aetiology of the initial mitral regurgitation (MR) in the present study is high.2–4 The incidence of post-procedural adverse events was relatively low, with only 2.2% of significant left ventricular outflow tract (LVOT) obstruction, <1% conversion to surgery, and 3.3% significant paravalvular leakage (PVL) at 30-day echo. The results of this study clearly show that mitral ViV is an attractive alternative method to treat a degenerated mitral bioprostheses. No differences have been observed according to the type of access used (trasapical vs. transseptal), suggesting that the transseptal route is feasible and should always be considered. In the absence of a randomized trial comparing redo surgery and ViV, in our opinion ViV should be considered the preferred method over redo surgery to treat a degenerated mitral bioprostheses, especially in the presence of intermediate to high surgical risk and functional aetiology, if anatomy is deemed to be favourable. In this regard, careful pre-procedural planning including transoesophageal echocardiogram (TOE) and angio-computed tomography (CT) is mandatory. If the results of ViV can be defined as excellent, in contrast the same cannot be said of ViR and ViMAC (‘the Bad and the Ugly’). Regarding ViR, incidence of procedural success is 57.4%, with an incidence of LVOT obstruction of 5% and of significant PVL of 18.4% immediately after the procedure and 12.6% at 30 days (the main predictor of mortality at follow-up). Thirty-day mortality was 9.9%, and 1-year mortality was 30.6%. This can be only be in part related to the lower ejection fraction of the ViR patients and to the higher incidence of associated ischaemic disease, while it is rather due to the high incidence of peri-procedural complications. The ‘suboptimal’ results observed in ViR indicate the several challenges of this procedure, that often is wrongly considered technically easy to perform. The high incidence of a need for a second valve implantation (∼12%) is a consequence of this and explains in part, together with the high incidence of residual MR, the high observed mortality. One limit of the study is that the type of annuloplasty device is not carefully reported. Similarly to aortic valve in valve procedures, where it is recognized that there are surgical bioprostheses which are associated with better results compared with others, the same differentiation has to be done for mitral ViR procedures, where some prosthetic rings are favourable and some others are not.5 If the type of annulus is not taken into consideration, the misconception that ViR is always associated with bad results can be generated. Different rings have different properties, and ‘good’ rings are associated with better results than ‘bad’ rings. In particular, a surgical ring which is ideal for the mitral ViR procedure should have the following features: ability to adapt to a circular shape, provide a good anchoring, radio opacity, and a proper size in the range of the currently available transcatheter aortic valve implantation (TAVI) devices. In general, complete radiopaque semi-rigid rings are ideal for ViR procedures. The use of the mitral ViV and ViR App is extremely useful, and provides all the information.6 Similarly to ViV, pre-procedural planning with multimodality imaging is the key for patient selection. Another important consideration is that, in contrast to ViV procedures where redo surgery is the only possible alternative, in most of the cases of failing ring annuloplasty transcatheter mitral repair with percutaneous edge-to-edge is a doable option.7 Although MitraClip-in-ring is a challenging procedure (mainly due to the echo shadowing due to the presence of the surgical ring that makes the proper visualization of the leaflets difficult during the grasping and due to the risk of mitral stenosis in the case of small rings), this has always to be considered, especially in the presence of rings that are suboptimal for a ViR procedure. Valve-in-MAC represents an even more challenging procedure. The results observed in the present study are largely disappointing, reflecting the high-risk population affected by mitral annular calcification (MAC) and the technical challenges of this procedure.8 Most of the time, patients with extensive MAC present with prohibitive surgical risk, and transcatheter repair is feasible only in a minority of the cases, leaving these patients almost without therapeutic alternatives. Reported procedural success is 41.4% (significantly reduced compared with ViV and ViR), with a conversion to surgery of 8.6% and a 30-day mortality of 34.5%. Importantly, the incidence of LVOT obstruction was 39.7%, valve embolization was 6.9%, and residual significant MR was ∼13%. Although a precise definition of post-procedural LVOT obstruction is not reported in the manuscript and a formal screening to assess the risk of this complication has not been performed routinely, the number of obstructions observed is impressively high and suggests that this is the main open issue of the ViMAC procedure (together with risk of valve embolization). Even more than in ViV and ViR, careful patient selection based on angio-CT is fundamental to ensure precise sizing, distribution of MAC, and risk of LVOT obstruction. Measuring the length of the anterior mitral leaflet, the mitro-aortic angle, the thickness of the interventricular septum, and the dimension of the LVOT and of the LV cavity allows prediction of the dimension of neo-LVOT in a quite accurate way with dedicated CT software.9 However, a safe cut-off point to exclude the risk of severe LVOT obstruction has never be validated; therefore, the risk of this complication cannot be completely ruled out, also due to the dynamic components playing a role in the pathophysiology of the obstruction (heart rate, preload, afterload, and medications). One-year mortality of ViMAC patient exceeds 60%, also related to the residual LVOT gradient observed in many patients, suggesting that pre-procedural identification of patients at high risk for LVOT obstruction is mandatory. The learning curve effect observed in this and other reports most probably reflects an improvement in patient selection.8 , 10 Preventing the risk of LVOT obstruction is the key to improve outcomes. Russel et al. recently reported a small series (eight patients) of ViMAC performed through direct open atrial access with excellent outcome (100% 30-day survival and 100% procedural success).11 Although the number of patients is really small, the results observed reflect the fact that with a direct surgical view, the operator is able to cut or remove the anterior mitral leaflet, in order to minimize the risk of LVOT obstruction. The feasibility of the replication of the same concept with a transcatheter method has been shown. The LAMPOON technique has been developed to split the anterior mitral valve leaflet percutaneously and prevent iatrogenic LVOT obstruction immediately before transcatheter mitral valve replacement.12 Regular adoption of the LAMPOON method is expected to improve the outcomes of ViMAC, ViR native transcatheter mitral valve replacement (TMVR), and high-risk cases of mitral ViV. The international NHLBI DIR LAMPOON Study (Intentional Laceration of the Anterior Mitral Leaflet to Prevent Left Ventricular Outflow Tract Obstruction During Transcatheter Mitral Valve Implantation, NCT03015194) is currently enrolling patients to investigate the feasibility, safety, and efficacy of the LAMPOON technique on a large scale. Since at the moment most of the procedures are performed using TAVI devices, another factor that could potentially improve the outcome of patients undergoing ViMAC and ViR is the use of dedicated mitral devices, possibly retrievable and repositionable. Having a retrievable device would allow the operators to remove the prostheses after the implantation in the case of significant unexpected LVOT obstruction and eventually to reposition it or to perform a LAMPOON technique. In this regard, the Tendyne prostheses (Abbott Vascular, USA) is a transapical fully retrievable and repositionable mitral valve prostheses. The feasibility of TMVR with Tendyne in MAC has been reported,13 and a multicentre feasibility study is currently ongoing in the USA (NCT03539458). In conclusion, mitral ViV, ViR, and ViMAC are feasible and the number of high-risk patients treated is increasing. ViV has excellent outcomes and has the potential to become the first-line therapy to treat degenerated mitral bioprostheses. ViV and ViMAC share several challenges that have to be overcome in order to improve the outcomes, including better procedural planning, improved techniques, and possibly better devices. Careful patient selection to identify subjects at risk and a high volume of patients treated in order to develop knowledge in the field are the fundamental elements to ensure reasonable results. Conflict of interest: none declared.