The Cardioband annuloplasty system: is it ready for prime time?

This editorial refers to ‘Transcatheter mitral valve repair for functional mitral regurgitation using the Cardioband System: 1 year outcomes’†, by D. Messika-Zeitoun et al., on page 466. Secondary mitral regurgitation (2ry MR) is a challenging problem, affecting patients with both ischaemic and non-ischaemic cardiomyopathies. Several surgical techniques have dealt with this challenge, the most widespread of which is undersized ring.1 A considerable proportion of patients with 2ry MR are at markedly increased surgical risk, due to advanced left ventricular dysfunction and/or associated cardiac and non-cardiac comorbidities. There is growing interest in the development of innovative catheter-based solutions for MR. The solutions are diverse, owing to the complexity of the disease. They include leaflet repair, chordal implant, neo-leaflet implant, indirect annuloplasty, direct annuloplasty, mitral annular anchor, and transcatheter mitral valve replacement. In this issue of the European Heart Journal, Messika-Zeitoun et al. report on 1 year outcomes of the Cardioband system (Edwards Lifesciences, Irvine, CA), an adjustable direct annuloplasty system that received a CE (Conformité Euròpéene) mark in 2015.2 This system is based on the surgical concept of an undersized ring. It is fully percutaneous and adjustable. As an important favourable feature, it allows further interventions at leaflet level, e.g. edge-to-edge clipping or catheter-based chordal implantation, or may serve as a rescue procedure after failure of such procedures.3 The paper by Messika-Zeitoun et al. follows two previous reports by the same group, reporting on early and 6 month outcomes.4 , 5 This was a single-arm study, conducted in patients with moderate or severe 2ry MR. Results were assessed by the mitral valve academic research consortium (MVARC) criteria.6 Based on these criteria, the technical, device, and procedural successes were 97, 72 and 68%, respectively. The reported rate of residual MR at 1 year is subject to the inclusion or exclusion of patients who required mitral re-intervention, died, withdrew their consent, or were lost to follow-up. In the best-case scenario, MR at one year was no more than moderate in 37/39 = 95%. If the more stringent MVARC definition of ‘optimal’ (rather than ‘acceptable’) residual MR was considered as success, mild MR at 1 year was 27/39 = 69%. With one-third of patients having more than mild MR 1 year after device implantation, these results are inferior to those achievable by a surgical undersized ring.7 These rather optimistic results are obviously misleading, since they exclude 19 early survivors, of whom 6 required re-intervention, 5 died later, and 8 were lost to follow-up or withdrew their consent. In the worst-case scenario, the rate of no more than moderate MR was 61%, and even less than that if moderate residual MR was not considered a favourable result. Importantly, MR worsened in 22% cases. There were no major safety issues. On average, patients showed clinical benefit in a variety of endpoints (New-York Heart Association (NYHA) class, Minnesota Living with Heart Failure Questionnaire (MLHFQ), and 6 min walk test). Some of the pros and cons of the Cardioband system are depicted in Take home figure. Although MR grade has improved overall, the Cardioband system is far from being perfect. Here are some plausible explanations. The first issue is the whole concept of a band. Cardioband is not a full ring. There are some data to show that a full ring might be better than a band in the surgical field. Kwon et al. showed that recurrent MR occurred less frequently in the complete vs. partial ring group (10 vs. 21%), despite lower pre-operative ejection fractions in the complete ring group.8 Some novel transcatheter innovations using a full ring are under way, e.g. Amend Valcare (Valcare Medical; Israel) and The Millipede IRIS Transcatheter Annuloplasty Ring (Millipede Inc; Santa Rosa, CA), and reports on their performance in human studies for both the immediate and longer term are awaited. Second, anchor detachment was quite prevalent in the early period (nine such cases), although the early reports of the same group somehow failed to address that issue.4 , 5 The third issue is the lack of standardization of the very point at which the system is released. The operators of Cardioband have a unique opportunity to optimize their results according to intraprocedural transoesophageal (TOE) findings on beating hearts. An important determinant of durable surgical annuloplasty is the coaptation length (CL), which was not tested in the current study. When Braun et al. implanted undersized annuloplasty rings for 2ry MR, they aimed to achieve at least 8 mm CL at the A2-P2 level (in addition to leaving no residual MR), with further downsizing if these criteria were not met.7 Such a strategy resulted in durable MR reduction, as 85% of survivors had less than 2+ MR grade at an average follow-up of 4.3 years. This 8 mm cut-off value for CL became quite standard in the surgical arena. Some studies used the coaptation length index (CLI), which is the CL divided by mitral antero-posterior diameter.9 Tozzi et al. tested an adjustable, percutaneously implanted mitral ring in a sheep model and showed that achieving a satisfactory CLI at the initial procedure was associated with a sustainable result over 6 months.10 A policy of active adjustment of annular size to achieve a satisfactory coaptation length seems quite obligatory in percutaneous annular reduction, and should be exercised in future studies. Selected pros and cons of the Cardioband system. What are the ideal haemodynamics in which the mitral repair system should be tested? These patients undergo the procedure under general anaesthesia, otherwise they would not be able to tolerate the TOE probe for a procedure that lasts, on average, more than 3 h. The degree of 2ry MR is reduced by general anaesthesia.11 In the operating room, it is advisable to assess MR under conditions that mimic an awake state. Shiran et al. re-achieved pre-operative MR degree by fluid loading (if pulmonary wedge pressure was < 15 mmHg) or by phenylephrine.12 Re-checking MR degree in the catheterization laboratory in haemodynamic conditions that resemble patients’ awake state might lead to further annular reduction, which, in turn, may provide improved durability of MR reduction. Patient selection issues merit some discussion. First, although MR degree was considered severe in 73% of patients, the average effective regurgitant orifice and regurgitant volume values (available for 34 patients) were quite modest (26 mm square and 37 cc/beat, respectively). This implies that MR degree was not extreme. Of note, about a quarter of patients had moderate MR, for which surgical intervention is not recommended by the current 2017 European Society of Cardiology/European Association for Cardio-Thoracic Surgery Guidelines for the management of valvular heart disease13. Second, patients were eligible if left-ventricular end-diastolic diameter (LVEDD) was ≤ 70 mm. This cut-off value might be too high, since there are some data on poor results in patients with LVEDD of > 65 mm undergoing surgical implantation of an undersized ring.7 Even if found effective, the Cardioband is not suitable for a considerable proportion of patients with 2ry MR. Significant mitral annulus calcification is a contraindication, and this is not rare in patients with coronary disease or in elderly patients.14 Extreme or highly asymmetric stenting with significant restriction of posterior leaflet mobility is also a contraindication. In both the above-mentioned conditions, therapies targeting the mitral leaflet (e.g. edge-to-edge solutions) might be more appropriate. In conclusion, these 1 year results of the Cardioband multicentre study, based on several dozens of patients, show MR reduction that is acceptable by the MVARC criteria, without serious safety concerns, disregarding the bothersome issue of anchor detachment in the early stages the study. These results may be further optimized by aiming at longer coaptation length at haemodynamic conditions mimicking an awake state. The procedure is cumbersome and lengthy, a feature that might discourage invasive cardiologists from its widespread adoption. It has a considerable way to go before being ready for prime time. Comparison with optimal medical therapy, surgical mitral repair, and other transcatheter modalities is mandatory. Lastly, the Cardioband may have a complementary role along with other catheter-based therapies, and particularly with leaflet-based therapy. Conflict of interest: none declared.