Mitral valve regurgitation (MR) is the second most common valve disease requiring intervention in developed countries.1 It can be either a disease of the mitral valve and/or subvalvular apparatus (primary MR) or arise secondary to left ventricular (LV) and/or left atrium (LA) geometry disruption, most frequently with chamber dilation (secondary MR). Primary MR is most often a consequence of degenerative MV disease, with prolapse and/or flail of one or more leaflet segments.1

According to the prevailing guidelines,1,2 surgical indication for primary severe MR is dictated by symptom status, in addition to the consequences of MR on LV and/or LA dimensions, LV systolic dysfunction, or pulmonary artery systolic pressure (PASP).

Mitral valve repair (MVr) is the preferred approach to treat severe MV disease secondary to leaflet flail or prolapse, supported by a long-term survival benefit over MV replacement.3 The surgical community has shifted from resection of abnormal leaflet tissue with precise valve anatomy reconstruction toward a more conservative approach, involving implantation of polytetrafluoroethylene neochordae to support the prolapsing segments’ free edge, with no or limited resection. Also, with patients being referred for surgery earlier, the need for routine adjunct annuloplasty has become debatable. Some authors state that in degenerative MR, early surgical intervention without annuloplasty, may preserve better the three-dimensional dynamic interplay of the ventriculo-annular continuity.4

Acute post-operative LV dysfunction observed in some patients after MV surgery may be directly related to cardiopulmonary bypass (CBP) and cardioplegia, which have known profound consequences in myocardial oxidative stress, free radical injury and inflammation. These observations, together with treated patients being evermore complex and frail, has raised interest in minimally invasive techniques for MVr.

A new minimally invasive MVr technique that allows transapical off-pump neochord implantation is now available and has been demonstrated to be reproducible and safe in short-term follow-up. The procedure is performed using the NeoChord DS1000TM system (NeoChord, Inc., Eden Praire, MN) using transesophageal echocardiography (TEE) guidance. The procedure has proven to be safe, effective and reproducible in selected patients, both in the short5 and in the medium term.6

In this study, we aim to characterize early Portuguese experience with this technique, providing early clinical and echocardiographic results.

Eighteen patients were treated with neochord implantation at our center over the above-mentioned period. The majority (72%) were male, and the mean age was 65±15 years (minimum 45 years, maximum 84 years).

The mean EuroSCORE II was 1.9±1.6. The prevalence of AF was 16.7%. All patients were in NYHA class≥II (N=18). Other relevant demographic and clinical variables are presented in Table 1.

In pre-procedural echocardiographic evaluation, mean effective regurgitant orifice area was 1.0±0.4 cm2, with a mean regurgitant volume 146±42 mL, and a mean leaflet-to-annulus index (LAI) of 1.29±0.14. MR was due to leaflet prolapse in 50% (N=9), and flail leaflet in 50% (N=9). Anatomic type A (isolated P2 defect) was the predominant form in 66.7% (N=12). Successful repair, defined by none, trace or mild mitral regurgitation, by implantation of two to four neochordae, was achieved in all 18 patients. No major complications arose intra-procedurally. Mean procedural time was 84±17 minutes. Mean intensive care unit stay was 1.5±0.3 days and mean hospitalization length was 4.0±0.8 days. Relevant technical and clinical peri-procedural outcomes are depicted in Table 2.

Median follow-up was 194 days. NYHA class was ≤II in 94.5% patients at six-month follow-up, which represented a significant improvement in symptomatic status (p=0.002). Temporal evolution of patient's functional class is presented in Figure 2.

Figure 2.

Distribution of patient's functional class according to New York Heart Association classification pre-procedure, by the time of discharge, and at six-month follow-up.

(0.09MB).

MR was quantified as grade ≤2 in 88.2% (N=15) patients by the time of discharge. In follow-up, 72% of patients (N=13) had grade ≤2 MR. Three patients had grade 3MR, and two patients had grade 4 MR.

There was a significant reduction in mean indexed LA volume (63±7 mL/m2 pre-procedure, compared to 45±6 mL/m2 in F-UP, p=0.038), mean indexed LV end-diastolic volume (87±7 mL/m2 vs. 79±9 ml/m2, p=0.001), and pulmonary artery systolic pressure (PASP) (44±4 vs. 31±8 mmHg, p=0.002); with no significant differences regarding tricuspid regurgitation severity or LV systolic function. Table 3 details relevant echocardiographic parameter evolution over the follow-up period.

No re-hospitalizations for heart failure were observed in follow-up. No major adverse cardiac or cerebrovascular events were registered. Table 4 details major clinical safety outcomes.

The re-intervention rate was 11.1% (N=2), driven by recurrence of severe MR. In one patient, the mechanism of procedure failure was related to re-prolapse of a treated leaflet, thought to be a consequence of LV reverse remodeling causing relative elongation of implanted neochordae. This patient underwent a re-do NeoChordTM with chord re-tensioning. In another patient, a rupture of three neochords at the apical attachment site was documented, and the patient later underwent standard surgical MVr with adjunctive annuloplasty, with good outcome.

Table 5 summarizes the main characteristics of patients with recurrent MR grade ≥3 at six-month follow-up.

We describe an early single-center Portuguese experience of a transapical off-pump MVr technique. This series demonstrates the feasibility of a new less invasive cardiac procedure, with a minithoracotomy approach without cardiopulmonary bypass. These results strengthen the technique efficacy and safety profile in short-term follow-up. The successful placement of two or more neochordae was achieved in all patients and resulted in a significant peri-operative reduction of MR severity (grade ≤2). Acute procedural success rates above 85% have been consistently reported in previous series9,12-14.

The rate of conventional MV surgery (repair or replacement) for NeoChordTM failure at 30-day follow-up was 0%, lower than that reported by Colli A. et al. (8%)9 or in the TACT Trial (27%).5

We observed a significant clinical improvement after the procedure, assessed by NYHA functional class (NYHA class≥III was 28% pre-procedure vs. 0% in early follow-up, i.e., at 30 days).

Left heart chamber positive remodeling is expected following the correction of the volume-overload status. We observed a significant reduction in mean indexed LA volume and mean indexed LV end-diastolic volume in early follow-up. A similar analysis was conducted by Kurnicka K. et al., with comparable results.15

Pulmonary arterial systolic pressure reduction, although inconsistently reported in previous studies, is also a physiopathological, plausible finding. We observed a significant PASP reduction early on in follow-up.

Left ventricular ejection fraction is highly load-dependent, and severe MR correction may result in spuriously lower values early on in follow-up, related to a higher afterload status. Also, LV apex “stunning” after ventriculotomy may affect echocardiographic appreciation of global systolic function. Furthermore, patients deemed candidates for this procedure have mostly preserved LV systolic function. Together, these factors may lessen the positive influence of postprocedural LV remodeling on systolic function. Other series failed to document a significant improvement in this parameter,15 and our work replicated this finding.

A proportion of patients had recurrence of grade ≥3 MR between 30 day and 6 months (28%, N=5), but only a minority had indication for re-intervention (11%, N=2).

Importantly, in the case where MVr with NeoChord™ implantation was unsuccessful in early follow-up (30 days to 6 months after the procedure), the patient was safely managed with either a re-do procedure (N=1) or conventional on-pump surgery (N=1). This is also in accordance to previously published data.12,15 It is noteworthy that in none of our patients was re-intervention clinically driven; instead, echocardiographic parameters of MR severity prompted surgical indication, along with an oligosymptomatic status that was comparable with the pre-procedure condition. Furthermore, in the patient who underwent conventional cardiac surgery, a re-repair was possible, highlighting one of the advantages of the technique, as the mitral valve is barely touched in the first procedure.

No major cardiovascular and/or cerebrovascular adverse events were recorded in our series over the abovementioned period.

Several mechanisms have been proposed to explain early recurrence of grade ≥3 MR, and are mainly related to patient selection criteria, technical issues of the procedure (mainly related to native chordae injury, not necessarily accompanied by acute rupture), or the so-called “curling” phenomenon of posterior leaflet (which displays postoperative posterior annular displacement causing exaggerated posterior systolic motion, with resultant end-systolic prolapse and an eccentric anteriorly directed regurgitant jet). Interestingly, in a case series of patients that experienced recurrent regurgitation, only a minority was attributable to reverse remodeling of the LV causing relative Neochord™ elongation (1.9%).16 In the early days, it was thought this would become the Achilles heel of the technique.

A variety of factors may explain such mismatch between MR grade ≥3 recurrence and the minimal impact on clinical status, possibly including a strong treatment placebo effect, reverse atrial and ventricular remodeling (with better dynamic systolic and diastolic performance on exertion), and small but clinically significant reductions in PASP.

Several procedure refinements are being widely adopted to minimize recurrence of moderate to severe MR, and include (1) placement of one to two more neochordae than the number deemed necessary in order to avoid excess chord tension; (2) overtension of implanted neochordae to account for LV reverse remodeling; (3) exclusion of patients with inadequate coaptation depth (LAI<1.15); (4) more posterolateral access at the left ventricle apex and careful device navigation in LV, along the posterior wall axis, to minimize the risk of native chord entanglement and injury; (5) treat preferentially anatomic type A and B patients to ensure more predictable results; and (6) perform more anterior left ventricular access in patients with borderline LAI to bring the MV coaptation line more anteriorly.17

Our results are consistent with the largest series concerning medium-term (12-month) follow-up, which included 144 patients.18 The authors reported the safety and feasibility of this approach, together with restoration of surgical-like MV functional status in 70% of all cases, and significant ventricular and atrial reverse remodeling.

To date, there are limited data on medium-term outcomes, consisting of a small series of six patients originally enrolled in the TACT trial. At five years, three patients experienced procedural failure and underwent conventional surgery, and three patients remained with residual MR and no symptoms over this period.6

A recently published analysis addressed the learning curve effect in this subset: An initial “learning phase” (up to 20 procedures) may have a probability of failure as high as 25%, whereas in the so-called “expert phase”, it may decrease to 5%.19 In this series, an overall failure rate of 11% at one year follow-up was found, which is similar to our results, even though we are still in the first twenty procedures.

The minimally invasive nature of the procedure, with short in-hospital and intensive care unit length of stay, low procedural complication rates, and encouraging medium term results may set the stage for the technique to grow in the near future. Given that a proportion of patients with severe MR recurrence over follow-up may still be treated with either a re-do procedure or conventional on-pump surgical MV repair, provides additional safety.

As the therapeutic armamentarium of minimally invasive techniques for off-pump MVr rapidly grows (either through transapical or trans-septal routes), careful patient selection remains the key. Multiparametric echocardiographic evaluation is of outmost importance, in procedural planning and in real-time guidance,20 with protocol refinements and technological improvements rendering evermore safe and predictable results.21

There are some important limitations of this study to acknowledge. First, we describe a small case sample with limited follow-up, which can preclude solid conclusions about infrequent short and medium-term clinical and/or echocardiographic adverse events. Second, we cannot ignore the learning curve effect of the assisting heart team, as it could have mitigated some advantages of the technique driven by sub-optimal results in the very early experience. This includes a non-negligible MR ≥3 recurrence at six-month follow-up: 28% (n=5 pts), albeit without significant worsening in the clinical status. Third, we applied some pathophysiological and anatomical exclusion criteria widely used in this subset, but it prevents the authors from drawing conclusions about the most complex cases of degenerative MV disease, once again emphasizing the major challenge of proper patient selection as the procedure technique matures and is disseminated. According to some authors, adjunctive mitral ring annuloplasty is a key element to maintaining annular stability, improving leaflet coaptation and durability of repair, as in most instances there is concomitant annular enlargement or dysfunction. This shortcoming should be acknowledged when selecting patients for this procedure. Finally, we must keep in mind that transapical neochord implantation lacks clinical data on long-term follow-up, and further studies with a larger patient population and longer follow-up periods are needed before widespread clinical application of this technique. It must be highlighted that this treatment strategy, and presented results, only apply to the treatment of primary (degenerative) MR. At present, the technique must be regarded as an alternative to conventional MV repair (gold standard). Careful inclusion and exclusion criteria should be applied, and patients must be informed about the probability of short-to-medium-term rates of success, concerning MR recurrence.

Minimally invasive MVr using the NeoChord™ system is safe, effective and reproducible in the short term. The technique may become a valuable treatment option for carefully selected patients with primary severe MR due to degenerative MV prolapse, more so in those with isolated leaflet disease. Early clinical and echocardiographic results suggest a significant symptom improvement, sustained MR grade decrease, and favorable left cardiac chamber remodeling, with low re-intervention rates. Owing to the minimal interference with valvular and subvalvular apparatus, re-doing MVr, either using minimally invasive or conventional surgery, is still possible in patients with suboptimal clinical and/or echocardiographic result. These results warrant further confirmation in larger cohorts, in longer period of follow-up.

The authors have no conflicts of interest to declare