Multidisciplinary decision-making in mitral valve disease: the mitral valve heart team

The concept of a multidisciplinary decision-making team is well established in various medical disciplines [1, 2] and has been associated with improved survival [3, 4]. Recently, such multidisciplinary teams have been introduced in the fields of cardiology and cardiac surgery, specifically to make decisions regarding coronary revascularisation and transcatheter aortic valve replacement [5‐7]. Although decision-making in the so-called heart team is apparently intuitive and has a class I recommendation in most recent guidelines [8‐10], supporting comparative data is still lacking [11‐13]. For mitral valve disease, only a few studies have reported first experience in multidisciplinary decision-making, limited to transcatheter mitral valve therapies [14, 15]. The variety in mitral valve treatment options is increasing with transcatheter and off-pump surgical interventions [16‐18]. Furthermore, surgical mitral valve repair has proven to be associated with a steep learning curve, and outcome is significantly procedural volume related [19, 20]. Therefore, a dedicated mitral valve care team seems even more warranted for treatment of mitral valve disease. Recently, we introduced the concept of a dedicated mitral valve heart team at our centre. This multidisciplinary approach focuses on a balanced treatment strategy for individual patients based on their specific mitral valve pathology, anatomical eligibility, comorbidities, background and wishes. The aim of the current study is to demonstrate standardised diagnostic pathways in mitral valve patients, give insight into our strategy of clinical decision-making for allocation of an individualised treatment pathway and to demonstrate the clinical outcome of all patients referred to the dedicated mitral valve heart team.

All consecutive patients whose mitral valve pathology was discussed by the dedicated mitral valve heart team between 1 January 2016 and 31 December 2016 were prospectively included in the current study. Patients were referred from four regional hospitals or by our own centre. Data were collected prospectively.

A total of 158 consecutive patients were discussed by the mitral valve heart team. Patients were allocated to their designated treatment modality. Sixty-seven patients were treated surgically, 20 with catheter-based interventions and 71 conservatively (Fig. 1).

The repair rate of MR based on degenerative disease was 100%. Within the surgically treated group, 46 patients underwent on-pump surgical mitral valve repair, 8 beating heart mitral valve repairs were performed, and 13 patients underwent biological or mechanical mitral valve replacement for rheumatic or ischaemic disease or systolic anterior motion.

In the catheter-based intervention group, 15 patients (75%) underwent a percutaneous edge-to-edge repair and 5 patients (25%) percutaneous annuloplasty. Reasons for conservative treatment were: MR not severe enough for intervention (30 cases, 42%), patient’s wish (17 cases, 24%), deteriorated clinical state (16 cases, 23%) and non-mitral surgical/interventional treatment (8 cases, 11%).

Baseline characteristics are depicted in Tab. 1. Surgically treated patients tended to be of younger age with fewer comorbidities and a lower surgical risk based on the EuroSCORE. Twenty-two patients (33%) underwent concomitant surgery. An endoscopic approach was used in 23 of 35 patients with isolated valve disease, whereas sternotomy was performed in 12 patients. Reoperations, endocarditis and non-elective cases were included in the analyses as well. Baseline echocardiographic parameters are presented in Tab. 2.

Thirty-day mortality was assessed for all groups. There was no mortality within 30 days for the catheter-based intervention group, whereas 3 patients died within 30 days of decision-making in the conservative group (4.2%). For surgically treated patients, a distinction was made between (1) the overall group, (2) the group treated by a dedicated mitral valve surgeon, and (3) elective, primary cases operated on by a dedicated mitral valve surgeon. For the overall group (n = 67, treated by 3 surgeons) 30-day mortality was 4.4% (3 cases, Fig. 2a). For the group treated by the dedicated surgeon (n = 60) 30-day mortality was 1.7% (1 case, a reoperation, Fig. 2b) and in the primary, elective group (n = 57) no 30-day mortality was observed (Fig. 2c).

A similar decrease in occurrence of MACCE within 30 days was found. In the catheter-based intervention group, 1 patient had to undergo a left ventricular assist device implant after edge-to-edge repair (5%). Sixteen percent of the patients in the surgically treated group had complications (Fig. 2d), 11.7% in the dedicated group (Fig. 2e) and 7.7% in the primary, elective group (Fig. 2f).

Postoperative echocardiography performed at 3 months after discharge revealed 23.5% of patients treated with a catheter-based intervention to have residual MR > grade 2, compared to 4.5% in the surgically treated group (3 patients, Fig. 3).

Survival was estimated using the Kaplan-Meier method at a median follow-up of 450 days (range 138–673 days) and is depicted in Fig. 4 for the various groups, demonstrating beneficial long-term survival for surgically treated patients. In addition to the stratification for surgically treated groups, Fig. 4d provides information on survival of the patient group with severe MR, revealing a poor short-term prognosis for the conservatively treated group.

Although intuitive, there is no supportive data for the use of the heart team in decision-making for cardiovascular disease [11].

Sixty-seven patients (43%) were treated surgically. Mitral valve reparability was assessed by the heart team preoperatively, and a repair rate of 100% was achieved for degenerative valves. Of all mitral pathologies, 84% were repaired, while 16% were replaced. Of note is the fact that isolated and concomitant mitral surgery were both included in the analysis. In selected cases, transapical beating heart valve repair (NeoChord) was performed when patients were deemed eligible. These older patients, with an overall elevated surgical risk and comorbidities, were eligible for repair but were expected to have a complicated postoperative course.

Furthermore, with the emergence of several multimodality imaging techniques, the current study provides an algorithm for the use of the modalities (CT, 3D anatomical reconstructions, TTE, TOE, CAG) in various stages of the standardised diagnostic pathway. This algorithm, provided in Fig. 1, could prevent unnecessary diagnostics and reduce associated costs, patient burden and exposure to radiation.

A routine CT thorax scan was performed in patients being evaluated for endoscopic mitral valve surgery. We were able to exclude 7 patients in the preoperative course because of inaccessibility or unsuitability of the vessels for this approach. Eventually 25 patients underwent endoscopic surgery, in which no conversions occurred.

Catheter-based interventions proved to be safe (no 30-day mortality, 5% MACCE), but had a relatively high probability of residual MR > grade 2 (23.5%) compared to the surgical group (4.5%), in line with prior studies [24]. Furthermore, after an initial uneventful course, these patients had shorter overall survival, presumably based on their age, poorer clinical state and diminished cardiac function.

A trend towards lower 30-day mortality with fewer major complications for patients treated by a dedicated surgeon was observed, confirming previous studies [25]. These studies demonstrated better outcomes and survival in mitral valve surgery when performed by a dedicated surgeon on a weekly basis after completion of the learning curve [19, 20, 26], indicating surgical volume to be a determinant of repair rate, freedom of reoperation and survival.

Survival was estimated with a median follow-up of 450 days. A relatively high cumulative mortality was observed in the conservatively treated group (25.4%). Most deaths occurred in the subgroup which was treated conservatively because of a deteriorated clinical state. This finding was also observed in a comparable revascularisation study [27]. Additionally, conservatively treated patients with severe MR had a poor short-term prognosis, potentially explained by a combination of a myriad of factors contributing to a higher surgical risk, such as advanced age, severe mitral stenosis with subsequent end-stage heart-failure and a high rate of severe pulmonary hypertension.

As an additional observation, we found a relatively high rate of incidentalomas on preoperative CT scans performed in the planning of endoscopic surgery. Out of 44 patients, 12 incidentalomas (27%) were found, of which 4 (9%) were actual carcinomas requiring further follow-up and/or treatment. Several other CT screening studies describe a lower prevalence of tumours on screening [28]. However, little is known yet about the complex interplay between cardiovascular disease and cancer, which could both be a different manifestation of common underlying risk factors [29], explaining this finding in a patient group with extensive cardiovascular disease.

The current study demonstrated the implementation of a multidisciplinary mitral valve heart team, gave insight into our strategy for clinical decision-making and treatment allocation, and demonstrated short-term clinical outcomes of patients with mitral valve disease. Our research group will focus on a comparative study with historical cohorts, potentially providing a scientific basis for the current recommendations in guidelines, as we believe a multidisciplinary approach will improve efficiency and patient outcome.