Predictability and durability of mitral valve repair in patients with severe degenerative mitral regurgitation in medium sized centres

Abstract

Introduction

Mitral valve repair (MVr) is now the treatment of choice to correct severe degenerative mitral regurgitation (MR). Repair rates vary greatly from centre to centre, and the concept of heart valve centres of excellence has been established.

Objective

The purpose of this study was to see whether large international centre repair rates, and outcomes, are transferrable to medium-sized centres with an interest in mitral repair.

Methods

Between 2011 and 2018, a total of 346 patients underwent mitral valve surgery by a single surgeon. Of these, 238 consecutive patients had repairs, or attempted repairs for degenerative MR, and are included in this study.

Results

The study sample consisted of 71% male patients and had a mean age of 64.4 ± 12.3 years; 66% of the study population had concomitant procedures. The overall repair rate in this cohort is 99%. Mean follow up was 3.7 ± 1.9 years. At 5 years, the freedom from MR ≥ 3+ was 95.9 ± 1.9% and at 7 years 91.1 ± 3.8%. Freedom from reoperation at 5 years was 92.9 ± 3.7%, while the 5 years actuarial survival was 89.1 ± 3.7%. On a multivariate analysis, predischarge echo grade was associated with higher risk of future reoperation (odds ratio (OR) = 21.82, p = 0.05). Only age (OR = 1.3, p = 0.03) was predictive of long-term survival.

Conclusions

In specialised medium-sized heart centres, where the surgical team have undergone specialist mitral training, favourable short- and long-term outcomes are achievable with mitral repair rates similar to those from large international centres of excellence. In these heart centres, early surgery should be considered for all patients with severe degenerative MR.

Introduction

Early surgical correction of severe mitral valve regurgitation (MR) caused by degenerative mitral valve disease is associated with improved long-term survival and diminished heart failure risks,^1,2 particularly when performed in high-volume specialist valve centres.^3,4 The benefits of mitral valve repair (MVr) over mitral valve replacement (MVR) for patients with severe degenerative MR has been well established^5,6 and include lower operative mortality, better long-term survival and fewer valve-related complications.⁷

Recently, the concept of centres of excellence in MVr has gained momentum and set the standards for best surgical practice in patients with severe MR.⁸ These centres should meet several criteria, such as large MVr surgery volume (centre and surgeon), availability of appropriate periprocedural imaging and willingness to provide data regarding outcomes (eg, repair rates, mortality, stroke rates and repair durability).⁸ Both the American and European guidelines define mitral centres of excellence as specialist centres with the likelihood of successful and durable repair >95% with an operative mortality of <1% or centres with high repair rates (≥50 annual repairs/annum), low operative mortality and a record of durable results.^9,10 In such centres, early surgical correction in asymptomatic patients with severe MR is now widely accepted, prophylactically averting the potential morbidities of chronic MR such as left ventricular (LV) dysfunction, atrial fibrillation (AF), pulmonary hypertension and reduced survival.¹¹

In a country such as Ireland, with a relatively small population of approximately 4.8 million, the annual number of mitral repairs a centre can feasibly facilitate is limited. Although volume is an essential criterion for a ‘centre of excellence’ designation, equally important are fundamentals such as repair rates, repair durability, long-term outcomes and individual surgeon training and experience. Therefore, in a smaller heart centre, assessment of surgical outcomes might be more practical in order to evaluate the reproducibility of results in MVr performed in large centres of excellence.

The aims of this study were therefore to assess the reproducibility of successful MVr in a medium-sized centre, to examine the factors that influence the success of mitral repair and to evaluate the short- and long-term outcomes of a single-surgeon series of mitral repair.

Methods

Eligibility criteria

The present study is based on an 8-year experience of MVr surgery. Inclusion criteria were first-time MVr or attempted repair for severe MR that was due to degenerative mitral valve disease, with MVr or MVR performed by a single surgeon in our institutions between 1 January 2011 to 31 December 2018. Institutional Review Board (IRB) approval was obtained from the local ethics committee on 12 June 2020 (IRB No: 1/378/2156TMR).

Surgical technique

The surgical approach used in the majority of cases in this series was the sternotomy approach with bicaval cannulation. Myocardial protection was performed with intermittent cold blood cardioplegia. Intraoperative transoesophageal echocardiography was used in all cases. The left atrium was opened through an incision in the interatrial groove and the mitral valve was optimally exposed.¹²

Repair strategy was based on Carpentier’s techniques and included assessment of leaflet quality, leaflet or commissural prolapse, leaflet height, condition of chordae, presence of leaflet or annular scarring or calcification.¹³ Posterior leaflet procedures included initially quadrangular and, more recently, triangular leaflet resection or no leaflet resection and/or utilisation of polytetrafluoroethylene neochords.¹⁴ Occasionally, a sliding technique was used if systolic anterior motion (SAM) was a risk,¹⁵ particularly if it was not dealt with by another technique. Commissural prolapse was generally dealt with by commissuroplasty or by using neochords. Anterior leaflet prolapse (ALP) was generally corrected with neochords or looped chords. Barlow’s disease was usually treated with leaflet resection and with neochord placement. A partial or complete annuloplasty prosthesis was implanted in almost all patients.

Since 2016, a minimally invasive technique has been performed in a subset of patients. Patients have a CT scan preoperatively to assess the aorta and iliac and femoral vessels. Cardiopulmonary bypass is established via femoral vessels and right minithoracotomy is used for access. The aorta is cross-clamped and antegrade cardioplegia is delivered with a cannula placed directly in the ascending aorta via the minithoracotomy. The mitral valve is then accessed via the left atrium and the valve is repaired using previously mentioned techniques. Postoperatively, the patient is managed in similar fashion to other repair patients.

Patient demographics

Between 2011 and 2018, a total of 238 patients underwent mitral repair surgery for degenerative valve pathology by a single surgeon at the Mater Misericordiae University Hospital, Mater Private Hospital and Blackrock Clinic, Dublin, Ireland. The preoperative data of the entire cohort are listed in Table 1. There was a male predominance (71%) in this cohort and the mean age of the entire cohort was 64.4 ± 12.3 years. Male patients were younger than their female counterparts (63.3 ± 0.9 vs 66.9 ± 1.6 years; mean difference (MD) = 3.6 ± 1.8 years; p = 0.05).

Table 1 .

Demographics, preoperative characteristics of the study population and surgical profile

Demographics
Age (Mean (years) ± SD)	64.4 ± 12.3
Male	63.3 ± 0.9
Female	66.9 ± 1.6
Gender (%)
Male	74.2
Female	25.8
NYHA status (%)
NYHA I	28.0
NYHA II	31.0
NYHA III	32.5
NYHA IV	8.5
LVEF status (%)
Normal (EF ≥ 55%)	81.9
Mild dysfunction (EF = 45–54%)	11.4
Moderate dysfunction (EF = 30–44%)	6.2
Severe dysfunction (EF < 30%)	0.5
Valve pathology (%)
ALP	15.5
PLP	55.0
Bileaflet prolapse	29.5
Atrial fibrillation/flutter	40.1
Surgical profile
Ring/band size (median)	34
Cross clamp time, minutes	62.4 ± 23.2
Bypass time, minutes	85.3 ± 24.9
Adjunct Alfieri stitch (%)	14.3
Type of prosthesis (%)
Band prosthesis	60.4
Ring prosthesis	38.8
No annuloplasty	0.8
Leaflet repair (%)
Posterior leaflet resection	59.0
Posterior chords	28.8
Anterior chords	23.5
Anterior leaflet resection	0.8
Commisuroplasty	0.8
Sliding plasty	5.0

ALP = anterior leaflet prolapse; EF = ejection fraction; LVEF = left ventricular ejection fraction; NYHA = New York Heart Association; PLP = posterior leaflet prolapse; SD = standard deviation

Preoperative morbidity

Data on preoperative New York Heart Association (NYHA) functional status is summarised in Table 1. The biggest subset in this group, at 32.5%, were patients with NYHA status III. The vast majority of patients had normal LVEF (ejection fraction (EF) ≥ 55%) at 81.9% followed by mild LV dysfunction (EF = 45–54%) at 11.4%, moderate dysfunction (EF = 30–44%) at 6.2% and severe dysfunction (EF < 30%) at 0.5%. ALP represents 15.5% of the study population, while posterior leaflet prolapse (PLP) represent the majority at 55.0% and bileaflet prolapse (BLP) at 29.5% (Table 1). Barlow’s valve disease represents 14.3% of the cohort.

Surgical procedures

The surgical procedures are summarised in Table 1. Most patients underwent repair surgery (99.2%, n = 236) while the remaining underwent replacement (0.8%, n = 2). The two patients who had a replacement had mixed ischaemic–degenerative disease. PLP was managed most commonly with a triangular leaflet resection. Sometimes neochords were also used in this situation, particularly if there was a risk of SAM,¹⁵ or if a leaflet resection was avoided. The majority of patients with ALP were managed with neochords either separately or as looped chords. Two patients with ALP had anterior leaflet resections. A sliding plasty was performed in 5% of patients to help prevent SAM.

A total of 234 patients had either a band or ring prosthesis (99.2%). Of these, the partial annuloplasty band was the most used (60.4%) followed by the complete ring (38.8%). The median annuloplasty size in this cohort was 34. An Alfieri stitch was employed as an adjunct for the repairs in 14.0% of cases, generally to manage a small residual jet of regurgitation, which might be a haemolysis risk, but occasionally to help prevent SAM; 66% of the study cohort had concomitant procedures and only 34% had isolated mitral surgery (Figure 1). Out of the 87 patients who had atrial fibrillation/flutter preoperatively, 96.6% (n = 84) also underwent concomitant left atrial appendage (LAA) occlusion and left atrial or biatrial maze procedure. Coronary artery bypass grafting (CABG) was performed in 16.8% of patients.

Figure 1 .

Summary of patients undergoing mitral valve surgery with other concomitant procedures such as LAA occlusion, maze procedure, CABG, ASD or VSD repair, TVR, AVR and PFO closure. ASD = atrial septal defect; AVR = atrial valve replacement; CABG = coronary artery bypass graft; LAA = left atrial appendage; PFO = patent foramen ovale; TVR = tricuspid valve repair; VSD = ventricular septal defect

Overall, the mean ± standard deviation (SD) cross clamp time in the entire cohort was 62.4 ± 23.2 minutes and cardiopulmonary bypass time was 85.3 ± 24.9 minutes. The mean ± standard error of the mean (SEM) cross clamp time was significantly longer in patients with concomitant procedures versus patients without additional procedures (68.07 ± 2.5 vs 51.73 ± 2.09 minutes, mean difference (MD) = 16.34 ± 3.77 minutes; p < 0.001). Similarly, cardiopulmonary bypass time was longer in patients with concomitant procedures compared with patients with isolated mitral valve surgery (90.93 ± 2.61 vs 74.54 ± 2.64 minutes; MD = 16.4 ± 4.07 minutes; p < 0.001).

Statistical methods

Data are presented as mean ± SD for continuous variables and as simple ratios (%) for categorical variables. Group comparisons were performed with standard t test or χ² test as appropriate. Estimates of survival and freedom from valve-related events were calculated by Kaplan–Meier method and expressed as the ratio (%) ± SEM of patients who remained event-free. Univariate and multivariate analyses were performed using the binomial logistic regression method to identify independent predictors of survival and reoperation. The potential variables evaluated in the study include age, gender, preoperative NYHA status, preoperative atrial fibrillation/flutter, pre- and immediate postoperative echocardiography results, type of leaflet prolapse, concomitant surgery and cross clamp and bypass times. A p value of <0.05 was considered statistically significant. Statistical analyses were performed using IBM SPSS v.23 and GraphPad Prism v.8.

Results

Postoperative outcomes

All patients had intraoperative transoesophageal echo and at least one transthoracic echo prior to hospital discharge. Intraoperatively, 91.5% had no, or only trace, MR, 6.8% had mild MR and 1.7% had moderate MR. The postoperative complications are summarised in Table 2. The majority of patients (90.8%) had no, or mild, complications following surgery. In-hospital mortality in this cohort was 0.4% (n = 1) involving an 85-year-old patient who developed acute cerebrovascular accident (CVA) postoperatively. The requirement for pacemaker insertion was the commonest postoperative complication at 2.5%, followed by resternotomy for bleeding at 2.1%.

Table 2 .

Immediate postoperative complications in patients undergoing mitral valve surgery

Postoperative complications	Overall, n (%)
Pacemaker insertion	6 (2.5%)
Resternotomy for bleeding	5 (2.1%)
Pleural/pericardial effusion with interventions	4 (1.6%)
Deep wound infection	3 (1.3%)
Renal failure requiring dialysis	1 (0.4%)
Sepsis	1 (0.4%)
CVA (with recovery)	1 (0.4%)
Early mortality	1 (0.4%)
CVA with in-hospital mortality	1 (0.4%)

CVA = cerebrovascular accident

Durability of repair

All surviving patients have been contacted successfully as part of this study after surgery. Overall, surgery improves NYHA grades (χ² = 64.3, p < 0.001). For patients who presented with NYHA grade I preoperatively, there was no significant change in their symptoms on the latest follow up (p = 0.08). However, for patients with preoperative NYHA grades II–IV, surgery significantly improved their symptoms and conditions (p < 0.001) (Table 3).

Table 3 .

Change in NYHA grade postoperatively based on the latest follow up

NYHA Grade	Mean difference postop (±SEM)	p value
I	−0.174 (0.06)	0.08
II	0.62 (0.11)	<0.001
III	1.62 (0.12)	<0.001
IV	2.4 (0.27)	<0.001

NYHA = New York Heart Association; SEM = standard error of the mean

Available serial postoperative echocardiographic data were analysed when available (82%). The mean ± SD follow up was 3.7 ± 1.9 years (range: 3 months–7 years). Taking the most recent echocardiographic data available, 52.8% had no significant, or only trivial, MR, 34.3% had 1+ MR, 11.1% had 2+ MR and 1.8% developed ≥3+ MR. At 5 years, freedom from MR ≥3+ was 95.9 ± 1.9% and at 7 years was 91.1 ± 3.8% (Figure 2).

Figure 2 .

Freedom from moderate-to-severe MR (3+ and 4+) on follow up. Freedom from MR ≥3+ was 95.9 ± 1.9% at 5 years and 91.1 ± 3.8% at 7 years. MR = mitral regurgitation

Reoperation was performed in 2.9% of patients (n = 7) during the mean follow-up period of 3.7 ± 1.9 years. In these cases, the mean ± SD interval time before reoperation was 10.6 ± 15.6 months. Freedom from reoperation at 3 years was 94.6 ± 3.0% and at 5 years was 92.9 ± 3.7% (Figure 3). The actuarial survival at 3 years in this cohort was 96.8 ± 1.8%, at 5 years was 89.1 ± 3.7% and at 7-year follow up, it was 85.2 ± 5.2% (Figure 4).

Figure 3 .

Freedom from reoperation on follow up. Freedom from reoperation was 94.6 ± 3.0% at 3 years and was 92.9 ± 3.7% at 5 years.

Figure 4 .

Long-term survival in the study population. The actuarial survival was 96.8 ± 1.8% at 3 years, 89.1 ± 3.7% at 5 years and 85.2 ± 5.2% at 7 years.

Predictors of reoperation

On a Pearson’s analysis, only age was weakly correlated with future reoperation (r = −0.16, p = 0.02) while variables such as gender, preop NYHA grade, LVEF status, type of leaflet prolapse, concomitant surgery and cross clamp/bypass times were not.

On a univariate analysis, only the predischarge echo grade was potentially a predictor of future reoperation (OR, 2.91, p = 0.13). On a multivariate analysis, only predischarge echocardiographic grade of MR was associated independently with higher risk of future reoperation (OR, 21.82; 95% confidence interval (CI), 1.02–467.7; p = 0.05) whereas gender, age, preop NYHA status, AF, LVEF, type of leaflet prolapse and cardiopulmonary bypass times were not predictive (Table 4).

Table 4 .

Determinants of MR reoperation and long-term survival

	Determinants of MR reoperation				Determinants of long-term survival
	Univariate analysis		Multivariate analysis		Univariate analysis		Multivariate analysis
Variables	OR (95% CI)	p value	OR (95% CI)	p value	OR (95% CI)	p value	OR (95% CI)	p value
Age	0.94 (0.89–0.99)	0.02	0.96 (0.89–1.04)	0.30	1.29 (1.10–1.51)	0.02	1.3 (0.043–2.88)	0.03
Male	1.02 (0.19–5.41)	0.98	6.67 (0.17–255.64)	0.31	1.45 (0.29–7.25)	0.65	1.07 (0.04–26.8)	0.97
Preop NYHA status	0.78 (0.34–1.76)	0.54	0.98 (0.25–1.45)	0.81	3.58 (1.33–9.67)	0.01	3.76 (0.31–46.2)	0.30
Preop atrial fibrillation	4.16 (0.49–35.19)	0.19	2.18 (0.88–14.89)	0.99	4.00 (0.96–16.63)	0.05	1.62 (0.08–32.65)	0.76
Bypass time	0.98 (0.96–1.01)	0.14	0.98 (0.94–1.02)	0.25	1.03 (1.01–1.05)	0.02	1.04 (0.97–1.11)	0.32
Predischarge echocardiographic grade	2.91 (0.72–11.78)	0.13	21.82 (1.02–467.7)	0.05	0.93 (0.34–1.57)	0.73	1.12 (0.22–12.90)	0.77
Localisation	–	0.57	–	0.15	–	0.63	–	0.71
Anterior versus posterior	2.46 (0.39–15.42)	0.34	2.56 (1.45–13.76)	0.22	1.2 (0.19–21.0)	0.04	1.14 (0.23–34.57)	0.90
Bileaflet versus posterior	1.25 (0.20–7.71)	0.81	1.34 (0.33–6.90)	0.36	0.15 (0.15–1.5)	0.11	0.11 (0.07–1.46)	0.87

CI = confidence interval; MR = mitral valve regurgitation; NYHA = New York Heart Association grading; OR = odds ratio

Predictors of survival

On a Pearson’s analysis, preoperative NYHA grade (r = 0.22, p = 0.05) and age (r = 0.25, p = 0.001) were associated with long-term survival, while the presence of preoperative AF has a weak correlation (r = 0.16, p = 0.04).

On a univariate analysis, preop NYHA status (OR, 3.58; p = 0.01), presence of AF preop (OR, 4.00; p = 0.05) and age (OR, 1.2; p = 0.02) were independent determinants for survival in this cohort. On a multivariate analysis, only age (OR, 1.3; p = 0.03) was predictive of long-term survival (Table 4).

Discussion

In this study, the overall repair rate of 99% seen in this cohort is comparable with rates reported in other published studies in high-volume centres (>100 mitral operations/annum).^3,16 Two patients could not be repaired. On average, a total of 29 MV repairs (for degenerative disease) per annum per surgeon was performed, just meeting the criterion for a centre of excellence.⁸ MVr continues to carry a low operative mortality and, in our cohort, the mortality rate of 0.4% compares favourably with the existing guidelines.^9,10

MVr has significant advantages over MVR, including a reduction in prosthetic material used, preservation of all of the subvalvular apparatus, the avoidance of anticoagulation and its complications and leads to improved survival. Mitral repair is now the procedure of choice for patients with severe MR due to degenerative disease. The published results of mitral repair surgery are now excellent and are often from high-volume academic centres with the best results. The results may not always extrapolate to all centres, and there is a recognition that there is a wide variety in rates of mitral repair among surgeons and institutions.^17,18 Thus, the concept of mitral repair centres of excellence has been fostered, with repairs being performed in high-volume specialised units.

It is recognised that the relationship between surgeon or institution volume and outcomes is complex, and actual surgeon-specific repair rates and outcomes are probably of greater value.⁹ With each surgeon-led team, the potential patient and the referring doctors need to know the results, which aids decision making. This is particularly important in smaller and medium-sized centres.

One of the perennial concerns with MVr is its long-term durability, which is evaluated by the freedom from reoperation. The freedom from reoperation in our series was 93 ± 3.7% at 5 years, which is similar to other reported series from larger centres.^5,16 Of the seven patients who required revision surgery in this cohort, the majority (n = 5) required early surgery due to haemolysis (n = 2), severe MR (n = 2) or SAM (n = 1). Not surprisingly, the reoperation rate was higher in the first year than in subsequent years. This has also been the experience in other studies.¹⁹ Echocardiographic follow up showed a 5-year freedom from MR ≥3+ of 95.9 ± 1.9%, in keeping with previous publications.¹⁹

The actuarial survival in this cohort was 97 ± 3.0% at 3 years, 89 ± 3.7% at 5 years and, at 7-year follow up, survival was 85 ± 5.2%. This is comparable with data from higher volume centres, that have reported 7-year survival rates of 88% and 10-year survival rates of 85%.^2,12 In our cohort, advancing age and advanced NYHA status preoperatively were predictors of adverse long-term survival. Possible predictors of mortality may have been ALP, or omission of an annuloplasty ring, but these were not evident in our population, although only two of our patients did not have an annuloplasty prosthesis. Chordal shortening has been previously linked with increased long-term mortality²⁰; however, this practice was not used routinely during the study period.

Approximately one-quarter of the patients in this study were classed as NYHA I preoperatively and remained so during follow up. This may indicate that the procedure was associated with low morbidity in this subgroup and can therefore be offered to asymptomatic patients with severe MR and normal LVEF. These results are in keeping with data from large registries that have shown improved long-term outcomes of mitral repair in patients with lower NYHA status.^9,12,18 Thus, a low-morbidity surgery that confers significant survival advantage can be offered to the patients in this region, and this outcome satisfies one of the main aims of this review.

While low NYHA status may predispose patients to lasting results, this study comprised a significant group of quite symptomatic patients, as the largest NYHA subgroup was class III. Patients with worse NYHA grades preoperatively appear to benefit the most from surgery (p < 0.001) (Table 3). These patients noted the biggest improvement in breathlessness postoperatively, indicating that low morbidity surgery can be offered to the relevant cohort in this region with positive results.²¹ Although this is not the ideal time to operate on these patients because they may have by this stage developed reduced LV function, it is gratifying to confirm their improvement in clinical status on follow up.

MVr with ring annuloplasty was performed in most cases in this cohort, and a partial ring was inserted in the majority of cases. The real-world difference in outcome between partial and complete ring annuloplasty remains debated in the current literature,^9,22,23 whereas the repair procedure with no ring has been linked with a higher rate of recurrent MR.¹² We could not identify a difference in our results due to the type of annuloplasty prosthesis used.

Interestingly, the rate (16.8%) of concomitant CABG observed in this study group was slightly higher than that observed in studies from high-volume centres.^2,3,22 This may be partly related to age of these patients and the preponderance of males in our cohort and coronary artery disease prevalence rates in the region. Despite this, the cross-clamp and the cardiopulmonary bypass times in the group with concomitant procedures were acceptable.

Tricuspid valve repair was performed on 6.4% of the population based on the quantity of TR (which is variable) and the measured tricuspid annular diameter.²⁴ While decisions to repair the tricuspid valve at the same time as performing a mitral repair are not an exact science, it is important to ask the question about the tricuspid valve in all patients undergoing mitral repair.

Over time, the mitral program has evolved. Mitral subspeciality training is important and the senior author was trained in mitral repair, initially in Dublin, and later at the Cleveland Clinic in the US. Numerous site visits to Mitral Centres have taken place over the years. Over time, more complex repairs are done with increasing frequency. Before starting the less-invasive approach, the Mitral Centre in Lyon was visited by the team on several occasions and proctoring was arranged in conjunction with professor Obadia’s team on the first cases. The initial less invasive approach patients were aware that they were being done at an early stage in the learning curve, and team training and proctoring was explained to them, which helped the consent process. Results have been audited and communicated with referring doctors and with patients, and this allows patients to have input into their planned surgery.

Ultimately, the aim is to have a successful and safe mitral repair operation. A less-invasive approach needs to maintain the established principles of using repair techniques that maximise the chances of having a repair that is durable over time. We managed to attain that goal in this series of patients albeit with slightly longer cross-clamp and bypass times than in the open group of patients. The heart team is central to attaining the safety goal for the patient. To date, patient satisfaction with the less invasive approach has been high. The team have also enjoyed the learning experience and this journey is applicable to many surgical teams embracing innovation in many disciplines of surgical practice.

The American College of Cardiology/American Heart Association and European Society of Cardiology/European Association of Cardiothoracic Surgery guidelines recommend that any institution offering a mitral repair program should provide surgery with low mortality and complication rates, and a high rate of repair that is durable.^9,10 This study has demonstrated that these outcomes can be achieved in smaller centres, by surgeons and teams with the appropriate specialist training. In the interests of their patients and referring doctors it is particularly important that these surgeons and centres know their mitral repair results, and do not depend entirely on results extrapolated from large academic centres.

Conclusion

In regions with relatively low population that have developed specialised medium-sized heart centres, where the surgical team have undergone specialist mitral training and where cardiology, imaging and critical care expertise is available, favourable short- and long-term outcomes are achievable with mitral repair rates similar to those from large international centres of excellence. In these heart centres, early surgery should be considered for all patients with severe degenerative MR.

Author contributions

LC: data collection; investigation; methodology; writing-original draft; writing-review and editing. MJ: data collection; formal analysis; methodology; writing-original draft. JM: methodology; supervision; validation. FK: methodology; project administration; supervision. JO’N: methodology; writing review. RB: methodology; writing review and editing. JMcC: conceptualisation; methodology; supervision; validation; writing-original draft; writing-review and editing.