Mid-term outcomes of right ventricular papillary muscle approximation for severe functional tricuspid regurgitation

Abstract

Open in new tabDownload slide

OBJECTIVES

Recurrence of tricuspid regurgitation (TR) after tricuspid annuloplasty can occur in cases where a dilated right ventricle exists and subsequent leaflet tethering follows. We previously reported a new technique of the right ventricular papillary muscle approximation (RV-PMA) for functional TR associated with leaflet tethering. The objective of this study is to elucidate the mid-term outcomes and evaluate the durability of RV-PMA.

METHODS

Between January 2014 and March 2023, we applied RV-PMA in 20 patients of advanced functional TR with severe leaflet tethering. The indication of the technique was severe TR with leaflet tethering height >8 mm, and/or a right ventricular end-diastolic diameter >45 mm. The patients were followed up with echocardiography before discharge and at annual interval thereafter.

RESULTS

There was no perioperative mortality. In the echocardiography performed before discharge, TR was decreased to mild or less in 85%, and a significant improvement in right ventricular end-diastolic diameter and tethering height were achieved (53–45 mm and 11.1–4.4 mm, respectively). Furthermore, during the median 3-year follow-up period, TR was kept controlled mild or less in 80% of the cases.

CONCLUSIONS

RV-PMA is considered to be a safe, effective and durable technique as an additional approach for tricuspid annuloplasty.

INTRODUCTION

Growing data of evidence have revealed that advanced functional tricuspid regurgitation (FTR) is a poor prognostic factor in patients with heart failure. FTR is associated with progressive remodelling of the right ventricle (RV) that further changes tricuspid valve anatomy. The standard procedure for the repair of FTR is tricuspid annuloplasty (TAP). However, recurrence of tricuspid regurgitation (TR) after TAP is frequently observed especially in patients with the severely dilatated RV and severe leaflet tethering [1–4]. The most suitable surgical technique for these high-risk patients is still under debate. We previously reported a new technique of the right ventricular papillary muscle approximation (RV-PMA) for FTR associated with leaflet tethering, and this study aimed to elucidate the mid-term outcomes and evaluate the durability of RV-PMA [5].

PATIENTS AND METHODS

Ethical statement

This study was approved by institutional review board of our hospital (Protocol No. 1250 and 3507, date of approvals: 26 June 2017 and 23 January 2020, respectively). The need for individual patient was waived.

Study population and echocardiographic assessment

Between January 2014 and March 2023, 291 patients underwent tricuspid valve surgery for FTR and 20 of whom underwent RV-PMA. The indications for RV-PMA were severe FTR with leaflet tethering height >8 mm, and/or right ventricular end-diastolic diameter (RVDd) >45 mm. TR was assessed using colour doppler imaging in apical four-chamber view of the transthoracic echocardiography, and classified into grades ranging from none-trivial, mild, moderate to severe based on the extent of the regurgitant jet's reach and the vena contracta. RVDd was measured as a linear diameter at the basal third of RV inflow tract in the end-diastolic phase. The tethering height was measured by the tip of the leaflets and the annular plane at the mid-systolic phase, when TR flow was at its maximum. The RV end-systole and end-diastole cavity was approximated by planimetry, and the % change in those area was defined as RV fractional area change (RVFAC). Tricuspid annular plane systolic excursion (TAPSE) was measured by placing M-mode cursor along the lateral tricuspid annulus near the free RV wall and tracing from the crest to the trough.

Surgical technique

After the annuloplasty stitches are placed on the tricuspid annulus under cardiac arrest, 2 sets of horizontal mattress suture using pledgeted 3–0 polypropylene are placed at the base and mid-level of papillary muscles (PMs). Each set is continuously and semicircumferentially sutured with multiple bites on the free wall of the RV in a clockwise direction; starting from the lateral side of the anterior PM and ending on the dorsal side of the posterior PM at the ventricular septum. Then the suture sets are tied down to approximate the displaced PMs in a side-by-side fashion (Fig. 1). A 28- or 30-mm semi-rigid partial ring is placed to complete the annuloplasty. For detailed information on the schema, please refer to the initial report paper [5].

In cases with more pronounced RV dilation, echocardiogram showed the interventricular septum shifts towards the D-shaped left ventricle, and then tethering of the septal leaflet becomes apparent. In such instances, additional patch augmentation of the septal leaflet with fresh autologous pericardium was performed. Septal leaflet was incised along the annulus and a semilunar autologous pericardial patch of 1.5–2 cm in height was anastomosed using a continuous interlocking suture of 5–0 polypropylene.

Follow-up

All the patients were followed up with echocardiography before discharge and at annual interval thereafter to compare the severity of TR. Additionally, RVDd, RVFAC and TAPSE were also evaluated as parameters of right ventricular function.

Statistical analysis

All statistical analysis was performed using JMP^® Pro 16 (SAS Institute, Cary, NC, USA). Continuous variables were expressed as mean ± standard deviation, while categorical variables are expressed as proportions. The comparison of pre- and postoperative blood test data and echocardiographic parameters was conducted using the paired t-test, and values of P < 0.05 were considered statistically significant. Survival curve was estimated using Kaplan–Meier analysis.

RESULTS

Demographics

Table 1 summarizes the clinical and echocardiographic characteristics of the patients. The mean age was 73 ± 5.2 years, and 10 were men. Nineteen patients (95%) had chronic atrial fibrillation (Af) or flutter and all of the cases were persistent arrhythmias with mean duration of 15 years. According to the preoperative New York Heart Association classification, there were 9 patients in class II, 10 patients in class III, and 1 patient in class IV. TRI-SCORE was 4–5 (intermediate risk) in 6 and ≧6 (high risk) in 14. Mitral valve (MV) operation was performed in 14 (70%) patients. Indication for MV surgery was severe mitral regurgitation (MR) in 6, severe mitral stenosis in 4 patients, although severe TR was the equivalent operative indication in all cases. In our early experiences, isolated tricuspid valve repair in cases with mild atrial functional MR resulted in immediate post-cardiopulmonary bypass worsening of MR, which required MV repair. We speculated that this could be associated with geometric changes in the interventricular septum and left ventricular papillary muscles following the decreased right ventricular volume and subsequent dilatation of left ventricle. Since those experiences, we subsequently adopted a strategy of addressing mild to moderate functional MR concurrently. Due to these considerations, other 4 cases underwent concomitant MV surgery, but main indication for the operation was the management of severe TR. In 10 patients including those 4 patients, the main indication of surgery was severe TR.

All the patients had severe FTR within 3 months before the operation, but in 2 cases, aggressive diuretic treatment including the use of tolvaptan (vasopressin antagonist) and intravenous carperitide (human atrial natrium peptide) reduced TR to a moderate-severe equivalent on preoperative echocardiograms. RVDd was markedly enlarged at 53 ± 9.3 mm, and the tethering height was also very high at 11 ± 2.6 mm.

Perioperative outcomes

Table 2 summarizes the operative characteristics, perioperative outcomes and concomitant procedures of the patients. The size of the prosthetic valve ring was 28 mm in 16 cases and 30 mm in 4 cases. Patch augmentation of the septal leaflet was performed in 3 cases. Although most of the patients had Af, Maze operation was not indicated because successful sinus rhythm recovery was considered unlikely due to severely enlarged right and left atrium and/or long history of Af. Postoperative central venous pressure decreased from 15.7 ± 5.6 to 8.3 ± 3.6 mmHg (P < 0.01) and cardiac index also showed an increase. The length of intensive care unit stay and hospital stay was 1.9 ± 1.1 and 25 ± 22 days, respectively, and there was no in-hospital death. Perioperative complications included 2 cases of re-exploration due to bleeding and 3 cases of arrhythmia requiring additional treatment; a pacemaker implantation (PMI) due to sick sinus syndrome (SSS), an implantable cardioverter defibrillator implantation due to sustained ventricular tachycardia, a catheter ablation due to atrial tachycardia. Echocardiography performed before discharge showed that TR was reduced at mild or less in 17 cases (85%). Furthermore, both RVDd and tethering height were significantly improved at 45 ± 8.7 mm (P < 0.01) and 4.4 ± 2.3 mm (P < 0.01), respectively. RVFAC worsened to 30 ± 7.4%.

Mid-term outcomes

In this study, it's important to note that the initial 3 cases experienced loss to follow-up or were discharged from further follow-up during the course of the study. Table 3 summarizes the Mid-term outcomes of the patients. During a median follow-up period of 36 ± 22 months (range, 6–74 months), there were 2 cases of all-cause death, with 1 case attributed to cardiac death caused by biventricular heart failure and the other to malignancy. The 3-year overall survival rate was 87%, and the rate of freedom from re-admission at 3 years was 84% (Fig. 2). The causes of re-admissions were SSS requiring PMI, rapid Af and biventricular heart failure. In several cases, there were temporary episodes of worsened TR during echocardiographic follow-up. This was attributed to elevated fluid retention or tachycardia/bradycardia resulting from Af, and after adjustment of diuretic medication or antiarrhythmic treatment, an improvement in the degree of TR was observed. Patients who developed moderate TR included one who developed SSS requiring subsequent PMI and the other with residual severe pulmonary hypertension from left side heart disease. The pre- and postoperative echocardiogram in 4 chamber view is shown in Video 1. Postoperative echocardiogram showed improved motion of leaflets, better coaptation of the tricuspid valve and only trivial TR although RV contractility remained poor. In the latest echocardiographic examinations, TR was controlled at mild or less in 80% of cases (Fig. 3). RVDd remained stable at 45 ± 8.1 mm, and tethering height was 4.2 ± 2.3 mm, showing no worsening compared to the perioperative period. RVFAC improved slightly to 36 ± 9.5 mm from the perioperative decline but remained similar to preoperative levels. TAPSE also did not show significant improvement and remained at low level.

In the most recent follow-up, there was a clinical improvement, with the mean New York Heart Association class decreasing from 2.6 to 1.3 (P < 0.01). In blood tests, the levels of total bilirubin, alkaline phosphatase and brain natriuretic peptide significantly decreased from 1.8 ± 0.9 to 1.0 ± 0.3 mg/dl (P < 0.01), 123 ± 60 to 88 ± 36 U/l (P = 0.01) and 191 ± 115 to 105 ± 105 pg/ml (P < 0.01), respectively. The platelet count was slightly increased from 142 ± 59 to 155 ± 71 × 10³/μl, and the albumin level remained almost stable at 4 g/dl.

DISCUSSION

Recurrent TR after TAP has been associated with numerous risk factors in the literature; female, renal failure, ischaemic heart disease, Af, reduced left ventricular systolic function, pulmonary hypertension and pacemaker placement are some examples [6, 7]. The existence of tricuspid leaflet tethering is also shown as a significant risk factor for recurrent TR. Fukuda et al. reported a tethering height >7.6 mm in 2D echocardiography was an independent predictor of residual TR, and we also previously reported a tethering height >7.2 mm measured by preoperative 3D computed tomography was identified as a predictor of recurrent TR [2, 3]. Dreyfus et al. [4] proposed a new staging system for FTR and defined a tethering height >8 mm measured by transthoracic echocardiography to be a most advanced stage. Therefore, we defined the indication of performing PV-PMA as advanced TR with leaflet tethering height >8 mm, and/or equivalent RV dilatation of RVDd >45 mm.

In cases of severe TR where control of regurgitation is difficult with TAP alone, efforts have been made to control TR by performing TVR or additional techniques to TAP. TVR is performed in about 20% of cases with severe FTR and up to 50% of those with isolated severe FTR [8]. However, TVR is associated with postoperative low-output cardiac and RV dysfunction, resulting in a high mortality rate of 10–20% [9, 10]. Choi et al. [11] compared changes in right ventricular volume and function following TAP and TVR on cardiac magnetic resonance imaging, and reported that right ventricular ejection fraction was decreased significantly in the TVR group. Reasons for the worse outcomes of TVR are still unclear, but Nguyen et al. [12] revealed that the implantation of a bioprosthetic valve increases a turbulence in RV, resulting in the formation of a negative vortex and energy loss by in vitro study using mock circulatory system. Limited durability of bioprosthesis at the tricuspid position due to pannus formation or native valve attachment has been another critical concern.

On the other hand, various additional techniques in conjunction with TAP have been reported, but the efficacy and durability of those techniques remain unclear. Dreyfus et al. [13] performed patch augmentation with autologous pericardium on the anterior leaflet in 15 cases, resulting in the complete disappearance of TR in all cases at the end of surgery. Pettinari et al. [14] applied a similar technique to 22 cases and reported mid-term outcomes with a median follow-up period of 2.1 years. The 30-day and 4-year survival rates were 81.1% and 71.6%, respectively, and the rate of freedom from moderate or greater TR during the observation period was 85.7%. De Bonis et al. reported surgical edge-to-edge repair of tricuspid leaflets so-called ‘clover technique’ in 96 cases, and TR was reduced to mild or less in 92% of cases before discharge. In the median follow-up period of 9 years, the 10-year survival rate was 78% and the rate of freedom from moderate or greater TR at the latest echocardiography was 78% [15].

Recently, transcatheter tricuspid valve repair is developing as an alternative approach. In the TRILUMINATE trial using TriClip™ (Abbott, Chicago, IL, USA) for edge-to-edge repair, the rate of freedom from severe or greater TR was 71% and significant improvement of clinical conditions was reported at 1 year postoperatively [16]. However, moderate TR was remained in 34% at 1 year postoperatively, and re-admission rate was high at 0.78 events/patient-year. Furthermore, Kresoja et al. [17] reported a higher risk of postoperative heart failure admissions and mortality following transcatheter tricuspid valve repair in cases with progressing right heart failure; decreased right ventricular ejection fraction or lower TAPSE, suggesting a need for careful consideration of intervention strategies in cases with advanced RV dysfunction.

Repair strategies at the subvalvular level have been also reported. Al-Attar et al. [18] first reported RV-PMA named the ‘right papillary muscle sling’ using e-PTFE tube in 5 cases, and TR was reduced to mild or less on echocardiography before discharge. We also previously reported the short-time outcomes of RV-PMA in 7 cases, resulting in mild or less TR and substantial improvement of the tethering height in all cases [4]. Some authors made modifications to RV-PMA like ‘papillary muscle septalization’ or ‘spiral suspension’ and reported favourable outcomes in short-term, but these reports do not provide outcomes regarding the changes of right ventricular function or the tethering improvement, and long-term effectiveness of TR control remains unclear [19–21]. When septal leaflet tethering is present, there is a possibility that TR cannot be controlled only with RV-PMA. In such cases, we have added septal leaflet patch augmentation. The septal leaflet is supported by multiple small chordae directly originated from the septal wall, so we believe it difficult to elevate a septal leaflet with papillary muscle suspension or triangle technique. To treat septal leaflet tethering, another report uses a vertical thread between septum and posteroseptal portion of the annulus to reposition the posteroseptal portion of the annulus down towards the RV apex, which leads to restoration of the original 3D configuration of the annulus [22]. However, a long-term durability is a concern because we suspect such a suture format might have a possibility to cause dehiscence.

As demonstrated by Gaweda et al. [23] in the animal experiments, the strategy of RV-PMA is to reduce tethering height by approximating the displaced anterior and posterior PMs, and this leads to an improvement in TR. Our current report clinically showed its effectiveness and favourable durability; significant improvement of tethering height and mild or less TR in 80% of the cases at 3 years after surgery.

Most of our patients had long history (mean duration of ∼15 years) of Af, normal left ventricular function, and no pulmonary hypertension. These features meet the criteria of atrial functional TR. However, recently proposed definition of atrial functional TR includes normal RV volume and function and no severe tethering of tricuspid valve leaflet [24]. It remains a major difficulty to differentiate later stage of atrial functional TR, which also include RV remodelling due to permanent volume overload of the right heart. Our patients had severe dilatation and poor systolic function of the RV and severe tethering of TV leaflet as a result of long-time volume overload. According to the recent definition of functional TR, those pathologies are rather classified as ventricular functional TR, although the aetiology of the disease progress may largely depend on long-standing Af. A more aggressive application of Maze procedure would enhance the durability of tricuspid valve repair if it can achieve a successful sinus conversion. However, in our patients, the atrial diameter was severely dilated due to long-standing Af and the likelihood of sinus rhythm conversion through Maze procedures was considered low. Furthermore, additional Maze procedure has a risk of inducing brady-arrythmia postoperatively, and if a new permanent PMI was required, it might result in device lead-induced recurrent TR. Therefore, we considered surgical ablation was not a good option for those patients.

While we acknowledge the emerging literature suggesting a survival advantage of performing tricuspid repair in the beating heart, we performed the procedure under cardiac arrest to ensure a secure suture, obtain a clear surgical field and perform a reproducible regurgitation test [25]. We employed the perioperative management option for right heart failure as needed. We prioritize early extubation and utilize small dose of inotropes (mainly milrinone or dobutamine), and carperitide (human atrial natriuretic peptide) for diuresis. We were not obliged to use mechanical circulatory support or haemodialysis in any patient. We consider that the most significant factors contributing to our favourable early postoperative outcomes appear to be the improvement in haemodynamics observed immediately after surgery, as evidenced by the improvement in cardiac index and central venous pressure.

There is considerable variation in tricuspid papillary muscle anatomy, particularly in the posterior papillary muscle. Despite this heterogeneity, our surgical approach allows for approximation regardless of the variations in the posterior papillary muscle, because we placed the approximation stitch into the base of papillary muscle but also into right ventricular wall behind the papillary muscle. Therefore, we could approximate small posterior papillary muscle as 1 bundle. Furthermore, in this procedure, plication of RV free wall could be expected to show a protective effect on RV remodelling. We have not encountered a late problem due to suture loosening or detachment such as acute TR recurrence or RV redilatation.

While notable improvement in clinical symptoms accompanied the control of TR, advanced right ventricular contractile dysfunction did not improve postoperatively. We considered the reason for the initial drop of RVFAC as multifactorial. In a manner similar to observations following MV surgery, we attribute the initial decrease in RVFAC before discharge to the increased afterload due to the reduction in regurgitation. Another explanation would be the stress to myocardium induced by cardiac arrest, which gradually returning to pre-operative levels as these burdens are alleviated. TAPSE also exhibited a postoperative decline, albeit showing gradual improvement. However, it is noteworthy that several papers do not recommend TAPSE as post-TAP evaluation of right ventricular function. Therefore, we did not construct a figure of the change of TAPSE.

Limitations

The study has a number of limitations. First, this study is retrospective design, single-centre trial, and the small number of patients included. Second, there is no control group in this study because we applied RV-PMA in all the patients who met our criteria. Third, the application of patch augmentation of septal leaflet is based on intraoperative judgement, and there are no specific numerical values determined for its application prior for surgery.

CONCLUSION

PV-PMA effectively addressed FTR and demonstrated satisfactory durability for a 3-year period. This technique may serve as a preferred option for an additional approach in cases of FTR associated with RV dilatation and leaflet tethering. However, further long-term follow-up is essential to validate its sustained effectiveness.

Conflict of interest: none declared.

DATA AVAILABILITY

All data underlying this article will be made available on reasonable request to the corresponding author.

Author contributions

Chihiro Ito: Conceptualization; Data curation; Formal analysis; Investigation; Methodology; Project administration; Resources; Software; Validation; Visualization; Writing—original draft; Writing—review & editing. Hiroki Kohno: Data curation; Formal analysis; Validation; Writing—review & editing. Kaoru Matsuura: Data curation; Formal analysis; Writing—review & editing. Michiko Watanabe: Data curation; Formal analysis; Validation. Tomohiko Inui: Data curation; Formal analysis; Validation. Goro Matsumiya: Conceptualization; Data curation; Formal analysis; Funding acquisition; Investigation; Methodology; Project administration; Resources; Supervision; Validation; Writing—review & editing.

Reviewer information

European Journal of Cardio-Thoracic Surgery thanks the anonymous reviewers for their contribution to the peer review process of this article.

Presented at the 37th EACTS Annual Meeting, Vienna, Austria, 4–7 October 2023.

REFERENCES

ABBREVIATIONS

 
• Af

  Atrial fibrillation

 
• FTR

  Functional tricuspid regurgitation

 
• MR

  Mitral regurgitation

 
• MV

  Mitral valve

 
• PMs

  Papillary muscles

 
• RV

  Right ventricle

 
• RVDd

  Right ventricular end-diastolic diameter

 
• RVFAC

  Right ventricular fractional area change

 
• RV-PMA

  Right ventricular papillary muscle approximation

 
• SSS

  Sick sinus syndrome

 
• TAP

  Tricuspid annuloplasty

 
• TAPSE

  Tricuspid annular plane systolic excursion

 
• TR

  Tricuspid regurgitation