Triangular resection versus neochordal replacement for posterior leaflet prolapse: a morphological assessment Free

Abstract

OBJECTIVES

To evaluate the morphological reconstruction as assessed by 3D transoesophageal echocardiography after triangular resection and neochordal replacement in the treatment of posterior leaflet prolapse.

METHODS

We retrospectively selected 46 patients with isolated posterior leaflet prolapse who were undergoing mitral valve repair using triangular resection (n = 20) and neochordal replacement (n = 26) techniques. Sixty patients without valvular heart disease were also included as the control group. Thorough 3D transoesophageal echocardiography inspections were performed for the entire cohort, and morphological parameters of mitral valve complex were measured and compared. Mid-term repair durability was followed up by transthoracic echocardiography.

RESULTS

The baseline parameters, including annular size, leaflet tenting, leaflet area, coaptation line lengths and aortomitral angle, were significantly larger in prolapsed valves. After repair, tenting volume, exposed posterior leaflet area and coaptation line lengths were restored to the normal range. Baseline clinical characteristics and 3D transoesophageal echocardiography parameters were comparable in patients treated with 2 techniques, and all parameters remained comparable between the resection and the non-resection groups after repair, except for exposed posterior leaflet area and posterior leaflet ratio. At 62.2 ± 18.5 months after surgery, degrees of residual regurgitation were similar between 2 techniques.

CONCLUSIONS

Triangular resection and neochordal replacement can achieve comparable restoration to structural normality and functional competency of mitral valves with posterior leaflet prolapse. Resection of prolapsed segment does not significantly affect coaptation geometry but instead may aid in achieving normal posterior leaflet ratio.

INTRODUCTION

Posterior leaflet prolapse, one of the most common forms of degenerative mitral regurgitation (MR), can be repaired with several established approaches [1]. The classic ‘resect and reattach’ technique has been long utilized and has proved to be efficacious and durable. Further modifications, focused on a smaller area of resection and avoidance of annular plication, may facilitate reverse remodelling of the left ventricle and recovery of cardiac function [2]. Meanwhile, experience in the use of neochordal replacement for the treatment of posterior leaflet prolapse has been accumulating, and this technique shows comparable surgical outcomes and durability [3, 4]. Advocates of this ‘respect’ technique have postulated that the preserved tissue may presumably yield a larger coaptation zone, and better leaflet kinematics and reverse remodelling, than leaflet resection [5, 6]. Meanwhile, comparable results have also been reported [7].

Over the past decade, with the advances in diagnostic capability of ultrasound imaging, 3D transoesophageal echocardiography (3DTOE) has become an increasingly important tool in morphological assessment of the mitral apparatus. Although previous studies of 3DTOE have revealed impacts of surgical reconstruction of MR on mitral valve morphology [8, 9], no extensive data exist regarding structural restoration after repair of posterior leaflet prolapse or morphological discrepancies between resectional and non-resectional repairing techniques.

In this study, we sought to evaluate the morphological characteristics of prolapsed mitral valves before and after surgery and compare the surgical reconstruction between triangular resection and neochordal replacement.

PATIENTS AND METHODS

Study design

The study was approved by the institutional review board at Shanghai Chest Hospital, and the requirement for patient consent was waived due to its retrospective nature. Between January 2008 and January 2012, a total of 455 consecutive patients with severe MR underwent mitral valve repair at Shanghai Chest Hospital. Routine TOE was performed for all these patients. The exclusion criteria included patients with non-degenerative mitral valve pathology (n = 141), anterior leaflet or bileaflet pathology (n = 116), concomitant cardiac procedures (n = 34), multiple repair attempts (n = 12) and combined leaflet repair techniques (n = 46). Forty-two patients with poor imaging or missing data on their 3DTOE reports were also excluded. To mitigate the confounding factors, 18 patients with full ring annuloplasty were further excluded. Eventually, a total of 46 patients treated with either triangular resection (n = 20) or neochordal replacement (n = 26) met the inclusion criteria. Furthermore, 3DTOE data from 60 patients who underwent TOE before radiofrequency catheter ablation for non-valvular atrial fibrillation were included to form the control group.

Operative approach

All operations were performed by 1 surgeon (W.Z.) with the use of mild hypothermic cardiopulmonary bypass through median sternotomy, and myocardial protection was achieved by intermittent antegrade/retrograde cold blood cardioplegia. The mitral valve was exposed via the traditional left atrial incision, parallel to the interatrial sulcus. The details of posterior leaflet resection and neochordal replacement had been thoroughly described in prior reports [3–7]. Briefly, after assessing the mitral valve, the structural characteristics of the entire mitral complex and the extent of the prolapsed segment were identified. The choice of technique was basically determined by the surgeon’s preference. Leaflet resection was performed by triangular resection of excess leaflet tissue and reattachment of leaflet remnants in the absence of annular plication. Neochords were implanted individually using expanded polytetrafluoroethylene sutures to repair the prolapsed leaflet or ruptured autologous chordae. After filling the left ventricle with saline, we deployed titanium clips on the neochord to control an appropriate length, tied knots and then removed the clips. Finally, a flexible C-shaped ring (Cosgrove ring; Edwards Lifesciences, Irvine, CA, USA) was applied for annuloplasty after measuring the inter-trigone distance. The strategy for choosing appropriate ring size was consistent between the 2 groups, and all annuloplasty rings were equally restrictive. Following favourable results of saline testing, the atrial incisions were closed using Prolene 4-0 running sutures (Ethicon, Somerville, NJ, USA) and the heart was deaired. The post-bypass 3DTOE parameters were not collected until mean arterial pressure reached 75 mmHg with adequate left ventricular filling.

Echocardiographic parameters and follow-up

In this study, the 3DTOE was performed by 1 experienced echocardiographer (W.W.) using a Philips iE33 echocardiograph (Philips Medical Systems, Andover, MA, USA) connected to an X7-2t transoesophageal probe before and after cardiopulmonary bypass. The 3D zoom included the entire mitral apparatus throughout a single cardiac cycle. Each 3D zoom data set was acquired intraoperatively and exported to MVQ QLAB software (Philips Medical Systems) for offline review. Assessment of the 3D reconstructions and the parameter measurement were performed between March 2016 and November 2016 by W.W. All parameters were measured at the early systolic phase. The 3DTOE parameters of annular dimension included anterior–posterior diameter, anterolateral–posteromedial diameter, non-planarity angle (angle subtended at the commissural diameter between the anterior and posterior horns of the mitral annulus), annulus circumference, annulus area, annulus height (distance on the same surface from the highest point to the lowest point of the annulus) and aortomitral angle. The parameters of leaflet and coaptation included tenting volume (volume enclosed between the annular plane and leaflets), exposed anterior and posterior leaflet areas, posterior leaflet ratio (exposed posterior leaflet area/total leaflet area) and lengths of anterior and posterior coaptation lines (from anterolateral to posteromedial commissure). For parameters that could be measured by either 2D or 3D methods, the 3D data were preferentially documented and used for analysis (Fig. 1). Degrees of MR were designated as none (0), trivial (1+), mild (2+), moderate (3+) and severe (4+), according to the current guidelines [10].

Routine follow-ups of transthoracic echocardiography (TTE) were performed at the discretion of the operating surgeons or the referring cardiologists after discharge. Each patient in this study cohort had at least 2 outpatient visits and 1 TTE performed. The degree of residual regurgitation was re-evaluated and compared between the 2 groups. The latest follow-up time was 62.2 ± 18.5 (range 50–92) months.

Statistical analysis

Continuous variables are shown as means ± standard deviations or medians with range, and they were compared using the Student’s t-test or the Mann–Whitney U-test, as appropriate. Categorical variables were described as numbers and percentages and compared with the χ² test or the Fisher’s exact test, as appropriate. A P-value of <0.05 was considered statistically significant. All statistical analyses were performed using SAS v8.0 (SAS Institute, Inc., Cary, NC, USA).

RESULTS

The baseline clinical and TTE characteristics of both groups are presented in Table 1. Of the 46 patients with posterior leaflet prolapse, the overall mean age was 56.5 ± 8.6 years (range 41–73 years), and 26 patients were men. All the baseline clinical and TTE variables, including body surface area, comorbidities, left chamber dimensions and location of prolapsed segment were comparable between the resection and the non-resection groups.

The preoperative and the postoperative 3DTOE data from the prolapse and the control groups are presented in Table 2. All parameters were significantly different between the prolapse and the control groups. The 3D annular size, leaflet tenting height and volume, exposed leaflet areas, posterior leaflet ratio, coaptation line lengths and aortomitral angle were found to be larger in the prolapse group than in the control group, whereas non-planarity angle was smaller in the surgical group. A difference between anterior and posterior leaflet coaptation lengths was also observed in the prolapse group. After surgical intervention, parameters of annular size and exposed anterior leaflet area were significantly lower in the prolapse group, whereas non-planarity angle and posterior leaflet ratio were above normal. Postoperative tenting volume, exposed posterior leaflet area, coaptation line lengths and aortomitral angle regressed to the normal range.

For patients treated with the 2 techniques, annular shape and size, exposed leaflet areas, coaptation line lengths and aortomitral angle were comparable between the non-resection group and the resection group before surgery. After surgical intervention, except for posterior leaflet area and posterior leaflet ratio, all parameters remained comparable between the 2 groups (Table 3).

All 46 patients underwent successful repair. TTE performed before discharge showed that 18 patients had no regurgitation, 26 had trivial regurgitation and 2 had mild regurgitation. The mean trans-mitral gradient was 3.0 ± 1.0 mmHg. The degree of residual regurgitation was comparable between the 2 surgical groups. At follow-up, TTE showed similar repair durability of the 2 techniques (Table 4).

DISCUSSION

Posterior leaflet prolapse is one of the most common degenerative lesions causing mitral valve insufficiency, and it should be treated preferably with mitral valve repair [1, 11]. Common techniques include posterior leaflet resection and neochordal replacement (and other non-resectional techniques). Extensive studies have investigated the difference between the 2 most common techniques in terms of durability and reverse remodelling of the left heart chambers [3, 6, 7]. However, 3DTOE data for the assessment of morphological alterations after repair of posterior leaflet prolapse remain relatively sparse. This investigation confirmed that (i) both repair techniques for posterior leaflet prolapse can resolve mitral insufficiency effectively by restriction of enlarged annuli and normalization of leaflet tenting and (ii) triangular resection and neochordal replacement can achieve comparable morphological reconstruction.

According to the 3DTOE data, mitral valves with posterior leaflet prolapse had larger annular size (anterior–posterior diameter, anterolateral–posteromedial diameter, annulus circumference, annulus area and annulus height), leaflet tenting volume, exposed anterior and posterior leaflet areas, posterior leaflet ratio, lengths of coaptation lines and aortomitral angle than the normal valves. Pathological alterations, including annular enlargement, left ventricular remodelling, prolapsed leaflet and ruptured/elongated chordae, were the main causes of the parametric abnormality. On the other hand, the non-planarity angle was smaller in the prolapse group than in the control group, indicating that the annuli of prolapsed valves enlarge in a 3D manner, with higher degree of enlargement in the annulus height than diameter. Typical changes after mitral valve repair have been reported in prior studies, including reduced size of mitral annulus, flattened annular plane and increased matching of anterior and posterior coaptation, which were also observed in the present study [8, 9]. The enlarged annular size and exposed leaflet areas were reduced significantly after annuloplasty, resulting in a smaller anterior leaflet area but similar posterior leaflet area compared with the control group. Influenced by leaflet repairing techniques and the annuloplasty ring, the tenting volume and height after surgery were found to be comparable to the normal values, suggesting structural restoration of the entire annulus–leaflet–chordae complex. Normalization of the aortomitral angle aids in avoidance of the residual MR and left ventricular outflow tract obstruction induced by systolic anterior motion.

The classic leaflet resection has been employed over the past several decades with efficacy and durability, and its surgical details have been modified for better reconstruction of the posterior leaflet, aiming to resect the redundant leaflet tissue only and to restore sufficient coaptation [2]. On the other hand, neochordal replacement, a non-resectional technique, has also been advocated for the preservation of leaflet mobility and a larger coaptation zone [12–14]. The core difference between the 2 techniques originates from the attitude towards prolapsed segments of the posterior leaflets. The question of whether to resect the redundant tissue to halt further degeneration or to respect the prolapse and use it for better coaptation remains under debate. Although numerous studies have investigated differences in long-term functional restoration between the 2 techniques, very few have evaluated the real-time morphological alterations after surgery. In this study, the postoperative data showed that all parameters but the exposed posterior leaflet area were similar between the 2 groups. Comparable annular size, leaflet tenting and lengths of coaptation lines indicated that triangular resection could achieve a similar geometry of coaptation to neochordal replacement. In addition to the correction of functional abnormality, both techniques could be performed with favourable morphological correction for mitral valves with prolapsed posterior leaflet.

Posterior leaflet ratio is a new concept raised in this study, and it represents the proportion of posterior leaflet area to total leaflet area and the relative positions of coaptation lines. In mitral valves with posterior leaflet prolapse, excess leaflet tissue led to anteriorly located coaptation lines and increased posterior leaflet ratio (prolapse versus control 49.1% vs 38.7%, P < 0.001). After surgery, posterior leaflet ratio remained higher than normal (44.8% vs 38.7%, P < 0.001). Inter-group analyses showed that compared with neochordal replacement, triangular resection was more effective in reducing posterior leaflet area, lowering the posterior leaflet ratio and restoring the curvature and positions of coaptation lines towards normal. However, whether such a morphological discrepancy would have impact on the functional outcomes, namely repair durability and left ventricular flow dynamics, merits further scrutiny.

With regard to coaptation area, it is conceivable that the non-resectional technique yields a larger coaptation zone than the resectional technique. One prospective study, by Falk et al. [6], suggested that the coaptation height at A2/P2 was significantly longer in the non-resection group (7.6 mm vs 5.9 mm, P = 0.03), which might indicate better repair durability. However, given that the coaptation height at A2/P2 has been reported to be 5.2–5.8 mm in normal valves [15, 16], the area–durability correlation may not be taken for granted. Moreover, because of the complexity of coaptation zone geometry, the coaptation height at 1 segment alone may not be sufficient to quantify overall coaptation strength. A comprehensive method of calculating the area of the total coaptation zone, provided by Cobey et al. [17], should be utilized for comparison of coaptation strength in future studies.

Although the morphological reconstruction achieved by the 2 techniques is comparable, it is noteworthy that the non-resectional technique may be more efficacious when treating extensive prolapse [18]. Extensive posterior prolapse complicates the repair by increasing the incidence of minor regurgitant jets after implantation of neochordae, which can be corrected by additional neochordae or combined repairing techniques. Extensive prolapse may also be a great challenge when determining the site and area for resection. Because triangular resection may not be applicable to patients with extensive prolapse, we excluded these patients in this study. Selection of the non-resectional and resectional techniques should be based on characteristics of leaflet pathology and surgeon’s preference.

Limitations

This study has limitations that include its retrospective methodology, centre-specific bias and disparities in medical and surgical management in the recent years. In addition, the study sample was relatively small, and the results may be biased by cohort selection. However, the homogeneous 3DTOE data of the entire cohort were measured by 1 experienced echocardiographer (W.W.), with the intention of minimizing the interobserver variability and increasing the reliability of the parametric data. Moreover, in this study, all patients underwent annuloplasty using the C-shaped Cosgrove ring. Whether mitral valve reconstruction remains comparable between the 2 techniques with full ring annuloplasty is still unknown. Further studies with larger multicentre cohorts are warranted.

Funding

This study was supported by the Shanghai Chest Hospital (2014YZDH10301 to W.W., 2014YZDH10302 and YZ2015-ZX03 to W.Z.) and the Shanghai Municipal Commission of Health and Family Planning (201540309 to W.Y.).

Conflict of interest: none declared.

REFERENCES

Author notes