Three-dimensional transesophageal echocardiographic morphological evaluation of the tricuspid valve

Abstract OBJECTIVES The morphology of the tricuspid valve (TV), particularly valves with two posterior leaflets, is attracting attention. The present study was performed to investigate the usefulness of three-dimensional transoesophageal echocardiographic data for morphological evaluation of the TV . METHODS Sixty patients underwent morphological evaluation of the TV by preoperative transoesophageal echocardiography followed by TV repair with median sternotomy, and each leaflet was measured intraoperatively. We analysed the TV morphology in 51 patients whose preoperative echocardiographic findings were consistent with intraoperative findings. RESULTS The mid-systolic echo data, which included the annulus diameter of each leaflet, were correlated with the intraoperative evaluation findings compared with those in the mid-diastole. The annulus and area of the posterior leaflet were larger in patients with two than one posterior leaflet valve (42.4 ± 13.5 vs 30.7 ± 9.1 mm, P < 0.001 and 327 ± 185 vs 208 ± 77 mm2, P = 0.006, respectively). In the severe tricuspid regurgitation patients, the annulus of the posterior leaflet was larger and the annulus of the anterior leaflet was smaller in patients with two than one posterior leaflet valve [posterior: 48 mm [95% confidence interval (CI), 41–54 mm] vs 36 mm (95% CI, 27–45 mm), respectively; P = 0.043 and anterior: 38 mm (95% CI, 33–42 mm) vs 46 mm (95% CI, 40–52 mm), respectively; P = 0.025]. CONCLUSIONS Patients who had a TV with two posterior leaflets had a larger annulus and area of the posterior leaflets. Preoperative three-dimensional transoesophageal echocardiography is useful for the morphological evaluation of the TV.


CONCLUSIONS:
Patients who had a TV with two posterior leaflets had a larger annulus and area of the posterior leaflets. Preoperative three-dimensional transoesophageal echocardiography is useful for the morphological evaluation of the TV.

INTRODUCTION
Tricuspid valve (TV) repair is often performed as a concomitant procedure of left cardiac surgery. Recently, following the establishment of the 2020 Japan JCS/JSCS/JATS/JSVS Guideline, the indications for surgical treatment of TV expanded to include mild or moderate tricuspid regurgitation (TR) in patients with tricuspid annular dilation or atrial fibrillation [1]. As TR treatment has become more widespread, the mechanism of TR and the anatomy of the TV have become growing concerns that are attracting increasing attention.
Regarding the morphological diversity of the TV, Silver et al. [2] reported that the posterior leaflet of the TV actually consists of several leaflets. In addition, Sakon et al. [3] reported that half of the autopsied hearts they examined had TVs with two or more posterior leaflets and that such valves showed anatomical differences in annular length. The TV annulus is a three-dimensional (3D) configuration with a saddle shape. However, as the TV annulus expands, the 3D structure is lost and becomes flattened [4]. In addition, enlargement of the annulus is greater on the freewall side, and the morphology of the TV changes depending on the degree of TR [5]. Therefore, preoperative evaluation of the TV morphology in patients with TR is important to improve the surgical outcome.
Recent advances in 3D echocardiographic technology have made it possible to evaluate the TV morphology using 3D imaging. Some previous studies of 3D transthoracic echocardiography or 3D transoesophageal echocardiography (TEE) indicated that preoperative 3D images were useful to assess predictors of residual TR. However, these reports only evaluated leaflet parameters such as tethering, the total area and the annulus [6,7]; no reports have focused on morphological aspects such as multiple posterior leaflets.
In the present study, we compared preoperative 3D image evaluation of the TV with intraoperative measurement of the TV and evaluated the accuracy of preoperative 3D TEE imaging. We also evaluated the structural characteristics of TVs with two posterior leaflets and their impact on TR severity.

Ethics statement
This study was approved by the Osaka City University Ethical Review Board (approval no. 2020-099). Informed consent was obtained from all patients using an opt-out method.

Patients
From April 2016 to November 2020, 60 patients were diagnosed with secondary TR associated with left heart disease and underwent TV repair with median sternotomy. All patients underwent preoperative TEE and measurement of the diameter of the TV annulus. The exclusion criteria were infective endocarditis and primary TR. Patients without perioperative echocardiographic data and patients who underwent re-do TV repair were also excluded.
Among the 60 patients, we collected data from 51 patients in whom the number of TV leaflets was correctly determined by preoperative TEE in comparison with intraoperative findings. We performed the following 3 analyses. (i) We compared the annulus diameter and the number of TV leaflets as determined by preoperative 3D TEE versus intraoperative measurement. In addition, we examined which of the 2 time phases (mid-systole or mid-diastole) of the measured parameters approximated the intraoperative measurement result. (ii) We compared the echocardiographic parameters of TVs with one versus two posterior leaflets. (iii) To determine whether a relationship was present between the TR grade and the degree of TV annular enlargement in patients who had a TV with one versus two posterior leaflets, we classified the patients into those with < _mild, moderate or severe TR and compared the size of each leaflet among the groups.

Transthoracic echocardiography
All patients underwent preoperative transthoracic echocardiography using an iE33 or Epiq system (Philips Medical Systems, Andover, MA, USA) at our echocardiography laboratory. The TR grade was defined using a multiparametric approach, including an assessment of the colour Doppler-derived jet area, the continuous-wave Doppler-derived jet density and contour and the hepatic vein flow velocity pattern [8]. Continuous-wave Doppler was used to obtain the TR peak velocity (m/s) and the transtricuspid systolic pressure gradient (mmHg), which was calculated as 4V 2 (where V is velocity). The right ventricular systolic pressure was then estimated as the sum of the estimated transtricuspid systolic pressure gradient and right atrial (RA) pressure. An estimated right ventricular systolic pressure of >40 mmHg was considered indicative of pulmonary hypertension [9]. The tricuspid annular diameter was measured at end-diastole, and an annular diameter of >40 mm or 21 mm/m 2 was considered significant annulus dilatation [1]. The RA dimension, left ventricular (LV) end-diastolic dimension, LV end-systolic dimension, LV ejection fraction and left atrial dimension were measured.

Real-time three-dimensional transesophageal echocardiography
All patients underwent preoperative real-time 3D TEE using an iE33 or Epiq system (Philips Medical Systems) at our echocardiography laboratory.
The annulus length of each leaflet was determined by measuring the distance of the tracing length. The area of each leaflet was determined by measuring the tracing area.
Horizontal and vertical diameters were obtained by the QLAB mitral valve navigation (MVN)/3D quantification software package (Philips Medical Systems).
Determination of the commissure area and measurement of all parameters were supervised by a cardiologist of our institution. The number of TV leaflets was determined by identifying each commissure of the TV in the mid-systolic phase and middiastolic phase using 3D TEE with the centre of the ventricular septum at the 6 o'clock position.
• We defined an indentation between two leaflets as a commissure.
Each leaflet annulus was traced and measured in both phases. The annulus length of each leaflet was determined by measuring the length of the tracing line. To determine the leaflet area, each leaflet was traced and measured in mid-systole (Fig. 1A). • In the quantification analysis using MVN, the line passing through the centre of the TV with reference to the intersection of the anterior and septal leaflets was defined as anterolateral-posteromedial, and the line perpendicular to this was defined as anteroposterior.
We measured the anteroposterior and anterolateral-posteromedial distance, total area of the leaflets and total annulus diameter of the leaflets in both phases (Fig. 1B).

Determination of the number of leaflets during operation
The number of leaflets was determined according to the definitions presented by Silver et al. [2]: (i) the commissure was defined as indentation of the leaflets by a fan-shaped chorda, (ii) the fan-shaped chorda forming the anteroposterior commissure arose from the anterior papillary muscle and (iii) the posteroseptal commissure was defined by the fan-shaped chorda arising from the most medially located papillary muscle on the posterior wall (Fig. 2).

Measurement between each commissure area during operation
The number of leaflets was confirmed, and each commissure was marked by pyoktanin blue (Sigma-Aldrich, St Louis, MO, USA). The distance between each parameter of the TV was measured directly using a paper measure (Fig. 2). All intraoperative evaluations were performed by the same surgical team.

Statistical analysis
Descriptive statistics for categorical variables are reported as absolute value and percentage, and continuous variables are shown as mean and standard deviation. Categorical data were compared using the X 2 test. Continuous variables were compared using the Wilcoxon signed-rank test. A regression model was performed to predict the annulus length and included an interaction of the TR level and the number of leaflets. All analyses were conducted with JMP version 13.0 (SAS Institute Inc., Cary, NC, USA), and a P-value of <0.05 was considered statistically significant.

Patients' background
The patients' profiles are shown in Table 1. Most of the patients (75%) had chronic or paroxysmal atrial fibrillation. Thirty-one patients (52%) had hypertension as a comorbidity. All patients had a history of congestive heart failure, and 22 patients (37%) had a history of hospitalization because of heart failure before surgery.

Preoperative echocardiographic data
The preoperative echocardiographic data are shown in Table 2.

Comparison of three-dimensional transesophageal echocardiography data with intraoperative measurements
Intraoperative examination showed that 31 patients (52%) had one posterior leaflet valve and 29 patients (48%) had two posterior leaflet valves. The number of leaflets measured intraoperatively was equal to that as determined by preoperative 3D TEE in 51 of 60 patients (85%). The number of posterior leaflets in 9 patients (15%) did not match the TEE findings.
A comparison of the echo parameters between the 51 patients with correct measurements and the 9 patients with incorrect measurements showed that it was difficult to determine the number of leaflets in patients with a smaller RA diameter (minimum: 44.2 ± 13.9 vs 35.5 ± 9.9 mm, P = 0.017; maximum: 63.8 ± 14.7 vs 54.3 ± 9.7 mm, P = 0.033).
We investigated the 3D TEE parameters of the 51 patients in whom the number of leaflets was correctly identified by preoperative TEE. The whole annulus length and each annulus length were measured in both mid-systole and mid-diastole, and we compared the 3D TEE parameters of the 2 phases with the intraoperative parameters. There was no significant difference in the whole annular length and each annular length between the midsystole data and intraoperative data. The annulus length of the anterior leaflet (TEE: 42.3 ± 9.7 mm vs intraoperation: 36.6 ± 9.6 mm, P < 0.001) and septal leaflet (TEE: 39.9 ± 7.8 mm vs intraoperation: 43.9 ± 7.6 mm, P = 0.003) showed significant  Data are presented as n (%) or mean ± standard deviation.
differences, although there was no significant difference in the whole annulus length in the mid-diastole (Fig. 3).

Comparison of parameters of tricuspid valves with 1 versus 2 posterior leaflets
The mid-systolic data, which did not differ from the intraoperative measurements, were used to compare the long and short diameters, the annulus length of each leaflet and the area of each leaflet between patients with one or two posterior leaflets (Table 3 and Fig. 4).
Six of 28 patients (21%) with one posterior leaflet had severe TR, whereas 11 of 23 patients (48%) with two posterior leaflets had severe TR. Patients with two posterior leaflets had a worse grade of TR before the operation (P = 0.046). In addition, more patients with two posterior leaflets had pulmonary hypertension (P = 0.034).
In patients with one posterior leaflet, the proportion of the anterior, posterior and septal leaflet was 35%, 27% and 38% of the total annular length, respectively (Fig. 4A). In patients with two posterior leaflets, the proportion of the anterior, posterior and septal leaflet accounted for 30%, 35% and 35%, respectively (Fig. 4B). The posterior leaflet area was also significantly larger and both the anterior and septal leaflet areas were smaller in patients with two posterior leaflets (posterior: P < 0.001, anterior: P < 0.001 and septal: P = 0.023).

Comparison of parameters between patients with one versus two posterior leaflets according to tricuspid regurgitation severity
We divided the 51 patients into 3 groups according to the preoperative TR grade (< _mild, n = 8; moderate, n = 26; and severe, n = 17) (Fig. 5).
Regarding the proportion of each leaflet by TR grade, there was no statistical difference between TVs with one versus two posterior leaflets. The proportion of TVs with one posterior leaflet did not change by the TR grade. However, the proportion of TVs with two posterior leaflets changed as the TR grade worsened (Fig. 5B).

DISCUSSION
In the present study, the mid-systolic parameters of 3D TEE were correlated with the intraoperative measurements. The ratio of the posterior leaflets' annulus and area to the whole TV in patients with two posterior leaflets was greater than that in patients with one posterior leaflet. In patients with two posterior leaflets with severe TR, the annulus length of the posterior leaflets was greater.

Usefulness and limitation by three-dimensional transesophageal echocardiography image evaluation
The quality and resolution of computed tomography images have improved remarkably, and there are many reports of computed tomography-based evaluation of cardiac structures of the LV system, including the aortic valve or mitral valve [10]. However, compared with valves of the LV system, morphological evaluation of structures of the right ventricular system by contrast-enhanced computed tomography is difficult, and echocardiography is commonly used for preoperative evaluation. In particular, real-time 3D TEE imaging for preoperative examination of the TV morphology has attracted attention [7]. A percutaneous catheter-based procedure with an edge-to-edge clip technique for severe TR was recently established [11], and the importance of evaluating the TV morphology has been indicated. However, no software is available to measure the TV structure, and it is difficult to estimate the TV morphology preoperatively. In this study, we used the QLAB MVN/3D quantification software package to evaluate the anatomy of the TV by real-time 3D TEE imaging according to a report by Utsunomiya et al. [7]. Only the total length and area of the TV annulus could be measured by this software; it was difficult to measure the diameter and area of each leaflet. Therefore, we measured the perivalvular diameter and area of each leaflet by tracing the 3D echo image. It would be difficult to accurately measure the circumferential diameter of each leaflet in a plane on the 3D echo image. However, when compared with the results of the intraoperative measurements, there was no significant difference in the diameters of the leaflets in mid-systole, suggesting that this technique could be useful for preoperative evaluation. The reason for the larger difference in diastolic parameters might be that it is difficult to identify the TV commissure during diastole, and it was therefore not possible to accurately measure the respective valve leaflet diameters.
Morphological diversity of the TV, in which more than one posterior leaflet exists, has been confirmed. Sakon et al. [3] reported that among 100 normal autopsied hearts, 42% of TVs had two posterior leaflets and 5% of TVs had three posterior leaflets. In the present  study, 29 patients (48%) had two posterior leaflets, suggesting that a similar percentage of patients with TR also had multiple posterior leaflets.
We could predict the number of posterior leaflets with 85% probability by preoperative 3D TEE imaging. The most difficult cases to evaluate the number of leaflets were those with a small RA dimension. In patients who have TR with a small RA dimension, the TV annulus is also small and the 3D structure is preserved, which may complicate image construction and judgement of the TV commissure.

Influence of tricuspid valve morphology on regurgitation
We examined patients in whom the number of posterior leaflets was correctly assessed to determine whether there was a difference in the annulus diameter and area depending on the number of posterior leaflets. There was a significant difference in the ratio of each leaflet to the whole annulus between TVs with one and two posterior leaflets, and patients with two posterior leaflets had a larger posterior leaflet annulus length and area. Furthermore, the proportion of posterior leaflets was larger in patients with two posterior leaflets as the TR grade increased. Despite the similar annulus diameter in both patients with one and two posterior leaflets, the TR grade was more severe in patients with two posterior leaflets, which indicated that the presence of multiple posterior leaflets might cause worsening of TR. This may be due to an increase in the number of valve leaflets, which may lead to an increase in the number of valve leaflet coaptations and a change in the ratio of the sizes of the anterior, posterior and septal leaflets, in turn causing an unbalanced form of the TV.

Surgical implications
We consider that selecting an appropriately sized annuloplasty ring and performing annuloplasty without deforming the TV annulus are important for accurate TV repair.
In this study, the number of TV leaflets was diverse and the size of each leaflet differed between TVs with one and two posterior leaflets. We also found the possibility that TVs with two posterior leaflets were easier to worse TR grade than TVs with one posterior leaflet.
In the clinical setting, annuloplasty rings or bands are used without considering the characteristics of TV morphology. The measuring points when selecting a ring size differ by ring type, and the marked lines on the rings used to suture the commissure area are inconsistent by ring type. Therefore, selecting a ring size by only measuring the anterior leaflet area might underestimate the ring size, especially in patients with two posterior leaflets (Fig. 6). Figure 5A indicates that the annulus length of septal leaflet has no difference between patients with one and two posterior leaflets by the TR grade, even though there were statistically significant differences in the annulus length of the posterior and anterior leaflets between patients with one and two posterior leaflets. Annulus length of septal leaflet might be useful as a decision-making element in case of TVs with two posterior leaflets. The TV is classified as a septal leaflet and a mural leaflet. Equal plication of the mural leaflet and part of the septal leaflet is useful especially in cases of two posterior leaflets not to deform mural leaflet.
By knowing in advance that there are two leaflets, the plication method for tricuspid annulus can be determined to some extent.
In addition, a method of fitting the marked line of the ring to the anteroposterior commissure might deform the TV shape when the annuloplasty ring is used in patients who have a TV with two posterior leaflets. To prevent the deformation of the TV for TV repair, it is necessary to recognize the morphological features of the TV.

Limitations
Our study has several limitations. First, it involved a small number of patients in a single facility. This issue should be addressed in future studies containing larger numbers of patients. Second, high-quality 3D TEE images are required to properly evaluate the number of leaflets and commissure area. For this reason, improvement of examination techniques and equipment is important. Third, the measurement method in this study was tracing the plain-view 3D image of the TV. The results were tolerated statistically, but there might have been a difference from the actual intraoperative measurements. Fourth, we did not measure the coaptation area but instead measured the leaflet area, which could be seen on the 3D image. Therefore, we should investigate the TV area using images of slices at various angles to accurately evaluate the whole leaflet size. Finally, measurement results might have differed between researchers. Our measurements Figure 6: Cases of one and two posterior leaflets with the same tricuspid valve diameter. When we measured the anterior leaflet as a reference, a 28-mm annuloplasty ring was selected in valves with one posterior leaflet and a 26-mm annuloplasty ring was selected in valves with two posterior leaflets.
were supervised by a cardiologist. Two or more researchers should perform several evaluations.

CONCLUSION
About half of patients with TR had a TV with two posterior leaflets, and the features of such TVs were a larger annulus and area of the posterior leaflets. Preoperative 3D TEE is useful for morphological evaluation and accurate TV repair.