A total of 317 patients with left-heart valvular disease and FTR who underwent left-heart surgery and TA repair between August 2017 and October 2020 in our hospital were enrolled. The surgical indications for TA repair included either moderate and severe tricuspid regurgitation, mild tricuspid regurgitation with annular dilation of ≥40 mm (or ≥21 mm/m²), or mild tricuspid regurgitation with AF. Surgical procedures of the left heart included mitral and aortic valve repair, mitral and aortic valve replacement, combined mitral and aortic valve replacement. If necessary, Cox-Maze IV procedure and coronary artery bypass grafting (CABG) were simultaneously performed in patients with AF and coronary heart disease, respectively. Patients who previously underwent left heart surgery or had right coronary artery disease were excluded. All patients were followed up after discharge.

Preoperative and postoperative echocardiographic data of all patients were recorded using transthoracic or transesophageal echocardiography. The FTR severity was classified into four degrees based on the vena contracta (VC) width in the apical four-chamber plane view and was qualitatively graded as trivial (VC <2 mm, jet area <1 cm²), mild (VC <3 mm), moderate (VC <7 mm), and severe (VC ≥7 mm). Left ventricular ejection fraction and heart chamber diameters were assessed according to the American College of Cardiology/American Heart Association (ACC/AHA) guidelines (14). Pulmonary artery systolic pressure (PASP) was calculated from the FTR jet velocity using the modified Bernoulli equation and was combined with an estimation of the right atrial pressure. FTR recurrence was defined as the presence of moderate or severe FTRs at follow-up.

The surgeries were performed via median sternotomy or right thoracotomy. In patients receiving median sternotomy, cardiopulmonary bypass was performed by arterial cannulation of the ascending aorta and vein cannulation for both the superior and inferior venae cavae. In patients who underwent right thoracotomy, cardiopulmonary bypass was established through the femoral artery and vein cannulation. Myocardial protection was performed through antegrade perfusion of cold blood cardioplegia or del Nido cardioplegia solution. The mitral and tricuspid valves were exposed through the left interatrial groove and right atrium pathways, respectively. The retractor was gently pulled to expose the entire tricuspid valve under cardiac arrest or on a beating heart with full-flow cardiopulmonary bypass. A silk suture was used to measure the length of the TA, and this suture was measured using a ruler. After the intraoperative assessment, tricuspid annuloplasty was performed using the Kay procedure, Edwards MC3 rigid ring (Edwards Life Sciences, Irvine, California, USA), and Sorin Sovering flexible ring (Sorin Biomedica Cardio, Saluggia, Vercelli, Italy) according to the manufacturers' protocol.

Categorical variables were expressed as the number and percentage of patients. Continuous data were presented as either means ± standard deviations if the variables were normally distributed or as medians with interquartile ranges (1st−3rd) if they were non-normally distributed. Fisher's exact test or chi-square test was used to compare categorical variables. The unpaired student's t-test, one-way analysis of variance, Mann–Whitney U test, and Kruskal–Wallis H test were used to compare continuous variables between groups. Factors correlated with preoperative severe FTR and postoperative recurrence of FTR were analyzed using binary logistic regression and Cox proportional hazards regression analysis. In univariate analysis, significant factors (p < 0.1) were included in the multivariate analysis, and forward conditional logistic regression was used to identify the risk factors. The validation of the prediction model was presented as receiver operating characteristic (ROC) curves. Kaplan–Meier (KM) survival analysis was performed to investigate the impact of the length of TA on postoperative FTR recurrence. All statistical analyses were performed using statistical package of social sciences (version 26.0; SPSS Inc., Chicago, Illinois, USA) and Prism 8.2.1 (GraphPad Software Inc., CA, USA) software, and a p-value of < 0.05 was considered statistically significant.

A total of 317 (132 male, 41.6%) patients with an average age of 57.67 ± 11.42 years were included in the study. They were divided into three groups according to the FTR severity (group 1, 139 patients with mild FTR; group 2, 126 with moderate FTR; group 3, 52 with severe FTR). Preoperative and operative data are summarized in Table 1. The demographic data, including the gender, height, weight, body mass index, and body surface area (BSA), were similar between the groups; however, patients in groups 2 and 3 were significantly older (group 1: 54.98 ± 10.46 years; group 2: 58.52 ± 11.80 years, group 3: 62.81 ± 11.07 years, p < 0.01). Patients in groups 2 and 3 had higher proportion of class III–IV of New York heart association (NYHA) classification; a greater proportion of AF; and higher levels of N-terminal pro-brain natriuretic peptide (NT-BNP), total bilirubin, and creatinine than those in group 1. In the analysis of echocardiographic parameters (Table 1), patients in groups 2 and 3 had significantly larger dimensions of the heart chamber and more dilated TA, PASP, and lower tricuspid annular plane systolic excursion (TAPSE). The etiologies of left heart disease were similar between the groups. The procedures of left heart surgery included combined mitral and aortic valve replacement (n = 107), mitral valve replacement (n = 167), mitral valve repair (n = 45), aortic valve replacement (n = 9), CABG (n = 4), and Cox-Maze IV (n = 5). The methods of TA repair included the Kay procedure and tricuspid annuloplasty using Sorin Sovering flexible ring and Edwards MC3 rigid ring. One patient each in group 2 and group 3 expired due to refractory heart failure on postoperative day 4 and cerebral hernia on postoperative day 1, respectively. The remaining patients were followed up until November 1, 2021, and the median follow-up time was 18.90 (11.39–26.03) months. FTR recurrence was observed in 27 patients (8.5%).

Intraoperative TA data are summarized in Table 2. We observed that each TA segment dilated with an increase in FTR severity (Figure 1). In the course of TA dilation from mild to severe FTR, the anterior annulus dilated ≈13.4%, the posterior annulus dilated ≈17.0%, and the septal annulus dilated ≈11.4%. Considering the difficulty in identifying the anteroposterior commissure in some cases, the ratio between the anteroposterior and septal annuli was calculated, which remained stable during the course of TA dilation. All these findings strongly suggested that each TA segment including septal annulus dilated in the course of TA dilation.

Age, preoperative AF, NYHA classification, NT-BNP, total bilirubin, creatinine, indices of the right ventricle (RV) and TA geometry, PASP, left atrium (LA) dimension, and etiology of mitral valve disease were found to be highly associated with severe FTR on univariate analysis (Table 3). Multivariate analysis revealed that the tricuspid annulus circumference index (TACI) [odds ratio (OR) 1.064, p < 0.01)], NYHA classification (NYHA III: OR 3.027, p < 0.01; NYHA IV: OR 9.410, p < 0.01), preoperative AF (OR 6.018, p < 0.01), and preoperative mitral stenosis (OR 3.165, p < 0.01) were independent predictors of severe preoperative FTR (Table 3). Using ROC curves, we observed that the sensitivity and specificity for predicting preoperative severe FTR were 66.0 and 78.0%, respectively, for a TACI of >72.95 mm/m² (area under the curve [AUC]: 0.714, 95% confidence interval [CI]: 0.627–0.802, p < 0.01). Moreover, when the TACI was combined with NYHA classification, preoperative mitral stenosis, and preoperative AF, the sensitivity and specificity for predicting preoperative FTR increased to 78.0 and 84.5%, respectively (AUC: 0.871, 95% CI: 0.815–0.926, p < 0.01) (Figure 2).

Univariate analysis identified 15 variables that correlated substantially with FTR recurrence during follow-up (Table 3). Interestingly, neither the methods of tricuspid repair nor size of the tricuspid ring correlated with FTR recurrence. Of these variables, preoperative severe FTR (OR 9.494, p < 0.05) and the TACI (OR 1.045, p < 0.01) independently correlated with FTR recurrence in Cox regression analysis. Patients were grouped into tertiles (T) based on the length of each annulus segment (T1, shortest; T3, longest). KM survival analysis demonstrated no significant difference between T1 and T3 when the patients were grouped based on the length of the anterior or posterior annulus (Figures 3A,B), while a significant difference was detected based on the length of the septal annulus (Figure 3C).

Based on the echocardiographic study, TA demonstrated increased annular area, planarity, and circularity with the development of FTR. According to our measurements, each TA segment dilated with an increase in the FTR severity. Compared with group 1, group 2 demonstrated increased dilation of the anterior, posterior, and septal annuli of ≈4.1, 6.8, and 4.1%, respectively. Similarly, compared with group 2, the anterior, posterior, and septal annuli in group 3 were significantly more dilated by ≈9.0, 9.6, and 6.8%, respectively. We observed that from mild to severe FTR, each TA segment dilated significantly ranging from 10 to 20%. Specifically, the posterior annulus was dilated with the largest ratio of ≈17.0%, followed by the anterior (13.4%) and septal annuli (11.4%). Such tricuspid annulus dilation is considered almost proportionate. Based on our data, we confirmed Deloche's findings that the posterior annulus dilated with the largest ratio in FTR. However, there was no significant difference in the dilation ratio between each TA segment, as reported previously (80, 40, and 20% for the posterior, anterior, and septal annuli, respectively) (5).

The reason for the above findings, especially septal annulus dilation, which are contrary to Deloche's theory, remains unknown. The possible reasons are as follows: first, Deloche's study had a small sample size with unknown inter-individual variations. Moreover, this study was executed with heart immersion in formalin fixation, which was known to cause morphometric changes in the preserved tissue, including the TA (20, 21).

Second, Kawada et al. stated that the posteroseptal commissural point was located on the free wall of the RV rather than on the interventricular septum (11). In addition, according to Seccombe et al. it was not unusual for the septal leaflet to “wrap around” onto the posterior portion of the free wall by 1 cm (22). These studies suggested that only the partial septal annulus was fixed on the fibrous trigones, while the remaining part would easily and synchronously dilate with the free wall of the RV in FTR, which explains the lower degree of dilation than the posterior annulus.

Third, in addition to the muscular part of the septal annulus (the part attached to the free wall of the RV), the fibrous part of the septal annulus (the part attached to the interventricular septum) might have also been affected by the dilation of the RV with pulmonary hypertension caused by any reason (23). Similar to the septal annulus, the fibrous portion of the mitral annulus has traditionally been considered a static part that does not dilate in functional mitral regurgitation. However, it was revealed that the fibrous part of the mitral annulus was proportionately dilated in patients with cardiomyopathy (24) and in animal models of biventricular failure (25). Therefore, it is not unreasonable to expect that the septal annulus of the tricuspid valve would dilate in FTR and a dilated RV.

Fourth, differences among the races and disease spectra should be considered. To the best of our knowledge, the studies which proposed notions contrary to Deloche's theory and Carpentier's technique were predominantly from Japan (10). Kondoh and colleagues introduced a new ring annuloplasty in the case of an extremely dilated septal annulus of 60 mm (10). Isomura et al. proposed a technique of septal plication in patients with FTR and significantly improved the intraoperative repair success rate from 62 to 93% (9). Two studies specified that the markers on tricuspid rings vary among manufacturers and suggested new techniques for tricuspid ring annuloplasty (11, 12). The underlying explanation remains unknown and requires further investigation.

The severity of FTR fluctuates greatly and is affected by various factors, such as the volume of systemic circulation, heart rate and rhythm, conscious state, and pulmonary artery pressure. Based on the European Society of Cardiology/European Association for Cardiothoracic Surgery and ACC/AHA guidelines, only a diameter of TA ≥40 mm or TA/BSA >21 mm/m² is considered an indication for tricuspid repair in patients undergoing left heart surgery (14, 26). Thus, it is evident that echocardiographic evaluation of FTR is not comprehensive and accurate, and reliable assessment of FTR continues to be a challenge. Our study reported that the TACI was an independent risk factor for preoperative FTR and postoperative FTR recurrence. In a study by Xu Meng et al., a tricuspid annulus circumference (TAC) of >11.45 cm and a TACI of >7.07 cm/m² were considered important factors of TA dilation in FTR (27). Our study concluded that a TAC of >11.35 cm and a TACI of >72.95 mm/m² were the risk factors for preoperative severe FTR. Thus, we highlight the importance of the measurement of TA intraoperatively, which could provide valuable and accurate evaluation of FTR severity.

Several factors have been considered to correlate significantly with postoperative FTR recurrence, such as preoperative severity of FTR, postoperative AF, size of the right atrium (RA) or RV, heart function, permanent pacemaker implantation, pulmonary hypertension, leaflet tethering, and rheumatic etiology of left heart disease (28, 29). This study reported that the preoperative severity of FTR and TACI independently correlated with postoperative FTR recurrence. Moreover, we observed that the septal annulus could dilate in FTR, and its length displayed higher sensitivity for predicting postoperative FTR recurrence than that of the anterior or posterior annulus. The possible reasons for such a result might be: (1) TA dilation usually accompanies enlargement of the right heart and decreased RV function. The longer the TAC is, the larger the right heart becomes, and the RV function worsens, which is attributed to recurrent regurgitation; (2) the entire posterior and major anterior annuli are attached to the free wall of the RV, which is prone to dilation when RV overload occurs. However, when the septal annulus is partially attached to the interventricular septum, it is less prone to dilation. Once the septal annulus is dilated, it is more likely that RV remodeling occurs and deterioration of RV starts, which explains why the length of the septal annulus displayed higher sensitivity for predicting postoperative FTR recurrence.

The concept of tricuspid dilation is as follows: dilation of the tricuspid annulus chiefly occurs in the posterior and anterior annuli. Currently available tricuspid rings have markers indicating the suturing position of the anteroposterior and septoposterior commissures. Such rings are designed especially for posterior annulus plication; however, based on our data, which is in accordance with other studies, we observed that each TA segment could dilate in FTR. These findings have significant clinical implications, as the current tricuspid ring might not entirely support the septal annulus, resulting in FTR recurrence. Thus, we highlight the potential benefits of septal annulus plication. We previously reported our 3-J technique for tricuspid annuloplasty to improve the FTR outcome (30). The early results of our 3-J technique were favorable, and further research is in progress to confirm our concept.

First, this was not a randomized controlled study; whether a concomitant tricuspid repair should be performed in patients with trivial or mild FTR while undergoing left heart surgery was decided by surgeons, which would cause selection bias. Second, not all patients were entirely followed up due to poor medical conditions of some patients who cannot return for a follow-up and no available international medical data-sharing platform. Finally, echocardiographic evaluation of the right heart and TV function was not comprehensive, as some parameters such as RA area, RA volume, and TV tenting height and area, were not routinely measured in our center.

Each TA segment could dilate with FTR severity (from trivial or mild to severe FTR) ranging from 10 to 20%, which varies from the previously reported results. The TACI is independently associated with severe preoperative FTR and postoperative FTR recurrence. The length of the septal annulus displays higher sensitivity to postoperative FTR recurrence than the anterior or posterior annulus. Based on our findings, septal plication should be performed during tricuspid annuloplasty to obtain favorable long-term results.