Data of all patients was recorded including preoperative transthoracic echocardiography, tricuspid regurgitation jet area, tricuspid annular plane systolic excursion (TAPSE), left ventricular ejection fraction (LVEF), heart chamber size, pulmonary artery pressure, etc. Tricuspid regurgitation was classified as 4 subtypes based on the ratio of TR jet area/right atrium (RA) area [5]: normal, mild (< 0.2), moderate (0.2–0.33) and severe (> 0.33). All patients were followed up with TTE periodically. The primary end-point was the recurrence of significant tricuspid regurgitation (defined as moderate and severe regurgitation).

The median sternotomy was selected for all patients. The cardiopulmonary bypass was done regularly, artery tube for ascending aorta and vein tubes for both superior and inferior vena cava. Myocardial protection was performed through antegrade perfusion of cold blood cardioplegia. Mitral valve was exposed through left interatrial groove pathway and tricuspid valve was exposed through right atrium pathway. The measurement of tricuspid annular circumference must be emphasized. Retractor was gently pulled to expose entire structure of tricuspid valve under cardiac arrest. We have tried to gently lift the atrial wall with two forceps or retractors to expose the entire tricuspid annulus instead of strenuous retraction with the hooks leading to deformation of the annulus and measurement variability. Taken special nonlinear shape of tricuspid valve annulus into consideration, instead of direct measurement, a soft silk line was used to measure indirectly and independently each margin of tricuspid annulus, and then a ruler was used to measure length of the silk line to obtain final size of corresponding margin, after triple attempts, three lengths were combined to calculate tricuspid annular circumference. The photos of the measurement are presented in Fig. 1a–c. Indications of tricuspid annuloplasty were shown as the presence of moderate or severe FTR, significant annular dilation (annulus > 40 mm or 21 mm/m² based on measurement under 4-chamber view), positive intraoperative check confirming the necessity of annuloplasty. Medtronic Contour 3D ring (Medtronic, Minneapolis, Minnesota, USA) and Edwards MC3 ring (Edwards LifeScience, Irvine, CA, USA) were used for tricuspid annuloplasty. Intraannular interrupted suture was generally initiated from center of septal annulus, along with right side of septal, posterior, and anterior annulus, and finally, completed at targeted location, 3–5 mm above anterior-septal commissure. The number of suture was usually 8–10 stitches. All of the patients were rechecked under transesophageal echocardiography to evaluate heart structural improvement after cardiac rehabilitation.

Continuous variables are expressed as the mean ± standard deviation and were analyzed by t test. Categorical variables are expressed as percentages and were analyzed by using the chi-square test. Risk factors of recurrence of significant tricuspid regurgitation were analyzed by binary logistic regression. In univariate analysis, significant factors with P < 0.05 were included in multivariate analysis, forward conditional logistic regression was used to identify risk factors. The optimal cut-off of the TAC/TACI to assess tricuspid annulus dilatation and validation of predictor of recurrent regurgitation were presented as receiver operating characteristic (ROC) curves. A value of p < 0.05 was considered as statistically significance. SPSS 15.0 software (SPSS Inc, Chicago, IL) was used for statistical analysis.

TAC and TACI (TAC index) are shown in Table 3. In order to assess tricuspid annulus dilatation, patients were divided into two groups according to the severity of FTR, group 1 (No significant tricuspid annulus dilatation): normal and mild regurgitation, group 2 (Significant tricuspid annulus dilatation): moderate and severe regurgitation. We found that both TAC and TACI were lager in group 2. Optimal cutoffs were obtained through receiver operating characteristic (ROC) curves analysis to confirm the threshold of annulus dilatation. The TAC with 11.45 cm and TACI with 7.07 cm/m² were considered as the optimal cutoffs defining significant tricuspid annulus dilatation (Table 4 and Figs. 2, 3).

Then the cutoffs (11.45 cm and 7.07 cm/m²) were used in group 1 for tricuspid annulus evaluation and there were 2 (under cutoff of 11.45 cm) and 6 (under cutoff of 7.07 cm/m²) significant tricuspid annulus dilatation patients, respectively.

During follow up longer than 1 year, 15 patients with the significant recurrent tricuspid regurgitation (moderate and severe tricuspid regurgitation) were found, among whom, except for necessary anticoagulation treatment of warfarin and aspirin, no any other medicine was administrated including diuretics.

It was demonstrated by univariable analysis that age, height, weight, BSA, LVEF, length of septal annulus, length of posterior annulus, length of anterior annulus, TAC, TACI, gender, preoperative AF, severity of preoperative FTR and postoperative AF were associated with recurrent tricuspid regurgitation (Table 5).

It was revealed by multivariable analysis that TACI (OR: 8.815, 95% CI: 2.599–25.773, p = 0.001), Postoperative Af (OR: 3.645, 95% CI: 1.078–16.437, p = 0.043) were risk factors of recurrent tricuspid regurgitation. TACI and Postoperative Af were considered as the independent predictor of postoperative recurrent tricuspid regurgitation (Table 6).

Optimal cutoff was obtained through receiver operating characteristic curves analysis to predict recurrent tricuspid regurgitation. The ROC curves showed the optimal TACI cutoff was 7.86 cm/m² (Table 7, Fig. 4).

In accordance with current European and American guidelines, surgical indications of FTR are recommended only based on the size of annulus measured by single 2-dimensional echocardiography under apical four-chamber view [11]. Obviously, this is not comprehensive and more accurate and objective parameters are required. Besides diameter of tricuspid annulus, both area and circumference may be valuable in evaluating tricuspid annular change. Mahia et al. [12], used three-dimensional echocardiography and commercial software to assess association between the tricuspid annular area and the severity of tricuspid regurgitation. Optimal cutoffs of tricuspid annulus dilatation were identified (3DA:10.4 cm², 6.5 cm²/m²). Addetia et al. [9], have tried to use three-dimensional echocardiography to observe tricuspid valve geometry as well as dynamic changes and preliminarily quantify the tricuspid valve parameters. Nevertheless, these findings are not involved in guidelines yet, even though better results are presented. Otherwise, obtained data information is collected based on Western model. Special Chinese data is still lacking. Under this prerequisite, the feasibility of assessment focusing on the association between the TAC and FTR is investigated. Generally, measurement of TAC is processed by 3-d transthoracic echocardiography or transesophageal echocardiography which is more comprehensive in evaluating tricuspid morphological changes. However, all these mentioned methods are cumbersome and special tools must be prepared which means it is difficult to perform measurements in clinical practice. Interestingly, in our study, a simple, repeatable and practical method with a single silk thread was used. The one of main findings in our study is that there is close association between TAC and severity of FTR. Both TAC > 11.45 cm and TACI > 7.07 cm/m² are considered as significant factors of tricuspid annulus dilatation for patients with FTR.

FTR will be worse if dilation of the tricuspid annulus progresses persistently. It has been indicated that during follow up, severity of FTR can be progressed from mild to severe. However, concomitant prophylactic tricuspid annuloplasty during mitral valve surgery is effective in decreasing recurrence of significant FTR at 5-year of follow up [13]. Sakata et al. [5] have also supported that concomitant tricuspid annuloplasty during left heart valve surgery benefits right ventricular geometry. So, tricuspid annuloplasty is recommended urgently and aggressively as excellent surgical strategy against FTR [14].

Surgical procedures of tricuspid annuloplasty varies which depends on severity of regurgitation and surgeons’ experience. A consensus has been reached that prosthetic ring is the preferred option with better postoperative outcome. While, the outcome of flexible or rigid rings is unclear. Nishi et al. [15] indicated that postoperative motion and shape of annulus depended on the special type of prosthetic rings. It has been demonstrated that both mid-term and long term outcome of Edwards and Medtronic prosthetic rings are positive and encouraged [16]. Although, occasionally, adverse events including III degree atrioventricular block and suture dehiscence after tricuspid annuloplasty were reported [17], mostly, concomitant tricuspid annuloplasty is considered as a safe and effective procedure without increasing surgical time and incidence of postoperative complications. Rigid rings were selected in our study, and only 1 patient with postoperative III degree atrioventricular block received implantation of permanent pacemaker. Among remaining patients, no other prosthetic ring related complications were found. Itzhaki Ben Zadok et al. [18] also have demonstrated that concomitant tricuspid annuloplasty is potential and optimal in assessment of long term outcome for patients with rheumatic mitral valve disease.

It has been shown that a few risk factors of the recurrence of significant tricuspid regurgitation after tricuspid annuloplasty, such as severity of tricuspid regurgitation, size of right atrium or ventricle, size of left atrium or ventricle, atrial fibrillation, heart function, permanent pacemaker, leaflet tethering, recurrent mitral regurgitation and pulmonary hypertension [19, 20].

The current study showed that during longer than 1 year of follow up, there were 15 patients presented significant tricuspid regurgitation, which suggested that tricuspid recurrent regurgitation was associated with preoperative tricuspid annular circumference dilation (TACI > 7.86 cm/m²). Possibly, there are 3 reasons: (1) Dilated tricuspid annulus may accompanied with severe damage of tricuspid valve lesion and low right heart function which attribute the presence of recurrent regurgitation. (2) Once tricuspid annulus is dilated, right ventricular remodeling will occur. After tricuspid annuloplasty, original tricuspid valve mismatches altered annulus due to heart remodeling which leads to possible regurgitation. (3) During procedure of tricuspid annuloplasty, the initial stitch is generally targeted at the middle point of septal annulus. Larger and longer tricuspid annulus is not covered by ring completely which also leads to recurrent regurgitation. So, dilatation of tricuspid annulus plays an important role in recurrence of FTR and furthermore, TACI (OR 8.815, 95% CI 2.599–25.773, p = 0.001) is an independent predictor of postoperative recurrent tricuspid regurgitation. In our center, several studies also have been conducted with the aim of exploring the association between tricuspid annular circumference and postoperative tricuspid regurgitation and as a result, there is a close association between the circumference and tricuspid regurgitation. Zhu et al. [21] have recommended that TACI of 83 mm/m² as the threshold of prophylactic tricuspid annuloplasty. Similarly, based on Mohammad Sharif Popal’s recommendation, the threshold is set as 80.2 mm/m² [22]. Although no clear association between the size of ring and postoperative recurrent regurgitation is found in current study, we also observe that the possibility of postoperative regurgitation is greater after implantation of regular rings with large size. Meanwhile, an undersized (size 26 or 28) rigid ring has also been proved to be a potential option for treatment of FTR [23]. At the same time, in this study, it has been demonstrated that postoperative AF (OR 3.645, 95% CI 1.078–16.437, p = 0.043) is also a risk factor of recurrence of regurgitation. Positive association between atrial fibrillation and tricuspid regurgitation is found. From previous clinical experience and literature, ablation of atrial fibrillation is recommended strongly, not only for patients with small left atrium but for patients with large left atrium, which maintains perioperative sinus rhythm and benefits hemodynamic management. Because of the adverse effect of atrial fibrillation on tricuspid regurgitation, tricuspid annuloplasty was performed in all patients with preoperative atrial fibrillation, regardless of severity of regurgitation and annular size.

This study only showed the risk factors of recurrent tricuspid regurgitation, but did not investigate the influence of recurrent tricuspid regurgitation on right heart function or survival rate, and did not provide any strategies dealing with these problems neither. So longer follow up data of these patients is still needed and handled. The sample of patients was relative small and some bias was inevitable. In future, the association between the severity of dilation from local lesions of tricuspid annulus and recurrent regurgitation should be emphazised.