Persistent Tricuspid Regurgitation After Tricuspid Annuloplasty During Redo Valve Surgery Affects Late Survival and Valve-Related Events

Transcription

1 Circulation Journal Official Journal of the Japanese Circulation Society Advance Publication by-j-stage Persistent Tricuspid Regurgitation After Tricuspid Annuloplasty During Redo Valve Surgery Affects Late Survival and Valve-Related Events Naoto Fukunaga, MD; Yukikatsu Okada, MD, PhD; Yasunobu Konishi, MD; Takashi Murashita, MD; Tadaaki Koyama, MD, PhD Background: Long-term survival and incidence of late tricuspid regurgitation (TR) were studied in patients who underwent tricuspid annuloplasty (TAP) during redo valve surgery. Methods and Results: We retrospectively analyzed 125 patients (mean age, 64.5±10.4 years) who underwent TAP using suture (n=54, group S) or ring (n=71, group R) implantation during redo surgery in a 20-year period. There was a significant difference in prevalence of preoperative atrial fibrillation (P=0.0199). More group S patients were in New York Heart Association functional class III or IV than group R patients (P=0.0066). Mean follow-up was 6.6±5.3 years. Mortality rate for group S was 9.3%, and for group R, 7.0% (P=0.6508). Survival at 5 and 10 years was 72.2±6.6% vs. 88.1±4.0%, and 66.4±7.3% vs. 61.0±7.2%, respectively (log-rank, P=0.7235). Less than moderate TR (hazard ratio [HR], 0.113; P=0.0198) before discharge was a predictor of late survival. There was no statistically significant difference in freedom from valve-related events (log-rank, P=0.5196). A predictor of freedom from valve-related events was less than moderate TR before discharge (HR 0.428; P=0.0100). It was also a positive predictor of freedom from late TR more than mild (HR 0.070; P<0.0001). Conclusions: Less than moderate TR before discharge after TAP during redo valve surgery was an independent risk factor for better long-term outcome. Key Words: Long-term outcome; Redo valve surgery; Tricuspid valve annuloplasty Some reports have described the late results and durability of tricuspid valve annuloplasty (TAP) for tricuspid regurgitation (TR) in the setting of an initial surgery. McCarthy et al reported a 5-year survival rate of 65% and a 97% rate of freedom from reoperation at 8 years. 1 Guenther et al found that the 10-year survival and freedom from reoperation after TAP with a prosthetic ring were 46% and 98%, respectively. 2 Given that the number of mitral valve operations is increasing, careful attention has been paid to functional TR. The number of TAP is also increasing because it has become common during initial mitral valve surgery. In the setting of redo valve surgery, however, TAP for TR has not been universal. Severe isolated TR was recognized in only 6.2% of the patients who underwent left-sided heart valve surgery. 3 Few publications have described results of late TR during redo surgery. TR tended to be ignored during redo surgery compared to the initial surgery because of a shortage of positive or negative data concerning TR in that setting. Also, redo surgery for late TR is associated with a high hospital mortality rate and poor survival, which may lead to avoidance of surgery to treat TR. 4 We reviewed our 20-year surgical experience with TAP performed to alleviate TR during redo surgery. We focused on the long-term survival and incidence of recurrent or progression of late TR in patients who had undergone TAP. Methods This study was approved by the institutional review board at Kobe City Medical Center General Hospital. We retrospectively analyzed 125 patients who underwent TAP using suture (n=54; group S) or ring (n=71; group R) implantation techniques during redo valve surgery over a 20-year period (January 1990 December 2010). During the same period, tricuspid valve replacement (TVR) was performed in 13 cases. The indications Received May 19, 2014; revised manuscript received August 17, 2014; accepted August 18, 2014; released online October 1, 2014 Time for primary review: 35 days Department of Cardiovascular Surgery, Kobe City Medical Center General Hospital, Kobe, Japan This study was presented orally at the 7 th Biennial Congress (Society of Heart Valve Disease) held on June 2013, Palazzo del Casino, Venice, Italy. Mailing address: Naoto Fukunaga, MD, Department of Cardiovascular Surgery, Kobe City Medical Center General Hospital, Minatojimaminamimachi, Chuo-ku, Kobe , Japan. naotowakimachi@hotmail.co.jp ISSN doi: /circj.CJ All rights are reserved to the Japanese Circulation Society. For permissions, please cj@j-circ.or.jp

2 FUKUNAGA N et al. Table 1. Clinical Patient Profile (n=125) Variable Group S (n=54) Group R (n=71) P-value Female 38 (70) 53 (75) Age (years) 62.1± ± Hypertension 9 (17) 17 (24) Hyperlipidemia 3 (6) 5 (7) Diabetes mellitus 7 (13) 10 (14) COPD 2 (4) 3 (4) PAD 3 (6) 1 (1) Stroke 14 (26) 14 (20) Liver cirrhosis 5 (9) 3 (4) Hemodialysis 2 (4) 1 (1) CAD 1 (2) 3 (4) Arial fibrillation 48 (89) 51 (72) Pulmonary hypertension 15 (28) 29 (41) NYHA FC III or IV 33 (61) 26 (37) EuroSCORE 8.7± ± Logistic EuroSCORE 16.2± ± Emergency/Urgent 3 (6) 3 (4) Second or more redo 11 (20) 18 (25) Operative interval 14.0± ± Data given as mean ± SD or n (%). CAD, coronary artery disease; COPD, chronic obstructive pulmonary disease; FC, functional classification; NYHA, New York Heart Association; PAD, peripheral artery disease. for TVR included structural valve deterioration in the tricuspid position, organic TR or redo surgery for tricuspid valve. Of these 13 patients, 3 died of multi-organ failure postoperatively (mortality rate, 23.1%), and there were also 2 late deaths (unknown cause, n=1; heart failure, n=1). There were no valverelated events during the follow-up period. Data were collected from the patients medical records. Data analysis was performed according to the Society of Thoracic Surgeons guideline. 5 Mean patient age was 64.5±10.4 years (range, years). Among the 125 patients, 91 (72.8%) were women. Also, 27 (21.6%) of the 125 patients were >70 years of age. Clinical patient profile is listed in Table 1. There were more elderly patients in group R than in group S (P=0.0871). There was also a significant difference in the number of patients with preoperative atrial fibrillation (P=0.0199). A total of 59 patients (47.2%) were at New York Heart Association (NYHA) functional class III or IV, with more patients at this level in group S than in group R (P=0.0066). There was no statistically significant difference between the 2 groups with regard to the value of the EuroSCORE or the logistic EuroSCORE. In all, 29 patients (23.2%) underwent redo heart valve surgery more than once. Table 2 lists the preoperative transthoracic echocardiographic data. TR was graded as follows: 0, no regurgitation; 1+, mild regurgitation; 2+, moderate regurgitation; 3+, moderately severe regurgitation; 4+, severe regurgitation. Completeness of predischarge transthoracic echocardiography was 93.7%. The left ventricular end-systolic dimension was likely to be smaller in group R than in group S (P=0.0690). In both groups, the degree (severity) of TR was more than moderate preoperatively, although it was more severe in group R than in group S (P=0.0030). Patients who underwent heart valve surgery were followed up every 6 12 months at the outpatient clinic. We were responsible for all decision-making regarding the timing of redo surgery. We also performed telephone interviews with the patients who could not visit the hospital. The mean follow-up period was 6.6±5.3 years (range, years). The follow-up rate for late outcome when the present database was created was 100%. Statistical Analysis All statistical analysis was conducted with StatView version 5.0 (SAS Institute, Cary, NC, USA). Categorical variables were analyzed using the chi-squared test and are expressed as percentages. Continuous variables were analyzed on Student s t-test or unpaired t-test and are expressed as mean ± SD. The Kaplan- Meier method was applied to calculate estimates of long-term survival and freedom from postoperative complications. Univariate analysis was performed with Fisher s exact probability and t-tests. Variables with univariate P 0.2, but which failed to meet the level of statistical significance, were submitted for multivariate logistic regression analysis to determine the independent multivariate factors pertaining to hospital mortality and morbidity. Risk factors for long-term survival were evaluated on Cox regression analysis. Results History of Valve Surgery Details of previous and current surgeries are listed in Table 3. Isolated mitral valve surgery or mitral valve plus other valve surgery accounted for approximately 90% of the cases in both groups. None of the present patients underwent previous isolated tricuspid valve surgery. The main indications for redo valve surgery included structural valve deterioration in the aortic or mitral position (n=100), paravalvular leakage (n=12), TR (n=6), and others (n=7). The mean interval between previous and current redo surgery was 15.4±7.7 years (range, years). Indications for TAP in the setting of redo valve surgery included more than

3 Persistent TR After TAP in Redo Valve Surgery Table 2. Serial Transthoracic Echocardiographic Data Variable Group S (n=54) Group R (n=71) P-value Before operation LVEF (%) 58.2± ± LVEDd (mm) 48.0± ± LVESd (mm) 32.4± ± LAD (mm) 54.9± ± RVP (mmhg) 47.9± ± Degree of TR 2.2± ± (26) 7 (10) (31) 11 (15) 3+ 3 (6) 5 (7) (37) 48 (68) Before discharge LVEF (%) 55.5± ± LVEDd (mm) 47.2± ± LVESd (mm) 33.6± ± LAD (mm) 51.0± ± RVP (mmhg) 39.1± ± Degree of TR 1.2± ± (77) 56 (79) 2+ 9 (17) 11 (15) 3+ 1 (2) 2 (3) 4+ 2 (4) 2 (3) Final follow-up LVEF (%) 55.8± ± LVEDd (mm) 49.2± ± LVESd (mm) 33.7± ± LAD (mm) 57.0± ± RVP (mmhg) 45.9± ± Degree of TR 2.0± ± (41) 47 (66) (28) 15 (21) 3+ 5 (9) 3 (4) (22) 6 (9) Data given as mean ± SD or n (%). LAD, left atrium diameter; LVEDd, left ventricular end-diastolic diameter; LVEF, left ventricular ejection fraction; LVESd, left ventricular end-systolic diameter; RVP, right ventricular pressure; TR, tricuspid regurgitation. mild TR, history of right heart failure, or pulmonary hypertension. A history of right-sided heart failure depended on whether the patients had physical signs such as neck vessel distension, jaundice, ascites or leg edema or not. The diagnosis of right heart failure by cardiologist was also included. We defined pulmonary hypertension as systolic pressure 50 mmhg. Therefore, even in patients with less than moderate TR, TAP was done for those with indications other than TR severity. Regarding suture-type TAP, the De-Vega technique was used in 29 patients (54%) and the Kay technique in 25 (46%) until December After that, TAP with a prosthetic ring was performed predominantly during the study period up to January A flexible ring or band was used based on surgeon preference and experience with mitral valve repair. Rigid rings were not used. Size selection of a band or ring was based on patient body surface area, with a 29-mm band or ring in men and a 27-mm band or ring in women. Concomitant mitral valve surgery accounted for 90% of the surgery in group S (P=0.0013). There was no concomitant aortic valve surgery in that group. Approximately 30% of the operations in group S were reoperation on the tricuspid valve (P=0.0028). In group R, there were 6 concomitant aortic valve operations. For atrial fibrillation, the maze procedure was used in 9 patients. Because most cases of atrial fibrillation were in the chronic phase, we did not perform surgical intervention at redo surgery. Isolated TAP with a prosthetic ring was done in 6 patients in the setting of redo surgery. The duration of cardiopulmonary bypass tended to be longer in group R than in group S. Early Outcome Overall hospital mortality was 8.0% (10/125). The mortality rate for group S was 9.3% (5/54), and for group R, 7.0% (5/71; P=0.6508). The causes of early death were multi-organ failure (n=5), hepatic failure (n=1), low-output syndrome (n=1), cerebral hemorrhage (n=1), sepsis (n=1), and arrhythmia (n=1). Major morbidity postoperatively occurred in 31 patients (24.8%). The morbidity rate for group S was 24.1% (13/54)

4 FUKUNAGA N et al. Table 3. Surgery History Variable Group S (n=54) Group R (n=71) P-value Previous valve surgery Isolated aortic 1 (2) 5 (7) Mitral 33 (61) 48 (68) Tricuspid 0 0 Aortic+mitral 1 (2) 8 (11) Aortic+tricuspid 0 0 Mitral+tricuspid 15 (28) 7 (10) Aortic+mitral+tricuspid 4 (7) 2 (3) Aortic valve surgery 6 (100) 15 (100) Repair 0 (0) 1 (7) Replacement 6 (100) 14 (93) Mitral valve surgery 53 (100) 65 (100) Repair 16 (30) 29 (45) Replacement 37 (70) 36 (55) Tricuspid valve surgery 19 (100) 9 (100) Repair 19 (100) 9 (100) Replacement 0 (0) 0 (0) Current concomitant surgery Aortic 0 6 (8) Mitral 50 (93) 49 (69) Aortic+mitral 4 (7) 10 (14) Aortic valve surgery 4 (100) 16 (100) Repair 1 (25) 0 (0) Replacement 3 (75) 16 (100) Mitral valve surgery 54 (100) 59 (100) Repair 4 (7) 6 (10) Replacement 50 (93) 53 (90) Isolated Tricuspid 0 6 (8) Reoperative Tricuspid valve 19 (35) 9 (13) Aorta cross-clamp time (min) 120.2± ± CPB time (min) 189.3± ± Data given as mean±sd or n (%). ASD, atrial septal defect; CABG, coronary artery bypass grafting; CPB, cardiopulmonary bypass. Table 4. Multivariate Risk Factors for Hospital Death Variable OR 95% CI P-value Small LVEDd Ventilator duration CI, confidence interval; LVEDd, left ventricular end-diastolic dimension; OR, odds ratio. Late Outcome Overall Survival Overall survival at 5, 10 and 15 years was 80.8±3.8%, 65.8±5.5%, and 54.8±7.5%, respectively. There were 26 late deaths during the follow-up period. Among them, cardiac-related death was identified in 18 (69.2%). Survival at 5 and 10 years for groups S and R were 72.2±6.6% vs. 88.1±4.0% and 66.4±7.3% vs. 61.0±7.2%, respectively (logrank, P=0.7235; Figure 1). On Cox regression analysis less than moderate TR before discharge (hazard ratio [HR], 0.113; 95% CI: ; P=0.0198) was a protective predictor of late survival. In this study, predischarge pulmonary hypertension was not a statistically significant predictor. Freedom From Valve-Related Events Freedom from valveand for group R, 25.4% (18/71; P=0.8698). The duration of ventilator support and intensive care unit stay were 72.6±159.3 h and 7.7±12.6 days, respectively. On logistic regression analysis, preoperative small left ventricular end-diastolic dimension (odds ratio [OR], 0.565; 95% confidence interval [CI]: ; P=0.0356) and duration of ventilator support (OR, 1.013; 95% CI: ; P=0.0158) were independent risk factors for hospital mortality. Predischarge pulmonary hypertension did not reach statistical significance on univariate analysis (Table 4). The degree of TR before discharge was 1.2±0.8 in group S and 1.2±0.7 in group R (P=0.5288). The right ventricular pressure at discharge was 39.1±12.0 mmhg in group S and 40.8±11.6 mmhg in group R (P=0.6639). There were no statistically significant differences for other items (Table 2). There were 27 patients with TR grade more than mild at discharge. On logistic regression analysis, second or more redo valve surgery (OR, 3.582; 95% CI: ; P=0.0418) was a risk factor for residual TR more than mild at discharge. In this analysis, left ventricular function or pulmonary pressure did not affect predischarge TR grade.

5 Persistent TR After TAP in Redo Valve Surgery Figure 1. Late survival between group S (suture implantation) and group R (ring implantation). Figure 2. Freedom from valve-related events. Group R, ring implantation; group S, suture implantation.

6 FUKUNAGA N et al. Figure 3. (A) Long-term survival and (B) freedom from valve-related events based on predischarge tricuspid regurgitation (TR) grade (more than mild or not). Table 5. Predictors of Late Outcome Variable Predischarge TR <2+ HR 95% CI P-value Survival Freedom from valve events Freedom from late TR < CI, confidence interval; HR, hazard ratio; TR, tricuspid regurgitation. related events at 5, 10 and 15 years was 68.0±4.8%, 47.9±6.0% and 28.2±6.9%, respectively. There were no statistically significant differences between the 2 groups (76.6±6.5% vs. 60.8±6.8% at 5 years and 58.2±8.2% vs. 36.4±9.1% at 10 years, respectively, log-rank, P=0.5196; Figure 2). Thus the only predictor of freedom from valve-related events was less than moderate TR before discharge (HR, 0.428; 95% CI: ; P=0.0100). Survival and Valve-Related Events Based on Predischarge TR Based on the aforedescribed analysis, we analyzed long-term survival and valve-related events based on predischarge TR grade less than moderate or not. Long-term survival rates at 5 and 10 years for the 2 groups were 86.6±3.6% vs. 61.8±10.3% and 76.3±5.4% vs. 17.4±14.0%, respectively (log-rank, P=0.0002; Figure 3A). With regard to freedom from valve-related events, there were statistically significant differences between the 2 groups (72.5±5.2% vs. 52.5±10.7% at 5 years and 67.2±5.7% vs. 14.6±12.6% at 8 years, respectively, log-rank, P=0.0010; Figure 3B). Freedom From Late TR >Mild Late TR was measured on late follow-up echocardiography. Patients with a TR grade more than mild at both discharge and late follow-up were included. The incidence of freedom from late TR more than mild at 5, 10 and 15 years was 70.7±5.0%, 45.3±6.5%, and 20.5±6.1%, respectively. Freedom from late TR more than mild at 5 and 10 years for groups S and R were 68.6±7.6% vs. 71.7±6.8%, and 53.4±8.4% vs. 28.8±10.2%, respectively (log-rank, P=0.3679). Predischarge TR grade less than moderate was a positive predictor of freedom from late TR more than mild (HR, 0.070; 95% CI: ; P<0.0001). Predischarge pulmonary hypertension, left ventricular function or TAP technique did not reach statistical significance. Final Follow-up Echocardiographic Data Completeness of final follow-up echocardiography was 94.8%. The severity of TR at final follow-up was 2.0±0.8 for group S, which was worse than for group R (P=0.0001). Other than the left ventricular end-systolic dimension, there were no statistically significant differences between the 2 groups (Table 2). Discussion We have described here the long-term outcome of patients who underwent TAP using the suture or ring implantation technique in the setting of redo valve surgery. There have been no publications on long-term outcome of TAP alone for redo valve surgery, therefore during this era in which redo surgery is dramatically increasing, the present data will be valuable for surgical decision-making. We did not find any differences between the suture and ring techniques with regard to late outcome, including survival and freedom from events. What is important in this study is that

7 Persistent TR After TAP in Redo Valve Surgery the predischarge severity of TR (more than mild TR) irrespective of the TAP techniques used was an independent risk factor for poor survival, and a negative predictor of freedom from valve-related events and late TR more than mild (Table 5). During the initial surgery, persistent TR after mitral valve replacement has a negative impact on freedom from congestive heart failure and late survival. 6 We found that residual TR following TAP irrespective of TAP techniques during redo valve surgery had a negative impact on late outcome, as described by Ruel et al. 6 Particularly, we must pay careful attention to the patients who undergo second-time or more redo surgery because the number of redo surgeries was a risk factor for predischarge TR more than mild. With regard to selecting a TAP technique for the initial surgery, there have been some studies on early and late results. Previously, we used suture methods, such as De-Vega or Kay techniques, for TR during both the initial and redo operations. The suture techniques were technically difficult, however, and recurrent TR followed TAP using suture techniques. In this study, final follow-up data after redo surgery showed that the severity of TR in group S was significantly worse than that in group R, although echocardiographic data at discharge were similar between the 2 groups. A prosthetic ring technique has recently been developed and is increasingly being used to address TR associated with leftsided heart valve disease. There is also evidence supporting ring annuloplasty over conventional De-Vega suture annuloplasty in terms of TR recurrence and long-term survival. 7 In accordance with current trends, we are alternatively using a prosthetic ring for TR as an initial surgery as described previously. 8 It has also been used in redo valve surgery at Kobe City Medical Center General Hospital. Flexible (not rigid) rings or bands have been used for TAP at Kobe City Medical Center General Hospital because they maintain tricuspid annular motion during the cardiac cycle. 9 Previously, we reported that the mitral annulus with a Duran total ring remained flexible and retained a non-planar configuration at 6 years after implantation. 10 Flexible rings or bands for TAP were originally designed to repair defects causing mitral regurgitation. Based on our experience, we thought that we could achieve tricuspid annular motion following TAP using a flexible ring or band, as has been done for mitral valve repair. We have continued to use a flexible band or ring for TAP. We also recognized the superiority of a flexible band over a rigid ring, regarding the risk of ring dehiscence. 11,12 The size selection of a band or ring is based on the patient s body surface area, in accordance with a report by Fukuda et al. 9 Fujita et al also selected ring size based on body surface area in TR surgery. 13 A 29-mm band or ring is used in men and a 27-mm band or ring in women. Although we performed TAP according to our own indications, there were 27 patients (21.6%) who had more than mild TR before discharge. In specific cases for example, stenotic tricuspid valve, history of previous tricuspid valve surgery, dilated right ventricle, right ventricular dysfunction, or repeat cardiac surgery, tricuspid valve replacement may be an option for managing less than moderate TR. 13,14 Tricuspid valve replacement does not necessarily influence early mortality compared to that related to TAP. 14,15 The factors associated with early death on univariate analysis were NYHA functional class III or IV and left ventricular ejection fraction If a patient who is undergoing TAP during redo surgery is found to have more than mild TR intraoperatively (on transesophageal echocardiography) and they do not have either risk factors noted earlier, it is reasonable to perform valve replacement to manage the TR. Chen et al reported that the 3-year survival rate following tricuspid valve repair during redo surgery was 84.1%, 4 which is significantly better than that after tricuspid valve replacement. Park et al reported that survival rates for TAP patients at 5 and 10 years were 93% and 81%, respectively. 15 These survival rates were better than those in present study. The reason is that their patients were much younger than the present ones. In fact, age has been reported to be a predictor of late mortality. 15 Also, Park et al had a smaller number of patients (n=14) who underwent TAP than in the present study (n=125). There have been no data regarding freedom from valverelated events or late TR after redo valve surgery. Therefore, the present data can be useful to cardiac surgeons and cardiologists for predicting future events during the follow-up period. The present study with its retrospective nature has some important limitations. The first is the large heterogeneous patient group because the present study included all patients who underwent TAP for varying etiologies. The S and R group surgery was performed in different time periods. There was also no uniformity of surgical indications or techniques for addressing TR during TAP in the setting of redo valve surgery, which might have affected the severity of TR and thus subsequent long-term survival or valve-related events. The second is the lack of data on echocardiography such as the right ventricular diameter and dysfunction or tricuspid valve tethering area or height because it was difficult to measure them during the longterm time frames or by the different cardiologists. We did include, however, a relatively large number of patients (n=125) who underwent TAP in the setting of redo surgery and who had a long duration of follow-up (up to 21.6 years). Additionally, an optimal follow-up rate for echocardiography was achieved. We investigated the impact of predischarge TR after TAP on late outcome. As redo valve surgery is increasing, our study provides important information for both cardiac surgeons and cardiologists. In particular, our study must draw attention in the setting of transcatheter valve implantation techniques, because even after successful valve implantation, the presence of TR affects the late outcomes. We must consider conventional redo surgery in patients who may suffer from residual TR after transcatheter implantation. Conclusions Less than moderate TR after TAP (and before discharge) irrespective of the TAP techniques used during redo valve surgery, was an independent risk factor for better survival and freedom from valve-related events and freedom from late TR more than mild. Particularly, in using TAP for patients who undergo second-time and more redo surgery, careful attention should be paid. Acknowledgments The authors state that they have no conflicts of interest. References 1. McCarthy PM, Bhudia SK, Rajeswaran J, Hoercher KJ, Lytle BW, Cosgrove DM, et al. Tricuspid valve repair: Durability and risk factors for failure. J Thorac Cardiovasc Surg 2004; 127: Guenther T, Mazzitelli D, Noebauer C, Hettich I, Tassani-Prell P, Voss B, et al. Tricuspid valve repair: Is ring annuloplasty superior? Eur J Cardiothorac Surg 2013; 43: Izumi C, Miyake M, Takahashi S, Matsutani H, Hashiwada S, Kuwano K, et al. Progression of isolated tricuspid regurgitation late after leftsided valve surgery. Circ J 2011; 75: Chen SW, Tsai FC, Tsai FC, Chao YK, Huang YK, Chu JJ, et al.

8 FUKUNAGA N et al. Surgical risk and outcome of repair versus replacement for late tricuspid regurgitation in redo operation. Ann Thorac Surg 2012; 93: Akins CW, Miller C, Turina MI, Kouchoukos NT, Blackstone EH, Grunkemeier GL, et al. Guidelines for reporting mortality and morbidity after cardiac valve interventions. Ann Thorac Surg 2008; 85: Ruel M, Rubens FD, Masters RG, Pipe AL, Bedard P, Mesana TG. Late incidence and predictors of persistent or recurrent heart failure in patients with mitral prosthetic valves. J Thorac Cardiovasc Surg 2004; 128: Khorsandi M, Banerjee A, Singh H, Srivastava AR. Is a tricuspid annuloplasty ring significantly better than a De Vega s annuloplasty stitch when repairing severe tricuspid regurgitation. Interact Cardiovasc Thorac Surg 2012; 15: Murashita T, Okada Y, Kanemitsu H, Fukunaga N, Konishi Y, Nakamura K, et al. Fate of functional tricuspid regurgitation after mitral valve repair for degenerative mitral regurgitation. Circ J 2013; 77: Fukuda S, Saracino G, Matsumura Y, Daimon M, Tran H, Greenberg NL, et al. Three-dimensional geometry of the tricuspid annulus in healthy subjects and in patients with functional tricuspid regurgitation: A real-time, 3-dimensional echocardiographic study. Circulation 2006; 114(Suppl I): I492 I Okada Y, Nasu M, Kanzaki J. Flexibility of the mitral annulus with the Duran ring at six years post-implantation. J Heart Valve Dis 2002; 11: Zhu TY, Wang JG, Meng X. Is a rigid tricuspid annuloplasty ring superior to a flexible band when correcting secondary tricuspid regurgitation? Interact Cardiovasc Thorac Surg 2013; 17: Pfannmuller B, Doenst T, Eberhardt K, Seeburger J, Borger MA, Mohr FW. Increased risk of dehiscence after tricuspid valve repair with rigid annuloplasty rings. J Thorac Cardiovasc Surg 2012; 143: Fujita S, Arai H, Tomita M, Mizuno T, Kawaguchi S, Manabe S, et al. Proposal of a novel index for selection of optimal annuloplasty ring size for tricuspid annuloplication. Circ J 2013; 77: Jeganathan R, Armstrong S, Al-Alao B, David T. The risk and outcomes of reoperative tricuspid valve surgery. Ann Thorac Surg 2013; 95: Park CK, Park PW, Sung K, Lee YT, Kim WS, Jun TG. Early and midterm outcomes for tricuspid valve surgery after left-sided valve surgery. Ann Thorac Surg 2009; 88: