Mechanical Tricuspid Valve Replacement Is Not Superior in Patients Younger Than 65 Years Who Need Long-Term Anticoagulation

Mechanical Tricuspid Valve Replacement Is Not Superior in Patients Younger Than 65 Years Who Need Long-Term Anticoagulation Ho Young Hwang, MD, PhD, Kyung-Hwan Kim, MD, PhD, Ki-Bong Kim, MD, PhD, and Hyuk Ahn, MD, PhD Department of Thoracic and Cardiovascular Surgery, Seoul National University Hospital, Seoul, Republic of Korea Background. We evaluated long-term results of tricuspid valve replacement (TVR) with mechanical valves and bioprostheses in patients less than 65 years old who needed anticoagulation therapy irrespective of TVR. Methods. Among 156 patients who underwent TVR from January 1996 to April 2010, 119 patients younger than 65 years who needed anticoagulation therapy independent of the TVR were studied. Anticoagulation therapy was needed owing to left-sided mechanical valves (n 11), atrial fibrillation (n 13), or both (n 95). Seventy patients underwent mechanical TVR, and 49 underwent bioprosthetic TVR. The follow-up duration was 68 38 months. Propensity score-adjusted analyses were performed. Results. Early postoperative outcomes were similar between groups, with an operative morality rate of 6.7% (8 of 119). Overall survival rates at 5 and 10 years were 86.4% and 80.3%, respectively. There were no differences in propensity score-adjusted overall survival rates between the two groups (p 0.291). Freedom from the composite endpoint of thrombosis, embolism, and bleeding was lower in the mechanical TVR group than in the bioprosthetic TVR group (76.3% and 70.0% at 5 and 10 years, respectively, versus 97.6% and 97.6%, p 0.020). Five- and 10-year valve-related event-free survivals were also lower in the mechanical TVR group than in the bioprosthetic TVR group, although not statistically significant (75.0% and 61.7% versus 89.2% and 80.3%, p 0.129). Conclusions. Even in younger patients who need anticoagulation therapy irrespective of TVR, mechanical TVR is not superior because of increased occurrence of valve-related events, especially the composite of thrombosis, embolism, and bleeding. (Ann Thorac Surg 2012;93:1154 61) 2012 by The Society of Thoracic Surgeons Tricuspid valve replacement (TVR) is still a challenging procedure with high operative and late mortality [1 4]. Patients who undergo TVR frequently have a history of previous cardiac surgery, atrial fibrillation, advanced symptoms of heart failure, and other cardiac problems needing concomitant correction [5]. Although there might be a consensus that a bioprosthetic valve is a reasonable option in patients older than 65 years, data are scarce whether mechanical TVR is superior compared with bioprosthetic TVR in patients younger than 65 years, especially those in need of life-long anticoagulation therapy due to other reasons such as left-sided mechanical heart valves or the presence of atrial fibrillation. The aim of this study was to evaluate whether mechanical TVR is appropriate in younger patients (less than 65 years old) for whom life-long anticoagulation therapy is needed irrespective of TVR. Accepted for publication Nov 30, 2011. Address correspondence to Dr Ahn, Department of Thoracic and Cardiovascular Surgery, Seoul National University Hospital, 28 Yeongeon-dong, Chongno-gu, Seoul 110-744, Republic of Korea; e-mail: ahnhyuk@ snu.ac.kr. Material and Methods Patient Characteristics The study protocol was reviewed by the Institutional Review Board and approved as a minimal risk retrospective study (approval number H-1106-087-366) that did not require individual consent based on the institutional guidelines for waiving consent. From January 1996 to April 2010, tricuspid valve procedures were performed in 659 patients (tricuspid valvuloplasty or annuloplasty, or both, in 503 patients and TVR in 156 patients) at our institution. Among them, 119 (aged 53.0 8.8 years, 28 men, 91 women) who were younger than 65 years, and needed anticoagulation therapy irrespective of the TVR procedure were enrolled in the present study. Long-term anticoagulation therapy was planned owing to preexisting or concomitantly replaced left-side mechanical valves (n 11), atrial fibrillation (n 13), or both (n 95). Seventy patients underwent mechanical TVR (MTV group) and 49 patients underwent bioprosthetic TVR (BTV group). Three types of mechanical valves were used: the On-X valve (On-X Life Technology, Austin, TX) in 34 patients, the St. Jude valve (St. Jude Medical, Minneapolis, MN) in 17 patients, and the Carbomedics valve (Sulzer Carbomedics, Austin, TX) in 19 patients. Two types of bioprosthesis were implanted: 2012 by The Society of Thoracic Surgeons 0003-4975/$36.00 Published by Elsevier Inc doi:10.1016/j.athoracsur.2011.11.075

Ann Thorac Surg HWANG ET AL 2012;93:1154 61 TVR IN YOUNGER PATIENTS NEEDING ANTICOAGULATION 1155 Table 1. Preoperative Characteristics and Risk Factors of the Study Patients Variables Total (n 119) MTV Group (n 70) BTV Group (n 49) p Value Age, years 53.0 8.8 50.4 8.5 56.8 7.8 0.001 Male/female 28/91 19/51 9/40 0.267 Risk factors, n (%) Smoking 3 (2.5) 1 (1.4) 2 (4.1) 0.568 Diabetes mellitus 16 (13.4) 10 (14.3) 6 (12.2) 0.748 Hypertension 12 (10.1) 5 (7.1) 7 (14.3) 0.229 History of stroke 12 (10.1) 5 (7.1) 7 (14.3) 0.229 BMI 25 kg/m 2 16 (13.4) 11 (15.7) 5 (10.2) 0.386 Dyslipidemia 2 (1.7) 1 (1.4) 1 (2.0) 0.999 COPD 4 (2.5) 1 (1.4) 3 (6.1) 0.305 Chronic renal failure 3 (2.5) 1 (1.4) 2 (4.1) 0.568 Atrial fibrillation 108 (90.8) 61 (87.1) 47 (95.9) 0.120 NYHA class III 59 (49.6) 37 (52.9) 22 (44.9) 0.393 Previous cardiac surgery 92 (77.3) 52 (74.3) 40 (81.6) 0.346 LVEF, % 57 8 57 8 58 9 0.287 BMI body mass index; BTV bioprosthetic tricuspid valve; COPD chronic obstructive pulmonary disease; LVEF left ventricular ejection fraction; MTV mechanical tricuspid valve; NYHA New York Heart Association. the Carpentier-Edwards Perimount (Edwards Lifesciences, Irvine, CA) in 44 patients, and the Hancock II (Medtronic, Minneapolis, MN) in 5 patients. Preoperative characteristics were similar between the two groups, except that patients in the MTV group were younger than those in the BTV group (Table 1). Surgical Procedures and Operative Data Indications for TVR were (1) organic changes precluding valve repair by rheumatic valve disease (n 55) or infective endocarditis (n 2); (2) reoperation after previous tricuspid valve repair (n 22); (3) failed attempt of tricuspid valve repair (n 11); and (4) a wish to avoid a high risk of future reoperation (n 29). Selection of a mechanical or bioprosthetic valve was at the discretion of the attending surgeon, although mechanical valves were preferred in the early study period. All operations were performed using a routine aortobicaval cannulation under the moderate hypothermia and cardioplegic arrest through a median sternotomy. Mean cardiopulmonary bypass and aortic cross-clamp times were 211 64 and 127 49 minutes, respectively. Seventy-one patients underwent concomitant cardiac procedures including mitral valve operations (n 50), aortic valve procedures (n 34), and arrhythmia surgery (n 28). There were no differences in operative data between the two groups (Table 2). Evaluation of Long-Term Clinical Outcomes Patients underwent regular postoperative follow-up through the outpatient clinic at 3-month or 4-month intervals, and were contacted by telephone for confirmation of their condition if the last clinic visit was not conducted at the scheduled time. Dose of anticoagulant was adjusted to obtain target international normalized ratio (INR) of 2.5 in patients with left-side heart valve or atrial fibrillation, or both, and 3.0 in patients with mechanical tricuspid valve. The same level of INR was targeted regardless of the types of mechanical valve. A specialized anticoagulation therapy service team composed of experienced pharmacists regularly followed up all patients. If a patient s INR was not in target ranges, follow-up duration was shortened to even less than 1- week intervals, until the patient obtained target INR. No patient underwent home INR monitoring. Clinical follow-up was closed on July 31, 2011. Follow-up was complete for 95.0% (113 of 119 patients), with a follow-up duration of 68 38 months. Operative mortality was defined as death within 30 days. Valverelated complications were recorded according to the Table 2. Operative Data of the Study Patients Variables Total (n 119) MTV Group (n 70) BTV Group (n 49) p Value ACC time, minutes 127 49 125 48 128 52 0.742 CPB time, minutes 211 64 203 61 223 67 0.097 Concomitant procedure, n (%) 71 (59.7) 40 (57.1) 31 (63.3) 0.503 Mitral valve procedure 50 (42.0) 30 (42.9) 20 (40.8) 0.824 Aortic valve procedure 34 (28.6) 21 (30.0) 13 (26.5) 0.680 Arrhythmia surgery 28 (23.5) 17 (24.3) 11 (22.4) 0.816 ACC aortic cross clamp; BTV bioprosthetic tricuspid valve; CPB cardiopulmonary bypass; MTV mechanical tricuspid valve.

1156 HWANG ET AL Ann Thorac Surg TVR IN YOUNGER PATIENTS NEEDING ANTICOAGULATION 2012;93:1154 61 previous guidelines [6]; the composite endpoint of thrombosis, embolism, and bleeding included occurrence of tricuspid valve thrombosis, right-side embolism, and bleeding that caused death, hospitalization, or permanent injury or necessitated transfusion. Tricuspid valve-related events included the following: (1) tricuspid valve-related mortality; (2) composite of thrombosis, embolism, and bleeding; (3) structural valve deterioration; (4) nonstructural valve dysfunction; (5) reintervention for tricuspid valve; and (6) congestive heart failure requiring readmission. Statistical Analysis Statistical analysis was performed using the SPSS software package (version 12.0; SPSS, Chicago, IL). Data were expressed as mean SD, median with ranges, or proportions. Comparison between the two groups was performed with the 2 test or Fisher s exact test for categorical variables and the Student t test for continuous variables. Multivariable analysis for operative mortality was performed with logistic regression analysis. Survival rates were estimated using the Kaplan-Meier method, and comparisons between groups were performed with the log rank test or Cox regression model. The Cox proportional hazards model was adopted for analysis of risk factors for time-related events. Type of prosthetic valve and other variables with p values of less than 0.1 were entered into multivariable analyses. To overcome baseline differences between the two groups, a propensity score of having bioprosthetic TVR was calculated and adjusted in the analyses. Underlying characteristics considered in estimation of propensity score were age, sex, smoking status, body mass index, diabetes mellitus, hypertension, dyslipidemia, history of stroke, chronic renal failure, New York Heart Association functional class of III or more, presence of left-side mechanical valve, atrial fibrillation, left ventricular ejection fraction, previous cardiac surgery, aortic cross-clamp time, and etiology of valve disease other than tricuspid. The p value of the Hosmer and Lemeshow test and the c statistic of the propensity score model, which reflected calibration and discrimination power of the propensity score model, were 0.831 and 0.852, respectively. A p value of less than 0.05 was considered statistically significant. Table 3. Early Mortality and Postoperative Complications Results Early Results The operative mortality rate was 6.7% (8 of 119); 8.6% (6 of 70 patients) in the MTV group and 4.1% (2 of 49 patients) in the BTV group (p 0.468). Postoperative morbidity included low cardiac output syndrome (n 17, 14.3%), acute renal failure (n 10, 8.4%), respiratory complication (n 8, 6.7%), postoperative bleeding (n 4, 3.4%), stroke (n 2, 1.7%), and mediastinitis (n 1, 0.8%). Complete atrioventricular conduction block occurred in 3 patients (2.5%). No patient underwent prophylactic placement of pacemaker leads at the time of surgery. Those patients underwent permanent pacemaker insertion 5 months, 6 months, and 46 months after surgery. There were no differences in postoperative complications between the two groups (Table 3). In a propensity score-adjusted multivariable analysis, the type of prosthetic valve was not associated with operative death. Longer cardiopulmonary bypass time was the only predictor of operative mortality after TVR (p 0.003; Table 4). Long-Term Survival Among 111 survivors, late death occurred in 11 (9.9%). Tricuspid valve-related death occurred in only 1 patient. That patient underwent TVR with a mechanical valve 15 years after an initial mitral valve replacement due to progression of tricuspid regurgitation; he died of rightside heart failure 38 months after TVR while waiting for heart transplantation. Overall survival rates at 5 and 10 years were 86.4% and 80.3%, respectively. There was no difference in overall survival between the two groups (p 0.350; Fig 1). This difference remained insignificant after propensity score adjustment (p 0.291). Multivariable analysis revealed that smoking was the only significant risk factor for long-term survival (p 0.048). However, after propensity score adjustment, there was no significant risk factor for long-term survival (Table 5). Freedom From Tricuspid Valve-Related Events Tricuspid valve thrombosis occurred in 3 patients in the MTV group 58, 85, and 175 months after surgery. Last INR before the development of thrombosis was 2.12, 2.15, Variables Total (n 119) MTV Group (n 70) BTV Group (n 49) p Value Mortality 8 (6.7) 6 (8.6) 2 (4.1) 0.468 Complications LCOS 17 (14.3) 11 (15.7) 6 (12.2) 0.595 Acute renal failure 10 (8.4) 6 (8.6) 4 (8.2) 0.999 Respiratory complication 8 (6.7) 4 (5.7) 4 (8.2) 0.716 Bleeding reoperation 4 (3.4) 2 (2.9) 2 (4.1) 0.999 Stroke 2 (1.7) 2 (2.9) 0 (0) 0.511 Complete AV block 3 (2.5) 1 (1.4) 2 (4.1) 0.568 Mediastinitis 1 (0.8) 0 (0) 1 (2.0) 0.412 Data are presented as n (%). AV atrioventricular; BTV bioprosthetic tricuspid valve; LCOS low cardiac output syndrome; MTV mechanical tricuspid valve.

Ann Thorac Surg HWANG ET AL 2012;93:1154 61 TVR IN YOUNGER PATIENTS NEEDING ANTICOAGULATION 1157 Table 4. Predictor of Early Mortality in a Propensity Score-Adjusted Multivariable Analysis Univariate Analysis Multivariable Analysis Variables p Value Odds Ratio (95% CI) p Value Age 0.001 0.956 (0.855 1.069) 0.427 Group 0.468 0.333 (0.021 5.214) 0.434 Sex 0.087 2.710 (0.328 22.406) 0.355 Overweight, BMI 25 kg/m 2 0.074 1.201 (0.064 22.702) 0.903 Smoking 0.999...... Diabetes mellitus 0.596...... Hypertension 0.999...... Chronic renal failure 0.190...... COPD 0.999...... Atrial fibrillation 0.551...... NYHA class III 0.163...... Previous cardiac surgery 0.196...... Severity of TR 0.827...... LVEF 0.287...... Concomitant operation 0.472...... ACC time 0.742...... CPB time 0.097 1.025 (1.009 1.042) 0.003 ACC aortic cross clamp; BMI body mass index; CI confidence interval; COPD chronic obstructive pulmonary disease; CPB cardiopulmonary bypass; LVEF left ventricular ejection fraction; NYHA New York Heart Association; TR tricuspid regurgitation. and 2.37 in those patients, respectively. Two patients underwent thrombolytic therapy, and 1 patient underwent tricuspid valve re-replacement. Bleeding events occurred in 13 patients in the MTV group and 1 patient in the BTV group at 29 19 months (range, 1 to 62 months) after surgery. There were 6 cases of intracranial hemorrhage; 6 cases of gastrointestinal bleedings and 2 cases of intraabdominal hematoma needed packed red blood cell transfusions. The INRs were greater than 3.0 in 6 of the 13 patients in the MTV group and 3.51 in the patient in the BTV group. No patient had tricuspid valve-related embolism during follow-up. Freedom from the composite endpoint of thrombosis, embolism, and bleeding events at 5 and 10 years was 84.0% and 78.3%, respectively, and was lower in the MTV group than in the BTV group (76.3% and 70.0% versus 97.6% and 97.6%, respectively; p 0.008; Fig 2). The differences were also significant in a propensity score-adjusted multivariable analysis (p 0.020; Table 6). Structural valve deterioration occurred in 2 patients in the BTV group; moderate to severe and severe tricuspid regurgitation were detected at 40 and 47 months after surgery, respectively. These patients were followed up on a regular basis, and reoperation was not considered because they had no symptoms and signs associated with tricuspid regurgitation. Reoperation was performed in only 1 patient in the MTV group. The patient underwent redo TVR owing to tricuspid valve thrombosis 58 months after initial TVR, as previously described. Five-year and 10-year tricuspid valverelated event-free survival rates were 80.2% and 66.1%, respectively, and were lower in the MTV group than in the BTV group although not significantly so (MTV group 75.0% and 61.7% versus BTV group 89.2% and 80.3%, respectively; p 0.145; Fig 3). The difference remained insignificant after propensity score adjustment (p 0.129). Fig 1. Overall survival curves in the bioprosthetic tricuspid valve replacement (BTV) group and the mechanical tricuspid valve replacement (MTV) group. Comment Our study demonstrated that even in patients younger than 65 years who need oral anticoagulation therapy

1158 HWANG ET AL Ann Thorac Surg TVR IN YOUNGER PATIENTS NEEDING ANTICOAGULATION 2012;93:1154 61 Table 5. Propensity Score-Adjusted Cox Proportional Hazard Analysis of Overall Survival Univariate Analysis Multivariable Analysis Variables p Value Hazard Ratio (95% CI) p Value Age 0.174...... Group 0.350 0.358 (0.080 1.599) 0.179 Sex 0.078 2.506 (0.928 6.796) 0.070 Overweight, BMI 25 kg/m 2 0.118...... Smoking 0.024 2.717 (0.394 18.742) 0.310 Diabetes mellitus 0.352...... Hypertension 0.576...... Chronic renal failure 0.305...... COPD 0.588...... Atrial fibrillation 0.349...... NYHA class III 0.092 1.751 (0.626 4.899) 0.286 Previous cardiac surgery 0.078 4.816 (0.625 37.126) 0.131 Severity of TR 0.999...... LVEF 0.415...... Concomitant operation 0.508...... ACC time 0.166...... CPB time 0.103...... ACC aortic cross clamp; BMI body mass index; CI confidence interval; COPD chronic obstructive pulmonary disease; CPB cardiopulmonary bypass; LVEF left ventricular ejection fraction; NYHA New York Heart Association; TR tricuspid regurgitation. irrespective of TVR, implantation of a mechanical valve at the tricuspid position was not superior to insertion of a bioprosthetic valve because of an increased occurrence of tricuspid valve-related events, especially the composite endpoint of thrombosis, embolism, and bleeding. Tricuspid valve surgery accounts for fewer than 10% of all cardiac procedures [7, 8]. In addition, most patients Fig 2. Freedom from composite endpoint of thrombosis, embolism, and bleeding events in the bioprosthetic tricuspid valve replacement (BTV) group and the mechanical tricuspid valve replacement (MTV) group. get satisfactory results from tricuspid annuloplasty with a suture plication or insertion of a prosthetic ring [9, 10]. However, 5% to 15% of patients with tricuspid valve disease need TVR for various reasons, including organic changes such as rheumatic valve disease and tricuspid valve endocarditis, failure of tricuspid valve repair, and a wish to avoid a high risk of future reoperation [5, 11]. Although there is no conclusive guideline for the selection of a prosthetic valve at the tricuspid position, bioprostheses are generally accepted in patients older than 65 years following guidelines for left-side heart valves [12]. However, there is still a controversy in selecting a suitable valve for TVR in patients younger than 65 years. Proponents of the mechanical TVR argue that the bioprosthetic TVR is associated with structural valve deterioration in the long term [5, 7, 13]. Opponents support bioprosthetic valves owing to a high incidence of valve-related complications after mechanical TVRs, including valve thrombosis and bleeding events related to anticoagulation therapy, limited life expectancy regardless of the type of TVR at long-term follow-up, and expected long-term durability of bioprostheses at the tricuspid position because of low pressures and, thus, low stress in the right-side heart chambers [1, 2, 7, 14 16]. The selection of a prosthetic valve might be more complicated when the patient should undergo long-term anticoagulation therapy for other reasons such as leftside mechanical heart valves and the presence of atrial fibrillation. Although data are limited, some investigators suggested that mechanical TVR should be preferred in patients who have mechanical prostheses in the left heart [5, 8, 17]. However, our data showed that even for patients who needed anticoagulation therapy irrespec-

Ann Thorac Surg HWANG ET AL 2012;93:1154 61 TVR IN YOUNGER PATIENTS NEEDING ANTICOAGULATION 1159 Table 6. Propensity Score-Adjusted Cox Proportional Hazard Analysis of Freedom From Composite of Thrombosis, Embolism, and Bleeding Univariate Multivariable Variables p Value Relative Risk (95% CI) p Value Age 0.555...... Group 0.008 0.089 (0.012 0.678) 0.020 Sex 0.189 Overweight, BMI 25 kg/m 2 0.096 2.713 (0.871 8.456) 0.085 Smoking 0.731...... Diabetes mellitus 0.487...... Hypertension 0.694...... Chronic renal failure 0.751...... COPD 0.627...... Atrial fibrillation 0.487...... NYHA class III 0.308...... Previous cardiac surgery 0.461...... Severity of TR 0.978...... LVEF 0.956...... Concomitant operation 0.229...... ACC time 0.573...... CPB time 0.301...... ACC aortic cross clamp; BMI body mass index; CI confidence interval; COPD chronic obstructive pulmonary disease; CPB cardiopulmonary bypass; LVEF left ventricular ejection fraction; NYHA New York Heart Association; TR tricuspid regurgitation. Fig 3. Five-year and 10-year tricuspid valve-related event-free survival in the bioprosthetic tricuspid valve replacement (BTV) group and the mechanical tricuspid valve replacement (MTV) group. tive of tricuspid valve selection, the composite endpoint of valve-related thrombosis, embolism, and bleeding was higher in the MTV group than in the BTV group. Although mechanical valves used in patients of the present study were bileaflet pyrolytic carbon prostheses, which were credited with improved hemodynamics, low gradients, reduced turbulence, and optimal durability [7], valve thrombosis occurred in 3 of 64 survivors. In addition, 13 patients in the MTV group had bleeding complications, in contrast to only 1 case of bleeding in the BTV group. Reported early mortality rates after TVR have ranged from 10% to 25% [1 4, 11, 13]. In the present study, the early mortality rate was 6.7%. This was better than mortality rates from previous studies. Improved early survival in the present study and in the recent report might be partly due to improved perioperative management including myocardial protection method, modified ultrafiltration, and postoperative intensive care [18]. As expected, multivariable analysis revealed that early postoperative outcome was not affected by the type of prosthetic valve. Previous studies demonstrated poor longterm results regardless of the type of TVR. Two studies performed in the late 1990s found that 5- and 10-year survival rates were less than 60% and less than 40%, respectively [7, 8]. As described above, this was one of the reasons that some authors preferred bioprosthetic valve implantation at the tricuspid position. However, most of the studies reported results of TVR performed before the year 2000. There have been improvements in perioperative management, cardiopulmonary bypass technique, and artificial valve technology since then. In addition, poor early and long-term results made surgeons, including the authors of the present study, perform operations earlier, before right-side heart enlargement and dysfunction became overt [18, 19]. In the present study, overall survival rates after TVR at 5 and 10 years were 86.4% and 80.3%, respectively. There were no differences in overall

1160 HWANG ET AL Ann Thorac Surg TVR IN YOUNGER PATIENTS NEEDING ANTICOAGULATION 2012;93:1154 61 survival between the two groups before and after propensity score adjustment. The advantage of low thromboembolic and bleeding risks after bioprosthetic TVR might be counterbalanced by a limited durability of bioprostheses in the long term, although previous studies suggested that bioprostheses at the tricuspid position have a lower degeneration rate than those located in the systemic circulation [5, 7].Inthe present study, structural valve deterioration occurred in 2 patients in the BTV group. However, reoperation has not been considered yet because they have no symptoms and signs associated with tricuspid valve regurgitation. Owing to improvement in long-term survival, structural valve deterioration could be a substantial problem beyond 10 to 15 years after bioprosthetic TVR. There are limitations to the present study that must be recognized. First, the present study was a nonrandomized, retrospective, observational study in a single institution. Although we enrolled only patients who were younger than 65 years, there were still baseline differences between the two groups, and these might have affected the results. To overcome these biases, we constructed a propensity score model including 16 variables and adjusted the score in multivariable analyses. Discrimination power was excellent with a c-statistic of 0.852. Second, the number of enrolled patients and follow-up duration was different between the two groups because we preferred to use mechanical valves in the early study period. Third, the follow-up duration was relatively short. Longer-term follow-up with a large cohort might be necessary to reach more definitive conclusions. Fourth, difference of clinical outcomes among three types of mechanical valves and two types of bioprosthesis were not evaluated because of small sample size. In conclusion, even in patients younger than 65 years who should undergo life-long anticoagulation therapy owing to left-side mechanical valves or atrial fibrillation, mechanical TVR should be performed with caution because it is associated with an increased occurrence of the composite endpoint of thrombosis, embolism, and bleeding, although there is no difference in overall survival between mechanical and bioprosthetic TVRs. References 1. Van Nooten GJ, Caes FL, François KJ, et al. The valve choice in tricuspid valve replacement: 25 years of experience. Eur J Cardiothorac Surg 1995;9:441 6. 2. Kawano H, Oda T, Fukunaga S, et al. Tricuspid valve replacement with the St. Jude Medical valve: 19 years of experience. Eur J Cardiothorac Surg 2000;18:565 9. 3. Mangoni AA, DiSalvo TG, Vlahakes GJ, Polanczyk CA, Fifer MA. Outcome following isolated tricuspid valve replacement. Eur J Cardiothorac Surg 2001;19:68 73. 4. Filsoufi F, Anyanwu AC, Salzberg SP, Frankel T, Cohn LH, Adams DH. Long-term outcomes of tricuspid valve replacement in the current era. Ann Thorac Surg 2005;80:845 50. 5. Kaplan M, Kut MS, Demirtas MM, Cimen S, Ozler A. Prosthetic replacement of tricuspid valve: bioprosthetic or mechanical. Ann Thorac Surg 2002;73:467 73. 6. Akins CW, Miller DC, Turina MI, et al. Guidelines for reporting mortality and morbidity after cardiac valve interventions. Ann Thorac Surg 2008;85:1490 5. 7. Rizzoli G, De Perini L, Bottio T, Minutolo G, Thiene G, Casarotto D. Prosthetic replacement of the tricuspid valve: biological or mechanical? Ann Thorac Surg 1998;66(Suppl): 62 7. 8. Ratnatunga CP, Edwards MB, Dore CJ, Taylor KM. Tricuspid valve replacement: UK Heart Valve Registry mid-term results comparing mechanical and biological prostheses. Ann Thorac Surg 1998;66:1940 7. 9. Calafiore AM, Iaco AL, Romeo A, et al. Echocardiographicbased treatment of functional tricuspid regurgitation. J Thorac Cardiovasc Surg 2011;142:308 13. 10. McCarthy PM, Bhudia SK, Rajeswaran J, et al. Tricuspid valve repair: durability and risk factors for failure. J Thorac Cardiovasc Surg 2004;127:674 85. 11. Iscan ZH, Vural KM, Bahar I, Mavioglu L, Saritas A. What to expect after tricuspid valve replacement? Long-term results. Eur J Cardiothorac Surg 2007;32:296 300. 12. Bonow RO, Carabello BA, Chatterjee K, et al. ACC/AHA 2006 guidelines for the management of patients with valvular heart disease: executive summary. A report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to Revise the 1998 Guidelines for the Management of Patients With Valvular Heart Disease). Circulation 2006;114:450-527. 13. Chang BC, Lim SH, Yi G, et al. Long-term clinical results of tricuspid valve replacement. Ann Thorac Surg 2006;81: 1317 23. 14. Nakano K, Ishibashi-Ueda H, Kobayashi J, Sasako Y, Yagihara T. Tricuspid valve replacement with bioprostheses: long-term results and causes of valve dysfunction. Ann Thorac Surg 2001;71:105 9. 15. Al-Ebrahim KE. Tricuspid valve replacement is an unfavorable operation. Eur J Cardiothorac Surg 2010;38:115. 16. Péterffy A, Szentkirályi I. Mechanical valves in tricuspid position: cause of thrombosis and prevention. Eur J Cardiothorac Surg 2001;19:735 6. 17. Scully HE, Armstrong CS. Tricuspid valve replacement. Fifteen years of experience with mechanical prostheses and bioprostheses. J Thorac Cardiovasc Surg 1995;109:1035 41. 18. Sung K, Park PW, Park KH, et al. Is tricuspid valve replacement a catastrophic operation? Eur J Cardiothorac Surg 2009;36:825 9. 19. Kim HK, Kim YJ, Park EA, et al. Assessment of haemodynamic effects of surgical correction for severe functional tricuspid regurgitation: cardiac magnetic resonance imaging study. Eur Heart J 2010;31:1520 8. INVITED COMMENTARY Over the past decade there has been an increased awareness among cardiac surgeons about the importance of addressing tricuspid valve (TV) pathology, especially when concomitant cardiac surgical procedures are being performed. A recent study based on the United States Nationwide Inpatient Sample database identified 28,726 TV operations over a 10-year period (1999 2008) and observed a doubling of TV procedures per year (1,712 cases in 1999 versus 4,072 cases in 2008) [1]. TV replacements are rarely performed, but as surgeons become more aggressive addressing a variety of complex and advanced TV pathologic conditions, these replacements 2012 by The Society of Thoracic Surgeons 0003-4975/$36.00 Published by Elsevier Inc doi:10.1016/j.athoracsur.2011.12.036