|
|
||||||||
Ann Thorac Surg 2006;82:1735-1741
© 2006 The Society of Thoracic Surgeons
a Divisions of Cardiac Surgery of the University of Toronto, Toronto, Ontario, Canada
b Toronto General Hospital, Toronto, Ontario, Canada
c The Hospital for Sick Children, Toronto, Ontario, Canada
Accepted for publication June 2, 2006.
* Address correspondence to Dr Borger, Toronto General Hospital, Room 4N-451, 200 Elizabeth Street, Toronto, Ontario, Canada M5G 2C4 (Email: michael.borger{at}uhn.on.ca).
Presented at the Forty-second Annual Meeting of The Society of Thoracic Surgeons, Chicago, IL, Jan 30Feb 1, 2006.
| Abstract |
|---|
|
|
|---|
METHODS: Two-hundred and fifty patients underwent surgery for organic TV disease at our institution from 1979 to 2003. Clinical and echocardiographic follow-up were obtained, were 99% complete, and were 5.2 ± 4.1 years long (mean ± standard deviation).
RESULTS: One hundred and seventy-eight patients (71%) underwent TV repair and 72 (29%) received TV replacement (54 bioprosthetic, 18 mechanical). Repair patients were more likely to have hypertension, rheumatic pathology, or elective surgery. Concomitant procedures included mitral (50% of patients), aortic (26%), and coronary bypass (6%) operations. Perioperative and midterm mortality were higher in the replacement group (both p < 0.001). Cox regression analysis revealed TV replacement as an independent predictor of midterm mortality (hazard ratio: 5.1, 95% confidence interval: 2.9 to 9.1, p < 0.001) and decreased event-free survival (hazard ratio: 2.0, 95% confidence interval: 1.1 to 3.6, p = 0.02). Follow-up echocardiography revealed more moderate to severe tricuspid regurgitation in repair patients (38% vs 5%, p < 0.001), but no difference in New York Heart Association functional class or reoperation rates.
CONCLUSIONS: Tricuspid valve repair is associated with better perioperative, midterm, and event-free survival than TV replacement in patients with organic tricuspid disease. Despite more tricuspid regurgitation in the repair group during follow-up, reoperation rates and functional class were similar. Repair should be performed whenever possible in patients with organic tricuspid disease.
| Introduction |
|---|
|
|
|---|
The optimal choice of surgical procedure for organic TV disease is not well-studied. The decision of whether to repair or replace the TV is based on several factors, including extent of the disease process, amount of salvageable leaflet tissue, concomitant procedures (repair or replacement) on other heart valves, patient age and comorbidities, and surgical expertise. Previous studies have revealed a high perioperative mortality rate associated with TV replacement, usually in the range of 20% [2, 4]. However, it is unclear whether the increased mortality is a consequence of associated patient comorbidities or of the procedure itself. Although TV repair is associated with better perioperative survival, it has relatively high recurrent rates of late TR [2]. Residual TR can lead to biventricular heart failure, death, or reoperation [2, 3]. The latter is associated with high mortality rates, approaching 40% [2, 3].
There is a lack of contemporary study in the literature to guide the choice of surgical management for organic TV disease. The aim of the current study was, therefore, to compare midterm survival, echocardiographic, and clinical outcomes in patients undergoing TV repair versus replacement for organic TV disease.
| Patients and Methods |
|---|
|
|
|---|
We performed a retrospective review of 250 patients who underwent surgery for organic TV disease between 1979 and 2003 at the Toronto General Hospital. Table 1 lists their pathologies, which included rheumatic, congenital (eg, Ebstein's, atrioventricular septal defect), prosthetic dysfunction, endocarditis, leaflet tears-prolapse, papillary muscle rupture, tumor, and myxomatous or degenerative disease. All patients with a diagnosis of secondary (functional) tricuspid disease were excluded from our analysis. Preoperative, operative, and postoperative data were entered into a database for analysis.
|
Cardiopulmonary bypass was instituted with venous cannulation of the superior and inferior vena cava. Tricuspid valve surgery was performed after concomitant cardiac procedures were completed, usually with the aortic cross-clamp in place. The TV repair techniques varied according to the specific valve pathology and included suture annuloplasty (n = 119), band annuloplasty (n = 59), and other techniques as required.
Follow-Up
Research personnel performed a cross-sectional follow-up, contacting all patients and(or) family members through mailed questionnaire and(or) telephone calls between February and September 2004. Clinical data on New York Heart Association (NYHA) functional status, postoperative morbidity, and mortality were tabulated. Patient clinical status and most recent transthoracic or transesophageal echocardiographic results were also obtained from the patients' cardiologists. Postoperative events were compiled and analyzed according to the American Association for Thoracic Surgery/Society of Thoracic Surgeons guidelines for reporting morbidity and mortality after cardiac valvular operations [5]. Follow-up was 99% complete with a mean (± standard deviation) follow-up time of 5.2 ± 4.1 years (range, 0 to 16 years).
Statistical Analyses
Categoric patient variables were compared using the
2 test or the Fisher exact test where appropriate, and are reported as percentages. The means ± standard deviation of continuous variables were compared using the Student t test for normally distributed variables and the Wilcoxon rank sum test was used for variables that had nonparametric distribution.
The propensity for valve replacement versus repair was performed using logistic regression with all preoperative characteristics as the independent variables, and TV replacements as the dependent outcome. Midterm survival and freedom from morbid events were compared using the Kaplan-Meier method; independent predictors were determined by means of Cox regression analysis. Backward elimination of nonsignificant determinants was employed in analyses. Significance was assumed for p less than 0.05. The SAS version 8.2 statistical software (SAS, Cary, NC) was used.
The variables entered in the multivariable model to determine the significant independent predictors of survival and event-free survival included the following: tricuspid surgery (replacement versus repair), age, male gender, diabetes, hypertension, cholesterol, left ventricular dysfunction, previous stroke, peripheral vascular disease, preoperative renal dysfunction, concomitant coronary artery bypass graft surgery, concomitant mitral valve surgery, concomitant aortic valve surgery, any redo-operation, and active endocarditis.
| Results |
|---|
|
|
|---|
Logistic regression analysis looked at the preoperative factors that increased the propensity for valve replacement versus repair. Significant, independent factors for replacement included urgent surgery, previous syncope or heart failure, hypertension, active endocarditis, and concomitant mitral valve surgery.
In-Hospital Outcomes
Both morbidity and in-hospital mortality were higher in the replacement patients (Table 2). Specifically, postoperative renal failure, duration of ventilatory support, and lengths of intensive care unit and hospital stays were significantly worse in the TV replacement group. There was a trend toward increased low cardiac output syndrome in the replacement group, but no differences were found for perioperative stroke, chest reopening, or insertion of permanent pacemaker.
|
|
|
|
|
Our results revealed no difference between the TV repair and replacement groups 10 years postoperatively with regard to freedom from valve thrombosis (99 ± 6% vs 98 ± 2%, p = 0.4) or bleeding rates (95 ± 2% vs 98 ± 2%, p = 0.5). Actuarial freedom from thromboembolic events at 10 years was also similar between groups (95 ± 2% vs 91 ± 7% for repair versus replacement, p = 0.5 by log rank). The use of anticoagulation at midterm follow-up was no different between the repair and replacement cohorts (55% vs 65%, p = 0.3).
Most recent echocardiographic follow-up showed that 95% of patients in the TV replacement group had mild or less TR, 2.5% had moderate TR, and 2.5% had severe TR. In comparison, 62% of those who had TV repair had mild or less TR, 26% had moderate TR, and 12% had severe TR. However, moderate to severe RV dysfunction on most recent echo was significantly higher in the replacement group (28% vs 9%, p = 0.003). Nonetheless, there was no difference in the functional status of patients in either group, as similar proportions were in NYHA class IIIIV on follow-up (repair: 17% vs replacement: 16%, p = 0.9).
| Comment |
|---|
|
|
|---|
Tricuspid valve repair techniques can be divided into those that are suture-based and those that utilize an annuloplasty ring. They are well-described in the literature in the setting of secondary (functional) TR. Such repair techniques usually do not add significantly to the operative time and can be performed with low rates of morbidity and mortality. Tricuspid valve repair is associated with improved long-term survival and event-free survival in patients with functional TR [7]. However, their long-term durability is questionable with relatively high published recurrence rates for TR [7, 8]. Tricuspid valve replacement operations, in contrast, are associated with marginally longer operative times and higher in-hospital mortality rates [2, 9, 10]. However, residual TR is not as frequent as in patients undergoing TV repair [2, 10]. Published TV replacement series contain a higher proportion of patients with organic pathology than published TV repair studies, making comparison of outcomes for the two surgical techniques difficult. For this reason, we chose to focus only on patients with organic TV pathology to compare repair with replacement.
Our results showed that TV repair for organic pathology resulted in significantly improved survival and event-free survival when compared with the TV replacement cohort, predominantly because of a higher early (perioperative) mortality rate in the TV replacement group. The survival curves for the two groups did not cross, nor were they parallel. The worsened mortality is possibly due to progressive RV dysfunction in the TV replacement group caused by a large, rigid object (the prosthetic valve) in a deformable, low-pressure cavity. Our results support this theory by the findings of increased perioperative low output syndrome and higher rate of RV dysfunction on follow-up echo in the TV replacement group. The association of acute RV failure post-cardiac surgery with increased mortality is well-supported in the literature [4].
Despite improved survival, midterm follow-up echocardiography revealed that the repair group had significantly more patients with recurrent moderate to severe TR. However, symptom severity and functional NYHA class were no different between the two groups. As noted above, RV function during follow-up was better in the repair group, despite the increased prevalence of recurrent TR. In addition, the incidence of overall TV reoperation rate, which is associated with significant risk of mortality, was not different between the two groups.
There is some related evidence in the echocardiography literature to compare with our findings. A study from the Cleveland Clinic examined 401 patients with TV dysfunction (etiology not stated) and followed them for 10 years post-surgery with serial echocardiography [11]. These investigators showed that early survival was worse in those that had TV replacement versus repair (relative risk 2.9). Late echocardiographic TV failure (moderate to severe TR) was associated with use of a TV repair-annuloplasty strategy, similar to our own results.
Our result of relatively high TR recurrence in the repair group is supported by other investigators, with reported recurrence rates up to 45% during long-term follow-up [1215]. The recurrence rate varies according to the type of repair performed, with higher recurrences for suture annuloplasty, particularly the De Vega repair [7, 8].
Tricuspid valve replacement results from the literature are also similar to those reported in our study. Filsoufi and colleagues [9] reported 81 consecutive cases from the Brigham and Women's hospital over 15 years, 52 (64%) with organic TV dysfunction. Overall operative mortality was 22%, the same as in our study. The largest and longest series on TV replacement is from the UK Heart Valve Registry [2], which looked at 425 patients who underwent TV replacement for all etiologies. They reported no significant survival benefit based on choice of prosthesis (bioprosthetic versus mechanical). Overall 30-day mortality was 17%. Survival rates at one, five, and 10 years postoperative were 72%, 60%, and 43%, respectively, similar to our reported rates of 76%, 63%, and 55%. However, a small proportion of the UK Heart Valve Registry patients (4.5%) were undergoing redo-cardiac surgery, and patients with organic and functional disease were included. Half of the patients in our study underwent redo-cardiac surgery and all had organic TV disease, which is associated with a worse long-term prognosis than functional disease.
Thromboembolism is well-described in the literature as a complication of TV replacement surgery, particularly for mechanical prostheses [10, 15, 16]. A meta-analysis reported rates of 1.28% patient/year (range, 0.1 to 4.6% patient/year) [10]. Our results revealed no difference between the TV repair and replacement groups fifteen years postoperatively with regard to freedom from thromboembolic events, valve thrombosis, or bleeding rates. This is likely due to the fact that our study population included only 18 mechanical TV replacements.
Study Limitations
Our study is limited by its retrospective nature with all of the inherent limitations of such investigations. However, the current study is the largest series to date comparing TV repair to replacement for organic TV pathology. In addition, our follow-up was nearly complete. Therefore we feel our results are important for cardiac surgeons to consider when making the difficult decision of whether to repair or replace a TV with structural leaflet pathology.
Conclusion
Tricuspid valve repair is associated with better early and midterm survival than TV replacement in patients with organic tricuspid disease. The increased mortality is possibly due to low output syndrome and RV dysfunction secondary to a rigid prosthesis in the tricuspid position. Despite more recurrent TR in TV repair patients during follow-up, NYHA functional class and reoperation rates are similar. Tricuspid valve repair should be performed whenever possible in patients with organic tricuspid disease.
| Discussion |
|---|
|
|
|---|
In addition, was there any preop assessment of the right ventricular function? A counterhypothesis could be that very poor right ventricular function requires tricuspid replacement, rather than repair. Most surgeons don't go into the operating room thinking they are going to replace a tricuspid valve. The valve was replaced for a reason, and it was probably because the right ventricle was very dilated and poorly contractile, and an attempt to repair was done initially. One of the reasons not to have a good outcome in tricuspid repair is the right ventricle is so massively dilated and the leaflets so severely tethered, that no matter how small a ring you put in, the leaflets will not reach the annulus.
DR SINGH: Those are excellent observations. Firstly, two-thirds of the repairs were non-ring annuloplasties, including De Vega and non-De Vega repair techniques. We did not specifically look at the incidence of late tricuspid regurgitation (TR) within the different types of repair techniques. However, we have another paper in press, in Circulation, that compares late TR rates for ring versus no-ring repair for functional TR, and patients who received a ring had less recurrent TR and improved event-free survival. Therefore, as you point out, the choice of repair approach may be related to the rate of late recurrent TR. However, what we observed in the current analysis was that despite the repair group having more TR at long-term follow-up, the incidence of reoperation or patient NYHA (New York Heart Association) functional class was no different compared with the replacement group, who had minimal late recurrent TR.
Regarding your second comment, unfortunately we did not have any preoperative assessment of RV (right ventricular) function. As you mention, a massively dilated RV may not allow for a perfect repair outcome. As such, surgeons may replace the valve, particularly if they have already replaced one of the other valves. However, our findings would suggest that it may be better to accept a less than ideal repair, rather than replace the tricuspid valve. This is because replacement, as we and others have shown, is a very risky operation with high perioperative mortality. Doing a repair avoids this high mortality risk. As we have shown, even if the repair was not ideal and residual TR or late TR developed, this does not significantly affect the need for a future reoperation, or worsen patient function class in the long term. So, surgeons should repair the valve if at all possible, and even if not ideal with some residual TR, subsequent replacement should be avoided during the operation.
DR FRANK W. SELLKE (Boston, MA): Your death rate and renal failure rate and rate of prolonged ventilation was much higher in the replacement group. Doesn't that suggest that the groups were not equal? The mortality in the replacement group was 20%. That seems pretty high.
DR SINGH: Tricuspid valve replacement does carry a high mortality rate. Our data revealed perioperative mortality to be 22%. However, this is comparable to other series in the literature, such as the UK Heart Valve Registry.
There certainly were differences in the preoperative and operative characteristics of the replacement patients versus the repair group. Many in-hospital outcomes were worse in the replacement group, which, as you comment, may suggest that the two groups were not equal. However, we did our best to account for these differences by performing multivariable regression analysis. From this analysis we determined independent risk factors for poor outcomes, and TV replacement was a poor independent predictor of short- and long-term outcomes.
DR JAMES W. FREDERIKSEN (Chicago, IL): Were you able to determine why patients who had tricuspid valve replacement had that procedure instead of repair?
DR SINGH: That is a good question. Unfortunately we weren't able to determine why repair versus replacement was chosen at the time of surgery in most cases. I am sure that the surgeon preference and patient comorbidities were taken into account. Likewise, the extent of the disease, whether there was sufficient salvageable tissue for a repair, and concomitant valvular operations were factored in to whether a repair or replacement was performed. As stated above, however, our take-home message is that an imperfect repair is probably preferable to a tricuspid valve replacement procedure.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
M. B. Izzat Spring Retractor: A New Adjunct for Aortic Valve Surgery Ann. Thorac. Surg., July 1, 2011; 92(1): 364 - 365. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Lauten, M. Ferrari, K. Hekmat, R. Pfeifer, G. Dannberg, A. Ragoschke-Schumm, and H. R. Figulla Heterotopic transcatheter tricuspid valve implantation: first-in-man application of a novel approach to tricuspid regurgitation Eur. Heart J., May 2, 2011; 32(10): 1207 - 1213. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. B. Irwin, M. Luckie, and R. S. Khattar Tricuspid regurgitation: contemporary management of a neglected valvular lesion Postgrad. Med. J., November 1, 2010; 86(1021): 648 - 655. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. A. Sarralde, J. M. Bernal, J. Llorca, A. Ponton, L. Diez-Solorzano, J. R. Gimenez-Rico, and J. M. Revuelta Repair of Rheumatic Tricuspid Valve Disease: Predictors of Very Long-Term Mortality and Reoperation Ann. Thorac. Surg., August 1, 2010; 90(2): 503 - 508. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. M. Bernal, A. Ponton, B. Diaz, J. Llorca, I. Garcia, J. A. Sarralde, J. Gutierrez-Morlote, C. Perez-Negueruela, and J. M. Revuelta Combined Mitral and Tricuspid Valve Repair in Rheumatic Valve Disease: Fewer Reoperations With Prosthetic Ring Annuloplasty Circulation, May 4, 2010; 121(17): 1934 - 1940. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. Vigano, A. Guidotti, M. Taramasso, A. Giacomini, and O. Alfieri Clinical mid-term results after tricuspid valve replacement Interact CardioVasc Thorac Surg, May 1, 2010; 10(5): 709 - 713. [Abstract] [Full Text] [PDF] |
||||
![]() |
B. M.S. Shrestha, S. Fukushima, M. Vrtik, I. H. Chong, L. Sparks, H. Jalali, and P. G. Pohlner Partial Replacement of Tricuspid Valve Using Cryopreserved Homograft Ann. Thorac. Surg., April 1, 2010; 89(4): 1187 - 1194. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. C. Lee, B. Desai, and D. D. Glower Results of 141 Consecutive Minimally Invasive Tricuspid Valve Operations: An 11-Year Experience Ann. Thorac. Surg., December 1, 2009; 88(6): 1845 - 1850. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. K. Park, P. W. Park, K. Sung, Y. T. Lee, W. S. Kim, and T.-G. Jun Early and Midterm Outcomes for Tricuspid Valve Surgery After Left-Sided Valve Surgery Ann. Thorac. Surg., October 1, 2009; 88(4): 1216 - 1223. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. H. Rogers and S. F. Bolling The Tricuspid Valve: Current Perspective and Evolving Management of Tricuspid Regurgitation Circulation, May 26, 2009; 119(20): 2718 - 2725. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. J. Bruce and H. M. Connolly Right-Sided Valve Disease Deserves a Little More Respect Circulation, May 26, 2009; 119(20): 2726 - 2734. [Full Text] [PDF] |
||||
![]() |
A. Shiran and A. Sagie Tricuspid Regurgitation in Mitral Valve Disease: Incidence, Prognostic Implications, Mechanism, and Management J. Am. Coll. Cardiol., February 3, 2009; 53(5): 401 - 408. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. J. Moraca, M. R. Moon, J. S. Lawton, T. J. Guthrie, K. A. Aubuchon, N. Moazami, M. K. Pasque, and R. J. Damiano Jr Outcomes of Tricuspid Valve Repair and Replacement: A Propensity Analysis Ann. Thorac. Surg., January 1, 2009; 87(1): 83 - 89. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. M. Bernal, A. Ponton, B. Diaz, J. Llorca, I. Garcia, A. Sarralde, C. Diago, and J. M. Revuelta Surgery for rheumatic tricuspid valve disease: a 30-year experience. J. Thorac. Cardiovasc. Surg., August 1, 2008; 136(2): 476 - 481. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. Guenther, C. Noebauer, D. Mazzitelli, R. Busch, P. Tassani-Prell, and R. Lange Tricuspid valve surgery: a thirty-year assessment of early and late outcome Eur J Cardiothorac Surg, August 1, 2008; 34(2): 402 - 409. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. H. Jones The Year in Cardiovascular Surgery J. Am. Coll. Cardiol., May 8, 2007; 49(18): 1887 - 1898. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |