| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Division of Cardiovascular Surgery, Mayo Clinic and Mayo Foundation, Rochester, Minnesota
b Department of Biostatistics, Mayo Clinic and Mayo Foundation, Rochester, Minnesota
Accepted for publication June 5, 2007.
* Address correspondence to Dr Sundt, Mayo Clinic and Mayo Foundation, 200 First St SW, Rochester, MN 55905 (Email: sundt.thoralf{at}mayo.edu).
Presented at the Fifty-third Annual Meeting of the Southern Thoracic Surgical Association, Tucson, AZ, Nov 8–11, 2006.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
Methods: Between January 1993 and March 2006, 481 adult patients underwent pericardiectomy at Mayo Clinic, Rochester, Minnesota. Excluding those with pericardiectomy for reasons other than constriction, previous tricuspid valve surgery, malignant infiltration, and those undergoing other concomitant cardiac operations, 261 patients remained for evaluation of echocardiographic TR before and after surgery as well as early and late survival.
Results: Tricuspid regurgitation was present in 71% of patients (185 of 261); in 20% (54 of 261), TR was graded moderate or severe. Operative mortality was higher (7 of 54, 13%) among those with moderate/severe TR (7 of 54, 13%, versus 9 of 207, 4.3%; p = 0.019), and by multivariate analysis, moderate/severe TR was an independent predictor of late mortality (hazard ratio: 2.9, 95% confidence level: 1.5 to 5.6; p < 0.001). After excluding patients with prior radiation, moderate/severe TR was no longer a predictor of operative risk, but remained associated with poorer late survival (5-year survival 47% with versus 87% without). Among those with moderate/severe TR, operative mortality was similar if repair was or was not undertaken (2 of 20, 10%, versus 5 of 34, 15%; p = not significant), and late survival was not impacted. Without intervention, however, TR improved in only 29% (8 of 28).
Conclusions: Tricuspid regurgitation frequently complicates constrictive pericarditis, and when moderate or severe, is associated with increased mortality. Although valve repair has little impact on late survival, TR seldom improves with pericardiectomy alone, and may be considered to reduce symptoms, as it can be undertaken without increasing operative risk.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
The incidence and impact of concomitant TR on outcome among patients undergoing pericardiectomy for constriction are unknown. Furthermore, the role of valve repair or replacement is undefined, as it has been reported that tricuspid regurgitation may worsen after pericardiectomy [4, 5]. We therefore reviewed our institutional experience with pericardiectomy for constrictive pericarditis to assess the incidence of TR and evaluate its impact on early and late survival. We also examined the impact of intervention on the tricuspid valve on operative mortality and late outcomes.
|
Material and Methods |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
Study Population
The operative notes of 481 pericardiectomy patients were reviewed to confirm the presence of constrictive pericarditis at the time of pericardiectomy as assessed by the operating surgeon. Three hundred and seventy-six patients had intraoperative evidence of pericardial constriction; the remaining 105 patients had a pericardiectomy performed for reasons other than constriction such as effusive, persistent, or relapsing pericarditis. An additional 99 patients were excluded from analysis as they underwent concomitant cardiac operations other than tricuspid valve surgery at the time of pericardiectomy, 13 patients because of a history of tricuspid valve surgery, and 3 patients because of malignant infiltration of the pericardium. The remaining 261 patients who underwent pericardiectomy for constrictive pericarditis with or without associated tricuspid valve surgery constitute the study population. Their medical records were retrospectively reviewed for details of preoperative characteristics, clinical presentation, and diagnostics, as shown in Table 1.
|
Etiology of Constrictive Pericarditis
The most common etiologies of constriction in this study population were postsurgical in 80 patients (31%), post–viral pericarditis in 33 (13%), and post–radiation therapy in 32 (12%). The etiology of constriction was defined as post–radiation therapy when associated with a history of mediastinal radiation, postsurgical if related to symptoms that followed a prior cardiac operation excluding pericardial windows or incomplete pericardiectomy, and post–viral pericarditis if there was a clear history of viral pericarditis before the onset of pericardial constriction. Other etiologies were encountered in 14 patients (5%) and included connective tissue disorders, post–myocardial infarction, and tuberculosis. Prior incomplete pericardiectomy with clinical evidence of residual constriction mandating a reoperation for completion pericardiectomy was seen in 16 patients (6%) No definitive etiology was identified in 102 (39%) of patients.
Assessment of Severity of Tricuspid Regurgitation
The severity of TR as assessed by preoperative TTE was based on the color flow Doppler characteristics of the tricuspid regurgitant jet. The severity of regurgitation was graded on a 5-point scale as none, trivial, mild, moderate, or severe. Patients who were reported at two levels of severity were assigned to the higher level of severity such that patients reported as mild/moderate TR were assigned to the moderate category. The degree of TR on the intraoperative transesophageal echocardiogram (TEE) and on a TTE performed at discharge was also recorded. Other additional echocardiographic information included tricuspid valve annular dilation, valve thickening or prolapse, right ventricular size and function, and left ventricular ejection fraction.
Surgical Technique
The standard pericardial resection at our institution is radical in nature, with removal of the diaphragmatic component as well as that posterior to the left phrenic nerve in addition to the anterior component from phrenic nerve to phrenic nerve. Both parietal and visceral pericardium are addressed as required. If no plane of dissection can be developed between visceral pericardium and the myocardium, the epicardial peel may be cross-hatched or "waffled," permitting expansion of the underlying ventricle. Small areas of calcified pericardium burrowing into the myocardium may be left unresected.
Radical pericardiectomy is conducted through sternotomy or left thoracotomy at the discretion of the operating surgeon. Whereas the latter may be preferred in a multiple redo setting, the former permits easy institution of cardiopulmonary bypass to facilitate complete pericardial resection as well as performance of concomitant procedures such as tricuspid valve repair. We are not reluctant to utilize cardiopulmonary bypass, as this adds little complexity to the procedure and is warranted in our view if it enables conduct of a more comprehensive procedure.
During the study interval, the need for tricuspid valve intervention at the time of pericardiectomy, the choice between tricuspid valve replacement or repair, and the type of repair was at the discretion of the operating surgeon. In general, when a tricuspid valve replacement is needed, we favor the use of bioprosthesis. Tricuspid valve repair is performed either by a De Vega anuloplasty or placement of a flexible mural anuloplasty band. The operative characteristics of the study population are shown in Table 2.
|
Statistical Analysis
In this study, the relationship between early mortality and categorical risk factors were analyzed using 
2 tests or Fisher’s exact tests, and the relationship between early mortality and continuous risk factors were examined by two-sample t tests or rank-sum tests when appropriate. Long-term survival was described by Kaplan-Meier survival curve, and log-rank tests or Cox regression models were used to identify the potential risk factors that may have impact on long-term survival. All analyses were performed using the SAS 9.1 statistical software (SAS Institute, Cary, North Carolina).
|
Results |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
Characteristics of Moderate to Severe TR Patients
As shown in Table 1, patients with moderate/severe TR were more often female and had a history of atrial arrhythmias, mediastinal radiation, prior cardiac surgery, or indwelling permanent pacemaker. As might be expected, echocardiographic evidence of right ventricular enlargement, tricuspid annular dilation, and right ventricular systolic dysfunction was more common in the group with greater severity of TR. Thickening of the tricuspid valve leaflets on echocardiographic imaging was present in only 13% of patients with moderate to severe TR, and only 1 patient in this group had documented leaflet prolapse.
The groups differed significantly in a number of procedural and postprocedural variables, again with the patients exhibiting TR being generally more severely ill. For obvious reasons, the groups differed with respect to the use of a sternotomy incision and cardiopulmonary bypass (Table 2). Postoperative need for inotropic and intra-aortic balloon support was more common among those with significant TR, as were postoperative atrial and ventricular arrhythmias, and renal failure. Mechanical ventilation for longer than 24 hours was also more common in this group (Table 3).
|
|
|
|
|
|
|
|
|
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
Our current approach to patients with surgically important tricuspid regurgitation at the time of pericardiectomy is a median sternotomy with bicaval cannulation, complete pericardiectomy, and repair of the tricuspid valve. Tricuspid valve replacement is performed in patients with structural abnormalities of the valve. Tricuspid valve intervention can be usually performed without cardioplegic arrest on the beating heart. Patients with severe TR are routinely addressed; patients with moderate TR are addressed selectively depending on the characteristics of the valve, degree of pulmonary hypertension, degree of tricuspid annular dilation, and right ventricular enlargement.
The operative mortality for pericardiectomy at our institution continues to decline over time in spite of an increased use of extracorporeal circulation during pericardiectomy. In our earliest series, McCaughan and coworkers [6] reported a 14% 30-day mortality among 231 patients operated on between 1936 and 1982. At that time, a left anterolateral thoracotomy was our preferred approach. In a subsequent report from our institution extending from 1985 to 1995, Ling and associates [7] reported a 6% 30-day mortality and a 10% operative mortality among 135 patients. Extracorporeal circulation was used in 33% of patients in this series [7]. In the current report, 30-day mortality has fallen to 4.9%, and the operative mortality reached 6.1%. A median sternotomy approach was used in 64% of the cases; and in 45% of our patients, cardiopulmonary bypass was used to aid in the dissection. Importantly, operative mortality in the current series was only 3.1% if patients with post–radiation constriction are excluded from analysis.
Our institutional approach to pericardiectomy also differs from others in the extent of resection we recommend. Although we are more aggressive and accordingly utilize cardiopulmonary bypass more often, our operative mortality is similar to that reported for the more commonly practiced "phrenic to phrenic" pericardiectomy. Bertog and coworkers [8] from the Cleveland Clinic reported a 6% mortality in a contemporary series of 163 patients with pericardial constriction, 143 patients (89%) were approached through a median sternotomy, and cardiopulmonary bypass was utilized in only 30 patients (18%).[8]. The operative mortality in most recent large series of "standard" pericardiectomy for constrictive pericarditis has ranged from 5.6% to 11.9% [9–14].
We believe that the value of our approach is evidenced by the number of patients (16 of 261, 6%) coming to us for "redo pericardiectomy." Furthermore, despite our aggressive attitude, a complete pericardiectomy by our standards could not be accomplished in 13 of 261 patients (5%). These patients had a greater operative mortality and an abbreviated long-term survival compared with patients in whom a complete resection was accomplished. This observation held true even when radiation patients were excluded from the analysis. Indeed, we believe that the more aggressive approach is an important element in reducing operative risk.
Patients with post–radiation constrictive pericarditis present a particularly challenging problem, with significant perioperative risk and diminished late survival, as previously reported by our institution [6, 7] and others [8]. Some degree of restrictive cardiomyopathy is common, if not the rule, as is radiation injury to the lungs. A detailed analysis of this complex group of patients is beyond the scope of this study.
This study is limited by its retrospective nature, and by its reliance on the course outcome variable of survival without details of symptomatic status or current medications. These weaknesses can be best addressed through a prospective study. We have recently created a specialty clinic devoted to this important and poorly understood entity at our institution, and that may help us in this regard. In the interim, however, these findings highlight the importance of attention to concomitant tricuspid valve dysfunction in the patient with constrictive pericarditis.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
You nicely noted the adverse effect on long-term survival of just having tricuspid regurgitation and state in the paper that, even if you repair or replace the tricuspid valve, there is no difference in the operative mortality or the late mortality or survival. But I would question that, in both of those instances you had, I think, a 50% increase in operative mortality for pericardectomy alone in the moderate to severe TR group if you did not operate on the tricuspid valve, which could represent a type II error, because that would give you an odds ratio of 1.5 for risk of death for not addressing the tricuspid valve. Secondly, on the long-term survival, you would have an odds ratio of 1.3 for not addressing the tricuspid valve, and again, this might be a type II error, and maybe you could comment on that as well.
And my final question would be, do you have any quality of life type data on this, because even if the borderline effect on mortality isn’t significant, what is the quality of life if you leave somebody with a moderate to severe tricuspid regurgitation unaddressed? Again, thanks for a nice paper.
DR GÓNGORA: Thank you for your comments, Dr Grover. In regard to the first question, I agree with your assessment. I think that moderate to severe tricuspid regurgitation in the setting of pericardial constriction is more a marker of right ventricular dysfunction, in the sense that most of these patients actually had tricuspid annular dilation, some degree of right ventricular dysfunction and some degree of right ventricular enlargement as the cause for their tricuspid regurgitation. In normal persons, as many as 10% to 20% may have some degree of tricuspid regurgitation, as has been shown in the past. In patients with constriction, as many as 70% have some degree of tricuspid regurgitation, 50% being trivial to mild. So just the constriction per se may impact also on creating some degree of regurgitation, although more than likely it is not of the moderate or severe kind. I think this is more related to ventricular function and dilation of the tricuspid annulus.
In regard to sample size, constrictive pericarditis is not very common, and over 10 years we have managed to see 264 patients who have been carefully selected to just look at the tricuspid regurgitation; and in those patients, 54 had moderate to severe tricuspid regurgitation and only 20 had an operation on the tricuspid valve. So I agree with you in the sense that maybe we are not able to see the differences in operative mortality and late survival just because of the small sample size. So we will continue to look at these patients and follow our experience to see if we can demonstrate differences in this regard.
In regard to quality of life, we are intervening on the tricuspid valves on these patients not with the hope of improving their late survival but with the hope of improving their quality of life. At our institution, we have a large experience in patients with right ventricular dysfunction and severe tricuspid regurgitation in other instances; and in those types of patients, fixing the tricuspid regurgitation improves their quality of life in the sense that edema is less, hepatomegaly is less, and the signs of right heart failure improve more. So we translated this experience to the patients with constrictive pericarditis, and we are operating on them in the hope of improving the control of right heart failure symptoms after surgery. In this paper, we don’t have those data, but that will be the grounds for further studies on this subset of patients.
DR ROBERT N. JONES (Saginaw, MI): I enjoyed this series of patients, who sometimes are very, very ill, very often. I have two questions, though. As I understand, this is a retrospective study. How did you decide who to repair or replace and not at the time of the operation? The second question is, if the outcome is basically the same, your long-term outcome, I am not sure what the take-home message here is, whether we should repair it or not? Is this done for quality of life, because it doesn’t look like it extends life.
DR GÓNGORA: With regard to repair or replacement, patients who had replacement had it because they had structure abnormalities of the tricuspid valve leaflets that would preclude a repair. Regarding patients with repair, they either had annuloplasty or a De Vega, patients who had severe tricuspid regurgitation. It has to do a lot with surgeon preference, but among patients in whom the tricuspid regurgitation is severe, I have noted that it is more than likely that they are going to try to fix it with an annuloplasty ring.
In regard to the take-home message, these data cannot state that tricuspid valve surgery should be undertaken on these patients, because we don’t have the quality of life data. We have established that moderate to severe tricuspid regurgitation increases operative mortality and decreases late survival, but we are still waiting for further study on these patients regarding the quality of life to emphatically state that we should approach these patients. Most patients who had tricuspid valve surgery compared with the ones who did not have tricuspid valve surgery, they had a greater severity of their TR in the sense that they had severe TR. So repairing that severe TR achieved equal early and late outcomes as for patients who didn’t have as severe TR.
DR STEVEN M. GUYTON (Seattle, WA): Enrique, I appreciate your presentation. Did you analyze your results with regard to the TR status after the operation? Do you have results of the group without any residual TR, and would that argue for valve replacement in all of these patients?
DR GÓNGORA: As of the follow-up data on the discharge echocardiogram, 72% of the patients who did not have intervention on their tricuspid valve and had moderate to severe tricuspid regurgitation did not have improvement. Patients who had surgical correction of their moderate to severe TR, 95% were improved and only 5% were in the same category, meaning moderate. Looking further at those patients, I looked at the severity of TR on discharge, and of patients who were discharged with the moderate to severe category, almost 40% who did not have surgery were discharged in the severe category and 60% in the moderate.
|
References |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  S. H. Rahimtoola The Year in Valvular Heart Disease J. Am. Coll. Cardiol., May 19, 2009; 53(20): 1894 - 1908. [Full Text] [PDF]
|
|
|
|
 |
|
 C. A. Anderson, E. Rodriguez, R. L. Shammas, and A. P. Kypson Early constrictive epicarditis after coronary artery bypass surgery. Ann. Thorac. Surg., February 1, 2009; 87(2): 642 - 643. [Abstract] [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 |
