|
|
||||||||
a Department of Cardiothoracic Surgery, Medical University Vienna, Vienna, Austria
b Department of General Surgery, Medical University Vienna, Vienna, Austria
c Department of Cardiothoracic and Vascular Anaesthesia and Intensive Care, Medical University Vienna, Vienna, Austria
Accepted for publication April 27, 2007.
* Address correspondence to Dr Seitelberger, Department of Cardiothoracic Surgery, Medical University Vienna, Waehringer Guertel 18-20, 1090, Vienna, Austria (Email: rainald.seitelberger{at}meduniwien.ac.at).
| Abstract |
|---|
|
|
|---|
Methods: Between October 1997 and July 2004, TV reconstruction was performed in 18 patients (mean age, 38 ± 17 years; 7 women, 11 men), and TV replacement in 4 patients (mean age, 48 ± 22 years; 2 women, 2 men). All patients presented with active endocarditis and severe TV incompetence. Reconstructive techniques included debridement of vegetations, complete resection of infected or destroyed leaflet tissue, leaflet reconstruction with pericardial tissue, sliding plasty of residual valve tissue and bicuspid valve formation with construction of a new commissure, and consecutive ring annuloplasty in all patients.
Results: There were no perioperative deaths. Late mortality was 0% for patients with TV reconstruction and 25% (n = 1) in the TV replacement group. At the latest follow-up (78% complete; mean, 53 ± 18 months), 11 patients had no recurrent TV incompetence. Three patients presented with TV incompetence grade I or II. Two patients with TV reconstruction had recurrent TV endocarditis between 3 and 18 month postoperatively, including new vegetations in both patients and an additional pleural empyema in one. In all cases, conservative treatment was successful and no reoperation was required.
Conclusions: The results of our study clearly demonstrate that in patients with severe TV endocarditis, complex reconstructive techniques yield excellent midterm results with regard to freedom of recurrence of endocarditis and valvular competence and should be considered as the primary surgical option in these patients. Tricuspid valve replacement should only be performed in cases of severe TV destruction that renders reconstructive techniques impossible.
| Introduction |
|---|
|
|
|---|
On the other hand, several reports on tricuspid reconstruction demonstrate that this treatment option offers good results with respect to hemodynamics and long-time survival. A number of different reconstructive techniques have been described, including debridement of vegetations, complete resection of infected or destroyed leaflet tissue, leaflet reconstruction with pericardial tissue, sliding plasty of residual valve tissue, bicuspid valve formation with construction of a new commissure, and ring annuloplasty [9,10]. In this study, we report our experience in patients with tricuspid endocarditis and the extensive use of various reconstructive techniques combined with the implantation of a tricuspid annuloplasty ring.
| Material and Methods |
|---|
|
|
|---|
Patient Characteristics
In 17 patients tricuspid valve endocarditis was isolated. Two patients were diagnosed with additional endocarditis of the aortic valve, and 1 patient had additional endocarditis of the aortic and mitral valves. In 1 of the patients with aortic valve endocarditis, a ventricular septal defect was also present. Two patients had concomitant significant mitral valve insufficiency not related to endocarditis. Three patients had prior cardiac surgery. One patient had prior aortic valve replacement, 1 patient had mitral valve replacement, and 1 patient had mitral valve replacement and DeVega annuloplasty of the tricuspid valve.
Before the operation all patients exhibited signs of acute infection, such as high fever and leukocytosis. Five patients presented with septic pulmonary emboli. Causes of tricuspid endocarditis were intravenous drug abuse in 12 patients, infected pacemaker leads in 5 patients, and congenital lesion (ventricular septal defect) in 1 patient. Factors predisposing to endocardial infections were present in the 3 patients who underwent prior cardiac surgery and valve replacement. In 1 patient no apparent cause for endocarditis could be detected. Blood cultures revealed Staphylococcus aureus (n = 10), Staphylococcus epidermidis (n = 2), Enterococcus faecalis (n = 1), Streptococcus bovis (n = 1), and Streptococcus milleri group (n = 1). In 7 patients no bacterial growth was detected, which was most likely because of previous antibiotic therapy. In the drug addict group, 6 patients were HIV positive and 9 patients were hepatitis C positive. Preoperatively, 11 patients were in New York Heart Association (NYHA) functional class II, 8 patients were in NYHA class III, and 3 patients were in NYHA class IV. In all patients, preoperative transthoracic echocardiography was performed, and the diagnosis of tricuspid valve endocarditis was established according to the Duke endocarditis service criteria [11]. All patients had significant tricuspid valvular incompetence with grade II in 2 patients, grade III in 15 patients, and grade IV in 5 patients. Infection with HIV did not influence our decision-making process, as several authors could demonstrate that mortality rates are equal compared with HIV-negative patients as long as CD4 counts are greater than 200/mm3 [12, 13].
Indication for Surgery
In our series the indication for surgery was largely based on the echocardiographic findings such as significant tricuspid valve insufficiency (grade III or IV) or vegetation size (larger than 1.5 cm) as well as vegetation morphologic features with additional hemodynamic compromise. If the clinical situation allowed, we tried to postpone surgery until the infection decreased.
Surgical Methods
In all patients tricuspid valve reconstruction was attempted and was successfully achieved in 18 patients (mean age, 38 ± 17 years; female/male, 7/11). In 4 patients (mean age, 48 ± 22 years; female/male, 2/2) reconstruction was not possible because of extensive valvular destruction. Those patients underwent tricuspid valve replacement. Three patients received a biologic valve, and 1 patient received a mechanical valve.
Reconstructive Techniques
Several reconstructive techniques were used depending on the site and extent of endocarditic lesions. In most of the patients with infected pacemaker leads (n = 4), the endocarditic lesion was limited to the posterior leaflet. In those patients, bicuspid valve formation of the tricuspid valve was performed as follows: the posterior leaflet was completely excised, the anterior and septal leaflets were partially mobilized, plication sutures were put in place, and a sliding plasty of the remaining two leaflets was performed (Fig 1).
|
|
|
In 2 patients, more than one of the described techniques were combined. In all patients, a prosthetic annuloplasty ring was implanted to stabilize the valve geometry and to prevent future ring dilatation. In this series the original Carpentier-Edwards tricuspid annuloplasty ring and, more recently, the Edwards Lifescience MC3 annuloplasty system were used. The predominant size of rings implanted was 30 mm.
Antibiotic Treatment
All patients received parenteral and oral antibiotics for at least 6 weeks postoperatively. Specific antibiotics were given in those patients who had positive blood or valve cultures, and broad-spectrum empirical therapy was performed in all other cases.
Follow-Up
Follow-up included transthoracic echocardiography and evaluation of functional recovery according to NYHA classification and was performed at 6 months after surgery and then every other year.
| Results |
|---|
|
|
|---|
Two patients with tricuspid valve reconstruction had recurring tricuspid valve endocarditis between 3 and 18 month postoperatively, including new vegetations in one and pleural empyema in the other. In both cases, conservative treatment was successful and no reoperation was required. At the latest follow-up transthoracic echocardiography showed no recurrent tricuspid incompetence in 11 patients and tricuspid valve incompetence grade I to II in 4 patients. Postoperatively NYHA classification improved in all patients from preoperative NYHA 2.6 ± 0.7 to NYHA 1.4 ± 0.5. Seven of the 12 patients who were preoperatively intravenous drug abusers discontinued this habit and were enrolled in a drug-substitution program. In patients with infected pacemaker leads new leads were implanted 2 to 3 weeks after the initial surgery.
| Comment |
|---|
|
|
|---|
Fortunately, most cases (approximately 80%) [6, 20] of uncomplicated right-sided endocarditis can be treated successfully with antibiotics. Endocarditis is termed uncomplicated in the absence of intracardial or extracardial complications, hemodynamic compromise, or infection with a highly virulent microorganism [21]. The abnormal cardiovascular hemodynamics associated with isolated tricuspid or pulmonary valve endocarditis often allows time for medical treatment because of the greater tolerance for tricuspid valve regurgitation and pulmonary embolization [22].
Successful surgical treatment of acute infective endocarditis should include radical debridement of infected tissue [23] and restoration of physiologic valve function, and several very diverse surgical approaches have been published [9, 10]. Valvulectomy without simultaneous replacement as proposed by Arbulu and associates [4, 7, 8] is the most aggressive treatment and was especially promulgated in intravenous drug abusers owing to the radical debridement and the avoidance of any prosthetic material, which is of special importance in patients with ongoing intravenous drug abuse. However, 20% of these patients do not tolerate this massive tricuspid regurgitation despite intensive medical treatment and need a second operation for implantation of a prosthetic valve as a result of progressive right heart failure [3, 8].
Valve replacement is another treatment option, and the development of bileaflet mechanical valves and biologic valves has greatly improved the prognosis of patients after tricuspid valve replacement [24]. Mechanical valves in the tricuspid position are at a significantly higher risk for thrombosis compared with valves in the aortic or mitral position [25] and, consequently, are at higher risk for early reoperation [26]. However, in long-term follow-up mechanical valves are favorable over biologic valves as a result of valvular degradation in biologic valves [27], and therefore mechanical valves should be chosen in younger patients with good long-term prognosis. Nevertheless, patients addicted to intravenous drug abuse and patients with transvenous pacemaker leads are at higher risk to become reinfected, and compliance to long-term anticoagulation is unpredictable.
During the past decades tricuspid valve reconstruction has become a more frequently used treatment alternative in active infective tricuspid valve endocarditis with a number of different treatment modalities such as Kay plasty [28], partial replacement of the tricuspid valve by mitral homografts [29], commissuroplasty, sliding plasty, the use of pericardial patches [9], the use of artificial chordae [10], and reconstruction of the tricuspid valve annulus with an autologous pericardial patch [30]. In some of these series [17, 31], tricuspid valve reconstruction was performed without the use of any foreign material to avoid recurring endocarditis even if significant residual tricuspid valve insufficiency was present. Both the latter studies agreed that the use of a prosthetic annuloplasty ring might have led to a better functional result in some of their patients. However, we think that the avoidance of prosthetic material should not compromise radical eradication of all infected tissue, long-term freedom of recurring endocarditis, or valvular competence.
In our series the main focus was on tricuspid valve reconstruction and an optimal long-term functional result with no or only limited tricuspid insufficiency. Therefore, we implanted an annuloplasty ring in all patients undergoing tricuspid valve repair to avoid future tricuspid ring dilatation and resulting tricuspid insufficiency. Initial moderate insufficiency is well tolerated by the patients undergoing tricuspid valve repair for infective tricuspid endocarditis and generally improves with time [17]. Nevertheless, it is, in our opinion, necessary to stabilize valve geometry to achieve a long-term competent valve. This is especially crucial in patients with massive destruction of one or two leaflets, which necessitates extensive repair with resection of a leaflet or a commissuroplasty.
Implantation of an annuloplasty ring should add only a little risk for recurring endocarditis because it could be shown that the rate for spontaneous prosthetic endocarditis is very low in the tricuspid position. This finding can be explained by the fact that the infection most often only involves the free margin of the valve, including the leaflets, but only rarely the annular region itself. Local factors, such as differences in blood supply of the annular area and the surrounding myocardium, may account for these different manifestations of endocarditis among mitral, aortic, and tricuspid valves [24].
Our approach is also supported by the growing number of successful reports on mitral valve repair involving ring annuloplasty in active endocarditis. Since Dreyfus and coworkers [32] reported in 1990 one of the first series of mitral valve repair in active endocarditis, many authors [33–36] have confirmed not only the feasibility of mitral valve repair in the active phase of endocarditis but could also show that mitral valve repair is superior to mitral valve replacement in regard to event-free long-term survival and recurrence of endocarditis. Although the fear is that implantation of an annuloplasty ring might present additional risk for recurrent endocarditis, all authors implanted annuloplasty rings in the vast majority of these patients to prevent future annular dilatation. In five studies with a medium follow-up of between 23 and 73 months, no recurrence of endocarditis could be observed [25–29].
In our series, 2 patients in the reconstruction group experienced recurring endocarditis. However, in both patients endocarditis could be successfully treated by antibiotics alone, and no reoperation for endocarditis was required. In all patients in the reconstruction group, an acceptable functional result with regard to tricuspid valve competence could be achieved, and none of these patients developed any signs of right heart failure during follow-up related to tricuspid insufficiency.
The greatest management problem in patients with tricuspid valve endocarditis is the question of whether and when they need surgical intervention. Our recommendation for surgery is to wait, if possible, until the peak of serum levels of systemic signs of infection as well as the peak of clinical infective illness has been transgressed. Because of our experience, pressor support during and after surgery is excessive when the operation is performed during the height of clinical infective illness and can be substantially aggravated by the sequelae of substantially high doses of norepinephrine. As stated by Chan and associates [3], indications for surgery can be divided into definite reasons like persistent sepsis and significant congestive heart failure, probable reasons like large vegetation size, involvement of left-sided heart valve(s), gram-negative organisms, or candida, and reasons that are no indication for surgery by itself like persisting fever, recurrent pulmonary embolizations, and polymicrobial infections. We are in agreement with those statements, but think that owing to the advancements in echocardiography during the last decades, our decision whether to perform surgery or not was more directed by echocardiographic findings than in older reported series.
Like other published series, our study is necessarily limited by its retrospective nature and relatively small sample size. Unfortunately we cannot make any statement on the number of patients receiving medical treatment alone during the study period, as many patients are referred to our center solely for surgical repair of their active infective endocarditis. Therefore, we have no conclusive information about the total numbers of patients with this disease in the referring centers. Furthermore, cardiologists and infectious disease specialists from our institution were not able to provide absolute numbers, as these patients often have multiple morbidities and acute infective endocarditis is not the main diagnosis of their admission. They usually are referred for fever of unknown origin, dyspnea, or cardiac decompensation, for example.
The results of our study clearly demonstrate that in the vast majority of patients with active infective endocarditis, tricuspid valve reconstruction in conjunction with the use of an annuloplasty ring can be safely performed with good midterm results. This is true for both freedom of recurring endocarditis and acceptable valvular competence. Consequently, valve replacement should be the last option, especially in younger patients, and should only be considered if extensive valvular destruction renders reconstruction impossible.
| References |
|---|
|
|
|---|
Related Article
This article has been cited by other articles:
![]() |
Endorsed by the European Society of Clinical Micro, Authors/Task Force Members, G. Habib, B. Hoen, P. Tornos, F. Thuny, B. Prendergast, I. Vilacosta, P. Moreillon, M. de Jesus Antunes, et al. Guidelines on the prevention, diagnosis, and treatment of infective endocarditis (new version 2009): The Task Force on the Prevention, Diagnosis, and Treatment of Infective Endocarditis of the European Society of Cardiology (ESC) Eur. Heart J., October 1, 2009; 30(19): 2369 - 2413. [Full Text] [PDF] |
||||
![]() |
J. S. Gammie Invited commentary Ann. Thorac. Surg., December 1, 2007; 84(6): 1949 - 1949. [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 |