| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Department of Cardiothoracic Surgery, Thoraxcentre, Erasmus Medical Centre Rotterdam, Rotterdam, The Netherlands
b Department of Cardiothoracic Surgery, University Hospital Maastricht, Maastricht, The Netherlands
c Department of Cardiology, Thoraxcentre, Erasmus Medical Centre Rotterdam, Rotterdam, The Netherlands
Received 12 November 2007; received in revised form 27 February 2008; accepted 21 March 2008.
* Corresponding author. Address: Department of Cardiothoracic Surgery, University Hospital Maastricht, P. Debeyelaan 25, PO Box 5800, 6202AZ Maastricht, The Netherlands. Tel.: +31 433877070; fax: +31 433875070. (Email: meindert_palmen{at}hotmail.com).
 |
Abstract |
|---|
 Top Abstract 1. Introduction2. Materials and methods3. Results4. DiscussionReferences |
|---|
Key Words: Cardiac congenital • Cyanotic • Acyanotic • Tricuspid valve
|
1. Introduction |
|---|
|
TopAbstract1. Introduction2. Materials and methods3. Results4. DiscussionReferences |
|---|
In 1988 we adopted the technique developed two decades ago by Carpentier et al. [1]. The repair consists of vertical plication of the atrialised part of the right ventricle and reimplantation of the anterior and posterior tricuspid valve leaflets at the level of the neo-tricuspid annulus. This method aims at restoration of right ventricular size and geometry, in addition to restoration of the tricuspid architecture and competence. Plication of the dyskinetic atrialised right ventricle seems vital to reach this goal [5]. We and others demonstrated promising short- and medium-term results using this repair technique [6–9]. In two recent papers by Chauvaud et al. [5,7], addition of a bidirectional partial cavo-pulmonary connection (PCPC) anastomosis is advocated in patients with impaired right ventricular function prior to surgery.
In this study we present our data of 20 years of experience with repair of Ebstein's anomaly using the vertical plication repair as described by Carpentier et al. [1].
|
2. Materials and methods |
|---|
|
TopAbstract1. Introduction2. Materials and methods3. Results4. DiscussionReferences |
|---|
|
2.3 Exercise testing
Exercise capacity was measured using symptom-limited bicycle ergometry with a standardised incremental protocol, as described previously [6]. Results were compared to a standardised population and corrected for age, sex and weight.
2.4 Arrhythmias
In our patient group, the incidence of pre- and postoperative arrhythmias is low. Only eight patients complained of palpitations preoperatively. Twenty-six patients were in sinus rhythm prior to operation, two patients were in atrial fibrillation. One of these (recently operated) patients received pulmonary vein isolation concomitant with repair of Ebstein's anomaly. The other patient had long-standing atrial fibrillation in which pulmonary vein isolation was not considered to be indicated. Two patients had a history of Wolff–Parkinson–White (WPW) syndrome and ablation of the accessory pathway several years prior to surgery. Only one patient complained of palpitations during follow-up, while two patients developed de novo atrial fibrillation during follow-up. One patient was treated pharmacologically, the other underwent pulmonary vein isolation at reoperation.
2.5 Preoperative echocardiography and cardiac catheterisation
All patients were preoperatively investigated by transthoracic echocardiography. When imaging was not considered satisfactory, additional transoesophageal echocardiography was performed. During echocardiography, attention was focused on morphological deformities, the degree of tricuspid regurgitation, right ventricular function (qualitatively classified as good, impaired or bad) and right-sided dimensions (either dilated or normal). When elevated right-sided pressures were suspected at echocardiography, right-sided cardiac catheterisation was performed (n
= 13). Right-sided pressures in this patient group showed a systolic pulmonary artery pressure of 20 ± 5 mmHg, a systolic right ventricular pressure of 21 ± 5 mmHg and a mean right atrial pressure of 9 ± 4 mmHg. Since this is a retrospective study spanning two decades, we were not able to collect more quantitative right-sided imaging using, for instance, cardiac MRI.
2.6 Exclusion criteria for vertical plication repair
In the same period, eight patients with Ebstein's anomaly did not fit the criteria for Carpentier repair. Five of these patients presented with severe Ebstein morphology with severely impaired right ventricular function or decreased pulmonary flow due to right ventricular outflow tract obstruction. One of these patients underwent tricuspid valve replacement (Carpentier type 3–4 with inadequate valvular tissue to obtain an adequate surface of coaptation). In four patients a univentricular approach was applied with creation of a Fontan circulation. These latter patients had presented in the neonatal period with a duct-dependent pulmonary circulation and were uniformly classified as Carpentier type 4. Additionally, in two Ebstein patients, a mild form (Carpentier type 1) was found with only a trace of tricuspid incompetence and minor right ventricular atrialisation. In one patient, this was associated with severely depressed right ventricular function. Therefore, only the secundum ASD was closed and a bidirectional cavopulmonary connection was constructed. The other patient presented with primary massive mitral incompetence and minimal tricuspid incompetence. Therefore, only a mitral valve procedure was performed after inspection of the tricuspid valve.
2.7 Operation
The procedure was performed as described extensively by Carpentier et al. [1] and Quaegebeur et al. [6] two decades ago. After institution of bicaval cardiopulmonary bypass and cold cardioplegic arrest, a right atriotomy was performed and the anomalous morphology inspected and classified according to Carpentier et al. [1]. Subsequently, the enlarged anterior (and when present the posterior) leaflet was detached from the tricuspid annulus, starting at the anteroseptal commissure and proceeding along the annulus in a clockwise fashion. All abnormal (fibromuscular) connections of the anterior and posterior leaflets with the ventricular wall were dissected as far as possible towards the apex of the true right ventricle, with great care to maintain adequate leaflet support near the edges (preventing flail of the leaflet). The atrialised right ventricular tissue is vertically plicated, using a continuous 5-0 prolene suture, as described in detail before [6]. Transmural suturing was actively avoided and the suture line remained caudal to the coronary sinus, in order not to compromise the coronary vasculature (posterior descending artery), or the conduction tissue, respectively. Subsequently, the detached tricuspid leaflet(s) were rotated clockwise and sutured back into the neo-tricuspid annulus, using 5-0 prolene. In only one patient there was enough leaflet tissue available to enable a full 360 degrees fit of the leaflet tissue without inducing stenosis, and a conus-like reconstruction was performed [9]. In nine (32%) patients the neo-tricuspid annulus was reinforced using a prosthetic ring (Carpentier–Edwards, average diameter of 32.8 mm). When a secundum ASD was present, this was closed, usually by primary suturing (96%). In four patients (14%), the right atrium was severely enlarged and right atrial reduction was performed by excision of part of the atrial wall. In only three (11%) patients (all operated after 2003) with anticipated impaired right ventricular function a bidirectional cavopulmonary connection was performed, as proposed by Chauvaud et al. [5,7]. After termination of CPB, transoesophageal echocardiography was performed and the function of the tricuspid valve and the right ventricle were assessed. When haemodynamically stable, the patients were transferred to the intensive care unit for recovery.
In 17 patients predischarge transthoracic echocardiographic studies were available for re-evaluation. All patients received coumarins for at least 3 months after operation. When patients were in sinus rhythm with an adequate right ventricular and right atrial function, coumarins were stopped and antiplatelet drugs were initiated. When indicated, coumarins were continued.
2.8 Follow-up
All patients were evaluated clinically and echocardiographically during routine visits to the outpatients’ clinic. We evaluated clinical status, exercise tolerance, NYHA functional class, electrocardiographic and echocardiographic results at 1 year and at the end of follow-up. The mean duration of follow-up was 10.7 ± 6.5 years with the longest duration of follow-up being 19.2 years. One patient was lost to follow-up.
2.9 Statistical analysis
Early mortality was defined as mortality within 30 days after operation. Actuarial survival and actuarial freedom from reoperation were analysed using the Kaplan–Meier method. Ranked variables were evaluated with Wilcoxon rank test. A p-value of p
< 0.05 was considered statistically significant. All testing was performed two-sided.
|
3. Results |
|---|
|
TopAbstract1. Introduction2. Materials and methods3. Results4. DiscussionReferences |
|---|
There was no operative mortality. One patient (3.6%) died on the third postoperative day due to right ventricular failure. This was a severely symptomatic patient (NYHA class IV) with preoperative right ventricular failure and severe right ventricular and right atrial dilation. Despite the addition of a PCPC and maximal inotropic support, right ventricular failure was inevitable. In the course of follow-up, no additional patients died, resulting in a 1 year and 18 years survival of 96.4% (96.3–96.5%). Fig. 1 demonstrates the overall actuarial survival.
|
3.1 Freedom from reoperation
Six patients (21%) required reoperation during the course of follow-up after an initially successful repair. No patients died following reoperation. Fig. 2
displays the overall actuarial freedom from reoperation. One year actuarial freedom from reoperation was 89% (CI 77–100%) and at end of follow-up 73% (CI 53–92%).
|
Three patients were reoperated beyond 1 year of follow-up. The first patient (Carpentier type 3) was reoperated 14 months after initially successful repair (tricuspid incompetence grade I/IV post-perfusion) because of symptomatic tricuspid incompetence. The vertical plication of the right ventricle was partially torn, resulting in a severe paravalvular tricuspid incompetence. The defect was closed and the valve repaired, after which a prosthetic ring was inserted to reinforce the annulus. Postoperative echo revealed no residual tricuspid incompetence. She is in NYHA functional class I, despite the fact that she suffered from pulmonary embolism, 13 years after reoperation, for which she was treated conservatively. The second patient (Carpentier type 4, impaired right ventricular function and tricuspid incompetence grade I/IV after repair) was reoperated 8 years after successful primary repair as a result of progressively symptomatic tricuspid incompetence and right ventricular failure. At reoperation it appeared that the re-inserted valve leaflet was partially detached from the annulus at the level of the septum possibly due to annular dilation. This was primary repaired and a bidirectional cavopulmonary shunt was added to alleviate the now impaired right ventricle. No residual tricuspid incompetence was noted after termination of bypass. Finally, one patient (Carpentier type 3, good right ventricular function and tricuspid incompetence grade II/IV after initial repair) was reoperated 10 years after initial repair because of decreased exercise intolerance and impaired right ventricular function in conjunction with mild (grade II/IV) tricuspid incompetence. At reoperation, a ring-annuloplasty and an additional PCPC were performed. At present, the patient is in NYHA functional class I and has some complaints of supraventricular tachycardia. Interestingly, all these patients were operated before 1992, and no prosthetic ring was included in the first repair at that time.
3.2 Freedom from MACE
Freedom from major adverse cardiac events (MACE) was classified according to Edmunds et al. [10]. We encountered eight MACE during the course of follow-up; six reoperations, one death and one third degree AV block necessitating a permanent pacemaker. Freedom from MACE after 1 year, 5 years and end of follow-up was 82 ± 7%, 78 ± 8% and 66 ± 10% respectively. Univariate analysis revealed no significant risk factors for MACE.
3.3 Echocardiographic results
Preoperatively, 0/28 patients (0%) had tricuspid incompetence grade I/IV (trace), one patient (4%) grade II/IV (mild), five patients (18%) grade III/IV (moderate), and the majority of patients (22/28, 79%) grade IV/IV.
After repair, tricuspid incompetence decreased following termination of bypass (p < 0.001), 19 patients (68%) had tricuspid incompetence grade I/IV, eight (28%) II/IV and one (4%) III/IV. Only 17 predischarge echocardiographic studies were available for re-evaluation. Because this number would result in incomplete data and because the available studies were essentially not different from the intra-operative evaluations, the results are not included.
At 1 year of follow-up, tricuspid incompetence was very much comparable to directly postoperatively. Ten (42%) patients had tricuspid incompetence grade I/IV, 11 (46%) grade II/IV, two (8%) grade III/IV and one (4%) grade IV/IV. At end of follow-up, in some patients tricuspid incompetence increased from grade II to grade III, but overall remained comparable to 1 year postoperatively. Ten (50%) patients had grade I/IV tricuspid incompetence, five (25%) grade II/IV, four (20%) grade III/IV and one (5%) grade IV/IV TI.
In 20/28 (71%) patients, right ventricular function was good prior to repair, in seven (25%) patients impaired and in one patient (4%) bad. Following surgery, right ventricular function was preserved in a vast majority (26, 93%) of the patients, although two patients (7%) demonstrated a slight decrease in right ventricular function, when compared to the echocardiographical findings before repair. At 1 year follow-up three (12%) patients had a bad right ventricular function, five (21%) impaired and 16 (67%) good. At the end of follow-up four (17%) patients had a bad right ventricular function, five (22%), impaired and 14 (61%)good. Interestingly, at end of follow-up half of the patients with a bad right ventricular function are in NYHA functional class II and the other half in class I.
3.4 Functional status and exercise tolerance during follow-up
Preoperatively, one patient (4%) was in NYHA functional class I, six patients (21%) in class II, 16 patients (57%) in class III and five patients (18%) in class IV. At 1 year of follow-up, patients demonstrated a significant improvement in NYHA functional class when compared to preoperative values (p
< 0.001). Only five (21%) of patients were in NYHA functional class II, while no patients were in functional class III or IV. At the end of follow-up, all patients remained in the same NYHA functional class when compared to 1 year postoperatively (p
< 0.001). Two of the NYHA class II patients did not have a biventricular result. One patient received a PCPC at the time of primary repair, while the other patient was ultimately converted to a univentricular circulation due to a failing right ventricle.
Preoperatively, exercise tolerance was markedly reduced in the study group, when compared to a standardised population. At end of follow-up, exercise tolerance was increased when compared with preoperative values, indicating a durable augmentation in functional capacity in this patient group (70 ± 19% vs 92 ± 9% of predicted values, p < 0.05).
|
4. Discussion |
|---|
|
TopAbstract1. Introduction2. Materials and methods3. Results4. DiscussionReferences |
|---|
Recently, the group of Danielson and co-workers [12] demonstrated long-term follow-up in a young patient population, averaging age 7. They observed an early mortality of 5.6%. However, at 15 years of follow-up, freedom of reoperation was 61%, while actuarial survival was 89.9%.
More recently, Komoda et al. introduced a new repair technique using the combination of a Hetzer operation and a Sebening stitch [13], posing the question whether or not ventricular reconstruction is essential to repair of Ebstein's anomaly. Although initial results are reported to be good, long-term functional results have to determined.
The efficacy of the vertical plication repair was demonstrated by Carpentier et al. two decades ago [1]. A higher percentage of patients seemed to be amenable for repair with good medium-term results. In a previous study, we presented similar short-term results and demonstrated that reduction in the grade of tricuspid incompetence was accompanied by improvement in clinical status and exercise tolerance [6].
In our present study, we demonstrated favourable long-term outcome for patients with Ebstein that present beyond the neonatal period, using the Carpentier technique of tricuspid valve and right ventricular repair, with vertical plication of the atrialised right ventricle. In our hands, this technique displays a low perioperative mortality and acceptable reoperation figures.
The low perioperative mortality rates (3.6%) are in the same range as in studies presented by Chauvaud et al. [7] (9%, 18/191 pts.) and Chen et al. [8] (0/25 pts.) in a similar patient population. However, in this study, we demonstrated that actuarial survival rates are considerably higher (96%) when compared to other patient populations. Chauvaud et al. described an 82% actuarial survival rate. In the results presented by Chen et al. [8], long-term follow-up showed a mortality rate of 8%. Da Silva et al. [9] demonstrated a perioperative mortality of 2.5% and a long-term mortality of only 5% using the cone reconstruction technique.
4.1 One and a half ventricle repair
In our series in 11% of patients (n
= 3) with impaired right ventricular function and residual tricuspid incompetence, a partial cavopulmonary connection was added to the repair. Chauvaud et al. advocated addition of a cavopulmonary connection in 2003 to unload the right ventricle [5,7]. Also the group of Danielson demonstrated that an addition of a partial cavopulmonary connection (1.5 ventricle repair) applied to patients with severe Ebstein's anomaly and impaired right ventricular function has good functional outcome and carries low perioperative mortality using the Danielson technique for repair of Ebstein's anomaly [14]. In our hands, we add a partial cavopulmonary connection to the repair, only when impaired function of the reconstructed right ventricle is anticipated or intra-operatively observed. It is interesting to see that this does not result in a much different outcome as compared to a more liberal or even standard inclusion of the partial cavopulmonary connection to the repair.
4.2 Residual tricuspid incompetence
In some of our patients (n
= 8), a mild (grade II (n
= 6) or moderate (grade II–III/IV, (n
= 2)) tricuspid regurgitation was present after surgery. When this was accompanied by a decrease in right ventricular function, these patients remained symptomatic, and reoperation was indicated in three patients, half of the number of reoperated patients. One preoperatively severely symptomatic patient with residual tricuspid regurgitation grade III/IV after surgery proved beyond repair and died because of right ventricular failure despite addition of a PCPC. Two other patients with residual tricuspid regurgitation showed progression over time and were eventually reoperated because of progressive complaints. The other four patients with residual TI are in good clinical condition and NYHA functional class with stable tricuspid incompetence grade II. The remaining three reoperations in this population had tricuspid regurgitation grade I after repair.
Whether or not the reinserted tricuspid valve is sufficient is dependent on the amount of functional valve tissue and on the size of the neo-annulus, which is dependent on the extent of vertical plication of the atrialised right ventricle. Too much plication results in a too small annulus and subsequent tricuspid stenosis, while insufficient valve tissue can result in regurgitation as well. We tend to accept some regurgitation, rather than having the risk of tricuspid stenosis or having to replace the valve. Furthermore, Chen et al. [8] demonstrated that even patients with moderate tricuspid regurgitation have good functional outcome. This is also underlined by the functional data in our study, as indicated above. In this regard, Da Silva's group [9] recently described a modified Carpentier technique, which they termed cone reconstruction. All of the available valve tissue is rotated clockwise for a 360-degrees fit into in the neo-ostium, thereby creating a bicuspid valve instead of a monocusp that coaptates with the septum, which is most often the case in the Carpentier technique. The cone reconstruction claims to result in better leaflet coaptation and improved valve competence, although tricuspid valve stenosis is a possible drawback of this technique [9].
4.3 Reoperation after Carpentier repair
In this study, we had a slightly higher reoperation rate (21% at the end of follow-up), when compared to similar studies in literature. Chauvaud et al. [7] demonstrated in a similar patient population an actuarial freedom from reoperation of 82% at the end of follow-up (mean follow-up duration 6.4 years), while Chen et al. [8] showed that 3/25 patients (12%) required reoperation during the course of follow-up, although mean duration of follow-up is shorter in the latter study (4.1 years) when compared to our data (10.7 years).
In the early reoperation group (n = 3), all patients ultimately received a prosthetic valve and in one patient an additional cavopulmonary anastomosis was added because of right ventricular failure. In one adult patient (Carpentier class II and without ring-annuloplasty at primary repair), the plication was torn near the suture line several days post-repair, probably due to too much tension on the suture. Because of inadequate valve tissue, a re-repair was impossible and the valve was replaced. The other two early reoperations (both Carpentier class III) had regurgitation grade II/IV after repair and demonstrated progression in the following months. One patient had a ring-annuloplasty at the primary repair, while the other patient was 4 years old at the time of primary repair and no ring-annuloplasty was performed because of growth potential reasons. In the first patient re-repair proved impossible due to inadequate valve tissue and this patient eventually received a prosthetic valve. The last patient suffered from progressive right ventricular failure and ended up with a Fontan circulation.
In the late reoperation group, all three patients underwent a successful re-repair of the valve. The mechanism of recurrent tricuspid incompetence in these patients was progression of right ventricular dilation and right ventricular failure, resulting in tricuspid annulus dilation and subsequent tricuspid incompetence. In one patient (without primary annuloplasty) stress on the neo-annulus may have resulted in tearing of the vertical plication, while in the two other patients (also without addition of ring-annuloplasty at the time of primary repair), tricuspid incompetence was associated with neo-annular dilation. In two of these patients, a ring-annuloplasty was performed at reoperation, strengthening the annulus, while also in two patients, a partial cavopulmonary connection was added to alleviate the failing right ventricle. In all of the late reoperations, no ring-annuloplasty was added during primary repair (All of them were operated within the first 4 years of this study, one patient was 23 years old at primary repair, while the other patients were 6 and 9 years old). Therefore, a ring-annuloplasty to reinforce the annulus is definitely an option in adults. In children, addition of a ring-annuloplasty is not always feasible because of the necessity of growth potential. After the addition of a ring-annuloplasty to the primary repair, no late reoperations occurred.
All six patients were successfully reoperated and all six patients are doing well at the end of follow-up. Four patients are in NYHA functional class I and the remaining two patients are in functional class II. One of these patients has a 1.5 ventricle repair, while the other patient has a Fontan circulation, so a slight increase in functional class could be anticipated.
4.4 Conclusion
After 1 year of follow-up, patients were generally in good clinical and functional condition and displayed a persistent decrease in tricuspid incompetence on echocardiography, in addition to an increase in exercise capacity. At the end of follow-up, no decline in clinical performance or echocardiographic measurements was observed, indicating a durable repair. This is in line with results from Chauvaud et al. [7] and Da Silva et al. [9], who described persistent reduction of tricuspid incompetence and good functional outcome. Chen et al. [8] described favourable early and medium-term results using the same repair technique with good functional outcomes despite residual tricuspid incompetence.
Therefore, we conclude that this technique of Ebstein repair can be performed with low mortality and morbidity and an acceptable rate of reoperation. This was associated with a persistent decrease in tricuspid incompetence and preservation of right ventricular function and good functional outcome.
|
References |
|---|
|
TopAbstract1. Introduction2. Materials and methods3. Results4. DiscussionReferences |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
