| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Eur J Cardiothorac Surg 2006;29:397-405
© 2006 Elsevier Science NL
a Paediatric Cardiac Unit, Paediatric Cardiology and Cardiac Surgery Department, Birmingham Children's Hospital, B4 6LT Birmingham, UK
b Paediatric Cardiology Department, Glenfield hospital, Leicester, UK
Received 5 August 2005; received in revised form 14 November 2005; accepted 15 November 2005.
* Corresponding author. Tel.: +44 121 3339437; fax: +44 121 3339441. (Email: david.barron{at}bch.nhs.uk).
 |
Abstract |
|---|
 Top Abstract 1. Introduction2. Methods3. Results4. Discussion5. ConclusionReferences |
|---|
Key Words: Bovine Contegra® valved conduit • Right ventricular outflow tract reconstruction • Right ventricular to left ventricular pressure ratio
|
1. Introduction |
|---|
|
TopAbstract1. Introduction2. Methods3. Results4. Discussion5. ConclusionReferences |
|---|
The recently developed Contegra® valved bovine conduit (Medtronic Inc., Minneapolis, MN, USA) has been advocated for its ‘off-the-shelf’ availability, small sizes, surgical pliability and encouraging short-term success in experimental animal studies [6–10].
Early results in humans have been encouraging [11–14], although there is limited information for smaller children and infants. A more recent study by Meyns et al. [15] has highlighted the potential problem of supravalvar conduit stenosis developing during intermediate follow-up, particularly in the smaller conduits. Most of these studies did not include significant numbers of infants undergoing primary repair of complex pulmonary atresia (i.e., involving unifocalisation of multiple aorto-pulmonary collateral arteries (MAPCAs)) where high RV/LV pressure ratios may be anticipated and pose many technical challenges. This paper analysed our experience with the Contegra conduit with particular focus on this more complex group of patients.
The objective of this study was to define the natural history of the Contegra® valved bovine conduit for RVOT reconstruction and to assess its performance, particularly with respect to younger children, smaller conduit sizes and in relation to post-operative right ventricular and pulmonary pressures.
|
2. Methods |
|---|
|
TopAbstract1. Introduction2. Methods3. Results4. Discussion5. ConclusionReferences |
|---|
Aspirin was used if additional patching had been performed on the pulmonary arteries; in 46 out of the 64 conduits (71.8%).
2.2 Study group
The patient group encompassed a broad spectrum of conditions and diagnoses. The patient details are given in Table 1
, dividing patients into the main diagnostic categories. Of note, 18 patients (29%) with PA/VSD/MAPCA were included in the study, reflecting the referral pattern at this institution.
|
Over half (60%) of the implanted conduits were of the smallest sizes (38% were 12 mm in size) (Fig. 1 ).
|
2.5 Echocardiography
The follow-up echocardiographic protocol was designed, taking into consideration the 1996 Guidelines from the American society of cardiothoracic surgeons [16]. To avoid error bias, a single paediatric cardiologist (S.S.) performed the echocardiograms at every time point in this institution. The protocol was sent out to every referring centre to guide the follow-up after discharge from our institution.
Studies were conducted intra-operatively in theatre, on the intensive care unit within 24 h of surgery, immediately prior to discharge from hospital and then subsequently at 1, 3, 6, 12 months, and annually thereafter.
Right ventricular pressure was quantitatively estimated from the tricuspid valve regurgitant velocity using the modified Bernoulli equation. Additionally, a subjective assessment of RV function was graded with a single observer as poor/moderate/good.
All the conduits were unsupported conduits; conduit diameter was measured at the level of the valve, above the valve and below the valve. A mean of the three measurements was taken and was expressed as the percentage of diameter difference between the actual diameter and the manufacturer-labelled dimensions at implantation (DD%).
The degree of conduit regurgitation was qualitatively assessed by colour flow Doppler, graded 1–4 regarding grades 3 and 4 as significant. Regurgitant fraction (RF) was quantitatively assessed by pulsed wave Doppler at a point just distal to the Contegra® conduit valve [17]. RF > 40% was considered significant.
Conduit valve stenosis was diagnosed if the mean pressure gradient across the valve was 
30 mmHg.
RV pressure was assessed regularly throughout the study. In the immediate post-operative period, it was measured directly in theatre and echocardiographically as above within 24 h post-operatively. Subsequently, RV pressure was assessed at each echocardiographic examination based on tricuspid regurgitation velocity and measured invasively when catheterisation was performed. Patients with RV/LV pressure ratio 60% at the end of the operation that persisted in the early post-operative period were put forward for elective catheterisation. The RV/LV pressure ratio was divided into quartiles from 1 to 4 which were equivalent to ratio of <0.4 for quartile 1 through to a ratio of >0.6 for quartile 4. Pulmonary artery pressure was the pressure measured distal to the conduit and was divided into four quartiles, the highest being quartile 4 (equivalent to 38–88 mmHg).
2.6 Cardiac catheterisation
This was performed either electively for the PA/VSD/MAPCA group or when there was evidence of RVOT obstruction, branch pulmonary artery stenosis, significant conduit dilatation or unexplained impairment of RV function.
The site of stenoses was classed as either conduit-related or branch pulmonary artery-related (i.e., distally). Further localisation of any stenosis within the conduit is described in Section 3.
The explanted conduits were sent for detailed histological examination by an independent pathologist who had specific experience with the Contegra® valves and who was anonymous to the patients and the purpose of the study.
2.7 Data analysis
Actuarial survival, freedom from reintervention based on age, diagnostic group, RV/LV pressure ratio and PAP were estimated by the Kaplan–Meier method, using SPSS for Windows (version 11 Inc.). Sigma-plot software was used to plot survival curves.
Continuous variables were expressed as mean ± SD or median range, and binomial or ordinal data were expressed as percentages. Comparative univariate analyses were performed with the 
2-test, two-sided Fisher exact test or binomial logistic regression, as appropriate. A probability value p
< 0.05 was taken to represent a statistically significant difference between groups.
The effect of pre-operative, operative and post-operative variables on the outcome was tested by univariate and multivariate analyses. Univariate analysis of early outcome measures was done using the tests as above and variables with p
0.1 were included in a stepwise logistic regression model. Results of the multivariate analysis were expressed as odds ratios (OR) with 95% CI for variables with p
< 0.05.
2.8 Outcome measures
Outcome measures were early mortality (defined as death within 30 days of surgery), late death, survival, freedom from reintervention and NYHA functional status.
|
3. Results |
|---|
|
TopAbstract1. Introduction2. Methods3. Results4. Discussion5. ConclusionReferences |
|---|
Early (30-day) survival was 94% (four deaths) and actuarial survival at 3 years was 84 ± 7% (three late deaths). The majority of patients (94.5%) were in NYHA class I at final follow-up (Fig. 2 ).
|
Early deaths: case 1 was a 3-week neonate with PA/VSD/MAPCAs who underwent repair with a 12 mm conduit and arrested suddenly on day 5 with profound hypoxia and was found to have a completely thrombosed conduit at post-mortem. Case 2 with PA/VSD/MAPCAs was 17 months old who died on post-operative day 7 from necrotising enterocolitis, gut perforation and sepsis. Case 3 was a 5.3 kg patient with Fallot's tetralogy, disconnected LPA and acquired pulmonary atresia that arrested on day 1 with low cardiac output. Case 4 was a patient with ccTGA, aortic atresia and severe ebsteinoid tricuspid valve who underwent Damus–Kaye/Senning/VSD closure with an RV-PA conduit of size 12 mm who died on day 1 from low cardiac output and RV failure. The Contegra® conduits in these three patients were patent with no thrombus.
Late deaths: the first patient was 6 months old with right atrial isomerism who died from overwhelming pneumococcal sepsis at 2 months post-operative (despite penicillin prophylaxis). The second patient was 2 months old following neonatal biventricular repair of aortic atresia/VSD who also had tracheo-oesophageal fistula repair and died suddenly on the paediatric ward; ECG and echocardiography had been satisfactory and post-mortem was declined. The third patient was 10 years old who had received a redo conduit replacement for PA/VSD 22 months earlier and had poor right ventricular function and normal conduit performance. She suffered a sudden arrhythmic death.
3.2 Freedom from reintervention
Overall freedom from reintervention (both catheter and surgical) was 53 ± 11% at 38 months. Reinterventions could be divided into those that were conduit-related and those that were related to distal branch pulmonary artery stenoses (exclusively in the PA/VSD/MAPCA group). There were 10 catheter reinterventions for distal (branch) pulmonary artery stenoses with no surgical reinterventions.
3.3 Conduit-specific freedom from reintervention
Conduit-specific freedom from reintervention was 66 ± 11% at 38 months. Six of these were percutaneous catheter interventions for acquired conduit stenosis and four were surgical replacements—one for (early) endocarditis and three for a combination of aneurismal dilatation with distal conduit stenosis.
Univariate analysis identified RV/LV pressure ratio at the end of the operation to be the strongest single factor associated with the need for reintervention (p = 0.0003).
Smaller conduit size (12 and 14 mm, p = 0.04), young age (<1 year, p = 0.04) and severe conduit regurgitation (regurgitant fraction > 40%, p = 0.016) were the only other significant factors. The only diagnostic group that had a significantly higher reintervention rate was the PA/VSD/MAPCA group which was mainly related to distal branch PA stenosis (p < 0.002, Fig. 3c).
|
3.4 Conduit stenoses
All stenosis requiring reintervention was an acquired stenosis in the distal conduit involving the site of the distal anastamosis to the branch pulmonary arteries. In addition, there were five conduits (8.6%) that developed mild to moderate valvular stenosis during the study. None required reintervention and none progressed during the period of the study.
3.5 Conduit dilatation and regurgitation
A total of 16 conduits (27.5%) developed significant dilatation during the period of the study (defined by DD% > 30%). All of these conduits developed severe regurgitation which appeared to be a secondary phenomenon to the dilatation as the valve cusps themselves were intact and there had not been any regurgitation prior to the dilatation. A high RV/LV pressure ratio was associated with the dilatation in 11 patients, 10 of whom required subsequent reintervention. Three of these patients had such severe dilatation (DD% > 60%) that it was regarded as the primary indication for urgent conduit replacement. All were associated with high RV pressures and stenosis in the distal conduit.
The remaining five conduits that developed dilatation in the absence of high RV pressures have not required any reintervention to date, but remain under close review. The relationship between conduit dilatation and time is shown in Fig. 4 , showing no significant correlation. Dilatation appeared to be influenced by RV/LV pressure rather than by any other factor.
|
3.7 Conduit endocarditis
There was one episode of endocarditis occurring just 2 weeks after implantation. This required urgent conduit replacement (with a second Contegra® conduit) and went on to make an uneventful recovery. The organism was a Staphylococcus aureus that was isolated on blood culture and subsequently from the conduit with no clearly apparent source.
3.8 Summary
Smaller conduit size (12 and 14 mm), young age at operation (<1 year), conduit dilatation/regurgitation and high RV/LV pressure ratio at the end of the procedure (>0.6) were the most significant factors in predicting the need for reintervention. The latter two are probably interrelated.
3.9 Histopathology of the explanted conduits
The three conduits explanted for aneurismal dilatation showed similar histopathological findings. There was a generalised layer of thrombus covering the neointima, degeneration of elastin fibers within the conduit itself, with additional infiltration with inflammatory and giant cells and diffuse but mild calcification. The patient whose conduit is illustrated in Fig. 5
had PA/VSD/MAPCAs and had undergone single stage repair at 8 months of age (14 mm conduit). RV/LV pressure ratio at the end of the operation was 60%. Elective cardiac catheterisation with balloon angioplasty of distal branch PA stenosis was undertaken 3 months later but there was progressive dilatation of the conduit (diameter difference 93%), and RV/LV pressure ratio increased to 80% leading to explantation at 9 months. Superficial external examination of the conduit was relatively unremarkable, with apparent mild intimal thickening superior to one valve cusp and mild thickening of vein conduit wall. There was no obvious thrombosis seen but microscopic examination showed diffuse conduit degeneration with multifocal elastic fiber fragmentation, calcification and dropout. There was also evidence of chronic neointimal and adventitial inflammation, with necrosis, luminal surface fibrin deposition and thickening of both neointima and adventitia. The valve cusps were covered by fibrin-rich thrombus layer (of varying thickness) associated with chronic inflammatory cells. At the distal anastamotic suture line, there was focal elastic fiber calcification.
|
|
4. Discussion |
|---|
|
TopAbstract1. Introduction2. Methods3. Results4. Discussion5. ConclusionReferences |
|---|
The current commercially available bovine graft, Contegra®, is available for clinical use in diameters between 12 and 22 mm (supported and unsupported versions). Overall length is 10–12 cm but the 12 mm graft has a length of around 7 cm and a conduit valve closer to the outflow. The grafts have been treated with buffered glutaraldehyde without either additional anticalcification or antithrombogenic preparation.
Initially published clinical data in humans have been encouraging [11–14]. Breymann et al. [12] compared the performance of 71 Contegra conduits with 52 homografts and 30 porcine xenografts. Their Contegra group showed no signs of conduit or valve degeneration over 27 months. There were five reoperations for peripheral branch pulmonary stenosis although no histopathology data were available. Breymann and co-workers [18,19] have recently updated their results, with a total of 108 Contegra conduits (the largest available single institute experience) extending the mean follow-up to 4 years, and still maintaining excellent results (see Table 2 ) although they did report 10 cases of supravalvar stenosis (9.3%). This is similar to the incidence of acquired conduit stenosis at the distal suture line reported in this series (11.6%).
|
Significant conduit dilatation was not originally reported as a concern, other than one case by Carrel et al. [14] in a series of 22 conduits and one case from Boudjemline et al. [20]. More recently, conduit dilation and thrombosis have featured more prominently [21,22] despite rather short follow-up. Subsequently, Meyns et al. [15] drew the attention to the neointimal proliferation at the distal anastomosis of the conduit, leading to severe stenosis, dilatation of the proximal conduits and a freedom from stenosis of only 49 ± 8% at 24 months.
We have specifically examined performance of the 12 and 14 mm conduits which are particularly valuable due to the scarcity of small homografts. Thirty-eight percent of the conduits were 12 mm and 60% were either 12 or 14 mm in diameter.
Conduit narrowing at the pulmonary anastomosis (distal suture line) was relatively common in this series, and predominantly occurring in the smaller sized conduits (12 and 14 mm). Similar findings have been reported by Myens et al. [15], together with the development of conduit dilatation. This acquired stenosis in the distal part of the conduit may be partly related to discrepancies in circumference between the smaller conduits and the native pulmonary arteries, as it did not seem to be operator dependent, nor related to the underlying anatomy or the RV/LV pressure ratio at the end of the surgery. The histopathological findings of elastic fiber proliferation and calcification at this point, on top of the neointimal proliferation and thrombus layer covering the entire conduit, suggest an active process occurring within the neointima which may accelerate the development of stenosis and create resistance to balloon angioplasty.
Almost one third (27.5%) of the conduits developed a degree of dilatation over time in this series. In the majority of cases, including all those that became clinically significant, the dilatation appeared to be secondary to distal branch pulmonary artery stenosis, reflecting the much larger proportion of patients with PA/VSD/MAPCAs. Although dilatation is significantly related to high RV and intra-conduit pressure rather than to time, there is clear histological evidence of a chronic inflammatory process in the three explanted aneurysmal conduits.
However, a small number of conduits (8.5%) demonstrated an unpredictable dilatation in the absence of raised intra-conduit pressure. None of these patients have required reintervention but there is regurgitation across the conduit as a result and these patients will need to be carefully followed up. A variable degree of inflammatory response may explain why these conduits dilated, and it will be valuable to examine the histopathology of this low RV pressure group when these conduits come up for replacement. The Contegra® conduits in this series have been subjected to higher pressures than in most previous reports and there is no doubt that the dilatation appears to be a secondary phenomenon and not primary conduit failure. Nevertheless, this degree of dilatation is not seen in homografts or prosthetic conduits and we would recommend that these should be used in the setting where high RV pressures are predicted post-operatively.
Echocardiographic evidence of conduit thrombosis was an unexpected finding in this study, although it has been subsequently reported [21,22]. The valve cusps of the jugular vein graft are deeper than normal semilunar valve cusps, which might predispose to thrombosis in situ. Also, the histopathological study of the three explanted (dilated) conduits showed a complete layer of thrombus covering the neointima of the valve cusps and the conduit itself, which was not obvious macroscopically. Biological implants cross-linked with glutaraldehyde are potentially thrombogenic [23]. The residual glutaraldehyde released from the implant long after insertion hinders host cells from colonising the luminal layer exposing the implant to thrombogenic process [24]. Aspirin therapy is well known to reduce the risk of thrombosis in vein grafts [25] and Boudjemline et al. [21] are now prescribing an anti-aggregant therapy in all patients for at least 6–12 months post-implantation.
In this study, the non-fatal valve thromboses occurred in patients already receiving aspirin (the mean dose used was 2.5 mg kg–1 day–1). In all cases the thrombus had disappeared on follow-up. Although aspirin's influence on the outcome of the conduit was not statistically significant, concerns about thrombogenicity appear genuine from clinical point of view and in combination with histological evidence of an inflammatory process. Serious consideration should be given to the evaluation of the use of more effective anti-inflammatory/antithrombotic medication after implantation of these grafts.
The small homograft is still considered by most as the ‘gold standard’ for neonatal and infant reconstructions. However, concern was about the long-term performance (which as yet, we cannot provide for Contegra grafts) and availability [3–5] which led to the original development of the Contegra. The main advantages of the Contegra conduit is the off-the-shelf availability in a wide size range, with more versatile handling properties compared to prosthetic conduits. There is little evidence to support that even the small Contegra® conduits do not perform as well as small homografts [12,18], although the tendency for the Contegra® to dilate (albeit in the setting of high pressure) and the incidence of valve thrombosis is not a feature of the homograft.
4.1 Limitations of the study
Follow-up is still too short to make conclusive evaluation of conduit performance or clear statements regarding durability. There is no direct control group in this study and it will be particularly valuable to evaluate these conduits against small homografts in the future, ideally as a randomised trial. The complex pulmonary atresia/MAPCAs group are a major confounding factor as to the need for non-conduit-related interventions and it is difficult to separate conduit performance from inherent problems with the branch pulmonary arteries. Although analysis may have been more straightforward by excluding these patients, it is important to evaluate Contegra performance in this difficult subset of patients.
|
5. Conclusion |
|---|
|
TopAbstract1. Introduction2. Methods3. Results4. Discussion5. ConclusionReferences |
|---|
High pressure in the conduit may lead to aneurysmal dilatation and resultant valvar regurgitation. However, there is also a small unpredictable risk of conduit dilatation that is unrelated to pressure and also of thrombosis in the valve sinuses, neither of which appear to be clinically important but remain a cause for concern and will need further evaluation.
The findings of this study suggest that these conduits perform poorly under higher pressure conditions (particularly with PA/VSD/MAPCAs repair). We would recommend that the conduits are used with caution in patients in whom a high RV pressure is anticipated post-operatively. The advantage of the Contegra conduit over a homograft in smaller patients (<1 year) also remains uncertain. The role of anti-inflammatory and antithrombotic therapy needs to be evaluated. At present we would recommend indefinite aspirin therapy.
|
Acknowledgments |
|---|
|
References |
|---|
|
TopAbstract1. Introduction2. Methods3. Results4. Discussion5. ConclusionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  C. D. Derby, J. Kolcz, S. Gidding, and C. Pizarro Outcomes following non-valved autologous reconstruction of the right ventricular outflow tract in neonates and infants Eur. J. Cardiothorac. Surg., October 1, 2008; 34(4): 726 - 731. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Frigiola, V. Tsang, C. Bull, L. Coats, S. Khambadkone, G. Derrick, B. Mist, F. Walker, C. van Doorn, P. Bonhoeffer, et al. Biventricular Response After Pulmonary Valve Replacement for Right Ventricular Outflow Tract Dysfunction: Is Age a Predictor of Outcome? Circulation, September 30, 2008; 118(14_suppl_1): S182 - S190. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Frigiola, V. Tsang, J. Nordmeyer, P. Lurz, C. van Doorn, A. M. Taylor, P. Bonhoeffer, and M. de Leval Current approaches to pulmonary regurgitation Eur. J. Cardiothorac. Surg., September 1, 2008; 34(3): 576 - 581. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. J. Hickey, B. W. McCrindle, E. H. Blackstone, T. Yeh Jr., F. Pigula, D. Clarke, C. I. Tchervenkov, J. Hawkins, and the CHSS Pulmonary Conduit Working Group Jugular venous valved conduit (Contegra(R)) matches allograft performance in infant truncus arteriosus repair Eur. J. Cardiothorac. Surg., May 1, 2008; 33(5): 890 - 898. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. F. Corno and E. S. Mickaily-Huber Comparative computational fluid dynamic study of two distal Contegra conduit anastomoses Interactive CardioVascular and Thoracic Surgery, February 1, 2008; 7(1): 1 - 5. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Sinzobahamvya, B. Asfour, M. Boscheinen, J. Photiadis, C. Fink, E. Schindler, V. Hraska, and A. M. Brecher Compared fate of small-diameter Contegras(R) and homografts in the pulmonary position Eur. J. Cardiothorac. Surg., August 1, 2007; 32(2): 209 - 214. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. J. Rastan, T. Walther, I. Daehnert, J. Hambsch, F. W. Mohr, J. Janousek, and M. Kostelka Bovine jugular vein conduit for right ventricular outflow tract reconstruction: evaluation of risk factors for mid-term outcome. Ann. Thorac. Surg., October 1, 2006; 82(4): 1308 - 1315. [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 |
