|
|
||||||||
J Thorac Cardiovasc Surg 2005;130:1071
© 2005 The American Association for Thoracic Surgery
Surgery for Congenital Heart Disease |
a Royal Brompton Hospital, London, United Kingdom
b National Heart and Lung Institute, Imperial College, London, United Kingdom
c Wessex Cardiothoracic Centre, Southampton General Hospital, Southampton, United Kingdom
d London School of Hygiene and Tropical Medicine, London, United Kingdom
e Institute of Child Health, University College, University of London, London, United Kingdom
f Children's Hospital of Pittsburgh, Pittsburgh, Pa.
Received for publication September 30, 2004; revisions received May 15, 2005; accepted for publication May 18, 2005. * Address for reprints: Piers Daubeney, MRCP, Royal Brompton Hospital, Sydney St, London, SW3 6NP. (Email: p.daubeney{at}rbh.nthames.nhs.uk).
| Abstract |
|---|
|
|
|---|
METHODS: The UK and Ireland Collaborative study of Pulmonary Atresia with Intact Ventricular Septum is an ongoing population-based study of all patients born with this disease from 1991 through 1995. All available clinical, morphologic, and investigative variables were directly reviewed, and risk factor analysis was performed for poor outcome.
RESULTS: One hundred eighty-three patients presented with pulmonary atresia with intact ventricular septum. Fifteen underwent no procedure, and all died. Of the remainder, 67 underwent a right ventricular outflow tract procedure (catheter or surgical), 18 underwent an outflow tract procedure with shunt, and 81 underwent a systemic-to-pulmonary shunt alone. One- and 5-year survival was 70.8% and 63.8%, respectively. Results from Cox proportional hazards model analysis showed that low birth weight (P = .024), unipartite right ventricular morphology (P = .001), and the presence of a dilated right ventricle (P < .001) were independent risk factors for death. The presence of coronary artery fistulae, right ventricular dependence, or the tricuspid valvar z score did not prove to be risk factors for death. After up to 9 years of follow-up, 29% have achieved a biventricular repair, 3% a so-called one-and-a-half ventricular repair, and 10.5% a univentricular repair, with 16.5% still having a mixed circulation (41% died).
CONCLUSIONS: This population-based study has shown which features at presentation place an infant in a high-risk group. This is important information for counseling in fetal life and for surgical strategy after birth.
| Introduction |
|---|
|
|
|---|
| Methods |
|---|
|
|
|---|
This was an observational study, and no attempt was made to influence management by investigators. The primary outcome measure was death. Risk factors were sought from the list in Appendix E1 online. Continuous risk factors were expressed as medians with ranges. The Cox proportional hazards model was used to investigate risk factors for death. A forward-step procedure was used to select prognostic factors. Risk factors with P values of less than .1 in the univariate analyses were included in the multivariate models. In the multivariate models P values of less than .05 were considered significant.
| Results |
|---|
|
|
|---|
Pathology
The range of morphology in this population has been described in a previous publication
7
and is summarized in Table 1.
|
years and died without undergoing an operation.
Balloon Atrial Septostomy
Balloon atrial septostomy was performed in 45 (25%) patients: 30 before the first surgical or catheter procedure, 14 after the first surgical or catheter procedure, and 1 as the only intervention. The majority (36/45) of these patients underwent construction of systemic-to-pulmonary shunts. Three underwent a transannular patch procedure, 3 underwent a valvotomy with shunt, 2 underwent catheter intervention, and 1 underwent no further procedure.
Of 81 patients undergoing a systemic-to-pulmonary shunt alone as the primary surgical procedure, 26 (32%) had a septostomy beforehand (2 of whom also underwent an atrial septectomy at the time of the shunt). After the initial shunt procedure, 10 patients underwent a septostomy between 0 and 161 days later, and 4 patients required atrial septectomy between 2 and 339 days later (2 of whom had a septostomy before the shunt and 1 who had a septostomy after). Freedom from septostomy-septectomy 4 and 8 months after a systemic-to-pulmonary shunt alone was 82.1% (95% confidence interval [CI], 71.4%-92.8%) and 75.8% (95% CI, 62.8%-88.7%), respectively, for patients who had not undergone a septostomy beforehand (censoring for subsequent outflow tract procedure, cavopulmonary anastomosis with septectomy, or death).
Two patients died as a consequence of a restrictive atrial septum, one of them 2 days after a shunt, having had an inadequate septostomy beforehand. Another who had not undergone a septostomy or septectomy presented 159 days after a shunt with a restrictive intra-atrial septum bulging to the left and died as a consequence. Both died immediately after an emergency atrial septectomy.
Primary Procedure
The primary procedure, excluding balloon atrial septostomy, comprised a systemic-to-pulmonary shunt in 81 patients, percutaneous transcatheter valvotomy in 40 patients, and surgical outflow tract reconstruction alone in 27 patients and with a concomitant shunt in another 18 patients. One patient had a stent placed in the arterial duct, and another underwent oversewing of the TV, plication of the right atrium, and construction of a systemic-to-pulmonary shunt (Starnes procedure
24
). These are shown in Table 2. None was listed for transplantation.
|
|
|
|
|
Outcome in High-Risk Groups
Patients with unipartite anatomy of the RV cavity at presentation who underwent a procedure had a particularly poor outcome, with 7 of 9 dying (some at subsequent procedures; Figure 1). These were procedure related in 2 patients (during insertion of a central shunt in 1 patient and after a Fontan procedure in another), sudden arrest in 2 patients, blocked shunt in 1 patient, after an RV outflow tract procedure in another patient having had a large shunt and coil occlusion of major RV-to-coronary connections, and necrotizing enterocolitis in 1 patient (after the placement of a large shunt). Similarly, 4 of 5 patients presenting with a significantly dilated RV and undergoing a procedure died (all early after the first procedure). Four of 6 patients with birth weights of less than 2 kg and undergoing a procedure died (1 as a consequence of a postoperative seizure, 1 as a consequence of an oversized shunt, 1 after an eventual catheter procedure at 4 months, and 1 suddenly at home 4 months after a successful initial procedure). Of these 6 patients, only 2 underwent an initial procedure within 7 days of birth (1 death), with the other 4 waiting up to 111 days (3 deaths). Three others with birth weights of less than 2 kg died without undergoing procedures (Cornelia de Lange, severe Ebstein's malformation, and severe and lethal apnea caused by prostaglandin).
The presence of an RV-dependent coronary circulation was not found to be an operative risk factor for poor outcome, although the number of patients in this group was small. Ten patients were found to have this morphology: 2 of 7 died after surgical intervention, and 3 did not undergo a procedure and died.
Achievement of Separated Circulations
At latest follow-up of up to 9 years, 53 (29%) of 183 patients have achieved a biventricular repair, 6 (3%) have achieved a one-and-a-half ventricular repair, 19 (10.5%) have achieved a univentricular repair, and 30 (16.5%) are still in a mixed circulation. Seventy-five (41%) have died. For those achieving a biventricular circulation, 44 of 53 underwent an initial RV outflow procedure (catheter or surgical), 5 underwent an outflow tract procedure with shunt, and 4 underwent an initial shunt alone. For those achieving a one-and-a-half ventricular repair (n = 6), 2 underwent an RV outflow procedure alone, 2 underwent an RV outflow procedure and shunt, and 2 underwent a shunt alone. For those achieving a univentricular repair, 18 of 19 had an initial shunt, and 1 had an RV outflow procedure.
| Discussion |
|---|
|
|
|---|
Because of its rarity, many studies of PAIVS contain relatively few patients but are contemporary; others achieve larger numbers by spanning many years, making them essentially historical. In addition, single-institution studies might have ascertainment bias because of specific referral patterns. Multi-institutional studies might be more powerful in delineating the relative merits of treatment strategies,
9
and they might allow identification of risk factors for specific outcome measures.
10,11,13
This is generally not possible in smaller single-institution studies. The strengths of the UK and Ireland Collaborative Study comprise inclusion of all children identified with the disease at birth, the large number of patients enrolled, its population-based nature (thus avoiding selection bias), and the high quality of the data collected facilitated by a single investigator visiting all study sites and directly reviewing investigations, ensuring completeness, as well as quality, of data collection.
Balloon Atrial Septostomy
Conventional management of infants undergoing a systemic-to-pulmonary shunt is that a septostomy should be performed or at least considered before the shunt. In this study clinicians elected in some cases not to perform this procedure, presumably on the grounds that there was an adequate passage across the atrial septum. One quarter of patients not undergoing a septostomy before a shunt nonetheless required a septostomy or septectomy alone by 8 months. More concerning was that 2 deaths were directly attributable to a restrictive atrial septum, one occurring 2 days after a shunt and the other occurring 5 months after a shunt.
Risk Factors at Presentation for Subsequent Death
Several clinical and morphologic features at presentation proved to be independent predictors of subsequent death: so-called unipartite anatomy (and, to some degree, bipartite anatomy), low birth weight, and significant RV dilatation. Systemic-to-pulmonary shunting as the primary procedure was also associated with subsequent risk of death in univariate, but not multivariate, analyses, but this is likely to be because it was primarily performed in patients with greater degrees of RV hypoplasia. It is unclear why unipartite RV morphology per se should be a risk factor for death, especially because neither RV-dependent coronary circulation nor fistulae between the RV and coronary arteries were identified as risk factors for death, and all were progressing toward a univentricular repair after an initial shunt procedure. Because this was an observational study, there was no requirement for clinicians to perform a cardiac catheterization before or after an initial shunt procedure (actually, a cardiac catheter was performed in 54% before the first procedure and in 85% at any time). Therefore it might be that important fistulae and stenoses were overlooked in those who died. Alternatively, the functionally unipartite ventricle itself might represent the most extreme morphologic subtype in this disease, with consequences as yet unknown.
Low birth weight is perhaps a more understandable risk factor for death, irrespective of whether the patient underwent an operation. Weight at the time of the procedure has been identified by all of the larger studies as an independent risk factor,
8-10,13
and low birth weight has been identified as a risk factor by the other major population-based study.
11
Only one third with a birth weight of less than 2 kg survived. In some centers there was a strategy of immediate operation. In others, an attempt was made for them to grow, both with poor outcome. This is a controversial topic in all forms of congenital cardiac disease. Recent reports suggest that early surgical intervention in infants born with low weight is technically possible with good results and indeed is preferable.
29
Significant RV dilatation was also found to be a poor prognostic indicator. Our previous study of the range of morphology in this condition revealed that RV dilatation was associated with abnormalities of the TV, such as dysplasia and Ebstein's malformation, and hypoplastic pulmonary arteries.
7
Coles and colleagues
8
found that the presence of Ebstein's malformation was an independent risk factor for poor outcome. Many such patients die in utero,
30
and when they survive to birth, their management is unclear. Some groups have advocated patching off the TV and performing a systemic-to-pulmonary shunt.
24
This was performed in 1 patient without success but perhaps should have been undertaken in more patients. Other investigators have removed this group from their analysis,
10
but we elected to include them because PAIVS is already a very heterogeneous phenotype.
It was perhaps surprising that morphologic features identified as risk factors for death by other groups were not confirmed in this study. Hanley and associates
10
reported that the TV z score could be used as a guide to initial procedure, but neither TV diameter or z score nor RV inlet length or z score predicted death in the current series. This might support those who have criticized basing an operative strategy on a single anatomic index and, in particular, might indicate that the TV might not always reflect the size of the RV. Ashburn and colleagues,
13
in the Congenital Heart Surgeons Study, the largest study to date, concluded that although TV is a primary determinant of type of repair, other variables need to be considered.
Alternatively, it might be that clinicians in the United Kingdom and Ireland have already been tailoring their approach to the initial procedure on the basis on the size of the TV. The reasons why RV dependence was not identified as a risk factor, unlike in some other studies,
10
have already been alluded to (RV angiography is not always performed before surgical intervention), but alternatively, clinicians might be already tailoring their approach to this feature by not decompressing the RV in these situations.
14,16
Medium-Term Outcome
The UK and Ireland Collaborative Study of PAIVS is ongoing, and at the most recent follow-up, only two thirds of patients have reached a definitive separated circulation. Therefore it is too early to identify risk factors at presentation for achievement of these circulations. This is an important difference compared with the Congenital Heart Surgeons Study, in which all had achieved a definitive repair by 15 years.
13
This might reflect a different philosophy in the timing of transition from a superior to a total cavopulmonary anastomosis in the United Kingdom and Ireland. By contrast, there is a remarkable similarity in the proportion of patients who have died or achieved a biventricular or one-and-a-half ventricular repair. It is of additional interest that in a study with up to 9 years' follow-up, there have been no deaths in patients older than 6 years.
Study Limitations
This was an observational study, with no attempt to influence management. This results in institutional variations, including differing strategies, nonconformity of surgical or transcatheter techniques, variation in procedural expertise, and standards of postoperative care. These differences can confound the interpretation of results in a multi-institutional study such as this. As discussed previously,
7
the incidence of PAIVS in our study was 4.5 cases per 100,000 live births, lower than other estimates. This was due to the development in the United Kingdom of fetal diagnosis and subsequent termination of pregnancy. Our previous work did not observe a difference, however, in the range of morphology at birth between those detected and not detected prenatally.
7
| Conclusion |
|---|
|
|
|---|
We thank the pediatric cardiologists, cardiac surgeons, pathologists, and staff of the participating institutions in the United Kingdom and Ireland for their cooperation and help during the study. Participating institutions are listed online in Acknowledgments E1.
| Appendix E1 |
|---|
|
|
|---|
Patient risk factors: Date of birth (168), birth weight (163), sex (168), and race (58); height (164), weight (164), and body surface area (164) at initial procedure.
Morphologic variables: Presence of RV-coronary fistulae (125), presence of coronary artery stenoses, ectasia or interruptions (121), Ebstein's malformation (168), type of atresia (membranous-muscular; 151), number of portions of RV not overgrown by muscular hypertrophy (so-called tripartite, bipartite, or unipartite ventricle; 135), RV dilatation (168), TV z score (145), RV inlet z score (145), angle that duct subtends on descending aorta (102).
Investigative variables: Cardiothoracic ratio on admission chest radiogram (110), right atrial (38) and ventricular (52) pressures before the initial procedure.
Clinical variables: Patient intubated on admission (114), balloon atrial septostomy performed before or after primary procedure (168).
Procedural variables: Initial surgical or catheter intervention (168).
| Acknowledgements E1 |
|---|
|
|
|---|
| Footnotes |
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
H M Gardiner The case for fetal cardiac intervention Heart, October 15, 2009; 95(20): 1648 - 1652. [Abstract] [Full Text] [PDF] |
||||
![]() |
M Marasini, P F Gorrieri, G Tuo, L Zannini, P Guido, M Pellegrini, S Bondanza, M G Calevo, and G Pongiglione Long-term results of catheter-based treatment of pulmonary atresia and intact ventricular septum Heart, September 15, 2009; 95(18): 1520 - 1524. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. Krasemann Radiofrequency perforation: the optimum treatment for pulmonary atresia? Heart, September 15, 2009; 95(18): 1473 - 1474. [Full Text] [PDF] |
||||
![]() |
R. L. Hannan, J. A. Zabinsky, R. M. Stanfill, R. A. Ventura, A. F. Rossi, D. G. Nykanen, E. M. Zahn, and R. P. Burke Midterm Results for Collaborative Treatment of Pulmonary Atresia With Intact Ventricular Septum Ann. Thorac. Surg., April 1, 2009; 87(4): 1227 - 1233. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. A Shinebourne, M. L Rigby, and J. S Carvalho Pulmonary atresia with intact ventricular septum: from fetus to adult Heart, October 1, 2008; 94(10): 1350 - 1357. [Full Text] [PDF] |
||||
![]() |
S. Mohammadi, O. Benhameid, A. Campbell, J. Potts, J. Joza, H. Al-Habib, S. Sett, and J. Le Blanc Could we still improve early and interim outcome after prosthetic systemic-pulmonary shunt? A risk factors analysis Eur. J. Cardiothorac. Surg., September 1, 2008; 34(3): 545 - 549. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. Bryant III, E. R. Nowicki, R. B.B. Mee, J. Rajeswaran, B. W. Duncan, G. L. Rosenthal, U. Mohan, M. Mumtaz, and E. H. Blackstone Success and limitations of right ventricular sinus myectomy for pulmonary atresia with intact ventricular septum J. Thorac. Cardiovasc. Surg., September 1, 2008; 136(3): 735 - 742. [Abstract] [Full Text] [PDF] |
||||
![]() |
H. M. Gardiner, C. Belmar, G. Tulzer, A. Barlow, L. Pasquini, J. S. Carvalho, P. E.F. Daubeney, M. L. Rigby, F. Gordon, E. Kulinskaya, et al. Morphologic and Functional Predictors of Eventual Circulation in the Fetus With Pulmonary Atresia or Critical Pulmonary Stenosis With Intact Septum J. Am. Coll. Cardiol., April 1, 2008; 51(13): 1299 - 1308. [Abstract] [Full Text] [PDF] |
||||
![]() |
J.-H. Yang, T.-G. Jun, P. W. Park, K. Sung, W. S. Kim, Y. T. Lee, J. Huh, and I-S. Kang Exclusion of the non-functioning right ventricle in children with pulmonary atresia and intact ventricular septum Eur. J. Cardiothorac. Surg., February 1, 2008; 33(2): 251 - 256. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. A. Webber New-Onset Heart Failure in Children in the Absence of Structural Congenital Heart Disease Circulation, January 1, 2008; 117(1): 11 - 12. [Full Text] [PDF] |
||||
![]() |
G. Butera Is there any difference between surgical outflow reconstruction and transcatheter valvotomy in patients with pulmonary atresia with intact ventricular septum? J. Thorac. Cardiovasc. Surg., August 1, 2006; 132(2): 444 - 444. [Full Text] [PDF] |
||||
![]() |
J. W. Salvin, D. B. McElhinney, S. D. Colan, K. Gauvreau, P. J. del Nido, K. J. Jenkins, J. E. Lock, and W. Tworetzky Fetal Tricuspid Valve Size and Growth as Predictors of Outcome in Pulmonary Atresia With Intact Ventricular Septum Pediatrics, August 1, 2006; 118(2): e415 - e420. [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 |