Ann Thorac Surg 2004;77:29-35
© 2004 The Society of Thoracic Surgeons
Original article: cardiovascular
Successful early surgical recruitment of the congenitally disconnected pulmonary artery
David N. Murphy, MBBSa,
David S. Winlaw, MBBS, MDa*,
Steve G. Cooper, MB ChBa,
Graham R. Nunn, MBBSa
a Adolph Basser Cardiac Institute, Children's Hospital at Westmead, Sydney, Australia
Accepted for publication July 29, 2003.
* Address reprint requests to Dr Winlaw, Adolph Basser Cardiac Institute, Children's Hospital at Westmead, Locked Bag 4001, Westmead NSW 2145, Australia
e-mail: davidw{at}chw.edu.au
 |
Abstract
|
|---|
BACKGROUND: Our purpose was to document our experience with early recruitment of congenitally disconnected pulmonary arteries and to assess subsequent pulmonary artery growth and function.
METHODS: Patients born in the 10-year period from 1989 to 1999 with a disconnected pulmonary artery diagnosed in infancy and treated in our unit were studied. To be included patients had nonconfluent pulmonary arteries with one or both completely disconnected from the main pulmonary artery. This series did not include patients with acquired stenosis causing occlusion of a pulmonary artery. Echocardiography, cardiac catheterization, MRI, lung perfusion scans, and intraoperative assessment were used to gauge pulmonary artery growth and function.
RESULTS: Seven patients with a disconnected pulmonary artery associated with intracardiac conotruncal congenital cardiac disease underwent successful early surgical recruitment of the affected pulmonary artery at 3 months of age or younger. Median follow-up from date of first operation was 4.2 years (range, 1.6 to 13.4). All 7 patients had postrecruitment lung perfusion scans showing a mean of 44% (range, 27% to 78%) of total pulmonary flow through the affected lung. Significant growth in the diameter of the recruited native pulmonary artery was demonstrated in all patients. There were no deaths reported in our series to date.
CONCLUSIONS: The rare possibility of a congenitally disconnected pulmonary artery needs to be considered in all patients with a conotruncal cardiac anomaly. To facilitate surgical correction, ensure subsequent growth of the pulmonary artery, and optimize associated lung development, early diagnosis and surgical recruitment is recommended.
|
Introduction
|
|---|
In 1868 Frantzel and associates [1] first described the rare condition of absence of a main branch of the pulmonary artery. Sotomora and Edwards [2] in a postmortem study in 1978 noted that the distal portion and intrapulmonary vessels of the so-called "absent pulmonary artery" were often in fact present. They commented that attempted surgery to form an anastomosis or conduit into this distal portion of disconnected pulmonary artery to recruit blood flow to the affected lung could be beneficial to the patient in the long term.
The importance of early recruitment in this condition has previously been recognized [1, 3, 4]. Previous reports have focused on an older group of patients [1, 3, 5], raising the possibility that some patients with a disconnected pulmonary artery have either been missed or diagnosed late (beyond the age of 6 months). These reports highlighted the surgical difficulties with late recruitments, which were largely unsuccessful. There is paucity of published documentation in the literature to date of the pulmonary artery growth and lung function of patients who have undergone attempted resurrection of a disconnected pulmonary artery.
Pfefferkorn and associates [6] in 1978 published a study of 9 patients over a 7-year period with unilateral absence of a pulmonary artery. This paper centered on the importance of previous ductal flow and the loss of pulmonary arterial flow with duct closure after birth. The prevalence of a diverticulum of the innominate artery seen on cardiac catheterization was viewed as evidence of previous duct-dependent arterial flow. They suggested that the diverticulum represented a fetal systemic blood supply to the affected lung through the distal part of the sixth aortic arch.
In this paper we document the incidence of congenitally disconnected pulmonary arteries in our population. We detail our experience with the recruitment of disconnected pulmonary arteries and the subsequent effect on pulmonary artery growth and lung perfusion.
|
Patients and methods
|
|---|
Patients born in the 10-year period from 1989 to 1999, with a disconnected pulmonary artery diagnosed in infancy and treated at our institution, were studied.
To be included patients had nonconfluent pulmonary arteries with one or both congenitally completely disconnected from the main pulmonary artery. At the time of the initial recruitment operation the affected pulmonary artery must have been either completely disconnected with no arterial supply or functionally disconnected. The latter group had arterial supply only from a closing ductus arteriosus (as once the ductus arteriosus occluded, the untreated affected lung would be isolated from normal arterial flow and soon lost from the pulmonary circuit).
We excluded patients with anomalous origin of the pulmonary artery directly from the aorta (true hemitruncus arteriosus). Patients with an acquired stenosis causing occlusion of a main branch pulmonary artery were excluded. Patients with lung agenesis were also excluded.
A combination of imaging modalities including serial echocardiography, cardiac catheterization, lung perfusion scans, and magnetic resonance imaging (MRI) as well as intraoperative assessment were used to diagnose and visualize the disconnected pulmonary artery and gauge its subsequent growth and function after recruitment.
During the 10-year period there were a total of 7 patients. All were diagnosed in early infancy by 2 months of age and underwent the first recruitment operation early, at 3 months of age or younger. Of the 7 patients (Table 1),
the primary cardiac lesion was truncus arteriosus in 2 and tetralogy of Fallot in 3, one being of the absent pulmonary valve type. In 1 exceptional case of pulmonary atresia with major aortopulmonary collateral arteries to the right lung there was a true nonconfluent disconnected left pulmonary artery present. Interestingly 3 of the 7 had phenotypic features of the DiGeorge anomaly or velocardiofacial syndrome with cytogenetic confirmation of the associated microdeletion of chromosome 22q11. No other syndromes were diagnosed in the patient series.
Ipsilateral pulmonary venous wedge injection at cardiac catheterization under general anesthesia was confirmatory of the diagnosis in 2 patients. Of these cases, 1 patient (case 3) underwent diagnostic cardiac catheterization at 3 months of age (Fig 1)
and the other (case 4) at 3 days of age. In 4 cases the aortic arch was right-sided. In both cases of truncus arteriosus with a disconnected pulmonary artery the affected pulmonary artery was on the same side as the aortic arch. In 2 patients (cases 3 and 7) there was an aberrant subclavian artery, which in both cases was on the contralateral side to the aortic arch and arose from the descending aorta.
View larger version (135K):
[in this window]
[in a new window]
|
Fig 1. Prerecruitment cardiac catheterization (case 3) with a retrograde left upper pulmonary vein wedge injection showing the disconnected hilar left pulmonary artery.
|
|
Six cases had a disconnected left pulmonary artery and 1 case had a disconnected right pulmonary artery. One patient (case 5) had a complex right isomerism with nonconfluent pulmonary arteries initially supplied at birth by bilateral ductus arteriosus, with the left one closing. In this patient the left ductus arteriosus occluded before an operative attempt to recruit the disconnected left pulmonary artery at 2 days of age. The right ductus arteriosus remained patent and a synthetic systemic to right pulmonary artery shunt was electively sited 1 week later. Two other patients (cases 6 and 7) initially had a closing ductus arteriosus supplying the affected pulmonary artery on echocardiography and were included, as the affected lung was functionally isolated from arterial flow. Of the 4 remaining cases with disconnection of a pulmonary artery, a further 2 patients (cases 2 and 4) had intraoperative evidence of the affected pulmonary artery being connected to a ligamentum arteriosum.
Where a central shunt was created, a short (< 1 cm) 4-mm systemic to pulmonary shunt was used. When a modified Blalock-Taussig shunt was employed a 5-mm shunt was used. This originated from the subclavian artery, which itself acts as a flow limiter that grows with the child, allowing a proportional increase in pulmonary blood flow.
|
Results
|
|---|
The median follow-up from the date of first operation is 4.2 years (range, 1.6 to 13.4 years). There have been no deaths recorded in the series and all patients with suitable intracardiac anatomy have achieved biventricular repair. Patient 5 is being managed along a single ventricle pathway to a Fontan circulation in the setting of heterotaxy, pulmonary atresia, and a complete atrioventricular septal defect with a small right ventricle. None of the children appear to be limited in their exercise capacity although this was not formally tested. None of the cohort requires regular cardiac medications.
All 7 patients have had postrecruitment follow-up lung perfusion scans. The scans were performed at 10 months to 4 years of age. They show that a mean of 44% (range, 27 to 78%) of total pulmonary blood flow is through the affected lung (Table 2).
One patient, case 2, had a preoperative lung perfusion scan confirming complete absence of perfusion to the affected lung. His subsequent scan showed more than one third of the pulmonary arterial supply passed through the recruited lung (Fig 2).
In case 3 the majority of pulmonary blood flow passes through the originally disconnected pulmonary artery. In this patient, owing to the poor contralateral pulmonary artery blood supply associated with unfavorable aortopulmonary collaterals and a severely hypoplastic pulmonary artery on that side, the importance of the recreating of an adequate sized ipsilateral pulmonary artery is highlighted.
View larger version (78K):
[in this window]
[in a new window]
|
Fig 2. Lung perfusion scan (case 2) showing pulmonary perfusion after surgical recruitment of a disconnected left pulmonary artery.
|
|
Serial measurements of both the native disconnected and contralateral pulmonary arteries have also been used to assess growth (Table 3). These measurements were obtained by echocardiography, MRI, and intraoperatively at subsequent operations. Progress postrecruitment cardiac catheterizations were also used to assess growth of the native pulmonary arteries and monitor for potential stenosis at conduit anastomoses. The attached catheterization picture (Fig 3)
shows the subsequent growth of the recruited native left pulmonary in case 3. The age range at latest follow-up measurements is from 6 months to 11 years. Intraoperative assessment of recruited native pulmonary artery growth can be seen in the attached photographic pictures of case 6 (Fig 4).
In all 7 patients studied there has been significant growth in the diameter of the recruited native pulmonary artery.
View this table:
[in this window]
[in a new window]
|
Table 3. Follow-Up Assessment of Native Pulmonary Artery Growth by Prerecruitment and Postrecruitment Measurementsa
|
|
View larger version (171K):
[in this window]
[in a new window]
|
Fig 3. Postrecruitment cardiac catheterization (case 3) with an aortogram showing a systemic shunt from the aberrant left subclavian artery into a well-developed left pulmonary artery. In this image the proximal stump of ductus diverticulum can be seen on the underside of the left common carotid artery.
|
|
View larger version (71K):
[in this window]
[in a new window]
|
Fig 4. Intraoperative assessment of pulmonary arteries (case 6) with this series of photographic images from a head-mounted camera at the time of the second operation showing (A) disconnected left pulmonary artery (LPA) mobilized and held in clamp and (B) anastomosis between LPA and main pulmonary artery (MPA).
|
|
|
Comment
|
|---|
Embryologically the proximal pulmonary artery branches develop from the proximal sixth pharyngeal arches. Nonconfluence of the pulmonary arteries may be due to regression of the sixth arch segment as part of conotruncal abnormalities [7]. Sotomora and Edwards [2] examined postmortem specimens in which there was no channel from the right ventricle to a main branch pulmonary artery. In 8 of 18 specimens there was a disconnected pulmonary artery with an atretic or absent proximal segment. However in 7 of these 8 dissection to the pulmonary hilus revealed an identifiable patent artery. The high incidence in this postmortem study of right aortic arch (6 of 18 cases) and an aberrant subclavian artery (5 of 18 cases) is consistent with the findings in our patient series.
Five of our 7 patients had evidence of either proven or probable preexistent ductal arterial supply to the disconnected pulmonary artery, supporting the findings of Pfefferkorn and associates [6]. In their work 5 of 6 patients had the aortic arch on the opposite side to the absent pulmonary artery. There are insufficient numbers in our series to suggest a relationship between the sidedness of the aortic arch and pulmonary artery disconnection.
A study by Presbitero and associates published in 1984 [1] reported 12 older patients with an angiographically absent pulmonary artery, aged from 5 days to 12 years at the time of first contact. Diagnostic pulmonary venous wedge injections were attempted at cardiac catheterization in 3 patients. These were unsuccessful, which was thought due to the contrast injection being a potent stimulus to coughing in the conscious patient. In our patient group cardiac catheterization had been performed under general anesthesia. Presbitero and associates suggested that ultimately, continuity between the hilar and main pulmonary artery may be surgically achieved. Eleven of their 12 patients underwent at least one surgical procedure at a median age of 3 years that aimed to look for the occult pulmonary artery and connect it to the main pulmonary artery or to the systemic circulation. Limited follow-up of selected patients was recorded in this series.
In 1997 Zhang and associates [8] reported a series of 24 older patients (from 4 years to 21 years of age) with tetralogy of Fallot and unilateral occlusion of a branch pulmonary artery. They focused on attempted primary repair of these patients presenting late. The majority of their patients (21 of 24) had suspected acquired stenotic occlusion, with only the central portion of the pulmonary artery absent rather than a congenital disconnection. Zachary and coworkers [3] published a study in 1998 of the hemodynamics and surgical outcome of one-lung Fontan operations compared with patients with a confluent pulmonary artery preparation. They found the postoperative systemic saturation lower in the one-lung group (87% versus 92%) who had completion of Fontan at a mean age of 23 months. Attempts were made to restore the continuity of the pulmonary arteries in 3 patients but were unsuccessful in all cases. This study highlights the importance of diagnosing a disconnected pulmonary artery early as a one-lung Fontan is not as good and it is harder to restore continuity in the older patient who has not had the disconnected pulmonary artery previously recruited.
Imanaka and associates [5] published a case report documenting the more extreme end of the spectrum in 1998. In this case, a boy with a disconnected left pulmonary artery and ventricular septal defect had left pulmonary vein wedge angiography on presentation at 2 months of age showing a hilar left pulmonary artery. He underwent intracardiac repair at 8 months of age, at which time reconstruction of the left pulmonary artery was not possible as it could not be visualized through the median sternotomy. He subsequently required an ipsilateral pneumonectomy at 11 months of age because of unremitting hemoptysis. The hilar left pulmonary artery, which had previously been imaged, was not found even by histologic examination. This case again highlights the need for early attempted surgical recruitment of a disconnected pulmonary artery.
There is an increasing role for MRI in the diagnosis of cardiac anomalies and on-going assessment of anatomic changes, such as those seen with pulmonary artery growth. Imanaka and associates [5] suggest that preoperative MRI to assess the distance from the main pulmonary artery to the disconnected hilar pulmonary artery is essential. We have not found preoperative MRI, which often requires a general anesthetic in the infant, to be essential in providing data to facilitate successful early recruitment. However in the postrecruitment follow-up when echocardiography of the native hilar pulmonary artery is difficult, MRI has proven to be a useful imaging tool. An example of the use of this informative imaging modality can be seen in the attached image from case 6 (Fig 5),
showing a left synthetic systemic shunt supplying a now well-developed recruited left pulmonary artery.
View larger version (143K):
[in this window]
[in a new window]
|
Fig 5. Magnetic resonance imaging (case 6) with a coronal plane image showing a right modified Blalock-Taussig shunt into a good-sized left pulmonary artery (LPA). (Ao = aorta.)
|
|
With improvements in imaging and surgical techniques, attempted recruitment may be achieved in premature neonates. This is exemplified in a case report by Salaymeh and associates [4] in which a 1 kg premature neonate with a nonconfluent disconnected left pulmonary artery supplied only by a ductus arteriosus was successfully repaired. We concur with the authors comments that early repair is the key in treating this anomaly because of the risk of early obstructive pulmonary disease and to avoid pulmonary hypertension or ultimate sacrifice of the affected lung.
Assuming our unit covers all of New South Wales, the estimated incidence of this condition is approximately 1 in 100,000 newborns. In our series of 7 patients over a period of 10 years with a truly disconnected pulmonary artery we have effectively followed the recommendation by Presbitero and associates [1] that in the presence of an intracardiac anomaly, adequate blood flow should be restored through the interrupted pulmonary artery as a first procedure. However we emphasize that this must be attempted early.
Our surgical strategy in the majority of patients has been to recruit the disconnected pulmonary artery by a synthetic systemic shunt to facilitate growth of the affected artery. We have performed this procedure through a thoracotomy. The second phase of the repair is performed at a time when the enhanced distal pulmonary artery is large enough to accommodate a prosthetic conduit of sufficient size to parallel growth in the patient. In some cases this may need to be reaugmented. The repair of the intracardiac defect can be electively performed after the initial operation as documented in our patients once both the previously disconnected and contralateral pulmonary vasculature has stabilized. We have shown that with early recruitment even with this staged technique there is evidence of good growth of the affected pulmonary artery with restoration of near normal perfusion to the affected lung. There are no deaths reported to date in our series. An alternative surgical strategy may be to perform early repair through a sternotomy, with early establishment of right ventricle to pulmonary artery continuity. This may be achieved by approximating the posterior walls of the disconnected branch pulmonary arteries, roofing the anastomosis with pulmonary homograft patch, into which a right ventricle-pulmonary artery conduit may be placed. Other groups have used various techniques [9].
We conclude that the rare possibility of a congenitally disconnected pulmonary artery needs to be considered in all patients with a conotruncal cardiac anomaly. To facilitate surgical correction, ensure subsequent growth of the pulmonary artery, and optimize associated lung development, early diagnosis and surgical recruitment is recommended.
|
References
|
|---|
- Presbitero P., Bull C., Haworth S.G., de Leval M.R. Absent or occult pulmonary artery. Br Heart J 1984;52:178-185.[Abstract/Free Full Text]
- Sotomora R.F., Edwards J.E. Anatomic identification of so-called absent pulmonary artery. Circulation 1978;57:624-633.[Free Full Text]
- Zachary C.H., Jacobs M.L., Apostolopoulou S., Fogel M.A. One-lung Fontan operation: hemodynamics and surgical outcome. Ann Thorac Surg 1998;65:171-175.[Abstract/Free Full Text]
- Salaymeh K.J., Kimball T.R., Manning P.B. Anomalous pulmonary artery from the aorta via a patent ductus arteriosus: repair in a premature infant. Ann Thorac Surg 2000;69:1259-1261.[Abstract/Free Full Text]
- Imanaka K., Shimizu S., Matsumoto J., Hashizume K., Tsuchiya K., Takemura T. Unilateral absence of pulmonary artery and ventricular septal defect in an infant. Ann Thorac Surg 1998;66:251-252.[Abstract/Free Full Text]
- Pfefferkorn J.R., Loser H., Pech G., Toussaint R., Hilgenberg F. Absent pulmonary artery. A hint to its embryogenesis. Pediatr Cardiol 1982;3:283-286.[Medline]
- McElhinney D.B., Hoydu A.K., Chin A.J., Weinberg P.M. Right-sided aortic arch with bilateral ductus: a rare case of nonconfluent pulmonary arteries without associated cardiac anomalies. J Thorac Cardiovasc Surg 2000;119:849-851.[Free Full Text]
- Zhang G.C., Wang Z.W., Zhang R.F., Zhu H.Y., Yi D.H. Surgical repair of patients with tetralogy of Fallot and unilateral absence of pulmonary artery. Ann Thorac Surg 1997;64:1150-1153.[Abstract/Free Full Text]
- Barbero-Marcial M., Atik E., Baucia J.A., Pradel H.O., Macruz R., Jatene A.D. Reconstruction of stenotic or nonconfluent pulmonary arteries simultaneously with a Blalock-Taussig shunt. J Thorac Cardiovasc Surg 1988;95:82-89.[Abstract]
This article has been cited by other articles:
|
|
|
|
 
E. A. Bacha, P. Lang, J. E. Mayer Jr, and D. B. McElhinney
Connection of Discontinuous Pulmonary Arteries in Patients With a Superior or Total Cavopulmonary Circulation.
Ann. Thorac. Surg.,
December 1, 2008;
86(6):
1948 - 1954.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
T. P. Graham Jr
The year in congenital heart disease
J. Am. Coll. Cardiol.,
June 2, 2004;
43(11):
2132 - 2141.
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
