| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J Thorac Cardiovasc Surg 1994;108:692-699
© 1994 Mosby, Inc.
SURGERY FOR CONGENITAL HEART DISEASE |
Leiden
Rotterdam
Groningen, The Netherlands
Supported by the Netherlands Heart Foundation (grant 87062).
Received for publication March 2, 1994. Accepted for publication May 23, 1994. Address for reprints: M. C. DeRuiter, PhD, Department of Anatomy and Embryology, University of Leiden, P.O. Box 9602, 2300 RC Leiden, The Netherlands.
Abstract
Now that systemic-pulmonary collateral arteries are used for unifocalization in patients with pulmonary atresia and ventricular septal defect, the question arises whether morphologic criteria of these collateral arteries could help to provide better results. In an attempt to classify the morphologic features of systemic-pulmonary collateral arteries, we studied 31 heart-lung autopsy specimens with pulmonary atresia and ventricular septal defect. The course of the systemic-pulmonary collateral arteries (origin, branching pattern, and conections with systemic and central pulmonary arteries) was related to their histologic characteristics. The results show that systemic-pulmonary collateral arteries cannot be classified according to their course related to the trachea and the main branches of the bronchi. The histologic features of these collateral arteries vary along their course to the lungs. Nearly all systemic-pulmonary collateral arteries contain a muscular or a musculoelastic segment. One type of collateral artery (complex loop anastomoses) is completelymuscular and resembles a bronchial artery. Nutritive branches (bronchial arteries) arise from all histologic types of systemic-pulmonary collateral artery segments. The size and number of intimal proliferations in muscular, elastic, and musculoelastic segments did not differ significantly. In 29 of 31 cases a ductus arteriosus did not coexist with large collateral arteries (two cases unknown). It is concluded that a classification of large systemic-pulmonary collateral arteries based on morphologic features results in a highly variable system, which does not facilitate decisions for the suitability of these arteries for unifocalization procedures. The variability of the systemic-pulmonary collateral arteries corresponds with the recent embryologic finding that during development, collateral artery formation is possible during extended periods. (J THORAC CARDIOVASC SURG 1994;108:692-9)
In patients with pulmonary atresia the blood supply of the lungs is derived from the systemic arterial circulation through a ductus arteriosus, through collateral arteries, or through both routes. 
1-5 Various terms have been suggested to describe these collateral arteries: bronchial arteries, 6 systemic arteries, 7 systemic collateral arteries, 8 systemic-pulmonary arteries, 5 and major aortopulmonary collateral arteries. 9 We prefer to use the term systemic-pulmonary collateral arteries, as suggested by McGoon and coworkers, 10 for any arterial vessel other than the ductus arteriosus, which compensates the obstructed blood flow from the right ventricle to the pulmonary arteries.
Many investigators distinguish two types of collateral supply to the lungs. The first consists of a group of small and tortuous systemic-pulmonary collateral arteries with a multifocal origin from the aorta and its branches. They are often associated with well-developed central pulmonary arteries. 7 These systemic-pulmonary collateral arteries are probably acquired postnatally to achieve a greater pulmonary flow. 2,11-13 The other collateral supply consists of large systemic-pulmonary collateral arteries, sometimes associated with absent or hypoplastic central pulmonary arteries, 7 and is considered to be of congenital origin. 14 In contrast to the small, tortuous systemic-pulmonary collateral arteries, 11 the large collateral arteries often connect with hilar or intrapulmonary arteries and are fewer in number. 7
Large systemic-pulmonary collateral arteries are believed to be persistent embryonic ventral splanchnic arteries or enlarged bronchial arteries. 15-17 This possibly different embryonic origin of the collateral arteries has led to a classification on the basis of the site of anastomosis with the pulmonary arteries. 18,19 However, such a classification appears to be an oversimplification in the light of other recent reports. For instance, Haworth and Macartney 9 stated that bronchial arteries and large systemic-pulmonary collateral arteries were two distinctly separate systems, although Liao and coworkers 2 showed that small nutritive branches to the bronchi (bronchial arteries) can also arise from the large collateral arteries.
A normal pulmonary blood flow can be obtained in case of pulmonary atresia with large systemic-pulmonary collateral arteries by surgical reconstruction of the obstructed right outflow tract in combination with unifocalization of the pulmonary blood supply. Unifocalization is a prerequisite for successful surgical treatment of pulmonary atresia with ventricular septal defect and large systemic-pulmonary collateral arteries. 20-22 Unifocalization can be defined as the surgical procedures directed toward elimination of the multiple extracardiac sources of pulmonary arterial blood flow, the restoration of complete lobar and segmental pulmonary arterial blood flow, adequate enhancement of hypoplastic segments or replacement of missing segments of the central pulmonary arteries, and the creation of a central, single, and accessible source of pulmonary flow. 21,22
Intimal proliferation can develop to varying degrees along the course between the origin and the peripheral pulmonary bed of the systemic-pulmonary collateral artery, eventually resulting in a stenosis. 21-25 Progressive intimal proliferation is thought to play a role, resulting in luminal constriction. 21,22 These stenoses may protect the intrapulmonary arterial bed from pulmonary hypertensive vasculopathy but can clearly prevent successful unifocalization. 22,25 Although surgical proceduresare described to deal with these stenoses, 24,25 it is not clear whether these result in long-term successful patency of the pulmonary vascular segment.
In the present study we attempted to clarify the morphology of the collaterals to possibly facilitate a better result in patients undergoing these complicated staged surgical procedures. The course of systemic-pulmonary collateral arteries was related with the histologic characteristics of the tunica intima and media. Specific attention was paid to the presence of a ductus arteriosus, normal bronchial arteries, nutritive branches from systemic-pulmonary collateral arteries to the bronchi, and to the presence of intimal proliferations. The results are discussed in view of recently described embryologic data on pulmonary and bronchial artery formation on the one hand and in view of the suitability for incorporation in surgical reconstruction of the pulmonary arterial system on the other hand.
MATERIAL AND METHODS
In the Children's Hospital of Pittsburgh the vascular patterns of large systemic-pulmonary collateral arteries in heart-lung autopsy specimens in 11 cases of pulmonary atresia with ventricular septal defect were studied. The ages of the patients from whom the specimens were retrieved ranged from 1 day to 22 months. Only two patients underwent an operation; both died within 3 days. It was possible to investigate the extrapulmonary course and the proximal part of the intrapulmonary course of the systemic-pulmonary collateral arteries and the pulmonary arteries relative to the trachea and bronchi. The presence of a ductus arteriosus, bronchial arteries, and anastomoses between the systemic-pulmonary collateral arteries and the pulmonary arteries were recorded. The histologic characteristics of the arteries could not be evaluated, however.
In the Mayo Clinic at Rochester, Minnesota, 20 heart-lung autopsy specimens of pulmonary atresia with ventricular septal defect were available. The ages of the patients ranged from 26 days to 30 years. This material belonging to a larger group (n = 31) of autopsy specimens has been described previously by Liao and coworkers. 2 The existing drawings of their work showed the course and branching pattern of the large systemic-pulmonary collateral arteries relative to the bronchi. The histologic features of the pulmonary arteries, ductus arteriosus, bronchial arteries, and the systemic-pulmonary collateral arteries were evaluated in microscopic sections taken from consecutive blocks that were taken along the branches. The sections were stained with hematoxylin-eosin and van Gieson elastic staining. These sections were restudied and compared with the working drawings. 2 For the present study some of these drawings were modified by the medical artist.
Definitions
We used the following definitions to describe the various types of arteries 26,27: An elastic artery contains more than seven elastic lamellae in the tunica media. Between the elastic lamellae smooth muscle cells are present. A loose elastic artery is a thin-walled elastic vessel with fewer than seven elastic lamellae alternating with smooth muscle cells. A muscular artery shows a thick muscular tunica media with only a few elastic lamellae. A musculoelastic artery is a large muscular vessel with many elastic lamellae in the tunica media. It is often present as a transitional segment between a muscular and an elastic artery. A ductus arteriosus is a muscular artery, but it is distinguished by the absence of an external elastic lamina in the tunica adventitia. The number, origin, and course of the complete muscular bronchial arteries are variable. The two left bronchial arteries mainly arise from the anterior wall of the descending aorta, and the right bronchial artery arises from the first posterior intercostal artery or the upper left bronchial artery. 27-30 The branches of the pulmonary arteries follow the course of the bronchi and show an elastic tunica media, if larger than 1 mm.
The anastomosis between a systemic-pulmonary collateral artery and the pulmonary system can be identified only in gross dissection when the (non-)confluent central pulmonary arteries are present. The systemic-pulmonary collateral arteries can communicate with the central pulmonary arteries by one single anastomosing vessel or by multiple vessels (definition: multiple simple anastomoses). The third possibility consists of one systemic-pulmonary collateral artery communicating more than once with the pulmonary artery system (definition: complex loop anastomoses). In addition, the site of anastomosis of a systemic-pulmonary collateral artery with the intrapulmonary system, when the central pulmonary artery is absent, can be detected in the Mayo collection from the transition of a muscular or musculoelastic tunica media into a complete elastic media.
Statistical methods
Discrete variables were compared with the 
2 test. A p value of 0.05 or less was considered to represent statistical significance.
RESULTS
Confluent central pulmonary arteries and systemic-pulmonary collateral arteries were present in 74% (23/31) of all specimens studied. Eight specimens (26%) contained systemic-pulmonary collateral arteries and nonconfluent or absent central pulmonary arteries. In specimens with confluent central pulmonary arteries, the systemic-pulmonary collateral arteries were connected to the pulmonary arteries by a single anastomosis (14 cases; Fig. 1, A), multiple simple anastomoses (5 cases; Fig. 1, B), or by complex loop anastomoses, that is, with more than one anastomosis (4 cases; Fig. 1, C).
|
Fifteen (12%) of all collateral arteries studied (n = 122) developed extrapulmonary branches. In 13 cases (87%) these collaterals supplied branches to both lungs, whereas only two extrapulmonary branching collateral arteries supplied either a left or a right lung.
In the Pittsburgh collection only the course of the systemic-pulmonary collateral arteries relative to the bronchi could be studied. Their course along the bronchi was variable. The proximal parts of the collateral arteries were mainly located posterior (59%, 22/39) and superior (31%, 12/39) to the bronchi. All possible courses of the systemic-pulmonary collateral arteries on the first branches of the bronchi in the left and right lungs were present. No differences were observed between the courses of the systemic-pulmonary collateral arteries to the left and right lungs.
In 71% of the systemic-pulmonary collateral arteries (59/83) of the 20 specimens of the Mayo collection, the proximal part was muscular or musculoelastic. In the other collateral arteries (29%), the proximal part was elastic. Only two systemic-pulmonary collateral arteries (2%) were completely elastic. The remaining 98% had a muscular segment or a musculoelastic segment or both. Seventy-three percent (61/83) of the systemic-pulmonary collateral arteries had a muscular segment ranging from a very short, noncircular segment to a very long segment. Of the remaining 22 specimens without a muscular segment, 20 had a musculoelastic segment.
In four specimens with complex loop anastomoses, 93% of the total number of systemic-pulmonary collateral arteries (14/15) had long muscular segments or were completely muscular (Fig. 1, C). One of these collateral arteries (7%) contained a short muscular segment. In all other specimens the histologic features of a systemic-pulmonary collateral artery were not related to its course, branching pattern, type of anastomosis (single or multiple simple), presence of a central pulmonary artery, place of connection with the intrapulmonary arteries, or total number of collaterals in the specimens.
In 29 of 31 autopsy specimens studied, neither a ductus arteriosus nor a ductus ligament was found (one unknown because of absent surgical data). In one case the vessel supplying the entire left lung arose from the subclavian artery, whereas the left central pulmonary artery was absent. Without histologic data the possibility that this vessel was a patent ductus arteriosus could not be evaluated.
Ten systemic-pulmonary collateral arteries arose from the aortic arch or from the contralateral subclavian artery to supply one lung, without communicating with the central pulmonary artery system. This origin and course are suggestive of the presence of a ductus arteriosus, but the typical ductal histologic characteristics, lacking an external elastic lamina, were not found. On the contrary, all the muscular (Fig. 2, A) and musculoelastic (Fig. 2, B) collateral arteries showed external elastic laminae and a relatively thin tunica media with regard to their lumina.
|
Intimal proliferations were present in all types of systemic-pulmonary collateral arteries. The difference in the frequency of intimal proliferations among muscular segments (32%, 20/62), elastic segments (23%, 21/90), and musculoelastic segments (24%, 18/75) was not significant. In 47% (39/83) of the systemic-pulmonary collateral arteries, the tunica intima was thickened. In the collateral arteries that had been surgically ligated (n = 21), intimal thickening and complete obliteration of the lumen was present in 12 cases (57%).
The histologic features of the intimal proliferations varied highly. Most of the proliferations were muscular. The severe intimal proliferations did not contain a subendothelial lamina, and the internal elastica lamina had disappeared or was fragmented (Fig. 3, A). Many of the mild intimal proliferations had a subendothelial lamina (Fig. 3, B), as seen in a persistent ductus arteriosus.
|
Large differences in age and patient history exist between the collections of the Children's Hospital of Pittsburgh and the Mayo Clinic. The Pittsburgh collection is characterized by a young age of the patients, who had not been operated on or had died within 3 days after the operation. The Mayo Clinic collection includes a large number of relatively old patients, who underwent surgery. Notwithstanding this dissimilarity between the two collections, significant differences in the arterial supply to the lungs were not observed. The age of the patients and the surgical interventions in the Mayo Clinic collection may have biased our observations on the presence or absence of a (ligamentous) ductus arteriosus and the extent of intimal thickening and obliteration, and it may also have changed the information on the extent of nutritive or bronchial arteries.
The studied autopsy specimens of the Mayo collection were published before by Liao and coworkers. 2 They described 31 cases of pulmonary atresia with a ventricular septal defect. Twenty (20/31) involved large systemic-pulmonary collateral arteries. In this group only one ductus arteriosus was described and no ligamentous ductus were seen. In the remaining 11 cases (11/31) without large collateral arteries, a ductus arteriosus was always found. In our study, only the 20 cases with large systemic-pulmonary collateral arteries were reinvestigated. This revealed that the ductus arteriosus described by Liao in this group lacked the typical histologic features of a ductus, because it had an external elastic lamina and thus could not be differentiated from other muscular systemic-pulmonary collateral arteries.
The macroscopic study of the Pittsburgh collection did not reveal the coexistence of a ductus arteriosus or a ligament with large systemic-pulmonary collateral arteries (two cases unknown.). Our data, therefore, suggest that the combination of a ductus arteriosus and large systemic-pulmonary collateral arteries does not occur. Data from the literature are less conclusive. Thiene and coworkers 5 described the coexistence of a ductus with large systemic-pulmonary collateral arteries, but in general not to the same lung. In some cases, a ductus and collateral arteries have been mentioned to run to the same lung, but then they supply different pulmonary units. 1 The vessel depicted (Fig. 1, page 348) in the workof Frescura and associates, 1 however, is not typical of a ductus arteriosus and, in our opinion, might well be a musculoelastic systemic-pulmonary collateral artery. The macroscopic work of Thiene and coworkers 5 is not substantiated by histologic evaluation.
The present results show that the systemic-pulmonary collateral arteries could not be classified according to their course relative to the trachea and the first branches of the bronchi. No differences were observed in either number or course of the collateral arteries between the left and right lungs. Although the proximal part of the collateral arteries is usually situated along the posterior surface of the bronchi, like the bronchial arteries, this does not indicate that the systemic-pulmonary collateral arteries must be considered to be enlarged bronchial arteries, 7 because most of them are already positioned posteriorly by their origin from the descending aorta. The histologic characteristics of the systemic-pulmonary collateral arteries vary along their course to the lungs, but almost all contain a muscular or a musculoelastic segment. Except in the specimens with complex loop anastomoses, which were completely muscular, no relationships were found between the histologic characteristics of systemic-pulmonary collateral arteries and their origin, course, branching patterns, or anastomoses to the pulmonary arteries.
The nutritive branches arising from some systemic-pulmonary collateral arteries suggest that these are enlarged bronchial arteries. Nutritive branches, however, arise not only from systemic-pulmonary collateral arteries that resemble the typical histologic characteristics of the bronchial arteries, such as those with complex loop anastomoses, but from all histologic types of segments.
Proposed classifications of different types of systemic-pulmonary collateral arteries 9,19 are based onembryologic theories. 15-17 These theories suggest a separate development of the (intra- and extra-) pulmonary arteries and the bronchial arteries, before the two systems start to anastomose. Classic embryologic observations, however, do not explain the great variability encountered in origin, course, arborization, and possible anastomoses with the pulmonary arteries of the systemic-pulmonary collateral arteries. From our recent study on the normal development of the pulmonary and bronchial vasculature, 31 it can be deduced that collateral artery supply develops from one splanchnic vascular plexus. This plexus, which covers the developing bronchi, trachea, and esophagus, gives rise to all types of lung vessels. During two periods separated by an interposed period, the plexus communicates by luminal connections developing with the aorta and the intercostal arteries. During the first period the caudal part of the splanchnic plexus is temporarily connected to the aorta. The pulmonary and bronchial arteries cannot be recognized as separate vessels with lumina within the plexus at that time. When these aortopulmonary connections disappear, pulmonary arteries with lumina start to develop in the splanchnic plexus. In the second, extended period when the pulmonary and aortic orifices are separated by the aortopulmonary septum, the pulmonary arteries are recognizable as vessels with lumina around the developing bronchi. The remnants of the splanchnic plexus connect again with the descending aorta. These communicating vessels with lumina ultimately give rise to the bronchial arteries. During further fetal development, the bronchial arteries remain connected with the pulmonary vessels by bronchopulmonary anastomoses. Because both sets of aortopulmonary connections originate subsequently from the same splanchnic plexus, this will not lead to an easy differentiation in the postnatal period between the various collateral arteries.
Surgical implications
Reconstructive surgery (e.g., unifocalization procedures) aims at the incorporation in the pulmonary vasculature of all systemic-pulmonary collateral arteries of importance. On the sole basis of pragmatic arguments, this usually means the incorporation of systemic-pulmonary collateral arteries providing the perfusion of more than a single pulmonary segment. 24 In this regard, bronchial arteries are usually too small to be incorporated in the reconstruction. However, surgical exposure could be inadequate because of overlying bronchial arteries. The latter can be sacrificed without any obvious clinical effect. The reconstruction of an unobstructed pulmonary vascular bed of optimal capacity is considered to be essential to provide adequate right ventricular decompression after complete repair. 32 Persistent or developing stenoses at different sites of the reconstructed pulmonary arterial tree may lead to less than optimal results and may even preclude the final step of complete repair. 24,32 In addition early rehabilitation of the pulmonary arterial tree by increasing blood flow of low pressure seems to improve the possibilities of complete repair. 33
On the basis of our morphologic and histologic data showing the variable origins of systemic-pulmonary collateral arteries being reflected in their variable histologic characteristics and the subsequent predisposition to stenosis or even occlusion, it is understandable that the surgical reports in unifocalization procedures cannot successfully predict the suitability of a specific systemic-pulmonary collateral artery for incorporation in the newly constricted pulmonary arterial bed.
In this setting, the preparing steps leading to repair should probably be done early in life, 33,34 to profit from tissue pliability in young patients and, more important, to prevent the complications of longstanding abnormal perfusion of lung segments. 20,25 However, risk-adjusted, time-related probabilities of morbidity and mortality are not yet available. 20 Nevertheless, adequate, preferably complete, unifocalization is therefore a prerequisite for successful repair. 20-22 In this regard, discrete stenoses of the pulmonary arterial bed may be treated at any stage by surgery or percutaneous balloon dilatation 20,21 or stenting. 35
These surgical and invasive cardiologic attempts definitely show the feasibility of incorporating these systemic-pulmonary collateral arteries into an adequately functioning pulmonary vascular bed, despite the uncertainties of the long-term fate of these collateral arteries and despite the risk of progressive stenoses in their vascular bed. On-going anatomic and surgical efforts are essential to improve the results of surgical attempts to treat pulmonary atresia with ventricular septal defect and pulmonary circulation dependent on systemic-pulmonary collateral arteries.
Acknowledgments
We appreciate the contribution of the Division of Anatomic Pathology, Mayo Clonic, and especially acknowledge the work of Pui-Kan Liao, MD, currently of The Children's Hospital of New Jersey, Newark, New Jersey. J.R. Zuberbuhler, MD, of the Children's Hospital of Pitsburgh, Department of Pathology, is thanked for his hospitality and the permission to study autopsy specimens. Moreover, we are indebted to J. Lens for his photographic work and to S. B. Blankevoort for his drawings.
References
This article has been cited by other articles:
|
|
 |
|
  M Chessa, G Butera, P Bonhoeffer, L Iserin, J Kachaner, S Lyonnet, A Munnich, D Sidi, and D Bonnet Relation of genotype 22q11 deletion to phenotype of pulmonary vessels in tetralogy of Fallot and pulmonary atresia-ventricular septal defect Heart, February 1, 1998; 79(2): 186 - 190. [Abstract] [Full Text]
|
|
|
|
 |
|
 F. D. Pagani, J. P. Cheatham, R. H. Beekman III, T. R. Lloyd, R. S. Mosca, and E. L. Bove THE MANAGEMENT OF TETRALOGY OF FALLOT WITH PULMONARY ATRESIA AND DIMINUTIVE PULMONARY ARTERIES J. Thorac. Cardiovasc. Surg., November 1, 1995; 110(5): 1521 - 1533. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
