- Email: [email protected]
The Outflow Tract in Transposition of the Great Arteries: An Anatomic and Morphologic Study
The Outflow Tract in Transposition of the Great Arteries: An Anatomic and Morphologic Study Shirin Lalezari, MD, Edris A. F. Mahtab, PhD, Margot M. Bartelings, MD, PhD, Lambertus J. Wisse, BS, Mark G. Hazekamp, MD, PhD, and Adriana C. Gittenberger-de Groot, PhD Departments of Anatomy and Embryology, and Cardiothoracic Surgery, Leiden University Medical Center, Leiden, the Netherlands PEDIATRIC CARDIAC
Background. Neoaortic root dilatation is observed after the arterial switch operation for transposition of the great arteries. Although structural differences in the vessel wall of these patients may be of influence, we hypothesize that a histomorphologic difference in composition and embedding of the fibrous annulus in transposition of the great arteries may play a role in neoaortic root dilatation. Methods. Two normal human hearts and two unoperated human hearts with transposition of the great arteries, 1 day postnatal, were studied. Histologic sections stained for collagen, myocardium, and elastin were prepared, and three-dimensional reconstructions of the outflow tracts were made to enable comparison of the morphologic structures between the normal hearts and those with transposition of the great arteries. Results. The amount of collagen in the arterial roots was diminished in hearts with transposition of the great
arteries compared with the normal hearts. In addition, the anchorage and embedding of both arterial roots in the myocardium was less extensive in transposition of the great arteries. The changed position of the arteries in the malformed hearts results in less support for the roots from the surrounding atrioventricular myocardium. Conclusions. The combination of the observed histomorphologic differences in amount of collagen and myocardial support may be an explanation for the neoaortic root dilatation observed after the arterial switch operation. The developmental background of the observed deficient fibrous annulus formation may originate from an epicardial problem.
T
ation depends on the capability of the native pulmonary root to function in the systemic circulation [2]. Neoaortic root dilatation is one of the late complications described after the ASO for TGA [2– 6, 9]. In a previous study we have shown that the pulmonary trunk and pulmonary sinus wall in unoperated hearts with TGA show a decrease of ␣-smooth muscle actin expression with increasing age [10]. Thus the structure of the vessel wall and sinus wall of the aorta and pulmonary trunk are different in hearts with TGA. As described above, after the ASO the pulmonary root remains on the left side and becomes the neoaortic root subjected to the systemic pressure. This may provide an explanation for the dilatation of the neoaortic root that is sometimes observed late after ASO. Apart from an abnormal and weaker ␣-smooth muscle actin expression of the wall of the pulmonary trunk [10] and the disconcordant ventriculoarterial connection described above, we hypothesize that in TGA the observed dilatation of the pulmonary root after ASO might also be attributable to histomorphologic defects. Therefore, we studied the morphology of the pulmonary and aortic roots in the left and right ventricles in hearts with TGA and compared our observations to the situation in normal hearts. We specifically paid attention to the amount
ransposition of the great arteries (TGA) is a common and severe congenital heart malformation, having an incidence of about 1 per 5,000 live births [1]. In TGA, the atria and ventricles are in the normal position (situs solitus and concordant atrioventricular connection), but the aorta arises from the right ventricle and the pulmonary trunk arises from the left ventricle (disconcordant ventriculoarterial connection), changing the serially connected pulmonary and systemic circulations into two parallel (separated) circulations. The arterial switch operation (ASO) is now the surgical procedure of choice for TGA [2– 6]. This operation involves transsection and reanastomosis of both pulmonary trunk and aorta above the sinuses of Valsalva and transplantation of the coronary arteries [7, 8]. After the ASO, the pulmonary root (neoaortic root) functions in the systemic circulation. The long-term success of this oper-
Accepted for publication June 22, 2009. Presented at the Poster Session of the Forty-third Annual Meeting of The Society of Thoracic Surgeons, San Diego, CA, Jan 29 –31, 2007. Address correspondence to Dr Gittenberger-de Groot, Department of Anatomy and Embryology, Leiden University Medical Center, PO Box 9600, Postal zone S-1-P, Leiden, 2300 RC, the Netherlands; e-mail: [email protected].
© 2009 by The Society of Thoracic Surgeons Published by Elsevier Inc
(Ann Thorac Surg 2009;88:1300 –5) © 2009 by The Society of Thoracic Surgeons
0003-4975/09/$36.00 doi:10.1016/j.athoracsur.2009.06.058
LALEZARI ET AL DILATATION OF THE NEOAORTIC ROOT
1301
Fig 1. Collagen distribution at the commissure level. Three-dimensional reconstruction is shown of the arterial roots of a normal heart in a right anterior view (A) and a heart with transposition of the great arteries (TGA) in a left anterior view (B). The amount of collagen (blue) is clearly decreased in the heart with transposition of the great arteries. Transverse sections (C–F) of the arterial root stained with Azan show the amount and distribution of collagen in the normal heart (C, D) and the heart with transposition of the great arteries (E, F). The intersection lines c in A and e in B refer to the respective sections below. At the commissure level, in contrast to the aorta (Ao) in the normal hearts (C, D), in the heart with transposition of the great arteries the collagenous network in the pulmonary trunk (PT) does not extend further into the adventitial layer (E, F). This collagenous layer in the heart with transposition of the great arteries is interrupted by the elastic lamellae (compare arrow in D with F). Scale bars: C, E ⫽ 600 m; D, F ⫽ 120 m. Color codes: aorta, transparent pink; atrial myocardium, light gray; mitral valve (MV), transparent yellow; pulmonary trunk, transparent green. (A ⫽ anterior; AoL ⫽ aortic lumen; L ⫽ left; LA ⫽ left atrium; P ⫽ posterior; PTL ⫽ pulmonary trunk lumen; R ⫽ right; RA ⫽ right atrium.)
of collagen and the distribution as well as myocardial support of the arterial roots using several histologic markers. A three-dimensional anatomic reconstruction was made to show the relationship of the arterial fibrous annulus with the surrounding atrial and ventricular myocardium.
Material and Methods Material Four unoperated human heart specimens aged 1 day (postpartum) were studied including two with normal anatomy and two with simple TGA. The hearts were
obtained from the Leiden Collection (Department of Anatomy and Embryology, Leiden University Medical Center, Leiden, The Netherlands). This study is in accordance with the institutional guidelines of the Leiden University Medical Center for the use of human tissue.
Methods All hearts were fixed in ethanol and glycerin and routinely processed for light microscopy. Transverse sections (10 m) were mounted serially onto glycerin-coated glass slides. The paraffin-embedded sections were deparaffinized and stained with hematoxylin-eosin, resorcin-fuchsin, modified
Fig 2. Collagen distribution at the mid-sinus level. Three-dimensional reconstruction (A, B) and transverse sections (C–F) are as described in Figure 1. At the mid-sinus level the pulmonary trunk (PT) in the heart with transposition of the great arteries shows a thin compact collagen layer at the outside and a less extensive anchorage to the adventitial layer, and both elastic lamella as well as collagen layers are relatively disorganized compared with the normal heart (compare C, D with E, F). Color codes are as in Figure 1. Scale bars: C, E ⫽ 200 m; D, F ⫽ 120 m. (A ⫽ anterior; AoL ⫽ aortic lumen; L ⫽ left; LA ⫽ left atrium; P ⫽ posterior; PTL ⫽ pulmonary trunk lumen; R ⫽ right; RA ⫽ right atrium; RCS ⫽ right coronary sinus; RFS ⫽ right-facing sinus.)
PEDIATRIC CARDIAC
Ann Thorac Surg 2009;88:1300 –5
1302
LALEZARI ET AL DILATATION OF THE NEOAORTIC ROOT
Ann Thorac Surg 2009;88:1300 –5
PEDIATRIC CARDIAC
Fig 3. Collagen distribution at the base of the sinus level. Three-dimensional reconstruction (A, B) and transverse sections (C–F) are as described in Figure 1. At the base of the sinus the collagen layer of the pulmonary trunk (PT) is thinner (E, F) and compared with the aorta in the normal heart (C, D) the fingerlike protrusions (arrowheads in D and F), which attach the sinus with the surrounding myocardium, are thinner and less extensive (E, F). Color codes are as in Figure 1. Scale bars: C, E ⫽ 200 m; D, F ⫽ 120 m. (A ⫽ anterior; AoL ⫽ aortic lumen; L ⫽ left; LA ⫽ left atrium; P ⫽ posterior; PTL ⫽ pulmonary trunk lumen; R ⫽ right; RA ⫽ right atrium; RCS ⫽ right coronary sinus; RFS ⫽ right-facing sinus.)
Van Gieson, and Azan for routine histologic assessment of the specimens according to the institutional guidelines.
Three-Dimensional Reconstruction One normal heart and one heart with TGA were selected for three-dimensional reconstruction based on the Azan-
stained sections, providing good discrimination between collagen fibers, myocardium, and vessel wall structures. The distance between the serial sections was 160 m. Digital photomicrographs were taken of the serial sections of the normal heart and the TGA specimen from just below until just above the arterial roots. The lower
Fig 4. Myocardial support. Three-dimensional reconstruction of the arterial root and parts of atrial and ventricular myocardium of a normal heart in a right anterior view (A, C) and a heart with transposition of the great arteries (TGA) in a left anterior view (B, D). Transverse sections C and D are cranial views and correspond to the intersection lines c and d in A and B, respectively. The aortic (Ao) annulus (blue) in the normal heart is supported by the atrial and ventricular myocardium forming a collar over most of its circumference (C). Both the aorta and pulmonary trunk (PT) in the heart with transposition of the great arteries appear to be placed on top of the heart instead of being surrounded by myocardium, as is seen in the normal heart (compare A with B). Only a small part of the circumference of the pulmonary trunk annulus is surrounded by myocardium (D). Color codes: aorta, transparent pink; atrial myocardium, light gray; mitral valve (MV), transparent yellow; pulmonary trunk, transparent green; ventricular myocardium, transparent dark gray. (A ⫽ anterior; L ⫽ left; LA ⫽ left atrium; LV ⫽ left ventricle; P ⫽ posterior; R ⫽ right; RA ⫽ right atrium; RV ⫽ right ventricle.)
Ann Thorac Surg 2009;88:1300 –5
1303
At the base of the aortic and pulmonary sinus the elastic lamellae are absent in the wall of the sinus, and the entire wall is composed of a collagen layer. In the normal hearts this collagen layer is thick and extends into the surrounding myocardium by means of several fingerlike protrusions, forming a strong anchorage (Figs 3C and 3D). In the TGA hearts, similar to the mid-sinus region, the collagen layer is thinner (Figs 3E and 3F) compared with the normal hearts. The collagen layer extends fingerlike protrusions into the surrounding myocardium; however, these protrusions are thinner and less extensive compared with the normal hearts (Figs 3E and 3F).
BASE OF THE SINUS LEVEL.
Myocardial Anchorage and Support Results Below we describe the differences in collagen distribution, myocardial anchorage, and embedding of the arterial root between the normal hearts and the hearts with TGA.
Collagen Distribution As described previously [11, 12], in the normal heart the relationship between collagen and vessel wall at the most distal part of the annulus (commissure) is different from that of the mid-sinus region and the base of the sinus. For this reason we have compared the collagen distribution of the arterial root of the normal hearts to hearts with TGA at three different regions of the sinus: (1) at the level of the commissure (Fig 1), (2) at the mid-sinus level (Fig 2), and (3) at the base of the sinus (Fig 3). Threedimensional reconstructions of both specimens show more clearly the distinct difference in collagen distribution of the annulus between the normal heart and the heart with TGA (Figs 1A, 1B, 2A, 2B, 3A, 3B). COMMISSURE LEVEL. At the distal level in the normal heart, collagen fibers arising from the aortic annulus continue into the valve commissures on the inside of the aortic root and then merge on the outside of the elastic aortic wall with the adventitial layer (Figs 1C and 1D). In TGA, the collagenous network forming the thin pulmonary annulus also concentrates distally to form the commissures. However, this layer does not extend any further to the adventitial layer surrounding the pulmonary trunk but only extends to the level just above the pulmonary annulus (Figs 1E and 1F).
At the mid-sinus level in the normal hearts the aortic sinus wall is surrounded by a thick collagen layer on the outside and by elastic lamellae at the luminal side (Figs 2C and 2D). Compared with the normal hearts, in the TGA hearts the compact collagen layer on the outside of the pulmonary trunk is thin, and both layers, elastic lamellae as well as collagen layer, are disorganized (Figs 2E and 2F). In addition, the collagen layer in the TGA hearts has less extensive anchorage to the adventitial layer compared with the normal hearts (compare Fig 2C with Fig 2E).
MID-SINUS LEVEL.
Above we have described in the normal hearts, at the base of the sinus the annulus has a stronger anchorage to the surrounding myocardium by means of fingerlike collagenous protrusions compared with the hearts with TGA. In the normal hearts the aorta is supported by its embedding in the myocardium of the left ventricular outflow tract, with the ventricular septum and atrial myocardium forming a partial collar around it, whereas the pulmonary trunk is positioned more superficially on top of the right ventricle, being only partially embedded by the myocardium of the right ventricular outflow tract and not being supported by atrial myocardium (Figs 4A, 4C). In the hearts with TGA most of the circumference of the pulmonary annulus appears to lie on top of the left ventricular outflow tract, rather than being embedded into it (compare Fig 4A with Fig 4B). Only a small part of the pulmonary annulus circumference is supported by the left ventricular myocardium (Fig 4D). Similar to the pulmonary annulus, the majority of the aortic annulus appears to be placed on top of the right ventricular outflow tract and is thus not embedded in the surrounding myocardium (Figs 4B and 4D). The atrial myocardium surrounds only a small part of the circumference of the pulmonary annulus compared with the aortic annulus of the normal hearts (compare Figs 4A and 4C with Figs 4B and 4D). Similar to the pulmonary annulus of the normal hearts, the aortic annulus has no support from the atrial myocardium (Figs 4B and 4D).
Comment The main focus of the current study concerns the question as to what could be responsible for the sometimes observed dilatation of the neoaortic root after ASO. Murakami and colleagues [13] describe a decrease of distensibility at the base of the neoaorta in post-ASO hearts, which they correlated with the disruption of vasa vasorum leading to the neoaortic dilatation later. In a previous study from our group we have shown that the arterial roots in TGA are structurally different from the arterial roots in a normal heart [10]. In the unoperated TGA specimens, dedifferentiation of smooth muscle cells was observed, especially in the pulmonary root. This may
PEDIATRIC CARDIAC
limit was at approximately three quarters of the baseapex distance. The upper limit was determined amply above the sinotubular junction of the vessel wall, which is the upper limit of the collagenous annulus. The serial photomicrographs were incorporated into the AMIRA V4.0 software package (Template Graphics Software, San Diego, CA). The different cardiovascular structures (ie, collagen, vessel wall, semilunar and atrioventricular valves, and myocardium) were color-labeled by means of manual drawing and processed for three-dimensional visualization. Processing conditions of the normal heart and TGA specimen were similar, thus providing good conditions for direct comparison between the reconstructions.
LALEZARI ET AL DILATATION OF THE NEOAORTIC ROOT
1304
LALEZARI ET AL DILATATION OF THE NEOAORTIC ROOT
be a factor leading to dilatation of the neoaortic root after the ASO. Although the pathogenesis of neoaortic root dilatation after ASO may be in part explained by the hypothesis of Murakami and colleagues [13], we postulate that the observed differences in collagen distribution and myocardial support of the arterial roots in TGA, as described in this study, as well as the structural differences of the roots in terms of the smooth muscle cell dedifferentiation [10], play a more significant role than the impaired blood flow of the vessel wall by disruption of the vasa vasorum. PEDIATRIC CARDIAC
Collagen Distribution The results observed in this study confirm the findings described by Bartelings and coworkers [12] concerning the fingerlike collagen structures extending from the annulus, as well as their anchorage into the ventricular myocardium in the normal heart. In the hearts with TGA, morphologic differences were noticed compared with the findings in the normal hearts. A striking difference between the TGA and the normal heart specimens was the distribution and quantity of collagen in the arterial roots. The amount of collagen appears to be less in the hearts with TGA than in the normal hearts. Furthermore, in the TGA specimens, the collagen did not spread to the adventitial layer at the site of the commissures, whereas this was seen in both arterial roots of the normal hearts. Additionally, both arterial roots in the TGA specimens appear to be less firmly supported by the myocardium of the corresponding ventricles than both arterial roots in the normal specimens. The development of TGA has been linked to neural crest migration abnormalities [14]. Neural crest cells are involved in initiating the septation of the cardiac outflow tract [15]. Exposing these cells to all-trans retinoic acid in animal models during development instigates several cardiac malformations, one of them being TGA [16 –19]. In TGA mouse embryos, Yasui and colleagues [20] observed an altered collagen distribution. They concluded that treatment with retinoic acid and the subsequent neural crest abnormalities leading to TGA or TGA-like anomalies induced the altered organization of extracellular matrix components, such as type I collagen and hyaluronic acid. A more recent study by Jenkins and associates [21] revealed malformed epicardium in retinoic acid receptor knockout mice. Formation of the fibrous skeleton of the heart, which consists of collagen, is linked to epicardial differentiation and migration [22]. When the epicardium is inhibited during development, the fibrous skeleton of the outflow tract is underdeveloped and its collagen appears to be distributed differently compared with the normal situation [23]. Thus, the difference in collagen distribution between TGA and normal hearts, as observed in this study, may also originate from an epicardial problem during early development, because altered fibrous skeleton formation is also seen in retinoic acid receptor knockout mice.
Ann Thorac Surg 2009;88:1300 –5
Myocardial Support In the normal heart, the aortic root is supported by atrial myocardium, left ventricular outflow myocardium, and the ventricular septum, forming a myocardial cuff surrounding the root. The pulmonary root in the normal heart is not supported by the atrial myocardium and only the pulmonary infundibular myocardium bulges slightly over the pulmonary root, forming a modest collar surrounding it and providing some support, Both roots are attached to the surrounding myocardium by fingerlike protrusions [11, 12]. By studying the three-dimensional reconstructions and the histologic sections of the hearts with TGA, it is apparent that none of the roots is supported by the atrial myocardium. Both roots are placed more superficially on top of the ventricles, missing the sufficient myocardial cuff surrounding the roots as was the case in the normal heart. The attachment of the roots to the surrounding myocardium by fingerlike protrusions was less extensive and thinner compared with the normal heart. The retrospective study by Vandekerckhove and associates [9], in which 39 patients were examined 20 years after ASO, revealed 60% dilatation of the neoaortic root in the operated population. Long-term follow-up of this population is necessary to observe whether the remaining 40% develop dilatation of the neoaortic root, which we relate to the morphologic changes of the arterial root of the hearts with TGA.
Conclusions In this study we observed an altered collagen distribution of the annulus at all three levels of both great arteries studied in TGA. Furthermore, both the aorta and the pulmonary trunk were found to be less firmly anchored to their corresponding ventricles in the TGA specimens than in the normal heart specimens. We speculate that the findings in the present study, together with the observed structural differences between the arterial roots in TGA and normal hearts, in combination with subjecting the neoaortic root to systemic pressures after ASO, might form the basis of the neoaortic root dilatation that occurs after ASO.
We thank Jan Lens for expert technical assistance with the figures.
References 1. Love BA, Mehta D, Fuster VF. Evaluation and management of the adult patient with transposition of the great arteries following atrial-level (Senning or Mustard) repair. Nat Clin Pract Cardiovasc Med 2008;5:454 – 67. 2. Hourihan M, Colan SD, Wernovsky G, Maheswari U, Mayer JE Jr, Sanders SP. Growth of the aortic anastomosis, annulus, and root after the arterial switch procedure performed in infancy. Circulation 1993;88:615–20. 3. Hazekamp MG, Schoof PH, Suys BE, et al. Switch back: using the pulmonary autograft to replace the aortic valve after arterial switch operation. J Thorac Cardiovasc Surg 1997;114:844 – 6.
4. Haas F, Wottke M, Poppert H, Meisner H. Long-term survival and functional follow-up in patients after the arterial switch operation. Ann Thorac Surg 1999;68:1692–7. 5. Formigari R, Toscano A, Giardini A, et al. Prevalence and predictors of neoaortic regurgitation after arterial switch operation for transposition of the great arteries. J Thorac Cardiovasc Surg 2003;126:1753–9. 6. McMahon CJ, Gauvreau K, Edwards JC, Geva T. Risk factors for aortic valve dysfunction in children with discrete subvalvar aortic stenosis. Am J Cardiol 2004;94:459 – 64. 7. Laks H. The arterial switch procedure for the neonate: coming of age. Ann Thorac Surg 1989;48:3– 4. 8. Castaneda AR, Mayer JE Jr, Jonas RA, Wernovsky G, di Donato R. Transposition of the great arteries: the arterial switch operation. Cardiol Clin 1989;7:369 –76. 9. Vandekerckhove KD, Blom NA, Lalezari S, Koolbergen DR, Rijlaarsdam ME, Hazekamp MG. Long-term follow-up of arterial switch operation with an emphasis on function and dimensions of left ventricle and aorta. Eur J Cardiothorac Surg 2009;35:582– 8. 10. Lalezari S, Hazekamp MG, Bartelings MM, Schoof PH, Gittenberger-de Groot AC. Pulmonary artery remodeling in transposition of the great arteries: relevance for neoaortic root dilatation. J Thorac Cardiovasc Surg 2003;126:1053– 60. 11. Hokken RB, Bartelings MM, Bogers AJJC, Gittenberger-de Groot AC. Morphology of the pulmonary and aortic roots with regard to the pulmonary autograft procedure. J Thorac Cardiovasc Surg 1997;113:453– 61. 12. Bartelings MM, Bogers AJ, Galantowicz ME, Gittenberger-de Groot AC. Anatomy of the aortic and pulmonary roots. Semin Thorac Cardiovasc Surg Pediatr Card Surg Annu 1998;1:157– 64. 13. Murakami T, Nakazawa M, Momma K, Imai Y. Impaired distensibility of neoaorta after arterial switch procedure. Ann Thorac Surg 2000;70:1907–10. 14. Costell M, Carmona R, Gustafsson E, Gonzalez-Iriarte M, Fassler R, Munoz-Chapuli R. Hyperplastic conotruncal en-
LALEZARI ET AL DILATATION OF THE NEOAORTIC ROOT
15. 16.
17. 18.
19.
20.
21. 22.
23.
1305
docardial cushions and transposition of great arteries in perlecan-null mice. Circ Res 2002;91:158 – 64. Yu J, Gonzalez S, Martinez L, ez-Pardo JA, Tovar JA. Effects of retinoic acid on the neural crest-controlled organs of fetal rats. Pediatr Surg Int 2003;19:355– 8. Waldo K, Miyagawa-Tomita S, Kumiski D, Kirby ML. Cardiac neural crest cells provide new insight into septation of the cardiac outflow tract: aortic sac to ventricular septal closure. Dev Biol 1998;196:129 – 44. Lammer EJ, Chen DT, Hoar RM, et al. Retinoic acid embryopathy. N Engl J Med 1985;313:837– 41. Yasui H, Nakazawa M, Morishima M, Miyagawa-Tomita S, Momma K. Morphological observations on the pathogenetic process of transposition of the great arteries induced by retinoic acid in mice. Circulation 1995;91:2478 – 86. Yasui H, Morishima M, Nakazawa M, Ando M, Aikawa E. Developmental spectrum of cardiac outflow tract anomalies encompassing transposition of the great arteries and dextroposition of the aorta: pathogenic effect of extrinsic retinoic acid in the mouse embryo. Anat Rec 1999;254:253– 60. Yasui H, Nakazawa M, Morishima M, Aikawa E. Altered distribution of collagen type I and hyaluronic acid in the cardiac outflow tract of mouse embryos destined to develop transposition of the great arteries. Heart Vessels 1997;12: 171– 8. Jenkins SJ, Hutson DR, Kubalak SW. Analysis of the proepicardium-epicardium transition during the malformation of the RXRalpha-/- epicardium. Dev Dyn 2005;233:1091–101. Gittenberger-de Groot AC, Vrancken Peeters M-PFM, Mentink MMT, Gourdie RG, Poelmann RE. Epicardium-derived cells contribute a novel population to the myocardial wall and the atrioventricular cushions. Circ Res 1998;82:1043–52. Eralp I, Lie-Venema H, DeRuiter MC, et al. Coronary artery and orifice development is associated with proper timing of epicardial outgrowth and correlated Fas ligand associated apoptosis patterns. Circ Res 2005;96:526 –34.
PEDIATRIC CARDIAC
Ann Thorac Surg 2009;88:1300 –5
Background. Neoaortic root dilatation is observed after the arterial switch operation for transposition of the great arteries. Although structural differences in the vessel wall of these patients may be of influence, we hypothesize that a histomorphologic difference in composition and embedding of the fibrous annulus in transposition of the great arteries may play a role in neoaortic root dilatation. Methods. Two normal human hearts and two unoperated human hearts with transposition of the great arteries, 1 day postnatal, were studied. Histologic sections stained for collagen, myocardium, and elastin were prepared, and three-dimensional reconstructions of the outflow tracts were made to enable comparison of the morphologic structures between the normal hearts and those with transposition of the great arteries. Results. The amount of collagen in the arterial roots was diminished in hearts with transposition of the great
arteries compared with the normal hearts. In addition, the anchorage and embedding of both arterial roots in the myocardium was less extensive in transposition of the great arteries. The changed position of the arteries in the malformed hearts results in less support for the roots from the surrounding atrioventricular myocardium. Conclusions. The combination of the observed histomorphologic differences in amount of collagen and myocardial support may be an explanation for the neoaortic root dilatation observed after the arterial switch operation. The developmental background of the observed deficient fibrous annulus formation may originate from an epicardial problem.
T
ation depends on the capability of the native pulmonary root to function in the systemic circulation [2]. Neoaortic root dilatation is one of the late complications described after the ASO for TGA [2– 6, 9]. In a previous study we have shown that the pulmonary trunk and pulmonary sinus wall in unoperated hearts with TGA show a decrease of ␣-smooth muscle actin expression with increasing age [10]. Thus the structure of the vessel wall and sinus wall of the aorta and pulmonary trunk are different in hearts with TGA. As described above, after the ASO the pulmonary root remains on the left side and becomes the neoaortic root subjected to the systemic pressure. This may provide an explanation for the dilatation of the neoaortic root that is sometimes observed late after ASO. Apart from an abnormal and weaker ␣-smooth muscle actin expression of the wall of the pulmonary trunk [10] and the disconcordant ventriculoarterial connection described above, we hypothesize that in TGA the observed dilatation of the pulmonary root after ASO might also be attributable to histomorphologic defects. Therefore, we studied the morphology of the pulmonary and aortic roots in the left and right ventricles in hearts with TGA and compared our observations to the situation in normal hearts. We specifically paid attention to the amount
ransposition of the great arteries (TGA) is a common and severe congenital heart malformation, having an incidence of about 1 per 5,000 live births [1]. In TGA, the atria and ventricles are in the normal position (situs solitus and concordant atrioventricular connection), but the aorta arises from the right ventricle and the pulmonary trunk arises from the left ventricle (disconcordant ventriculoarterial connection), changing the serially connected pulmonary and systemic circulations into two parallel (separated) circulations. The arterial switch operation (ASO) is now the surgical procedure of choice for TGA [2– 6]. This operation involves transsection and reanastomosis of both pulmonary trunk and aorta above the sinuses of Valsalva and transplantation of the coronary arteries [7, 8]. After the ASO, the pulmonary root (neoaortic root) functions in the systemic circulation. The long-term success of this oper-
Accepted for publication June 22, 2009. Presented at the Poster Session of the Forty-third Annual Meeting of The Society of Thoracic Surgeons, San Diego, CA, Jan 29 –31, 2007. Address correspondence to Dr Gittenberger-de Groot, Department of Anatomy and Embryology, Leiden University Medical Center, PO Box 9600, Postal zone S-1-P, Leiden, 2300 RC, the Netherlands; e-mail: [email protected].
© 2009 by The Society of Thoracic Surgeons Published by Elsevier Inc
(Ann Thorac Surg 2009;88:1300 –5) © 2009 by The Society of Thoracic Surgeons
0003-4975/09/$36.00 doi:10.1016/j.athoracsur.2009.06.058
LALEZARI ET AL DILATATION OF THE NEOAORTIC ROOT
1301
Fig 1. Collagen distribution at the commissure level. Three-dimensional reconstruction is shown of the arterial roots of a normal heart in a right anterior view (A) and a heart with transposition of the great arteries (TGA) in a left anterior view (B). The amount of collagen (blue) is clearly decreased in the heart with transposition of the great arteries. Transverse sections (C–F) of the arterial root stained with Azan show the amount and distribution of collagen in the normal heart (C, D) and the heart with transposition of the great arteries (E, F). The intersection lines c in A and e in B refer to the respective sections below. At the commissure level, in contrast to the aorta (Ao) in the normal hearts (C, D), in the heart with transposition of the great arteries the collagenous network in the pulmonary trunk (PT) does not extend further into the adventitial layer (E, F). This collagenous layer in the heart with transposition of the great arteries is interrupted by the elastic lamellae (compare arrow in D with F). Scale bars: C, E ⫽ 600 m; D, F ⫽ 120 m. Color codes: aorta, transparent pink; atrial myocardium, light gray; mitral valve (MV), transparent yellow; pulmonary trunk, transparent green. (A ⫽ anterior; AoL ⫽ aortic lumen; L ⫽ left; LA ⫽ left atrium; P ⫽ posterior; PTL ⫽ pulmonary trunk lumen; R ⫽ right; RA ⫽ right atrium.)
of collagen and the distribution as well as myocardial support of the arterial roots using several histologic markers. A three-dimensional anatomic reconstruction was made to show the relationship of the arterial fibrous annulus with the surrounding atrial and ventricular myocardium.
Material and Methods Material Four unoperated human heart specimens aged 1 day (postpartum) were studied including two with normal anatomy and two with simple TGA. The hearts were
obtained from the Leiden Collection (Department of Anatomy and Embryology, Leiden University Medical Center, Leiden, The Netherlands). This study is in accordance with the institutional guidelines of the Leiden University Medical Center for the use of human tissue.
Methods All hearts were fixed in ethanol and glycerin and routinely processed for light microscopy. Transverse sections (10 m) were mounted serially onto glycerin-coated glass slides. The paraffin-embedded sections were deparaffinized and stained with hematoxylin-eosin, resorcin-fuchsin, modified
Fig 2. Collagen distribution at the mid-sinus level. Three-dimensional reconstruction (A, B) and transverse sections (C–F) are as described in Figure 1. At the mid-sinus level the pulmonary trunk (PT) in the heart with transposition of the great arteries shows a thin compact collagen layer at the outside and a less extensive anchorage to the adventitial layer, and both elastic lamella as well as collagen layers are relatively disorganized compared with the normal heart (compare C, D with E, F). Color codes are as in Figure 1. Scale bars: C, E ⫽ 200 m; D, F ⫽ 120 m. (A ⫽ anterior; AoL ⫽ aortic lumen; L ⫽ left; LA ⫽ left atrium; P ⫽ posterior; PTL ⫽ pulmonary trunk lumen; R ⫽ right; RA ⫽ right atrium; RCS ⫽ right coronary sinus; RFS ⫽ right-facing sinus.)
PEDIATRIC CARDIAC
Ann Thorac Surg 2009;88:1300 –5
1302
LALEZARI ET AL DILATATION OF THE NEOAORTIC ROOT
Ann Thorac Surg 2009;88:1300 –5
PEDIATRIC CARDIAC
Fig 3. Collagen distribution at the base of the sinus level. Three-dimensional reconstruction (A, B) and transverse sections (C–F) are as described in Figure 1. At the base of the sinus the collagen layer of the pulmonary trunk (PT) is thinner (E, F) and compared with the aorta in the normal heart (C, D) the fingerlike protrusions (arrowheads in D and F), which attach the sinus with the surrounding myocardium, are thinner and less extensive (E, F). Color codes are as in Figure 1. Scale bars: C, E ⫽ 200 m; D, F ⫽ 120 m. (A ⫽ anterior; AoL ⫽ aortic lumen; L ⫽ left; LA ⫽ left atrium; P ⫽ posterior; PTL ⫽ pulmonary trunk lumen; R ⫽ right; RA ⫽ right atrium; RCS ⫽ right coronary sinus; RFS ⫽ right-facing sinus.)
Van Gieson, and Azan for routine histologic assessment of the specimens according to the institutional guidelines.
Three-Dimensional Reconstruction One normal heart and one heart with TGA were selected for three-dimensional reconstruction based on the Azan-
stained sections, providing good discrimination between collagen fibers, myocardium, and vessel wall structures. The distance between the serial sections was 160 m. Digital photomicrographs were taken of the serial sections of the normal heart and the TGA specimen from just below until just above the arterial roots. The lower
Fig 4. Myocardial support. Three-dimensional reconstruction of the arterial root and parts of atrial and ventricular myocardium of a normal heart in a right anterior view (A, C) and a heart with transposition of the great arteries (TGA) in a left anterior view (B, D). Transverse sections C and D are cranial views and correspond to the intersection lines c and d in A and B, respectively. The aortic (Ao) annulus (blue) in the normal heart is supported by the atrial and ventricular myocardium forming a collar over most of its circumference (C). Both the aorta and pulmonary trunk (PT) in the heart with transposition of the great arteries appear to be placed on top of the heart instead of being surrounded by myocardium, as is seen in the normal heart (compare A with B). Only a small part of the circumference of the pulmonary trunk annulus is surrounded by myocardium (D). Color codes: aorta, transparent pink; atrial myocardium, light gray; mitral valve (MV), transparent yellow; pulmonary trunk, transparent green; ventricular myocardium, transparent dark gray. (A ⫽ anterior; L ⫽ left; LA ⫽ left atrium; LV ⫽ left ventricle; P ⫽ posterior; R ⫽ right; RA ⫽ right atrium; RV ⫽ right ventricle.)
Ann Thorac Surg 2009;88:1300 –5
1303
At the base of the aortic and pulmonary sinus the elastic lamellae are absent in the wall of the sinus, and the entire wall is composed of a collagen layer. In the normal hearts this collagen layer is thick and extends into the surrounding myocardium by means of several fingerlike protrusions, forming a strong anchorage (Figs 3C and 3D). In the TGA hearts, similar to the mid-sinus region, the collagen layer is thinner (Figs 3E and 3F) compared with the normal hearts. The collagen layer extends fingerlike protrusions into the surrounding myocardium; however, these protrusions are thinner and less extensive compared with the normal hearts (Figs 3E and 3F).
BASE OF THE SINUS LEVEL.
Myocardial Anchorage and Support Results Below we describe the differences in collagen distribution, myocardial anchorage, and embedding of the arterial root between the normal hearts and the hearts with TGA.
Collagen Distribution As described previously [11, 12], in the normal heart the relationship between collagen and vessel wall at the most distal part of the annulus (commissure) is different from that of the mid-sinus region and the base of the sinus. For this reason we have compared the collagen distribution of the arterial root of the normal hearts to hearts with TGA at three different regions of the sinus: (1) at the level of the commissure (Fig 1), (2) at the mid-sinus level (Fig 2), and (3) at the base of the sinus (Fig 3). Threedimensional reconstructions of both specimens show more clearly the distinct difference in collagen distribution of the annulus between the normal heart and the heart with TGA (Figs 1A, 1B, 2A, 2B, 3A, 3B). COMMISSURE LEVEL. At the distal level in the normal heart, collagen fibers arising from the aortic annulus continue into the valve commissures on the inside of the aortic root and then merge on the outside of the elastic aortic wall with the adventitial layer (Figs 1C and 1D). In TGA, the collagenous network forming the thin pulmonary annulus also concentrates distally to form the commissures. However, this layer does not extend any further to the adventitial layer surrounding the pulmonary trunk but only extends to the level just above the pulmonary annulus (Figs 1E and 1F).
At the mid-sinus level in the normal hearts the aortic sinus wall is surrounded by a thick collagen layer on the outside and by elastic lamellae at the luminal side (Figs 2C and 2D). Compared with the normal hearts, in the TGA hearts the compact collagen layer on the outside of the pulmonary trunk is thin, and both layers, elastic lamellae as well as collagen layer, are disorganized (Figs 2E and 2F). In addition, the collagen layer in the TGA hearts has less extensive anchorage to the adventitial layer compared with the normal hearts (compare Fig 2C with Fig 2E).
MID-SINUS LEVEL.
Above we have described in the normal hearts, at the base of the sinus the annulus has a stronger anchorage to the surrounding myocardium by means of fingerlike collagenous protrusions compared with the hearts with TGA. In the normal hearts the aorta is supported by its embedding in the myocardium of the left ventricular outflow tract, with the ventricular septum and atrial myocardium forming a partial collar around it, whereas the pulmonary trunk is positioned more superficially on top of the right ventricle, being only partially embedded by the myocardium of the right ventricular outflow tract and not being supported by atrial myocardium (Figs 4A, 4C). In the hearts with TGA most of the circumference of the pulmonary annulus appears to lie on top of the left ventricular outflow tract, rather than being embedded into it (compare Fig 4A with Fig 4B). Only a small part of the pulmonary annulus circumference is supported by the left ventricular myocardium (Fig 4D). Similar to the pulmonary annulus, the majority of the aortic annulus appears to be placed on top of the right ventricular outflow tract and is thus not embedded in the surrounding myocardium (Figs 4B and 4D). The atrial myocardium surrounds only a small part of the circumference of the pulmonary annulus compared with the aortic annulus of the normal hearts (compare Figs 4A and 4C with Figs 4B and 4D). Similar to the pulmonary annulus of the normal hearts, the aortic annulus has no support from the atrial myocardium (Figs 4B and 4D).
Comment The main focus of the current study concerns the question as to what could be responsible for the sometimes observed dilatation of the neoaortic root after ASO. Murakami and colleagues [13] describe a decrease of distensibility at the base of the neoaorta in post-ASO hearts, which they correlated with the disruption of vasa vasorum leading to the neoaortic dilatation later. In a previous study from our group we have shown that the arterial roots in TGA are structurally different from the arterial roots in a normal heart [10]. In the unoperated TGA specimens, dedifferentiation of smooth muscle cells was observed, especially in the pulmonary root. This may
PEDIATRIC CARDIAC
limit was at approximately three quarters of the baseapex distance. The upper limit was determined amply above the sinotubular junction of the vessel wall, which is the upper limit of the collagenous annulus. The serial photomicrographs were incorporated into the AMIRA V4.0 software package (Template Graphics Software, San Diego, CA). The different cardiovascular structures (ie, collagen, vessel wall, semilunar and atrioventricular valves, and myocardium) were color-labeled by means of manual drawing and processed for three-dimensional visualization. Processing conditions of the normal heart and TGA specimen were similar, thus providing good conditions for direct comparison between the reconstructions.
LALEZARI ET AL DILATATION OF THE NEOAORTIC ROOT
1304
LALEZARI ET AL DILATATION OF THE NEOAORTIC ROOT
be a factor leading to dilatation of the neoaortic root after the ASO. Although the pathogenesis of neoaortic root dilatation after ASO may be in part explained by the hypothesis of Murakami and colleagues [13], we postulate that the observed differences in collagen distribution and myocardial support of the arterial roots in TGA, as described in this study, as well as the structural differences of the roots in terms of the smooth muscle cell dedifferentiation [10], play a more significant role than the impaired blood flow of the vessel wall by disruption of the vasa vasorum. PEDIATRIC CARDIAC
Collagen Distribution The results observed in this study confirm the findings described by Bartelings and coworkers [12] concerning the fingerlike collagen structures extending from the annulus, as well as their anchorage into the ventricular myocardium in the normal heart. In the hearts with TGA, morphologic differences were noticed compared with the findings in the normal hearts. A striking difference between the TGA and the normal heart specimens was the distribution and quantity of collagen in the arterial roots. The amount of collagen appears to be less in the hearts with TGA than in the normal hearts. Furthermore, in the TGA specimens, the collagen did not spread to the adventitial layer at the site of the commissures, whereas this was seen in both arterial roots of the normal hearts. Additionally, both arterial roots in the TGA specimens appear to be less firmly supported by the myocardium of the corresponding ventricles than both arterial roots in the normal specimens. The development of TGA has been linked to neural crest migration abnormalities [14]. Neural crest cells are involved in initiating the septation of the cardiac outflow tract [15]. Exposing these cells to all-trans retinoic acid in animal models during development instigates several cardiac malformations, one of them being TGA [16 –19]. In TGA mouse embryos, Yasui and colleagues [20] observed an altered collagen distribution. They concluded that treatment with retinoic acid and the subsequent neural crest abnormalities leading to TGA or TGA-like anomalies induced the altered organization of extracellular matrix components, such as type I collagen and hyaluronic acid. A more recent study by Jenkins and associates [21] revealed malformed epicardium in retinoic acid receptor knockout mice. Formation of the fibrous skeleton of the heart, which consists of collagen, is linked to epicardial differentiation and migration [22]. When the epicardium is inhibited during development, the fibrous skeleton of the outflow tract is underdeveloped and its collagen appears to be distributed differently compared with the normal situation [23]. Thus, the difference in collagen distribution between TGA and normal hearts, as observed in this study, may also originate from an epicardial problem during early development, because altered fibrous skeleton formation is also seen in retinoic acid receptor knockout mice.
Ann Thorac Surg 2009;88:1300 –5
Myocardial Support In the normal heart, the aortic root is supported by atrial myocardium, left ventricular outflow myocardium, and the ventricular septum, forming a myocardial cuff surrounding the root. The pulmonary root in the normal heart is not supported by the atrial myocardium and only the pulmonary infundibular myocardium bulges slightly over the pulmonary root, forming a modest collar surrounding it and providing some support, Both roots are attached to the surrounding myocardium by fingerlike protrusions [11, 12]. By studying the three-dimensional reconstructions and the histologic sections of the hearts with TGA, it is apparent that none of the roots is supported by the atrial myocardium. Both roots are placed more superficially on top of the ventricles, missing the sufficient myocardial cuff surrounding the roots as was the case in the normal heart. The attachment of the roots to the surrounding myocardium by fingerlike protrusions was less extensive and thinner compared with the normal heart. The retrospective study by Vandekerckhove and associates [9], in which 39 patients were examined 20 years after ASO, revealed 60% dilatation of the neoaortic root in the operated population. Long-term follow-up of this population is necessary to observe whether the remaining 40% develop dilatation of the neoaortic root, which we relate to the morphologic changes of the arterial root of the hearts with TGA.
Conclusions In this study we observed an altered collagen distribution of the annulus at all three levels of both great arteries studied in TGA. Furthermore, both the aorta and the pulmonary trunk were found to be less firmly anchored to their corresponding ventricles in the TGA specimens than in the normal heart specimens. We speculate that the findings in the present study, together with the observed structural differences between the arterial roots in TGA and normal hearts, in combination with subjecting the neoaortic root to systemic pressures after ASO, might form the basis of the neoaortic root dilatation that occurs after ASO.
We thank Jan Lens for expert technical assistance with the figures.
References 1. Love BA, Mehta D, Fuster VF. Evaluation and management of the adult patient with transposition of the great arteries following atrial-level (Senning or Mustard) repair. Nat Clin Pract Cardiovasc Med 2008;5:454 – 67. 2. Hourihan M, Colan SD, Wernovsky G, Maheswari U, Mayer JE Jr, Sanders SP. Growth of the aortic anastomosis, annulus, and root after the arterial switch procedure performed in infancy. Circulation 1993;88:615–20. 3. Hazekamp MG, Schoof PH, Suys BE, et al. Switch back: using the pulmonary autograft to replace the aortic valve after arterial switch operation. J Thorac Cardiovasc Surg 1997;114:844 – 6.
4. Haas F, Wottke M, Poppert H, Meisner H. Long-term survival and functional follow-up in patients after the arterial switch operation. Ann Thorac Surg 1999;68:1692–7. 5. Formigari R, Toscano A, Giardini A, et al. Prevalence and predictors of neoaortic regurgitation after arterial switch operation for transposition of the great arteries. J Thorac Cardiovasc Surg 2003;126:1753–9. 6. McMahon CJ, Gauvreau K, Edwards JC, Geva T. Risk factors for aortic valve dysfunction in children with discrete subvalvar aortic stenosis. Am J Cardiol 2004;94:459 – 64. 7. Laks H. The arterial switch procedure for the neonate: coming of age. Ann Thorac Surg 1989;48:3– 4. 8. Castaneda AR, Mayer JE Jr, Jonas RA, Wernovsky G, di Donato R. Transposition of the great arteries: the arterial switch operation. Cardiol Clin 1989;7:369 –76. 9. Vandekerckhove KD, Blom NA, Lalezari S, Koolbergen DR, Rijlaarsdam ME, Hazekamp MG. Long-term follow-up of arterial switch operation with an emphasis on function and dimensions of left ventricle and aorta. Eur J Cardiothorac Surg 2009;35:582– 8. 10. Lalezari S, Hazekamp MG, Bartelings MM, Schoof PH, Gittenberger-de Groot AC. Pulmonary artery remodeling in transposition of the great arteries: relevance for neoaortic root dilatation. J Thorac Cardiovasc Surg 2003;126:1053– 60. 11. Hokken RB, Bartelings MM, Bogers AJJC, Gittenberger-de Groot AC. Morphology of the pulmonary and aortic roots with regard to the pulmonary autograft procedure. J Thorac Cardiovasc Surg 1997;113:453– 61. 12. Bartelings MM, Bogers AJ, Galantowicz ME, Gittenberger-de Groot AC. Anatomy of the aortic and pulmonary roots. Semin Thorac Cardiovasc Surg Pediatr Card Surg Annu 1998;1:157– 64. 13. Murakami T, Nakazawa M, Momma K, Imai Y. Impaired distensibility of neoaorta after arterial switch procedure. Ann Thorac Surg 2000;70:1907–10. 14. Costell M, Carmona R, Gustafsson E, Gonzalez-Iriarte M, Fassler R, Munoz-Chapuli R. Hyperplastic conotruncal en-
LALEZARI ET AL DILATATION OF THE NEOAORTIC ROOT
15. 16.
17. 18.
19.
20.
21. 22.
23.
1305
docardial cushions and transposition of great arteries in perlecan-null mice. Circ Res 2002;91:158 – 64. Yu J, Gonzalez S, Martinez L, ez-Pardo JA, Tovar JA. Effects of retinoic acid on the neural crest-controlled organs of fetal rats. Pediatr Surg Int 2003;19:355– 8. Waldo K, Miyagawa-Tomita S, Kumiski D, Kirby ML. Cardiac neural crest cells provide new insight into septation of the cardiac outflow tract: aortic sac to ventricular septal closure. Dev Biol 1998;196:129 – 44. Lammer EJ, Chen DT, Hoar RM, et al. Retinoic acid embryopathy. N Engl J Med 1985;313:837– 41. Yasui H, Nakazawa M, Morishima M, Miyagawa-Tomita S, Momma K. Morphological observations on the pathogenetic process of transposition of the great arteries induced by retinoic acid in mice. Circulation 1995;91:2478 – 86. Yasui H, Morishima M, Nakazawa M, Ando M, Aikawa E. Developmental spectrum of cardiac outflow tract anomalies encompassing transposition of the great arteries and dextroposition of the aorta: pathogenic effect of extrinsic retinoic acid in the mouse embryo. Anat Rec 1999;254:253– 60. Yasui H, Nakazawa M, Morishima M, Aikawa E. Altered distribution of collagen type I and hyaluronic acid in the cardiac outflow tract of mouse embryos destined to develop transposition of the great arteries. Heart Vessels 1997;12: 171– 8. Jenkins SJ, Hutson DR, Kubalak SW. Analysis of the proepicardium-epicardium transition during the malformation of the RXRalpha-/- epicardium. Dev Dyn 2005;233:1091–101. Gittenberger-de Groot AC, Vrancken Peeters M-PFM, Mentink MMT, Gourdie RG, Poelmann RE. Epicardium-derived cells contribute a novel population to the myocardial wall and the atrioventricular cushions. Circ Res 1998;82:1043–52. Eralp I, Lie-Venema H, DeRuiter MC, et al. Coronary artery and orifice development is associated with proper timing of epicardial outgrowth and correlated Fas ligand associated apoptosis patterns. Circ Res 2005;96:526 –34.
PEDIATRIC CARDIAC
Ann Thorac Surg 2009;88:1300 –5