I of logical pathological and surgical considerations, a

Transcription

1 Fibrous Skeleton and Ventricular Outflow Double-Outlet Right Ventricle C. Eric Howell, MD, Siew Yen Ho, PhD, Robert H. Anderson, MD, and Martin J. Elliott, MD Tracts in Department of Paediatrics, National Heart & Lung Institute. Brornpton Hospital, and the Cardiothoracic Unit, The Hospital for Sick Children, London, United Kingdom Twenty-four hearts in which both great arteries arose from the right ventricle were studied to establish variations present within the fibrous skeleton and infundibular morphologies. Variations were also noted in the location of the ventricular septal defect and measurements were obtained of the outlet septum and the circumferences of the arterial valves. Completely muscular subarterial infundibulums were present in only 9 (37.5%) of the hearts, with varying degrees of fibrous continuity between the leaflets of the arterial and atrioventricular valves in the remainder. The aorta was rightward and posterior in 12 (50%) of the hearts, and subaortic and subpulmonary ventricular septal defects were present in equal numbers in this group. No subaortic defects were present when the aorta was side-by-side and right-sided. No subpulmonary defects were present in hearts with a posterior aorta. The mean ratio of 0.91 f 0.36 for the subpulmonary to subaortic length of the outlet septum was significantly less than the value of 1.54 & 0.41 noted previously in hearts with tetralogy of Fallot (p < 0.001). ( ) n 1972, Lev and associates [ 11 propounded, on the basis I of logical pathological and surgical considerations, a concept of defining the heart with double-outlet right ventricle in terms of the type and location of the ventricular septal defect. This concept excluded from consideration neither those hearts with incomplete origin of both arterial trunks from the right ventricle (provided the majority of both arterial circumferences arose from the right ventricle) nor those hearts with fibrous continuity between the leaflets of the arterial and atrioventricular valves. This was in contrast to the "prevalent concept at the present time" [l], namely that double outlet from the right ventricle should only be diagnosed when both arteries arose exclusively from the right ventricle, with the leaflets of each arterial valve supported exclusively by completely muscular infundibular structures. Since then, subsequent morphological and surgical studies [2-51 have argued in favor of using the "50% rule" to define hearts having a double-outlet right ventricle ventriculoarterial connection. Controversy persists, nonetheless, with other authors [6, 71 still strongly supporting the restrictive definition of double-outlet right ventricle for cases with exclusively right ventricular origin of both great arteries in the presence of discontinuity bilaterally between leaflets of the arterial and atrioventricular valves. In the light of this continuing debate, particularly because the concept of a bilateral infundibulum is held to represent prevailing opinion [6], we have studied the morphology of the ventricular outflow tracts in those Accepted for publication Oct 29, Address reprint requests to Prof Anderson, Department of Paediatrics, National Heart & Lung Institute, Dovehouse St, London SW3 6LY, United Kingdom. hearts within our collection having exclusively right ventricular origin of both arterial trunks, comparing our findings in this respect with those of a similar study performed on hearts with tetralogy of Fallot [8]. This latter point is of major importance, because much of the debate concerning the role of infundibular morphology in categorization of double-outlet right ventricle centers upon the similarities and differences between such hearts and those with tetralogy of Fallot [6]. Material and Methods Hearts were selected from the cardiopathological collection of the National Heart and Lung Institute, Brompton Hospital, London, choosing the 24 specimens with usual atrial arrangement, concordant atrioventricular connections, and, as judged by us, exclusive origin of both arterial trunks from the morphologically right ventricle. The study was limited, therefore, to those hearts in which there was no extension of an arterial trunk into the cavity of the left ventricle. In addition, hearts with the overall morphology of tetralogy of Fallot (subaortic ventricular septal defect and anterior deviation of the outlet septum with infundibular pulmonary stenosis) were excluded even if the aorta also arose exclusively from the right ventricle. The 24 selected hearts were then studied with particular attention directed toward the morphology of the areas of fibrous continuity between the leaflets of the arterial and atrioventricular valves, the orientation and position of the great arteries, the type of ventricular septal defect, the dimensions of the outlet septum, and the circumferences of the great arteries, noting in this respect the presence of any subarterial infundibular stenosis. Integrity of the by The Society of Thoracic Surgeons /91/$3.50

2 HOWELL ET AL 395 Fig 1. Schematic representation of patterns of morphology recognized in the hearts within this study, vimed from the ventricular aspect in the short axis. (A = aortic valve; ALMV = aortic leaflet of mitral valve; ALTV = anterosuperior leaflet of tricuspid valve; ' I = pulmonary valve; SLTV = septal leaflet of tricuspid valve; VS = ventricular septum; = region of valvar jibrous continuity.) specimen was another important factor in selection, and we excluded some that would otherwise have qualified but were excessively distorted by surgical or postmortem alteration. Measurements of the subaortic and subpulmonary infundibular lengths were taken from the crest of the outlet septum to the closest point of attachment of the leaflets of the arterial valves. Aortic and pulmonary internal circumferences were measured at the superior level of the attachment of the valvar leaflets. Ratios were derived for infundibular lengths and truncal circumferences by using the values for the aorta as the denominator. Results The basic patterns of morphology observed in the hearts are shown schematically in Figure 1. Of the 24 hearts examined, 15 (62.5%) exhibited some degree of continuity between the leaflets of the arterial and atrioventricular valves. Six (40%) of the 15 showed continuity between the pulmonary and atrioventricular valvar leaflets. One of these hearts displayed continuity only between the pulmonary and tricuspid valve, with a muscular ventricular septal defect being present in a noncommitted location (Fig 2). A further 9 hearts (60%) demonstrated continuity between the leaflets of the aortic and atrioventricular valves. In 1 heart with pulmonary-mitral valvar continuity, and in 1 with aortic-mitral continuity, the fibrous band between the leaflets was incomplete, a muscle bar of 3 to 4 mm being interposed between the attachments (Fig 3). Five of the remaining 9 hearts that did not have any fibrous continuity between the leaflets of the arterial and atrioventricular valves did, however, display continuity between the leaflets of the mitral and tricuspid valves (Fig 4). This is the criterion we use to designate a ventricular septal defect as being perimembranous in the setting of double-outlet right ventricle. The other 4 specimens exhibited discontinuity between all valvar structures. Taken together, only just over one-third of this selection of hearts satisfy the supposed "prevalent" [6] definition for the heart with double-outlet right ventricle, as exemplified by the specimen shown in Figure 4. Within the overall series, the ventricular septal defect was subaortic in 9 hearts, subpulmonary in 7, and noncommitted in 7; the remaining ventricular septal defect was doubly committed in the presence of fibrous continuity between the leaflets of the aortic and pulmonary valves (Fig 5). The ventricular septal defects were perimembranous in 14 hearts, in that there was continuity between the leaflets of the mitral and tricuspid valves in the posteroinferior rim of the defect as described in the preceding paragraph (see Fig 4). The posteroinferior rim was muscular in the remaining 10 hearts, the ventriculoinfundibular fold fusing with the posterior limb of the septomarginal trabeculation to create discontinuity between the leaflets of the mitral and tricuspid valves. All hearts with aortic-mitral valvar continuity had subaortic or doubly committed ventricular septal defects, and all

3 396 HOWELL ET AL Fig 2. Specimen demonstrating fibrous continuity only between the leaflets of the pulmonary and tricuspid valves (stippled margin). A noncommitted defect is present within the trabecular septum. (ALTV = anterosuperior leaflet of tricuspid Valve; AOV = aortic valve; 0s = outlet septum; PV = pulmonary valve; SLTV = septal leaflet of tricuspid valve; SMT = septomarginal trabeculation; VSD = ventricular septa1 defect.) Fig 3. Specimen demonstrating fibrous continuity between the leaflets of the pulmonary valve and the mitral valve at the margins (arrowheads) with a central muscular separation (stippled margin). (ALMV = aortic leaflet of mitral valve; = aorta; SMT = septomarginal trabeculation; TV = tricuspid valve; VIF = ventriculoinfundibular fold; pv = pulmonary valve; VSD = ventricular septa1 defect.) hearts with pulmonary-mitral valvar continuity had subpulmonary ventricular septal defects. The position of the aorta relative to the pulmonary trunk is shown in Figure 6, with subdivision of the type of septal defects related to each aortic location. An aorta located rightward and posterior in relation to the pulmonary trunk was most common, being seen in 12 hearts (50%), followed by right (in side-by-side orientation) and posterior locations, with 5 hearts in each category. One heart had the aorta in a left-sided and anterior position, whereas 1 had a directly anterior aorta. Of the 5 hearts with a posterior aorta, all had aortic-mitral valvar continuity with subaortic ventricular septal defects (1 was doubly committed). The 2 hearts with anterior and left-anterior positioning of the aorta had noncommitted perimembranous ventricular septal defects and arteriabatrioventricular valvar discontinuity. No subaortic ventricular septal defects were found with rightward positioning of the aorta, and no subpulmonary defects were noted with posterior aortic location. None of the hearts had serious subpulmonary stenosis. In all hearts in which the origin of the orifices of the coronary arteries could be noted, they came from the aortic sinuses facing the pulmonary trunk irrespective of Fig 4. Specimen with double-outlet right ventricle, complete subaortic and subpulmona ry muscular infundibulums, and a perimembranous ventricular septal deject. The asterisk indicates the fibrous posteroinferior margin of the defect produced by fibrous continuity between the leaflets of the mitral and tricuspid valves. (ALTV = anterosuperior leaflet of tricuspid valve; AoV = aortic valve; 0s = outlet septum; PV = pulmona ry valve; SLTV = septal leaflet of tricuspid valve; VSD = ventricular septal defect.)

4 HOWELL ET AL 397 VARIATIONS IN DOUBLE-OUTLET FXHT VENTRICLE 1 noncommitted I 0 subaortic 5 subpulmonary 2 noncommitted 4 subaortic 1 doubly committed 0 subpulmonary Post. Fig 6. Schematic diagram of aortic positions in relation to the pulmonary trunk, with division of the types of ventricular septal defects found with each aortic position. Fig 5. Specimen with double-outlet right ventricle, a ventricular septal defect that is subarterial and doubly committed but with a muscular posteroinferior rim, and fibrous continuity between the leaflets of the aortic and pulmonary valves (stippled margin). (ALTV = anterosuperior leaflet of tricuspid valve; AoV = aortic valve; PV = pulmom y valve; SLTV = septal leaflet of tricuspid valve; SMT = septomarginal trabeculation; VIF = ventriculoinfundibular fold; VSD = ventricular septal defect.) truncal orientation. One heart had a single coronary artery that arose from the left-hand facing aortic sinus. No coronary artery took origin from the pulmonary trunk. Measurement of subpulmonary and subaortic lengths yielded a subpulmonary to subaortic ratio of 0.91 f 0.36 (n = 18; range, 0.42 to 1.56; median, 0.79). The ratio of the pulmonary to aortic circumference was 1.21 f 0.72 (n = 20; range, 0.25 to 2.67; median, 1.12). Of 6 hearts that exhibited subpulmonary ventricular septal defects, and in which the great arteries were sufficiently intact to allow measurement, all had circumferential ratios greater than 1.0 (pulmonary trunk larger than aorta). Of 5 hearts with subaortic ventricular septal defects and great arteries suitable for measurement, all had circumferential ratios less than 1.0 (aorta larger than pulmonary trunk). Comment Despite restricting our selection of hearts studied to those having virtually exclusive connection of both great arteries to the cavity of the right ventricle, considerable varia- tion was seen in the morphology of the fibrous skeleton and the orientation of the great arterial trunks. This was particularly marked with regard to the infundibular morphologies observed. In light of these findings, it seems too restrictive to maintain [6] that a heart must have both subaortic and subpulmonary infundibulums before it may be described as having double outlet from the morphologically right ventricle. Stringent application of this traditional criterion would have rendered almost two-thirds of this selection of hearts unclassifiable as double-outlet right ventricle despite the fact that both great arteries arose exclusively, as far as we could judge, from the right ventricle. The origination of these "traditional" criteria has previously been brought into question by Bartelings [5], who noted the inconsistency between the early morphological descriptions of hearts with double-outlet right ventricle and studies that have subsequently employed the criterion of the bilateral infundibulum to define the entity. The landmark works of Neufeld and associates [9, 101, which provided some of the earliest detailed descriptions of hearts with double-outlet right ventricle, clearly established the existence of hearts with fibrous continuity between the leaflets of the aortic and mitral valves as a part of the spectrum of hearts with unequivocal connection of both great arteries to the right ventricle. As discussed by these authors, the co-existence of these two features resulted in elongation of the aortic leaflet of the mitral valve. It appears that it was the detailed description of the so-called Taussig-Bing heart [ll, 121, the original of which clearly possessed discrete subaortic and subpulmonary infundibulums (although the precise connection of the pulmonary trunk remained controversial [ 13]), that provided the impetus for the application of the features of the bilateral infundibulum to all hearts classified as double-

5 398 HOWELL ET AL outlet right ventricle. This criterion then received support from the radiographic studies of Baron [14] and Hallerman and co-workers [15], which put reliance upon separation of the leaflets of the mitral and aortic valves for the differentiation between tetralogy of Fallot and double outlet. This was despite, however, the reference by Baron [14] to the work of Neufeld and associates [lo] that, as Bartelings [5] indicated, described a subgroup of hearts with double outlet that exhibited continuity between the leaflets of the aortic and mitral valves. Presumably, the desire to retain the criterion of the bilateral infundibulum reflects a temptation to preserve nomenclature that refers to a specific morphological entity and that allows instantaneous and consistent recognition of that entity as well as its separation from others (61. It is difficult now to find either morphological, clinical, or embryological bases for this approach. If used, confusion is the inexorable result. What name, then, is to be given to the hearts not possessing a bilateral infundibulum by using this approach? We should emphasize in this context that none of the hearts included within our present study showed the morphology of tetralogy of Fallot. Thus, our discussions Concerning the significance of a bilateral infundibulum in the diagnosis of double-outlet right ventricle are peripheral, albeit important, to the ongoing debate concerning the distinction of the lesion from tetralogy of Fallot. It should also be noted in this context that no specimen had subpulmonary stenosis. This simply reflected the morphology of the hearts at our disposal. The debate over traditional separations between hearts with double-outlet right ventricle and those with tetralogy of Fallot also has unclear foundations in light of the early report by Witham [16], the recognized initiator of the term double-outlet right ventricle. He grouped the hearts studied as those with true persistent truncus, the Eisenmenger type, and the Fallot type! Although he did not address valvar fibrous continuity as a morphological feature, he made a clinical and pathological distinction between the Fallot type of double-outlet right ventricle and hearts with tetralogy of Fallot that lack complete origin of the aorta from the right ventricle. This was primarily on the basis that the former displayed hypertrophy of both ventricles, rather than only of the right ventricle, as in the latter. It must always be remembered that these earlier reports came in an era when diagnostic techniques were often insufficient to demonstrate all details of a given lesion, so that often it was necessary to make inferences. This is no longer the case. Currently, all pertinent features can now accurately be determined during life. In this light, we submit that clarity and consistency in description must account for all features of a malformed heart. Thus, tetralogy of Fallot with double-outlet ventriculoarterial connection becomes more descriptive and clinically relevant than simply tetralogy of Fallot when the greatest part of the aorta is connected to the right ventricle in a patient whose heart bears all the morphological features of tetralogy. Similarly, because of the well-recognized variability in location of the ventricular septal defect and its effect on clinical presentation, mere description of double-outlet right ventricle in the traditional sense now retains little value. Numerous surgical and morphological studies [14, 17, 181 have testified to the importance of characterization of the differences in these hearts in terms of the orientation of the great arteries and the type of ventricular septal defect (subaortic, subpulmonary, doubly committed, and noncommitted) for surgical decision-making and prognosis. In this study, we have confirmed that, even when both arterial trunks are unequivocally connected to the right ventricle, there is marked inconsistency in the extent of obliteration of the muscular inner heart curvature ( absorption of the conoventricular flange ) in producing continuity or discontinuity between the leaflets of the arterial and atrioventricular valves. This feature, therefore, has utility in morphological description but is unreliable as a criterion of morphological classification. As variations continue to be detailed, as in this and other studies, the need for less confusing and more descriptive terminology becomes more pronounced. Such efforts can be facilitated by using the term double outlet to describe a specific ventriculoarterial connection, to which it is suited, rather than to describe infundibular morphology, in which role it remains both confusing and counterproductive. In terms of morphogenesis, the observed variations in arterial position and formation of the subarterial musculature support a role for the processes of absorption, rotation, and medial translocation of the ventricular outflow tracts. As observed by Van Mierop and Gessner [19] in studies of human embryos, there is leftward translocation of the outflow tracts during horizons 12 through 14 as described by Streeter [20, 211 (23 to 26 days gestational age). Over this period, there is a shift from an exclusive origin of the outlet component from the developing right ventricle to that producing an outlet in part from the left ventricle. This process occurs before the development of the swellings responsible for septation of the outflow tracts and the arterial pole. Then, as normal septation proceeds, the aortic orifice is in position to become connected with the developing outlet component of the left ventricle. A defect in the process of leftward shift would leave an exclusive connection to the right ventricle, or, if incomplete, predominant connection to the right ventricle. Further anomalous development in the process of septation could then produce the orientational variations of the great arteries as we have observed. The more recent embryological studies of Bartelings and Gittenberger-de Groot [22] have shown that the process of septation of the ventricular outflow tract proceeds before that of truncal septation. The aortic orifice is already in close proximity to the superior atrioventricular cushion within the atrioventricular orifice, suggesting the minimal presence of a subaortic infundibulum in this area which must, effectively, be resorbed to produce valvar continuity. As regression of the muscular inner heart curvature proceeds in the normal and abnormal heart, and based on precedent normal or abnormal translocation or septation, our observed variations in the fibrous skeleton suggest that it is the orifice of the arterial trunk

6 1991 :51: HOWELL ET AL 399 initially in closest proximity to the superior atrioventricular endocardia1 cushion that achieves fibrous continuity with the atrioventricular valve(s). And, as the variations suggest, the irregularity of this regressive process produces a spectrum, in these selected hearts, from complete discontinuity to extensive areas of valvar continuity, with intermediate forms that include muscular remnants of the inner curve within those areas of fibrous continuity (see Fig 3). Experiments with irradiation of mammalian embryos by Okamoto and associates [23] have shown that development of the ventricular outflow tracts is a complex process dependent on the formation of the cardiac loop; the volume and length of the trunks; the arrangement of the valvar cushions and outflow ridges; and the position, amount, and timing of appearance and disappearance of focuses of cell death within the ridges. Recognition of the many variables involved should presage the expectation of the broad variations in pathology that have been historically observed. Although our sample of specimens was small and selected, we feel justified in making other inferences from our findings. From the relationships between aortic position and type of defect it appears that, in the presence of a right aorta (in side-by-side orientation), the finding of a subaortic defect would be unlikely, and similarly, in the presence of a posteriorly positioned aorta, a subpulmonary defect would be unlikely. Measurements of the components of the outlet septum revealed virtual comparability in subaortic and subpulmonary lengths. Such a relationship was also alluded to in the reports by Neufeld and associates [9, 101 and Witham [16], with their observations that the distinguishing feature between hearts having tetralogy of Fallot and those with origin of both great vessels from the right ventricle was that in the latter, both semilunar valves are in the same cross-sectional and coronal body planes [9]. This is a finding that contrasts with those in our recent study of hearts with tetralogy of Fallot [8]. In the latter hearts, using similar measurements, we noted a mean ratio of subpulmonary to subaortic lengths of 1.54 & The mean of obtained in this selection of hearts gives a difference that is statistically significant from that obtained in the hearts with tetralogy (Student s t test, p < 0.001). Another finding in the embryologic study of Bartelings and Gittenberger-de Groot [22], correlating with their observation that the aortic orifice achieved proximity to the atrioventricular orifice concomitant with the beginnings of septation of the outflow tract but before that of truncal septation, was that this preexisting proximity produced a longer subpulmonary than subaortic length in the outlet septum. It is the infundibular morphology, therefore, that determines the existence of tetralogy of Fallot, even when both arterial trunks spring almost exclusively from the right ventricle. Due to the minute dimensions of the outlet septum in the mature heart, comparisons between hearts with tetralogy and normal hearts are difficult, if not impossible. The differential ratios, nonetheless, suggest differing temporal relationships in the defects producing hearts with unequivocal double-outlet right ventricle in the absence of tetralogy as opposed to those with the morphology of tetralogy. This supports the suggestion by Van Mierop and Gessner [19] that the basic defect in double outlet is at the stage of leftward shift of the outflow tract before septation and its attendant development of a shorter subaortic infundibular length. The differential in length found in hearts with tetralogy then supports the suppositions by Goor and co-workers [24] and Becker and colleagues [3] that the morphogenesis of tetralogy is due to anomalous septation and inversion, steps that occur subsequent to the leftward shift of the outflow tracts and, according to the work by Bartelings and Gittenberger-de Groot [22], also subsequent to the establishment of the differential subarterial lengths as in normal embryos. The finding in this study of a larger circumference of the great vessel to which the ventricular septa1 defect was committed may also be of developmental importance. This, too, reflects the early stage of development in which the overall anomaly is produced. It supports the importance of hemodynamic factors in growth of the great vessels, because the committed vessel would receive the majority of the outflow from both ventricles. References 1. Lev M, Bharati S, Meng CCL, Liberthson RR, Paul MH, Idriss F. A concept of double-outlet right ventricle. J Thorac Cardiovasc Surg 1972;64: Wilcox BR, Ho SY, Macartney FJ, Becker AE, Gerlis LM, Anderson RH. Surgical anatomy of double-outlet right ventricle with situs solitus and atrioventricular concordance. J Thorac Cardiovasc Surg 1981;82: Becker AE, Conner M, Anderson RH. Tetralogy of Fallot: a morphometric and geometric study. Am J Cardiol 1975;35: Kirklin JW, Pacific0 AD, Bargeron LM, Soto 8. Cardiac repair in anatomically corrected malposition of the great arteries. Circulation 1973;48: Bartelings MM. The outflow tract of the heart. Embryologic and morphologic correlations (Thesis). Leiden: University of Leiden, 1990:9!L Van Praagh R. Etienne-Louis Arthur Fallot and his tetralogy: a new translation of Fallot s summary and a modern reassessment of this anomaly. Eur J Cardiothorac Surg 1989;3: Judson JP, Danielson GK, Puga FJ, Mair DD, McGoon DC. Double-outlet right ventricle. Surgical results J Thorac Cardiovasc Surg 1983;85: Howell CE, Ho SY, Anderson RH, Elliott MJ. Variations within the fibrous skeleton and ventricular outflow tracts in tetralogy of Fallot. 1990;50: Neufeld HN, DuShane JW, Wood EH, Kirklin JW, Edwards JE. Origin of both great vessels from the right ventricle. I. Without pulmonary stenosis. Circulation 1961;23: Neufeld HN, DuShane JW, Edwards JE. Origin of both great vessels from the right ventricle. 11. With pulmonary stenosis. Circulation 1961;23: Taussig HB, Bing RJ. Complete transposition of the aorta and a levoposition of the pulmonary artery. Am Heart J 1949; Van Praagh R. What is the Taussig-Bing malformation [Editorial]? Circulation 1968;38: Hinkes P, Rosenquist GC, White RI Jr. Roentgenographic re-examination of the internal anatomy of the Taussig-Bing heart. Am Heart J 1971;81:335-9.

7 400 HOWELL ET AL 1991;51: Baron MG. Angiographic differentiation between tetralogy of Fallot and double-outlet right ventricle. Relationship of the mitral and aortic valves. Circulation 1971;53: Hallerman FJ, Kincaid OW, Ritter DG, Titus JL. Mitralsemilunar valve relationships in the angiography of cardiac malformations. Radiology 1970; Witham AC. Double outlet right ventricle. A partial transposition complex. Am Heart J 1957;53:92& Kirklin JW, Pacific0 AD, Blackstone EH, Kirklin JK, Bargeron LM Jr. Current risks and protocols for operations for doubleoutlet right ventricle. Derivation from an 18 year experience. J Thorac Cardiovasc Surg 1986; Musumeci F, Shumway S, Lincoln C, Anderson RH. Surgical treatment for double-outlet right ventride at the Brompton Hospital, J Thorac Cardiovasc Surg 1988;96: Van Mierop LHS, Gessner IH. Pathogenetic mechanisms in congenital cardiovascular malformations. Prog Cardiovasc Dis 1972;15: Streeter GL. Developmental horizons in human embryos. Description of age group XI, 13 to 20 somites, and age group XII, 21 to 29 somites. Carnegie Inst Wash Publ 541, Contrib Embryol 1942; Streeter GL. Developmental horizons in human embryos. Description of age group XIII, embryos about 4 or 5 millimeters long, and age group XIV, period of indentation of the lens vesicle. Carnegie Inst Wash Publ 541, Contrib Embryol 1945;31: Bartelings MM, Gittenberger-de Groot AC. The outflow tract of the heart-embryologic and morphologic correlations. Int J Cardiol 1989;22: Okamoto N, Satow Y, Hidaka N, Akimoto N, Miyabara S. Morphogenesis of congenital heart anomaly-bulboventricular malformations. Jpn Circ J 1978;42: Goor DA, Dische R, Lillehei CW. The conotmncus. I. Its normal inversion and conus absorption. Circulation 1972;46: