Clinical Anatomy 28:477–486 (2015)
REVIEW
The Importance of Being Isomeric ROBERT H. ANDERSON,1* NIGEL A. BROWN,2 CHIKARA MENO,3 1
AND
DIANE E. SPICER4,5
Institute of Genetic Medicine, Newcastle University, Newcastle upon Tyne, United Kingdom 2 Division of Biomedical Sciences, St George’s University, London, United Kingdom 3 Graduate School of Medical Sciences, Kyushu University, Fukuoka, Kyushu, Japan 4 Department of Pediatric Cardiology, University of Florida, Gainesville, Florida 5 Congenital Heart Institute of Florida, St. Petersburg, Florida
In the normal individual, the parietal components of the body are mirrorimaged and appropriately described as isomeric. The thoraco-abdominal organs, in contrast, are lateralized. However, in “visceral heterotaxy,” the thoraco-abdominal organs also show some degree of isomerism, best seen in the arrangement of the bronchial tree. Whether isomerism can be found within the heart remains controversial. One of two recent publications in this journal emphasized the crucial features of bronchial isomerism; the other, in contrast, confused the situation of isomerism within the heart. In this review, we show how the topic of cardiac isomerism is clarified by concentrating on the anatomical features of the cardiac components and determining how best they can be described. Appropriate manipulation of developing mice produces unequivocal evidence of isomerism of the atrial appendages, but with no evidence of ventricular isomerism. In hearts from patients with so-called “heterotaxy,” only the atrial appendages, distinguished on the basis of the pectinate muscles lining their walls, are uniformly isomeric, permitting the syndrome to be differentiated into the subsets of left as opposed to right atrial appendage isomerism. Thus, controversies are defused by simply describing the isomerism of the atrial appendages rather than “atrial isomerism,” recognizing the frequency of abnormal venoatrial connections in these settings. Any suggestion of ambiguity is removed by the equally simple expedient of describing all the variable cardiac features, describing the arrangements of the thoracic and abdominal organs separately should there be discordances. Clin. Anat. 28:477–486, 2015. VC 2015 Wiley Periodicals, Inc. Key words: visceral heterotaxy; situs solitus; left isomerism; right isomerism; morphological method
INTRODUCTION In a review in a recent issue of this journal, Spentzouris and colleagues discussed the anatomy of the inferior caval vein and included the abnormalities found in so-called “situs ambiguus” (Spentzouris et al., 2014). They suggested that this syndrome, also known as visceral heterotaxy, “is a severe variant of situs inversus which involves a complete breakdown in the classical asymmetry found in the abdominal viscera.” It is questionable whether the constellation of abnormal findings is a “severe variant of situs inversus.” Rather, the “breakdown in classical asymmetry” is of greatest significance, but this involves not only
C V
2015 Wiley Periodicals, Inc.
the abdominal but also the thoracic organs (Jacobs et al., 2007). Indeed, in the article immediately preceding the review relating to the inferior caval vein, in the same issue of the journal, another group of authors including two of those who collaborated with
*Correspondence to: Robert H. Anderson, Institute of Genetic Medicine, Newcastle University, Newcastle upon Tyne, Tyne and Wear NE1 7RU, United Kingdom. E-mail: [email protected] Received 14 January 2015; Accepted 18 January 2015 Published online 17 March 2015 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/ca.22517
478
Anderson et al.
Fig. 1. The appendages of the normal heart are shown as photographed from the right (A) and left (B) lateral aspects. The morphologically right appendage (A) is triangular, and has a broad junction with the remainder of the atrial chamber (double-headed red arrow). The
morphologically left appendage (B), in contrast, is tubular and hooked, with a narrow junction with the body of the left atrium (compare size of double-headed red arrows). [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
Spentzouris discussed the clinical significance of variations in tracheobronchial tree morphology (Wooten et al., 2014). These authors indicated that an isomeric arrangement of the bronchi is a key feature of socalled heterotaxy, and rightly identified this feature as “a disorder of laterality.” Commenting on the correlation between abnormalities of the abdominal and thoracic organs, they suggested that left bronchial isomerism is found in three-fifth of patients with “polysplenia syndrome,” and right bronchial isomerism in seven-tenths of those with asplenia. The source cited in support of this statement is a chapter from a textbook devoted to pediatric cardiology (Gutgesell, 1998). Here lies the rub. As stated by Spentzouris and colleagues, heterotaxy “carries a 90–99% association with significant cyanotic cardiac defects” (Spentzouris et al., 2014). In reality, not all cardiac lesions produce cyanosis, in particular those in the variant associated with multiple spleens. Nevertheless, so-called heterotaxy harbors the most complex combinations of lesions in patients with congenitally malformed hearts. The analysis of such abnormal hearts becomes easier when it is recognized that some of their parts, such as the tracheobronchial tree, show evidence of isomerism in contrast to the expected lateralization (Jacobs et al., 2007). However, the presence of isomeric features within the heart remains controversial (Van Praagh and Van Praagh, 1990). When assessed appropriately in terms of anatomy, particularly in mice with targeted mutations, isomerism within the heart is an unequivocal finding (Uemura et al., 1995; Meno et al., 1998; Bamforth et al., 2004). However, only the atrial appendages are uniformly isomeric (Uemura et al., 1995). It is
Fig. 2. The short axes of the atrioventricular junctions of the normal heart are viewed from the atrial aspect after removal of most of the atrial walls. It can still be seen that the pectinate muscles lining the morphologically right appendage extend round the atrioventricular junction to the crux of the heart (white arrow with red borders), while the left-sided pectinate muscles are confined within the tubular appendage (double-headed red arrow). Note also the coronary sinus occupying the left atrioventricular groove. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
Cardiac Isomerism
479
Fig. 3. The scanning electron microscopic images show the changes in the structure of the developing heart tube as material grows in from the heart-forming areas. The initial migration, from the first heart field, forms little more than the primordium of the left ventricle, as shown in the linear tube in a developing mouse embryo early in E8.5, with seven to eight somites (A). Additional migra-
tion of tissues from the so-called second heart field causes elongation of the tube, as shown in the dissection of a mouse embryo later in E8.5, with 11 somites (B). The new material has produced the primordium of the atrial chamber, the ventricular loop, and the outflow tract. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
therefore unfortunate that Wooton and colleagues described polysplenia as associated with “left atrial isomerism” (Wooten et al., 2014). They described the association between asplenia syndrome and isomerism of the right atrial appendages correctly. Attention to such details helps to resolve existing controversies. More importantly, when assessed relative to known isomerism of the atrial appendages, there is an excellent correlation between this feature and bronchial isomerism (Uemura et al., 1995a,b). This correlation can serve as the starting point for sequential segmental analysis in patients with congenitally malformed hearts who have so-called heterotaxy (Jacobs et al., 2007).
ered necessary to justify the alleged “univentricular” nature of the congenitally malformed heart with a double-inlet right ventricle. Keeton and colleagues (1979) had argued that the small chamber in such hearts was no longer a ventricle because it lacked its inlet portion. Van Praagh et al. (1980), in contrast, argued correctly that it could still be identified as a ventricle since it retained its most constant component, namely its apical trabecular part. Extending these observations, van Praagh and associates (1980) suggested that any structure within the heart that was subject to congenital malformation is best defined on the basis of its most constant intrinsic component, rather than of other parts that could be variable. The logic they proposed is inescapable. Therefore, their concept has now become the basis for analyzing the abnormal chambers or components in congenitally malformed hearts. It obviously requires knowledge of the most constant components of the cardiac chambers in such settings. Experience has shown that the most constant parts of the atrial chambers are the appendages. The venous connections are frequently abnormal, while defects in the septum distort the anatomy. In the ventricles, as emphasized by Van Praagh et al. (1980), the apical trabecular components are most constant, the inlets and outlets being variably shared between the apical parts in congenitally malformed hearts (Spicer and Anderson, 2013). No intrinsic features permit distinction of the arterial trunks, but fortunately the patterns of branching of
HOW DO WE DETERMINE LATERALITY WITHIN THE HEART? The most important anatomical principle by which leftness, as opposed to rightness, is determined morphologically in the cardiac components was established some time ago by Van Praagh et al., (1980). They proposed their concept, which has become known as the “morphological method,” in a letter pointing to the inappropriate logic used by one of the current authors and his colleagues when they sought to disqualify the left ventricle from ventricular status when there was a double inlet to the right ventricle (Keeton et al., 1979). This approach had been consid-
480
Anderson et al.
Fig. 4. The scanning electron microscopic images show hearts from developing mice at E9.5 and E10.5. The image in Panel A is at E9.5. It shows the formation of the ventricular loop, with the primordia of the developing left and right ventricles (LV, RV) seen in series. Both atrial chambers (R, L), in contrast, will develop in parallel from the caudal atrial component of the heart tube. The image in Panel B, at E10.5, was prepared by sectioning along the long axis of the ventricular loop. By this stage, it can be seen how the atrial appendages have developed by ballooning in parallel (white arrows with red borders) from
the initial atrial component of the heart tube. The atrial component is now beginning to divide into its right and left atrial components as the primary atrial septum grows from the atrial root (single-headed white arrow). The ventricular apical components, in contrast, have ballooned in parallel (red arrow and blue arrow with white borders) from the inlet and outlet components of the ventricular loop. It is the ballooning that produces the primordium of the apical ventricular septum (white star with red borders). [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
the aortic, pulmonary, common and solitary arterial trunks enable them to be recognized. The question then remains, whether the differences in the most constant ventricular and atrial components are sufficiently distinct to permit their recognition when the heart is congenitally malformed. For the ventricles, the coarseness of the apical trabeculations is sufficient to fulfill the criterion of distinction (Spicer and Anderson, 2013), particularly for the morphologist, who can compare patterns of apical trabeculation in the same heart. For the atrial chambers, it was initially thought that the shape of the appendages would be sufficient to distinguish them, the morphologically right appendages being triangular with a broad junction with the remainder of the atrium, while the tubular and hooked left atrial appendage has a very narrow neck joining it to the body of the left atrium (Sharma et al., 1988—Fig. 1). In their critique of the existence of isomeric features within the heart, Van Praagh and Van Praagh (1990) nevertheless pointed out that shape and junctional size could be altered by abnormal hemodynamics, disqualifying their use as arbiters of rightness or leftness when the heart was malformed. But one feature of the appendages is not affected by hemodynamic changes. This is the extent of the pectinate muscles relative to the atrioventricular junctions. As
Uemura et al. (1995a) showed, this feature always allows the morphologically right appendage to be distinguished from its morphologically left counterpart (Fig. 2).
HOW DO THE DISTINCTIVE CHAMBER COMPONENTS DEVELOP? It is now recognized that the original linear component of the developing heart tube forms little more than the definitive left ventricle (Fig. 3A). The atrial components, along with the primordium of the right ventricle and the outflow tract, are formed as further material migrates into the pericardial cavity from the heart-forming areas (Moorman and Christoffels, 2003—Fig. 3B). When first formed, the atrial component of the heart tube (the ventricular loop) and the outflow tract have relatively smooth walls, with endocardial cushions developing within the atrioventricular canal and the outflow tract. These components are formed from the so-called primary myocardium. The chamber myocardium then develops by ballooning from the cavities of the primary tube (Moorman et al., 2007). There is a significant difference in the pattern of ballooning between the atrial component and the ventricular loop, since the appendages balloon in
Cardiac Isomerism
Fig. 5. The scanning electron microscopic image shows the atrial chambers from a mouse in which the c isoform of the Pitx2 gene has been knocked out. The atrial chambers have been removed from the heart and are imaged from the ventricular aspect. Both the appendages have right morphology, and bilaterally symmetrical venous valves have formed to guard the entrances of the systemic venous tributaries. The pulmonary vein opens in a midline position. The image shows isomerism of the morphologically right appendages, along with—in the mouse—isomerism of the septum spurium and the venous valves. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
parallel while the ventricular apical components balloon in series from the inlet and outlet parts of the loop (Fig. 4). These differences in ballooning mean that the two halves of the dividing primary atrial component of the heart tube are influenced in different ways by the genes responsible for producing morphological leftness as opposed to rightness in the overall embryo. It is also well established that a cascade of genes, including nodal and Pitx2, is responsible for producing the morphologically left characteristics (Bamforth et al., 2004). Within the heart, Pitx2 expression is restricted to the morphologically left atrium, along with a small part of the right atrium extending to the location of the left venous valve. Within the ventricles, in contrast, the genes are expressed throughout the ventricular mass, with no differentiation between the developing right and left ventricular components. When Pitx2c is knocked-out in genetically modified mice (Liu et al., 2001), the mice develop with appendages bilaterally of right morphology, the two appendages in the same mouse being isomeric, or mirrorimages of each other (Fig. 5). There is also bilateral formation of the venous valves, along with the septum spurium (Fig. 5), but these features are not usually found in clinical situations (see later). The processes responsible for producing the left– right body axis originate in the node, a bilaminar
481
structure at the cranial end of the primitive streak in early embryos. The information is then transferred to the left lateral plate mesoderm, in which there is local induction of Nodal. Nodal expression then spreads through the entirety of the left lateral plate mesoderm along the cranio-caudal axis of the developing embryo (Meno et al., 1998). Nodal signaling not only induces its own expression, thus expanding its domain of expression, but also induces Pitx2 expression whereby the visceral organs acquire their consistent left–right asymmetry. In the developing heart, Nodal influences the direction of cardiac looping and induces the asymmetric expression of Pitx2. Since Nodal also has the property of self-expansion, Lefty genes, feedback inhibitors of Nodal, restrict its expression domain to the left side. When Lefty1 is knocked-out, both Nodal and Pitx2 are expressed ectopically on the right side, so the genetically modified mice have isomeric left atrial appendages within the heart (Meno et al., 1998—Fig. 6). While the atrial appendages of these mutant mice, and in many instances the patterns of systemic venous tributaries too, show evidence of bilateral symmetry and hence are isomeric in the same individual, the ventricles and the arterial trunks show no evidence of symmetry. Nevertheless, the ventricles are rarely normal, being typically associated
Fig. 6. The histological section, prepared in so-called four-chamber orientation, is from a mouse in which Lefty1 has been knocked out. Both the atrial appendages are of left morphology and there is a common atrioventricular junction (double-headed red arrow). However, the ventricles are morphologically right and left, and show righthanded topology. Similarly, Pitx2c knockout mice show no evidence of isomerism within the ventricular mass. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
482
Anderson et al.
Fig. 7. The images show the atrial chambers in the heart obtained from a patient with so-called asplenia, recognized as a subset of visceral heterotaxy. Panel A shows the opened left-sided atrioventricular junction. The leftsided atrium has a triangular appendage (star) and receives the superior caval vein, and the pectinate muscles within the appendage extend to the crux. Note that the atrium connects to a left-sided morphologically right ventricle (Morph RV). The right-sided atrial chamber (Panel B) also has a triangular appendage, and the pectinate muscles within the appendage again extend to the
crux. The right-sided atrial chamber receives the pulmonary veins in anatomically abnormal fashion. It connects to the right-sided morphologically left ventricle (Morph LV). The venoatrial connections are mirror-imaged, but the heart has isomerism of the morphologically right appendages. The atrioventricular connections are biventricular and mixed, with left-handed ventricular topology. The arrangement should not be mistaken for an overall mirror-image arrangement. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
with a common atrioventricular junction, so the junctions are not divided into the normal tricuspid and mitral valvar orifices (Fig. 6). Therefore, gene knockout evidence shows that it is possible to recognize the features of isomerism within the heart, but the isomeric features involve the atrial rather than the ventricular chambers.
the basis of “splenic syndromes” (Van Mierop et al., 1972), or “situs ambiguus” (Van Praagh, 1972). As we have emphasized, the latter term remains a popular designation (Spentzouris et al., 2014). It was therefore unfortunate that when it was suggested that congenitally malformed hearts from patients known to exhibit heterotaxy should be analyzed on the basis of bilateral symmetry within the atrial chambers, the condition was described in terms of “atrial isomerism” (Macartney et al., 1980). As Van Praagh and Van Praagh (1990) pointed out, patients with right atrial isomerism might be expected to have bilateral symmetry of both caval veins, along with dual coronary sinuses. Patients with left atrial isomerism would be expected to have eight pulmonary veins. However, as discussed earlier, van Praagh et al. had previously emphasized that structures within the heart should be defined on the basis of their most constant features, not according to components that might themselves be variable (Van Praagh et al., 1980). It was well known that abnormal venous connections were an integral part of the so-called splenic syndromes (Van Praagh, 1972; Van Mierop et al., 1972). Hence, venoatrial connections should not have been used to identify atrial anatomy in this setting. It had also been
IS THERE EVIDENCE OF CARDIAC ISOMERISM IN SO-CALLED “HETEROTAXY” CASES? It has been known for many years that there is some degree of isomerism in the syndrome originally described on the basis of splenic abnormalities. As long ago as 1962, van Mierop and Wiglesworth described the isomerism of the sinus node in asplenia (Van Mierop and Wiglesworth, 1962). Shortly thereafter, Moller and colleagues pointed to the evidence of bilateral left-sidedness in the syndrome then usually described as polysplenia (Moller et al., 1967). Despite these observations, it remained conventional to describe the cardiac findings in heterotaxy cases on
Cardiac Isomerism
483
Fig. 8. The images show the atrial chambers in the heart from a patient with polysplenia, another subset of visceral heterotaxy. Both appendages (white stars with red borders in panels A and B) are tubular and have narrow junctions with the common atrium (red double-headed arrows). [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
shown that the appendages, even when analyzed on the basis of their shape, showed evidence of isomerism in these syndromes (Sharma et al., 1988). As Uemura et al. (1995a) then showed, analysis of the appendages on the basis of the extent of the pectinate muscles permitted hearts obtained from patients with splenic syndromes to be segregated completely into groups with isomerism of either the morphologically right or morphologically left atrial appendages. Our continuing studies have confirmed the observations of Uemura and associates (1995a): only the appendages are truly isomeric (Figs. 7–9). The objections raised by Van Praagh and Van Praagh (1990) to the concept of “atrial isomerism” are therefore readily defused by simply describing isomerism of the atrial appendages.
ONLY THE APPENDAGES ARE TRULY ISOMERIC The introduction of the “segmental approach” by Van Praagh (1972) revolutionized the diagnosis of patients with congenitally malformed hearts. The introduction of the morphological method was equally important in refining the criteria for distinguishing the components of the cardiac segments (Van Praagh et al., 1980). It is therefore surprising that Van
Praagh and Van Praagh (1990) argued against the concept of isomerism within the heart. As we have emphasized, the morphological method shows that the venoatrial connections cannot serve to distinguish the morphologically right from the morphologically left atrial chambers. This point is especially important since the venoatrial connections themselves can be essentially normal or mirror-imaged in the presence of either isomeric right or left atrial appendages (Figs. 7–9). In many other instances, if not most, the venoatrial connections are grossly abnormal when the appendages are isomeric. Interruption of the inferior caval vein, with continuation through the azygos venous system, is a frequent finding when there are isomeric left atrial appendages. The pulmonary venous connections are always abnormal in the presence of isomeric right atrial appendages, even when the veins connect to one or other of the atrial chambers (Fig. 9). However, in half of hearts from patients with isomeric right atrial appendages, the pulmonary veins connect anomalously to an extracardiac site. Bilateral superior caval veins connecting to the atrial roofs are also frequent in the setting of right isomerism, absence of the coronary sinus being a uniform finding. In contrast, when there are bilateral caval veins in left isomerism, one of them usually drains to the opposite atrial chamber through a coronary sinus (Fig. 10).
Fig. 9. The chambers shown in Figure 8 have been opened to reveal the morphology of the atrioventricular junctions. Both appendages have narrow necks connecting them to the cavities of the right-sided and left-sided atrial chambers (stars). Both vestibules are smooth. The systemic veins drain to the right-sided atrial chamber while the pulmonary veins drain to the left-sided chamber. Therefore, the venoatrial connections are hemodynamically and physiologically normal, and there is righthanded ventricular topology, with panel A showing how
the right-sided atrium connects to the morphologically right ventricle (Morph RV) while the left-sided atrium connects to the morphologically left ventricle (Morph LV). Nonetheless, the heart lacks the usual atrial arrangement and concordant atrioventricular connections. Instead, there are isomeric left atrial appendages, with mixed atrioventricular connections, and right-handed ventricular topology. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
Fig. 10. As shown by the frontal view of the heart (Panel A), which was located in the right chest with its apex pointing to the right, both the atrial appendages are narrow and tubular, with narrow junctions to the bodies of the right- and left-sided atrial chambers. There is isomerism of the left atrial appendages. There are bilateral superior caval veins (SCV). The posterior view of the heart (Panel B) shows that the pulmonary veins connect to the right-sided atrial chamber, while the right superior caval
vein extends through the right-sided atrioventricular groove (white arrows with red borders) and opens to the left-sided atrium through a coronary sinus. The left-sided superior caval vein and the hepatic veins also connect to the left-sided atrium. The left-sided atrium also receives the venous return from the abdomen via the azygos venous system, which drains to the left superior caval vein. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
Cardiac Isomerism All these features require description. The recognition of isomeric atrial appendages is but the starting point for sequential segmental description of the congenital cardiac malformations that are so frequent in so-called visceral heterotaxy. Furthermore, in hearts with isomeric appendages, the atrioventricular connections can never be concordant or discordant. Even when there are biventricular atrioventricular connections, half the heart will be concordantly connected and the other half discordantly connected. These features are clear in Figs. 7 and 9, demonstrating that the ventricular mass itself in these settings can show either right-handed (Fig. 9) or left-handed (Fig. 7) topology. These findings confirm that the ventricles do not show features of isomerism. In the clinical setting, particularly when the morphologically right appendages are isomeric, there is often a double-inlet ventricle, but even then an incomplete second ventricle is usually present. Nonetheless, solitary and indeterminate ventricles, themselves exceedingly rare, are more frequent when there is right isomerism (Smith et al., 2006). Any suggestion of ambiguity is removed by simply describing the various malformations. Such descriptions are also needed to account for arrangements at the ventriculo-arterial junctions. This is particularly important when the ventriculo-arterial connections are discordant, since this can give the impression of congenitally-corrected transposition when it is associated with left-handed ventricular topology. Similarities with congenitally-corrected transposition will also appear when there is mirror-imagery of the venous returns, but with right-handed ventricular topology and discordant ventriculo-arterial connections. Isomeric atrial appendages are also potentially confusing if accompanied by normally-arranged venous returns, left-handed ventricular topology, and mirrorimaged spiraling of concordantly-connected arterial trunks. This combination will produce the hemodynamic picture of transposition, but the key to understanding the anatomical arrangement is recognition of the isomeric atrial appendages, since these determine the location of the conduction tissues (Smith et al., 2006). Sequential description of the anomalies present will therefore always clarify the anatomical arrangement. The same applies to arrangements of other systems of organs. The lack of lateralization involves all thoracic and abdominal organs. As shown in the recent review, isomerism of the bronchial tree is readily recognizable (Wooten et al., 2104). This feature is perhaps the best guide to the likelihood of isomerism of the atrial appendages within the heart (Uemura et al., 1995a). Nevertheless, analysis of the findings in heterotaxy cases can produce significantly different results when approached from the basis of splenic abnormalities rather than the presence of isomeric atrial appendages (Uemura et al., 1995b). Absence of the coronary sinus, for example, is a uniform finding when there are isomeric right atrial appendages, but sometimes occurs when the spleen is absent. The latter association emphasizes that not all patients with absence of the spleen, or “asplenia,” have isomeric right atrial appendages. This does not indicate that
485
they have “situs ambiguous.” On the contrary, as with analysis of the heart itself, ambiguity is removed by simply describing the findings in each system of organs specifically. From the viewpoint of the heart, analysis should begin with establishing the arrangement of the atrial appendages. In this respect there are only four possibilities: the usual arrangement, its mirror-image variant, and the arrangements with isomeric right or left appendages.
ACKNOWLEDGMENTS The authors are indebted to the families of patients who did not survive their birth with congenitally malformed hearts for their philanthropy in granting permission for the hearts to be archived for ongoing research and educational purposes.
REFERENCES Bamforth SD, Braganc ¸a J, Farthing CR, Schneider JE, Broadbent C, Michell AC, Clarke K, Neubauer S, Norris D, Brown NA, Anderson RH, Bhattacharya S. 2004. Cited2 controls left-right patterning and heart development through a Nodal-Pitx2c pathway. Nat Genet 36:1189–1196. Gutgesell HP. 1998. Cardiac malposition and heterotaxy. In: Garson AJ, Bricker JT, Fisher DJ, Neish SR, editors. The Science and Practice of Pediatric Cardiology. Vol. 2. Baltimore: Williams and Wilkins. p 1539–1563. Jacobs JP, Anderson RH, Weinberg PM, Walters HL, Tchervenkov CI, land MJ, Colan SD, Del Duca D, Franklin RCG, Aiello VD, Be Gaynor JW, Krogmann ON, Kurosawa H, Maruszewski B, Stellin G, Elliott MJ. 2007. The nomenclature, definition and classification of cardiac structures in the setting of heterotaxy. Cardiol Young 17:S4 1–28. Keeton BR, Macartney FJ, Hunter S, Mortera C, Rees P, Shinebourne EA, Tynan MJ, Wilkinson JL, Anderson RH. 1979. Univentricular heart of right ventricular type with double or common inlet. Circulation 59:403–411. Liu C, Liu W, Lu M-F, Brown NA, Martin JF. 2001. Regulation of leftright asymmetry by thresholds of Pitx2c activity. Development 128:2039–2048. Macartney FJ, Zuberbuhler JR, Anderson RH. 1980. Morphological considerations pertaining to recognition of atrial isomerism. Consequences for sequential chamber localisation. Br Heart J 44: 657–667. Meno C, Shimono A, Saijoh Y, Yashiro K, Mochida K, Ohishi S, Noji S, Kondoh S, Hamada H. 1998. Lefty-1 is required for left-right determination as a regulator of lefty-2 and nodal. Cell 94:287– 297. Moller JH, Nakib A, Anderson RC, Edwards JE. 1967. Congenital cardiac disease associated with polysplenia: A developmental complex of bilateral ’left-sidedness’. Circulation 36:789–799. Moorman AFM and Christoffels VM. 2003. Cardiac chamber formation: Development, genes, and evolution. Physiol Rev 83:1223– 1267. Moorman AFM, Christoffels VM, Anderson RH, van den Hoff MJB. 2007. The heart-forming fields—One or multiple? Phil Trans R Soc B 362;1257–1265. Sharma S, Devine W, Anderson RH, Zuberbuhler JR. 1988. The determination of atrial arrangement by examination of appendage morphology in 1842 heart specimens. Br Heart J 60:227– 231. Smith A, Ho SY, Anderson RH, Connell MG, Arnold R, Wilkinsion JL. 2006. The diverse cardiac morphology seen in hearts with isomerism of the atrial appendages with reference to the disposition
486
Anderson et al.
of the specialized conduction system. Cardiol Young 16:437– 454. Spentzouris G, Zandian A, Cesmebasi A, Kinsella CR, Muhleman M, Mirzayan N, Shirak M, Tubbs RS, Shaffer K, Loukas M. 2014. The clinical anatomy of the inferior vena cava. A review of common congenital anomalies and considerations for clinicians. Clin Anat 27:1234–1243. Spicer DE, Anderson RH. 2013. Methodological review of ventricular anatomy—The basis for understanding congenital cardiac malformations. J Cardiovasc Transl Res 6:145–154. Uemura H, Ho SY, Devine WA, Kilpatrick LL, 1995a, Anderson RH. Atrial appendages and venoatrial connections in hearts from patients with visceral hetertotaxy. Ann Thor Surg 60:561–569. Uemura H, Ho SY, Devine WA, Anderson RH. 1995b. Analysis of visceral heterotaxy according to splenic status, appendage morphology, or both. Am J Cardiol 76:846–849. Van Mierop LHS, Wiglesworth FW. 1962. Isomerism of the atria in the asplenia syndrome. Lab Invest 11:1303–1315.
Van Mierop LHS, Gessner IH, Schiebler GL. 1972. Asplenia and polysplenia syndromes. Birth Defects: Original Article Series 8:36– 52. Van Praagh R. 1972. The segmental approach to diagnosis in congenital heart disease. In: Bergsma D, editors. Birth Defects Original Article Series, Vol. VIII, No. 5. The National Foundation—March of Dimes. Baltimore: Williams and Wilkins. p 4–23. Van Praagh R, Van Praagh S. 1990. Atrial isomerism in the heterotaxy syndromes with asplenia, or polysplenia, or normally formed spleen: An erroneous concept. Am J Cardiol 66:1504– 1506. Van Praagh R, David I, Wright GB, Van Praagh S. 1980. Large RV plus small LV is not single RV. Circulation 61:1057–1058. Wooten C, Patel S, Cassidy L. Watanabe K, Matusz P, Tubbs RS, Loukas M. 2014. Variations of the tracheobronchial tree: Anatomical and clinical significance. Clin Anat 27:1223–1233.
REVIEW
The Importance of Being Isomeric ROBERT H. ANDERSON,1* NIGEL A. BROWN,2 CHIKARA MENO,3 1
AND
DIANE E. SPICER4,5
Institute of Genetic Medicine, Newcastle University, Newcastle upon Tyne, United Kingdom 2 Division of Biomedical Sciences, St George’s University, London, United Kingdom 3 Graduate School of Medical Sciences, Kyushu University, Fukuoka, Kyushu, Japan 4 Department of Pediatric Cardiology, University of Florida, Gainesville, Florida 5 Congenital Heart Institute of Florida, St. Petersburg, Florida
In the normal individual, the parietal components of the body are mirrorimaged and appropriately described as isomeric. The thoraco-abdominal organs, in contrast, are lateralized. However, in “visceral heterotaxy,” the thoraco-abdominal organs also show some degree of isomerism, best seen in the arrangement of the bronchial tree. Whether isomerism can be found within the heart remains controversial. One of two recent publications in this journal emphasized the crucial features of bronchial isomerism; the other, in contrast, confused the situation of isomerism within the heart. In this review, we show how the topic of cardiac isomerism is clarified by concentrating on the anatomical features of the cardiac components and determining how best they can be described. Appropriate manipulation of developing mice produces unequivocal evidence of isomerism of the atrial appendages, but with no evidence of ventricular isomerism. In hearts from patients with so-called “heterotaxy,” only the atrial appendages, distinguished on the basis of the pectinate muscles lining their walls, are uniformly isomeric, permitting the syndrome to be differentiated into the subsets of left as opposed to right atrial appendage isomerism. Thus, controversies are defused by simply describing the isomerism of the atrial appendages rather than “atrial isomerism,” recognizing the frequency of abnormal venoatrial connections in these settings. Any suggestion of ambiguity is removed by the equally simple expedient of describing all the variable cardiac features, describing the arrangements of the thoracic and abdominal organs separately should there be discordances. Clin. Anat. 28:477–486, 2015. VC 2015 Wiley Periodicals, Inc. Key words: visceral heterotaxy; situs solitus; left isomerism; right isomerism; morphological method
INTRODUCTION In a review in a recent issue of this journal, Spentzouris and colleagues discussed the anatomy of the inferior caval vein and included the abnormalities found in so-called “situs ambiguus” (Spentzouris et al., 2014). They suggested that this syndrome, also known as visceral heterotaxy, “is a severe variant of situs inversus which involves a complete breakdown in the classical asymmetry found in the abdominal viscera.” It is questionable whether the constellation of abnormal findings is a “severe variant of situs inversus.” Rather, the “breakdown in classical asymmetry” is of greatest significance, but this involves not only
C V
2015 Wiley Periodicals, Inc.
the abdominal but also the thoracic organs (Jacobs et al., 2007). Indeed, in the article immediately preceding the review relating to the inferior caval vein, in the same issue of the journal, another group of authors including two of those who collaborated with
*Correspondence to: Robert H. Anderson, Institute of Genetic Medicine, Newcastle University, Newcastle upon Tyne, Tyne and Wear NE1 7RU, United Kingdom. E-mail: [email protected] Received 14 January 2015; Accepted 18 January 2015 Published online 17 March 2015 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/ca.22517
478
Anderson et al.
Fig. 1. The appendages of the normal heart are shown as photographed from the right (A) and left (B) lateral aspects. The morphologically right appendage (A) is triangular, and has a broad junction with the remainder of the atrial chamber (double-headed red arrow). The
morphologically left appendage (B), in contrast, is tubular and hooked, with a narrow junction with the body of the left atrium (compare size of double-headed red arrows). [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
Spentzouris discussed the clinical significance of variations in tracheobronchial tree morphology (Wooten et al., 2014). These authors indicated that an isomeric arrangement of the bronchi is a key feature of socalled heterotaxy, and rightly identified this feature as “a disorder of laterality.” Commenting on the correlation between abnormalities of the abdominal and thoracic organs, they suggested that left bronchial isomerism is found in three-fifth of patients with “polysplenia syndrome,” and right bronchial isomerism in seven-tenths of those with asplenia. The source cited in support of this statement is a chapter from a textbook devoted to pediatric cardiology (Gutgesell, 1998). Here lies the rub. As stated by Spentzouris and colleagues, heterotaxy “carries a 90–99% association with significant cyanotic cardiac defects” (Spentzouris et al., 2014). In reality, not all cardiac lesions produce cyanosis, in particular those in the variant associated with multiple spleens. Nevertheless, so-called heterotaxy harbors the most complex combinations of lesions in patients with congenitally malformed hearts. The analysis of such abnormal hearts becomes easier when it is recognized that some of their parts, such as the tracheobronchial tree, show evidence of isomerism in contrast to the expected lateralization (Jacobs et al., 2007). However, the presence of isomeric features within the heart remains controversial (Van Praagh and Van Praagh, 1990). When assessed appropriately in terms of anatomy, particularly in mice with targeted mutations, isomerism within the heart is an unequivocal finding (Uemura et al., 1995; Meno et al., 1998; Bamforth et al., 2004). However, only the atrial appendages are uniformly isomeric (Uemura et al., 1995). It is
Fig. 2. The short axes of the atrioventricular junctions of the normal heart are viewed from the atrial aspect after removal of most of the atrial walls. It can still be seen that the pectinate muscles lining the morphologically right appendage extend round the atrioventricular junction to the crux of the heart (white arrow with red borders), while the left-sided pectinate muscles are confined within the tubular appendage (double-headed red arrow). Note also the coronary sinus occupying the left atrioventricular groove. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
Cardiac Isomerism
479
Fig. 3. The scanning electron microscopic images show the changes in the structure of the developing heart tube as material grows in from the heart-forming areas. The initial migration, from the first heart field, forms little more than the primordium of the left ventricle, as shown in the linear tube in a developing mouse embryo early in E8.5, with seven to eight somites (A). Additional migra-
tion of tissues from the so-called second heart field causes elongation of the tube, as shown in the dissection of a mouse embryo later in E8.5, with 11 somites (B). The new material has produced the primordium of the atrial chamber, the ventricular loop, and the outflow tract. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
therefore unfortunate that Wooton and colleagues described polysplenia as associated with “left atrial isomerism” (Wooten et al., 2014). They described the association between asplenia syndrome and isomerism of the right atrial appendages correctly. Attention to such details helps to resolve existing controversies. More importantly, when assessed relative to known isomerism of the atrial appendages, there is an excellent correlation between this feature and bronchial isomerism (Uemura et al., 1995a,b). This correlation can serve as the starting point for sequential segmental analysis in patients with congenitally malformed hearts who have so-called heterotaxy (Jacobs et al., 2007).
ered necessary to justify the alleged “univentricular” nature of the congenitally malformed heart with a double-inlet right ventricle. Keeton and colleagues (1979) had argued that the small chamber in such hearts was no longer a ventricle because it lacked its inlet portion. Van Praagh et al. (1980), in contrast, argued correctly that it could still be identified as a ventricle since it retained its most constant component, namely its apical trabecular part. Extending these observations, van Praagh and associates (1980) suggested that any structure within the heart that was subject to congenital malformation is best defined on the basis of its most constant intrinsic component, rather than of other parts that could be variable. The logic they proposed is inescapable. Therefore, their concept has now become the basis for analyzing the abnormal chambers or components in congenitally malformed hearts. It obviously requires knowledge of the most constant components of the cardiac chambers in such settings. Experience has shown that the most constant parts of the atrial chambers are the appendages. The venous connections are frequently abnormal, while defects in the septum distort the anatomy. In the ventricles, as emphasized by Van Praagh et al. (1980), the apical trabecular components are most constant, the inlets and outlets being variably shared between the apical parts in congenitally malformed hearts (Spicer and Anderson, 2013). No intrinsic features permit distinction of the arterial trunks, but fortunately the patterns of branching of
HOW DO WE DETERMINE LATERALITY WITHIN THE HEART? The most important anatomical principle by which leftness, as opposed to rightness, is determined morphologically in the cardiac components was established some time ago by Van Praagh et al., (1980). They proposed their concept, which has become known as the “morphological method,” in a letter pointing to the inappropriate logic used by one of the current authors and his colleagues when they sought to disqualify the left ventricle from ventricular status when there was a double inlet to the right ventricle (Keeton et al., 1979). This approach had been consid-
480
Anderson et al.
Fig. 4. The scanning electron microscopic images show hearts from developing mice at E9.5 and E10.5. The image in Panel A is at E9.5. It shows the formation of the ventricular loop, with the primordia of the developing left and right ventricles (LV, RV) seen in series. Both atrial chambers (R, L), in contrast, will develop in parallel from the caudal atrial component of the heart tube. The image in Panel B, at E10.5, was prepared by sectioning along the long axis of the ventricular loop. By this stage, it can be seen how the atrial appendages have developed by ballooning in parallel (white arrows with red borders) from
the initial atrial component of the heart tube. The atrial component is now beginning to divide into its right and left atrial components as the primary atrial septum grows from the atrial root (single-headed white arrow). The ventricular apical components, in contrast, have ballooned in parallel (red arrow and blue arrow with white borders) from the inlet and outlet components of the ventricular loop. It is the ballooning that produces the primordium of the apical ventricular septum (white star with red borders). [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
the aortic, pulmonary, common and solitary arterial trunks enable them to be recognized. The question then remains, whether the differences in the most constant ventricular and atrial components are sufficiently distinct to permit their recognition when the heart is congenitally malformed. For the ventricles, the coarseness of the apical trabeculations is sufficient to fulfill the criterion of distinction (Spicer and Anderson, 2013), particularly for the morphologist, who can compare patterns of apical trabeculation in the same heart. For the atrial chambers, it was initially thought that the shape of the appendages would be sufficient to distinguish them, the morphologically right appendages being triangular with a broad junction with the remainder of the atrium, while the tubular and hooked left atrial appendage has a very narrow neck joining it to the body of the left atrium (Sharma et al., 1988—Fig. 1). In their critique of the existence of isomeric features within the heart, Van Praagh and Van Praagh (1990) nevertheless pointed out that shape and junctional size could be altered by abnormal hemodynamics, disqualifying their use as arbiters of rightness or leftness when the heart was malformed. But one feature of the appendages is not affected by hemodynamic changes. This is the extent of the pectinate muscles relative to the atrioventricular junctions. As
Uemura et al. (1995a) showed, this feature always allows the morphologically right appendage to be distinguished from its morphologically left counterpart (Fig. 2).
HOW DO THE DISTINCTIVE CHAMBER COMPONENTS DEVELOP? It is now recognized that the original linear component of the developing heart tube forms little more than the definitive left ventricle (Fig. 3A). The atrial components, along with the primordium of the right ventricle and the outflow tract, are formed as further material migrates into the pericardial cavity from the heart-forming areas (Moorman and Christoffels, 2003—Fig. 3B). When first formed, the atrial component of the heart tube (the ventricular loop) and the outflow tract have relatively smooth walls, with endocardial cushions developing within the atrioventricular canal and the outflow tract. These components are formed from the so-called primary myocardium. The chamber myocardium then develops by ballooning from the cavities of the primary tube (Moorman et al., 2007). There is a significant difference in the pattern of ballooning between the atrial component and the ventricular loop, since the appendages balloon in
Cardiac Isomerism
Fig. 5. The scanning electron microscopic image shows the atrial chambers from a mouse in which the c isoform of the Pitx2 gene has been knocked out. The atrial chambers have been removed from the heart and are imaged from the ventricular aspect. Both the appendages have right morphology, and bilaterally symmetrical venous valves have formed to guard the entrances of the systemic venous tributaries. The pulmonary vein opens in a midline position. The image shows isomerism of the morphologically right appendages, along with—in the mouse—isomerism of the septum spurium and the venous valves. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
parallel while the ventricular apical components balloon in series from the inlet and outlet parts of the loop (Fig. 4). These differences in ballooning mean that the two halves of the dividing primary atrial component of the heart tube are influenced in different ways by the genes responsible for producing morphological leftness as opposed to rightness in the overall embryo. It is also well established that a cascade of genes, including nodal and Pitx2, is responsible for producing the morphologically left characteristics (Bamforth et al., 2004). Within the heart, Pitx2 expression is restricted to the morphologically left atrium, along with a small part of the right atrium extending to the location of the left venous valve. Within the ventricles, in contrast, the genes are expressed throughout the ventricular mass, with no differentiation between the developing right and left ventricular components. When Pitx2c is knocked-out in genetically modified mice (Liu et al., 2001), the mice develop with appendages bilaterally of right morphology, the two appendages in the same mouse being isomeric, or mirrorimages of each other (Fig. 5). There is also bilateral formation of the venous valves, along with the septum spurium (Fig. 5), but these features are not usually found in clinical situations (see later). The processes responsible for producing the left– right body axis originate in the node, a bilaminar
481
structure at the cranial end of the primitive streak in early embryos. The information is then transferred to the left lateral plate mesoderm, in which there is local induction of Nodal. Nodal expression then spreads through the entirety of the left lateral plate mesoderm along the cranio-caudal axis of the developing embryo (Meno et al., 1998). Nodal signaling not only induces its own expression, thus expanding its domain of expression, but also induces Pitx2 expression whereby the visceral organs acquire their consistent left–right asymmetry. In the developing heart, Nodal influences the direction of cardiac looping and induces the asymmetric expression of Pitx2. Since Nodal also has the property of self-expansion, Lefty genes, feedback inhibitors of Nodal, restrict its expression domain to the left side. When Lefty1 is knocked-out, both Nodal and Pitx2 are expressed ectopically on the right side, so the genetically modified mice have isomeric left atrial appendages within the heart (Meno et al., 1998—Fig. 6). While the atrial appendages of these mutant mice, and in many instances the patterns of systemic venous tributaries too, show evidence of bilateral symmetry and hence are isomeric in the same individual, the ventricles and the arterial trunks show no evidence of symmetry. Nevertheless, the ventricles are rarely normal, being typically associated
Fig. 6. The histological section, prepared in so-called four-chamber orientation, is from a mouse in which Lefty1 has been knocked out. Both the atrial appendages are of left morphology and there is a common atrioventricular junction (double-headed red arrow). However, the ventricles are morphologically right and left, and show righthanded topology. Similarly, Pitx2c knockout mice show no evidence of isomerism within the ventricular mass. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
482
Anderson et al.
Fig. 7. The images show the atrial chambers in the heart obtained from a patient with so-called asplenia, recognized as a subset of visceral heterotaxy. Panel A shows the opened left-sided atrioventricular junction. The leftsided atrium has a triangular appendage (star) and receives the superior caval vein, and the pectinate muscles within the appendage extend to the crux. Note that the atrium connects to a left-sided morphologically right ventricle (Morph RV). The right-sided atrial chamber (Panel B) also has a triangular appendage, and the pectinate muscles within the appendage again extend to the
crux. The right-sided atrial chamber receives the pulmonary veins in anatomically abnormal fashion. It connects to the right-sided morphologically left ventricle (Morph LV). The venoatrial connections are mirror-imaged, but the heart has isomerism of the morphologically right appendages. The atrioventricular connections are biventricular and mixed, with left-handed ventricular topology. The arrangement should not be mistaken for an overall mirror-image arrangement. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
with a common atrioventricular junction, so the junctions are not divided into the normal tricuspid and mitral valvar orifices (Fig. 6). Therefore, gene knockout evidence shows that it is possible to recognize the features of isomerism within the heart, but the isomeric features involve the atrial rather than the ventricular chambers.
the basis of “splenic syndromes” (Van Mierop et al., 1972), or “situs ambiguus” (Van Praagh, 1972). As we have emphasized, the latter term remains a popular designation (Spentzouris et al., 2014). It was therefore unfortunate that when it was suggested that congenitally malformed hearts from patients known to exhibit heterotaxy should be analyzed on the basis of bilateral symmetry within the atrial chambers, the condition was described in terms of “atrial isomerism” (Macartney et al., 1980). As Van Praagh and Van Praagh (1990) pointed out, patients with right atrial isomerism might be expected to have bilateral symmetry of both caval veins, along with dual coronary sinuses. Patients with left atrial isomerism would be expected to have eight pulmonary veins. However, as discussed earlier, van Praagh et al. had previously emphasized that structures within the heart should be defined on the basis of their most constant features, not according to components that might themselves be variable (Van Praagh et al., 1980). It was well known that abnormal venous connections were an integral part of the so-called splenic syndromes (Van Praagh, 1972; Van Mierop et al., 1972). Hence, venoatrial connections should not have been used to identify atrial anatomy in this setting. It had also been
IS THERE EVIDENCE OF CARDIAC ISOMERISM IN SO-CALLED “HETEROTAXY” CASES? It has been known for many years that there is some degree of isomerism in the syndrome originally described on the basis of splenic abnormalities. As long ago as 1962, van Mierop and Wiglesworth described the isomerism of the sinus node in asplenia (Van Mierop and Wiglesworth, 1962). Shortly thereafter, Moller and colleagues pointed to the evidence of bilateral left-sidedness in the syndrome then usually described as polysplenia (Moller et al., 1967). Despite these observations, it remained conventional to describe the cardiac findings in heterotaxy cases on
Cardiac Isomerism
483
Fig. 8. The images show the atrial chambers in the heart from a patient with polysplenia, another subset of visceral heterotaxy. Both appendages (white stars with red borders in panels A and B) are tubular and have narrow junctions with the common atrium (red double-headed arrows). [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
shown that the appendages, even when analyzed on the basis of their shape, showed evidence of isomerism in these syndromes (Sharma et al., 1988). As Uemura et al. (1995a) then showed, analysis of the appendages on the basis of the extent of the pectinate muscles permitted hearts obtained from patients with splenic syndromes to be segregated completely into groups with isomerism of either the morphologically right or morphologically left atrial appendages. Our continuing studies have confirmed the observations of Uemura and associates (1995a): only the appendages are truly isomeric (Figs. 7–9). The objections raised by Van Praagh and Van Praagh (1990) to the concept of “atrial isomerism” are therefore readily defused by simply describing isomerism of the atrial appendages.
ONLY THE APPENDAGES ARE TRULY ISOMERIC The introduction of the “segmental approach” by Van Praagh (1972) revolutionized the diagnosis of patients with congenitally malformed hearts. The introduction of the morphological method was equally important in refining the criteria for distinguishing the components of the cardiac segments (Van Praagh et al., 1980). It is therefore surprising that Van
Praagh and Van Praagh (1990) argued against the concept of isomerism within the heart. As we have emphasized, the morphological method shows that the venoatrial connections cannot serve to distinguish the morphologically right from the morphologically left atrial chambers. This point is especially important since the venoatrial connections themselves can be essentially normal or mirror-imaged in the presence of either isomeric right or left atrial appendages (Figs. 7–9). In many other instances, if not most, the venoatrial connections are grossly abnormal when the appendages are isomeric. Interruption of the inferior caval vein, with continuation through the azygos venous system, is a frequent finding when there are isomeric left atrial appendages. The pulmonary venous connections are always abnormal in the presence of isomeric right atrial appendages, even when the veins connect to one or other of the atrial chambers (Fig. 9). However, in half of hearts from patients with isomeric right atrial appendages, the pulmonary veins connect anomalously to an extracardiac site. Bilateral superior caval veins connecting to the atrial roofs are also frequent in the setting of right isomerism, absence of the coronary sinus being a uniform finding. In contrast, when there are bilateral caval veins in left isomerism, one of them usually drains to the opposite atrial chamber through a coronary sinus (Fig. 10).
Fig. 9. The chambers shown in Figure 8 have been opened to reveal the morphology of the atrioventricular junctions. Both appendages have narrow necks connecting them to the cavities of the right-sided and left-sided atrial chambers (stars). Both vestibules are smooth. The systemic veins drain to the right-sided atrial chamber while the pulmonary veins drain to the left-sided chamber. Therefore, the venoatrial connections are hemodynamically and physiologically normal, and there is righthanded ventricular topology, with panel A showing how
the right-sided atrium connects to the morphologically right ventricle (Morph RV) while the left-sided atrium connects to the morphologically left ventricle (Morph LV). Nonetheless, the heart lacks the usual atrial arrangement and concordant atrioventricular connections. Instead, there are isomeric left atrial appendages, with mixed atrioventricular connections, and right-handed ventricular topology. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
Fig. 10. As shown by the frontal view of the heart (Panel A), which was located in the right chest with its apex pointing to the right, both the atrial appendages are narrow and tubular, with narrow junctions to the bodies of the right- and left-sided atrial chambers. There is isomerism of the left atrial appendages. There are bilateral superior caval veins (SCV). The posterior view of the heart (Panel B) shows that the pulmonary veins connect to the right-sided atrial chamber, while the right superior caval
vein extends through the right-sided atrioventricular groove (white arrows with red borders) and opens to the left-sided atrium through a coronary sinus. The left-sided superior caval vein and the hepatic veins also connect to the left-sided atrium. The left-sided atrium also receives the venous return from the abdomen via the azygos venous system, which drains to the left superior caval vein. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
Cardiac Isomerism All these features require description. The recognition of isomeric atrial appendages is but the starting point for sequential segmental description of the congenital cardiac malformations that are so frequent in so-called visceral heterotaxy. Furthermore, in hearts with isomeric appendages, the atrioventricular connections can never be concordant or discordant. Even when there are biventricular atrioventricular connections, half the heart will be concordantly connected and the other half discordantly connected. These features are clear in Figs. 7 and 9, demonstrating that the ventricular mass itself in these settings can show either right-handed (Fig. 9) or left-handed (Fig. 7) topology. These findings confirm that the ventricles do not show features of isomerism. In the clinical setting, particularly when the morphologically right appendages are isomeric, there is often a double-inlet ventricle, but even then an incomplete second ventricle is usually present. Nonetheless, solitary and indeterminate ventricles, themselves exceedingly rare, are more frequent when there is right isomerism (Smith et al., 2006). Any suggestion of ambiguity is removed by simply describing the various malformations. Such descriptions are also needed to account for arrangements at the ventriculo-arterial junctions. This is particularly important when the ventriculo-arterial connections are discordant, since this can give the impression of congenitally-corrected transposition when it is associated with left-handed ventricular topology. Similarities with congenitally-corrected transposition will also appear when there is mirror-imagery of the venous returns, but with right-handed ventricular topology and discordant ventriculo-arterial connections. Isomeric atrial appendages are also potentially confusing if accompanied by normally-arranged venous returns, left-handed ventricular topology, and mirrorimaged spiraling of concordantly-connected arterial trunks. This combination will produce the hemodynamic picture of transposition, but the key to understanding the anatomical arrangement is recognition of the isomeric atrial appendages, since these determine the location of the conduction tissues (Smith et al., 2006). Sequential description of the anomalies present will therefore always clarify the anatomical arrangement. The same applies to arrangements of other systems of organs. The lack of lateralization involves all thoracic and abdominal organs. As shown in the recent review, isomerism of the bronchial tree is readily recognizable (Wooten et al., 2104). This feature is perhaps the best guide to the likelihood of isomerism of the atrial appendages within the heart (Uemura et al., 1995a). Nevertheless, analysis of the findings in heterotaxy cases can produce significantly different results when approached from the basis of splenic abnormalities rather than the presence of isomeric atrial appendages (Uemura et al., 1995b). Absence of the coronary sinus, for example, is a uniform finding when there are isomeric right atrial appendages, but sometimes occurs when the spleen is absent. The latter association emphasizes that not all patients with absence of the spleen, or “asplenia,” have isomeric right atrial appendages. This does not indicate that
485
they have “situs ambiguous.” On the contrary, as with analysis of the heart itself, ambiguity is removed by simply describing the findings in each system of organs specifically. From the viewpoint of the heart, analysis should begin with establishing the arrangement of the atrial appendages. In this respect there are only four possibilities: the usual arrangement, its mirror-image variant, and the arrangements with isomeric right or left appendages.
ACKNOWLEDGMENTS The authors are indebted to the families of patients who did not survive their birth with congenitally malformed hearts for their philanthropy in granting permission for the hearts to be archived for ongoing research and educational purposes.
REFERENCES Bamforth SD, Braganc ¸a J, Farthing CR, Schneider JE, Broadbent C, Michell AC, Clarke K, Neubauer S, Norris D, Brown NA, Anderson RH, Bhattacharya S. 2004. Cited2 controls left-right patterning and heart development through a Nodal-Pitx2c pathway. Nat Genet 36:1189–1196. Gutgesell HP. 1998. Cardiac malposition and heterotaxy. In: Garson AJ, Bricker JT, Fisher DJ, Neish SR, editors. The Science and Practice of Pediatric Cardiology. Vol. 2. Baltimore: Williams and Wilkins. p 1539–1563. Jacobs JP, Anderson RH, Weinberg PM, Walters HL, Tchervenkov CI, land MJ, Colan SD, Del Duca D, Franklin RCG, Aiello VD, Be Gaynor JW, Krogmann ON, Kurosawa H, Maruszewski B, Stellin G, Elliott MJ. 2007. The nomenclature, definition and classification of cardiac structures in the setting of heterotaxy. Cardiol Young 17:S4 1–28. Keeton BR, Macartney FJ, Hunter S, Mortera C, Rees P, Shinebourne EA, Tynan MJ, Wilkinson JL, Anderson RH. 1979. Univentricular heart of right ventricular type with double or common inlet. Circulation 59:403–411. Liu C, Liu W, Lu M-F, Brown NA, Martin JF. 2001. Regulation of leftright asymmetry by thresholds of Pitx2c activity. Development 128:2039–2048. Macartney FJ, Zuberbuhler JR, Anderson RH. 1980. Morphological considerations pertaining to recognition of atrial isomerism. Consequences for sequential chamber localisation. Br Heart J 44: 657–667. Meno C, Shimono A, Saijoh Y, Yashiro K, Mochida K, Ohishi S, Noji S, Kondoh S, Hamada H. 1998. Lefty-1 is required for left-right determination as a regulator of lefty-2 and nodal. Cell 94:287– 297. Moller JH, Nakib A, Anderson RC, Edwards JE. 1967. Congenital cardiac disease associated with polysplenia: A developmental complex of bilateral ’left-sidedness’. Circulation 36:789–799. Moorman AFM and Christoffels VM. 2003. Cardiac chamber formation: Development, genes, and evolution. Physiol Rev 83:1223– 1267. Moorman AFM, Christoffels VM, Anderson RH, van den Hoff MJB. 2007. The heart-forming fields—One or multiple? Phil Trans R Soc B 362;1257–1265. Sharma S, Devine W, Anderson RH, Zuberbuhler JR. 1988. The determination of atrial arrangement by examination of appendage morphology in 1842 heart specimens. Br Heart J 60:227– 231. Smith A, Ho SY, Anderson RH, Connell MG, Arnold R, Wilkinsion JL. 2006. The diverse cardiac morphology seen in hearts with isomerism of the atrial appendages with reference to the disposition
486
Anderson et al.
of the specialized conduction system. Cardiol Young 16:437– 454. Spentzouris G, Zandian A, Cesmebasi A, Kinsella CR, Muhleman M, Mirzayan N, Shirak M, Tubbs RS, Shaffer K, Loukas M. 2014. The clinical anatomy of the inferior vena cava. A review of common congenital anomalies and considerations for clinicians. Clin Anat 27:1234–1243. Spicer DE, Anderson RH. 2013. Methodological review of ventricular anatomy—The basis for understanding congenital cardiac malformations. J Cardiovasc Transl Res 6:145–154. Uemura H, Ho SY, Devine WA, Kilpatrick LL, 1995a, Anderson RH. Atrial appendages and venoatrial connections in hearts from patients with visceral hetertotaxy. Ann Thor Surg 60:561–569. Uemura H, Ho SY, Devine WA, Anderson RH. 1995b. Analysis of visceral heterotaxy according to splenic status, appendage morphology, or both. Am J Cardiol 76:846–849. Van Mierop LHS, Wiglesworth FW. 1962. Isomerism of the atria in the asplenia syndrome. Lab Invest 11:1303–1315.
Van Mierop LHS, Gessner IH, Schiebler GL. 1972. Asplenia and polysplenia syndromes. Birth Defects: Original Article Series 8:36– 52. Van Praagh R. 1972. The segmental approach to diagnosis in congenital heart disease. In: Bergsma D, editors. Birth Defects Original Article Series, Vol. VIII, No. 5. The National Foundation—March of Dimes. Baltimore: Williams and Wilkins. p 4–23. Van Praagh R, Van Praagh S. 1990. Atrial isomerism in the heterotaxy syndromes with asplenia, or polysplenia, or normally formed spleen: An erroneous concept. Am J Cardiol 66:1504– 1506. Van Praagh R, David I, Wright GB, Van Praagh S. 1980. Large RV plus small LV is not single RV. Circulation 61:1057–1058. Wooten C, Patel S, Cassidy L. Watanabe K, Matusz P, Tubbs RS, Loukas M. 2014. Variations of the tracheobronchial tree: Anatomical and clinical significance. Clin Anat 27:1223–1233.
