Ann Thorac Surg 1996;62:710-716
© 1996 The Society of Thoracic Surgeons
Original Articles: Cardiovascular
How Should We Optimally Describe Complex Congenitally Malformed Hearts?
Robert H. Anderson, MD
National Heart & Lung Institute, Imperial College School of Medicine, London, United Kingdom
Accepted for publication April 15, 1996.
 |
Abstract
|
|---|
Background. There is still no agreement on how best to describe the arrangement of complex congenitally malformed hearts. Some favor a system using short coded segmental combinations. Others favor a more descriptive approach.
Methods and Results. I have reviewed my own experience in examining congenitally malformed hearts. This shows that it is possible to select in all such hearts the most constant anatomic components to define the morphology of the atrial and ventricular chambers and the arterial trunks. These components are the appendages for the atriums, the apical trabecular regions for the ventricles, and the branching pattern for the arterial trunks. Thereafter, it is possible in all hearts to base description on the fashion in which the chambers and trunks are or are not joined together, describing separately any pertinent additional morphologic features or abnormal relationships.
Conclusions. By using this simple descriptive approach, which owes nothing to concepts of development, it is possible to give a basic description of each heart, and predict the location of the conduction tissues, even if the heart in question has never previously been encountered.
|
Introduction
|
|---|
In a recent issue of The Annals, Santini and colleagues [1] described a very rare congenitally malformed heart, which they labeled "tetralogy of Fallot {S,D,I}." During the review process for this work, I was asked to prepare an accompanying editorial comment, which I gladly did [2]. Subsequent to the publication of report and comment, however, Van Praagh [3] (one of the authors of the report), perhaps in the belief that the original description had not done justice to the case, felt compelled to respond to my comments with a further editorial outlining an analytic approach to complex congenital heart disease. To my eyes, obviously biased to some degree, this new editorial leaves as many questions unanswered as it attempted to answer. At the same time, it reveals the extent to which differences have emerged in the fashion in which the Boston group, in contrast to the European school of which I am part, now describes complex congenitally malformed hearts. As the approach and philosophy of the European school were unequivocally derived from the segmental concept initially advocated by Van Praagh [4], together with the similar method promoted by de la Cruz and her colleagues [5], and because I am often asked if there are, indeed, differences between these various systems, it seemed appropriate to respond to the more recent editorial [3]. Therefore, in this review I will outline the current approach of the European school for describing complex congenital heart disease and emphasize the differences that do exist in the approach to this important topic.
|
Recognition of Chambers
|
|---|
As Wilcox [6] indicated in his introduction to my recent Sloan lecture [7], the basic principles underscoring the philosophy of the European School are simplicity and pragmatism. Each case is treated on its merits, describing the anatomy in each of the cardiac components as it is observed. Following the initial suggestion of Van Praagh [4], the heart itself is considered to have atrial, ventricular, and arterial segments. Each segment is then known to contain two components (or sometimes to have a common arrangement of the components), these being the right and left atrial chambers, the right and left ventricular chambers, and the aortic and pulmonary trunks, respectively. Following another important principle promoted by Van Praagh, and dubbed the "morphological method" [8], the most constant parts of these components are used to identify malformed atriums as being morphologically right or left, abnormal ventricles as being of right, left, or indeterminate morphology, and malformed arterial trunks as being aortic, pulmonary, common, or solitary (Fig 1
). This suggestion of Van Praagh, namely that variable features of any structure should not be used as a criterion for its identification [8], has more than proved its worth in expunging illogical aspects of our initial approach [9]. In my experience, using this method with the heart in the hands, it has always proved possible to identify the morphologic nature of different chambers and arterial trunks in all congenitally malformed hearts thus far studied, no matter how abnormal they be. Thus, the nature of the junction of the atrial appendage with the remainder of the atrium serves to distinguish rightness and leftness in the atrial segment. It is the pattern of the apical trabeculations that distinguishes between types of ventricles. The branching patterns then serve to identify morphologically the different types of arterial trunk (Fig 1). Once the chambers and arterial trunks have been identified in this fashion, it is then an easy matter to describe any heart in terms of the way in which they are or are not joined together, accounting separately when appropriate for the pertinent specific morphology in each segment, and for the interrelationships of the components of each segment.
View larger version (31K):
[in this window]
[in a new window]
|
Fig 1. . The three segments of the heart are made up of strictly limited components. At the atrial level, as judged by the extent of the pectinate muscles (shown by the bracket lines) within the appendages, atrial chambers can be of only right or left morphology, although the venoatrial connections (not shown) can vary markedly. According to the nature of apical trabeculations, ventricles may be of right, left, or indeterminate morphology. Four types of arterial trunk can be recognized according to the pattern of branching, the solitary variant existing only in the setting of complete absence of the intrapericardial pulmonary arteries.
|
|
In this respect, I should emphasize that, within our system, the ventricular mass is considered to extend from the atrioventricular to the ventriculoarterial junctions, and is analyzed in terms of its inlet, apical trabecular, and outlet components. This is now one very important difference from the Boston approach where, as Van Praagh explained in his recent editorial [3], the ventricular outlets are considered as if they belong to the arterial segment. It is also the case that, in the analysis favored in Boston, the interrelationship between segments is considered in terms of harmony or disharmony between their topologic arrangements, rather than according to how the chambers or great arteries within each segment are or are not joined together. Concordance or discordance within segmental notation, for example, accounts for harmony between atrial and ventricular situs {S,D,*} and {I,L,*} representing concordance irrespective of whether the atriums are joined each to its own ventricle or to only one ventricle. Within the Boston system, the atrial segment is also considered to show either solitus or inversus according to the nature of the venoatrial connections, as isomerism of the atrial appendages is held not to exist [10].
In contrast, our analysis of a large number of hearts with visceral heterotaxy has shown that, when attention is directed to the appendages, these being the most constant atrial components and those, therefore, most suited for the arbitration of atrial morphology according to Van Praagh's morphological method, the atriums do show evidence of isomerism. Thus, in the setting of right isomerism, the pectinate muscles within the triangular appendages extend bilaterally around the atrioventricular junctions to meet at the crux. In the many hearts with common atrioventricular junctions in this setting, the pectinate muscles truly encircle the junction. In contrast, in left isomerism the posterior aspects of both atrial vestibules are smooth and devoid of pectinate muscles, these muscles being largely confined within the tubular appendages [11].
|
Analysis of the Atrioventricular Junctions
|
|---|
When the feature of morphology of the appendages used as the arbiter of atrial sidedness is combined with the way in which the segments of atrial myocardium are or are not joined with the ventricles across the atrioventricular junctions, we find that only a proportion of hearts, albeit the greater majority, can be described as showing concordant or discordant atrioventricular connections (Fig 2). It is not possible for hearts with isomeric appendages to have both junctional components of atrial and ventricular myocardium joined concordantly or discordantly. Instead, they must have biventricular and ambiguous connections (Fig 3), or else univentricular connections (Fig 4). The univentricular pattern of union between the atrial and ventricular myocardial masses can also be found in hearts with usually arranged or mirror-imaged atrial appendages. It is the hearts with double inlet, or absence of one atrioventricular connection, that make up this important "univentricular" group (Fig 4). Very few of these hearts in which the atrial chambers are joined across the atrioventricular junctions to only one ventricle have a solitary chamber within the ventricular mass. In this respect, therefore, very few can logically be described as being "single ventricles" or "univentricular hearts." By following the morphological method as advocated by Van Praagh and colleagues [8], nonetheless (this principle dictating that the ventricles be identified on the nature of their most constant component, namely the apical trabeculations), the majority of hearts in this group can be readily and accurately described as having one big and one small ventricle. The apical parts of the two chambers are of complementary morphologically right and left trabecular patterns, the smaller ventricle of necessity being rudimentary and incomplete when the junctional segments of atrial myocardium are joined only to the dominant ventricle (Fig 4). The European school, therefore, no longer finds it necessary to resort to illogical descriptions such as "univentricular heart" or "single ventricle" in the situations where there are two chambers within the ventricular mass. Although it remains entirely reasonable to describe such hearts as having functionally single ventricles, anatomic accuracy is achieved when an incomplete and rudimentary nature is recognized for those ventricles that are not joined directly to the atriums. In addition to restoring logic, therefore, this change has brought greater accuracy into the system used for description. Furthermore, when diagnosing the double- inlet nature of the union of the atriums and one ventricle around the atrioventricular junctions, emphasis is directed at the nature of the segments of myocardium rather than at the morphology of the valve or valves guarding the junctions (Fig 5). If the valvar leaflets are removed from a heart with double inlet atrioventricular connection, there is no way of knowing whether initially the junctions where guarded by separate right and left valves or by a common valve. This concentration on the precise nature of the unions of the myocardial segments also permits a logical approach to hearts with straddling and overriding atrioventricular valves. In the initial approach of the European school, it was necessary to deny ventricular status to a chamber connected to less than half of an atrioventricular orifice in the setting of an overriding valve. Then, suddenly, when the degree of override was such that more than half of the overriding segment was joined to the chamber, it became a ventricle, although the nature of its apical component had not changed within this spectrum [9]. The lack of logic in this approach was rightly emphasized by Brandt [12]. Now, by appropriate application of the morphological method, the chambers are described as ventricles throughout this sequence of overriding, but are recognized as being more rudimentary and incomplete in the setting of double inlet as opposed to concordant or discordant atrioventricular connections. Careful analysis of hearts with overriding atrioventricular valves also reveals one further group requiring special attention. This is the group in which one atrioventricular connection is absent ("valvar atresia"), but the solitary atrioventricular junction overrides so that the atrial myocardium is joined to both ventricles (Fig 6). In these hearts, which can take various forms depending on the atrial arrangement and ventricular morphology and topology, the union across the atrioventricular junction is uniatrial but biventricular (Fig 7). Both of the ventricles, sharing only one segment of atrial myocardium, are rudimentary and incomplete to greater or lesser extent.
View larger version (38K):
[in this window]
[in a new window]
|
Fig 2. . Hearts with usually arranged or mirror-imaged atrial appendages can be joined to the ventricles in either concordant or discordant fashion. (LV = left ventricle; RV = right ventricle.)
|
|
View larger version (42K):
[in this window]
[in a new window]
|
Fig 3. . When the appendages are isomeric, then irrespective of how the atrial chambers are joined each to their own ventricle the arrangement must be biventricular and ambiguous, as of necessity half of the heart will always show a concordant union whereas the other half will be joined in discordant fashion. (LV = left ventricle; RV = right ventricle.)
|
|
View larger version (46K):
[in this window]
[in a new window]
|
Fig 4. . Hearts with all patterns of arrangement of the atrial appendages (upper row) can exist with the atrial chambers joined to only one ventricle, this arrangement being produced by either double-inlet connection or absence of one atrioventricular (AV) connection (middle row). The ventricle to which the atrial chambers are joined can be a dominant left ventricle (LV), with an anterosuperior rudimentary right ventricle (RV), a dominant left ventricle with a posteroinferior rudimentary left ventricle, or a solitary and indeterminate ventricle (Ind. V) lacking a second chamber within the ventricular mass (bottom row). The rudimentary ventricles can be right- or left-sided irrespective of their morphology.
|
|
View larger version (31K):
[in this window]
[in a new window]
|
Fig 5. . Double-inlet ventricle exists when both segments of atrial myocardium (right-sided or left-sided) are joined to the same ventricle, shown in this diagram as a dominant left ventricle (LV). As shown, this arrangement can be found irrespective of whether the two junctions (right-sided and left-sided) are guarded by separate right and left atrioventricular (AV) valves (R and L) or by a common atrioventricular valve. (RV = rudimentary and incomplete right ventricle.)
|
|
View larger version (37K):
[in this window]
[in a new window]
|
Fig 6. . The heart exhibits complete absence of the left-sided atrioventricular connection, with the right atrial myocardium joined to both the left and right ventricles because of overriding of its orifice. (AV = atrioventricular; LV = left ventricle; RV = right ventricle.)
|
|
View larger version (39K):
[in this window]
[in a new window]
|
Fig 7. . The arrangement shown in Figure 6 gives a uniatrial but biventricular union between the atrial and ventricular myocardial segments. Both ventricles are incomplete and rudimentary in that they possess less than all their normal component parts. According to the specific atrial arrangement and ventricular topology in any individual case, various combinations must be anticipated with this uniatrial but biventricular junctional pattern.
|
|
|
Analysis of the Ventriculoarterial Junctions
|
|---|
A silmilar overall approach to that described for the atrioventricular junctions is adopted for analysis at the ventriculoarterial junctions. According to the way that the arterial trunks are supported by the underlying ventricles, unions across these junctions can only be concordant, discordant, double outlet from one ventricle, or else single outlet from the heart. When an arterial valve overrides a ventricular septal defect so that its leaflets are attached within two ventricles, from the stance of classification the overriding arterial trunk is considered to belong to the ventricle supporting the greater part of its circumference. The diagnoses in terms of ventriculoarterial connections are made irrespective of the nature of the subarterial ventricular infundibulums, and irrespective of the way the arterial trunks are themselves interrelated (when two trunks are present). This is not to suggest that infundibular morphology or arterial relationships are unimportant, simply that they are accounted for separately as and when appropriate. As I have already emphasized, when Van Praagh promulgated the morphological method [8], he established the principle that one variable feature should not be used as a criterion for defining another aspect of the malformed heart. To use either infundibular morphology or arterial relationships to define and describe the way that the arterial trunks are attached to the ventricular mass, therefore, would be to abrogate this important principle.
|
Is the System of Analysis of Surgical Import?
|
|---|
I believe now that the differences that have emerged over the years in the way that the European school, as compared to the Boston group, analyze structure in the congenitally malformed heart have become of major surgical importance. Not only do the extent of the pectinate muscles distinguish between the morphologically right and left atriums, they also guide the surgeon toward the location of the sinus node. All our studies thus far have indicated that, when there are isomeric right appendages ("asplenia syndrome"), then the sinus node is similarly present bilaterally [13]. In contrast, the sinus node is grossly hypoplastic and located away from major venoatrial junctions when there is left isomerism. This is why it is so important to recognize the morphology of the appendages in hearts from patients with visceral heterotaxy, in addition to describing the venous drainage [11]. Furthermore, we have seen several hearts from patients with heterotaxy in which the venous connections suggested usual or mirror-imaged drainage, yet the appendages were unequivocally isomeric. If the surgeon made assumptions based on the venoatrial connections, then not only might the site of the sinus node be misinterpreted, but cases of isomerism with left hand ventricular topology may incorrectly be diagnosed as having mirror-imagery in the setting of atrioventricular concordance. This would have potentially catastrophic consequences in terms of inferences regarding the location of the atrioventricular conduction axis. In patients with mirror-image atrial arrangement and concordant atrioventricular connections, the atrioventricular node is regularly positioned in the mirror-imaged triangle of Koch. In contrast, in patients with isomeric atrial appendages and left hand ventricular topology ("L-loop"), the atrioventricular node is either anteriorly located, or else a sling of conduction tissue encircles any accompanying ventricular septal defect.
Further crucial inferences can be made from appropriate analysis of ventricular morphology. As indicated, the European school now recognizes that hearts with a dominant left ventricle and an incomplete and rudimentary left ventricle are no longer anatomically "univentricular," although supporting a functionally univentricular circulation. A corollary of this recognition is that the septum separating the apical components of the ventricles is the remnant of the primary ventricular septum [14]. It carries, therefore, the ventricular conduction axis. When viewed from the rudimentary right ventricle, the conduction axis is known to bear a constant relationship to the ventricular septal defect, irrespective of whether the heart has double inlet or absence of the right or left-sided atrioventricular connection and irrespective of whether the rudimentary right ventricle is right or left-sided [15]. Recognition of the univentricular arrangement of the atrioventricular junctions, coupled with the precise morphology of the dominant ventricle, therefore, produces an accurate guide to the location of the atrioventricular node and conduction axis.
|
Comment
|
|---|
It is now the case that we have two markedly different approaches to the description of complex congenitally malformed hearts. As Van Praagh [3] rightly indicated, such complex disease is dominated by abnormal and unusual unions between the chambers and arterial trunks that make up the heart. Van Praagh [3] advocates a system of description that analyzes these hearts in terms of segmental mismatches. In this way, it is possible to imagine multiple and different segmental sets, and to ponder how many different variants might exist. My approach, developed over the years in collaboration with my clinical and morphological colleagues, is to produce a system that accounts for the hearts that do exist, even if they have never been seen before by the individual surgeon. This is achieved by providing simple rules for recognition of the morphology of the individual components of each segment, and then analyzing how these components are joined to their neighbors, as well as how they relate to each other. This approach does, indeed, mean that, on occasions, it is necessary to use 48 words in description where a segmental set can be constructed using three words and one group of symbols [3]. In this respect, I am reminded of the exchange purported to have occurred between Emperor Frans Joseph and Wolfgang Amadeus Mozart. "Too many notes, Mozart," the Emperor is reported to have stated. "No, Majesty, just sufficient!" was the purported reply. Another pertinent exchange is the oft-quoted one between the Mad Hatter and Alice in Alice in Wonderland. After a debate on whether words can mean whatever they are wanted to mean, the Mad Hatter sums up by commenting that the real question is which is to be the master. So it is with descriptions of malformed hearts, where surely the hearts themselves must take the magisterial role. And the practicing surgeon must decide on the system most suited for practical description. Van Praagh [3] concluded that the answers to all his questions "are unknown at the present time." In contrast, I submit that our own system, as it has now evolved from the original segmental approach, is fully capable of answering all the questions posed by the surgeon seeking to diagnose, describe, and repair any congenitally malformed heart, even if the heart in question may not have been encountered previously. At the same time, as well as providing logical description, the approach also guides the surgeon to the location of the crucially important components of the conduction system.
It is also, perhaps, pertinent to comment on the words I now use to describe the congenitally malformed hearts. In the process of review for this work, one reviewer commented "it would also help if some of the words are used correctly." The reviewer then took exception to my use of "atriums", "usual arrangement", "superior caval vein", and "oval fossa" among others. I have to confess that I find these "minor irritations" to be surprising. The beauty of the English language, and its American variant, in my eyes, is its constant commitment to change. The language has become enriched over the years by its incorporation of many foreign words, which have subsequently been subjected to English grammatical rules, rather than always being treated grammatically as if still belonging to their original language. This is exemplified by music, where the listener is now frequently treated to two or more concertos in the same program, but rarely hears a comparable number of concerti! And increasingly, sports aficionados congregate in stadiums rather than stadia! Surely the time has come for a similar relaxed attitude to be taken with the description of congenitally malformed hearts. There are then further advantages to be gained in terms of style and syntax. Flow through the superior caval vein is venous-is it correct, therefore, to describe such flow as "caval" (cava translates to hollow, and could, in any case, be said to be tautologous). American plurals are easy-simply add an "s". How many times do we see the appropriate plural of "ductus" (which is ductus, as this is a fourth declension noun). "Correctness " on use of words, surely, is in the eye of the beholder. I find it difficult to believe that "usual arrangement" can be considered harder to understand than "situs solitus," but I recognize my bias. Eventually, the words to be adopted will be decided by usage. I am prepared to bide my time.
|
Acknowledgments
|
|---|
I am indebted to all my clinical and morphological colleagues who were involved in the evolution of the system described in this review. The original members of the "European School" deserve especial credit. They were Anton E. Becker, MD, Michael J. Tynan, MD, Fergus J. Macartney, MD, Elliot A. Shinebourne, MD, and Manuel Quero-Jimenez, MD. As far as I am aware, they continue to espouse the "European " approach, but I accept full responsibility for any subtle points within this version with which they might disagree. I am particularly indebted to Dr Siew Yen Ho, who also contributed markedly over the years and, additionally, provided the illustrative material. Throughout my studies, I have been supported by the British Heart Foundation together with the Joseph Levy Foundation.
|
Footnotes
|
|---|
Address reprint requests to Dr Anderson, Department of Paediatrics, National Heart & Lung Institute, Imperial College School of Medicine, Dovehouse St, London SW3 6LY, United Kingdom.
|
References
|
|---|
- Santini F, Jonas RA, Sanders SP, Van Praagh R. Tetralogy of Fallot {S,D,I}: successful repair without a conduit. Ann Thorac Surg 1995;59:562–4.[Free Full Text]
- Anderson RH. What is meant by tetralogy of Fallot (S,D,I)? Ann Thorac Surg 1995;59:508–15.[Abstract/Free Full Text]
- Van Praagh R. Tetralogy of Fallot {S,D,I}: A recently discovered malformation and its surgical management. Ann Thorac Surg 1995;60:1163–5.[Free Full Text]
- Van Praagh R. The segmental approach to diagnosis in congenital heart disease. In: Bergsma D, ed. Birth defects: original article series, Vol VIII, No. 5. The National Foundation-March of Dimes. Baltimore: Williams and Wilkins, 1972:4–23.
- De la Cruz MV, Nadal-Ginard B. Rules for the diagnosis of visceral situs, truncoconal morphologies and ventricular inversions. Am Heart J 1972;84:19–32.[Medline]
- Wilcox BR. Introduction of Professor Anderson. Ann Thorac Surg 1995;59:1366–7.[Abstract/Free Full Text]
- Anderson RH, Wilcox BR. Understanding cardiac anatomy- the prerequisite for optimal cardiac surgery. J Thorac Cardiovasc Surg 1995;59:1366–75.
- Van Praagh R, David I, Wright GB, Van Praagh S. Large RV plus small LV is not single RV. Circulation 1980;61:1057–8.[Medline]
- Shinebourne EA, Macartney FJ, Anderson RH. Sequential chamber localization-the logical approach to diagnosis in congenital heart disease. Br Heart J 1976;38:327–40.[Abstract/Free Full Text]
- Van Praagh R, Van Praagh S. Atrial isomerism in the heterotaxy syndromes with asplenia, or polysplenia, or normally formed spleen: an erroneous concept. Am J Cardiol 1990;66:1504–6.[Medline]
- Uemura H, Ho SY, Devine WA, Kilpatrick LL, Anderson RH. Atrial appendages and venoatrial connections in hearts from patients with visceral hetertotaxy. Ann Thorac Surg 1995;60:561–9.[Abstract/Free Full Text]
- Brandt PWT. Cineangiography of atrioventricular and ventriculo-arterial connexions. In: Godman MJ, ed. Paediatric cardiology, Volume 4. Edinburgh: Churchill Livingstone, 1981:199–220.
- Ho SY, Seo J-W, Brown NA, Cook AC, Fagg NLK, Anderson RH. Morphology of the sinus node in human and mouse hearts with isomerism of the atrial appendages. Br Heart J 1995;74:437–42.[Abstract/Free Full Text]
- Lamers WH, Wessels A, Verbeek FJ, et al. New findings concerning ventricular septation in the human heart. Implications for maldevelopment. Circulation 1992;86:1194–205.[Abstract/Free Full Text]
- Cheung HC, Lincoln C, Anderson RH, et al. Options for surgical repair in hearts with univentricular atrioventricular connection and subaortic stenosis. J Thorac Cardiovasc Surg 1990;100:672–81.[Abstract]
Top
Abstract
Introduction
