| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 1995;59:1366-1375
© 1995 The Society of Thoracic Surgeons
Department of Paediatrics, National Heart & Lung Institute, London, England, and Division of Cardiothoracic Surgery, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina
Abstract
If advances in cardiac surgery are to continue into the twenty-first century, it will be necessary to concentrate on details of matters such as the anatomy of the heart. This will be achieved best when anatomy is described as it is observed. This approach is obscured when words are used in inappropriate fashion, or else assigned a function separate from their everyday meaning. Examples of how the congenitally malformed heart and the normal heart have been described in the past are examined within the framework of using words in their vernacular meaning. Suggestions are made to improve descriptions and understanding for the 21st century. Using the example of the ``univentricular heart'', it is shown how conventions debarring ventricular status to discrete chambers within the ventricular mass are, of necessity, artificial. Similar examples are used to distinguish septal from parietal structures within the heart and to elucidate the structure of some congenital malformations. For valves, it is shown how proper description requires assessment of these structures in their closed as well as their open positions. Understanding of cardiac anatomy, truly a prerequisite for successful cardiac surgery, will be facilitated in future if words are used in their generally accepted sense, and if artificial conventions are avoided.
Introduction of Professor Anderson by President Benson R. Wilcox
The 1995 Herbert Sloan lecturer is Robert H. Anderson, the Joseph P. Levy Professor of Paediatric Cardiac Morphology at the Royal Brompton Hospital and National Heart and Lung Institute in London. I am proud to say that he is also a Clinical Professor of Surgery at the University of North Carolina at Chapel Hill. Then again, it is hard to pin Bob down to any one spot, for he has been generous with his time, traveling the world over, visiting many of the great institutions represented in this audience today. Indeed, some have called him the peripatetic Pan American Professor of Pathology.
Let me tell you a little about him and his career. He was born north of London-a veritable ``Shropshire Lad'' (Fig 1
). As you can see, life was not easy in the Midlands in those days, but he overcame those earlier hardships and graduated from Wellington Grammar School in Shropshire and the University of Manchester. In 1974 he began his most illustrious career in London at the Royal Brompton Hospital.
|
I said, ``Hugh, who is the very best person in the world that if I were to take 3 to 6 months off, take my family with me, and go study congenital heart disease, who is the very best person in the world?'' And he said, without hesitation, ``Ben, there's no doubt about it, it's Dr so and so,'' fortunately I don't remember his name, ``in Omaha, Nebraska.'' And I said, ``Hugh, who is the second best?'' because although I would have loved to have gone to Omaha, I was not sure I could get my family to move out there in the wintertime. And Hugh said, ``Well, there's a young man in London who ... ,'' and I said, ``Ah, London, tell me about that.'' And so I investigated the school in London, and as a result I spent 6 months working in Bob's laboratory. And this experience formed the basis of a most pleasant friendship and a productive, professional collaboration over the ensuing years.
We are delighted that Christine Anderson is here with us today. She is literally Bob's helpmate at home and at work. Lucinda and I have been honored to have Bob and Christine in our home on occasion and to regularly bask in the warmth of their hospitality in London.
Lest any of you think our guest has attained his successes by focusing his life narrowly, let me tell you he is an accomplished musician, an oenophile of the first water, and an ardent athlete. In spite of the apparent competitive element in our long friendship, I am proud to say we are just that, good friends.
In working with Bob I have found him to be a man of many qualities. He has an apparently inexhaustible store of energy, so his capacity for work is prodigious. He is often brilliant, perhaps less often than he thinks he is. He is occasionally misguided, certainly less often than his detractors think, and as you will hear, when found in error, he is willing to reappraise his position without rancor.
But most of all he is always interesting to be around. Evidence for that is the international group he has gathered about him to form a truly modern school of anatomy. Because of its members' parallel interest in fine wines, I have labeled it the ``Bacchanalian School of Cardiac Morphology'' (Fig 2). This helps me remember what they believe in, for this school believes that any effective system for classifying congenital heart disease must be not only broad and accurate but also clear and consistent, and to be helpful, the terminology used should be unambiguous and as simple as possible. This is a system that is particularly useful to those of us interested in correcting congenital cardiac anomalies.
|
It would be presumptuous for a morphologist to suggest that cardiac surgeons do not understand the anatomy of the heart, and that is not our intent. It remains a fact, nonetheless, that some of the simplest aspects of the anatomy of the normal, the diseased, and the congenitally malformed heart remain controversial. Most of these disagreements reflect differences in definitions rather than representing discrepancies in morphologic observations, although some anatomic features do themselves remain contentious. These are the areas that will be discussed in this review, making a plea for description on the basis of anatomy as it is observed using simple and unambiguous words.
To illustrate the fact that these problems confronting the surgeon are not as trivial as they may seem, we will commence by reviewing briefly some of the trials and tribulations we have encountered in our own studies, using the example of hearts unified by having their atrial chambers connected to only one ventricle. Thereafter, we will illustrate how this descriptive morphologic method may be applied to areas of anatomy regularly encountered by the surgeon, namely, the septal components of the heart and the arterial and atrioventricular valves.
Hearts With Univentricular Atrioventricular Connections
For years, those hearts with both atriums connected through separate atrioventricular valves to the left ventricle have been considered to represent, and described as, single ventricles. Because the greater majority of such hearts also possess a second, albeit much smaller, chamber within the ventricular mass, the reasons underscoring such descriptions remain unclear. One of us argued previously for such hearts to be considered ``univentricular'', basing these arguments on speculative embryologic premises [1]. The dangers inherent in such an approach subsequently have been highlighted by investigations showing the initial developmental hypothesis to be fallacious [2]. This emphasizes the fact that descriptions based on anatomy as it is observed provide much stronger foundations, being independent of the advances currently being made in our understanding of embryologic concepts. This is the more significant because the anatomic observations made in the initial study [1] remain entirely valid, irrespective of the fact that the morphogenetic hypothesis derived from these observations proved fallacious. A major problem with the initial approach was that the concept of a ``single'' or ``univentricular'' heart was inherently illogical when there were obviously two chambers in the ventricular mass. The words chosen to describe the anatomic observations, therefore, could not withstand rigorous examination. Their use has only compounded the problems faced by the morphologist.
The object of the initial study [1] had been to show that, when examples of tricuspid atresia and double-inlet left ventricle with comparable ventriculoarterial connections were studied, the morphologic similarities were striking in the structure of the hypoplastic chamber within the ventricular mass (Fig 1). It was unfortunate that attempts to establish this similarity were made by arguing that the obvious anterior chamber in the hearts with tricuspid atresia was not a ventricle, but rather that tricuspid atresia was a univentricular heart! This was done despite the warnings of close colleagues, who could see that tricuspid atresia had a big left and a small right ventricle. The existing conventional wisdom was that double-inlet left ventricle represented a single ventricle [3–5]. In retrospect, it is abundantly clear that double-inlet left ventricle, as well as tricuspid atresia, can be described morphologically as being univentricular only if some arbitrary device is established whereby the rudimentary and incomplete anterior chamber is denied ventricular status [6]. This subterfuge was achieved initially on the basis of a lack of an inlet component in the hypoplastic chamber. Some still seek to deny this chamber its ventricular status, again basing their criteria for exclusion on embryologic speculations [7]. These speculations fly in the face of all the existing and emerging evidence that the structure separating the dominant left from the incomplete right ventricle (Fig 2) is the apical trabecular component of the muscular septum [2]. The important lessons that emerged for us from this sequence of events were (1) that it is preferable to use anatomic rather than embryologic premises as the basis for classification and (2) that description should be accomplished whenever possible using words employed in their everyday meanings. These principles now will be discussed in the context of the structure of the septal structures and valves.
|
|
Of all the parts of the heart that are least apparent to the cardiac surgeon, the septal structures are probably the most important. The surgeon is unlikely to open chambers so as to establish which parts of the wall are truly septal. Several procedures, nonetheless, require this knowledge, such as the extent of the atrial septum when performing atrial operations, or the extent of the outlet component of the ventricular septum relative to the Ross procedure. Knowledge of the septal components also clarifies the understanding of some congenital lesions, such as the sinus venosus interatrial communication, atrioventricular septal defects, and the doubly committed and juxtaarterial ventricular septal defect (the ``supracristal'' defect). To understand fully these various entities, it is essential to know the boundaries of the atrial, ventricular, and atrioventricular septal structures, but first it is necessary to establish what is meant by a septal structure. We define septums, from both surgical and morphologic viewpoints, as those parts of the cardiac walls that separate the cavities of two cardiac chambers. This is in contrast to the parietal, or outside, walls of the heart.
The Atrial Septum
On opening the right atrium, there is, at first sight, an extensive septal surface surrounding the oval fossa. Dissection shows that, in reality, much of this apparent septum is no more than the infolded parietal walls of the right and left atriums (Fig 3a). The mouth of the coronary sinus also seems to be a septal component, yet dissection shows that the sinus is also a separate venous structure that runs adjacent to, but separate from, the wall of the left atrium in the left atrioventricular groove (Fig 3b). The true atrial septum, therefore, is the floor of the oval fossa together with its immediate margins, particularly the inferior rim adjacent to the atrioventricular septum (see below). This limited extent of the septum explains several apparent paradoxes. It shows why so-called lipomas of the atrial septum are not truly septal at all, but merely abnormal growth of the fat that normally occupies the space between the infolded rims of the fossa. It explains why the so-called sinus venosus defects allow an interatrial communication but are not true septal defects. They exist because of abnormal connection of either the superior or inferior caval veins to the atriums [8, 9]. Observation of these defects reveals the fact that the caval veins are not properly aligned with the superior or inferior aspect of the atrial septum. Also, often the pulmonary veins are seen to be abnormally connected so that they drain into either the caval vein itself or the right atrium.
|
). And, because of the presence of the free-standing sleeve of subpulmonary infundibular musculature, the larger part of the right ventricular outlet beneath the attachments of the pulmonary valve is a parietal structure rather than septal (Fig 4b). It is this fact, of course, that permits the surgeon to remove the pulmonary valve in its entirety for use as an autograft in the Ross operation. It is similar support of the aortic valve by free-standing infundibular musculature in hearts with complete transposition that underscores the success of the Nikaidoh procedure [11] and its derivatives. This initial misinterpretation of the extent of the septal structures, however, in no way invalidates the categorization of ventricular septal defects as perimembranous, muscular, or doubly committed and juxtaarterial [10]. Rather, it means that perimembranous defects (those abutting directly on the central fibrous structures of the heart) are best described as opening into the inlet or outlet components of the right ventricle [12], rather than ``excavating'' the nonexistent inlet and outlet septal components. Indeed, the muscular outlet septum is easily seen in its own right only when there is malalignment between it and the remainder of the muscular ventricular septum, as is the case in tetralogy of Fallot. It is then significantly larger than in the normal heart, although even in lesions such as tetralogy of Fallot it remains the case that the pulmonary valvar leaflets are supported by a sleeve of freestanding infundibular musculature. This sleeve of musculature, in contrast, is totally lacking in those hearts that have doubly committed and juxtaarterial defects. This infundibular musculature is lacking in addition to the outlet septum, thus permitting the leaflets of the aortic and pulmonary valves to be in fibrous continuity, this feature being the anatomic hallmark of these particular lesions [12].
|
). It is deficiencies of these septal components that underscore production of the different types of atrioventricular septal defect. Very rarely, the fibrous component of the atrioventricular septum can be defective in an otherwise normally structured heart. This produces a membranous atrioventricular septal defect extending between the cavities of the right atrium and the normally positioned left ventricular outflow tract. Much more usually, however, the entire atrioventricular septal structures are lacking. This septal deficiency then produces a heart with a common atrioventricular junction. It is the combination of the common junction together with the atrioventricular septal deficiencies that produces the characteristic morphology of the so-called atrioventricular canal malformations or endocardial cushion defects. Because of the crucial role of the deficiency of the atrioventricular septums in these lesions, it is our preference to describe them as atrioventricular septal defects, but we recognize fully the significance of the common junction in producing the common atrioventricular valve with its trifoliate left ventricular component (Fig 6). It is understanding of the arrangement of the bridging leaflets of the atrioventricular valve that is probably most crucial in determining the optimal surgical approaches to this important lesion [13].
|
|
Interest in the structure of the valves of the heart dates back to at least the middle ages. Leonardo da Vinci, as noted by Robiscek [14], was well aware of the importance of the sinuses of Valsalva for the proper functioning of the aortic valve, and he also illustrated a bicuspid valve. It was the anatomist Vesalius who probably first likened the atrioventricular valve of the left ventricle to the episcopal miter [15]. In this respect, however, Hyrtl [16] commented more than 100 years ago that, in terms of adequate description, the term ``mitral'' came into the heart like the devil into the baptismal font. Indeed, discussions still continue concerning the number of leaflets in, and the best descriptions for, both atrioventricular valves [17, 18]. For the arterial valves, most surgeons, although accepting the absence of a collagenous ring in the sense of a planar circle, continue to describe an ``annulus''. The meaning of annulus as a ring is well accepted, but rings can take different shapes and forms. Such problems with the annulus parallel those existing with the commissure. Consider the definitions of a commissure given in the Shorter Oxford English Dictionary [19]. Of the four definitions, three are pertinent to the valves of the heart. All three define a commissure as the zone of apposition between adjacent structures, such as the line of closure of the lips or the eyelids. Taking the lips or eyelids as exemplars for the cardiac valves, those with two leaflets, when closed, would have one commissure, and would exhibit the two ends of this commissure when open. When interpreted in this fashion, valves with three leaflets would have three commissures, each extending from the periphery of the orifice to its centroid seen with the valve in its closed position (Fig 7). This definition of ``commissure'' in a standard dictionary, nonetheless, already has been eroded by decades of medical use. Thus, in Dorland's Illustrated Medical Dictionary [20], the term is used to describe sites of junction between adjacent cusps of the valves of the heart but also to describe the lateral margins of the zone of apposition between structures, such as the nasal or temporal commissures of the eyelids, or the junction of upper and lower lips at either side of the mouth. This latter definition is also the one given in Stedman's Medical Dictionary [21], and this is the conventional use of the term when describing heart valves, with the mitral valve usually described as having medial and lateral commissures. It is similarly the circumferential ends of the zones of apposition between arterial valvar leaflets that are named in this way. This usage will certainly continue. But, as we will see below, it tends to obscure the fact that the valves must function when closed. To describe appropriately the structures of the valve in closed position, it is essential to recognize the zone, or zones, of apposition between adjacent leaflets.
|
). This zone of primary apposition between the mural and aortic (anatomically more accurate adjectives than ``anterior'' and ``posterior'') leaflets of the mitral valve would be described appropriately as the valvar commissure if this term was used in its vernacular sense. As we have seen, nonetheless, it is the two ends seen when the valve is opened in typical pathologic fashion that are, and will continue to be described as the commissures. What, then, of the zone of apposition between the leaflets? Surprisingly, a recent review of the surgical anatomy of the mitral valve, designed to facilitate its description, ignored totally the existence of this zone and concentrated only on the commissural areas [25]. This is a significant omission. Once the importance of the solitary primary zone of apposition is appreciated, then the additional slits in the mural leaflet can be compared nicely to the pleats placed in a skirt or kilt, thus enabling it to ``sit'' in elegant fashion [17].
|
) or is divided into separate valvar orifices for the right and left ventricles (``ostium primum'' or ``partial'' defect) (Fig 9b). Furthermore, describing the closure point of the bridging leaflets themselves as a zone of apposition then defuses arguments concerning its existence as a ``commissure''. There can be no question concerning the fact that the left ventricular components of the bridging leaflets close together over this area. Whether this left valve in atrioventricular septal defects then is described suitably as a ``mitral'' valve is a matter of taste. Analysis in our suggested fashion certainly shows that the valve differs from the normal mitral valve in all anatomic respects apart from its residence within the left ventricle. Even in this respect, however, the bridging nature of the superior and inferior leaflets indicates their biventricular commitment. It also should be emphasized that, just because the valve has three leaflets and the junction of the bridging leaflets, when closed, is part of the orifice of the valve, these facts cannot in themselves be taken as evidence that the surgeon does not need to repair, or partially close, this zone of apposition. It is the case that the left valve is often competent before operation. After the operation, the leaflets are distorted to some extent, and partially immobilized, by closure of the septal defects. It may well be necessary, and expedient, for the surgeon to close partially the ``cleft'' [13]. This is the more so because, usually, the valvar orifice of the left atrioventricular junction is larger in atrioventricular septal defects than in the comparably proportioned normal heart.
|
). The mitral valve, in contrast, has its aortic leaflet in fibrous continuity with two of the leaflets of the aortic valve, thus conferring a kidney shape to the atrioventricular annulus of the left ventricle (Fig 10b). Further evidence for the trifoliate nature of the tricuspid valve comes from the observation that, in Ebstein's malformation, it is the mural and septal components that are displaced downwardly, whereas the anterosuperior leaflet retains its hinge point at the atrioventricular junction. This consistent deformity would be most unexpected if, as suggested by Victor and Nayak [18], the valve possessed only septal and mural leaflets.
|
). In this way, part of the wall of the supporting ventricular structures is incorporated within the three valvar sinuses. More significantly, three triangles of arterial wall are incorporated within the ventricular outflow tracts, being located directly beneath the apices of the zones of apposition of the leaflets at their attachments to the sinutubular ridge. This combination of interdigitating components of the arterial wall within the sinuses and the ventricle results in the generally recognized coronet, or ``annulus'', of the arterial valves. The interleaflet triangles, however, do not support exclusively the hinge points of the leaflets in the fashion of the annulus of the atrioventricular valves. Rather, they occupy the spaces between the arterial sinuses. The structure and location of these triangles are important to surgeons wishing to enlarge the outflow tract. Incisions within the triangles widen the outflow tract without placing at risk the valvar leaflets, but such incisions open directly into the pericardial space (Fig 12). It is knowledge of these relatively forgotten triangles [27] that leads to proper understanding of the functional anatomy of the arterial valves, and to the structure of the valvar annulus.
|
|
We have tried, in this review, to highlight certain areas that should be familiar to all surgeons who operate within the heart. Our goal has been to emphasize features and nuances that may not immediately be apparent to the operating surgeon. We also have tried to demonstrate that the continued use of jargon, particularly when sufficiently established by general use to become conventional wisdom, is not always of benefit to those coming new into the field. The advances and strides made in the practice of cardiac surgery over the past 20 years have been truly remarkable. By concentrating on the more subtle points, such as emphasized in this review, we will continue to make progress as we move into the twenty-first century.
Acknowledgments
The concepts outlined in this review have drawn heavily on our interactions with numerous colleagues, and we thank them all. The illustrations, however, could not have been produced without the help provided over and above the call of duty by Dr Siew Yen Ho at the National Heart and Lung Institute, and Drs James R. Zuberbuhler and William H. Neches and Mr William A. Devine at Children's Hospital of Pittsburgh. To them, we express our particular thanks for their ongoing support. Throughout the period of study, Prof Anderson has been supported by the British Heart Foundation, the Joseph Levy Foundation, and the Patrick Dick Memorial Fund.
Footnotes
Presented at the Thirty-first Annual Meeting of The Society of Thoracic Surgeons, Palm Springs, CA, Jan 31–Feb 2, 1995.
Address reprint requests to Prof Anderson, Department of Paediatrics, National Heart & Lung Institute, Dovehouse St, London SW3 6LY, United Kingdom.
References
This article has been cited by other articles:
|
|
 |
|
  R. H. Anderson and M. Kanani Mitral valve repair: critical analysis of the anatomy discussed MMCTS, February 19, 2007; 2007(0219): 2147. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. H. Anderson and B. R. Wilcox Reply Ann. Thorac. Surg., June 1, 2000; 69(6): 1990 - 1990. [Full Text] [PDF]
|
|
|
|
|
|
R. H. Anderson How Should We Optimally Describe Complex Congenitally Malformed Hearts? Ann. Thorac. Surg., September 1, 1996; 62(3): 710 - 716. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
