| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 1995;60:1163-1165
© 1995 The Society of Thoracic Surgeons
Departments of Pathology and Cardiology, Children's Hospital, Boston, Massachusetts
Tetralogy of Fallot {S,D,I} is a recently discovered form of tetralogy that was first reported in 1988 by Foran and associates [1] on the centennial of Fallot's classic description of the typical form of this anomaly [2]. Recently, Santini and colleagues [3] reported the first surgical repair of this rare variant of tetralogy without the use of an external conduit. In an accompanying editorial, Anderson [4] raised several questions of general interest that I shall endeavor to answer briefly here.
What is tetralogy of Fallot {S,D,I}? This has to be the first question of anyone who has never seen this rare anomaly. As the name indicates, there is situs solitus of the viscera and atria (S), D-loop ventricles (D) with atrioventricular concordance, inverted normally related great arteries (I) with ventriculoarterial concordance, and inverted (mirror-image) tetralogy of Fallot. There is pulmonary infundibular and valvar stenosis or atresia, aortic–mitral fibrous continuity, and a large subaortic conoventricular type of ventricular septal defect with anterosuperior malalignment of the conal septum. Because the inverted aorta lies to the left of the inverted pulmonary artery, the right coronary artery runs across the obstructed subpulmonary infundibulum from left to right to reach the right-sided atrioventricular groove. This course of the right coronary artery across the obstructed pulmonary outflow tract has been treated by means of an external right ventricle (RV) to pulmonary artery conduit [1]. However, as described in a recent report [3], Jonas and colleagues were able to establish satisfactory RV to pulmonary arterial continuity by myocardial excision beneath the right coronary artery, thereby avoiding an external conduit that would have to undergo repeated changes over time.
Does tetralogy of Fallot {S,D,I} ``really have typical morphology of tetralogy of Fallot?'' Yes. The morphology of the infundibulum and great arteries was typical of inverted (mirror-image) tetralogy of Fallot, as the angiocardiograms showed well (Fig 2 in [3]).
``Does the chosen title convey the morphology present to the working surgeon?'' Yes, as long as the surgeon is familiar with this rare form of tetralogy of Fallot [1, 3].
Can one make the diagnosis of double-outlet right ventricle when one of the semilunar valves is in direct fibrous continuity with an atrioventricular valve? Yes. For example, when double-outlet right ventricle coexists with hypoplastic left heart (mitral stenosis or atresia), it is not rare for either the aortic or the pulmonary valve to be in direct fibrous continuity with the tricuspid valve, but not with the mitral valve [5]. This is what Dr Richard Rowe called the ``infantile type'' of double-outlet right ventricle (because the natural history usually was death in infancy).
What is the difference between tetralogy of Fallot {S,D,I} and anatomically corrected malposition of the great arteries {S,D,L}? In anatomically corrected malposition (ACM), there typically is a well-developed muscular subaortic conus (not just a 1-mm-long subaortic conus suggested by Anderson [4]), with or without a well-developed muscular subpulmonary conus [6, 7]. Also, in ACM the great arteries are very abnormally related, externally resembling transposition of the great arteries with an anterior aorta and a posterior pulmonary artery [6, 7].
Hence, tetralogy {S,D,I} and ACM {S,D,L} are both similar and different. The similarity is that both have concordant ventriculoarterial alignments. In both, the morphologically left ventricle (LV) is aligned with the aorta and the morphologically RV is aligned with the pulmonary artery. The differences concern (1) the anatomy of the conus (normal subpulmonary type in tetralogy {S,D,I} and abnormal subaortic or bilateral [subaortic and subpulmonary] type in ACM {S,D,L}) and (2) the relations between the great arteries (normal or tetralogy type in tetralogy {S,D,I} but transposition-like in ACM {S,D,L}). Hence, although the ventriculoarterial alignments are concordant in both anomalies, the ventriculoarterial connections (the anatomic types of conal connector) are very different: of the normal subpulmonary type in tetralogy {S,D,I} and of the abnormal subaortic, or subaortic and subpulmonary type in ACM {S,D,L}. In view of the anatomic differences, we think that these two anomalies should continue to be classified separately.
Is the infundibulum part of the ventricles, or is it really part of the great arteries? ``My own preference,'' Anderson says [4], ``is to define the arterial segment as starting at the ventriculoarterial junction. Within my definition, the ventricular outlet components are considered as part of the ventricular mass.'' Within our approach, as Anderson adds [4], ``the infundibular structures are taken to belong to the distal cardiac segment, which is then defined as `infundibuloarterial'.''
Anderson is essentially right about this difference. He regards the third of the three major cardiac segments to be the great arteries (only), whereas we consider this third major cardiac segment to consist of the distal or subsemilunar infundibulum (parietal band portion) plus the great arteries, ie, the infundibuloarterial segment or the conotruncus. We prefer the latter view for many reasons: Embryologists have long understood that this third segment is the conotruncus, as the work of Keith [8] illustrates. Comparative anatomists have long held this view, as the work of Robb [9] shows.
Our own anatomic observations of straddling conus and left ventricular conus in humans, plus our own comparative anatomic observations all support the view that the conus arteriosus is (as its name indicates) an arterial structure that in mammals becomes incorporated into the ventricles, mostly into the RV and to a minor degree into the LV. In humans, the subaortic conus can straddle the ventricular septum to any degree and can be located predominantly above the LV sinus, as in ACM [6, 7] and as in double-outlet left ventricle [10]. The distal or subsemilunar part of the conus (the parietal band part) definitely is not an intrinsic and inseparable part of the RV. The subsemilunar conus can and does dissociate from the RV. In other words, the conal septum and parietal band can be located above the LV sinus, because the conus is not really part of the RV. Instead, the conus is part of the infundibuloarterial segment. The subsemilunar conus forms the outflow tract of both ventricles, because the conus really is part of the great arterial segment, not an intrinsic and inseparable part of the RV, as many have thought. Nor is the conus simply a part of the subsemilunar ``ventricular mass'' (RV and LV). The subsemilunar conus really is the connector by which the great arteries and the ventricular sinuses (pumps) attach to each other.
When you examine the heart of an amberjack or ray, you will see that the musculature of the conus arteriosus extends from the ventricle all the way forward to the gill arches (the branchial or aortic arches). Only then will you understand what conus arteriosus really means. In evolutionary terms, the ventral aorta (homologous with our ascending aorta) begins as a muscle-bound vessel. It was only with the evolution of air-breathing and land-living by the Amniota that radiated into reptiles, birds, and mammals that the musculature of the conus arteriosus receded toward the ventricular portion of the heart, beneath the semilunar valves, and became incorporated mostly into the RV and somewhat also into the LV. This retreat of the musculature of the conus arteriosus from the aortic arches to a subsemilunar location was also important because the now fibroelastic ventral aorta—no longer muscle-bound—could evolve into the mammalian ascending aorta and main pulmonary artery. No longer encased in the musculature of the conus arteriosus, the fibroelastic great arteries could untwist about each other, thereby dissipating the rotation introduced by D-loop formation and asymmetric development of the subsemilunar conus, the latter being Mother Nature's arterial switch operation [11].
Conceptually, one cannot simply detach the great arteries and plug each one into the opposite semilunar valve, as Anderson [4] suggests, even though this may appear to be what has happened during abnormal development. Instead, one must also ask, How did the great arteries become malpositioned during abnormal development? We think that the short answer is: conal maldevelopment [8, 11].
Hence, the muscular conus arteriosus and the fibroelastic great arteries really are two different parts of the same cardiac segment—the infundibuloarterial segment. This is why variations in the development of the conal connector have such an impact on the alignments and spatial relations of the great arteries [8, 11].
Is a long verbal description better than a brief diagnostic name? Anderson suggests that a long verbal description (his was 48 words, or more, in length) is better than a brief diagnostic label: tetralogy of Fallot {S,D,I} (four ``words'' in length, counting the segmental set as one word). We like the description with which Anderson ended his editorial [4]. In fact, it was very similar to the description with which we began our article [3]. So, it is not that we do not like Anderson's long description; we do. Instead, from the practical standpoint, we think that brief diagnostic names are very useful. Once a diagnostic name is understood, little or nothing more need be said. For example, the diagnostic labels tetralogy of Fallot or double-outlet right ventricle of the Taussig-Bing type ordinarily suffice. Names and symbols are convenient shorthand, or thought packages. The three words tetralogy of Fallot make it unnecessary to spell out the four anatomic features of the classic tetrad. Diagnostic labels are helpful, and should not be denigrated in favor of long verbal descriptions. However, no diagnostic name is of any use until one knows what it means. Understanding and familiarity are key.
Anderson says, ``The problem that I find with the concept of segmental notation ...is its cryptic nature when applied to rare and unusual lesions'' [4]. We think that this just indicates lack of familiarity. No diagnostic label or symbolic notation is helpful until one becomes familiar with it. Nonetheless, Anderson is clearly becoming familiar with segmental notation. He correctly points out that the usual form of tetralogy of Fallot has the segmental anatomy {S,D,S} and the inverted or mirror-image form of tetralogy has the segmental set {I,L,I}. Tetralogy {S,D,I} is now a third documented segmental set with a tetralogy type of anomaly of the infundibulum and great arteries [1, 3]. Because a tetralogy type of malformation of the infundibulum and great arteries can occur with solitus (S) or inverted (I) normally related great arteries, one wonders how many other rare forms of tetralogy remain to be discovered.
Rare segmental sets with atrioventricular discordance and normally related great arteries (S or I) that have been documented and could have a tetralogy of Fallot type of infundibuloarterial malformation include {S,L,S}, {S,L,I}, and {I,D,S} [12]. To our knowledge, the following segmental sets have not as yet been documented: {I,D,I} and {I,L,S}.
Hence, the advantages of segmental symbolic anatomy include brevity, convenience, and universality. The latter means that this approach always works. For example, even with visceroatrial situs ambiguus, the diagnosis of D-loop or L-loop ventricles always applies, even when the atrial situs is uncertain and the concepts of atrioventricular concordance or discordance do not apply. Of course, one can always fall back on a long verbal description.
However, there is more than just convenience as a reason for becoming familiar with segmental symbols. They facilitate segmental set analysis. Many segmental sets have distinctive and different associated malformations. Consequently, segmental set analysis is illuminating and helpful, both diagnostically and surgically. Consider the segmental set {S,D,I}. For reasons that are still unclear, almost all known cases of {S,D,I} have had a tetralogy of Fallot type of malformation of the infundibulum and great arteries. Why this conotruncal malformation is associated with the segmental set of {S,D,I} is unknown.
Complex congenital heart disease largely consists of segmental situs mismatches. Segmental symbols and sets facilitate the analysis of these segmental situs mismatches. How many different segmental sets are there in human congenital heart disease? What are their relative frequencies? How many significantly different constellations of associated malformations exist in association with the various segmental sets (types of congenital heart disease)? The answers to all of these questions are unknown at the present time.
Footnotes
Address reprint requests to Dr Van Praagh, The Cardiac Registry, Children's Hospital, Harvard Medical School, 300 Longwood Ave, Boston, MA 02115.
References
This article has been cited by other articles:
|
|
 |
|
  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 |
