Common Atrioventricular Canal

Acknowledgment

I would like to thank Silvio Litovsky, M.D., for his great help in analyzing the data on which this chapter is based.

Definition

The anomaly known as common atrioventricular canal (or atrioventricular septal defect) is characterized anatomically by incomplete separation of the atrioventricular canal into mitral and tricuspid canals. Hence, the atrioventricular canal is incompletely subdivided, or in common.

The morphology and classification of these malformations are described in detail below. But first we should mention the history and terminology of these anomalies.

History and terminology

Common atrioventricular canal has been known for more than a century. In 1899, Griffith presented an “example of a large opening between the two auricles of the heart, unconnected with the fossa ovalis.” This was an ostium primum type of atrial septal defect, a partial form of common atrioventricular canal. Griffith wrote that a Dr. Norman Moore had presented such a case to the Pathology Society of London in 1881.

In 1936, Maude Abbott of McGill University in Montreal, Canada, described the “ostium primum atrial septal defect” and the “common atrioventricular canal defect.” Following World War II, in 1948, Rogers and Edwards at the Mayo Clinic in Rochester, Minnesota, recognized that incomplete division of the atrioventricular canal resulted in patency of the interatrial foramen primum, also known as persistent common atrioventricular ostium . They reported five cases and reviewed the literature. The older Latin name of this malformation had been atrioventricularis communis , persistent common atrioventricular ostium being an English translation. By 1956 and 1958, Wakai and Edwards ^, had described and classified persistent common atrioventricular canal into complete , partial , and transitional forms.

As is shown under Morphology and Classification, the complete form of common atrioventricular canal has a complete atrioventricular septal defect and a common atrioventricular valve. The partial form typically has an ostium primum type of atrial septal defect—which is a partial atrioventricular septal defect—and a cleft anterior leaflet of the mitral valve, but no ventricular septal defect. The transitional form characteristically is like the partial form of common atrioventricular canal, except that in addition there are one or more small ventricular septal defects beneath the anterior and/or posterior endocardial cushion components of the cleft anterior leaflet of the mitral valve. Hence, the transitional form is an “almost” incomplete form, except that it also has one or more tiny ventricular septal defects.

Thus, by the mid to late 1950s, Dr. Jesse Edwards and his colleagues had made a very good start toward the description and classification of the anomalies that they called common atrioventricular canal.

At about the same time in London, England, Bedford and his colleagues—one of whom was Dr. Walter Somerville—coined the term atrioventricular defect for these anomalies, as was later recounted by Dr. Jane Somerville in 1968. Hence, it is not too surprising that the British pediatric cardiology community tended to prefer the designation atrioventricular defect, which was used by Piccoli et al in 1979.

However, by 1982, our British colleagues ^, had concluded that Bedford and Somerville’s term atrioventricular defect should be modified to atrioventricular septal defect, perhaps in order to make it more specific, and this remains their preference to this day.

It is noteworthy that the designation atrioventricular (AV) septal defect is an example of medical synecdoche . Linguistically, synecdoche is a figure of speech in which a part is used to represent a whole, or vice versa. AV septal defect, which is part of common AV canal, is used to represent the whole anomaly, which typically has both septal and leaflet malformations.

Another widely used term for common atrioventricular canal is endocardial cushion defect. This designation is somewhat passé now, probably because it is so very nonspecific. Almost any anomaly of the atrioventricular valves can be regarded as an endocardial cushion defect. Tricuspid atresia, stenosis, regurgitation, and Ebstein’s anomaly may all be regarded as different kinds of endocardial cushion defect. So, too, can mitral atresia, stenosis, and regurgitation. But none of the aforementioned malformations has of necessity common atrioventricular canal. Thus, although common atrioventricular canal is an endocardial cushion defect, the latter term is far too broad and nonspecific to denote the anomalies that we are considering precisely.

We prefer the term common atrioventricular (AV) canal for several reasons. This designation includes both the septal defect component and the leaflet malformation component of this anomaly. In other words, common AV canal includes much more than an AV septal defect. Surgically, the leaflet malformations are the more difficult part of common AV canal.

Proponents of the AV septal defect terminology agree with the foregoing, of course, and they consequently are now talking and writing about AV septal defect with a common AV valve, and AV septal defect with two AV valves, a considerable improvement, we think.

Bharati, Lev, and their colleagues have always preferred common atrioventricular orifice, which is very close to Edwards’ common atrioventricular canal.

In addition to the brevity, accuracy, and inclusiveness (septal and leaflet anomalies) of Edwards’ designation, we think it appropriate to honor the pioneer who first described and classified these anomalies by using his terminology. If there were something wrong with this designation common AV canal, then it should be changed. But this is not the case: it is brief, anatomically accurate, and all-inclusive—embracing both the septal and the leaflet malformations.

What about common AV canal defect? This, too, is very close to Dr. Jesse Edwards’ original designation. We think that defect is redundant: common AV canal is obviously a defect. No one thinks we are talking about human embryos, in which a common AV canal is normal prior to Streeter’s horizon XVII, that is, prior to 34 to 36 days of gestational age. Earlier than horizon XVII in humans, common AV canal is not a defect; it is normal. But thereafter, common AV canal is a defect, as is generally understood. Thus, the term AV canal defect is certainly not wrong. However, our preference is to avoid redundancy in medical terminology whenever possible.

It should also be understood that not all partial forms of common AV canal have an AV septal defect. For example, isolated cleft of the mitral valve can have a cleft of the AV canal type, but without an AV septal defect. This means that AV septal defect is not a fully satisfactory common denominator for the entire spectrum of common AV canal.

From a developmental viewpoint, common AV canal really means that the superior and inferior endocardial cushions of the AV canal are not completely fused in the normal way. It must be remembered that the concept of common AV canal includes both the septal defect and the leaflet defect components. Consequently, isolated cleft of the mitral valve of the AV canal type represents a partial form of this anomaly in the sense that the superior and inferior endocardial cushions are not completely fused in the normal way.

From the anatomic standpoint, however, isolated cleft of the mitral valve of the AV canal type is not a partial form of common AV canal in the sense that the AV canal is completely divided into mitral and tricuspid canals.

Conversely, not all AV septal defects have common AV canal. For example, left ventricular–to–right atrial shunting (Gerbode defect) has a hole in the AV portion of the membranous septum. This is why left ventricular–to–right atrial shunt has been called an AV septal defect. But this type of AV septal defect is very different from that seen in common AV canal.

In somewhat greater detail, although the so-called Gerbode defect has been regarded as a defect in the AV portion of pars membranacea septi, in our experience this membranous ventricular septal defect (VSD) really does not extend above the tricuspid annulus. The Gerbode defect does not have an ostium primum defect component. Instead, the blood shunts through and between the tricuspid valve leaflets and their chordal attachments at the anteriosuperior commissure of the tricuspid valve. Not all membranous VSDs have left ventricular–to–right atrial shunting. Those that do (Gerbode defects) have an incompletely occlusive anterosuperior commissure of the tricuspid valve. The tricuspid component of the Gerbode defect is indeed above the membranous VSD component. This is why Gerbode defects have been described as having an atrio ventricular septal defect. However, this atrioventricular septal defect is very different from that seen in common AV canal, as mentioned. In the typical membranous VSD, the anterior leaflet of the tricuspid valve closes the anterosuperior commissure. This portion of the anterior tricuspid leaflet bulges upward, forming a tricuspid pouch—produced by the left-to-right jet of blood striking the ventricular side of the anterior tricuspid leaflet.

Thus, the real difference between a typical membranous VSD and a Gerbode defect is not the nature of the septal defect. Both have membranous VSDs. But, in addition, the Gerbode defect has an incompetent (regurgitant) anterosuperior commissure of the tricuspid valve. Consequently, we think that the Gerbode defect should not be regarded as an AV septal defect.

To summarize, the foregoing are the reasons why we think that it is very difficult to improve on common atrioventricular canal as the name for the fascinating malformations that we are about to consider. To the best of our present knowledge, the term and concept of common AV canal is the only one that includes all forms of this anomaly, and excludes all malformations that are not part of the spectrum of common AV canal.

We also fully understand and respect the preferences of others: AV septal defect is an important part of common AV canal. The complete form of common AV canal certainly does have a common AV orifice. And common AV canal is indeed an endocardial cushion defect. All of the foregoing terms should be understood.

However, the main focus of this chapter—and of this book—is not terminology, but data. This chapter presents a detailed study of 443 postmortem cases of complete and partial forms of common AV canal. But first we must consider morphology and classification in greater detail.

Morphology and Classification

1.
The complete form of common AV canal typically has a complete AV septal defect, that is, an ostium primum type of atrial septal defect lying above and behind the leaflets of a common AV valve, which is confluent with a VSD of the AV canal or inlet type lying below and in front of the leaflets of the common AV valve ( Fig. 11.1 ).
2.
The partial form of common AV canal typically has a partial AV septal defect, usually an ostium primum type of atrial septal defect (ASD), with no VSD of the AV canal type, and with a cleft in the anterior leaflet of the mitral valve.
3.
The transitional form of common AV canal is like the partial form, except that it has a few small interstices between the fibrous attachments of the leaflets of the mitral and tricuspid valves to the crest of the muscular ventricular septum. These few small interstices result in a very small restrictive ventricular septal defect(s) of the AV canal type. As mentioned, the transitional form of common AV canal lies between the complete and partial forms. The VSD of the AV canal type has been obliterated almost totally, so that the development of the AV canal has proceeded almost to the partial common AV canal stage; hence the transitional form is almost a partial common AV canal—except for the coexistence of a few small, slit-like VSDs.
4.
The intermediate form of common AV canal indicates the rare type in which the superior (anterior) and inferior (posterior) leaflets of the common AV valve (Fig.11.1) have fused in the midline, dividing the common AV valve into left-sided and right-sided AV valves. In addition to a large ostium primum type of ASD, the rare feature is the coexistence of what can be a large VSD of the AV canal type. As Dr. Dwight McGoon commented to me as we were both examining heart specimens of the intermediate type at one of Dr. Maurice Lev’s exhibits at a meeting of the American Heart Association, “The VSD is big enough to drive a Mack truck through.” Usually, when the superior and inferior leaflets of the common AV valve have fused in the midline, the VSD component is absent (as in the partial form of common AV canal), or very small (as in the transitional form). This is why the intermediate form is rare.
5.
The left ventricular type ^, of common AV canal indicates that the common AV valve opens predominantly into the morphologically left ventricle (LV). This in turn indicates that the morphologically right ventricle (RV) is hypoplastic or absent, which is why the common AV valve opens predominantly or entirely into the LV. This is also known as the LV dominant type of common AV canal.
6.
The right ventricular type ^, of common AV canal means that the common AV valve opens predominantly or entirely into the RV—because the LV is hypoplastic or absent. This is also known as the RV dominant type of common AV canal.
7.
The balanced form ^, of common AV canal indicates that the common AV valve opens approximately equally into the RV and the LV. This in turn means that both the RV and the LV are well developed. When the AV canal is described as unbalanced to the left , ^, this means that the common AV valve opens predominantly into the LV. Hence, unbalanced to the left is synonymous with the LV type of common AV canal.

Similarly, when the AV canal is described as unbalanced to the right, ^, this means that the common AV valve opens predominantly into the RV. Consequently, unbalanced to the right is synonymous with the RV type of common AV canal.

The LV type of common AV canal, or AV canal unbalanced to the left, indicates that the common AV valve is predominantly left-sided with D-loop ventricles, but predominantly right-sided with L-loop ventricles—because the LV is left-sided with a D-loop and right-sided with an L-loop.

Similarly, the RV type of common AV canal, or a common AV canal unbalanced to the right, means that the common AV valve is right-sided with D-loop ventricles (because the RV is right-sided), but left-sided with L-loop ventricles (because the RV is left-sided).

Hence, the right-sided and left-sided types of common AV canal are used morphologically, not positionally. The morphologically RV may be right-sided (D-loop ventricles) or left-sided (L-loop ventricles), and the morphologically LV may be left-sided (D-loop ventricles) or right-sided (L-loop ventricles).

It is also noteworthy that it is not the common AV valve or the common AV canal that is unbalanced. Instead, it is the development of the ventricles that is unbalanced. Hence, in so-called unbalanced common AV canal, one may say (with a wink) that it is not the AV canal that needs to see the psychiatrist. It’s the ventricles.

It should be pointed out that the right ventricular type and the left ventricular type of unbalanced common AV canal were two of Dr. Maurice Lev’s most important contributions to the understanding of common AV canal. Another was the description of the intermediate type of common AV canal. In both of these achievements, Dr. Saroja Bharati ^, played a significant role.

In the complete form of common AV canal, the Rastelli classification is widely used and hence merits description:

In type A, the superior (anterior) leaflet of the common AV valve is divided and attached to the crest of the muscular ventricular septum ( Fig. 11.2 ).
In type B, the superior leaflet of the common AV valve is partially divided and is not attached to the ventricular septal crest. Instead, the superior leaflet is attached to the anterior papillary muscle of the right ventricle ( Fig. 11.3 ). Rastelli et al said that the superior leaflet attaches to a papillary muscle that arises from the right ventricular septal surface. In the rare type B, this is how it looks—because the moderator band is so short. Consequently, the anterior papillary muscle of the RV appears to originate directly from the RV septal surface.
In type C, the superior leaflet of the common AV valve is undivided and unattached to the ventricular septal crest ( Fig. 11.4 ).

Type A Completely Common AV Canal

In somewhat greater detail, let us carefully examine completely common AV canal type A ( Fig. 11.2 ). In Rastelli’s drawing ( Fig. 11.2A ), note that above the ventricular septum, the superior (anterior) leaflet of the common AV valve is divided into two components (A,A), which are attached to the crest of the muscular ventricular septum by numerous short chordae tendineae. The inferior (posterior) leaflet of the common AV valve is also attached to the crest of the muscular ventricular septum by numerous short chordae tendineae. Despite these chordal attachments, a VSD of the AV canal type exists through the chordae tendineae. The components of the mitral valve (MV) and tricuspid valve (TV) are readily recognizable, even though the AV valve is in common (undivided, or unfused in the midline). Note the left-sided and right-sided lateral leaflets (L) of the common AV valve, that lie between the superior (A, A) and the inferior (P) leaflets. The potentially TV opens into the morphologically right ventricle (RV) and the potentially MV opens into the morphologically left ventricle (largely unseen).

Fig. 11.2B shows a photograph of the complete form of common AV canal type A that is very comparable to Rastelli’s drawing ( Fig. 11.2A ). Note the atrioventricular septal defect (D) that lies between the “scooped-out” or concave crest of the muscular ventricular septum anteroinferiorly and the nude margin of the atrial septum posterosuperiorly. The superior leaflet of the common AV valve (S, S) is divided and attached to the crest of the muscular ventricular septum by short chordae tendineae.

How can one be sure that common AV canal is present, given only the right atrial (i.e., surgical) view? Because the anteroinferior rim of the atrial septum is nude or bare. Anteroinferiorly, the atrial septum does not have its tricuspid valve “pants” on: it is naked. Once you see the bare anteroinferior margin of the atrial septum, you know that common AV canal is present. In front of and below this bare rim of the atrial septum lies the AV septal defect, that is, the defect in the septum of the AV canal.

So now you say to yourself, “Okay, a common AV canal is present.” The next question is: “Is it a complete type or a partial type of common AV canal?” Can you make this diagnosis from the right atrial view? The answer is yes. Is there a VSD component? If so, this is a complete type of common AV canal—or a complete AV septal defect. Look at the superior leaflet ( Fig. 11.2B ). You can easily see that a VSD component is present between the short chordae tendineae attaching the superior leaflet to the ventricular septal crest. This VSD component extends in front of or below the divided superior leaflet of the common AV valve. So this definitely is a complete form of common AV canal. And it’s a Rastelli type A, because the superior leaflet is divided and attached to the crest of the muscular ventricular septum. The large space behind and above the leaflets of the common AV valve is the ostium primum component of this complete atrioventricular septal defect. Note that the ostium primum type of atrial septal defect and the VSD of the AV canal or inlet type are confluent, which is typical of complete forms of common AV canal.

Is there a VSD beneath the inferior leaflet (I, Fig. 11.2B )? There may or may not be. Often there isn’t.

Note also the small ostium secundum type of atrial septal defect, the normal opening of the coronary sinus, and the right atrial and right ventricular hypertrophy and enlargement ( Fig. 11.2B ).

Hence, it is readily possible for a surgeon to make a highly accurate diagnosis of completely common AV canal type A, given only a right atrial view ( Fig. 11.2A ).

The left ventricular views in completely common AV canal type A ( Figs. 11.2C and D ) are also highly informative. These views can be well imaged with two-dimensional and three-dimensional echocardiography and with magnetic resonance imaging (MRI). Again one can see that the divided superior leaflet is attached to the scooped-out crest of the muscular ventricular septum by a thicket of short chordae tendineae. There is a VSD component beneath (in front of, or anterior to) the superior leaflet in both cases ( Figs. 11.2C and D ). There is a VSD component beneath the inferior leaflet in Fig. 11.2D , but not in Fig. 11.2C .

Note that the inferior leaflet is much closer to the ventricular septal crest than is the superior leaflet. We think that this fact may well be of morphogenetic importance, as is explained subsequently.

It is noteworthy that the left ventricular inlet dimension (from the inferior leaflet component to the left ventricular apex) is much shorter than is the left ventricular outlet dimension (from the left ventricular apex to the aortic valve). This is particularly well seen in Fig. 11.2D , where the ratio of the left ventricular inlet dimension to the left ventricular outlet dimension equals approximately 0.6. Normally, the left ventricular inlet/outlet dimension ratio is approximately 1.0. This marked shortening of the left ventricular inlet dimension relative to the left ventricular outlet dimension is characteristic of common AV canal, both complete and incomplete. This typical shortening of the left ventricular inlet dimension contributes to the aforementioned “scooped-out” appearance, and reflects failure of formation of the septum of the AV canal, which contributes to the inlet part of the normal definitive interventricular septum.

The outlet portion of the septum also is not normal. Although approximately normal in apex-base length, the outlet part of the septum is much shallower in superior-inferior dimension than normal.

The “goose-neck” appearance seen angiocardiographically (and with other imaging modalities) during atrial systole and ventricular diastole is related in part to the high horizontal “floor” of the subaortic left ventricular outflow tract during atrial systole, which results in the goose-neck appearance. The superior leaflet component of the anterior mitral leaflet makes an angle of approximately 90° with the left ventricular septal surface. And the anterosuperior leaflet of the common AV valve inserts into the ventricular septum very high (very superiorly). Consequently, the subaortic left ventricular outflow tract is very shallow—very lacking in superoinferior depth.

Normally, with division of the atrioventricular canal into mitral and tricuspid valves, the angle formed by the anterior leaflet of the mitral valve and the left ventricular septal surface is much smaller, often about 45°. And the superoinferior dimension of the subaortic LV outflow tract is much deeper. The “floor” of the LV outflow tract is greatly lowered, increasing the superoinferior dimension of the LV outflow tract, thereby “degoosing” the goose-neck appearance.

This is why the normal subaortic LV outflow tract does not at all resemble a goose-neck appearance. Why not? Because the superoinferior dimension of the normal subaortic LV outflow tract is so much greater—so much deeper. Why? Normally, the superior and inferior leaflets of the common AV valve attach to each other—thereby closing the cleft, instead of attaching high up to the LV septal surface. By attaching to each other and closing the cleft in the anterior leaflet of the mitral valve, the superior and inferior leaflets of the common AV valve greatly lower the floor of the subaortic LV outflow tract. The process of superior cushion–inferior cushion fusion (cleft closure) also greatly reduces the angle between the endocardial cushions and the LV septal surface from about 90° in common AV canal to about 45° normally after cleft closure.

Thus, cleft closure is important not only for the formation of the anterior mitral leaflet, but also for the creation of a deep and nonstenotic subaortic LV outflow tract. Discrete fibrous subaortic stenosis is caused by residual, unremodeled endocardial cushion fibrous tissue (see the section on fibrous subaortic stenosis later in this chapter).

The ventricular systolic image with common AV canal is “the scallops.” The thicket of chordae tendineae in type A completely common AV canal creates a scalloped appearance in front of the anterior leaflet of the common AV valve during ventricular systole.

In Fig. 11.2C , it is noteworthy that subaortic insertions of short chordae tendineae can create, or contribute to, subaortic left ventricular outflow tract stenosis in type A completely common AV canal—and also in partially common AV canal in which these chordal insertions are dense enough to obliterate the VSD.

Why does the electrocardiogram in common AV canal typically have a counterclockwise and superior frontal QRS loop? Because the atrioventricular bundle of His has to enter the ventricular septum in a very low or inferior position—because of the relatively huge hole in the center of the heart, the AV septal defect. The accession wave of ventricular depolarization must therefore travel from inferiorly to superiorly as it progresses from the ventricular septum to the ventricular free walls. If the heart is pointing leftward (levocardia), and if the common AV canal is balanced (both ventricles well developed), then depolarization typically inscribes a counterclockwise and superior frontal QRS loop. It’s like left anterior hemiblock that can be produced experimentally in the dog lab. The accession wave travels from inferiorly to superiorly, creating a counterclockwise QRS loop as projected on the frontal plane.

The foregoing was realized many years ago at the Mayo Clinic by a young physician from Mexico City, Dr. Toscano-Barbosa, one of Professor Sodi-Pallares’ bright young men. The foregoing realization made it possible to diagnose common AV canal with the help of scalar electrocardiography, prior to the advent of good angiocardiography.

When you look at the opened left ventricle ( Figs. 11.2C and D ) from a developmental perspective, what do you see? I’ll try to tell you some of the things that I see. Dr. David Kurnit put it to me very nicely years ago: “The fibroblasts of the endocardial cushions of the AV canal take random walks in space, looking for something to attach to.” When they contact the crest of the muscular ventricular septum, short chordae tendineae are formed, which are characteristics of type A completely common AV canal. This is really a normal stage in the morphogenesis of the AV canal which, for reasons still not completely understood (but probably genetic), becomes arrested at this stage of development in the complete form of common AV canal, type A.

Normally, the spaces between these short chordae tendineae then get filled in with fibrous tissue, converting a type A completely common AV canal into a partially common AV canal with no VSD, a cleft mitral valve, and an ostium primum type of atrial septal defect.

Thus, the short chordae tendineae, particularly well seen in Fig. 11.2D , are one of the stages in the normal morphogenesis of the AV canal. As will be seen, other types of common AV canal represent other normal stages in the morphogenesis of the AV septum and of the AV valves. Hence, the spectrum of common AV canal may be viewed as a series of natural experiments illustrating the various stages in the normal development of the AV canal region—a topic to which we shall return.

Type A completely common AV canal often shows you very clearly where the normal definitive mitral and tricuspid valve leaflets come from ( Fig. 11.5 ). In this diagram, the common AV valve is viewed from above. Note the superior cushion component (SCC) and the inferior cushion component (ICC) that together normally form the anterior leaflet of the mitral valve. The superior cushion component and the inferior cushion component are separated by a cleft. The cleft does not run horizontally toward the left ventricular septal surface. Instead, it slants superiorly.

Fig. 11.5, Where do the mitral and tricuspid valve leaflets come from? Common atrioventricular (AV) canal type A is very illuminating in this regard. The AV canal is surrounded by endocardial cushion tissue indicated by cross-hatching. Conventionally, the anterior (A) or superior cushion is said to extend from the anterolateral papillary muscle (ALP) of the left ventricle (LV) to the anterior papillary muscle (AP) of the right ventricle (RV) . The posterior (P) or inferior endocardial cushion is said to lie between the posteromedial papillary muscle (PMP) of the LV and the posterior papillary muscle (PP) of the RV. The left (L) lateral endocardial cushion extends between the ALP and the PMP of the LV. The right (R) lateral endocardial cushion lies between the AP and PP muscles of the RV. These conventional designations of the endocardial cushions are convenient for describing the ring of endocardial cushion tissue that internally lines the AV canal. What does the anterosuperior endocardial cushion give rise to? The superior cushion component (SCC) of the anterior mitral leaflet above the cleft; the membranous septum (MS) ; and the anterior leaflet (AL) of the tricuspid valve. What does the posteroinferior endocardial cushion contribute to? The inferior cushion component (ICC) of the anterior mitral leaflet below the cleft and the septal leaflet (SL) of the tricuspid valve. What does the left lateral endocardial cushion give rise to? The posterior leaflet (PL) or mural leaflet of the mitral valve between the ALP and the PMP muscles of the LV. What does the right lateral endocardial cushion of the AV canal give origin to? To the posterior leaflet (PL) of the tricuspid valve. Thus, common AV canal type A, diagrammed above, is very informative concerning the origins of the leaflets of the definitive normal AV valves. VS, ventricular septum.

Note that the anterior leaflet of the normal mitral valve is composed of two endocardial cushions (the superior and the inferior). The other AV leaflets are formed essentially by one endocardial cushion only.

Why is this anterior mitral leaflet normally a doublet, whereas the other AV valve leaflets are singletons? Our hypothesis is that in order to form a deep semicircular anterior mitral leaflet to occlude the approximately circular systemic AV orifice, the contributions of two endocardial cushions are needed. By contrast, the pulmonic AV orifice (tricuspid) is semicircular (not round) prenatally and is crescentic postnatally. Consequently, a deep semicircular septal leaflet is not needed in the tricuspid valve in order to achieve valvar competence.

In fact, this hypothesis is tested and substantiated by the findings in common AV canal. A normally formed two-component anterior mitral leaflet is not present in common AV canal because of the cleft, that is, the failure of the superior and inferior endocardial cushions to fuse and zipper closed the cleft from medially (at the ventricular septum, VS in Fig. 11.5 ) to laterally at the free margin of the anterior mitral leaflet. Failure of the normal zippering fusion results in the persistence of a cleft that is often associated with significant mitral regurgitation, as will be seen under Findings later in this chapter.

In Fig. 11.5 , note also that the membranous septum (MS) is formed from the superior endocardial cushion. The tricuspid valve also often has a small cleft between the superior and the inferior endocardial cushions. This tricuspid cleft ( Fig. 11.5 ) is seldom of major hemodynamic significance and hence is often ignored.

The anterolateral papillary muscle of the left ventricle (ALP, Fig. 11.2 ) usually is larger than the posteromedial papillary muscle (PMP, Fig. 11.2 ), reflecting the fact that the superior leaflet of the common AV valve typically is larger than the inferior leaflet.

When potentially parachute mitral valve occurs in the setting of common AV canal, typically the anterolateral papillary muscle of the left ventricle is large and receives all of the primary chordae tendineae of the abnormal mitral valve, whereas the posteromedial papillary muscle of the left ventricle is hypoplastic or absent. By potentially parachute mitral valve we mean that a parachute mitral valve would be present postoperatively because there is only a single focus of chordal insertion in the morphologically left ventricle.

When the AV canal is normally divided into mitral and tricuspid valves, parachute mitral valve typically has the opposite papillary muscle anatomy: a large posteromedial papillary muscle and a small or absent anterolateral papillary muscle.

The size of the posterior or mural leaflet of the mitral valve depends on the distance between the anterolateral and posteromedial papillary muscles ( Fig 11.5 ). In potentially parachute mitral valve, there is no discrete posterior or mural leaflet of the mitral valve, and the interchordal spaces are often poorly formed or absent. Hence the cleft typically is the major component of the mitral orifice and hence should not be sutured closed, in order to avoid or minimize iatrogenic mitral stenosis postoperatively.

Note also that normal aortic-mitral fibrous continuity with normally related great arteries is really aortic-to-superior endocardial cushion direct fibrous continuity ( Fig. 11.2D ).

The tricuspid cleft is about two-thirds of the way up as one proceeds from inferiorly to superiorly; that is, the tricuspid cleft lies below (or inferior to) the superior commissure of the tricuspid valve between the anterior leaflet (AL) and the septal leaflet (SL) ( Fig. 11.5 ).

The septal leaflet of the tricuspid valve (SL, Fig. 11.5 ) consists mostly of inferior endocardial cushion tissue—below the tricuspid cleft. Above the cleft lies the membranous septum (MS, Fig. 11.5 ), which consists of superior endocardial cushion tissue. In the normal definitive tricuspid valve, the membranous septum may or may not be covered by tricuspid septal leaflet tissue. Thus, although the septal leaflet of the tricuspid valve normally consists mostly of inferior endocardial cushion or leaflet tissue—below the tricuspid cleft or fusion line ( Fig. 11.5 ), the septal leaflet of the tricuspid valve may or may not have a contribution from the superior endocardial cushion—above the tricuspid cleft in completely common AV canal type A ( Fig. 11.5 ), or when development is normal—above the tricuspid fusion line (closed cleft).

Thus, careful study of the common AV valve in type A from above helps one to understand where all of the components of the normal mitral and tricuspid valve leaflets come from ( Fig. 11.5 ).

Grammatical Note: Please observe that we are talking about completely common AV canal, not complete common AV canal, and partially common AV canal, not partial common AV canal. Why? Because, as your elementary school grammar teacher no doubt taught you, an adverb (completely, or partially) is required to modify an adjective (common). One adjective cannot modify another adjective.

Since it has been a long time since most of us were in elementary school, let’s briefly review the grammar: completely (adverb) common (adjective) atrioventricular (adjective) canal (noun). The adverb completely modifies the adjective common . The adjectives common and atrioventricular both modify the noun canal . The meaning is not whether the canal is complete or partial. Instead, the meaning is whether the canal is completely common, or only partly common, that is, completely undivided, or only partly undivided.

By contrast, complete and partial —the adjectives—are grammatically correct in complete AV defect and partial AV defect, because both adjectives are modifying the noun defect.

However, grammatically it should be morphologically right ventricle, not morphologic right ventricle, and morphologically left ventricle, not morphologic left ventricle. Why? Because the adverb (morphologically) is required to modify the adjective (right, or left). The question is: Is this ventricle morphologically right, or morphologically left? The meaning is not: Is this structure a morphologic ventricle—as opposed, for example, to a morphologic atrium?

The foregoing note concerning adjectives and adverbs is written in the hope that it may help to improve our terminology.

Type B Completely Common AV Canal

Let us now consider type B completely common AV canal in greater detail. The superior (anterior) leaflet is divided partially ( Fig. 11.3 ), not completely as in type A ( Fig. 11.2 ). Also note that the partial division of the superior leaflet in type B occurs above the right ventricular cavity ( Fig. 11.3 )—further to the right than the complete division of the superior leaflet in type A that occurs above the ventricular septum ( Fig. 11.2 ). Perhaps this is why the margins of the incomplete division of the superior leaflet are not tethered to the crest of the ventricular septum in type B ( Fig. 11.3 ), as they are in type A ( Fig. 11.2 ).

In the diagram of Rastelli et al ( Fig. 11.3a ), the inset confirms that there are no attachments between the partly divided superior leaflet (A,A) and the ventricular septal crest. Normally, when the leftward A and the leftward part of P fuse, the composite anterior leaflet of the MV is formed. L becomes the posterior leaflet of the mitral valve. The rightward A becomes the anterior leaflet of the TV. The rightward portion of P forms the septal leaflet of the TV. The posterior leaflet of the potential tricuspid valve is not shown in Fig. 11.3A . Note that an ostium primum type of atrial septal defect (above the leaflets of the common AV valve) is confluent with a ventricular defect of the AV canal type (below the leaflets of the common AV valve); hence a complete AV septal defect is present, indicating that this is a complete form of common AV canal.

As indicated above, we found that Rastelli’s right ventricular muscle that originates from the right ventricular septal surface is indeed the anterior papillary muscle of the right ventricle which is associated with a very short moderator band ( Fig. 11.3B ). This creates the impression that the anterior papillary muscle is originating directly from the right ventricular septal surface. As will be confirmed in the Findings section, type B completely common AV canal is rare.

Type C Completely Common AV Canal

Let us now have a somewhat more detailed look at the pathologic anatomy of type C completely common AV canal ( Fig. 11.4 ). As the inset in Rastelli’s drawing shows, there were no fibrous attachments between the superior (anterior) leaflet of the common AV valve (A, Fig. 11.4A ) and the underlying ventricular septum. In type C, the superior leaflet is a free-floating, bridging leaflet: it is like a bridge arching unsupported over the ventricular septum.

In the photograph of a heart specimen of type C ( Fig. 11.4C ), note that there is one lone chorda tendinea running from the central surface of the superior leaflet to the ventricular septal crest. Nonetheless, this case was still classified as a type C because the superior leaflet was undivided, and no chordae ran from the free margin of the superior leaflet to insert into the ventricular septal crest.

In the opened view of the left ventricle ( Fig. 11.4C ), note that the anterolateral papillary muscle (the upper one) is distinctly larger than the posteromedial (the lower one), which may be due to the fact that the superior leaflet of the common AV valve (SL, Fig. 11.4C ) is much larger than the inferior leaflet (IL, Fig. 11.4C ). The superior leaflet of the common AV valve inserts only into the anterolateral papillary muscle of the left ventricle, and the inferior leaflet inserts only into the posteromedial papillary muscle, these being the usual leaflet attachments in the complete form of common AV canal—but by no means the only ones that occur, as we shall see.

Also noteworthy in Fig. 11.4C : the “scooped-out” (concave) crest of the muscular ventricular septum; the reduced left ventricular inlet length/outlet length ratio (0.83), the normal being approximately 1.0; and the aortic valve-to-superior leaflet direct fibrous continuity, typical of normally related great arteries. Note also that the inferior leaflet of the common AV valve is much closer to the crest of the muscular ventricular septum that is the superior leaflet: this is a point of morphogenetic significance, to which we shall return.

In Fig. 11.4B , one can see the ostium primum defect well. This brings up the question: Is an ostium primum defect really an atrial septal defect? The answer, we think, is both yes and no. Yes, it has long been regarded as an atrial septal defect because the defect lies above the leaflets of the AV valve(s) ( Fig. 11.4B ). No, because an ostium primum defect is really a defect in the septum of the AV canal; that is, it is a partial AV septal defect. The AV septum lies in front of (or below) the atrial septum, and behind (or above) the ventricular septum ( Fig. 11.4B ). This is why the AV septum used to be known as the septum intermedium —because it is intermediate between the atrial septum behind (or above) and the ventricular septum in front (or below). Hence, accurately speaking, an ostium primum defect is not really an atrial septal defect. Instead, it is a partial atrioventricular septal defect. This is why we prefer to speak of an ostium primum defect (not an ostium primum type of atrial septal defect).

However, the term ostium primum type of atrial septal defect (ASD I) is deeply ingrained by common usage. The question becomes: Is this time-honored designation really wrong? We think the answer is no, in the following sense. From an anatomic standpoint, there are three main developmental components to the anatomic atrial septum: septum primum, septum secundum, and the interatrial part of the atrioventricular septum. When the latter is absent, the ostium that results resembles the first opening between the atria, prior to the development of the atrial septum in the embryo. Hence, this defect has been called an ostium primum type of ASD—because it is reminiscent of the first hole (ostium primum), the original interatrial foramen prior to atrial septation.

So, whenever we use the term ASD I, we do so with the understanding that it is a partial AV septal defect, the presence of which results anatomically in a typical low defect in the interatrial septum.

Is the atrioventricular septum in front of the atrial septum, or below it ( Fig. 11.4B )? Is the atrioventricular septum behind the ventricular septum, or above it ( Fig. 11.4B )? The answer depends on the position of the heart. In the fetus, the newborn, and young children, the cardiac position typically is horizontal, with the atria behind and the ventricles in front. Hence in the infant and child, the AV septum lies in front of the atrial septum and behind the ventricular septum ( Fig. 11.4B ).

However, with growth—when one becomes an adolescent or an adult—the heart position becomes semivertical or vertical. The ventricular apex no longer points anteriorly and to the left; instead, it points inferiorly and to the left. This is how the atria come to be above the ventricles. Hence, in an adolescent or an adult, and AV septal defect lies below the atria and above the ventricles. Similarly, the superior bridging leaflet of an infant or child with type C completely common AV canal becomes the anterior bridging leaflet of an adult patient. The inferior leaflet of the infant and child becomes the posterior leaflet of the adolescent and adult ( Fig. 11.4 ). Hence, heart position is an important consideration for accurate cardiac description.

Partial Forms of Common AV Canal

The most frequent partial form of common AV canal is ostium primum atrial septal defect with cleft mitral valve ( Fig. 11.6 ). The VSD of the AV canal type has been obliterated, but the ventricular septum is still “scooped-out” or deficient superiorly ( Fig. 11.6B ).

Fig. 11.6, Partial form of common atrioventricular (AV) canal with an ostium primum type of atrial septal defect (ASD I) , no ventricular septal defect (VSD) of the AV canal type, and a cleft mitral valve. (A) Right lateral view of opened right atrium (RA) and right ventricle (RV) . The ASD I is a partial AV septal defect extending from the bare lower rim of the atrial septum posterosuperiorly to the septal leaflet (SL) of the tricuspid valve (TV) anteroinferiorly. The anterior leaflet (AL) and the SL of the TV are fused at the superior commissure. There is no VSD beneath the SL of the TV. (B) Partial form of common AV canal, left ventricular (LV) view of another case (not the same patient as in A). There is a relatively large gap (more than a cleft) between the superior leaflet component (SLC) and the inferior leaflet component (ILC) of the unfused and tissue deficient anterior leaflet of the mitral valve. The thickening and rolling of the leaflet margins about the huge cleft (gap) in the anterior leaflet of the MV indicate severe mitral regurgitation. The posterior leaflet (PL) of the MV is attached to a third papillary muscle between the anterolateral papillary muscle above and the posteromedial papillary muscle below. There is no VSD; the ventricular septum (VS) is “scooped out,” that is, tissue deficient superiorly; and the LV inlet/outlet ratio is very subnormal (0.63, normal = 1.0). FW, Free wall.

Let us examine such a case in somewhat greater detail, beginning on the right side. A large ostium primum type of atrial septal defect is seen between the intact atrial septum above and behind, and the intact ventricular septum below and in front ( Fig. 11.6A ). The septal leaflet of the tricuspid valve is firmly adherent to the crest of the ventricular septum, and both the anterior and the posterior leaflets of the tricuspid valve appear unremarkable.

The bare or nude anteroinferior margin of the atrial septum, which does not have its tricuspid valve “pants” on, is the posterosuperior margin of the partial AV septal defect (the ostium primum defect) ( Fig. 11.6A ). This nude anteroinferior margin of the atrial septum is a highly reliable marker that a common AV canal (an AV septal defect) is present.

Is this a complete or a partial form of common AV canal? Look under the septal leaflet of the tricuspid valve. Absence of a VSD of the AV canal type beneath the septal leaflet of the tricuspid valve indicates that a partial form of common AV canal (a partial AV septal defect) is present. The tricuspid valve’s septal leaflet forms the anteroinferior margin of this partial AV septal defect. (When the AV septal defect is complete, the anteroinferior margin of the defect is the muscular crest of the ventricular septum, as in Figs. 11.2 to 11.4 ).

An ostium primum type of defect is the only type of interatrial communication that typically is confluent with the septal leaflet of the tricuspid valve ( Fig. 11.6A ). The ostium secundum type of atrial septal defect ( Fig. 11.3B ) is located in the central portion of the atrial septum, and is not confluent with the septal leaflet of the tricuspid valve.

Is this tricuspid valve cleft ( Fig. 11.6A )? No. There is no gap between the septal and the anterior leaflets of the tricuspid valve in this case, as there can be in other cases.

As Dr. Jesse Edwards pointed out, in the partial form of common AV canal, the superior (anterior) and the inferior (posterior) leaflets of the common AV valve have fused in the midline and have obliterated the VSD.

Now let us examine the opened left ventricle of this case ( Fig. 11.6B ). Note that the superior leaflet component (SLC) of what normally becomes the anterior mitral leaflet inserts exclusively into the anterolateral papillary muscle (the upper one, Fig. 11.6B ). The inferior leaflet component (ILC) of what normally becomes the anterior leaflet of the mitral valve inserts exclusively into the inferior papillary muscle (the lower one, Fig. 11.6B ). The posterior leaflet (PL) or mural leaflet inserts into a third papillary muscle that we call the middle papillary muscle.

The middle papillary muscle of the left ventricle lies between the anterolateral papillary muscle above and the posteromedial papillary muscle below ( Fig. 11.6B ). The middle papillary muscle normally is attached by chordae tendineae to the posterior (mural) leaflet only. In one form of double-orifice mitral valve, chordae tendineae run from the middle papillary muscle to both the posterior and the anterior mitral leaflets, thereby subdividing the mitral orifice into two approximately equal-sized openings.

The middle papillary muscle is noteworthy for two reasons:

1.
It is a normal left ventricular structure.
2.
Its existence is largely unknown.

The conventional description is that there are two papillary muscles in the left ventricle: the anterolateral and the posteromedial. In fact, there are three papillary muscles in the left ventricle: the anterolateral, the posteromedial, and the middle ( Fig. 11.6B ). However, the middle papillary muscle can be fused with the anterolateral or the posteromedial papillary muscle, obscuring the independent existence of the middle papillary muscle. This may well be why people often speak of the anterolateral or the posteromedial papillary muscle group . For example, the anterolateral papillary muscle group suggests that more than the anterolateral papillary muscle is present. The additional papillary musculature can be the middle papillary muscle—when the anterolateral and the middle papillary muscles are confluent.

The middle papillary muscle is seen with unusual clarity in Fig. 11.6B because of the coexistence of incompletely common AV canal with a large gap (more than a cleft) in the anterior mitral leaflet. This tissue-deficient gap in the anterior mitral leaflet resulted in severe congenital mitral regurgitation. Note the thickening and rolling of the free margins of the superior and inferior leaflet components of the anterior mitral leaflet above and below this very large cleft or gap. This thickening or rolling of the leaflet free margins indicates the presence of congenital mitral regurgitation ( Fig. 11.6B ).

A partial cleft in the anterior mitral leaflet is presented in Fig. 11.7 . The patient was a 35-year-old woman with Down syndrome who had an ostium primum type of atrial septal defect, a partially cleft anterior leaflet of the mitral valve, and no ventricular septal defect. Note that the superior component (SC) and the inferior component (IC) are fused paraseptally (F). The cleft is partial because it does not extend all the way from the free margin of what should normally have been the anterior mitral leaflet to the left ventricular septal surface. This heart specimen illustrates with unusual clarity that the anterior leaflet of the mitral valve normally is a composite structure, and how the zippering closed of the cleft normally occurs, beginning at the left ventricular septal surface medially, and proceeding laterally to the free margins of the superior and inferior components of the definitive (fused) anterior leaflet of the mitral valve ( Fig. 11.7 ).

Fig. 11.7, Partial cleft of the anterior leaflet of the mitral valve, left ventricular (LV) view. The patient was a 35-year-old woman with Down syndrome, an ostium primum atrial septal defect, and a partially cleft anterior leaflet of the mitral valve. The cleft is partial because it does not extend from what normally is the free margin of the anterior mitral leaflet all the way to the left ventricular septal surface (VS) . The superior component (SC) and the inferior component (IC) of what normally is the anterior mitral leaflet are fused paraseptally (F) . The anterior mitral leaflet is a composite structure. Normally, the AML = SC + IC. Cleft closure normally occurs from medially (paraseptally) to laterally (at the free margin of the AML).

Are we looking at a cleft, or at a commissure ( Fig. 11.7 )? At a cleft. Why? Because the anterior mitral leaflet (AML) normally is composed of the superior endocardial cushion component (SC) and the inferior endocardial cushion component (IC), as partial fusion of the cleft proves ( Fig. 11.7 ): briefly, AML = SC + IC. The heart specimen shown in Fig. 11.7 is a very fortunate natural experiment: It shows the normal process of cleft closure, only partly completed, indicating that this is how the anterior mitral leaflet normally forms, and also showing what happens when fusion of the superior and inferior endocardial cushions fails to occur: a cleft results.

Note also that the superior component and the inferior component of the anterior mitral leaflet are pulled away from each other during ventricular systole, because the superior component is attached to the anterolateral papillary muscle (the upper one) and the inferior component is attached to the posteromedial papillary muscle (the lower one) ( Fig. 11.7 ). Thus, the leaflets bounding a cleft are pulled away from each other during ventricular systole, a feature that may predispose to mitral regurgitation.

By contrast, at a commissure—such as the anterolateral or posteromedial commissure of the normal mitral valve—the leaflets on either side of the commissure are pulled together, helping to prevent mitral regurgitation, because both leaflets are attached to the same papillary muscle group. This is what commissure literally means. Commissure comes to us from commissura (Latin), which is derived from commissus, the past participle of committere, which means to bring together: com- or cum-, together or with, + mittere , to send. Hence etymologically, commissure means “to send together.” That is what happens to the AV valve leaflets at a commissure between leaflets. By contrast, a cleft is a defect within a leaflet, where the leaflet components are pulled apart—in diverging directions during ventricular systole, not in a converging direction as occurs at a commissure.

Consequently, in the complete and partial forms of common AV canal, we think that the left-sided AV valve is a malformed mitral valve, not a trileaflet or tricuspid nonmitral valve.

Where does the idea that the mitral cleft is really a commissure come from? This line of thought runs essentially as follows: Let’s look at the left-sided AV valve, particularly in the incomplete form of common AV valve. Anatomically, we see three leaflets ( Fig. 11.6B ). We don’t really know anything about embryology. We are not going to be misled by embryologic theories. We are going to stick with the pathologic anatomy that we can see and know for sure. In incompletely common AV canal, we see three leaflets. Hence, the so-called cleft we regard as a commissure in a trileaflet left-sided AV valve.

What’s wrong with this view? Why is it wrong? Why not rely on what we can see with our own eyes, and forget embryologic theories?

The problem with the aforementioned view is that the embryology of the AV canal region is well documented and well known ( Figs. 11.8 and 11.9 ). In the normal human embryo, the AV canal is in common during Streeter’s horizon (stage) XV (estimated gestational age 30 to 32 days) and horizon XVI (estimated gestational age 32 to 34 days), as is shown by Asami’s microdissections ( Fig. 11.8 ). ^, In horizon XVII (estimated gestational age 34 to 36 days), the superior and inferior endocardial cushions of the AV canal normally fuse ( Fig. 11.8 ), dividing the common AV canal into mitral and tricuspid canals. The same process of dividing the common AV canal into mitral and tricuspid canals also normally occurs in the rat between 13.75 and 14.5 days of gestational age ( Fig. 11.9 ). ^,

Fig. 11.8, Septation of the atrioventricular (AV) canal in human embryos from Streeter’s horizon (stage) XV (30 to 32 days of age) to Streeter’s horizon XVIII (36 to 38 days of age). The atria have been removed to permit visualization of the AV valve(s). The truncus arteriosus or great arteries have been removed to facilitate viewing of the semilunar valve(s). The hearts are viewed from behind (dorsally) and slightly from above (somewhat cranially); the superior surfaces face the top of the photograph; the inferior surfaces face the bottom of the photograph; and the right and left sides of these embryonic hearts are on the viewer’s right right-hand side and left-hand side, respectively. The AV canal normally remains in common (undivided) until horizon XVII (34 to 36 days of age following ovulation), when the superior and inferior endocardial cushions of the AV canal fuse, which is part of the process of dividing the common AV canal into mitral and tricuspid canals. Late in horizon XVI, that is, in horizon XVIb (33 to 34 days of age), the common AV valve opens predominantly into the left ventricle and only slightly into the right ventricle. Even by horizon XVIII (36 to 38 days of age), the anterior leaflet of the mitral valve of this embryo remains cleft. Note also the associated morphogenetic movements of the developing pulmonary valve (p) and aortic valve (ao) . The aortic valve moves almost not at all, except that it turns to keep on facing the anteriorly moving pulmonary valve. In horizon XV (30 to 32 days), the pulmonary part of the undivided truncal valve is posterior and to the left of the aortic part of the truncal valve, similar to the semilunar interrelationship in D-transposition of the great arteries. Early in horizon XVI, that is, XVIa (32 to 33 days of age), the pulmonary valve (p) has moved anteriorly on the left and is now side-by-side relative to the aortic valve (ao), similar to the semilunar interrelationship in the Taussig-Bing type of double-outlet right ventricle. In the second half of horizon XVI, that is, XVIb (33 to 34 days of age), the pulmonary valve is now mildly anterior and left-sided relative to the aortic valve, similar to the semilunar relationship seen in tetralogy of Fallot with type C completely common AV canal. By horizon XVII (34 to 36 days of age), the common AV valve is divided into mitral and tricuspid valves, and the semilunar interrelationship is as in solitus normally related to great arteries. Maldevelopment of the AV canal and of the conotruncus often appear to be linked, as in the heterotaxy syndrome with asplenia, in which the findings resemble those normally found in horizons XV and XVI (30 to 34 days of age). However, the linkage between AV canal and conotruncal development often appears to be broken in the heterotaxy syndrome with polysplenia: the great arteries typically are normally related, or nearly normally related as in tetralogy of Fallot. But common AV canal, which may be of the partial type, is present. Thus, our hypothesis is that the heterotaxy syndrome with asplenia represents an earlier developmental arrest, resembling normal developmental stages XV and XVIa (30 to 33 days); whereas heterotaxy with polysplenia represents a slightly later developmental arrest, resembling normal horizon XVIb and perhaps early horizon XVII (33 days to early day 34 in normal development). It is understood that development in the heterotaxy syndromes with asplenia or polysplenia is not normal, and hence the above-mentioned timing based on normal development may well not be accurate.

Fig. 11.9, Septation of the AV canal and morphogenetic movements of the conotruncus in the rat, dorsal view, same orientation as in Fig. 11.8 . (A) 12.5 days of age; (B) 12.75 days; (C) 13 days; (D) 13.75 days; (E) 14.5 days; (F) 15.5 days. When the pulmonary valve (p) has moved from posteriorly to anteriorly to the left of the aortic valve (ao) , the pulmonary valve riding atop the developing subpulmonary conus and achieving normally related great arteries, only then do the superior and inferior endocardial cushions of the AV canal fuse, dividing the common AV canal into mitral and tricuspid canals. This linked development of the AV canal and conotruncus normally is accomplished in the rat embryo by 13.75 days (D). Although septation of the common AV canal has begun, the anterior leaflet of the mitral valve is still cleft. However, by 14.5 days of age (E), the mitral cleft has disappeared. Sap, Aortopulmonary septum.

In the human embryo, the mitral valve appears somewhat cleft even during horizon XVIII (estimated gestational age 36 to 38 days) ( Fig. 11.8 ). ^, In the rat, the mitral cleft is clearly visible by 13.75 days, but has been fused closed by 14.5 days ( Fig. 11.9 ). ^,

Hence, the embryology of septation of the common AV canal and of cleft closure of the mitral valve has been well documented in human and in comparative mammalian embryos, and is now well established and not controversial. We still have a lot to learn concerning exactly how these fusional processes occur, plus how and why they may go awry. But the phenomenology—what happens normally, and what can happen abnormally—is not hypothetical.

The foregoing is why it can no longer be claimed that we know nothing about the relevant embryology and therefore that we must rely only on what our eyes can see of the pathologic anatomy. This attitude is what the late Dr. Tomas Pexeider called “declaring war on embryology.” He regarded this as a very misguided approach, and we agree. The function of embryology is to make pathologic anatomy readily and accurately comprehensible. The dream of our field has long been to build an accurate bridge from the genome to the operating room. Embryology is an essential part of this bridge. Embryology is to pathology as vectorcardiography is to electrocardiography: both embryology and vectorcardiography are means of understanding.

Surgically, one can justify a failure to suture closed the cleft in the anterior leaflet of the mitral valve in the incomplete form of common AV canal, particularly if the cleft is not regurgitating at the time of surgery, by claiming that the left-sided AV valve is really a trileaflet valve and not a malformed mitral valve. One could even assert that one should not suture the “cleft” closed, because it is really a commissure. However, our surgical colleagues at Children’s Hospital Boston have found that if one does not suture the cleft at the time of initial reparative surgery because it is not regurgitating, later on one often wishes one had closed the cleft—because the development of late mitral regurgitation in this situation is frequent. We are not surprised at this course of events because the view that the mitral cleft is really a commissure is anatomically and embryologically incorrect, for the reasons presented heretofore ( Figs. 11.6 to 11.9 ).

Embryology

The normal morphogenesis of the atrioventricular canal in man and in the rat has been illustrated by the elegant microdissections of Asami ( Figs. 11.8 and 11.9 ). ^,

An understanding of the normal and abnormal morphogenesis of the atrioventricular canal can be augmented by a consideration of the spatial geometry of the superior and inferior endocardial cushions ( Fig. 11.10 ).

Fig. 11.10, Spatial geometry and the morphogenesis of common atrioventricular (AV) canal: diagram of the complete form of common AV canal, viewed from the left. Note that the inferior endocardial cushion (IC) of the AV canal is closer to the crest of the muscular ventricular septum (VS) than is the superior endocardial cushion (SC) . Both endocardial cushions are much closer to the crest of the VS than they are to the anteroinferior rim of the atrial septum (AS) . Our hypothesis is that these spatial geometric considerations are important determinants of the normal sequence of morphogenesis and the classification of common AV canal: (1) Because IC is closest to the VS, the VSD beneath the IC is obliterated first. (2) Because the SC is farther away from the VS, the VSD beneath the SC is obliterated as the second step in normal morphogenesis. In this way, the complete form of common AV canal is converted to the most frequent partial form of common AV canal. (3) Because the SC and the IC are so much farther away from the AS than from the VS, closure of the ostium primum type of atrial septal defect and closure of the cleft in the anterior leaflet of the mitral valve together constitute the third and last step in the normal morphogenesis of the septation of the AV canal and the formation of the mitral valve. Closure of the VSD component of the AV septal defect may well stabilize the SC and IC relative to each other, facilitating cleft closure from medially to laterally, despite the fact that the SC and the IC are moving back and forth 140 to 160 times/minute because of the heartbeat. LA, Left atrium; LV, left ventricle; Sept 1, septum primum; AV septal defect shown in black.

The superior and inferior endocardial cushions (leaflets) of the common AV valve form an anteriorly pointing V-shaped wedge ( Figs. 11.2C and D , 11.4C , 11.6B, and 11.7 ). The superior and inferior endocardial cushions are not oriented supero-inferiorly, like two little bricks standing on their ends, as they are often portrayed ( Fig. 11.1 ).

The inferior endocardial cushion is much closer to the crest of the muscular interventricular septum than is the superior endocardial cushion ( Figs. 11.2C and D , 11.4C ).

Let us call the distance between the inferior cushion and the ventricular septal crest χ ( Fig. 11.10 ). The distance between the superior endocardial cushion and the ventricular septal crest may then be approximately 2χ to 3χ ( Fig. 11.10 ).

The distance between the superior and inferior endocardial cushions and the lower margin of the atrial septum posterosuperiorly is much greater than either χ or 2χ to 3χ. Let us call this distance between the endocardial cushions and the atrial septum approximately 5χ to 10χ.

Now let us consider the above-mentioned spatial geometry and its influence on the normal and abnormal morphogensis of the AV canal ( Fig. 11.10 ):

1.
When the fibroblasts of the endocardial cushions of the AV canal take their random “walks” in space, these fibroblasts first encounter the crest of the muscular ventricular septum beneath the inferior endocardial cushion—because this is the shortest distance (χ) between the endocardial cushions and either septum. So, obliteration of the VSD beneath the inferior endocardial cushion is the first step in the formation of the AV septum. This spatial geometry also explains why many cases of completely common AV canal have no VSD component beneath the inferior leaflet of the common AV valve—only beneath the superior leaflet.
2.
Fibroblasts from the superior endocardial cushion of the AV canal then encounter the muscular ventricular septal crest and obliterate this greater space (2χ to 3χ). This second step completes the closure of the VSD of the AV canal type. A completely common AV canal has now been converted into a partially common AV canal—with an ostium primum type of atrial septal defect, a cleft anterior leaflet of the mitral valve, and an intact ventricular septum. In other words, a complete AV septal defect has now become a partial AV septal defect with a cleft anterior mitral leaflet.
3.
Step 3 in the normal morphogenesis of the AV canal has two parts that appear to occur approximately simultaneously: (a) cleft closure, and (b) closure of the ostium primum type of atrial septal defect. Let us consider each part of step 3 in turn.
- a.
  Why does cleft closure occur now? It must be remembered that the superior and inferior endocardial cushions (leaflets) are moving back and forth between 130 and 150 times per minute because of the heartbeat. How can the cleft possibly close when the superior and inferior cushion components are moving back and forth an average of 140 beats/minute? Our hypothesis is that once the VSD has been closed by fibrous tissue, then the superior and inferior endocardial cushion components have been anchored and stabilized relative to each other. Although both cushions move rapidly back and forth because of the heartbeat, they now are anchored medially at the septum, and consequently the superior and inferior leaflet components move very little relative to each other. This minimal movement of the superior and inferior cushions relative to each other permits the medial-to-lateral “zippering” process of cleft fusion to occur. Since cleft closure appears to depend on the great reduction or elimination of significant movement of the superior and inferior cushions relative to each other, closure or near closure of the AV septal defect may occur before cleft closure; that is, in addition to the closure of the VSD component, closure of much if not all of the ASD component may well be necessary to stabilize the superior and inferior cushions relative to each other. Note that in man (Fig.11.8) and in the rat ( Fig. 11.9 ), the mitral cleft persists after fusion of the superior and inferior endocardial cushions of the AV canal. If indeed mitral cleft closure is the last act in this drama, as Figs. 11.8 and 11.9 suggest, such timing would help to explain the existence of isolated cleft of the mitral valve of the AV canal type, that is, isolated cleft of the anterior leaflet of the mitral valve with no septal defect. This is a topic to which we shall return.
- b.
  At approximately the same time, the relatively large ostium primum type of atrial septal defect (5χ to 10χ) is being closed as the posteriorly growing fibroblasts from the endocardial AV cushions encounter and fuse with the anterorinferior margin of the atrial septum. Why is closure of the ostium primum part of step 3 perhaps the last part of the septational process? Again, we think that spatial geometry is key ( Fig. 11.10 ): the anteriorly pointing V-shaped wedge of the endocardial cushions of the AV canal is much farther away from the atrial septum than from the ventricular septum.

To summarize, the three normal morphogentic steps in the development of the AV canal—that become arrested in persistent common AV canal—are as follows ( Figs. 11.8, 11.9, and 11.10 ):

1.
inferior VSD closure; then
2.
superior VSD closure; then
3.
ostium primum ASD closure and mitral cleft closure (approximately simultaneously).

It should be added that this usual morphogenetic sequence can occasionally get out of order, as in VSD of the AV canal type, with or without straddling tricuspid valve. Usually, closure of the AV canal type of VSD is the first step in the morphogenetic sequence. But in VSD of the AV canal type with or without straddling tricuspid valve, step one in the usual morphogenetic sequence fails to occur, while the subsequent steps are accomplished normally (normal AV valves and no ostium primum type of atrial septal defect). Hence, although the aforementioned morphogenetic sequence is the usual progression of developmental events, it is by no means invariable.

It is fascinating to note that the foregoing morphogentic sequence helps to explain the pathologic anatomic classification of common AV canal:

1.
In the complete form of common AV canal, step one (inferior VSD closure) may or may not have occurred.
2.
When steps one and two have been accomplished (complete VSD closure), then the most frequent form of partially common AV canal has been reached (ostium primum ASD with cleft mitral valve).
3.
Both parts of morphogenetic step three may not be completed, resulting in isolated mitral cleft (of the AV canal type—other types do exist); or resulting in isolated ostium primum ASD (without cleft mitral valve and with intact ventricular septum). The latter is very rare.

It is also interesting to understand that the Rastelli classification of the complete type of common AV canal is backwards from a developmental standpoint:

Type C is the most primitive, with its undivided and unattached anterosuperior leaflet ( Fig. 11.4 ).
Type B is slightly more advanced, the anterosuperior leaflet being partly divided, but unattached to the ventricular septum ( Fig.11.3 ).
Type A is even more advanced, the anterosuperior leaflet being divided and attached by short chordae tendineae to the ventricular septal crest ( Fig. 11.2 ).

Ironically, Rastelli, Kirklin, and Titus did not intend to classify the complete form of common AV canal into types A, B, and C. Instead, their photographic figures 2 and 4 were labeled A, B, and C. And so, inadvertently, the medical artist who labeled these photographs made history because subsequently, these were known as types A, B, and C of the complete form of common AV canal.

Etiology

An increase in cell-surface adhesiveness ^, has been proposed as a possible etiologic factor in the common AV canal that occurs with Down syndrome. The idea essentially is that when the endocardial cushion fibroblasts take their random “walks” in space, if they have an increase in cell surface adhesiveness, these fibroblasts will tend to stick together too much, will migrate out subnormally, and hence will fail to form a normal AV septum and will also fail to close the cleft between the superior endocardial cushion component and the inferior endocardial cushion component of what should normally develop into the anterior leaflet of the mitral valve.

Cardiac morphogenesis appears to be strongly influenced by genetics and also by chance. The stochastic (probabilistic) single-gene hypothesis now appears more appealing than does the multifactorial polygenic hypothesis. Identical twins with Down syndrome, but discordant for congenital heart disease, are more the rule than the exception. This suggests that although genetic factors are very important, random (chance, or stochastic) events also appear to play a significant role in determining the organic phenotype.

The importance of genetics is emphasized by the work of Korenberg and colleagues. ^, These investigators think that the Down syndrome congenital heart disease genes are located in a small region at the distal end of chromosome 21, from q22.1 to q22.3.

Is the foregoing relevant to the etiology of common AV canal? The answer is yes, because common AV canal is by far the most frequent form of congenital heart disease associated with Down syndrome . In 100 randomly selected cases of Down syndrome with congenital heart disease, common AV canal occurred in 63%: The complete form of common AV canal was found in 52% (type A in 29% and type C in 23%) and a partial form of common AV canal was present in 11%.

Comparison of common AV canal with Down syndrome versus common AV canal without Down syndrome revealed several statistically highly significant differences:

1.
The complete form of common AV canal was much more frequent in patients with Down syndrome (83%) than in patients without Down syndrome (45%) ( p < .001).
2.
Conversely, the partial form of common AV canal was much more frequent in patients without Down syndrome (55%) than in patients with Down syndrome (17%) ( p < .001).
3.
Type A completely common AV canal was more frequent in patients without Down syndrome (83%) than in those with Down syndrome (56%) ( p < .005).
4.
Type C completely common AV canal was much more frequent in patients with Down syndrome (44%) than in those without Down syndrome (17%) ( p < .005).
5.
Hypoplastic left ventricle (i.e., the hypoplastic left heart syndrome) was more frequent in patients without Down syndrome (19%) than in those with Down syndrome (8%) ( p < .01).
6.
Left ventricular outflow tract obstruction was more frequent in individuals without Down syndrome (35%) than in those with Down syndrome (6%) ( p < .001).

Thus, common AV canal with Down syndrome was more primitive than common AV canal without Down syndrome. Down syndrome cases had significantly higher prevalences of complete (as opposed to partial) canals, and type C (as opposed to type A).

There are two major groups of common AV canal:

1.
nonsyndromic common AV canal, that is, not associated with any identified syndrome; and
2.
syndromic common AV canal, that is, an integral part of an identified syndrome such as Down syndrome or the heterotaxy syndromes with asplenia, with polysplenia, and with a normally formed but right-sided spleen.

Regarding syndromic common AV canal, we have never seen Down syndrome in association with situs inversus totalis, nor with the heterotaxy syndromes with asplenia or polysplenia. Trisomy 21 has many pernicious effects, but there is at least one good thing about it: Trisomy 21 appears to guarantee situs solitus (the normal pattern of anatomic organization) at all segmental levels—visceroatrial situs, AV valves, ventricles, infundibulum (conus), and great arteries. The segmental situs set is always {S,D,S}: situs solitus of viscera and atria (S), D-loop ventricles (D), and solitus normally related great arteries (S).

To the best of our present knowledge, this means that Down syndrome never has atrial inversion or atrial situs ambiguus, ventricular inversion (L-loop formation), or a well-developed subaortic conus with transposition of the great arteries or double-outlet right ventricle. (We are continuing to search for exceptions to this generalization.) Down syndrome can have underdevelopment of the subpulmonary conus and its sequelae, that is, tetralogy of Fallot, and a lot of aortic overriding can lead to the diagnosis of double-outlet right ventricle. But double-outlet right ventricle with a well-developed muscular subaortic conus or a well-developed bilateral conus (subaortic and subpulmonary) we have never seen with Down syndrome. In other words, the conus is always subpulmonary─normally developed or underdeveloped (tetralogy of Fallot).

Other trisomies, such as 13-15 and 17-18, also seem always to have situs solitus at all segmental levels.

The other major form of syndromic common AV canal, that is, the heterotaxy syndromes with asplenia and polysplenia, is the exact opposite. Segmental situs discordance is the rule: they are a segmental situs salad or potpourri. One should expect anything: the atrial situs may be solitus, inversus, or ambiguus (undiagnosed).

Atrial isomerism—right isomerism or left isomerism—is an erroneous concept . Accurately speaking, atrial isomerism does not exist. See Chapter 29 for more details. D-loop ventricles or L-loop ventricles may be present. The conus may be subpulmonary, subaortic, or bilateral. The great arteries may be normally related or transposed, or double-outlet right ventricle may be present.

These anatomic findings suggest that trisomies such as Down syndrome may result in a genetic “overdosage.” Despite their obviously deleterious effects, chromosomal trisomies appear to guarantee normal segmental situs at all levels. In other words, trisomies are characterized by visceroatrial situs solitus, segmental situs concordance at all levels, and concordant atrioventricular and ventriculoarterial alignments. Such genetic “overdosage” appears to guarantee situs solitus totalis.

By contrast, the heterotaxy syndromes may well be characterized by genetic “underdosage,” that is, lack of the genetic information that should specify situs solitus at all segmental levels. The result of such lack of apparent genetic control of segmental situs is the aforementioned segmental situs discordances that result in what we call complex congenital heart disease.

Lack of genetic information (perhaps at the level of genes or gene regulators) appears to be the fundamental problem in Kartagener syndrome, which is characterized by situs inversus totalis, sinusitis, and bronchiectasis. The bronchiectasis part of this triad is now known to be due to lack of dynein side arms in the nasal sinus and tracheobronchial cilia, which consequently do not beat properly and hence do not cleanse the nasal sinuses and the tracheobronchial tree in the normal way. Men with Kartagener syndrome may be infertile, because their sperm tails also lack dynein, and consequently their sperm do not swim normally.

Lack of a specific protein such as dynein is strong presumptive evidence that this protein is not being coded for. In turn, this strongly suggests a lack of genetic information. Lack of genetic information, resulting in segmental situs deregulation, suggests that segmental situs may be random, or occurring by chance. This is the Layton hypothesis . ^, Dr. William M. Layton developed this hypothesis while working with the iv/iv mouse model. Indeed, Dr. Layton was really the discoverer of the congenital heart disease in the iv/iv mouse model that has proved so helpful in clarifying the molecular genetics of laterality defects.

iv is the gene symbol for situs inversus. As it turned out, the so-called iv/iv mouse model is really a model not of situs inversus totalis, but of visceral heterotaxy with situs ambiguus, many of these animals having polysplenia, and a few having asplenia. Phenotypically, some also had situs inversus.

In approximately 3000 experiments with the iv/iv mouse model, Layton found that D-loop ventricles and L-loop ventricles were very nearly 50/50: half D-loops, half L-loops (i.e., randomized). This suggested that cardiac loop formation was occurring by chance, like a coin flip (with a true coin). The only model of inheritance that these data appeared to fit was a chance or stochastic model. ^, Hence, the data suggested a true or apparent lack of genetic information. The Layton hypothesis ^, is what we mean by genetic “underdosage.”

Perhaps I should add that I served as the murine “cardiologist” in these early experiments with the iv/iv mouse model. My job was to help make the diagnosis, for example, transposition of the great arteries (in about 20%) and double-outlet right ventricle (in about 22%).

Consequently, our impression is that both syndromic forms of common AV canal have a strong genetic component in their etiology, but that these genetic components are very different (“overdosage” versus “underdosage”), as are the anatomic findings. Much more remains to be learned about the specific genetic aspects of both forms of syndromic common AV canal.

Are the heterotaxy syndromes with asplenia and polysplenia totally chaotic? Are there no patterns amid these segmental situs mismatches? Comparison of 95 postmortem patients with the asplenia syndrome versus 67 postmortem patients with the polysplenia syndrome revealed several statistically highly significant differences between these two different heterotaxy syndromes:

1.
Common AV canal was more than twice as frequent with asplenia (96%) as with polysplenia (45%) ( p < .0001).
2.
The complete form of common AV canal occurred much more frequently with asplenia (81%) than with polysplenia (33%) ( p < .0001).
3.
Normally divided AV canal with normally formed AV valves were much more frequent with polysplenia (55%) than with asplenia (4%) ( p < .0001).

Parachute Mitral Valve

Potentially parachute mitral valve in common AV canal requires a more detailed presentation in order to be well understood. First, what is meant by potentially parachute mitral valve in the setting of common AV canal? Potentially means following surgical repair, that is, following surgical division of the common AV canal into mitral and tricuspid canals in the complete form of common AV canal by closure of the AV septal defect, and following surgical reconstruction of the AV valves. Prior to surgical repair, in completely common AV canal, one cannot accurately speak of parachute mitral valve or double-orifice mitral valve—because there is no mitral valve, only a common AV valve. However, in the partial form of common AV canal, there is a malformed (cleft) mitral valve and hence the qualifier potentially is unnecessary.

What is “normal,” that is, usual, in common AV canal? In the left ventricle, both in the complete and partial forms of common AV canal, there usually are two main foci of chordal insertion ( Fig. 11.11 ): the anterolateral papillary muscle, the upper one; and the posteromedial papillary muscle, the lower one.

Fig. 11.11, Incomplete form of common atrioventricular canal (CAVC), that is, ostium primum type of atrial septal defect (ASD I) with cleft anterior leaflet of mitral valve. Note the typical cleft (CL) between the anterior leaflet component (AL) and the posterior leaflet component (PL) of what should normally be the anterior (noncleft) leaflet of the mitral valve. The AL component, derived from the anterior (superior) endocardial cushion of the AV canal, typically inserts only into the anterolateral (superior) papillary muscle group of the left ventricle (ALP) . The PL component, derived from the posterior (inferior) endocardial cushion of the common AV canal, typically inserts only into the posteromedial (inferior) papillary muscle group of the left ventricle (PMP) . The left lateral (mural) leaflet of mitral valve (LL) lies between the ALP and the PMP muscles of the left ventricle. Usually, the ALP and the PMP muscles constitute the two foci of chordal insertion of the left ventricle. AoV, Aortic valve; FW, free wall; LV, morphologically left ventricle; VS, ventricular septum.

The superior (anterior) leaflet of the common AV valve or the mitral valve usually inserts only into the superior (anterolateral) papillary muscle ( Fig. 11.11 ).

The inferior (posterior) leaflet of the common AV valve or the mitral valve usually inserts only into the inferior (posteromedial) papillary muscle group ( Fig. 11.11 ).

Note that these two main foci of chordal insertion—the anterolateral and posteromedial papillary muscles—are widely separated. Consequently, the left lateral leaflet of the common AV valve—that will form the posterior or mural leaflet of the mitral valve postoperatively—lies between the anterolateral and the posteromedial papillary muscles of the left ventricle ( Fig. 11.11 ).

The essential features of the potentially normal mitral valve are shown diagrammatically as seen from above in Fig. 11.12 . The most important details are the two widely spaced papillary muscles (anterolateral and posteromedial), which makes possible the existence of a posterior or mural leaflet of the mitral valve, which is derived from the left lateral leaflet of the common AV valve. As mentioned, note that the superior (anterior) leaflet of the common AV valve inserts only into the anterolateral papillary muscle; the inferior (posterior) leaflet inserts only into the posteromedial papillary muscle; the posterior or mural leaflet (derived from the left lateral leaflet of the common AV valve) inserts into both papillary muscles; and there are no accessory orifices in any of the leaflets of the common AV valve.

Fig. 11.12, Diagrammatic view of common atrioventricular (AV) canal without potentially parachute mitral valve, that is, with two widely spaced foci of chordal insertion, as seen from the atria—similar to the surgical view. The anterolateral papillary muscle (ALP) and the posteromedial papillary muscle (PMP) are widely spaced. Consequently there is a good-sized left lateral or mural leaflet (Lt. Lat. Leaflet) of the potentially mitral valve postoperatively. AP, Anterior papillary muscle group of the morphologically right ventricle (RV) ; LV, morphologically left ventricle; PP, posterior papillary muscle of the RV; VS, ventricular septum.

Potentially parachute mitral valve in completely common AV canal type A is presented in Fig. 11.13 . Note that both the superior (anterior) leaflet and the inferior (posterior) leaflet of the common AV valve insert only into the anterolateral papillary muscle. There is only one focus of chordal insertion in this left ventricle (the anterolateral papillary muscle), not the normal two foci of chordal insertion (the anterolateral and the posteromedial papillary muscles).

Fig. 11.13, Complete form of common AV canal, type A, with potentially parachute mitral valve, opened left ventricular view. All of the primary chordae tendineae from both the anterior leaflet component (AL) and the posterior leaflet component (PL) of the cleft (CL) anterior leaflet of the potentially mitral valve insert only into the anterolateral papillary muscle group (ALP) . Although a small posteromedial papillary muscle may be present (?PMP) , it does not receive primary chordal insertions of the PL component of the anterior mitral leaflet. Potentially parachute mitral valve is present because there is only one focus of left ventricular chordal insertion, not two (cf. Figs. 11.11 and 11.12 ). With only one focus of chordal insertion, there can be no left lateral (mural, or posterior) leaflet of the potentially parachute mitral valve postoperatively (cf. Figs. 11.11 and 11.12 ). Because the CL is the main orifice of the potentially parachute mitral valve, the cleft must not be sutured closed in order not to create iatrogenic mitral stenosis postoperatively. ALM, Anterolateral muscle of the left ventricle, also known as the muscle of Moulaert; AoV, aortic valve; FW, free wall; VS, ventricular septum; VSD, ventricular septal defects.

Note that there may well be a hypoplastic posteromedial papillary muscle in this left ventricle (?PMP, Fig. 11.13 ). However, this does not matter. What matters is how many foci of chordal insertion are present—not how many papillary muscles are present. Also, we are talking about the number of foci of chordal insertion of primary chordae tendineae—those that control the free margins of the superior and inferior leaflets of the common AV valve.

To summarize, this is regarded as a case of potentially parachute mitral valve ( Fig. 11.13 ) because there is only a single focus of chordal insertion of the primary chordae tendineae that control both the superior and the inferior leaflet margins of the common AV valve.

In common AV canal with potentially parachute mitral valve, the single focus of left ventricular chordal insertion typically is the anterolateral papillary muscle, the posteromedial papillary muscle being hypoplastic or absent, as in this patient ( Fig. 11.13 ).

As mentioned heretofore, the anomalies of the papillary muscles of the left ventricle in common AV canal with potentially parachute mitral valve ( Figs. 11.11 to 11.13 ) are the opposite of those found in typical cases of parachute mitral valve with a normally divided AV canal in which the posteromedial papillary muscle of the left ventricle is present and hypertrophied, while the anterolateral papillary muscle of the left ventricle is hypoplastic or absent ( Fig. 11.14 ).

Fig. 11.14, Typical parachute mitral valve (MV) with divided (as opposed to common) atrioventricular (AV) canal. Note that the posteromedial papillary muscle group (PMP) is present and hypertrophied, whereas the anterolateral papillary muscle group (ALP) is absent. The papillary muscles typically are the opposite with common AV canal and potentially parachute mitral valve. With divided AV canal, parachute MV typically is severely stenotic, as in this figure. But with common AV canal, potentially parachute MV often is not demonstrably stenotic preoperatively (cf. Fig. 11.13 ).

The salient features of common AV canal with a single left ventricular focus of primary chordal insertion is shown diagrammatically in Fig. 11.15 . This is a diagram of the case shown in Fig. 11.13 . Note that the diagram shows that both the superior and the inferior leaflets insert only into the anterolateral papillary muscle. There is no functional posteromedial papillary muscle receiving the primary chordae tendineae from the inferior leaflet.

Fig. 11.15, Diagram of potentially parachute mitral valve (MV) with completely common atrioventricular (AV) canal type A, viewed from the atria. This is the same patient as is shown photographically in Fig. 11.13 . Note that there is only one focus of left ventricular chordal insertion, the anterolateral papillary muscle group (ALP) . The orifice of the potential MV postoperatively is the cleft of the common AV valve. LV, left ventricle; VS, ventricular septum; RV, right ventricle; AP, anterior papillary muscle; PP, posterior papillary muscle.

Since there is only one functional papillary muscle group in this left ventricle, there can be no posterior or mural leaflet of the mitral valve. There can be no interpapillary muscle distance—an important component of the normal mitral orifice. Consequently, from a surgical standpoint, one must not suture closed the mitral cleft because the cleft is the major part of the postoperative mitral orifice. Surgical closure of the cleft of the potential mitral valve, when there is only one functional left ventricular papillary muscle, results in iatrogenic mitral stenosis postoperatively and can lead to the death of such patients. This is why a clear understanding of potentially parachute mitral valve is so important both diagnostically and surgically.

The anatomy of potentially parachute mitral valve can be even more misleading diagnostically. Two papillary muscles can be present, the anterolateral large and the posteromedial small and juxtaposed to the larger one ( Fig. 11.16 ). Despite the presence of two papillary muscles, admittedly not well spaced, most of the chordae inserted into the larger anterolateral group, and the potential posterior or mural leaflet of the mitral valve was very small—because of the markedly reduced interpapillary muscle distance ( Fig. 11.17 ). This may be regarded as a forme fruste of potentially parachute mitral valve in the setting of common AV canal ( Figs. 11.16 and 11.17 ). Although a forme fruste anatomically, this case functions like congenital mitral stenosis postoperatively unless handled expertly by the cardiac surgeon.

Fig. 11.16, Potentially parachute mitral valve in common atrioventricular (AV) canal with two juxtaposed papillary muscle groups. Note that the anterolateral papillary muscle group (APL) and the posteromedial papillary muscle group (PMP) are very close together (touching each other). The single focus of left ventricular chordal insertion consists of this clump of unseparated papillary musculature. Other abbreviations as previously noted.

Fig. 11.17, Diagram of potentially parachute mitral valve in common atrioventricular (AV) canal with two juxtaposed papillary muscle groups and one mitral orifice. The larger anterolateral papillary muscle (ALP) and the smaller posteromedial papillary muscle (PMP) are very close together (touching). This is a diagram of the heart shown in Fig. 11.16 . AP , anterior papillary muscle of the right ventricle; LV , left ventricle; PP , posterior papillary muscle of the right ventricle; RV , right ventricle; VS , ventricular septum.

Finally, potentially parachute mitral valve in common AV canal can be even more misleading diagnostically. It is even possible to have potentially parachute mitral valve with two well-developed left ventricular papillary muscles—if all of the primary chordae tendineae insert only into the anterolateral papillary muscle ( Fig. 11.18 ). This fortunately rare anatomic type of potentially parachute mitral valve is shown diagrammatically in Fig. 11.19 .

Fig. 11.18, Potentially parachute mitral valve in common atrioventricular (AV) canal with two papillary muscle groups and one mitral orifice. In this opened left ventricle, note that there are two well developed and well separated papillary muscle groups, the anterolateral (ALP) and the posteromedial (PMP) . However, the chordae tendineae from both the anterior (superior) leaflet (AL) above the cleft (CL) and from the posterior (inferior) leaflet (PL) below the cleft insert into the ALP muscle group. Why the free margin of the PL did not insert with primary chordae tendineae into the PMP muscle group is unknown. FW, Free wall; VS, ventricular septum.

Fig. 11.19, Diagram of potentially parachute mitral valve in common atrioventricular (AV) canal with two papillary muscle groups and one potentially mitral orifice, viewed from the atrial (surgical) aspect. This is a diagram of the heart specimen shown in Fig. 11.18 . Note that the anterolateral papillary muscle (ALP) and the posteromedial papillary muscle (PMP) are both present, well developed, and well separated. Nonetheless, both the anterior and the posterior leaflets insert chordae tendineae only into the ALP muscle group, thereby creating a potentially parachute mitral valve. The mitral orifice is the cleft and there is no posterior mitral leaflet. This is why we define potentially parachute mitral valve with common AV canal as a single focus of left ventricular insertion of primary chordae tendineae, that is, of the chordae that control the margins of the cleft. We do not define potentially parachute mitral valve in terms of the number of left ventricular papillary muscles, because this can be diagnostically misleading, as these cases ( Figs. 11.16 to 11.19 ) illustrate. AP , anterior papillary muscle of the right ventricle; LV , left ventricle; PP , posterior papillary muscle of the right ventricle; RV , right ventricle; VS , ventricular septum.

Would it be possible to diagnose such a case accurately preoperatively ( Figs. 11.18 and 11.19 )? We think that the presence of two large left ventricular papillary muscles could be very confusing diagnostically. This, in turn, is why the final diagnosis may well have to be made by the well-educated cardiac surgeon, under direct vision.

This, too, is why we have been saying that potentially parachute mitral valve in common AV canal is not just a question of the number of left ventricular papillary muscles. It is really a question of the number of functional left ventricular papillary muscles that receive primary chordae tendineae. In other words, it is a question of the number of left ventricular foci of primary chordal insertion. Unfortunately, the diagnosis of potentially parachute mitral valve is not just a question of counting left ventricular papillary muscles. Obviously, if there is only one left ventricular papillary muscle—the anterolateral—one should certainly make this diagnosis. However, the real problem is that this diagnosis is more subtle: potentially parachute mitral valve can be present with two papillary muscle groups ( Figs. 11.16 to 11.19 ). Can we resolve and accurately visualize the chordae tendineae by imaging, and thus make this diagnosis preoperatively? That is the diagnostic challenge.

Double-Orifice Mitral Valve

Potentially double-orifice mitral valve also occurs with common AV canal and hence needs to be understood. In the study of Baño et al concerning double-orifice mitral valve with a divided AV canal and separate mitral and tricuspid valves, we learned how very important the tensor apparatus is in this anomaly. The same principles apply in potentially double-orifice mitral valve in common AV canal.

Double-Orifice Mitral Valve Principles

AV valve tissue is reminiscent of the law of conservation of mass and energy in the sense that there is only so much “stuff”: it is either “mass” or “energy,” but it cannot be both at the same time. By analogy, AV valve tissue can be either leaflet, or chorda tendinea, but cannot be both at the same place. The meaning of this metaphor will become clear on studying Fig. 11.20 .

Fig. 11.20, Potentially parachute mitral valve with common atrioventricular (AV) canal and with an accessory orifice in the anterosuperior leaflet. In this opened left ventricle, note that the primary chordae tendineae of the anterosuperior leaflet (AL) and of the posterioinferior leaflet (PL) both insert into the anterolateral papillary muscle group (ALP) . There may be a posteromedial papillary muscle group present (?PMP) ; but it does not receive the primary chordae tendineae of the PL that instead insert into the ALP. The primary chordae from the AL—from the superior margin of the cleft (CL) —also insert into the ALP muscle group. Hence, a potential parachute mitral valve is present, with one left ventricular focus of insertion of primary chordae tendineae into the ALP muscle group. In the AL, there is an accessory orifice (AO) . The superior chordae of this accessory orifice insert into the anterolateral muscle (ALM) or muscle of Moulaert. The inferior chordae from this accessory orifice insert into the ALP muscle group. Postoperatively, this patient would have not only a parachute mitral valve, but also a double-orifice mitral valve, that is, a parachute and double-orifice mitral valve. The cleft is the main orifice— MO (CL) —of this double-orifice parachute mitral valve. VSD , ventricular septal defect.

Potentially parachute mitral valve is present in this case of completely common AV canal because both the superior leaflet (AL) and the inferior leaflet (PL) send primary chordal insertions only to the anterolateral papillary muscle (ALP) of the left ventricle. A rudimentary posteromedial papillary muscle also appears to be present (?PMP), but it receives no primary chordae tendineae from the free margins of the inferior leaflet of the common AV valve. The inferior (or posterior) leaflet displays crossing chordae: the chordae tendineae of the inferior leaflet cross over from inferiorly to superiorly and insert only into the superior (anterolateral) papillary muscle—instead of inserting into the inferior (posteromedial) papillary muscle, which would be “normal” or usual for common AV canal ( Fig. 11.11 ). Crossing chordae tendineae (from inferiorly to superiorly) are typical of potentially parachute mitral valve.

However, please note that this patient does not have a single focus of left ventricular chordal insertion ( Fig. 11.20 ). It is true that there is only a single focus of primary chordal insertion (into the anterolateral papillary muscle); hence potentially parachute mitral valve is indeed present.

But there is a secondary focus of chordal insertion ( Fig. 11.20 ): from the superior leaflet (AL) into the anterolateral muscle (ALM) of the left ventricle (also known as the muscle of Moulaert).

So now the superior leaflet has a “problem”: it has two foci of left ventricular chordal insertion—the usual one into the anterolateral papillary muscle, and an unusual one into the muscle of Moulaert (ALM).

So what does the superior leaflet do ( Fig. 11.20 )? Its substance is used to make these two chordal insertions. Consequently, the superior leaflet cannot make leaflet tissue at the same site. (Remember the rule: chordae, or leaflet, but not both at the same site, apparently because the amount of endocardial cushion tissue is limited.) Since the superior endocardial cushion tissue has been used to make closely adjacent chordae tendineae, consequently there is no leaflet tissue between these two foci of chordal insertion. The result is an accessory orifice in the superior leaflet ( Figs. 11.20 and 11.21 ).

Fig. 11.21, Potentially parachute mitral valve in common atrioventricular (AV) canal with an accessory orifice (AO) in the anterosuperior leaflet. This is a diagram of the heart specimen shown in Fig. 11.20 . ALP, anterolateral papillary muscle of the left ventricle; AO, accessory orifice; AP , anterior papillary muscle of the right ventricle; LV , left ventricle; PP , posterior papillary muscle of the right ventricle; RV , right ventricle; VS , ventricular septum.

It must be emphasized that the aforementioned analogy or metaphor is presented in order to make this potentially double-orifice mitral valve in common AV canal comprehensible. The real morphogenesis of such an accessory orifice may be different from the foregoing explanatory analogy or metaphor.

However, it is anatomically factual that the main orifice of this potentially parachute mitral valve is the cleft orifice; and that there is an accessory orifice within the superior leaflet, between closely adjacent chordal insertions ( Figs. 11.20 and 11.21 ).

Is it possible for there to be an accessory orifice in the inferior leaflet of the common AV valve, resulting in a different anatomic type of potentially double-orifice mitral valve? The answer, of course, is yes ( Figs. 11.22 and 11.23 ). (In congenital heart disease, almost anything that one can imagine in fact does occur; hence the fascination of our field, which certainly isn’t boring.) How does it work?

Fig. 11.22, Potentially parachute mitral valve in common atrioventricular (AV) canal with an accessory orifice in the posteroinferior leaflet. In this opened left ventricle, note that the cleft (CL) between the anterior leaflet (AL) and the posterior leaflet (PL) has been sutured closed, greatly narrowing the main orifice (MO) . The accessory orifice (AO) in the posterior leaflet sends some chordae tendineae to a small posteromedial papillary muscle (PMP) and other chordae tendineae to the larger anterolateral papillary muscle group (ALP) . VS, Ventricular septum; FW, free wall.

Fig. 11.23, Potentially parachute mitral valve in common atrioventricular (AV) canal with an accessory orifice in the posteroinferior leaflet of the common AV valve, diagram of the heart presented in Fig. 11.22 . This case is an example of potentially parachute mitral valve with two papillary muscles and two orifices. ALP , anterolateral papillary muscle of the left ventricle; AO , accessory orifice; AP , anterior papillary muscle of the right ventricle; LV , left ventricle; PP , posterior papillary muscle of the right ventricle; RV , right ventricle; VS , ventricular septum.

Examine Fig. 11.22 carefully. The main cleft orifice (CL) of this potentially double-orifice mitral valve looks quite small—because it was sutured closed surgically, greatly narrowing its main orifice (MO). The inferior leaflet (PL) sends a few small chordae to insert into a small posteromedial papillary muscle. However, most of the chordae of the inferior leaflet insert via crossing chordae into the superior (anterolateral) papillary muscle. So, now it is the inferior leaflet of the potentially parachute mitral valve that has an unusual “problem”: two foci of chordal insertion.

Remembering our analogy with the law of mass and energy, we have two closely adjacent foci of chordal insertion involving the inferior leaflet ( Fig. 11.22 ). Since the endocardial cushion tissue has been used up making chordae tendineae, it is not surprising that there is no leaflet tissue at the same site. Hence, there is an accessory orifice in the inferior leaflet between these closely adjacent chordae tendineae ( Figs. 11.22 and 11.23 ).

Concerning the morphogenesis of potentially double-orifice mitral valve, there is the aforementioned “competition” between chordae and leaflets: one, or the other, but not both at the same site. Another way of expressing the same idea is as follows:

There is an inverse relationship between leaflet tissue and chordae tendineae. Think of the normal anterior leaflet of the mitral valve. It is semicircular in shape. Toward its center, leaflet tissue is maximal and chordae are minimal. Then, as one approaches either commissure, the leaflet tissue becomes less and less, while the chordae become more and more prominent.

Now think of the normal posterior leaflet of the mitral valve with its lateral, middle, and medial scallops. In the center of each scallop, the leaflet tissue is predominant and the chordae tendineae are less prominent. But between the scallops, the reverse situation is present: the chordae are predominant, and the leaflet tissue is minimized.

Hence, the inverse relationship between leaflet tissue and chordae tendineae is seen both in the normal anterior and posterior leaflets, and at the commissures.

Accessory orifices in AV valve leaflets (mitral, tricuspid, and common) appear to reflect this inverse relationship between leaflets and chordae, in which chordae are prominent, and the leaflet tissue between or among the chordae is minimal, that is, absent.

There may also be a simple mechanical explanation for accessory orifices: When there is a ring of chordae tendineae inserting into the ventricular surface of mitral leaflet tissue, the chordae pull on the endocardial cushion tissue with each ventricular systole. Traction by chordal rings may help to produce holes (accessory orifices) in AV leaflets.

Whatever the correct morphogenesis of potentially double-orifice mitral valve may ultimately prove to be, whenever you find a superior or an inferior leaflet of a common AV valve with two adjacent foci of chordal insertions, or with an abnormal chordal ring, you should also expect to find an accessory orifice between the two chordal foci, or within the chordal ring. One chordal focus is from the free margins of the leaflet: these are the primary chordae. The other chordal focus arises from the ventricular surface of the superior or inferior leaflet, away from the leaflet’s free margins. These chordae may be called secondary chordae.

Consequently, the accessory orifices of a potentially parachute mitral valve often lie between the primary and secondary chordal insertions ( Figs. 11.20 to 11.23 ).

The diagnostic implications of the foregoing include the following:

1.
In common AV canal with potentially parachute mitral valve, although there is only one focus of primary chordal insertion, there may be an additional focus of secondary chordal insertion.
2.
Between the primary and secondary foci of chordal insertion, a careful imaging search should be made for an accessory orifice, in order to establish or exclude the diagnosis of potentially parachute mitral valve with a primary (cleft) orifice, and with an additional accessory orifice in either the superior or inferior leaflet of the common AV valve.
3.
The focus of diagnostic imaging should be primarily on the tensor apparatus—not just on the number of papillary muscles, but also on the number and type (primary or secondary) of chordal insertions.
4.
Diagnostic attention should also be focused on the superior, inferior, and lateral leaflets of the common AV valve. One should use the tensor apparatus to understand and to explain the leaflet findings.

In other words, potentially double-orifice mitral valve is not just about double-orifice mitral valve. It is really about tensor apparatus abnormalities. The same principle also applies to potentially parachute mitral valve, with which double-orifice mitral valve may coexist. The key to diagnostic understanding is not just the leaflets: it really is the tensor apparatus.

Findings

Complete Forms of Common AV Canal

Frequency: This study is based on 266 postmortem cases of completely common AV canal.
Percentage of Congenital Heart Disease: 266/3216 cases, that is, 8.27% of all patients with congenital heart disease (95% confidence interval [CI] 7.32% to 9.22%).
Gender: Males/females approximately equal, 126/130 (0.97/1). The gender was not known in 10 cases.
Age at Death: The mean age at death was 926 days (2.54 years) ± 2062 days (5.6 years), ranging from 0 (fetal demise) to 40.8 years. The median age of death was 120 days (4 months).

Associated Anomalies

Associated anomalies were frequent and important:

1.
Down syndrome was the most frequent associated anomaly: 82/266 cases (30.8%).
- a.
  Down syndrome was associated with multiple congenital anomalies in 4/82 patients (4.9%).
- b.
  Down syndrome occurred with Hirschprung disease in 1/82 (1.2%).
- c.
  A Down mosaic was found in 1/82 (1.2%).
- d.
  Familial Down syndrome occurred in 1/82 (1.2%).
2.
The heterotaxy syndrome with asplenia was the second most frequent associated anomaly: 51/266 cases (19.2%).
- a.
  The asplenia syndrome with additional multiple congenital anomalies occurred in 2/51 patients (3.9%).
- b.
  Familial asplenia syndrome was found in 2/51 (3.9%).
- c.
  Asplenia syndrome was observed in one of twins: 1/51 (2.0%).
3.
Multiple congenital anomalies was the third most frequent type of associated anomalies: 21/266 (7.9%). Trisomy 13 occurred in 1/21 (4.8%).
4.
The heterotaxy syndrome with polysplenia was found in 16/266 patients (6.0%). Of these 16 cases of the polysplenia syndrome, 1 had congenital heart block (6.25%) and 1 had additional multiple congenital anomalies (6.25%).
5.
Familial congenital heart disease was encountered in 2/266 cases (0.75%).
6.
Conjoined twins were observed twice: 2/266 (0.75%). Both were thoracopagus.
7.
One patient (1/266, 0.38%) had, to the best of our knowledge, a previously unknown group of associated malformations: Noonan syndrome with Meckel diverticulum, microcardia, heterotaxy with a normally formed spleen and an accessory spleen, left ventricular type of completely common AV canal with tricuspid stenosis postoperatively, infant of a diabetic mother, isolated ventricular inversion {S,L,S}, familial Holt-Oram syndrome, and Ellis-van Creveld syndrome.

Heart Position

Levocardia in 221/266 cases (83%); dextrocardia in 40/266 patients (15%); and mesocardia in 5/266 cases (2%).

When levocardia (a left-sided heart) was present, the segmental anatomy of the heart was usually normal, that is, {S,D,S} in 166/221 cases (75%).

When levocardia was present, complex congenital heart disease (abnormal segmental anatomy) was present in 55/221 cases (25%).

In contrast, when dextrocardia (a right-sided heart) was present, the segmental anatomy was always abnormal: 40/40 cases, 100% ( p < .0001).

When mesocardia (a centrally located heart) occurred, the segmental anatomy and alignments were normal, that is, {S,D,S}, in 3/5 patients (60%), and were abnormal in 2/5 cases (40%).

Is the sidedness of the heart (levocardia, or dextrocardia, or mesocardia) related to the type of ventricular loop that is present?

When dextrocardia was present, L-loop ventricles were present in 27 of 40 patients (67.5%), while D-loop ventricles were found in 13 of 40 (32.5%).

By contrast, when levocardia or mesocardia was present, L-loop ventricles occurred in only 10 of 226 cases (4.42%), while D-loop ventricles were found in 216 of 226 cases (95.58%).

Thus, L-loop ventricles were associated with dextrocardia (67.5%) and D-loop ventricles were associated with levocardia and mesocardia (95.58%) ( p < .0001). From a developmental viewpoint, L-loop ventricles “should” have dextrocardia and D-loop ventricles “should” have levocardia (see Chapter 2 )—and this is what was found in this study of the complete form of common AV canal.

Is the sidedness of the heart (levocardia, or dextrocardia, or mesocardia) related to the type of visceroatrial situs that is present?

With levocardia, the types of visceroatrial situs were:

situs solitus in 190 of 221, 86%;
situs inversus in 3 of 221, 1%;
situs ambiguus in 11 of 221, 5%;
situs ambiguus, probably solitus, in 8 of 221, 4%; and
situs ambiguus, probably inversus, in 9 of 221, 4%.

Thus, solitus atria [including S + A (S)] occurred in 198 of 221 patients (90%). Inversus atria [including I + A (I)] were found in 12 of 221 cases (5%). Situs ambiguus atrialis (undiagnosed atrial situs) occurred in 11/221 cases (5%).

With dextrocardia, the types of visceroatrial situs were:

situs solitus in 12 of 40, 30%;
situs inversus in 11 of 40, 27.5%;
situs ambiguus, probably situs solitus, in 2/40, 5%;
situs ambiguus, probably situs inversus, in 14/40, 35%; and
situs ambiguus, 1 of 40, 2.5%

Thus, the atrial situs was:

solitus in 14 of 40, 35%;
inversus in 25 of 40, 62.5%; and
ambiguus in 1 of 40, 2.5%.

Conclusions

1.
Situs solitus of the viscera and atria was much more frequent with levocardia (198/221, 90%) than with dextrocardia (14/40, 35%) ( p < .0001).
2.
Situs inversus of the viscera and atria was much more frequent with dextrocardia (25/40, 62.5%) than with levocardia (12/221, 5%) ( p < .0001).
3.
Situs ambiguus of the viscera and atria was somewhat more common with levocardia (11/221, 5%) than with dextrocardia (1/40, 2.5%) ( p = not significant [NS]) (Fisher’s exact test).

Summary

Thus, cardiac position (levocardia or dextrocardia) was significantly related (1) to the type of visceroatrial situs and (2) to the type of ventricular loop, as above.

Types of Ventricular Loop

In these 266 cases of the complete form of common AV canal, D-loop ventricles were present in 229 patients (86%) and L-loop ventricles were found in 37 cases (14%).

Types of AV Alignment

In these 266 cases, AV concordance was present in 220 (83%), AV discordance was observed in 33 (12%), and the AV alignment was indeterminate in 13 patients with visceral heterotaxy (5%). When the AV alignment was in determinate, the atrial and ventricular segmental anatomy was {A,D,-} in 10 patients and {A,L,-} in 3.

Ventriculoarterial Alignments

The types of ventriculoarterial alignment in levocardia ( n = 221) and in dextrocardia ( n = 40) are presented in Table 11.1 .

Table 11.1

Ventriculoarterial Alignments

	Levocardia ( n = 221)	Dextrocardia ( n = 40)	p -Value
Solitus normal	168 (76.02%)	3 (7.5%)	<.0001
Inversus normal	3 (1.36%)	4 (10%)	NS
TGA	16 (7.24%)	7 (17.5%)	NS
DORV	33 (14.93%)	25 (62.5%)	<.0001
DOLV	1 (0.45%)	0 (0%)
ACM	0 (0%)	1 (2.5%)
Truncus	2 (0.90%)	0 (0%)

ACM, Anatomically corrected malposition of the great arteries; DOLV, double-outlet left ventricle; DORV, double-outlet right ventricle; NS, not significant; TGA, transposition of the great arteries.

Mesocardia was present in 5 cases. Solitus normally related great arteries were present in 3 cases (60%), and double-outlet right ventricle was found in 2 cases (40%).

Conclusions

1.
Solitus normally related great arteries were significantly more frequent in completely common AV canal with levocardia (76%) than with dextrocardia (7.5%) ( p < .0001, Table 11.1 ).
2.
Double-outlet right ventricle was significantly more common in completely common AV canal with dextrocardia (62.5%) than with levocardia (15%) ( p < .0001, Table 11.1 ).
3.
Other VA (ventriculoarterial) alignments were not significantly different with levocardia and with dextrocardia ( Table 11.1 ).

Conus

The anatomic status of the conus arteriosus or infundibulum, which is the connector between the ventricular sinuses and the great arteries, is presented in Table 11.2 .

Table 11.2

Anatomic Types of Conus in 266 Cases of Completely Common AV Canal

Type of Conus	No. of Cases	% of Series
1. Subpulmonary	199	74.81
2. Subaortic	20	7.52
3. Bilateral (subaortic and subpulmonary)	45	16.92
4. Bilaterally deficient (neither subaortic nor subpulmonary)	2	0.75

Thus, a normal type of conus (subpulmonary, with aortic-mitral fibrous continuity) was present in almost three-quarters of the cases (74.81%, Table 11.2 ). Second in frequency was a bilateral conus (both subaortic and subpulmonary, preventing semilunar-atrioventricular fibrous continuity) in 16.92% ( Table 11.2 ). Third in frequency was a subaortic conus (with pulmonary-atrioventricular fibrous continuity) in 7.52% ( Table 11.2 ). Least frequent was a bilaterally deficient conus in 0.75% ( Table 11.2 ), “bilaterally deficient” meaning that there were both aortic-atrioventricular and pulmonary-atrioventricular fibrous continuity—permitted by the bilateral muscular deficiency of the conus.

Perhaps even more interesting were the correlations between the anatomic types of conus on the one hand (subpulmonary, subaortic, bilateral, and bilaterally deficient) and the types of ventriculoarterial alignment on the other (normally related great arteries, transposition of the great arteries, double-outlet right ventricle, etc.). These correlations were as follows:

1.
A subpulmonary conus was usually associated, as expected, with normally related great arteries (180/199 cases, 90.45%). Seven of these 180 cases had inverted normally related great arteries. However, double-outlet right ventricle (DORV) was found in 19 cases (9.55%); this was a surprise.

One is used to thinking of DORV as being associated with a bilateral conus (not with a subpulmonary conus). What do we think of this finding of a subpulmonary conus associated with DORV in almost 10% of the cases with the complete form of common AV canal? Many of these cases of DORV with a subpulmonary conus had the polysplenia syndrome. For reasons that are still unknown, cases with visceral heterotaxy and polysplenia never have had a well-developed muscular subaortic conus, in our experience, whereas patients with visceral heterotaxy and the asplenia syndrome almost always have had a well-developed subaortic conus, or a bilateral conus. Consequently, DORV with a subpulmonary conus is quite typical of the heterotaxy syndrome with polysplenia.

The other situation in which DORV typically has a unilateral conus (as opposed to a bilateral conus) is with what Dr. Richard D. Rowe used to call the infantile type of DORV (“infantile” because it has such an unfavorable natural history), that is, DORV with the hypoplastic left heart syndrome (e.g., with mitral atresia or severe congenital mitral stenosis). In DORV with the hypoplastic left heart syndrome, the unilateral conus can be supulmonary (only), with aortic-atrioventricular fibrous continuity, or subaortic (only), with pulmonary-atrioventricular fibrous continuity.

Thus, DORV with a subpulmonary conus should suggest two diagnostic possibilities: (1) the heterotaxy syndrome with polysplenia; and (2) DORV with the hypoplastic left heart syndrome.

One patient had truncus arteriosus, which we think is typically associated with an atretic subpulmonary conus.

2.
A subaortic conus was often associated with transposition of the great arteries (TGA), just as one would expect (14/20, 70%). However, DORV coexisted with a subaortic conus in 6/20 patients (30%). These findings serve as a reminder that a subaortic conus is certainly not associated exclusively with TGA. Again, DORV does not have to have a bilateral conus; the conus can be subaortic (only).
3.
A bilateral conus, again as expected, was usually associated with DORV (35/45, 77.77%). TGA was associated with a bilateral conus in 7 of 45 patients (15.55%), also not surprising. But to find 2 of these 45 patients (4.44%) with a bilateral conus and normally related great arteries—this raised our eyebrows.

Morphogenetically, how can one understand normally related great arteries with a bilateral (subpulmonary and subaortic) muscular conus (with no semilunar-AV fibrous continuity)? Our best hypothesis is as follows. A bilateral conus is how it starts, early in embryogenesis. Then what normally happens is that the subpulmonary part of the conus grows and develops, whereas the subaortic infundibular free wall normally undergoes absorption—probably due to apoptosis (programmed cell death), thereby permitting the normal aortic-mitral fibrous contiguity and continuity. However, if subaortic conal free wall resorption is subnormal, some subaortic conal free wall can persist, resulting in normally related great arteries with a small tongue of subaortic conal free wall musculature separating the aortic and mitral valves. So, a bilateral conus can coexist with normally related great arteries if (1) the subpulmonary part of the conus is well developed, and (2) if a relatively small amount of subaortic conal free wall musculature persists and develops—but not enough subaortic conal free wall musculature to disrupt significantly the normal aortic-mitral and aortic-left ventricular approximations.

We obviously hope that our morphogenetic hypothesis is correct. But in any case, it is important to know that rarely, normally related great arteries can have a bilateral conus.

Finally, there was one patient with anatomically corrected malposition (ACM) of the great arteries with a bilateral conus (1/45, 2.22%). Is this surprising? No, just rare. ACM means that despite the malposition of the great arteries, nonetheless the great arteries originate above the anatomically correct ventricles—aorta above the morphologically left ventricle, and pulmonary artery above the morphologically right ventricle. In our 1967 paper that established the existence of anatomically corrected malposition of the great arteries, all three cases had a bilateral conus (with subaortic and subpulmonary conal musculature) and therefore with no semilunar-atrioventricular fibrous continuity.

4.
A bilaterally deficient conus was found in 2 of these 266 patients with the complete form of common AV canal (0.75%). One had TGA, and the other had DORV. Interestingly, neither had double-outlet left ventricle—that one often associates with a bilaterally deficient conus.

Segmental Sets or Combinations

In addition to the foregoing segment-by-segment analysis, what we now need is segment-by-segment synthesis. What segmental combinations (or sets) occurred? Only when one knows what the segmental combinations were is it possible to understand clearly the various types of heart in which the complete form of common AV canal occurred. For clarity and brevity, the segmental sets are presented in Table 11.3 . This is a form of multivariable analysis: the study of three segments at a time, in triplets—{atria, ventricles, great arteries}—as they actually occur in life. The foregoing may be called segmental set or combination analysis.

Table 11.3

Segmental Combinations in Complete Form of Common AV Canal ( n = 266)

	Segmental Combination or Set	No. of Cases	% of Series
1.	{S,D,S}	173	65.04
2.	{S,D,I}	1	0.38
3.	{I,L,I}	1	0.38
4.	{I,D,I}	2	0.75
5.	{A (S),D,S}	1	0.38
6.	{A,D,S}	1	0.38
7.	{A (S),L,I}	1	0.38
8.	{A,L,I}	1	0.38
9.	{A (I),L,I}	1	0.38
10.	TGA {S,D,D}	4	1.50
11.	TGA {S,L,L}	3	1.13
12.	TGA {I,D,D}	1	0.38
13.	TGA {I,D,L}	1	0.38
14.	TGA {A (S),L,L}	3	1.13
15.	TGA {A (I),L,L}	1	0.38
16.	TGA {A (I),D,D}	4	1.50
17.	TGA {A,D,D}	2	0.75
18.	TGA {A,L,L}	1	0.38
19.	TGA {A,L,D}	1	0.38
20.	DORV {S,D,D}	17	6.39
21.	DORV {S,L,L}	4	1.50
22.	DORV {S,L,D}	1	0.38
23.	DORV {I,L,L}	7	2.63
24.	DORV {I,D,D}	2	0.75
25.	DORV {A (S),D,D}	4	1.50
26.	DORV {A (S),L,L}	2	0.75
27.	DORV {A (I),L,L}	7	2.63
28.	DORV {A (I),L,D}	2	0.75
29.	DORV {A (I), D,D}	9	3.38
30.	DORV {A,D,D}	6	2.26
31.	DOLV {S,D,D}	1	0.38
32.	ACM {A,D,L}	1	0.38

For explanation, please see the text.

Table 11.3 answers the question, What was the segmental anatomy?

Of these 32 different segmental combinations with completely common AV canal ( Table 11.3 ), the first eight combinations have normally related great arteries, either solitus normally related, that is, {-,-,S}, or inversus normally related, that is, {-,-,I}. Thus, normally related great arteries comprised 68% of this series.

1.
The normal segmental combination, {S,D,S}, was, of course, by far the most common segmental situs set (65%).
2.
The inverted normal segmental combination, {I,L,I}, was noteworthy for its rarity (0.38%). This patient also had asplenia, so that the visceroatrial situs inversus was not “pure,” but was admixed with visceral heterotaxy.
3.
The second segmental combination, {S,D,I}, was also rare (0.38%). It consists of situs solitus of the viscera and atria {S,-,-}, D-loop ventricles {-,D,-}, and inverted normally related great arteries {-,-,I}. As the segmental set suggests, there were AV concordance and VA concordance. In words, {S,D,I} is known as isolated infundibuloarterial inversion. Only the infundibulum and the great arteries were inverted {-,-I}, whereas the atria and the ventricles were not inverted {S,D,-}. Because of the AV and VA concordances, the circulations can be physiologically normal (corrected), unless associated malformations such as tetralogy of Fallot vitiate the potential physiologic correction. A tetralogy of Fallot type of conotruncal anomaly was not present in this 63-day-old female infant with a type A completely common AV canal and a relatively small left ventricle.
4.
The fourth segmental combination ( Table 11.3 ), {I,D,I}, is also rare (0.75% of the series) and consists of situs inversus of the viscera and atria, D-loop ventricles, and inverted normally related great arteries. Because there is atrioventricular discordance {I,D,-}, but with VA concordance {-,D,I}, the systemic and pulmonary circulations are physiologically uncorrected (one segmental alignment discordance, at the level of the AV junction). The segmental combination {I,D,I} may be called isolated ventricular noninversion. Only the ventricles were not inverted, that is, {-D,-}, whereas both the atria and the infundibuloarterial part of the heart were inverted, that is, {I,-,I}. As in all forms of AV discordance with VA concordance, an atrial switch procedure (Senning or Mustard) is needed to correct the circulations physiologically.

It is noteworthy that in the study by Pasquini et al, {I,D,I} was the only type of AV discordance with VA concordance that we did not know to have been documented ( Fig. 11.24 ). So now it may be said that all six theoretically predictable types of AV discordance with VA concordance have been documented ( Fig. 11.24 ):

1.
{S,L,S}, that is, isolated ventricular inversion;
2.
ACM {S,L,D}, that is, anatomically corrected malposition of the great arteries with solitus atria, L-loop ventricles (and hence AV discordance), and D-malposition of the great arteries (and thus VA concordance by the definition of ACM);
3.
{S,L,I}, that is, solitus atria, L-loop ventricles (and hence AV discordance), and inverted normally related great arteries (and consequently VA concordance by the definition of normally related great arteries);
4.
{I,D,I}, that is, inversus atria, D-loop ventricles (and thus AV disconcordance), and inverted normally related great arteries (with VA concordance in view of the definition of inverted normally related to great arteries);
5.
ACM {I,D,L}, that is, anatomically corrected malposition of the great arteries with the segmental anatomic combination of inversus atria, D-loop ventricles (and hence AV discordance), and L-malposition of the great arteries (with VA concordance because of the definition of ACM); and
6.
{I,D,S}, that is, situs inversus of the viscera and atria, D-loop ventricles (and thus AV discordance), and solitus normally related great arteries (and consequently VA concordance).

Fig. 11.24, Anatomic types of atrioventricular (AV) discordance with ventriculoarterial (VA) concordance. The heart diagrams are viewed from below, similar to a subxiphoid 2-dimensional echocardiogram. A, Anterior; P, posterior; R, right; L, left; Inf, infundibulum (conus); LA, morphologically left atrium; LV, morphologically left ventricle; RA, morphologically right atrium; RV, morphologically right ventricle. The aortic valve is indicated by coronary ostia, and the pulmonary valve by absence of coronary ostia. Braces {} mean “the set (or combination) of.” The set of {atria, ventricles, great arteries} is shown. The atria may be in situs solitus, as in the top row: {S,-,-}, or in situs inversus, as in the bottom row: {I,-,-}. The ventricles may have D-loop situs with RV to the right and right-handed: {-,D,-}; or the ventricles may have L-loop situs with RV to the left and left-handed: {-,L,-}. The great arteries may be solitus normally related: {-,-,S}; or the great arteries may be inversus normally related: {-,-,I}. If anatomically corrected malposition (ACM) of the great arteries is present, then VA alignment concordance is present, but the infundibular connector is abnormal (subaortic, or subaortic and subpulmonary), rather than normal (subpulmonary)—as it is with solitus and inversus normally related great arteries. When the great arteries are normally related, the VA alignments and connections (type of conal connector) are both normal. With ACM, the VA alignments are concordant (but not fully normal), and the VA connections (anatomic type of conal or infundibular connector) are abnormal. With ACM, the direction of ventricular looping and the direction of infundibuloarterial twisting always are opposites (to the best of our present knowledge).

To our knowledge, this is the first time that {I,D,I} — isolated ventricular noninversion — has been reported.

5. and 6.
{A(S),D,S}and {A,D,S} are also rare (each 0.38% of this series) and noteworthy. Infrequently, it is possible to have the heterotaxy syndrome with situs ambiguus of the viscera, but situs solitus of the atria, that is, {A(S),-,-}, and with D-loop ventricles and solitus normally related great arteries. Hence, the segmental was {A(S), D,S}. When we were unable to diagnose the atrial situs with confidence, we made the diagnosis of visceroatrial situs ambiguus, as in {A,D,S}.

To summarize this point, when we are able to diagnose the atrial situs, we note this fact, for example, as follows: {A(S),D,S} ( Table 11.3 ). This notation indicates that visceroatrial situs ambiguus is present, but that we think that the atrial situs is situs solitus. Both {A(S),D,S} and {A,D,S} had the heterotaxy syndrome with polysplenia, and interruption of the inferior vena cava with a prominent azygos vein to a superior vena cava. When the segmental set is {A,(S)D,S}, there is AV concordance. But when the segmental combination is {A,D,S}, indicating that the atrial situs is not diagnosed, then the AV alignment (concordant/discordant) also remains undiagnosed.

7. and 8.
{A,(S),L,I} with polysplenia, and {A,L,I}with asplenia (sister also with asplenia) were both infrequent (1 case each, 0.38%, Table 11.3 ). These similar segmental combinations are noteworthy. In the patient with polysplenia, we were able to make the diagnosis of atrial situs solitus, that is, {A(S),L,I}. However, in the patient with familial asplenia, we were not able to diagnose the atrial situs, as the segmental combination indicates: {A,L,I}.

In the patient with {A,(S),L,I}, there was AV discordance: {-(S),L,-}. In this polysplenic patient with discordant L-loop ventricles, the usual corrected transposition was not present; that is, this patient did not have the more usual segmental combination of TGA {S,L,L}. Instead, this patient had the much more rare segmental combination of {A,(S),L, I }. Why? No one fully understands this at the present time. But remember that with the polysplenia syndrome, solitus or inversus normally related great arteries is the rule. We have never seen a well-developed muscular subaortic conus, typical of TGA {S,L,L}, in the heterotaxy syndrome with polysplenia.

The patient with {A,L,I} and familial asplenia was similarly unusual. DORV or TGA is the rule with asplenia. But this rare case had inverted normally related great arteries.

9.
{A(I),L,I} with polysplenia occurred in only 1 patient (0.38%), serving to emphasize how rare hearts resembling situs inversus totalis are. This case, as in the patient mentioned above with {I,L,I} and the asplenia syndrome (category 3, Table 11.3 ), did not have so-called pure situs inversus. Instead, both patients had some degree of visceroatrial heterotaxy; however, the patient mentioned in category 3 of Table 11.3 was phenotypically {I,L,I}.

This mixing or blending of visceroatrial situs inversus on the one hand and visceral heterotaxy with situs ambiguus on the other makes us think that there really may be only two basic types of visceroatrial situs: (1) normal (situs solitus); and (2) abnormal (situs inversus and situs ambiguus), that share overlapping phenotypes. We hope that molecular genetics will be able to resolve this question.

10.–19.
Ten different phenotypes of TGA were found ( Table 11.3 ) in 21 patients, accounting for 8% of this series. Note how infrequent typical D-TGA was: TGA{S,D,D} in only 4 patients (1.5% of this series of 266 cases of the complete form of common AV canal). Similarly, classical corrected L-TGA was even less frequent: TGA{S,L,L} in 3 patients (1.13%, Table 11.3 ).
20.–30.
Eleven different segmental combinations of DORV were found in 61 patients (23%, Table 11.3 ).
31. and 32.
Double-outlet left ventricle (DOLV) and anatomically corrected malposition of the great arteries (ACM) were both represented by 1 case each (0.38% each, Table 11.3 ).

Segmental combination analysis is helpful and convenient because it makes it possible to see clearly the many different types of heart in which completely common AV canal occurred. In this way, the segmental anatomic combinations are conveyed quickly, briefly, and precisely, no matter how complex they may be ( Table 11.3 ).

The Common AV Valve and the AV Septal Defect

The salient features of the complete form common AV canal are presented in Table 11.4 .

Table 11.4

Salient Anatomic Features of Completely Common AV Canal ( n = 266)

Anatomy	No. of Cases	% of Series	95% CI
Type A	115	43.23	37–49
Type B	3	1.13	0–2
Type C	108	40.60	35–47
Not typed	40	15.04
RV type	55	20.68	16–26
LV type	21	7.89	5–11
Regurgitation of CAVV	13	4.89
Potentially parachute MV	10	3.76
Double-orifice MV	7	2.63
Ebstein’s anomaly of TV component of CAVV	3	1.13
Congenital MS	3	1.13
Mitral atresia	3	1.13
Double-orifice TV	2	0.75
Tricuspid atresia	1	0.38
Tricuspid stenosis	1	0.38
Mitral regurgitation	1	0.38
Muscular MV	1	0.38
Muscular “island” in superior leaflet	1	0.38

CAVV, Common atrioventricular valve; CI, confidence interval; MS, mitral stenosis, i.e., congenital stenosis of mitral component of CAVV; MV, mitral valve, i.e., mitral component of CAVV; TV, tricuspid valve, i.e., tricuspid component of CAVV.

Type A ( Fig. 11.2 ) completely common AV canal (43%) was only slightly more common than type C ( Fig 11.4 ) (41%), whereas type B ( Fig. 11.3 ) was rare (1%) ( Table 11.4 ).

A sizeable minority of cases (15%) was not assigned a Rastelli type for a variety of reasons (e.g., surgery, artifact, or inapplicability).

Inapplicability? Yes. For the Rastelli classification to apply, a relatively well developed and normally positioned ventricular septum must be present. The Rastelli classification of the complete form of common AV canal breaks down when a relatively well developed and approximately normally positioned ventricular septum is not present. Examples of inapplicability include common AV valve with common-inlet RV, or common-inlet LV, or tricuspid atresia, or mitral atresia, or single LV, or single RV.

The right ventricular type of completely common AV canal (21%) (the situation in which the common AV valve opens predominantly into the morphologically RV, because the morphologically LV is underdeveloped or absent) was almost three times as common as the left ventricular type (8%) (the situation in which the common AV valve opens predominantly into the morphologically LV, because the sinus, body, or inflow tract of the morphologically right ventricular sinus is hypoplastic or absent).

The foregoing data focus on the contributions to the classification of completely common AV canal of Rastelli and his colleagues, Edwards and his associates, and Bharati and Lev. ^, ^, Infrequently, however, we encountered other anatomic findings of considerable diagnostic and surgical importance that often are not mentioned in association with completely common AV canal ( Table 11.4 ):

Significant regurgitation of the common AV valve was encountered in 13 patients (5%).
Potentially parachute mitral valve, mentioned previously, was found in 10 patients (4%) ( Figs. 11.13, 11.15, and 11.16 ).
Double-orifice mitral valve was present in 7 cases (3%) ( Figs. 11.20 and 11.22 ).
Ebstein’s malformation of the tricuspid component of the common AV valve was found in 3 patients (1%).
Mitral stenosis , that is, congenital stenosis of the mitral component of the common AV valve, was present in 3 cases (1%).
Mitral atresia, that is, atresia of the mitral component of the common AV valve, was found in 3 patients (1%).
- How does “mitral atresia” occur in the setting of common AV canal? If the common AV valve opens essentially only into the RV, and if the leftward or mitral component of the common AV valve becomes adherent to the crest of the muscular ventricular septum, then no blood can pass directly from the left atrium into the left ventricle. Because there is no left-sided atrioventricular inlet, all of the left atrial (LA) blood has to shunt from left-to-right into the right atrium (RA) via the ostium primum type of atrial septal defect and thence into the RV. Blood can enter the small LV only by right-to-left shunting at the ventricular level. Hence, absence of a direct left atrial–to–left ventricular communication is what we mean by mitral atresia with common AV canal.
Double-orifice tricuspid valve , that is, two orifices in the tricuspid component of the common AV valve, was observed in 2 cases (0.75%).
Tricuspid atresia , that is, atresia of the tricuspid component of the common AV valve, was found in 1 patient (0.38%). This type of tricuspid atresia is characterized by adherence of the tricuspid component of the common AV valve to the crest of the ventricular septum, preventing direct communication between the RA and the RV. Hence, the blood has to shunt right-to-left at the atrial level (RA to LA), and left-to-right at the ventricular level (LV to RV).
Tricuspid stenosis , that is, congenital stenosis of the tricuspid component of the common AV valve, was found in 1 case (0.38%).
Congenital mitral regurgitation , that is, regurgitation of the mitral component of the common AV valve, was present in 1 patient (0.38%).
A muscular mitral valve component of the common AV valve was observed in 1 patient (0.38%). Another had an “island” of muscular tissue in the superior leaflet of the common AV valve.

The observation that the mitral component of the common AV valve can be muscular is both fascinating and perplexing. This observation suggests that the mitral valve is, in part, myogenic. Hence, these fascinating cases pose the question: Are the endocardial cushions derived in part from the myocardium (as well as from the endocardium)?

Additional Interatrial Communications

In addition to an ostium primum type of atrial septal defect (an incomplete AV septal defect), many of these 266 cases of completely common AV canal had additional interatrial communications:

1.
A secundum type of atrial septal defect (i.e., an ostium secundum type of atrial septal defect) was present in 120 patients (45.11%) (95% CI 39% to 51%).
2.
A common atrium (an essentially absent interatrial septum) was found in 47 cases (17.67%) (95% CI 13% to 22%). Dr. Jesse Edwards used to say that common atrium is the forgotten type of common AV canal. We would agree and would add that there are several different settings in which common atrium occurs:
- a.
  with common AV canal, complete or partial, in visceroatrial situs solitus;
- b.
  with the Ellis-van Creveld syndrome, in which the common atrium is associated with a divided AV canal (separate mitral and tricuspid valves), but the mitral valve may have a cleft anterior leaflet of the AV canal type, and a persistent left superior vena cava may be unroofed into the LA;
- c.
  with visceral heterotaxy and polysplenia, with two AV valves, and often with a cleft anterior mitral leaflet of the AV canal type; and
- d.
  with visceral heterotaxy and asplenia, typically with a common AV valve.
3.
Coronary sinus septal defect (unroofing of the coronary sinus into the LA) was found in 1 patient (0.38%). Although a coronary sinus septal defect is not an atrial septal defect, it acts like one physiologically. Typically, bright red blood from the left atrium passes through the defect in the posterior wall of the LA and the anterior wall of the coronary sinus. In this way, the left atrial blood enters the coronary sinus and passes into the RA through an only mildly enlarged right atrial coronary sinus ostium; hence, an ASD-like left-to-right shunt is present. If the surgeon has the patient on cardiopulmonary bypass when the right atrial septal surface is inspected, he or she may well not see the tell-tale bright red blood issuing from the mouth of the coronary sinus. The surgeon may then conclude (correctly) that the patient does not have an atrial septal defect, but may not suspect the presence of a coronary sinus septal defect that cannot be seen from the right atrial perspective. Preoperative or intraoperative diagnostic evaluation—such as transesophageal echocardiography with color-flow Doppler interrogation—should prevent misdiagnosis. (Parenthetically, although the right atrial ostium of the coronary sinus is only slightly enlarged with typical coronary sinus septal defect, the coronary sinus ostial enlargement with totally anomalous pulmonary venous connection to the coronary sinus is not subtle: it’s huge. This difference is helpful in the differential diagnosis of atrial level left-to-right shunts without an atrial septal defect.)
4.
The Raghib syndrome was present in 1 patient (0.38%). This syndrome is characterized by a large low posterior opening in the atrial septum (interpreted as an enlarged right atrial ostium of the coronary sinus), with extensive unroofing of a persistent left superior vena cava to coronary sinus into the LA (see Chapter 5 ).
5.
The primum atrial septal defect was tiny in 1 of these 266 cases of completely common AV canal (0.38%). This patient was a 6-year-old boy with visceral heterotaxy, polysplenia, and DORV {A(S),D,D} with a subpulmonary conus. In addition to a tiny ostium primum type of ASD, he also had a moderate sized secundum ASD. He had aortic outflow tract stenosis, had preductal coarctation of the aorta, and was status post–subclavian flap aortoplasty for coarctation of the aorta.
6.
Leftward deviation of septum primum was found in 3 cases (1.13%).
7.
Septum primum attached to the right of septum secundum in 1 patient who was thought to have visceroatrial situs solitus. This 3-day-old female infant had DORV {S,D,D}, asplenia, bilateral conus, subpulmonary stenosis, unicommissural pulmonary valve, and hypoplastic pulmonary arteries.

Why did we think this asplenic patient had atrial situs solitus? Because the inferior vena cava was right-sided; the superior vena cava was right-sided; there was no left superior vena cava; and the left hepatic veins connected with an unroofed coronary sinus. There was totally anomalous pulmonary venous connection below the diaphragm. The right atrial appendage was enlarged, and the left atrial appendage was small. However, the right lung was bilobed and the left lung was trilobed, suggesting situs inversus of the lungs.

Whenever the right-left relationship of septum primum relative to septum secundum has been surprising, as in this case, visceral heterotaxy has always been present. We have concluded that septum primum occasionally can lie on the right atrial sided of septum secundum in the heterotaxy syndromes. Why? Perhaps because the superior limbic band of septum secundum is often poorly developed in the heterotaxy syndromes, and consequently septum primum may not attach normally to septum secundum. This is our best present hypothesis to explain surprising septum primum/septum secundum right-left relationships, as in this case.

An alternative hypothesis, that we doubt, is that our diagnosis of the atrial situs in such heterotaxic cases is wrong.

To summarize this intriguing observation in a factual, nonspeculative way: Occasionally in the heterotaxy syndrome (with asplenia or polysplenia), septum primum can be on the apparently right atrial side of septum secundum—instead of vice versa, which is normal.

The Ventricular Septal Defects

1.
An isolated VSD of the atrioventricular canal (AVC) type was found in 233 of these 266 cases of completely common AV canal (88%) (95% CI 84% to 92%). The “scooped-out” crest of the muscular ventricular septum forms the anteroinferior rim of the VSD below the plane of the AV valve leaflets ( Figs. 11.2 to 11.4 ). Above the plane of the AV valve leaflets, the superoposterior rim of the ostium primum type of atrial septal defect (ASD I) is formed by the anteroinferior margin of the atrial septum, that is, the so-called inferior limbic band of septum secundum. Hence, the AV septal defect in completely common AV canal is formed by the VSD of the AV canal type plus the ASD I, these defects being confluent ( Figs. 11.2 to 11.4 ).

We prefer not to call a VSD of the AV canal type an “inlet VSD” because one can also have a muscular VSD in the inlet portion of the muscular ventricular septum. In other words, there are two very different anatomic types of “inlet” VSD, whereas VSD of the AV canal type is specific and therefore clear.

In somewhat greater detail, what is the difference between a VSD of the AV canal type and a muscular inlet VSD? A VSD of the AV canal type is confluent with the tricuspid valve or with the common AV valve, whereas a muscular inlet VSD is not confluent with the tricuspid valve or with the common AV valve—because the muscular inlet VSD is separated from the tricuspid valve or the common AV valve by septal myocardium. Thus, there are two very different anatomic types of “inlet” VSD, which the above-mentioned terminology helpfully distinguishes.

2.
VSD of the AV canal type confluent with a conoventricular VSD was present in 17 of these 266 cases of completely common AV canal (6%) (95% CI 3% to 9%). A conoventricular VSD is one that extends superiorly beneath the conal septum, as in the tetralogy of Fallot in which the conal septum is malaligned anterosuperiorly.
3.
VSD of the AV canal type plus muscular VSD was found in 16 patients (6%) (95% CI 3% to 9%).

Regarding the first type of VSD mentioned above, isolated VSD of the AV canal type means (1) that there is no anterosuperior extension beneath the conal septum (no conoventricular VSD component) and (2) that there is no additional muscular VSD(s). Thus, an isolated VSD of the AV canal type is purely an “inlet” VSD, with no “outlet” component (no conoventricular VSD component), and with no additional muscular VSD(s).

Hence, in this series of 266 cases of the complete form of common AV canal (complete AV septal defect), all had a VSD of the AV canal type (100%), 6% had outlet extension of the VSD, and 6% had additional muscular VSD(s).

The Right Ventricle

1.
Hypoplasia of the morphologically RV sinus was present in 40 of these 266 patients with completely common AV canal (15%) (95% CI 10% to 19%).
2.
Absence of the RV sinus , resulting in single LV with common-inlet LV was found in 2 patients (0.75%).
3.
Common AV canal with tricuspid atresia was observed in 1 case (0.38%).
4.
Superoinferior ventricles with RV superior, LV inferior, and ventricular septum approximately horizontal were present in 4 patients (1.5%).
5.
Double-chambered RV , also known as anomalous muscle bundles of the RV, was observed in 1 patient (0.38%).

Thus, the RV sinus (body or inflow tract) was hypoplastic or absent in 42 patients (16%). Consequently, the RV was well developed in the great majority of these patients with completely common AV canal (84%).

The Left Ventricle

1.
The morphologically LV was hypoplastic in 66 of these 266 cases of completely common AV canal (25%) (95% CI 20% to 30%).
2.
Potentially parachute mitral valve was found in 5 patients (1.88%). They all had only one focus of insertion of the primary chordae tendineae: into the anterolateral papillary muscle group of the LV, not into the posteromedial papillary muscle group ( Fig. 11.13 ).
3.
A small diverticulum of the LV was present in 1 case (0.38%).
4.
Criss-cross AV relations were observed in 1 patient (0.38%).

In conclusion, LV sinus hypoplasia (66/266 patients, 25%) was significantly more frequent than RV sinus hypoplasia, or atresia, or absence (43/266 patients, 16%) in the complete form of common AV canal ( p < .01).

The Pulmonary Outflow Tract

1.
Pulmonary outflow tract stenosis was found in 71 of these 266 cases of the complete form of common AV canal (27%) (95 CI 21% to 32%).
2.
Pulmonary outflow tract atresia occurred in 22 of the 266 patients (8%) (95% CI 5% to 12%).
3.
Tetralogy of Fallot was present in 18 of these 266 patients (7%) (95% CI 4% to 10%). Inverted tetralogy of Fallot (TOF) was found in 1 case, a 2{11/12}-year-old girl with polysplenia and TOF {A(S),L,I}, interruption of the inferior vena cava, prominent azygos vein to the right superior vena cava, suprahepatic segment of the inferior vena cava connecting with the RA, and coronary sinus opening into the LA.
4.
Truncus arteriosus type A2 was present in 1 of these 266 cases of completely common AV canal (0.38%).

The Aortic Outflow Tract

1.
Aortic outflow tract stenosis was found in 47 of 266 cases of the complete form of common AV canal (18%) (95% CI 13% to 22%).
2.
Subaortic stenosis produced by the anterolateral muscle of the left ventricle , also known as the muscle of Moulaert, was found in 3 cases (1.13%).

Hence, aortic outflow tract obstruction was present in a total of 50 of 266 cases (19%).

In conclusion , pulmonary outflow tract obstruction (93/266 cases, 35%) was significantly more frequent than aortic outflow tract obstruction (50/266, 19%) in the complete form of common AV canal ( p < .0001).

The Pulmonary Valve

1.
Pulmonary valvar stenosis was present in 46 of 266 cases of completely common AV valve (17%).
2.
Pulmonary valvar atresia (congenital) occurred in 28 of the 266 cases (11%).
3.
Acquired pulmonary atresia was observed in 2 patients (0.75%). Acquired means that the atresia was not present at birth, but was acquired postnatally.
4.
Bicuspid pulmonary valve was found in 39 of 266 cases (15%).
5.
Unicuspid (unicommissural) pulmonary valve occurred in 7 of 266 patients (3%).
6.
Hypoplasia of the pulmonary valve was observed in 6 cases (2%).
7.
A domed pulmonary valve was described in 4 patients (1.5%).
8.
A myxomatous pulmonary valve was found in 2 cases (0.75%).

Thus, pulmonary valvar obstruction was present in 76 patients (29%) (95% CI 23% to 34%): (a) valvar pulmonary stenosis in 46 of 266 patients (17.29%); and (b) valvar pulmonary atresia in 30 of 266 patients (11.28%).

You're Reading a Preview

Become a Clinical Tree membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here

Acknowledgment

Definition

History and terminology

Morphology and Classification

Type A Completely Common AV Canal

Type B Completely Common AV Canal

Type C Completely Common AV Canal

Partial Forms of Common AV Canal

Embryology

Etiology

Parachute Mitral Valve

Double-Orifice Mitral Valve

Double-Orifice Mitral Valve Principles

Findings

Complete Forms of Common AV Canal

Associated Anomalies

Heart Position

Conclusions

Summary

Types of Ventricular Loop

Types of AV Alignment

Ventriculoarterial Alignments

Conclusions

Conus

Segmental Sets or Combinations

The Common AV Valve and the AV Septal Defect

Additional Interatrial Communications

The Ventricular Septal Defects

The Right Ventricle

The Left Ventricle

The Pulmonary Outflow Tract

The Aortic Outflow Tract

The Pulmonary Valve

You're Reading a Preview

Related Posts

Post Scriptum

Understanding Normally and Abnormally Related Great Arteries

Conclusions

The Heterotaxy Syndromes: Asplenia, Polysplenia, and With Normally Formed but Right-Sided Spleen