Definition
The morphologically right atrium (RA) opens into both the morphologically right ventricle (RV) and the morphologically left ventricle (LV) via two atrioventricular (AV) valves resulting in double-outlet right atrium (DORA), or via a common AV valve resulting in common-outlet right atrium (CORA).
Anatomic Types of Double-Outlet Right Atrium/Common-Outlet Right Atrium
Three anatomic types of DORA or CORA are known at the present time: (1) leftward malalignment of the atrial septum relative to the atrioventricular valve(s); (2) rightward malalignment of the ventricular septum relative to the atrial septum; and (3) an accessory valve between the coronary sinus and the LV.
Double-Outlet Right Atrium and Common-Outlet Right Atrium With Leftward Malalignment of the Atrial Septum
The concepts of DORA and double-outlet left atrium (DOLA) were first proposed in 1977 by Van Mierop in the setting of common atrioventricular (AV) canal, also known as endocardial cushion defect. His concept was that DORA is present when the RA opens in the RV and into the LV via a common AV valve. Van Mierop did not distinguish between DORA and CORA.
Equally noteworthy is that left atrial outlet atresia is present, and that the only exit from the LA is an atrial septal defect (ASD) that can be hemodynamically restrictive, resulting in pulmonary venous obstruction. Also noteworthy is the fact that the ASD can be of the ostium primum type (consistent with common AV canal), or an ASD of the ostium secundum type, or even of the sinus venosus defect type. Van Mierop did not present any cases of DORA (or of DOLA); instead, he coined these terms and defined these concepts.
We have regarded this anatomic type of DORA as common AV canal with left atrial outlet atresia (see Chapter 11 ).
By now it has become clear that this anatomic type of DORA is associated with (or perhaps caused by) leftward malposition of the atrial septum relative to the AV valve(s). This is why there can be LA outlet atresia with significant pulmonary venous obstruction, depending on the patency (or otherwise) of the atrial septum.
DORA with levomalposition of the atrial septum relative to the AV valves in visceroatrial situs solitus appears to be the most common form of this rarity. To the best of our knowledge, Utley, Noonan, Walters, and Frist first reported DORA with an anomalous position of the atrial septum that was displaced posteriorly and to the left of the mitral valve in a 2½-year-old boy ( Fig. 12.1 ). The interatrial communication was small, was probably restrictive, and was interpreted as an ostium primum type of atrial septal defect. The mitral valve was thought not to be cleft, but angiocardiography revealed a goose-neck deformity suggestive of common AV canal. The goose-neck deformity is the diastolic angiocardiographic picture in common AV canal. Consequently, their patient appears to have a rare partial form of common AV canal with a restrictively small ASD I, no mitral cleft, an intact ventricular septum, and a goose-neck deformity. No coronary sinus ostium was found in the right atrium. Instead, the coronary sinus, the left upper pulmonary vein, and a left superior vena cava were confluent with a common channel that opened into the roof of the left atrium. This patient had bilateral superior venae cavae with absence of the left innominate vein.
Was the left atrial appendage part of the left atrium? This important point in differential diagnosis remains unclear. If the left atrial appendage lay below the leftwardly displaced atrial septum, then cor triatriatum (see Chapter 8 ) was present, not DORA.
Did this patient have a levoatrial cardinal vein decompressing this obstructed LA? This interesting diagnostic possibility also remains unresolved.
With the aforementioned uncertainties borne in mind, I think it is reasonable to regard Utley’s case as the first known example of DORA with leftward displacement of the atrial septum. The malpositioned atrial septum was excised and replaced by a normally positioned pericardial patch, and the persistent left superior vena cava was ligated above the aberrant left upper pulmonary vein. It should be added that whether or not the correct diagnosis was DORA or cor triatriatum, the surgical management of Utley et al was appropriate for both: excision of the obstructive and leftwardly malaligned atrial septum or intra-left-atrial membrane, and replacement with a normally located atrium septum.
In 1983, Corwin and his colleagues reported the first indubitable case of DORA with leftward deviation of the lower portion of the atrial septum, again in a 2½-year-old boy. These investigators noted that the levomalpositioned atrial septum inserted into the left atrial wall to the left of and posterior to the mitral valve. The only outlet from the obstructed left atrium was via a small secundum atrial septal defect. These authors stated that the left atrial appendage was in the high-pressure left atrium along with the pulmonary veins, thereby excluding the diagnosis of cor triatriatum sinistrum. The RA opened into the RV and the LV via a common AV valve. There was no ventricular septal defect (VSD).
Parenthetically, it is noteworthy that common AV canal was present in the form of a common AV valve. Also, both the mitral valve and the tricuspid valve were described as cleft. But no atrioventricular septal defect was present; that is, there was no ostium primum type of atrial septal defect, and there was VSD. This is just one example of why we prefer the diagnosis of common AV canal to that of AV septal defect. The concept of common AV canal is more inclusive: it embraces partial forms of common AV canal, even when an AV septal defect is not present.
It should be added that we like the diagnosis of AV septal defect when this anomaly is present, completely or in part. But we prefer the overall diagnosis of common AV canal because this diagnosis is more inclusive, more flexible, and more sensitive than AV septal defect, which is only one part of common AV canal, as the patient of Corwin, Singh, and Karlson demonstrates.
Despite the absence of a VSD, this little boy had moderate pulmonary artery hypertension (57/19 mm Hg), congestive heart failure, and failure to thrive—all of which were thought to be related to the presence of significant pulmonary venous obstruction (left atrial outlet atresia with a small secundum ASD).
The patient also had multiple congenital anomalies with dysmorphic facies: bilateral epicanthal folds, blepharophimosis (narrowing of the opening of the eyelids), mild ptosis of the eyelids, and a coloboma of the right iris.
Successful surgical management of this anatomic type of DORA involved excision of the levodeviated atrial septum, and its replacement with a normally located pericardial atrial septum.
Thus, in 1983, Corwin et al confirmed the existence of DORA with marked leftward deviation of the atrial septum, left atrial outlet atresia, and pulmonary venous obstruction, and reported successful surgical management of this rare anomaly. Further confirmation and description subsequently occurred with the publications of Pérez-Martínez et al (1984), Alivizatos et al (1985), Westerman et al (1986), Suzuki et al (1994), and Radermecker et al (1995). What else did we learn? Much, as is summarized below.
Such patients could present almost as teenagers, not just as very young children: a 12-year-old boy and a 12-year-old girl.
The right superior vena cava could be atretic, with a persistent left superior vena cava (LSVC), making it necessary surgically to tunnel the LSVC to the RA, as well as excising the levomalpositioned atrial septum and replacing it.
In the 12-year-old girl, the only outlet from the LA was a small ostium primum type of ASD (6 mm in diameter). The mean left atrial pressure was 25 mm Hg. At surgery, lung biopsy interpreted by Dr. Sheila Haworth showed large thick-walled pulmonary veins with pronounced intimal fibrosis. The small pulmonary arterial changes were much milder than the pulmonary venous histopathology: an increase in arterial medial muscularity, but without intimal proliferation (grade 1 Heath-Edwards changes in the small pulmonary arteries). Mr. Jaroslav (“Jarda”) Stark found severe regurgitation through the mitral cleft into the RA (none into the left); hence he closed the cleft with sutures. He also excised the leftwardly deviated atrial septum and replaced it with a correctly positioned pericardial patch.
Westerman et al documented the fact that the segmental anatomy in this anatomic type of DORA is not always normal, that is, {S,D,S}—as it had been in all previously published cases. In Westerman’s 3-year-old male patient, the segmental anatomy was tetralogy of Fallot (TOF) {S,D,S}. TOF has nearly normally related great arteries, but not entirely normally related great arteries (see Chapter 20 ).
These authors found that the RA opened through a common AV valve into both ventricles. Accurately speaking, this is common-outlet right atrium, not double -outlet right atrium, because a common AV valve connected the RA with both ventricles. Two AV valves were not present—only one, a common AV valve. This 3-year-old boy had no ostium primum type of atrial septal defect. There was left atrial outlet atresia, the only outlet from the LA being an ostium secundum type of atrial septal defect that was 1 cm in diameter. The atrial septum deviated to the left of the common AV valve, and the coronary sinus was noted to drain into the RA, which it did not do in the cases of Utley and Pérez-Martínez. Westerman’s patient also had a brain abscess. The common AV valve was type C of Rastelli, and there was an underlying VSD of the AV canal type.
Hence, this patient had an interesting form of partial common AV canal: no primum atrial septal defect, but with a VSD of the confluent AV canal and TOF types, with a type C common AV valve. We have noted repeatedly that when common AV canal is associated with tetralogy of Fallot, the primum ASD is often unusually small or absent, as in Westerman’s patient, resulting in an unusual form of partially common AV canal.
In addition to repairing the stenotic pulmonary outflow tract, these authors enlarged the restrictive atrial septal defect and used one patch to close the ASD and the VSD and to subdivide the common AV valve.
Thus, the report of Westerman, Norton, and Van Devanter was important because it documented CORA opening into both ventricles, and because it showed that an important conotruncal malformation (TOF) can coexist with malalignment of the atrial septum to the left of the AV junction and valve(s).
The 18-month-old patient reported by Starc, Bierman, Bowman, Steeg, Wang, and Krongrad (Case 3) had Down syndrome with completely common AV canal and CORA. The atrial septum deviated to the left and attached to the lateral wall of the LA above the plane of the AV groove. The left atrial appendage was part of the high-pressure left atrium (excluding the diagnostic possibility of cor triatriatum). A persistent left superior vena cava returned to the coronary sinus. The levomalposed atrial septum was excised and double-patch closure with repositioning was used to repair the complete form of common AV canal that was present. The thrust of this report by Starc et al was that DORA or CORA and cor triatriatum are surgically correctable forms of pulmonary venous obstruction.
Ahmadi and colleagues reported a 4-year-old boy with the Ellis-van Creveld syndrome. The atrial septum was deviated far to the left resulting in DORA, left atrial outlet atresia, and severe pulmonary venous obstruction. The primum ASD was described as small and the ventricular septum was intact. The mean left atrial pressure was 28 mm Hg, and the mean right atrial pressure was 6 mm Hg. Systemic arterial desaturation was noted (89%). The cleft anterior leaflet of the mitral valve permitted a Gerbode-like LV-to-RA shunt. A persistent left superior vena cava was present.
The authors made the most interesting observation that the left atrium as a whole was malpositioned. The LA was located to the left, above and posterior to the RA. In other words, this form of DORA or CORA that has been described as having leftward malposition of the atrium septum may well in fact have malposition of the whole LA, or malposition of the atrial segment of the heart relative to the ventriculoarterial part of the heart (the bulboventricular loop).
It is probably inevitable that this anatomic type of DORA/CORA would be described as having leftward malposition of the arterial septum relative to the AV valve(s)—because this is how it looks. But one should remember that cardiac septa usually are where they are because of what is happening on either side of the septa, or below the septa, or elsewhere. I am thinking of the location of the ventricular septum in single ventricle, and in straddling tricuspid and mitral valves, and of the configuration of the aortopulmonary septum in conotruncal malformations (see Chapter 17, Chapter 22, Chapter 23, Chapter 24, Chapter 25 ). Suffice it to say that malposition of a cardiac septum typically does not indicate malseptation as a primary morphogenetic mechanism. Malposition of a cardiovascular septum is usually related to an abnormality somewhere else.
If the same principle applies to DORA and CORA, it may well be that the atrial septum is in fact normally located and that the ventriculoarterial parts of the heart are malaligned—displaced to the right relative to the normally located atria and atrial septum . Developmentally, the atria are relatively fixed in position by the great veins (the venae cavae) and the pulmonary veins; whereas the bulboventricular parts of the heart are “professional contortionists” and hence prone to malalignment (see Chapter 2, Chapter 3, Chapter 4 ).
This hypothesis of rightward malalignment of the ventricles and the AV valves relative to the atria and the atrial septum is mentioned here in the interests of in-depth understanding. No change in terminology is being suggested. Leftward malposition of the atrial septum relative to the AV valve(s) should be understood as an anatomic description, not as an embryologic hypothesis or developmental mechanism.
Another problem merits mention: In this anatomic type of DORA or CORA with leftward malalignment of the atrial septum relative to the AV valve(s) and left atrial outlet atresia with pulmonary venous obstruction, what is/are the AV valve/valves? Does left atrial outlet atresia mean that mitral atresia is present? If so, then the AV valve or valves that underlie the RA and open into the RV and the LV are in fact the tricuspid valve that straddles the ventricular septum. If the straddling tricuspid valve, in this interpretation, attaches to the ventricular septum, the result is two AV valves that resemble the tricuspid valve (entering the RV) and the mitral valve (entering the LV). But if the straddling tricuspid valve, in this interpretation, does not become adherent to the ventricular septum, then it morphologically resembles a common AV valve.
So the problem in this anatomic type of DORA or CORA is: What is the AV valve (or valves)? Are we dealing with a common AV valve or a divided AV valve (tricuspid and mitral)? Or are we dealing with mitral atresia and a straddling tricuspid valve that may or may not be subdivided into two AV valves—one opening into the RV and the other opening into the LV?
I don’t really know the answer to this question. However, I suspect that we are not really dealing with mitral atresia, that is, with atresia of the mitral valve per se . Note that the left ventricle typically is well developed in DORA and CORA, whereas in “real” mitral atresia, the LV typically is diminutive. This is why it is generally known as left atrial outlet atresia (due to malposition of the atrial septum), rather than as mitral atresia. However, as indicated above, we don’t really know for sure. I think it is very important to realize what we know, and also what we don’t know. There is so much that we really don’t know for sure. Intellectual honesty requires that this be stated clearly. That said, I think we are dealing with left atrial outlet atresia (different from typical mitral atresia) and with a common AV valve or an AV valve that has divided into mitral and tricuspid valves. These two AV valves could also be called by other names, such as the RA-to-RV valve and the RA-to-LV valve. Instead, we usually call the AV valve entering the RV the tricuspid valve and the AV valve entering the LV the mitral valve.
To summarize, Ahmadi et al observed that the entire LA is abnormally located relative to the RA in this anatomic type of DORA/CORA—a very important observation (as above). Surgically, these authors resected the malpositioned atrial septum, replaced it with an appropriately positioned Dacron atrial septum, and suture closed the mitral cleft.
Suzuki and colleagues reported an 18-month-old boy with CORA. The common AV valve had a free floating anterior leaflet (type C of Rastelli). The marked leftward malalignment of the atrial septum relative to the common AV valve resulted not only in CORA, with the RA opening into both the RV and the LV, but also in left atrial outlet atresia with severe pulmonary venous obstruction and systemic pulmonary arterial hypertension. At 13 months of age, this patient with Down syndrome underwent palliative surgical management with atrial septal resection and inadequate banding of the main pulmonary artery (MPA). Complete repair was not attempted because of the coexistence of severe pulmonary vascular disease. The small pulmonary arteries showed not only medial hypertrophy, but also intimal proliferation or total luminal occlusion (grade 3 Heath-Edward changes). The pulmonary veins displayed medial hypertrophy. Adequate pulmonary artery banding was prevented by the development of cyanosis and bradycardia. Following what was thought to be inadequate MPA banding, the pressure in the MPA was 58/22 mm Hg and systemic arterial pressure was 75/44 mm Hg.
Thus, the patient of Suzuki et al indicates that very severe occlusive disease of the small pulmonary arteries can occur by 12 months of age in a patient with Down syndrome, CORA, left atrial outlet atresia, and marked pulmonary venous obstruction produced by the malalignment of the atrial septum to the left of the common AV valve. The ASD was only 1 to 2 mm in diameter.
Radermecker, Chauvaud, and Carpentier reported the oldest patient to date with this anatomic type of DORA: a 14-year-old girl. She had multiple congenital anomalies consisting of pectus excavatum and mild psychomotor impairment. Preoperatively she was misdiagnosed as having congenital mitral stenosis. At surgery, Carpentier and his colleagues found that their patient had a partial form of common AV canal ( Fig. 12.2 ). However, the ostium primum atrial septal defect was restrictively small (1 cm in diameter) and the atrial septum was displaced posteriorly and to the left of the left AV valve (a cleft mitral valve). Radermecker and his colleagues prefer to regard this cleft as a commissure. They also prefer to call this the left atrioventricular valve (not the mitral valve) because this left AV valve is trifoliate: it has three leaflets, if one regards the cleft as a commissure. (With respect, we think the cleft is a cleft, not a commissure; see Chapter 11 .) This patient also had a persistent left superior vena cava returning to the coronary sinus.
Surgically, the authors resected the malpositioned atrial septum and replaced it with a normally located atrial septum using glutaraldehyde-treated pericardium.
Radermecker, Chauvaud, and Carpentier stated: “We present a case to draw attention to this rare diagnosis (double-outlet right atrium) as an unusual cause of impediment to left AV valve inflow and to include it in the differential diagnosis of mitral stenosis with basically normal subvalvar apparatus.”
Exactly right. The differential diagnosis of patients that appear physiologically to have congenital mitral stenosis should include the following:
- •
congenital mitral stenosis;
- •
Shone syndrome;
- •
cor triatriatum (sinistrum);
- •
double-outlet right atrium; and
- •
common-outlet right atrium.
There are, of course, additional conditions such as stenosis of individual pulmonary veins.
The problem that Radermecker and colleagues were getting at, although they were perhaps too polite to say so, is that still many pediatric cardiologists, radiologists, and cardiac surgeons have never heard of DORA or CORA with leftward malposition of the atrial septum, left atrial outlet atresia, and pulmonary venous obstruction ( Figs. 12.1 to 12.2 ).
Understanding the disorder is at least 50% of the cure, which is why pathology is so important. In our field, the disorder typically is the pathology—and its physiologic sequelae. The pathology is what we try to diagnose with accuracy. And the pathology is what we try to modify surgically and interventionally, in a physiologically advantageous way.
Next we are going to consider an even less well known form of double-outlet right atrium.
Double-Outlet Right Atrium With Rightward Malalignment of the Ventricular Septum
This is the straddling tricuspid valve type of DORA, except that the straddling tricuspid valve has become adherent to the crest of the muscular, rightwardly malaligned ventricular septum. Consequently, the straddling tricuspid valve (TV) has become subdivided into two AV valves: the rightward that opens from the RA into the small RV; and the leftward TV that opens from the RA into the large LV. The adherence of the straddling TV to the underlying rightwardly malaligned ventricular septum has occluded the VSD of the AV canal (inlet) type that usually is associated with typical straddling TV. Consequently, there are three AV valves: the right-sided TV connecting the RA with the small RV; the left-sided TV connecting the RA with the large LV; and the mitral valve connecting the LA with the large LV. DORA and double-inlet left ventricle (DILV) may both be present. The mitral valve can be unobstructed, or stenotic, or atretic. If the mitral valve is atretic, then DILV is not present in this anatomic type of DORA.
Why is the ventricular septum malaligned to the right, and why is this rightward malalignment important? We think that the ventricular septum is malaligned to the right relative to the atrial septum (with D-loop ventricles) because the RV sinus (inflow tract) is underdeveloped or poorly expanded. So rightward malalignment of the ventricular septum appears to be of basic importance in this anatomic type of DORA, just as it is with typical straddling TV (in which the TV does not adhere to the ventricular septum).
DORA with rightward malalignment of the ventricular septum was first reported by Büchler, Rabelo, Marino, David, and Van Praagh at the World Congress of Paediatric Cardiology in London, England, in 1980 ( Fig. 12.3 ). Another very similar example of DORA with rightward malposition of the ventricular septum was reported by Birincioğlu and colleagues in 1999 ( Fig. 12.4 ).
