1274
6
Disharmony Between Atrioventricular Connections and Segmental
Combinations: Unusual Variants of "Crisscross" Hearts
ROBERT H. ANDERSON, MD, FRCPATH, AUDREY SMITH, PHD,
JAMES L. WILKINSON, MB, CHB, FRCP
Liverpool, United Kingdom
Two congenitally malformed hearts are described, one
studied in life and the other at autopsy, in which the
topology of the ventricular mass was not as expected for
the atrioventricular (AV) connection present. In Case 1,
studied at autopsy, there was the usual atrial arrange-
ment with a concordant AV connection. The morpho-
logically right ventricle, however, was left-sided and the
ventricular mass was of left-hand topology (I loop) rather
than the expected right-hand pattern (d loop). In Case
2, studied during life, there was the usual atrial ar-
When Van Praagh et al. (1-3) introduced the concept of
atrioventricular (AV) concordance and discordance, they
used the terms to describe segmental harmony between the
topological arrangements of the atrial chambers and the
ventricular mass. Thus, concordance described the combi-
nation of usual atrial arrangement (solitus) with right-hand
ventricular topology (d loop) or mirror image atrial arrange-
ment (inversus) with left-hand topology (lloop). Atrioven-
tricular discordance accounted for the reverse combinations.
These diagnoses were made irrespective of the precise con-
nections between atria and ventricles, which were described
additionally.
When Kirklin et al. (4) modified the Van Praagh ap-
proach, they used concordance and discordance to describe
connections, and the European school (5,6) followed this
lead when promoting the importance of connections between
the three cardiac segments. Within this approach, concord-
ance was used only to describe cases with the morpholog-
ically right atrium connected to the morphologically right
From the Institute of Child Health, Alder Hey Children's Hospital,
Liverpool, United Kingdom. Dr. Anderson is the Joseph Levy Professor
of Paediatric Cardiac Morphology, Cardiothoracic Institute, Brompton
Hospital, London, United Kingdom, and is supported by the British Heart
Foundation, London, United Kingdom. Drs. Smith and Wilkinson are
supported by the National Heart Research Fund, London, United Kingdom.
Manuscript received December 22, 1986; revised manuscript received
July 9, 1987, accepted July 20, 1987.
Address for reprints: Professor Robert H. Anderson, MD, Department
of Paediatrics, Cardiothoracic Institute, Brompton Hospital, Fulham Rd,
London, SW3 6HP, United Kingdom.
1987 by the American College of Cardiology
lACC Vol. 10, No.
December 1987:1274-7
rangement and a discordant AV connection. Rather than
the anticipated left-hand topology (I loop), the ventric-
ular mass was arranged in right-hand fashion (d loop)
and the morphologically right ventricle was right-sided.
The cases emphasize that for a full description of a con-
genitally malformed heart, it is often necessary to ac-
count for the topology of each segment as well as the
connections (or alignment) among the segments.
(J Am Coli CardioI1987;10:1274-7)
ventricle and the morphologically left atrium connected to
the morphologically left ventricle, irrespective of the topo-
logic arrangement of the segments. Furthermore, this sys-
tem of analysis did not utilize the terms concordance and
discordance to describe hearts with other AV connections,
such as double-inlet ventricle or the absent connection var-
iant of AV valve atresia. Separate terms were used to de-
scribe these other connections.
At the time, there seemed to be no major conflict among
the systems because all cases examined with concordant or
discordant AV connections also had concordance or dis-
cordance, respectively, in the sense that these terms were
used by the Boston group (3,7). Then, at the First World
Congress of Paediatric Cardiology in 1980, Weinberg (8),
Van Praagh (9) and their colleagues described several cases
with disharmony between the topologic combinations and
the AV connections. One similar case had previously been
encountered in our collection (10). Since that time, we have
clinically studied an additional case that indicates the ne-
cessity of distinguishing AV connections and segmental to-
pology. We have also reexamined our previously published
case (10), and in this report describe it together with our
new case to emphasize this point.
Case Reports
Case 1. This premature infant (birth weight I. 9 kg)
presented with cyanosis and cardiac failure. The chest ra-
0735-1097/87/$3.50
lACC Vol. 10, No.6 ANDERSON ET AL. 1275
December 19X7: 1274-7 SEGMENTAL DISHARMONY

Tricuspid
valve Probe through inlet VSD

Figure 1. Case I. a, The morphologically right

ventricle has its apical trabecular part located to
the left of the ventricular mass. The tricuspid
valve connects the ventricle to the right-sided
right atrium ("crisscross heart"). The valve
straddles an inlet ventricular septal defect (VSD),
I and there is also an apical muscular defect. b,
I
The ventricle with probes through the two ven-
tricular septal defects. The probe shown in (a)
has been advanced through the tricuspid valve.
PT =' pulmonary trunk.

/
~
,Left-sided apical trabecular portion
a) b)

diograph showed a right-sided heart with its apex pointing Case 2. This infant presented at the age of 6 months
to the right in the setting of normally arranged abdominal with cyanosis. Examination revealed a soft, nonspecific
and thoracic organs (solitus). The electrocardiogram showed murmur and no signs of cardiac failure. The chest radiograph
right atrial hypertrophy with a P axis of + 90 and a QRS
axis of 0, The pattern of the precordial leads suggested
Figure 2. Case I. These illustrations show the left-sided left atrium
right ventricular hypertrophy, The infant's condition dete- connected to the morphologically left ventricle through a mitral
riorated and she died before further investigation could be valve. The ventricle is seen without (a) and with (b) the probes
performed, through the two ventricular septal defects (VSD) (see Fig. I). The
Autopsy demonstrated the presence of the usual atrial upper probe marks the straddling component of the tricuspid valve
arrangement (situs solitus) with the heart in the right side in the left ventricle.
of the chest and its apex pointing to the right. There was
left-sided juxtaposition of the atrial appendages, The venous
connections were normal, and a large defect was noted in Left atrium
the oval fossa, The right atrium connected to the morpho-
logically right ventricle through a superiorly located mor-
phologically tricuspid valve (Fig, I), The valve straddled
and overrode the inlet part of the ventricular septum. being
"
connected by 25% of its annular circumference to the mor-
phologically left ventricle, The dominant left ventricle was
also connected to the left atrium by an inferiorly located Mitral valve
mitral valve (Fig. 2). The inlet to the right ventricle was
anterior to that of the left ventricle, but the trabecular and Left ventricle
outlet portions of the morphologically right ventricle were a)
left-sided. The latter ventricle gave rise to both great arter-
ies, each arterial valve supported by a complete muscular
infundibulum. with the aorta being right-sided relative to
the pulmonary trunk (Fig. 3). In addition to the malalign-
ment inlet defect of the ventricular septum straddled by the
tricuspid valve, there was an apical muscular defect (Fig.
I and 2). Despite the concordant AV connection, the mor-

'/
phologically right ventricle was unequivocally of left-hand
topology (I loop). Figuratively speaking. only the palmar
surface of the observer's left hand could be placed on the -,
septal surface of the morphologically right ventricle with . through
Probe Probe through
the thumb in the inlet. the wrist in the apical component
and the tingers in the outlet portion (Fig. 3).
straddling tricuspid valve
h)
apical VSD
.
1276 ANDERSON ET AL. lACC Vol. 10, No.6
SEGMENTAL DISHARMONY December 1987:1274-7

Figure3. Case I. Themorphologically right

ventricle is left-sided and is connected to the
right atrium (concordant atrioventricular con-
nection). The ventricular mass is unequivo-
cally of left-hand topology. Pulm. = pul-
monary.

showed normal cardiac size, but a prominent right atrium
and mildly oligemic lungs. The arrangement of the thoracic
and abdominal organs was normal (solitus), as was the elec-
trocardiogram.
Investigations including cross-sectional echocardiog-
raphy and cardiac catheterization showed the atrial arrange-
ment to be normal (solitus). The right atrium was connected
to the left ventricle by way of a large AV valve (Fig. 4a).
The left atrium, entered through a small atrial septal defect
(Fig. 4b), communicated with the right ventricle by way of
a restrictive AV valve. A second communication existed
between the left atrium and left ventricle through a separate
stenotic orifice in the AV valve (straddling and overriding
tricuspid valve with the components committed separately
to the two ventricles). The right ventricle was mildly hy-
poplastic, but unequivocally had a right-hand pattern (Fig.
-lc), It was situated with its inlet component superiorly and
to the right. It was also positioned anterosuperiorly relative
to the dominant left ventricle (Fig. 5). Its outlet swung
leftward to give rise to a concordantly connected pulmonary
trunk in an anterosuperior and left-sided position (normal
relations). There was a completely muscular subpulmonary
infundibulum. The aorta was connected to the left ventricle.
The angiographic and echocardiographic sections sug-
gested the presence of right-sided juxtaposition of the atrial
appendages (Fig. 4b). Thus, despite the presence of a dis-
cordant AV connection with the usual atrial arrangement,
there was a right-hand ventricular topology.
Discussion
The terms "concordance," "discordance" and "con-
cordant AV connection." These cases along with those
described by Weinberg (8), Van Praagh (9) and their col-
leagues show that for a full description of congenitally mal-
formed hearts, it is necessary to account separately for seg-

Figure4. Case 2. Posteroanterior an-

giocardiogram. a, The catheter has been
advanced through the right atrioven-
tricular (AV) valve into the morpho-
logically (Morph.) leftventricle (LV),
which gives rise to the aorta. b, The
catheter (Cath.) has passed across the
atrial septum toopacify theleftatrium.
Note the juxtaposed left atrial appen-
dage. c, The left atrium is connected
to the morphologically right ventricle.
Despite the discordant A V connec-
tion, there is a right-hand (RH) pattern
of ventricular topology. The tricuspid
valve is straddling andoverriding, but
each of thecomponents is exclusively
connected to themorphologically right
and left ventricles, respectively. Pulm.
= pulmonary.
JACC Vol. io, No.6
December 1987:1274-7
Figure 5. Case 2. Lateral angiograms. These showthe concordant
ventriculoarterial connection. This along with the discordant atrio-
ventricular connection give the circulatory pattern of complete
transposition. Abbreviations as in Figure 4.
mental topology and intersegmental connections. When we,
as part of the European school, first used the terms "con-
cordance" and "discordance," we used them to account
for two of the six AV connections we had identified. In
contrast, Van Praagh et al. (1-3) used "concordance" and
"discordance" to describe only segmental harmony or dis-
harmony, irrespective of the precise fashion in which the
atrial chambers were connected to the ventricles. Thus, they
would describe a patient with a double-inlet ventricle as
having a concordant ventricular loop if the segmental com-
binations were of either {S,D-} or {I,L,-}. We had used
concordance exclusively to describe the setting of the mor-
phologically right atrium connected to the morphologically
right ventricle and the morphologically left atrium connected
to the morphologically left ventricle. We now circumvent
this potential problem area by describing specifically a con-
cordant or discordant connection.
We accept that there is a difference between, for ex-
ample, AV concordance as defined by Van Praagh et al.
(1-3) and a concordant AV connection. It costs nothing to
achieve the much needed specificity by using the precise
term "concordant AV connection." At the same time, many
others now use the term "concordance" to describe the
connection rather than segmental harmony and, presumably,
will continue to do so. In this respect, some might prefer
to promote a concept of "alignment concordance" as op-
posed to "situs concordance." We find this confusing. We
believe it better to describe the way in which the atrial
myocardium is connected to the ventricular myocardium as
the AV connection. This is then qualified as being concor-
dant, discordant, double-inlet and so on. Segmental topol-
ANDERSON ET AL. 1277
SEGMENTAl. DISHARMONY
ogy, when necessary, is described separately without resort
to the adjectives concordance and discordance. Nonetheless,
this approach is simply our preference. We fully recognize
that others choose to cope with the problem in a different
fashion (9).
Segmental topology versus AV connections. The con-
ceptual point of significance is that, almost always, hearts
with a concordant AV connection have the anticipated seg-
mental harmony as originally described by Van Praagh et
al. (1-3). The same is true for hearts with a discordant AV
connection. Very rarely, as exemplified by the two cases
described here and those reported by Weinberg, Van Praagh
and colleagues (8,9), the segmental combinations are not
as anticipated. When one specifies the segmental topology
and the connections among the segments separately, it is an
easy matter to account for these exceptional cases, irre-
spective of the terms used for description. Thus, in our first
case, it is easy to describe the usual atrial arrangement
existing with a concordant AV connection along with the
unexpected left-hand topologic arrangement of the ventric-
ular mass. In the second case, there is an unexpected right-
hand ventricular topology despite the usual atrial arrange-
ment and a discordant AV connection.
References
I. Van Praagh R, Ongley PA, Swan HJC. Anatomic types of single or
common ventricle in man: morphologic and geometric aspects of sixty
necropsied cases. Am J Cardiol 1964;13:367-86.
2. Van Praagh R, Van Praagh S, Vlad P, Keith 10. Anatomic types of
congenital dextrocardia: diagnostic and embryologic implications. Am
J Cardiol 1964;13:510-31.
3. Van Praagh R. The segmental approach to diagnosis in congenital
heart disease. In: Birth Defects: Original Article Series, Vol 8. Bal-
timore: Williams & Wilkins, 1972:4-23.
4. Kirklin JW, Pacifico AD, Bargeron LM. Soto B. Cardiac repair in
anatomically corrected malposition of the great arteries. Circulation
1973:48:153-9.
5. Shineboume EA, Macartney FJ, Anderson RH. Sequential chamber
localization: logical approach to diagnosis in congenital heart disease.
Br Heart J 1976;38:327-40.
6. Macartney FJ, Shineboume EA, Anderson RH. Connexions, relations,
discordance and distorsions. Br Heart J 1976;38:323-6.
7. Van Praagh R. Terminology of congenital heart disease: glossary and
commentary. Circulation 1977;56:139-43.
8. Weinberg PM, Van Praagh R, Wagner HR, Cuaso Cc. New form of
criss-cross atrioventricular relation: an expanded view of the meaning
of D- and Lloops (abstr). In: Abstract Book of World Congress of
Paediatric Cardiology. London, 1980:319.
9. Van Praagh R, David I, Gordon D, Wright GB and Van Praagh S.
Ventricular diagnosis and designation. In: Godman MJ, ed. Paediatric
Cardiology, Vol. 4. 1981;17:153-68.
10. Otero Coto E, Wilkinson JL. Dickinson DF, Rufilanchas 11, Marquez
J. Gross distortion of atrioventricular and vemriculo-arterial relations
associated with left juxtaposition of atrial appendages: bizarre form
of atrioventricular criss-cross. Br Heart J 1979;41 :486-92.