Editor—Tricuspid atresia, an inflow anomaly, does not classically belong to the embryological group of conotruncal malformations involving the outflow tract. We report here five families with tricuspid atresia in one member and conotruncal malformation in a relative. We believe that this apparently non-concordant recurrence is not fortuitous according to the mechanistic classification of congenital heart malformations.1

The family pedigrees are shown in fig 1. All cases of tricuspid atresia were of the same subtype with muscular ventricular septal defect, subpulmonary stenosis, and ventriculoarterial concordance (type Ib). All patients had a normal karyotype. No chromosome 22q11 microdeletion could be identified by fluorescence hybridisation with the Sc11.1 probe in any of the patients.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Pedigrees of families. Subjects shaded in black had tricuspid atresia. Subjects shaded in grey had a conotruncal malformation, either tetralogy of Fallot (ToF) or a truncus arteriosus communis (TAC).

One of the major group of congenital cardiac malformations defined in Clark’s classification is the conotruncal cardiac malformation group.1 Cells derived from occipital neural crest contribute to the formation of the conotruncal region and participate in the process of aortopulmonary septation.2Indeed, depleting the heart of cranial neural crest cells result in outflow tract malformations.2 While Clark’s classification sheds light on the relationship between anatomically different cardiac defects, such as truncus arteriosus, tetralogy of Fallot, and interrupted aortic arch, a number of cardiac malformations cannot be classified into any group. Among these, tricuspid atresia can be considered either as a malformation of the inflow tract or as an orphan disease. We believe, however, that tricuspid atresia may belong to the group of conotruncal malformations. The arguments for this hypothesis are the following. First, neural crest ablation may result in inflow anomalies, including tricuspid atresia, tricuspid stenosis, and double inlet left ventricle.3 These defects usually occur with either truncus arteriosus or dextroposed aorta and are much less predictable than the outflow tract anomalies with regard to the site and length of the neural crest lesion. Secondly, the association of tricuspid atresia with persistent truncus arteriosus has been described in humans, as well as malformation of the tricuspid valve associated with tetralogy of Fallot in which the aorta is dextroposed.4 5 Thirdly, 22q11 deletion has been recently reported in two cases of tricuspid atresia.6 Children with a 22q11 deletion exhibit a wide spectrum of conotruncal malformations and it remains unknown why this haploinsufficiency shows such a wide range of penetrance and expressivity. The prevalence of the various conotruncal anomalies in a population of patients with 22q11 deletion is apparently randomly distributed. This might be explained by the “stochastic single gene model”.7 In this model, assuming that conotruncal malformations are single gene defects inherited as a mendelian trait or that a single gene is responsible for the constellation of cardiac features of 22q11 deletion, the amplification of a small disturbance in the migration of neural crest cells will result in the variability which different conotruncal malformations show in different affected subjects, even in the same kindred in which the genetic abnormality at the affected locus is identical.8

We conclude that the apparently discordant recurrence of cardiac defects in our families could be regarded as concordant from an embryological and genetic viewpoint.