Pulmonary atresia/ventricular septal defect associated with facial port‐wine stain and retinal vascular abnormality: A new constellation or deletion in chromosome 22q11.2?

We have read with interest the report by Tan et al. [2003], on an individual with a congenital heart defect (pulmonary atresia with multiple MAPCA’s and VSD), associated with vascular anomalies, i.e., a facial port-wine stain and tortuous retinal vessels. The authors suggest that this may represent a novel entity. However, we believe that a deletion in chromosome 22q11.2 should first be excluded in this individual, before one can accept their suggestion. First, the type of heart defect is very suggestive of a 22q11.2 microdeletion. In individuals with pulmonary atresia, ventricular septal defect, and major aortopulmonary collateral arteries the incidence of a del22q11.2 is over 40% [Chessa et al., 1998; Hofbeck et al., 1998]. Second, tortuosity of the retinal vessels is a common finding in patients with a del22q11.2 [Mansour et al., 1987] (see Fig. 1). Tortuous retinal vessels is also a common finding in preterm children, in familial retinal arteriolar tortuosity, and has been described in many other syndromes [Hellstrom et al., 1997; Sutter and Helbig, 2003]. In the del22q11.2, it is present in both individuals with and without a heart malformation, and therefore this reflects a primary developmental defect of these vessels. There is evidence that neural crest cells contribute at least indirectly to the del22q11 phenotype, and for this reason, it is of interest that the periendothelial cells of the retinal vessels are also derived from the cranial neural crest cells [Etchevers et al., 2001]. Alternatively, a primary vascular malformation is suggested by the finding that these retinal vascular anomalies are also observed in a recently described animal model of the DiGeorge syndrome, i.e., mice lacking the VEGF-164 isoform [Stalmans et al., 2002, 2003]. Third, even though the facial features in the reported patient are difficult to appreciate from the publication, the long face with relatively small mouth and long nose are compatible with a del22q11.2. Port-wine stains are not a common feature in the del22q11.2. In our study group of 203 consecutively diagnosed individuals with a del22q11.2, not a single patient has a facial portwine stain, and this association has not been reported before to the best of our knowledge. However, given the frequent occurrence of a del22q11.2 (estimated at 1/5,000–1/4,000) one can expect that several patients have additional somatic manifestations purely by chance. When these features are prominent or dominate the clinical presentation, they can be misleading.