Research on behavioral phenotypes: velocardiofacial syndrome (deletion 22q11.2) Paul P Wang* MD; Michael F Woodin PhD; Rachel Kreps-Falk BA; Edward M Moss PhD, Children’s Seashore House, Philadelphia, PA, USA. *Correspondence to first author at Children’s Seashore House, 3405 Civic Center Boulevard, Philadelphia, PA 19104, USA. E-mail: wangp@mail.med.upenn.edu Recent molecular genetics research has established that most cases of velocardiofacial syndrome (VCF) and of DiGeorge syndrome result from a submicroscopic deletion at chromosome 22q11.2 1–3 . The medical features of this dis- order include hypocalcemia, immunodeficiency, cleft palate, subtle facial dysmorphism, ‘conotruncal’ cardiac malforma- tions (e.g. truncus arteriosus, tetralogy of Fallot, interrupted aortic arch, and certain types of ventricular septal defects), and other congenital malformations 4 . While none of these complications is found in all patients with the 22q11.2 microdeletion, each is sufficiently common to be considered part of the disorder’s medical phenotype. Recent research also has established the existence of a behavioral phenotype for VCF. Investigators from the fields of Developmental-Behavioral Pediatrics, Child Neuropsychology, Developmental Psychology, Child Psychiatry, and other fields all contributed to the description of this phenotype. Investigators employed multiple approaches for examining VCF, including psychometric and psychiatric diagnostic methods, as anticipated by Dykens 5 . The resultant phenotype spans multiple domains of behavior. In this paper, we review the behavioral phenotype research on VCF, and identify cur- rent and future directions for this work. This research has implications not only for the clinical care of patients with VCF, but also for broader issues related to developmental cogni- tive psychology and to the pathogenesis of psychiatric dis- ease. The research on VCF also illustrates several important methodological considerations for behavioral phenotype research that others have noted 5–7 . Psychoeducational profile of VCF The earliest behavioral descriptions of VCF were based on patients with clinically diagnosed cases of the syndrome, but without molecular confirmation of the diagnosis. As a conse- quence, these reports cannot be considered definitive, though they certainly offered valuable guidance for later investigations 8 . After the advent of widespread molecular diagnosis by fluorescent in-situ hybridization testing, a char- acteristic pattern of developmental delay and learning differ- ences became apparent. In infancy and early childhood, children with VCF typically show delays that span the motor, cognitive, and linguistic domains. First words generally do not emerge until age 2 years or later, hypotonia is present in over half of all patients, and most do not walk until after 15 months of age. Over 75% of all children tested on the Bayley Scales of Infant Development 9 were in the delayed range, with Motor Development Indices often lower than Psychomotor Development Indices 10,11 . Most also have speech abnormalities, including wet hoarseness, high pitch, strained-strangled voice, dysarthria, and compensatory artic- ulation errors (secondary to the alteration of air flow due to palatal anomalies). Feeding disorders also are present in most young patients, even in the absence of palatal defects. Despite the conspicuous delays and abnormalities in early speech-language development, school-aged children with VCF typically show higher Verbal IQ than Performance IQ scores 11,12 . Reading and Spelling 13 achievement scores are higher than Mathematics 13 scores in almost 90% of patients, with a mean difference of almost 10 points 12 . As a corollary, the dichotomy between Verbal Comprehension and Perceptual Organization factor scores from the Wechsler Intelligence Scales for Children 13 is more dramatic and more consistent than the split between Verbal IQ and Performance IQ. (This is because poor Arithmetic subtest scores tend to lower Verbal IQ scores, but do not contribute to the calculation of Verbal Comprehension factor scores.) Full-Scale IQ is typically in the low-normal to borderline range among school-aged children with the disorder, with Verbal IQ averaging close to 80, 422 Developmental Medicine & Child Neurology 2000, 42: 422–427 Annotation