Abstract 1470 Poster Session III, Monday, 5/3 (poster 5)

OBJECTIVES: To assess the intelligence quotient (IQ) and academic achievement of a cohort of very-low-birth-weight (VLBW) subjects born at <1000 g now in their late teens or early adult years. POPULATION: In 1976-81 the neonatal survival rate of VLBW infants born in or referred to this hospital was 82/351 (23%). Three handicapped children died in childhood, 9 VLBW subjects were lost, 16 refused to participate and 54 survivors (68%) were examined. Controls were 42 term normal birthweight (NBW) subjects born in our hospital matched for age, birth order, race and sex. Subjects were traced with the help of government assistance. METHODS: Neonatal data had been collected prospectively. VLBW children had an IQ assessment at 5-9 years with a Weschler Preschool and Primary Scale of Intelligence or a Weschler Intelligence Scale for Children Revised (Dev Med Child Neurol 1988;30:162-9). At 16-21 years of age subjects had the Weschler Adult Intelligence Scale assessment and an interview about school achievement: need for special schooling, grades repeated, last grade succeeded, obtainment of secondary school (11th grade) diploma. Neurosensory disabilities consisted of cerebral palsy, blindness or bilateral hearing loss. Socio-economic status (SES) was measured by the mother's Hollingshead index. RESULTS: Neonatal data for the 54 VLBW and 42 NBW infants were: mean (±SD) birthweight 908±82 vs 3419±440 g, gestational age 28.3±2.3 (range 25-34) vs 39.9±0.8 (range 38-42) weeks, intrauterine growth retardation (<2SD) 15 (28%) vs 0, girls 33 (61%) vs 29 (69%). VLBW and NBW subjects were seen at a mean age of 18.2±1.7 vs 18.2±1.8 years. SES of VLBW subjects was 49±16 vs 37±16 for NBW (p<0.001). There were significant differences when comparing VLBW to NBW subjects in mean adult global IQ (95±11 vs 107±14, p<0.001), verbal IQ (94±11 vs 107±15, p<0.001) and performance IQ (98±13 vs 108±16, p<0.001), in incidence of IQ <85 (9=17% vs 1=2%, p<0.05), of neurosensory disabilities (6=11% vs 0, p<0.05), of grade repetition in regular classes during elementary school, ie kindergarden and the 6 elementary years (28/51=55% vs 7/42=17%, p<0.001) or during the entire schooling (33/51=65% vs 14/42=33%, p<0.005), of requirement for special classes (9=17% vs 1=2%, p<0.05) and of obtainment of secondary school diploma, ie after 5 years of secondary school, for those aged ≥18 years (20/33=61% vs 20/23=87%, p<0.05). Multivariate analysis revealed that 33% of the IQ variance could be explained by gestational age and SES. Covariance analysis showed that mean IQs remained significantly different between VLBW and NBW subjects when adjusted for SES. Linear regression showed that 50% of the variance in adult IQ in VLBW subjects was explained by IQ at 5-9 years. CONCLUSION: VLBW subjects had a mean adult IQ in the normal range but lower than NBW subjects and had significantly more school failures but despite these difficulties the majority of VLBW subjects obtained their secondary school diploma.