Abstract

All 155 surviving children from a cohort of 200 very low birthweight infants originally studied in 1984-5 were traced. These infants had careful sequential ultrasound examinations in the neonatal period. The children were examined again at entry into school at 5 years of age. The test of motor impairment (TOMI) and the vocabulary subscale of the Wechsler preschool and primary scale of intelligence (WPPSI) were administered to 152 of the index cohort and 144 control children of the same age in the same class at school. Twelve of the cohort had cerebral palsy, but eight of these were in mainstream schools. The index group scored significantly higher on both the TOMI and the WPPSI subscale compared with the controls. The index cases were subdivided on the basis of their neonatal ultrasound scans into four groups: group 1, consistently normal; group 2, 'prolonged flare'; group 3, germinal matrix haemorrhage-intraventricular haemorrhage (GMH-IVH), without parenchymal haemorrhage, but no evidence of prolonged flare; and group 4, both GMH-IVH and prolonged flare. The group of index children with consistently normal ultrasound scans had a higher TOMI and lower WPPSI compared with their controls. There was a statistically significant increase in the TOMI subscore 1 (manual dexterity) in group 4 infants compared with group 1, but not differences between the other groups. Regression analysis suggests that neither prolonged flare nor GMH-IVH has an important individual contribution to the variation, but the low birth weight does have a significant relationship with motor impairment. It appears that relatively minor ultrasound appearances such as prolonged flare and GMH-IVH are associated with motor impairment (clumsiness) at 5 years, but this has a small effect compared with low birth weight.