A review of: Stewart AL, Rifkin L, Amess PN, et al. 1999 Brain structure and neurocognitive and behavioural function in adolescents who were born very premature. Lancet 353:1653–1657.

STEWART ET AL. recently publ ished an important study relating brain structure using magnetic resource imaging (MRI) with neurocognitive and behavioral function in adolescents who were born very preterm. Moreover, the report compares ultrasound findings in the neonatal period to neurodevelopmental outcome at 8 years and brain images, neurocognitive, and behavioral function in adolescence.

Since the introduction of neonatal intensive care, the survival of preterm infants has increased significantly. However, the prevalence of major disabilities found in these toddlers and pre-school children is around 10 percent. This percentage has been stable over the years and differences in prevalence seem to be largely due to variable definitions of disabilities. There is, however, a growing body of evidence that adolescents who were born very preterm are at risk for neurodevelopmental problems including mild cognitive impairments and neurological abnormalities, as well as learning and behavioral problems that will impair their potential to develop into independent adults. It is, however, difficult to establish the prevalence of these problems, because of the lack of standardized outcome measures.

Stewart and her group, in a study of 72 adolescents born before 33 weeks of gestation, showed that 76% had abnormal or equivocal MRI brain scans at the age of 14–15 years. Abnormal MRI scans in the subjects were 10 times as frequent as in controls. These lesions were related to the more severe neurological impairments, epilepsy, and behavioral disturbances found at the same age.

Surprisingly, the abnormal MRI lesions were not related to the milder neurological impairments and learning problems that were frequently found in premature infants with normal or abnormal brain scans. Of interest, there was a high correlation between behavioral problems and abnormal MRI scans. It would have been interesting to know whether combinations of impairments were related to the MRI scans, because it is the combination of several mild impairments that have a large impact on every day life.

The authors also showed that ultrasonography in the neonatal period, even when early linear array with 5 MHz or 7 MHz probes was only available, could predict MRI lesions when there was evidence of hypoxic-ischemic injury.

Neurodevelopmental problems are probably very frequent in adolescents and have a large impact on the lives of these children and their families. The finding of a high prevalence of structural brain lesions in adolescents who were born prematurely, stresses the need for standardized outcome measures in order to evaluate perinatal care. Moreover, now that advanced methods of brain imaging in the neonatal period are available and hypoxic-ischemic injuries can readily be recognized, research can be focused on the prevention of these lesions.

Therefore, Stewart and her colleagues conclude that “If the prognosis of the very preterm infant is to be improved, structural brain abnormalities must be prevented or, at least, the prevalence reduced, in order to enhance the neurodevelopmental outcome of the very premature infant.” This challenge should encourage future research in the next millennium, to improve the outcome of these vulnerable infants.