Cognitive function in preterm infants: no simple answers.

Abstract

Listen

Translate article

IMPROVED SURVIVAL RATES HAVE HEIGHTENED EMPHASIS ON the long-term outcome of infants born at very low-birthweight (VLBW) and extremely low-birth-weight (ELBW). Medical or physical, neurological, cognitive, academic, neuropsychological, motor, social, emotional or behavioral, functional, and health-related quality of life are all outcomes of interest. Previously, the incidence of major disabilities, such as moderate or severe mental retardation, cerebral palsy, epilepsy, blindness, and sensorineural hearing loss, was the greatest concern because of the severity of developmental morbidity. Over the last decade, the incidence of these disabilities has remained constant, ranging from 15% to 20%, with higher incidence corresponding to lower birth weight and gestational age. In addition, more subtle cognitive impairments have been detected in survivors without major disabilities. These high-prevalence/ low-severity dysfunctions include learning disabilities, borderline intellectual functioning, attention-deficit/ hyperactivity disorders, specific neuropsychological deficits, and behavioral problems, and reportedly occur in 50% to 70% of VLBW infants. In contrast with major disabilities, cognitive dysfunctions are moderated by environmental factors, such as the social, ethnic, and educational background of caretakers, neighborhoods, schooling, and socioeconomic status. Even though major disabilities are often identified during infancy, high-prevalence/low-severity dysfunctions become more obvious at school entry and later. Unfortunately, studies of children who were born with VLBW contain methodological problems and conflicting data, which preclude drawing conclusions with respect to these outcomes. Whether early cognitive deficits worsen, are stable, or improve over time is an unanswered question. The article by Ment and colleagues in this issue of THE JOURNAL attempts to address this issue. Serial assessments of cognitive function (the Peabody Picture Vocabulary Test– Revised [PPVT-R] and age-appropriate IQ tests) were performed on 296 VLBW infants at 36, 54, 72, and 96 months of corrected age. The median PPVT-R score increased from 88 to 99, the full-scale IQ from 90 to 95, verbal IQ from 91 to 97.5, and the performance IQ improved from 89 to 92. These findings suggest some reasons for optimism regarding the cognitive outcome in VLBW children. However, the article by Ment et al also highlights several potentially problematic issues found in follow-up studies. First, Ment et al relied on controls used for the standardization of the cognitive tests rather than using a concurrent control group of full-term infants. A concurrent comparison group would clarify whether the reported changes are specific to VLBW infants as a whole or this specific follow-up population. Second, the test instruments used and the area of function assessed can affect outcome interpretation. The authors’ primary outcome was receptive verbal abilities, measured with the PPVT-R. They chose this measure because it could be used at all test ages and does not require verbal responses, allowing children with motor deficits to participate. However, receptive vocabulary represents one facet of cognitive function, and one area of function that may not be as strongly influenced by biological risks such as prematurity. Some language functions are relatively intact in VLBW children, particularly recognition vocabulary and receptive language. However, other more complex verbal processes, such as understanding of syntax, abstract verbal skills, verb production, and word fluency, have been found to be deficient. Moreover, language is highly influenced by environmental factors such as socioeconomic status and maternal education. This was demonstrated by Ment et al, where improvement in scores over time was related to positive environmental influences such as residence in a 2-parent household and higher levels of maternal education. Intelligence quotients are also addressed in the article by Ment et al. The IQ scores improved by approximately 5 points from 36 to 96 months but still were in the average to low-average range. Earlier meta-analyses of cognitive deficits among those born with low birth-weight revealed a 5to 7-point deficit in comparison with those born fullterm. More recent comparisons indicate a 0.3 to 0.6 SD decrease in IQ, the reported decrement ranging from 3.8 to 10.9 points. In those born with VLBW who are free of major disabilities, mean group IQs generally fall in the bor-