Although observational findings linking breast milk to higher scores on cognitive tests may be confounded by factors associated with mothers' choice to breastfeed, it has been suggested that one or more constituents of breast milk facilitate cognitive development, particularly in preterms. Because cognitive scores are related to head size, we hypothesized that breast milk mediates cognitive effects by affecting brain growth. We used detailed data from a randomized feeding trial to calculate percentage of expressed maternal breast milk (%EBM) in the infant diet of 50 adolescents. MRI scans were obtained (mean age ϭ 15 y 9 mo), allowing volumes of total brain (TBV) and white and gray matter (WMV, GMV) to be calculated. In the total group, %EBM correlated significantly with verbal intelligence quotient (VIQ); in boys, with all IQ scores, TBV and WMV. VIQ was, in turn, correlated with WMV and, in boys only, additionally with TBV. No significant relationships were seen in girls or with gray matter. These data support the hypothesis that breast milk promotes brain development, particularly white matter growth. The selective effect in males accords with animal and human evidence regarding gender effects of early diet. Our data have important neurobiological and public health implications and identify areas for future mechanistic study. (Pediatr Res 67: 357-362, 2010) N umerous studies report that breastfeeding is associated with higher scores on tests of neurodevelopment and cognition in later life (1,2), suggesting that breast milk may impact early brain development, with potentially important biologic, medical, and social implications. These findings are often questioned because other factors associated with mothers' choice to breastfeed (e.g. higher socioeconomic status and level of education, different child-rearing attitudes) might promote cognitive development. In a study using data from a national database, adjustment for maternal intelligence quotient (IQ) eliminated the breastfeeding effect (3). All these studies, however, are potentially flawed by lack of experimental design.More recently, a large cluster randomized trial of breastfeeding promotion using an experimental design demonstrated a large effect of breastfeeding on cognition, adding credence to the evidence for beneficial effects seen in past observational studies (4). Anderson et al. (5) showed in a meta-analysis that, after appropriate adjustments, breastfeeding was associated with an advantage of around three points on tests of cognition in children born at term and around five points in those born preterm, both large effects in population terms. The implication is that, over and above social factors, one or more constituents of breast milk benefit neurodevelopment, particularly so in those born preterm, at a more sensitive stage of brain development.Our current study was prompted by two factors, the now likely causal relationship between breastfeeding and cognition and the previous observation that cognitive scores in preterms are rela...

Children born preterm and of very low birth weight have an increased incidence of learning difficulties, but little is known about the specific nature of their cognitive deficits and the underlying neuropathology. We hypothesized that their vulnerability to hypoxic, metabolic, and nutritional insults would lead to reduced hippocampal volumes and to deficits in memory because of the role of the hippocampus in this domain of cognition. Neuropsychological and magnetic resonance imaging methods were used to investigate this hypothesis in adolescents born preterm (< or = 30 wk gestation, n = 11) or full-term (n = 8). The preterm group had significantly smaller hippocampal volumes bilaterally, despite equivalent head size, and showed specific deficits in certain aspects of everyday memory, both on objective testing and as indicated by parental questionnaires. The preterm group also had a specific deficit in numeracy. The reduced hippocampal volumes and deficits in everyday memory have previously been unrecognized, but their prevalence in a group of neurologically normal children is striking.

Learning difficulties, including problems with numeracy, are common in Western populations. Many children with learning difficulty are survivors of preterm birth. Although some of these children have neurological disabilities, many are neurologically normal, and the latter group provides us with an important opportunity to investigate the neural bases of learning problems. We have conducted a neuroimaging study of adolescent children who had been born preterm at 30 weeks gestation or less, to investigate the relationship between brain structure and a specific difficulty in arithmetic calculation. Using voxel-based morphometry, we have been able to demonstrate that there is an area in the left parietal lobe where children without a deficit in calculation ability have more grey matter than those who do have this deficit. To our knowledge, this is the first report establishing a structural neural correlate of calculation ability in a group of neurologically normal individuals.

Early nutrition in animals affects both behavior and brain structure. In humans, randomized trials show that early nutrition affects later cognition, notably in males. We hypothesized that early nutrition also influences brain structure, measurable using magnetic resonance imaging. Prior research suggested that the caudate nucleus may be especially vulnerable to early environment and that its size relates to IQ. To test the hypothesis that the caudate nucleus could be a neural substrate for cognitive effects of early nutrition, we compared two groups of adolescents, assigned a Standard-or High-nutrient diet in the postnatal weeks after preterm birth. Groups had similar birth status and neonatal course. Scans and IQ data were obtained from 76 adolescents and volumes of several subcortical structures were calculated. The High-nutrient group had significantly larger caudate volumes and higher Verbal IQ (VIQ). Caudate volumes correlated significantly with VIQ in the Standardnutrient group only. Caudate volume was influenced by early nutrition and related selectively to VIQ in males, but not in females. Our findings may partly explain the effects of early diet on cognition and the predominant effects in males. They are among the first to show that human brain structure can be influenced by early nutrition.

Rolandic epilepsy is regarded as the classic example of benign focal epilepsy. However, neuropsychological deficits have been noted in affected children. As Rolandic discharges are mainly distributed over the centrotemporal region, specific interference with language function might be suspected. Therefore, we conducted a standardized neuropsychological assessment in children with Rolandic epilepsy which covered all important aspects of language processing. We measured intelligence Quotient, verbal memory, auditory discrimination, vocabulary, grammar and literacy in 20 children with an active Rolandic focus. Information about performance at school was obtained from teachers by means of a questionnaire. Patients with Rolandic epilepsy failed five of the twelve standardized language tests significantly more often than the normative population and consequently showed impairment of the following functions: reading, spelling, auditory verbal learning, auditory discrimination with background noise and expressive grammar. Thirteen of the 20 children showed language dysfunction with difficulties in two or more of the twelve standardized language tests. In eight of these 13 children the Full Scale Intelligence Quotient was within average range, indicating a specific language deficit. Language dysfunction was closely associated with learning difficulties at school. This study documents a consistent pattern of language dysfunction in children with Rolandic epilepsy which suggests interictal dysfunction of perisylvian language areas.