UC San Diego

Rationale. Very preterm (VPT) birth, birth before 33 weeks gestational age, puts children at increased risk for a variety of neurodevelopmental, cognitive, and behavioral difficulties as they grow. Severe complications of VPT birth (e.g., Grade 3-4 intraventricular hemorrhage, cystic periventricular leukomalacia, bronchopulmonary dysplasia) are associated with more serious functional deficits and often prompt extensive follow-up. However, milder brain injury may result in subtle deficits that may only become apparent when children begin to struggle in elementary school, particularly with regard to neuromotor functioning, mathematics, and related skills. A better understanding of the neuropsychological and neuroanatomical variations related to VPT birth as these children are entering kindergarten is therefore needed to develop screening procedures and guide appropriate interventions in the future.

Design. 60 children born VPT without severe birth-related complications and 40 children born full term (FT) were enrolled in this study within six months of beginning kindergarten. Each child underwent structural magnetic resonance imaging (MRI), diffusion-tensor imaging (DTI), and a battery of neuropsychological measures. The aims of this study were to (1) evaluate mathematics abilities and determine the set of skills across a set of a priori measures from related cognitive domains that mediate the group differences between performance on a mathematics measure using a multiple mediation model, (2) utilize structural MRI to conduct an exploratory analysis to compare cortical morphometry measures of gray matter volume, gray matter thickness, surface area, and sulcal depth between groups using a voxel-wise, surface-based method, and (3) examine overall group differences in motor task performance, and compare an a priori set of regional white matter integrity measures and brain structural volumes in VPT children with and without motor deficits.

Results. Groups did not differ significantly on age, sex distribution, or socioeconomic status (a combination of maternal education and household income). Study 1 found that children born VPT scored significantly lower on the mathematics measure. The difference between VPT and FT groups was mediated by scores on tests of visual-motor integration, verbal comprehension, and phonological awareness but not but by motor skills, parent-reported executive functioning, spatial working memory, or phonological working memory. In Study 2, VPT-born children showed a widespread pattern of reduced cortical thickness in the temporal and lateral parietal lobes and increased cortical thickness in the medial occipital lobe. Cortical surface area was larger in the medial cingulate in the VPT-born group but smaller in the fusiform area. Cortical volume was larger in the medial occipital lobe for the VPT-born group and smaller in the fusiform area. In Study 3, the VPT-born children performed significantly worse on the motor measure, with 50% demonstrating clinically significant deficits. The VPT-born children with motor deficits demonstrated lower fractional anisotropy (FA) and higher mean diffusivity (MD) in the forceps major compared to the VPT group without motor deficits. The VPT group with motor deficits also showed higher MD in the corpus callosum and lower FA compared to the FT group in the forceps major, anterior thalamic radiations (ATR), and cingulum. Both VPT groups had significantly reduced volumes of the thalamus, brainstem, and cerebellar white matter compared to the FT group. Additional regions that were initially hypothesized to be related but were not significant between-group predictors were volume of the corpus callosum, forceps minor, ATR, and cortical-spinal tract (CST), as well as volumes of the ventral diencephalon and basal ganglia. White matter diffusion measures (FA and MD) of the inferior frontal-occipital fasciculus, thalamus, basal ganglia, superior cortical striatal fibers, and CST were not significant predictors of group.

Conclusion. Advances in medical care have resulted in decreased mortality and reduced rates of more severe neurological complications in children born preterm. However as shown here, even children born very preterm with a relatively benign early health history often demonstrate difficulties in mathematics and motor skills. These difficulties are related to subtle differences in neuroanatomical areas often affected by very preterm birth. Pre-kindergarten screening of motor functioning, mathematics, and related cognitive skills may assist in early detection of difficulties, leading to increased opportunities for appropriate support and interventions, and potentially improvement of long-term outcomes.