Global and regional differences in brain anatomy of young children born small for gestational age

Abstract

In children who are born small for gestational age (SGA), an adverse intrauterine environment has led to underdevelopment of both the body and the brain. The delay in body growth is (partially) restored during the first two years in a majority of these children. In addition to a negative influence on these physical parameters, decreased levels of intelligence and cognitive impairments have been described in children born SGA. In this study, we used magnetic resonance imaging to examine brain anatomy in 4- to 7-year-old SGA children with and without complete bodily catch-up growth and compared them to healthy children born appropriate for gestational age. Our findings demonstrate that these children strongly differ on brain organisation when compared with healthy controls relating to both global and regional anatomical differences. Children born SGA displayed reduced cerebral and cerebellar grey and white matter volumes, smaller volumes of subcortical structures and reduced cortical surface area. Regional differences in prefrontal cortical thickness suggest a different development of the cerebral cortex. SGA children with bodily catch-up growth constitute an intermediate between those children without catch-up growth and healthy controls. Therefore, bodily catch-up growth in children born SGA does not implicate full catch-up growth of the brain.

Figure 1. Global brain parameters in AGA, SGA+ and SGA− children.

Only data from the right hemisphere are shown. Bars depict the mean within each group. P-values for trend derived from polynomial contrast analyses are shown (see also Table 2). P-values of the left hemisphere are mentioned only when statistically different compared to p-values of the right hemisphere. AGA: appropriate for gestational age; SGA+: small for gestational age with catch-up growth; SGA−: small for gestational age without catch-up growth; n.s.: not signifcant. *: P-value of left hemisphere not significant.

Figure 2. Cortical thickness of the cerebral cortical mantle in AGA, SGA+ and SGA− children.

Figure shows significant cortical thickening in children born SGA in comparison to normal AGA children. Specifically, most pronounced thickening is found in frontal brain regions, overlapping medial frontal and superior frontal cortices. Figure 2A shows the cortical thickness of the cerebral cortical mantle in the AGA, SGA+ and SGA− group, respectively. Figure 2B shows the effect-size difference maps between the AGA and SGA+ and AGA and SGA− group, showing strong thickening of the medial frontal and superior frontal regions in both SGA children. Figure 2C shows the statisical difference maps between AGA vs SGA+ and AGA vs SGA− , thresholded at p<0.001. Both SGA+ and SGA− children showed wide-spread signifiant higher thickness of the cortical mantle, most pronounced in frontal (as marked as the frontal cluster) and parietal regions, surviving cluster-wise correction for multiple testing (see materials and methods). Regions a–g refer to regions in Table 3 (AGA vs SGA+: a = Superior frontal, b = Lateral orbitofrontal; AGA vs SGA−: c = Superior frontal, d = Pericalcarine, e = Superior frontal, f = Posterior cingulated, g = Superior parietal). AGA: appropriate for gestational age; SGA+: small for gestational age with catch-up growth; SGA−: small for gestational age without catch-up growth; c–p: p-value after cluster-wise correction for multiple comparisons.

Figure 3. Cortical thickness of the medial prefrontal cortex of both hemispheres in AGA, SGA+ and SGA− children.

P-values for trend derived from polynomial contrast analyses are shown. Bars depict the mean within each group. AGA: appropriate for gestational age; SGA+: small for gestational age with catch-up growth; SGA−: small for gestational age without catch-up growth.