Developmental cortical thinning in fetal alcohol spectrum disorders

Abstract

Regional cortical thickness was evaluated using CIVET processing of 3D T1-weighted images (i) to compare the variation in cortical thickness between 33 participants with fetal alcohol spectrum disorders (FASD) aged 6–30 years (mean age 12.3 years) versus 33 age/sex/hand-matched controls, and (ii) to examine developmental changes in cortical thickness with age from children to young adults in both groups. Significant cortical thinning was found in the participants with FASD in large areas of the bilateral middle frontal lobe, pre- and post- central areas, lateral and inferior temporal and occipital lobes compared to controls. No significant cortical thickness increases were observed for the FASD group. Cortical thinning with age in a linear model was observed in both groups, but the locations were different for each group. FASD participants showed thinning with age in the left middle frontal, bilateral precentral, bilateral precuneus and paracingulate, left inferior occipital and bilateral fusiform gyri; while controls showed decreases with age in the bilateral middle frontal gyrus, right inferior frontal gyrus, bilateral precuneus gyrus, and bilateral occipital gyrus. A battery of cognitive assessments of memory, attention, motor, and verbal abilities was conducted with many of the FASD participants, but no significant correlations were found between these cognitive scores and regional cortical thickness. Non-invasive measurements of cortical thickness in children to young adults with FASD have identified both key regions of cortex that may be more deleteriously affected by prenatal alcohol exposure as well as cortical changes with age that differ from normal developmental thinning.

Introduction

Maternal alcohol consumption during pregnancy can lead to a variety of developmental disorders (Jones and Smith, 1973, Lemoine et al., 1968). Fetal alcohol syndrome (FAS) refers to a specific set of abnormalities associated with prenatal alcohol exposure (PAE), including facial dysmorphology, growth deficiency and cognitive disabilities (Bertrand et al., 2005), but frequently, children with prenatal exposure to alcohol who do not have the facial characteristics of FAS also suffer severe cognitive and neurological deficits. These children may be diagnosed with a variety of disorders falling under the umbrella term fetal alcohol spectrum disorders (FASD) (Astley, 2004, Chudley et al., 2005). Individuals with FASD may experience cognitive difficulties, which include lower overall intelligence, hyperactivity, behavioral and adaptive difficulties, and deficits in motor function, attention and executive function, learning and memory, expressive and receptive language, executive function and visuospatial skills (for a review, see Kodituwakku, 2007, Mattson et al., 1998, Niccols, 2007).

Brain morphological abnormalities in individuals with FASD have been observed in autopsy (Clarren et al., 1978, Jones and Smith, 1975), but also have been identified by a variety of in vivo quantitative neuroimaging techniques (for a review, see Norman et al., 2009). For example, overall brain volume is reduced in FASD, as are gray and white matter volumes (Lebel et al., 2008a). Whether the volume changes are more severe in certain cortical regions than others has been investigated by voxel based morphometry (VBM) of gray matter density and surface based shape analysis (Sowell et al., 2001a, Sowell et al., 2001b, Sowell et al., 2002). In 21 participants with FASD, bilateral inferior parietal/perisylvian regions (more prominent in the left) and left ventral frontal cortex were shown to have increased gray matter density and smaller brain growth (as measured by radial distance from the center of the brain) (Sowell et al., 2001a, Sowell et al., 2001b, Sowell et al., 2002).

Algorithms have been developed in order to measure regional cortical thickness as the distance between two surfaces (pial–cortical surface and gray–white interface), usually on 3D, isotropic T1-weighted images (Fischl and Dale, 2000, MacDonald et al., 2000, Sowell et al., 2004). From typically developing children to young adults, the cortex generally has been shown to thin across development (Shaw et al., 2008, Tamnes et al., 2010).

There has only been one study reporting cortical thickness measurements in individuals with prenatal exposure to alcohol (Sowell et al., 2008) where thicker cortices in the bilateral temporal, bilateral inferior parietal, and right frontal regions were observed in 21 FASD participants (14 of these with FAS) aged 8–22 years relative to healthy controls. It was proposed that the thicker cortices may be explained by an absence of the cortical thinning that usually occurs in healthy development. Notably, cortical thickness in the right dorsal frontal and left occipital regions was related to verbal recall and visuospatial measures, respectively; however, these were positive correlations where thinner cortex was associated with worse performances in the FASD group. This latter observation is more in line with animal models of PAE showing thinner cortices in rat (Aronne et al., 2008) and fewer neurons in the somatosensory cortex of the macaque brain (Miller, 2007). There is a need to further study cortical thickness in human FASD participants to evaluate the consistency of findings, and to explore their relationship with cognitive abilities. Furthermore, age-related changes of cortical thickness have not previously been evaluated in participants with FASD.

Therefore, the purpose of the current research was to assess 1) the regional variation in cortical thickness in a cohort of 33 FASD participants versus age/sex/handedness matched healthy controls, 2) cortical thickness differences with age from childhood to young adulthood (6 to 30 years) in both FASD and controls, and 3) correlations between cortical thickness and cognitive functioning in FASD.