Dissociations in Cortical Morphometry in Youth with Down Syndrome: Evidence for Reduced Surface Area but Increased Thickness

Abstract

Detailed descriptions of cortical anatomy in youth with Down syndrome (DS), the most common genetic cause of intellectual disability (ID), are scant. Thus, the current study examined deviations in cortical thickness (CT) and surface area (SA), at high spatial resolution, in youth with DS, to identify focal differences relative to typically developing (TD) youth. Participants included 31 youth with DS and 45 age- and sex-matched TD controls (mean age ∼16 years; range = 5-24 years). All participants completed T1-weighted ASSET-calibrated magnetization prepared rapid gradient echo scans on a 3-T magnetic resonance imaging scanner. Replicating prior investigations, cortical volume was reduced in DS compared with controls. However, a novel dissociation for SA and CT was found-namely, SA was reduced (predominantly in frontal and temporal regions) while CT was increased (notably in several regions thought to belong to the default mode network; DMN). These findings suggest that reductions in SA rather than CT are driving the cortical volume reductions reported in prior investigations of DS. Moreover, given the link between DMN functionality and Alzheimer's symptomatology in chromosomally typical populations, future DS studies may benefit from focusing on the cortex in DMN regions, as such investigations may provide clues to the precocious onset of Alzheimer's disease in this at-risk group.

Keywords: Alzheimer's disease; Trisomy 21; cerebral cortex; intellectual disability; magnetic resonance imaging.

Figure 1.

Reductions in cortical surface area and increases in cortical thickness in DS relative to typical controls. A series of regression analyses were completed at 40 962 vertices in each hemisphere using the following regression equation: Cortical thickness or surface area (vertex_j) ∼ intercept + β1(Sex) + β2 (age centered) + β3 (group). T-statistics associated with the group term that exceeded the FDR-corrected threshold of ±2.2 were projected onto the cortex. Panel (a) displays surface area findings; panel (b) displays cortical thickness findings. Note that the range of T-values for surface area is negative 2.2 to negative 11 while the range for thickness is positive 2.2 to positive 11. This is because DS is associated with largely decreased surface area relative to controls but largely increased cortical thickness. To identify regions in which cortical surface area and thickness were most deviant from controls, the t-scores associated with the group term in the regression equation were ranked and those vertices above the 90th percentile for thickness and below the 10th percentile for surface area were identified. The approximate locations of the peak vertices within these regions are marked with small red circles in panels (a) and (b).