Dynamic Development of Regional Cortical Thickness and Surface Area in Early Childhood

Abstract

Cortical thickness (CT) and surface area (SA) are altered in many neuropsychiatric disorders and are correlated with cognitive functioning. Little is known about how these components of cortical gray matter develop in the first years of life. We studied the longitudinal development of regional CT and SA expansion in healthy infants from birth to 2 years. CT and SA have distinct and heterogeneous patterns of development that are exceptionally dynamic; overall CT increases by an average of 36.1%, while cortical SA increases 114.6%. By age 2, CT is on average 97% of adult values, compared with SA, which is 69%. This suggests that early identification, prevention, and intervention strategies for neuropsychiatric illness need to be targeted to this period of rapid postnatal brain development, and that SA expansion is the principal driving factor in cortical volume after 2 years of age.

Keywords: brain development; cerebral cortex; gray matter; human; magnetic resonance imaging.

Figure 1.

Regional growth of CT and SA over time. (A) CT anatomical LS means at neonate (blue), 1 year (red), 2 years (green), and adult (black); (B) SA anatomical LS means at neonate (blue), 1 year (red), 2 years (green), and adult (black). Only left hemispheric ROIs are shown; ROIs for both CT and SA exhibited symmetry. Quantitative results can be found in Supplementary Table 2 (CT), Supplementary Table 4 (SA), Supplementary Table 8 (Adult CT), and Supplementary Table 9 (Adult SA). (Note: SA values are determined by the size of the ROI as defined by the parcellation atlas).

Figure 2.

Early cortical surface measures development from birth to 2 years. Regional percent change for CT. (A–C) and SA (D–F) from birth to 1 year (A, D), 1 to 2 years (B, E), and birth to 2 years (C, F). Warmer colors indicate larger percent changes than cooler colors. Quantitative results can be found in Supplementary Tables 2–5.

Figure 3.

Comparison of 2 year old to adult cortical surface measure values. Percent of adult values for CT (A) and SA (B). For each ROI, we calculated the percent of adult value at 2 years of age. Warmer colors indicate larger percent of adult values than cooler colors. Quantitative results can be found in Supplementary Tables 8 and 9.

Figure 4.

Relationship of CT and SA to cortical volume. Raw data values for each cohort are plotted at each time point. (A) The log–log relationship of mean CT to cortical volume is shown for neonate singletons (red), neonate twins (orange), 1-year singletons (blue), 1-year twins (pink), 2-year singletons (green), 2-year twins (purple), and adults (black). Regression analysis r² values at neonate (singleton: r² = 0.0061; twin: r² = 0.1327), 1 year (singleton: r² = 0.0033; twin: r² = 0.0099), 2 years (singleton: r² = 0.0196; twin: r² = 0.0001), as well as in the adult cohort (r² = 5.98E−5) are not significantly different from zero. (B) The log–log relationship of total cortical SA to cortical volume is shown for neonate singletons (red), neonate twins (orange), 1-year singletons (blue), 1-year twins (pink), 2-year singletons (green), 2-year twins (purple), and adults (black). Linear regression r² values at neonate (singleton r² = 0.7177; twin r² = 0.8691), 1 year (singleton r² = 0.5924; twin r² = 0.9324), 2 years (singleton r² = 0.8395; twin r² = 0.9558), and adult (r² = 0.8907) are each significantly different from zero (P < 0.0001).