Cortical morphology and cognitive impairments in adolescents with complex congenital heart disease

Individuals with complex congenital heart disease (CHD) are at increased risk for cognitive impairments linked to altered brain structure, including reduced cortical volume. Cortical volume can be interpreted in relation to other morphological properties, such as cortical thickness (CT), surface area (SA) and cortical folding (gyrification index: GI). CT and SA exhibit distinct developmental, genetic and evolutionary origins, while cortical folding emerges from developmental and mechanical processes associated with cortical SA expansion. Exploring these individual cortical morphometric changes may provide additional insight into cortical volume changes in CHD and their relationships with cognitive function. High-resolution 3D T1-weighted images, intelligence quotient (IQ) and executive function (EF) scores were acquired from a final sample of 49 patients with CHD (mean age = 13.65, male: 63%) and 80 controls (mean age = 12.86, male: 49%). Cortical reconstruction and volumetric analyses were performed using Freesurfer version 7.1. Group differences in cortical volume, total SA, mean CT and mean GI, as well as their associations with cognitive measures, were investigated through vertex-wise multivariate general linear modelling. Age, sex and parental education level were adjusted in all analyses. While both SA and CT were significantly reduced in the CHD group, both globally (P < 0.001) and regionally, the CHD group exhibited higher and lower local GI than controls in several regions, while showing no statistically significant difference in mean GI (P = 0.27). Among the three cortical measurements, SA was the most widely affected across the brain, and total SA showed a stronger and more consistent correlation with cortical volume (R ² = 0.91) than mean CT (R ² = 0.30) or mean GI (R ² = 0.38). Furthermore, SA and GI, but not CT, were widely associated with IQ and EF across the brain. These relationships were especially prominent in the frontal and occipitotemporal cortices, where patients with CHD exhibited stronger associations between SA and cognitive performance compared to controls. These findings indicate that cortical volume reductions observed in CHD primarily reflect reduced cortical SA rather than cortical thinning or altered cortical folding. While structure-function associations between cortical morphology and cognition are established in healthy populations, our data suggest that these relationships are accentuated in CHD, likely due to disrupted neurodevelopmental processes. This study underscores the importance of differentiating cortical morphometric features to improve our understanding of brain-behaviour associations and the neurobiological underpinnings of cognitive impairment in CHD.