Sex differences in atypical fronto-subcortical structural connectivity among children with attention-deficit/hyperactivity disorder: Associations with delay discounting

Abstract

Purpose: Atypical fronto-subcortical neural circuitry has been implicated in the pathophysiology of attention-deficit/hyperactivity disorder (ADHD), including connections between prefrontal cortical regions involved in top-down cognitive control and subcortical limbic structures (striatum and amygdala) involved in bottom-up reward and emotional processing. The integrity of fronto-subcortical connections may also relate to interindividual variability in delay discounting, or a preference for smaller, immediate over larger, delayed rewards, which is associated with ADHD, with recent evidence of ADHD-related sex differences.

Methods: We applied diffusion tensor imaging to compare the integrity of the white matter connections within fronto-subcortical tracts among 187 8-12 year-old children either with ADHD ((n = 106; 29 girls) or typically developing (TD) controls ((n = 81; 28 girls). Analyses focused on diagnostic group differences in fractional anisotropy (FA) within fronto-subcortical circuitry implicated in delay discounting, connecting subregions of the striatum (dorsal executive and ventral limbic areas) and amygdala with prefrontal regions of interest (dorsolateral [dlPFC], orbitofrontal [OFC] and anterior cingulate cortex [ACC]), and associations with two behavioral assessments of delay discounting.

Results: Children with ADHD showed reduced FA in tracts connecting OFC with ventral striatum, regardless of sex, whereas reduced FA in the OFC-amygdala and ventral ACC-amygdala tracts were specific to boys with ADHD. Across diagnostic groups and sex, reduced FA in the dorsal ACC-executive striatum tract correlated with greater game time delay discounting.

Conclusions: These results suggest a potential neurobiological substrate of heightened delay discounting in children with ADHD and support the need for additional studies including larger sample sizes of girls with ADHD to further elucidate ADHD-related sex differences in these relationships.

Keywords: Amygdala; DTI; Prefrontal; Reward; Striatum.

Figure 1.

Cortical and subcortical ROIs used for fiber tracking within the fronto-striatal executive (top row), fronto-striatal limbic (middle row), and fronto-amygdala limbic (bottom row) pathways.

Figure 2.

Estimated marginal means from the 2 Diagnosis x 2 Sex x 2 Hemisphere x 7 ROI RM-ANCOVA covarying for age and framewise displacement (FD) for each fronto-subcortical tract among the diagnosis x sex subgroups. OFC=orbitofrontal cortex, AMG=amygdala, vACC=ventral anterior cingulate cortex, lPFC = lateral prefrontal cortex, STR_limbic =limbic striatum, dlPFC=dorsolateral prefrontal cortex, STR_exec=executive striatum, dACC=dorsal anterior cingulate cortex *p<.05, d=Cohen’s d effect size for comparisons within sex (significant among boys only) after correcting for multiple comparisons.

Figure 3.

Estimated marginal means from the 2 Diagnosis x 2 Sex ANCOVA covarying for age and general ability index (GAI) for the game time (left) and money (right) delay discounting tasks among girls and boys with ADHD and typically developing (TD) controls. *p<.05, d=Cohen’s d effect size for comparisons within sex (significant among girls only).

Figure 4.

Scatterplot of Spearman partial correlation between ranked residual scores for game time delay discounting and FA in the dACC-STR_exec tract.