Limbic System White Matter in Children and Adolescents With Attention-Deficit/Hyperactivity Disorder: A Longitudinal Diffusion Magnetic Resonance Imaging Analysis

Background: Attention-deficit/hyperactivity disorder (ADHD) is increasingly being recognized as a disorder linked to atypical white matter development across large-scale brain networks. However, current research predominantly focuses on cortical networks, leaving the developmental trajectories of many subcortical networks, including the limbic system, largely unexplored. The limbic system is crucial for emotion and cognition, making it a key area of interest in ADHD research. Methods: In this study, we used multishell high-angular resolution diffusion magnetic resonance imaging in a sample of 169 participants (72 with ADHD and 97 control participants) across 3 time points between ages 9 and 14 years to map the development of limbic system white matter. Diffusion kurtosis imaging and graph theory metrics were used to characterize limbic system white matter, alongside assessments of emotional regulation and ADHD symptom severity. Results: Compared with control participants, individuals with ADHD exhibited significantly lower microstructural organization, particularly in kurtosis anisotropy, within the bilateral cingulum bundle from childhood to adolescence. There were no significant group-by-time interactions in limbic system white matter metrics, suggesting no significant differences in developmental trajectories between the ADHD and control groups. Furthermore, brain-behavior analyses revealed that higher ADHD symptom severity was associated with fewer limbic system white matter connections, notably decreased routing efficiency and network density. Conclusions: These findings offer novel insights into the role of disrupted limbic system white matter, particularly the cingulum bundle, in ADHD pathophysiology, thereby broadening our understanding of the disorder’s neural mechanisms and opening promising avenues for future exploration of subcortical brain networks.