Abstract
Background: Pathophysiological models of pediatric bipolar disorder (PBD) are lacking. Multimodal approaches may provide a comprehensive description of the complex relationship between the brain and behavior. Aim: To assess behavioral, neuropsychological, neurophysiological, and neuroanatomical alterations in youth with PBD. Methods: Subjects with PBD (n = 23) and healthy controls (HCs, n = 23) underwent (a) clinical assessments encompassing the severity of psychiatric symptoms, (b) neuropsychological evaluation, (c) analyses of event-related potentials (related to the passive viewing of fearful, neutral, and happy faces during electroencephalography recording, and (d) cortical thickness and deep gray matter volume measurement using magnetic resonance imaging. Canonical correlation analyses were used to assess the relationships between these dimensions. Results: Youth with PBD had higher levels of anxiety (p < 0.001) and borderline personality features (p < 0.001), greater commission errors for negative stimuli (p = 0.003), delayed deliberation time (p < 0.001), and smaller risk adjustment scores (p = 0.002) than HCs. Furthermore, they showed cortical thinning in the frontal, parietal, and occipital areas (all p < 0.001) and greater P300 for happy faces (p = 0.29). In youth with PBD, cortical thickening and P300 amplitude positively correlated with more commission errors for negative stimuli, longer deliberation times, reduced risk adjustment, higher levels of panic and separation anxiety, and greater levels of negative relationships, whereas they negatively correlated with levels of depression (overall loadings > or <0.3). Limitations: Small sample size, cross-sectional design, and limited variables investigated. Conclusions: This preliminary work showed that multimodal assessment might be a viable tool for providing a pathophysiological model that unifies brain and behavioral alterations in youth with PBD.