Background
Inflammation is implicated in the neuropathology of bipolar disorder (BD). The association of C-reactive protein (CRP) with brain structure has been examined in relation to BD among adults but not youth.
Methods
Participants included 101 youth (BD, n=55; control group [CG], n=46; aged 13-20 years). Blood samples were assayed for levels of CRP. T1-weighted brain images were acquired to obtain cortical surface area (SA), volume, and thickness for three regions of interest (ROI; whole-brain cortical gray matter, prefrontal cortex (PFC), orbitofrontal cortex (OFC)) and for vertex-wise analyses. Analyses included CRP main-effects and interaction effects controlling for age, sex, and intracranial volume.
Results
In ROI analyses, higher CRP was associated with higher whole-brain SA (β=0.16; p=0.03), lower whole-brain (β=-0.31; p=0.03) and OFC cortical thickness (β=-0.29; p=0.04) within the BD group, and was associated with higher OFC SA (β=0.17; p=0.03) within CG. In vertex-wise analyses, higher CRP was associated with higher SA and lower cortical thickness in frontal and parietal regions within BD. A significant CRP-by-diagnosis interaction was found in frontal and temporal regions, whereby higher CRP was associated with lower neurostructural metrics in the BD group but higher neurostructural metrics in CG.
Conclusions
This study found that higher CRP among youth with BD is associated with higher SA but lower cortical thickness in ROI and vertex-wise analyses. The study identified two regions in which the association of CRP with brain structure differs between youth with BD vs. CG. Future longitudinal, repeated-measures studies incorporating additional inflammatory markers are warranted.