Several disease and treatment factors in pediatric Crohn’s disease (CD) carry the potential to affect the brain and associated cognitive and emotional development. These include intestinal and systemic inflammation during active disease as well as anti-inflammatory treatments such as corticosteroids which have known neurotoxicity. In adults with CD, structural brain changes have been recently documented for both gray and white matter, particularly in fronto-limbic and subcortical regions—areas mediating both cognitive and emotional functions. Very little is known in pediatric CD. We previously demonstrated decreased white matter integrity in temporo-limbic fiber tracts in children with CD compared to healthy controls, with differential impact of steroids and inflammatory markers (Mrakotsky et al., 2012). We previously also found inflammation to be inversely correlated with cognition and mood in pediatric CD. We here provide further evidence for the effects of inflammation versus steroids on brain gray matter volume in pediatric CD.
Structural magnetic resonance imaging (MRI) was carried out in 35 children age 9–14 years (CD patients: n = 12, age-matched healthy controls: n = 23) using a 3 Tesla Siemens Trio MRI system. Anatomical whole brain T1-weighted magnetization-prepared rapid gradient-echo images were acquired at 1 mm3 voxel size. All subjects also underwent neuropsychological assessments (tests of attention, memory, and IQ; parent and self-report of pain, sleep/fatigue, emotional functions), and phlebotomy to obtain serum markers of inflammation (CRP, ESR, IL-6, TNFR2). CD patients were scanned and tested during steroid treatment for an acute flare (prednisone dose: M = 32.0, SD = 10.1 mg/day). Subcortical volumes (mm3) were measured using Freesurfer. Correlations and regression were performed for associations among measures, ANOVAs and t-tests for patient group comparisons. Demographics, SES and general cognitive ability were comparable between groups. Inflammation was higher in CD patients than healthy controls.
CD patients demonstrated smaller gray matter volumes than controls in basal ganglia, particularly the left (P = 0.001) and right putamen (P = 0.01), as well as poorer memory performance (P < 0.05), lower reported competence (academic, social) (P < 0.05), and lesser behavioral function (P < 0.05). Smaller putamen volume was associated with higher inflammation, higher steroid dose, poorer memory and poorer overall competence (see Table). Correlations remained significant after adjusting for steroid. Regression including several disease (inflammation, pain, sleep) and treatment (steroid, 6-mp, infliximab) variables in 1 model showed both inflammation and steroid (but no other variables) to be significant predictors for gray matter volume.
Results provide further preliminary evidence of an association between pediatric CD and brain structure, and suggest impact of both steroids and inflammatory markers on gray matter volume, particularly the putamen. As part of the basal ganglia and dorsal striatum, the putamen plays a critical role in learning and emotional processes and has been linked to dementia, ADHD, depression, and anxiety disorders. The negative correlation between putamen volume, steroids and inflammation is thus of clinical relevance, particularly in context of cognitive, academic and emotional functions found to be affected. Larger neuroimaging studies including CD patients off steroids and longitudinal designs are warranted to elucidate these initial findings.
 |
|
|
|
