Corticosteroids, inflammation, and memory in pediatric Crohn's disease.

Abstract

PURPOSE: Pediatric Crohn's disease (CD) is a remitting-relapsing illness with chronic intestinal and systemic inflammation still commonly brought into remission with corticosteroids. Both inflammation and steroids can disrupt brain systems critical for memory, executive functions and mood. Children with CD therefore may be at particular risk for cognitive and emotional problems based on both their underlying disease and the therapies used to treat them. Although emotional problems (anxiety, depression) have frequently been associated with pediatric CD, little is known about cognitive outcomes in CD, particularly following steroid and immune challenges. Subsequent to our earlier reports of acute steroid effects on cognition, we investigate the effects of steroids and inflammation on memory and executive functions in a longitudinal on-off-steroid design. [Table Prsented] PARTICIPANTS AND METHODS: Behavioral and biomarker data from this ongoing NIH study were available on 48 (out of 74 enrolled) children age 8-16 years who have completed the study thus far. CD patients treated with steroids (> = 0.75 mg/kg/day prednisone) (n=26) were compared to inflammation-free functional abdominal pain controls (n=22) on standardized measures of memory (verbal and spatial), executive functions, IQ, mood, sleep, and pain severity at baseline (on steroids) and 6 months post-treatment (off steroids). Serum markers of inflammation included erythrocyte sedimentation rate (ESR), C-reactive protein (CRP), and pro-inflammatory cytokines. Demographics, IQ and pain severity were comparable between groups. RESULTS: During steroid treatment, the 'CD Steroid' group showed poorer word list recall, spatial learning and recall, and reported more emotional control problems than controls. Steroid effects remained after adjusting for inflammation and other disease covariates. Longitudinal repeated measures analyses showed that post-treatment, CD patients improved more than controls in emotion regulation (time x treatment interaction p=.016). In contrast, albeit all patients improved over time, CD patients previously treated with steroids did not improve significantly more than controls in memory performance. Higher steroid dose predicted poorer memory outcome post treatment. Immune marker analyses showed higher levels of inflammation to be variably associated with poorer neurobehavioral outcome independent of steroids (ESR: spatial memory, mood problems; IL-6: verbal memory, working memory, school and mood problems; TNF-a: executive and behavior problems; IFN-c: school and behavior problems). CONCLUSION: Results suggest differential impact of steroids and inflammatory markers on memory and emotion regulation in children with CD. Emotional side effects of steroids may be transient, whereas effects on memory may persist post-treatment which carries implications for counseling patients about potential neurobehavioral toxicity of this widely used drug. However, underlying inflammation itself can affect cognitive and emotional development warranting further investigation in pediatric CD.