O-017 YI Inflammatory Versus Non-inflammatory Predictors of Specific Depressive Symptoms in a Large Pediatric Cohort with IBD.

BACKGROUND: Youth with inflammatory bowel disease (IBD) are at increased risk of developing depression. Clinical depression is a heterogeneous disorder comprising different symptom clusters, which may derive from different psychopathologic etiologies. Previous studies identified distinct depressive subtypes among youth with IBD, with nearly 20% experiencing somatic-symptom predominant depression. We aimed to identify depressive symptoms related to inflammatory disease activity in a large sample of youth with Crohn's disease (CD), the subtype of IBD associated with more systemic inflammation.

METHODS: Youth (ages 9-17) with IBD (n = 765) were recruited at Boston Children's Hospital and Children's Hospital Pittsburgh. Subjects with ulcerative colitis were excluded. Subjects were screened for depression with the Children's Depressive Index (CDI). CDI scores ≥12 were considered to represent clinically significant depressive symptoms (CSDS). Disease activity was measured using the Pediatric Crohn's Disease Activity Index (PCDAI). For analysis, PCDAI scores were divided into subjective, objective laboratory, and objective exam subscales. Data regarding demographics, inflammatory markers, IBD medications, and prior surgery were obtained from the EMR. Exploratory factor analysis (EFA) was performed to identify symptom profiles and create factor subscales. Multiple regression modeling was used to identify predictors of each subscale.

RESULTS: Five hundred fifty CD patients (51.8% male, average age 14.4 years) were included. Mean PCDAI score was 13.6, indicating mild disease. Using validated cutoffs, 40.7% had scores indicating remission, 32.7% mild, 6.4% moderate, and 7.3% severe disease. At baseline, 38.8% met criteria for CSDS, of whom 9.4% had prior surgery, 12.3% ostomy, and 28.1% were on corticosteroids. EFA revealed 4 symptom profiles: (1) somatic-affective, (2) low self-esteem, (3) suicidality, and (4) anhedonia. Univariate analysis suggested that females, patients on corticosteroids, those with elevated PCDAI scores or ESR, or lower hematocrit or albumin were more likely to have somatic-affective symptoms (P < 0.01). Multiple regression modeling explained 27.9% of the variance in somatic-affective symptoms (P < 0.0005). Subjective PCDAI symptoms (pain, stool frequency, and functional impairment) made unique significant contributions to the model. Patients with elevated PCDAI subjective and objective lab scores, ESR, CRP, and those on corticosteroids were more likely to experience anhedonia (P < 0.01). Multiple regression modeling explained 8.4% of the variance in anhedonia, with subjective PCDAI symptoms making unique significant contributions. Low self-esteem and suicidality correlated with the subjective PCDAI subscale (P < 0.01), but not with inflammatory markers.

CONCLUSIONS: Inflammatory markers and disease activity are significant predictors of somatic-affective symptoms and anhedonia in youth with IBD. Symptoms associated with other profiles may be driven more by functional symptoms than inflammation. Social support, coping styles, childhood trauma, and disease course should be evaluated as potential predictors of these clusters. Identification of symptom profiles can guide treatment of comorbid IBD and depression. Even if somatic-affective symptoms stem from inflammation, referral to psychotherapy can improve symptoms more rapidly.