Clinical Features and Treatment Responses in Pediatric Lymphocytic and Collageneous Colitis

Abstract

Purpose: Microscopic colitis (MC) has become increasingly prevalent in adult patients investigated for chronic watery diarrhea, yet characterization of pediatric MC is limited. We aim to describe our center's experience. Methods: The Mayo Clinic pathology database was searched from January 1995 to December 2011 for pediatric cases (<18 years) given a diagnosis of lymphocytic colitis (LC, >20 intraepithelial lymphocytes per 100 enterocytes with normal architecture) or collagenous colitis (CC, subepithelial collagen band >10 microns). Those with chronic diarrhea (>2 weeks) and visually normal colonoscopy were accepted as cases. Data on demographic information, labs (hemoglobin, albumin, autoimmune and inflammatory makers, and stool tests for enteric pathogens), medications within 3 months of presentation, past medical or family history of autoimmune disease, and response to treatments was abstracted. Results: The colon histologies of twenty-seven patients were consistent with LC or CC. Clinical review showed that one patient had Crohn's disease, one had culture positive Salmonella infection, and three presented with abdominal pain only; these patients were excluded. Among the remaining 22 (19 LC and 3 CC), 59% were female, and mean age at diagnosis was 11.6 years [range: 2.0-17.6 years]. Two had type 1 DM, three were ANA positive, and three had immunodeficiency. Of 19 patients who underwent an EGD, one had collagenous sprue and 4 had celiac disease. One presented after clearance of recurrent Clostridium difficile infection. Prior drug exposures included: nonsteroidal anti-inflammatory drugs (n=5), proton-pump inhibitors (n=7), and selective serotonin re-uptake inhibitors (n=3). Of 8 patients treated with steroids, 4 responded completely and 4 had partial response. Of 7 patients treated with mesalamines, 3 had complete, 1 had partial, and 3 had no response. Of 7 patients treated with bismuth subsalicylate, 2 had complete, 1 had partial, and 4 had no response. One patient with CVID, celiac sprue, and DM proved difficult to manage medically, and had frequent relapses. Conclusion: This is the largest single-center cohort of pediatric patients with microscopic colitis to date. LC was much more common than CC. As described in adults with MC, we observed associations with celiac disease, type 1 DM, and medications. We also saw an association with immunodeficiency. The majority of children responded to steroid treatment, with less response to mesalamine and bismuth subsalicylate. We also observed histologic evidence of MC in patients with isolated abdominal pain. It is unclear whether this represents a nonspecific finding, or whether MC in children may cause abdominal pain without diarrhea.