Correlation of Colonoscopic and Histopathological Findings in the Diagnosis and Clinical Course of Pediatric Inflammatory Bowel Disease

Background: Clinical remission in pediatric inflammatory bowel disease (IBD) may not coincide with microscopic inflammation resolution. This study evaluated the relationship between recorded colonoscopic and histopathological findings in children with IBD and examined changes in reported microscopic activity among patients with paired pre- and post- treatment biopsy assessments.Methods: Medical records of children with ulcerative colitis (UC), Crohn disease (CD), or indeterminate colitis/IBD- unclassified (IBD-U) followed at a tertiary pediatric gastroenterology center between 2003 and 2016 were retrospectively reviewed. The original pathology report's categorical assessment of active or inactive inflammation was abstracted; validated histological scores were not available. Endoscopic findings were abstracted descriptively without retrospective application of validated endoscopic scores. PUCAI was used for UC. For Crohn disease, disease activity was assessed using the Pediatric Crohn's Disease Activity Index (PCDAI). Exploratory colonoscopic (0-5) and histopathological (0-8) feature counts were correlated using Spearman's rank coefficient; these transparent counts are not validated activity scores. Paired changes in reported microscopic activity were assessed using the exact McNemar test.Results: Seventy-one children were included: 34 (47.9%) had UC, 32 (45.1%) had CD, and 5 (7.0%) had IBD-U. Lower gastrointestinal histopathology was available for 64 patients; cryptitis was reported in 25 (39.1%), crypt abscesses in 16 (25.0%), basal plasmacytosis in 15 (23.4%), and increased plasma-cell infiltration in 15 (23.4%). Among 64 patients with complete component data, colonoscopic and histopathological feature counts were moderately correlated (Spearman rho=0.487, p<0.001); correlations were rho = 0.485 (p = 0.004) in UC and rho=0.654 (p<0.001) in CD. Among 47 patients with paired pre- and post-treatment biopsy assessments, reported microscopic activity decreased from 28/47 (59.6%) to 15/47 (31.9%; exact McNemar p = 0.011). The corresponding changes were 17/27 (63.0%) to 11/27 (40.7%) in UC (p = 0.146) and 11/19 (57.9%) to 4/19 (21.1%) in CD (p = 0.065).Conclusion: Recorded colonoscopic and histopathological feature burdens were related, but microscopic inflammation could persist despite clinical improvement. These domains should therefore be interpreted as complementary rather than interchangeable. Because the exploratory counts were not validated activity scores and treatment timing and combinations were incompletely captured, the results require cautious interpretation.