Ulcerative colitis extent varies with time but endoscopic appearances may be deceptive

Question What are the changes in the endoscopic and histological extent of ulcerativ e colitis (UC) if colonoscopy is repeated one year after diagnosis? Design Follow up study of a population based cohort of incident cases. Methods All new cases of UC occurring in a defined population were identified durin g a four year period (n=496) and when possible were subjected to a second c olonoscopy (n=408) with histological material available in 384 (78%). Results After initial treatment, 47% of UC patients had remained in clinical remiss ion until the time of follow up colonoscopy performed a median of 14 months after diagnosis. A total of 399 UC patients of whom 8% were symptomatic sh owed changes in the macroscopic distribution of colitis with progression in 14%, regression in 22%, and a normal macroscopic appearance in 30%. In com parison, histology showed progression in 20%, regression in 24%, and normal appearances in 24%. According to the Kappa statistics there was poor agree ment between extent assessed by histology and colonoscopy both at diagnosis and at follow up, At diagnosis, histology showed more extensive disease th an endoscopy in 4% and less extensive disease in 18% whereas at follow up, histology showed more extensive disease in 28% and less extensive disease i n 12%. Correlation between histology and endoscopy was greatest for pancoli tis both at diagnosis and follow up. Conclusions Agreement between colonoscopic and histological findings is better at diagn osis than at follow up. One year after diagnosis only a third of UC patient s will have the same endoscopic disease extent as that found at diagnosis a nd in a third endoscopic appearances will be normal.