Incidence of diverticular bleeding: a population-based study.

OBJECTIVE: To determine the incidence of diverticular bleeding (DB) and examine the time trend of the incidence. Furthermore to study prognosis with regard to therapy and rebleeding.

METHODS: A retrospective, population-based study of patients with DB in a National University Hospital from 2006 to 2016. Patients were identified in an electronically stored colonoscopy database. Definite diverticular bleeding was defined as active bleeding, a nonbleeding visible vessel or adherent clot. Presumptive diverticular bleeding was defined as acute painless rectal bleeding leading to hospitalization with visible diverticula but no evidence of bleeding and no other colonic lesions or bleeding sites identified on endoscopy. A 30-day re-bleeding was determined after discharge.

RESULTS: A total of 3683 colonoscopy reports were reviewed, including 345 patients (males 51%) with presumptive 95% (n = 327) or definitive 5% (n = 18) diverticular bleeding. Overall 96% were treated conservatively, 3% endoscopically and 0.3% surgically. Only 5.8% of patients had a 30-day rebleed. After exclusion, 315 patients were included in the incidence calculations. The mean cumulative incidence of diverticular bleeding was 14/100,000 inhabitants per year. A time trend analysis of the incidence of DB revealed no significant change in incidence during the study period.

CONCLUSIONS: The mean incidence of colonic diverticular bleeding was found to be approximately 14 cases per 100,000 inhabitants and year. The incidence does not seem to have changed in the past decade. The vast majority of patients with diverticular bleeding did not require endoscopic therapy and could be managed with conservative treatment.