Risk of fracture in ulcerative colitis: a population-based study from Olmsted County, Minnesota.

BACKGROUND & AIMS: Osteopenia is common in patients with ulcerative colitis (UC), but less is known about fracture risk. Previously we were unable to demonstrate increased fractures in a population-based cohort with Crohn's disease. METHODS: Medical records of 273 Olmsted County, Minnesota residents initially diagnosed with UC in 1940-1993 were reviewed for evidence of subsequent fractures, as were records of a control cohort of county residents matched on age and gender. Cumulative fracture incidence after diagnosis was estimated by using the Kaplan-Meier method. The hazard ratio of fracture in cases relative to control subjects was estimated by Cox proportional hazards regression, which was also used to evaluate potential risk factors for fracture. RESULTS: Median follow-up was 13 years (range, 1 day-53 years). The cumulative incidence of any fracture from time of diagnosis was 40% at 25 years versus 42% in control subjects (P=0.615). The hazard ratio in cases compared to control subjects was 1.1 (95% confidence interval, 0.8-1.6) for any fracture and 1.3 (95% confidence interval, 0.6-2.8) for any osteoporotic fracture (hip, spine, or wrist as a result of moderate trauma). Other causes of secondary osteoporosis were associated with increased fracture risk, whereas estrogen use was protective. One hundred three cases received any corticosteroids (38%), and 34 (12%) had taken corticosteroids for 6 months or longer. Corticosteroids and bowel resection were not associated with fracture risk. CONCLUSIONS: In this population-based cohort of patients with UC, fracture risk was not elevated relative to matched community control subjects. Use of corticosteroids did not appear to significantly influence the risk of fracture.