METHODS AND RESULTS: A total of 125 418 Danish AF patients initiating OAC
therapy were identified using Danish administrative registers. Non-parametric estimation and semi-parametric absolute risk regression were used to estimate the absolute risks of
colorectal cancer in patients with and without lower GI-
bleeding. During a maximum of 3 years of follow-up, we identified 2576 patients with lower GI-
bleeding of whom 140 patients were subsequently diagnosed with
colorectal cancer within the first year of lower GI-
bleeding. In all age groups, we observed high risks of
colorectal cancer after lower GI-
bleeding. The absolute 1-year risk ranged from 3.7% [95% confidence interval (CI) 2.2-6.2] to 8.1% (95% CI 6.1-10.6) in the age groups ≤65 and 76-80 years of age, respectively. When comparing patients with and without lower GI-
bleeding, we found increased risk ratios of
colorectal cancer across all age groups with a risk ratio of 24.2 (95% CI 14.5-40.4) and 12.3 (95% CI 7.9-19.0) for the youngest and oldest age group of ≤65 and >85 years, respectively.
CONCLUSION: In anticoagulated AF patients, lower GI-
bleeding conferred high absolute risks of incident
colorectal cancer. Lower GI-
bleeding should not be dismissed as a benign consequence of OAC
therapy but always examined for a potential underlying malignant cause.