Background: We do not routinely discontinue clopidogrel before colonoscopy because we have judged the cardiovascular risks of that practice to exceed the risks of post-polypectomy bleeding (PPB).
Aims: The aim of this study was to compare the rates of PPB for clopidogrel users and non-users.
Methods: We performed a retrospective, case-control study of patients who had colonoscopic polypectomy at our VA hospital from July 2008 through December 2009. We compared the frequency of delayed PPB (within 30 days) for patients on uninterrupted clopidogrel therapy with patients not taking clopidogrel. To minimize confounding from differences between groups in conditions that might contribute to PPB, propensity scoring was used to match clopidogrel users with controls based on numerous factors including age, aspirin use, number and size of polyps removed.
Results: A total of 1,967 patients had polypectomy during the study period; 118 were on clopidogrel and 1,849 were not. Logistic regression analysis revealed no significant difference in frequency of PPB between clopidogrel users and non-users (0.8% vs. 0.3%, P = 0.37, unadjusted OR = 2.63, 95% CI 0.31-22). Matched analyses using propensity scoring also revealed no significant difference in PPB rates between clopidogrel users and non-users (0.9% vs. 0%, P = 0.99).
Conclusions: The delayed PPB rate for our patients on clopidogrel was less than 1%, and PPB rates did not differ significantly between users and non-users. Our conclusions are limited by differences in therapeutic methodology between the groups, and our findings are most applicable to small polyps (<1 cm). We speculate that cardiovascular risks of routinely discontinuing clopidogrel before elective colonoscopy may exceed any excess risk of PPB.