BACKGROUND It is rarely known whether antiplatelets increase the risk of bleeding after endoscopic submucosal dissection (ESD).

OBJECTIVE To evaluate the effect of antiplatelets on post-ESD bleeding.

DESIGN Retrospective study.

SETTING Single, tertiary-care referral center.

PATIENTS This study involved 1591 gastric neoplasms (815 adenomas and 776 early gastric cancers) in 1503 patients who had ESD between April 2005 and April 2010.

INTERVENTION ESD.

MAIN OUTCOME MEASUREMENTS Overt hematemesis/hematochezia, a drop of hemoglobin >2 g/dL from baseline, or requirement of endoscopic hemostasis, angiographic embolization, and/or transfusion.

RESULTS Of 1591 subjects, 274 took antiplatelets, among whom 102 discontinued them for 7 days or more before ESD. Post-ESD bleeding occurred in 94 subjects including 20 from the continuation group, 6 from the withdrawal group, and 68 from the no-antiplatelet group. In univariate analysis, antiplatelets, early gastric cancer (EGC), comorbidity, and specimen diameter were related to post-ESD bleeding. In multivariate analysis, EGC (odds ratio [OR] 1.839; 95% confidence interval [CI], 1.168-2.896; P = .009), comorbidity (OR 2.246; 95% CI, 1.280-3.939; P = .005), and specimen diameter (OR 2.315; 95% CI, 1.282-4.180; P = .005) were independent risk factors of post-ESD bleeding, whereas antiplatelet usage was not (OR 1.596; 95% CI, 0.877-2.903; P = .126). In subgroup analysis, continuous antiplatelet usage was not found to be an independent risk factor of post-ESD bleeding in multivariate analysis (OR 2.027; P = .146). Among 102 subjects who discontinued antiplatelets, 1 developed an acute cerebral infarction (1.0%).

LIMITATION A retrospective, single-center analysis.

CONCLUSION In ESD for antiplatelet users, continuous administration was not found to have an independent significant association with bleeding.