Factors predicting adverse events associated with therapeutic colonoscopy for colorectal neoplasia: a retrospective nationwide study in Japan.

Abstract

Few large studies have evaluated the adverse events associated with therapeutic colonoscopy for colorectal neoplasia, including bleeding and bowel perforation. Our aim was to investigate factors associated with these events, using a Japanese national inpatient database. We extracted data from the nationwide Japan Diagnosis Procedure Combination database for patients who underwent therapeutic colonoscopy for colorectal neoplasia between 2013 and 2014. Therapeutic colonoscopy included endoscopic submucosal dissection (ESD), EMR, and polypectomy. Outcomes included bleeding, perforation, cerebro-cardiovascular events, and in-hospital death. A multivariable logistic regression model was used to evaluate factors associated with bleeding and bowel perforation. We analyzed 345,546 patients, including 16,812 (4.9%) who underwent ESD, 219,848 (63.6%) who underwent EMR, and 108,886 (31.5%) who underwent polypectomy. The rates of bleeding, bowel perforation, cardiovascular events, cerebrovascular events, and death were 32.5, 0.47, 0.05, 0.88, and 1.32 per 1000 patients, respectively. In the multivariate analysis, a higher bleeding rate was associated with being male, comorbid diseases, ESD, tumor size≥2 cm, and use of drugs including low-dose aspirin, thienopyridines, non-aspirin antiplatelet drugs, novel oral anticoagulants, warfarin, non-steroidal anti-inflammatory drugs (NSAIDs), and steroids. A higher bowel perforation rate was associated with being male, renal disease, ESD, tumor size≥2 cm, and drugs including warfarin, NSAIDs, and steroids. Although the incidence of adverse events after therapeutic colonoscopy was low, several patient-related factors were significantly associated with bleeding and bowel perforation.