Prediction and Prevention of Delayed Post-polypectomy-induced Ulcer Hemorrhage Based Upon an Anatomic Study and Multivariable Analyses

BACKGROUND: Better ways to predict and prevent delayed post-polypectomy-induced ulcer hemorrhage (DPPIUH) are needed. AIMS: (1) To correlate Doppler endoscopic probe (DEP) detection of PPIU blood flow with polyp artery size on histology. (2) To report risk factors which predict DPPIUH and prevent it. METHODS: 96 high-risk patients with benign colon polyps and 10-40 mm PPIUs were included in this cohort study which analyzed prospectively collected data. Coded polyps were analyzed by expert GI pathologists to report artery size at resection margins. 28 patients with severe DPPIUH's all DEP positive (+) for arterial blood flow including 15 with prior empiric hemoclip (HC) closure were compared with 68 similar high-risk patients without DPPIUHs: 34 DEP+ treated to obliterate blood flow; 22 DEP negative not treated; and 12 DEP+ not treated. 16 potential risk factors for DPPIUH were simultaneously assessed using multivariable logistic analysis and a classification tree model. RESULTS: The odds ratio (95% CI's) of DEP positivity predicting medium or large polyp arteries was 23.4 (5.73, 11.0)-p < 0.001. Risk factors predicting DPPIUH were DEP positivity, medium, or large size polyp artery, right colon polyp location, and empiric HC closure. DEP-guided treatment predicted the absence of DPPIUH. CONCLUSIONS: (1) DEP positivity and medium or large artery size on polyp histology highly correlated and were new risk factors predicting DPPIUH. (2) Other significant risk factors predicting DPPIUH were right colon polyp location and empiric PPIU HC closure. (3) DEP-guided treatment prevented DPPIUH.