Mo1108 CAN THE ORBIT SCORE PREDICT DELAYED BLEEDING AFTER COLORECTAL ENDOSCOPIC SUBMUCOSAL DISSECTION IN PATIENTS RECEIVING ORAL ANTICOAGULATION THERAPY?

Abstract

Oral anticoagulation therapy increases the risk of delayed bleeding after colorectal endoscopic dissection (ESD). The Outcomes Registry for Better Informed Treatment (ORBIT) score was recently proposed as a simple predictive model for major bleeding in patients with atrial fibrillation receiving warfarin or direct oral anticoagulants (DOACs). This study aimed to investigate the usefulness of the ORBIT score for predicting delayed bleeding after colorectal ESD in patients receiving oral anticoagulation therapy. This was a retrospective, single-center study. Colorectal ESD was performed on 581 consecutive patients with 635 lesions at our hospital between April 2005 and August 2017. Of these, we enrolled 23 patients receiving oral anticoagulation therapy. In patients with multiple lesions, the first treated or submucosal invasive lesion was considered the representative lesion. The ORBIT score was calculated from the total score of the following five factors: reduced hemoglobin/hematocrit and bleeding history (2 points each); and insufficient kidney function, antiplatelet treatment, and age >74 years (1 point each). In the original report, the high-risk group was defined as patients with a score of ≥4 points. Delayed bleeding was defined as bleeding requiring endoscopic intervention after ESD. Major delayed bleeding was defined as a hemoglobin loss of >2 g/dL or a transfusion requirement. The primary outcome was the incidence of delayed/major delayed bleeding associated with a high risk based on the ORBIT score. The secondary outcomes were predictive factors and diagnostic accuracy of the ORBIT score for delayed/major delayed bleeding. Of 23 patients, 9 were categorized as high-risk patients. The incidence of delayed bleeding and major delayed bleeding were 39.1% (warfarin, 3/8; DOAC, 6/15) and 17.4% (warfarin, 1/8; DOAC, 3/15), respectively. Delayed bleeding occurred more frequently in the high-risk group than in the non-high-risk group (66.7% vs. 21.4%, p=0.077). Major delayed bleeding was significantly more frequent in the high-risk group than in the non-high-risk group (44.4% vs. 0.0%, p=0.014). ORBIT score ≥4 was the only predictive factor for delayed bleeding in logistic regression analysis including age, sex, diabetes mellitus, hypertension, warfarin/DOAC, ORBIT score ≥4, location, macroscopic type, tumor diameter, procedure time, submucosal fibrosis, and histologic type (odds ratio, 7.33; 95% confidence interval, 1.11-48.26; p=0.038). The sensitivity and specificity of ORBIT score 4 for delayed/major delayed bleeding were 78.6%/73.7% and 66.7%/100.0%, respectively. All delayed bleeding cases were managed without surgery. ORBIT score might be useful for predicting delayed bleeding, especially major delayed bleeding, in colorectal ESD for patients receiving oral anticoagulation therapy.