Analysis The Risk Factors For Postoperative Bleeding Of Endoscopic Submucosal Dissection

Objective Aim: This study aimed to retrospectively analysis the risk factors for postoperative bleeding of endoscopic submucosal dissection for gastrointestinal mucosal or submucosal lesions.Methods: This is a retrospective study to analysis a total of 473 lesions of 463 patients diagnosed with gastrointestinal mucosal or submucosal lesions from 2008 June to 2015 February and treated by ESD in our Digestive endoscopy center. Patients with or without postoperative bleeding were compared on the basis of age, gender, underlying diseases(hypertension, diabetes mellitus,coronary disease), location of lesions, procedure time, size of lesions, depth of invasion, pathology and so on to analysis the risk factors for postoperative bleeding of ESD.Results: During the treatment of the 473 lesions, in total of 24 cases of post-ESD bleeding occurred(5.1%)(24/473). There were 83 in esophagus,2cases of post-ESD bleeding occurred(0.4%).And 237 in gastric,16 cases of post-ESD bleeding occurred(3.4%).Then one hundred and forty three in colorectum,6cases of post-ESD bleeding occurred(1.3%). There were no statistical differences in the different regions between the post-ESD bleeding and nonpost-ESD bleeding(P>0.05).There were 12 patients of post-ESD bleeding treated by emergency endoscopy,6 patients needed blood transfusion treatment,2 patients were treated by emergency surgery, one patients suffered a fatal bleeding. The post-ESD bleeding team, there were 15 men and 9womem.In underlying diseases,9patients were hypertension,3patients were diabetes mellitus and 1patients is coronary disease,the average age was 60±12.949.Tere were no statistical differences in those variables(P>0.05).There were 9patients of low level intraepithelial neoplasia,11 patients of early cancer,3patients of the submucosal neoplasm, and one patient of mucosal inflammation/polyp occurred in post-ESD bleeding. There were no statistical differences in pathology and depth of invasion(P>0.05).The averageprocedure time was 107.13±85.130 min, and the average size of lesion was 4.646±2.3659 cm in the post-ESD bleeding team.Univariate analysis revealed that procedure time(P<0.001)and size of lesions(P=0.01)were significantly associated with post-ESD bleeding in the patients with or without bleeding. Multivariate analysis indicated that procedure time(P<0.001,odds ratio [OR]:7.050,95% confidence interval[CI]:2.410-20.625)was associated with a significantly higher risk of post-ESD bleeding.Conclusions: In conclusion, from the present retrospective study we found that, 1.procedure time is the independent risk factor for postoperative bleeding following ESD for gastrointestinal mucosal or submucosal lesions. 2.Patients with or without postoperative bleeding in age, gender, underlying diseases(hypertension, diabetes mellitus,coronary disease).3. Patients with or without postoperative bleeding in pathology and depth of invasion were not statistically significant.4. Size of lesions was not associated with a significantly higher risk of post-ESD bleeding. Information regarding operation risk factors should be useful for planning strategies for ESD.