The Establishment Of A Risk Assessment Model For Delayed Bleeding By ESD In Early Colorectal Cancer And Precancerous Lesions

Objective: Delayed bleeding is one of the most common adverse events of colorectal ESD.In this study,we analyzed the clinical data of patients who had the operation of ESD because of colorectal mucosal lesions and explored the incidence rate of delayed bleeding and its risk factors after the treatment of early colorectal cancer and precancerous lesions.On this basis of risk factors,we developed a risk assessment model for delayed bleeding.Methods: In this study,we retrospectively reviewed the medical records of 131 patients from The general hospital of western theater command pla from October 2015 to October 2019 and 56 patients from Sichuan Mian Yang 404 hospital from January 2018 to August 2019 who had the operation of ESD because of early colorectal cancer or precancerous lesions.We mainly analyzed the patients-related,tumor-related and procedure-related factors.Patients are divided into the delayed bleeding（+） group and the delayed bleeding（-） group according to the fact whether delayed bleeding occurred after colorectal ESD and the incidence of delayed bleeding is calculated.The risk factors of delayed bleeding after endoscopic mucosal dissection（ESD）were analyzed by univariate analysis and Multivariate Logistic regression analysis.And we finally establish a risk assessment model based on risk factors.Results: In the total of 187 patients,intraoperative bleeding occurred in 64 patients（34.8%）and delayed bleeding occurred in 13 patients（7.0%）respectively.Univariate analysis revealed that there was statistical difference in lesion long diameter,specimen long diameter,the degree of submucosal fibrosis,the depth of lesion infiltration and whether intraoperative bleeding occurred between the delayed bleeding（+）group and the delayed bleeding（-）group（P<0.05）.Multivariate Logistic regression analysis revealed that specimen long diameter which is greater than or equal to 3cm（OR=9.400,95%CI 1.118-79.009,P=0.039）,existence of severe submucosal fibrosis（OR=38.663,95%CI 2.502-597.414,P=0.009）and the occurrence of intraoperative bleeding（OR=9.213,95%CI 1.848-45.939,P=0.007）are the independent risk factors which is related to the delayed bleeding（P<0.05）.A regression model for evaluating delayed bleeding was established based on independent risk factors: P（Y=1）=1/（1+exp[-（-15.661+2.241X₁+3.665X₃+2.221X₄）]）,the X₁ represents the specimen long diameter,X₃ represents the degree submucosal fibrosis and X₄ represents intraoperative bleeding.Hosmer-Lemeshow test was used to analyze goodnessof-fit of the model and the results showed that χ²=0.912,P=0.923,indicating that the assessment model fits well.The performance of the model was assessed by calculating the area under the receiver operating characteristic（ROC）curve with the 95% confidence interval（CI）.And the results showed that the area under the curve was 0.890（P<0.001,95%CI 0.817-0.963）.The best cut-off value was screened by the Youden Index,and it was found that when the cut-off value is greater than 0.823,the incidence of delayed bleeding increased.At this time,the prediction sensitivity is 92.3%,the prediction specificity is 75.3%,and the prediction accuracy is 93.0%.In order to simplify the evaluation model,the three independent risk factors were assigned scores according to the β coefficient of the regression model.The scoring model is tested internally,and as the score increases,the risk of bleeding increases.Conclusion: 1.Endoscopic mucosal dissection is an effective treatment method for early colorectal cancer and precancerous lesions,and the cure rate is high.2.The specimen long diameter is greater than or equal to 3cm,the existence of severe submucosal fibrosis and the occurrence of intraoperative bleeding are the independent risk factors of delayed bleeding after the treatment of colorectal ESD.3.During the treatment of colorectal ESD,if it is found that the specimen long diameter is greater than or equal to 3cm or the existence of severe submucosal fibrosis,adequate measures should be taken to prevent delayed bleeding.Besides,strengthen management after colorectal ESD is necessary.Clinicians should try to identify the delayed bleeding as soon as possible,so that the adverse effects of delayed bleeding can be controlled to the minimum.4.When intraoperative bleeding occurs during ESD treatment,it is not only necessary to deal with the immediate bleeding actively,but also handle the blood vessel stump carefully,which can help prevent delayed bleeding.5.The risk predictive model of delayed bleeding after the treatment of colorectal ESD can effectively predict the risk of bleeding.The model is simple and easy to use.It is believed that it can help identify the patients who have a high risk of delayed bleeding in the early clinical stage if continuous improvement is put in.so that clinicians can take active and effective preventive measures to he occurrence of bleeding.