Analysis Of Influencing Factors Of Anastomotic Leakage After Laparoscopic Dixon Operation For Rectal Cancer And Construction And Validation Of Prediction Model

Objective: To explore the influencing factors of anastomotic leakage after laparoscopic Dixon operation for rectal cancer and build a visual prediction model,in order to provide a theoretical basis for clinical effective prevention and timely identification of anastomotic leakage after laparoscopic Dixon operation for rectal cancer.Methods: Clinical data of 537 patients who underwent laparoscopic Dixon surgery for rectal cancer in the Department of Gastrointestinal Surgery of Yijishan Hospital from August 2017 to July 2021 were retrospectively collected as the training set.These patients were divided into two groups according to whether postoperative anastomotic leakage occurred,and the difference analysis,univariate and multivariate logistic regression analysis were carried out to find the independent influencing factors of postoperative anastomotic leakage,and the correlation column graph prediction model was established.In addition,clinical data of 160 patients who underwent laparoscopic Dixon surgery for rectal cancer in the Department of Gastrointestinal Surgery of Yijishan Hospital from August 2021 to September 2022 were retrospectively collected as validation set.The subjects’ working characteristic curve,calibration diagram and clinical decision curve were drawn respectively in the training set and verification set to conduct internal and external verification of the model’s differentiation,calibration and clinical effectiveness.Results: Preoperative neoadjuvant therapy(OR: 3.43,95%CI: 1.70-6.92),tumor distance from anal margin <7 cm(OR: 3.78,95%CI: 1.78-8.00),preoperative albumin<35g/L(OR: 4.16,95%CI: 2.07 ～ 8.38)is an independent risk factor for anastomotic leakage after laparoscopic Dixon rectal cancer surgery.Preoperative clean enema(OR:0.48,95%CI: 0.24-0.94)is an independent protective factor for anastomotic leakage after laparoscopic Dixon rectal cancer surgery.A risk prediction model for anastomotic leakage after laparoscopic Dixon rectal cancer surgery was established based on the above four indicators,and the model was verified internally and internally.The area under the curve of the subjects’ operating characteristic curve was 0.774(P<0.05,95%CI:0.774-0.847)and 0.769(P<0.05,95%CI: 0.636-0.902),the calibration curve and the ideal curve in the calibration chart all fit well,and the clinical decision curves all prove that the model has clinical practicability.Conclusions: Preoperative neoadjuvant therapy,tumor distance from anal margin<7 cm and preoperative albumin <35g/L are independent risk factors for anastomotic leakage after laparoscopic Dixon rectal cancer surgery,and preoperative cleaning enema is an independent protective factor for anastomotic leakage after laparoscopic Dixon rectal cancer surgery.The prediction model established in this study has good differentiation and calibration degree and clinical practicability,which can provide certain help for clinical prevention and monitoring of anastomotic leakage after laparoscopic Dixon rectal cancer surgery.