| Objective:This article aims to analyze the surgical indications,surgical methods,postoperative complications and prognosis of EN,explore its clinical application value,and establish an effective pancreatic fistula prediction model.Methods:The clinical data of 130 patients with Da Vinci robotic pancreas enucleation completed between December 2010 and December 2017 in our department were retrospectively analyzed.The patients'general conditions,preoperative conditions,intraoperative conditions,and postoperative conditions were reviewed.Factors such as univariate analysis were performed.Statistically relevant factors before surgery were selected and included in the binary logistic regression model for multivariate analysis.A scoring system was established to predict the occurrence of pancreatic fistula.Results:According to the four independent risk factors assigned and the establishment of predictive scoring system assigned 1 point for diabetic patients,the distance from the main pancreatic duct is less than 2.5mm assignment 1 minute,the tumor in the pancreatic surface cross-sectional area greater than 7.5cm2 recorded 1 minute,the tumor is located in the right pancreas Record 1 point for the sidelines and 0 points for the rest.This study selected ROC curve to evaluate the effectiveness of the prediction system.The AUC value was 0.804.The best critical point was 4 points.The specificity was 88.9%,the sensitivity was 79.5%,the missed diagnosis rate was 20.5%,and the misdiagnosis rate was 11.1%.Conclusion:The incidence of pancreatic fistula after removal of pancreatic cancer is high,but the prognosis is good after standardized diagnosis and treatment.The distance between the main pancreatic duct is small,the tumor has a large cross-sectional area,and the history of diabetes.The tumor located on the right side indicates a high incidence of pancreatic fistula.Preoperative predictive scoring system can better evaluate the possibility of postoperative pancreatic fistula. |
PDF Full Text Request