Development Of A Predictive Model For The Risk Of Parasigmoid Hernia Based On Radiologic Predictors

Object: We analyzed risk factors for permanent sigmoid colostomy hernias.Based on these risk factors,we developed and validated this nomograph prediction model for the risk of permanent sigmoidostomy hernia(PSH).Methods: This study is a retrospective study.A total of 291 patients with permanent sigmoidostomy were included in the affiliated hospital of Qingdao University.Collect general clinical data of patients and relevant muscle and fat index of abdominal CT 2 months before surgery.Then,single factor analysis and binary logistic regression analysis of SPSS software were used to screen independent risk factors for parastomal hernia after permanent sigmoid colon.Based on independent risk factors,a nomograph prediction model of the incidence risk of permanent sigmoidostomy hernia was constructed.The external validation of the model is based on the data of 63 patients with permanent sigmoidostomy in Qingdao Municipal Hospital.We use the Bootstrap method for internal validation and external validation in the validation group.Results: 291 patients underwent permanent sigmoidostomy.61 cases of periostal hernia occurred within 2 years.There were 230 patients without periostal hernia.The median subcutaneous fat index measured by CT was 51.13 cm2/m2.The median rectus abdominis index was 3.65 cm2/m2.After 2 years of follow-up,61 patients had PSH,of which 27(44.26%)had PSH symptoms such as abdominal discomfort,pain and even intestinal obstruction.According to univariate and binary logistic regression analysis,gender,rectus abdominis index,subcutaneous fat index,age,BMI(body mass index),albumin and the maximum diameter of sigmoidostomy are independent risk factors for the occurrence of periostal hernia in patients with permanent sigmoidostomy.However,diabetes,hypertension,respiratory related diseases(including emphysema and chronic obstructive pulmonary disease),smoking history,chemotherapy history,drinking history,waist circumference,psoas major muscle index and total abdominal wall muscle index have no significant correlation with the occurrence of peristomal hernia.In both internal and external validation of ROC curves.We found that the sensitivity and specificity of internal validation were 0.757 and 0.967,respectively.The sensitivity and specificity of external validation were 0.744 and 0.900,respectively.The area under the working characteristic curve of subjects in the training group and the validation group was 0.941 and 0.887,respectively.According to the Hosmer-Lemeshaw test and the calibration curve of the model,we found that the model has a good prediction ability for the risk of permanent sigmoidostomy hernia.Conclusion: This nomogram helps with the preoperative identification and guidance of high-risk patients with parastomal hernia.This nomogram can be generalized to patients undergoing surgery for stoma by specialized surgeons.This provides a more visualized way for preoperative patients to account for the risk of postoperative parastomal hernia.Provides personalized and disease-related estimates for patients undergoing colostomy.