The Construction And Application Of A Nomogram Model For The Risk Prediction Of Adult Parastomal Hernia

ObjectiveThe influencing factors of parastomal hernia are analyzed and determined based on large number of databases.On the basis of determining the influencing factors of parastomal hernia,a parastomal hernia risk prediction nomogram model is constructed to identify high-risk patients with parastomal hernia,providing a theoretical basis for the individualized scientific prevention of parastomal hernia,and promoting the health of patients.Methods1、Determination of influencing factors: Search the literature on the factors affecting parastomal hernia in major databases,the time limit is from the establishment of the database to June 2019.Software RevMan5.3 conducts Meta analysis on the retrieval results.To obtain more reliable parastomal hernia influencing factors,fixedeffects model and random-effects model is used to calculate the combined OR value and 95% CI.2、Model establishment: On the basis of determining the influencing factors of parastomal hernia,the questionnaire on the factors affecting parastomal hernia is determined through the discussion of experts of the research group.This article collects the clinical data of 262 patients undergoing ostomy who met the inclusion and exclusion criteria from January 2016 to December 2019 in a tertiary A general hospital in a city.Univariate and multivariate Logistic regression analysis are used to determine the independent influencing factors of parastomal hernia in stoma patients,and the R software is used to convert them into nomograms.3、Model verification: According to the same standard of the modeling group,data of 60 patients with stoma in another tertiary general hospital from January 2016 to December 2019 are collected.In the modeling group,internal verification is applied on the model.And the external verification is applied on the model in the verification group.In the both groups,the model’s discrimination is evaluated with area under the curve(AUC),and the calibration is evaluated using the calibration curve.Result1、Meta analysis results: A total of 13 articles are included,with a cumulative total of 1857 cases in the case group and a cumulative total of 5888 cases in the control group.The results shows that the risk factors for PSH are high BMI,female,COPD or chronic cough,hypertension,laparoscopic surgery,and stoma infection.Extraperitoneal stoma is a protective factor for parastomal hernia.Smoking,diabetes,preoperative radiotherapy/chemotherapy,previous history of abdominal surgery,and timing of surgery have nothing to do with the occurrence of parastomal hernia.Stoma pore size,hypoproteinemia,transrectus abdominis stoma,past hernia repair history,and the use of steroids need more research to further verify.2、Modeling results: The incidence of parastomal hernia in the modeling group is37.80%(99/262),and the incidence of parastomal hernia in the verification group is21.67%(13/60).The final factor that entered the prediction model are BMI,Postoperative time,stoma aperture size,stoma location and postoperative intestinal obstruction(P<0.05).3、Model verification results: The area under the ROC curve of the nomogram model in the modeling group and verification group are 0.910(95%CI:0.8741～0.9452)and 0.923(95%CI: 0.8498～0.9963),respectively.In the modeling group,the best cutoff probability is 0.329,the sensitivity is 82.8% and the specificity is 84.8% at the best cut point.In the verification group,the best cut-off probability is 0.501 and at the best cut point,the sensitivity and specificity are 87.2% and 84.6%,respectively,showing good discrimination.The agreement of the predicted value of the nomogram model and the actual observation value is verified by the two sets of calibration curves.ConclusionMeta-analysis has determined the influencing factors of parastomal hernia.By collecting clinical data of stoma patients,a nomogram model that can effectively predict the risk of parastomal hernia in stoma patients has been established.For patients whose score exceeds the cut-off probability,that is,high-risk patients,timely targeted individual interventions should be carried out to reduce the incidence of parastomal hernia,improve the quality of life of patients,control medical costs,and promote patient health.