Influencing Factors Of High Risk Of Aspiration During Painless Gastroscopy In Patients After Esophagectomy

Objective:The purpose of the study was to find the influencing factors of gastric contents with high risk of aspiration during painless gastroscopy in patients after esophagectomy,build a model and visualize it,so as to facilitate early detection of patients with high risk of aspiration,adjust anesthesia plan,and prepare interventions.Methods:This study was a cross-sectional,observational study.The subjects were adult patients for painless gastroscopy in the endoscopy room of a provincial cancer hospital from January 2021 to October 2021,who had undergone esophagectomy for esophageal cancer.The independent variables were gender,age,body mass index（BMI）,postoperative time,pyloroplasty,anastomosis site,surgical method,history of radiotherapy,history of chemotherapy,diabetes,gastric mucosal lesions,history of gastric retention,acid-suppressive drugs,gastric motility drug,fasting time,last meal type.The dependent variable was risk of aspiration.High risk of aspiration was defined as the presence of solid in the stomach or the ratio of the volume of liquid to the patient’s body weight greater than 1.5 ml/kg.The sample data was divided into training set and validation set according to the ratio of 7:3.The selection of variables and development of the model were performed by Least absolute shrinkage and selection operator regression（LASSO）.The prediction accuracy of the model was checked by the area under the receiver operating characteristic curve（ROC）.Decision curve analysis（DCA）was used to evaluate the performance of the model in supporting clinical decision-making.Finally,the model was visualized for clinical application using a nomogram.Results:A total of 304 patients were included,and the incidence of high risk of aspiration was 19.7%.LASSO regression screened out 7 variables,namely age[Beta coefficient（β）=0.108,odds ratio（OR）=1.114],BMI（β=0.204,OR=1.227）,pyloroplasty（β=0.905,OR=2.471）,diabetes（β=0.215,OR=1.240）,type of last meal（β=1.733,OR=5.868）,history of radiotherapy（β=-1.101,OR=0.333）,gastric mucosal lesions（β=0.502,OR=0.605）.The areas under the ROC curve of the training set and validation set were respectively 0.813 and 0.803,indicating high prediction accuracy.The DCA showed that when the probability threshold was set to 0.19,the net benefit of the training set and validation set were 0.115 and 0.119,respectively.Conclusion:During the painless gastroscopy in patients after esophagectomy,the incidence of gastric contents with high risk of aspiration was 19.7%.The results of the LASSO regression model showed that age,BMI,pyloroplasty,diabetes,and type of last meal were risk factors of high-risk gastric contents,while history of radiotherapy and gastric mucosal lesions were protective factors of high-risk gastric contents.The prediction accuracy and clinical benefit of this model were satisfactory,and it was converted into a nomogram,which is convenient for anesthesiologists and nurses to use.