Application of neck/chest CT to establish the evaluation perforation model of esophageal foreign body

Objective: Establish a predictive model with the perforation risk of esophageal foreign bodies (EFBs)--jujube pit, based on the structure of jujube pit and the relevant diameter of neck/chest computed tomography (CT). Methods: A 6-year retrospective study enrolled 162 patients from November 2016 and October 2021, who were diagnosed as esophageal foreign bodies with jujube pits, proved by neck/chest CT and endoscopy. Clinical characters contained age, gender, and the time from ingestion to perforation. Neck/chest CT were collected for analysis. Those patients with incomplete clinical data or no imaging data have been eliminated. The jujube pit section (Intermediate Sheet Separator) has been stretched to maximum after 3D reconstruction, the longest diameter of jujube pit is defined as jujube pit size. The impaction angle was the angle between the longest diameter and the esophagus wall. Multifactor logistic regression and Receiver operating characteristic (ROC) curve analysis were used to calculate the predictive model of perforation risk of jujube pits. Results: There are 48 cases complicated with perforation among the 162 cases with complete data. Patients in the perforated group had higher age, jujube pit size and impaction angle than those in the non-perforated group ( P <0.05). Multifactorial logistic regression analysis showed that patients age, jujube pit size, impaction angle were the risk factors for concomitant perforation in patients with EFBs (OR=0.926, 95%Cl: 0.886-0.968, P =0.001; OR=0.075, 95%Cl: 0.020-0.281, P =0.000; OR=0.926, 95%Cl: 0.899-0.954, P =0.000) and based on this result the regression equation of the risk assessment model was established Logit(p)=15.919-0.077*patients age-2.586*jujube pit size-0.077*impaction angle. ROC curve analysis revealed that the areas under the ROC curve (AUC) values of patient age, jujube pit size, impaction angle and comprehensive prediction model were 0.65,0.73,0.82 and 0.89, with a sensitivity of 54%,79%,92% and 81%, and a specificity of 68%,58%,68% and 82%, respectively. Conclusion: In patients with EFBs, those who experienced perforation were significantly older and had larger jujube pit sizes and greater impaction angles compared to those without perforation. These factors were identified as risk factors for the development of perforation in EFBs patients. A risk assessment model that incorporates these variables can effectively predict the likelihood of perforation.