Abstract

<h3>Background</h3> Predicting the risk of perforation during the endoscopic resection (ER) of gastric submucosal tumors (SMTs) originating from the muscularis propria (MP) could remind us to be more cautious and be prepared ahead of time. Moreover, we could select the appropriate patients for the inexperienced endoscopists. But for now, the predictive model was rare. Therefore, we conducted this study to figure out important parameters in the endoscopic ultrasonography (EUS) images to predict perforation, and try to build predictive models. <h3>Methods</h3> From May 1st, 2013, to January 15th,2021, consecutive patients with gastric SMTs originating from MP and received ER in our hospital were retrospectively reviewed. The patients were classified into two groups based on whether they had perforation (the case group) or without (the control group). To evaluate the depth of the SMT locates using the quantification index, we came up with a parameter called the ratio of inner distance to outer distance (I/O ratio). Logistic multivariate analysis was used to identify potential variables and predictive models. <h3>Results</h3> 199 EUS procedures (194 patients) were finally chosen for analysis. The median age of the patients was 53 years old, and 88 (44.22%) were males. The en bloc resection rate had no difference in the two groups, while patients in the case group had longer procedure time, higher probability of infection, and longer postoperative hospital stay. Multivariate analysis showed that age, I/O ratio, and the pathology of the tumors were risk factors for perforation, and we further built two models to predict perforation, with an AUC of 0.836 (for Model 1) and 0.755 (for Model 2). <h3>Conclusions</h3> EUS was useful in predicting perforation of ER for gastric SMTs originating from MP, and we have identified two models to predict perforation, which might serve as a useful tool in clinical practice.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call