Complete closure prevents delayed adverse events after endoscopic treatment of superficial duodenal epithelial tumors: a multicenter retrospective study

Aims Delayed adverse events following endoscopic resection (ER) of superficial duodenal epithelial tumors (SDET) have occurred less frequently with complete closure; however, there are insufficient data. This study aimed to elucidate whether there were differences in the frequency of adverse events based on the degree of prophylactic closure.

Methods Of the 3047 SDETs with ER performed at 18 Japanese centers between 2008 and 2018, 2914 neoplastic lesions (adenomas and cancers), were excluding non-neoplastic lesions, included in the study. The incidence of adverse events was analyzed according to the degree of prophylactic closure. Subsequently, multivariate analysis using a logistic regression model was conducted to identify the risk factors associated with incomplete closure.

Results Of the 2914 cases, there were 156 CSP, 1238 EMR, 554 UEMR, and 966 ESD. The patients were categorized into three groups (2432 complete closure, 115 incomplete closure, and 367 non-closure).The rates of delayed bleeding (DB) in the complete/incomplete/non-closure were as follows: CSP, 1.3/0/0%; EMR, 1.4/7.7/16.3%; UEMR, 1.9/13.3/7.7%; ESD, 3.0/8.6/12.2%. For ESD, EMR, and UEMR cases, the DB rate was significantly lower in the complete closure compared to the incomplete/non-closure (ESD and EMR, p<0.001; UEMR, p=0.024). For CSP, there was no significant difference in the DB rate. According to the tumor location, the DB rate in the complete/incomplete/non-closure was observed 0.4/3.4/5.3% on the 1st portion (1st), 2.5/12.9/12.3% on 2nd portion (2nd), and 1.6/7.1/19.2% on the 3rd portion (3rd). Regardless of tumor location, the DB rate was significantly lower in the complete closure compared to the incomplete/non-closure (p<0.001).The rates of delayed perforation (DP) in the complete/incomplete/non-closure were as follows: CSP, 0/0/0%; EMR, 0/0/1.1%; UEMR, 0.2/0/0%; ESD, 1.2/10.0/3.5%. For ESD, the DP rate was significantly lower in the complete closure than in the incomplete/non-closure (p<0.001). For EMR, UEMR, and CSP, there was no significant difference in the DP rates. The DP rates in the complete/incomplete/non-closure were 0.2/3.4/0.7% on the 1st, 0.5/4.8/2.5% on the 2nd, and 0.3/14.3/3.8% on the 3rd. The complete closure had significantly fewer DPs than the incomplete/non-closure (2nd, p<0.001; 3rd, p=0.003).Multivariate analysis showed that lesions located on the oral side of the papilla (OR 1.663, 95% CI 1.114-2.483, p=0.013), lesion diameter>20 mm (OR 2.28, 95%CI 1.722-4.489, p<0.001), piecemeal resection (OR 2.157, 95%CI 1.261-3.691, p=0.005), and intraoperative perforation (OR 4.365, 95%CI 2.352-8.1, p<0.001) were independent risk factors for incomplete closure.

Conclusions Complete closure was preferable for ESD/EMR/UEMR of the SDET, regardless of the location. In particular, the following factors were associated with incomplete closure: lesions located on the oral side of the papilla, lesion diameter>20 mm, piecemeal resection, and intraoperative perforation.

 

MissingLinkText

Conflicts of Interest

Authors do not have any conflict of interest to disclose.

Publication History

Article published online:
27 March 2025