Evaluation of endoscopic submucosal dissection as a treatment option for duodenal tumors ≥ 20 mm: a systematic review and meta-analysis by lesion size.

The role of endoscopic submucosal dissection (ESD) in treating superficial non-ampullary duodenal epithelial tumors (SNADETs) remains controversial. This study aimed to evaluate the role of ESD for SNADETs by lesion size. A systematic review and meta-analysis were conducted to compare outcomes between ESD and endoscopic mucosal resection (EMR). For duplicate publications, the study with the largest sample was included. Key outcomes included rates of en bloc resection, intraoperative perforation, and delayed adverse events (AEs) (bleeding, perforation). Data were analyzed according to lesion size categories (< 20 mm and ≥ 20 mm). Corresponding authors were contacted for data clarification when necessary. Fifteen retrospective observational studies met the inclusion criteria. For lesions < 20 mm, there was no significant difference in the en bloc resection rates between both groups (OR: 1.12 [95% CI 0.50-2.51]); however, the intraoperative perforation (OR: 9.74 [95% CI 5.07-18.69]) and delayed AEs rates (OR: 4.21 [95% CI 2.54-6.98]) were significantly higher for the ESD group. For lesions ≥ 20 mm, ESD was associated with a significantly higher en bloc resection rate (OR: 6.17 [95% CI 3.52-10.80]) and intraoperative perforation rate (OR: 5.33 [95% CI 2.23-12.76]), whereas the rate of delayed AEs was not significantly different from that of EMR (OR: 1.34 [95% CI 0.58-3.11]). For SNADETs ≥ 20 mm, ESD demonstrates a higher en bloc resection rate while maintaining a comparable rate of delayed AEs compared to EMR. These findings suggest that ESD may be considered a potential treatment option for large SNADETs, with careful patient selection and risk assessment.