RESUMO
BACKGROUND & AIMS: Conventional endoscopic mucosal resection (C-EMR) is established as the primary treatment modality for superficial nonampullary duodenal epithelial tumors (SNADETs), but recently underwater endoscopic mucosal resection (U-EMR) has emerged as a potential alternative. The majority of previous studies focused on Asian populations and small lesions (≤20 mm). We aimed to compare the efficacy and outcomes of U-EMR vs C-EMR for SNADETs in a Western setting. METHODS: This was a retrospective multinational study from 10 European centers that performed both C-EMR and U-EMR between January 2013 and July 2023. The main outcomes were the technical success, procedure-related adverse events (AEs), and the residual/recurrent adenoma (RRA) rate, evaluated on a per-lesion basis. We assessed the association between the type of endoscopic mucosal resection and the occurrence of AEs or RRAs using mixed-effects logistic regression models (propensity scores). Sensitivity analyses were performed for lesions ≤20 mm or >20 mm. RESULTS: A total of 290 SNADETs submitted to endoscopic resection during the study period met the inclusion criteria and were analyzed (C-EMR: n = 201, 69.3%; U-EMR: n = 89, 30.7%). The overall technical success rate was 95.5% and comparable between groups. In logistic regression models, compared with U-EMR, C-EMR was associated with a significantly higher frequency of overall delayed AEs (odds ratio [OR], 4.95; 95% CI, 2.87-8.53), postprocedural bleeding (OR, 7.92; 95% CI, 3.95-15.89), and RRAs (OR, 3.66; 95% CI, 2.49-5.37). Sensitivity analyses confirmed these results when solely considering either small (≤20 mm) or large (>20 mm) lesions. CONCLUSIONS: Compared with C-EMR, U-EMR was associated with a lower rate of overall AEs and RRAs, regardless of lesion size. Our results confirm the possible role of U-EMR as an effective and safe technique in the management of SNADETs.
RESUMO
OBJECTIVE: To evaluate the risk factors for lymph node metastasis (LNM) after a non-curative (NC) gastric endoscopic submucosal dissection (ESD) and to validate and eventually refine the eCura scoring system in the Western setting. Also, to assess the rate and risk factors for parietal residual disease. DESIGN: Retrospective multicentre multinational study of prospectively collected registries from 19 Western centres. Patients who had been submitted to surgery or had at least one follow-up endoscopy were included. The eCura system was applied to assess its accuracy in the Western setting, and a modified version was created according to the results (W-eCura score). The discriminative capacities of the eCura and W-eCura scores to predict LNM were assessed and compared. RESULTS: A total of 314 NC gastric ESDs were analysed (72% high-risk resection (HRR); 28% local-risk resection). Among HRR patients submitted to surgery, 25% had parietal disease and 15% had LNM in the surgical specimen. The risk of LNM was significantly different across the eCura groups (areas under the receiver operating characteristic curve (AUC-ROC) of 0.900 (95% CI 0.852 to 0.949)). The AUC-ROC of the W-eCura for LNM (0.916, 95% CI 0.870 to 0.961; p=0.012) was significantly higher compared with the original eCura. Positive vertical margin, lymphatic invasion and younger age were associated with a higher risk of parietal residual lesion in the surgical specimen. CONCLUSION: The eCura scoring system may be applied in Western countries to stratify the risk of LNM after a gastric HRR. A new score is proposed that may further decrease the number of unnecessary surgeries.
