Association of preoperative endoscopic ultrasound assessment with outcomes of endoscopic resection for gastrointestinal subepithelial lesions

BACKGROUNDGastrointestinal subepithelial lesions (SELs) are often asymptomatic in early stages, yet some carry malignant potential.Accurate assessment of their nature and layer of origin is crucial for treatment decisions.Although endoscopic ultrasound (EUS) clearly visualizes lesion characteristics, evidence on its efficacy and safety in endoscopic resection of 1.5-4.0cm SELs remains limited. AIMTo investigate the role of EUS assessment in minimally invasive endoscopic treatment of 1.5-4.0cm gastrointestinal SELs selected for endoscopic resection. METHODSIn this single-center retrospective observational cohort study, data were collected from patients with gastrointestinal SELs treated at Jiujiang No. 1 People's Hospital from January 2020 to December 2022.Of 210 patients screened, 81 were excluded according to the study criteria and 129 were included.Patients were classified according to routine clinical practice into an EUS-assessed group (87 cases) and a non-EUS-assessed group (42 cases).All 129 included patients completed 30-day follow-up, and no data were missing for the major study variables.Because allocation to EUS assessment was not randomized, the comparison was potentially affected by confounding by indication.Procedure-related outcomes, postoperative pathological findings, and complications were compared between groups.Given the small number of incomplete resections and complications, multivariable logistic regression analyses were interpreted as exploratory. RESULTS / 31EUS diagnostic accuracy varied according to the lesion layer of origin, with high accuracy for lesions originating in the muscularis mucosae and submucosa and lower accuracy for lesions originating in the muscularis propria (P < 0.05).The EUS-assessed group had higher complete resection and negative-margin rates than the non-EUSassessed group; mean operative time, intraoperative blood loss, and length of hospital stay were lower; and the overall postoperative complication rate was also lower (all P < 0.05).Multivariate analysis showed that the absence of EUS assessment, tumor diameter > 3.0 cm, origin in the muscularis propria/serosa, and gastrointestinal stromal tumor histology were associated with incomplete tumor resection; the absence of EUS assessment, tumor diameter > 3.0 cm, and operative time > 60 minutes were associated with postoperative complications (P < 0.05). CONCLUSIONIn this single-center retrospective cohort of patients with 1.5-4.0cm gastrointestinal SELs selected for endoscopic resection, preoperative EUS assessment was associated with higher complete resection and negative-margin rates and a lower overall postoperative complication rate.These findings support the potential value of EUS for preoperative characterization and treatment planning, but prospective multicenter studies are needed before routine use can be recommended for all patients.