A retrospective study on the safety, diagnostic yield, and therapeutic effects of endoscopic unroofing for small gastric subepithelial tumors.

Abstract

Accurate diagnosis of small gastric subepithelial tumors (SETs) is essential to assess their malignant potential. Endoscopic unroofing has been reported to yield sufficient tissue samples for histologic evaluation. This study aimed to evaluate the safety, diagnostic yield, and potential therapeutic effects of this technique over time. This retrospective analysis of prospectively collected clinical data identified patients who underwent endoscopic unroofing at the Medical University of Vienna from January 2003 to December 2012. Demographic data, indications for endoscopic unroofing, intraprocedural adverse events, hospital stay, histologic results, and follow-up procedures were reviewed. A total of 14 patients (7 men; 7 women; median age, 70 years; range, 51-95 years) underwent endoscopic unroofing of 14 gastric SETs with a mean diameter of 26 ± 13mm at EUS. In 9 of 14 cases, endoscopic unroofing was done exclusively for diagnostic purposes; in the remaining cases, it was performed with therapeutic intent because of bleeding from the gastric SETs. Unroofing was technically successful in 13 of 14 cases and revealed 8 cases of GI stromal tumor (GIST) and 1 case each of leiomyoma, fibroid polyp, glomus tumor, pancreatic rest, and nondiagnostic material at histology. Intraprocedural bleeding was the only adverse event (4 cases) and could be managed endoscopically. A follow-up EUS was available (median, 8 months) for 10 of the 14 patients. Notably, most patients showed complete regression of their gastric SETs after unroofing (on white light and EUS), including the glomus tumor, the leiomyoma, and 6 of the 8 cases of GIST. Endoscopic unroofing was safe and had a very favorable diagnostic yield in this study. Unexpectedly, it led to complete regression in most gastric SETs. Although it is not an oncologically curative treatment, endoscopic unroofing can be a valuable option to treat local adverse events in patients unfit for surgical therapy. (Clinical trial registration number: NCT02587923.).