Introduction It is very difficult to establish an accurate diagnosis for upper GI submucosal tumours. Biopsy during endoscopy cannot go deep enough. EUS is unable to give a tissue diagnosis. The risks of surgical resection are higher than the benefits as the lesion may very well be benign. As a result most of these patients keep having endoscopic surveillance as ‘possible’ GISTs.

Methods A retrospective cohort study of patients undergoing ESD for upper GI submucosal tumours. They were all referred to us as possible GISTs that were found to be growing in size on surveillance. ESD was carried out in all these cases. As these lesions are mostly bulky, gravity and patient positioning were utilised as traction during ESD to achieve deroofing and enucleation of these tumours. Any complications were recorded. Endoscopic follow up was performed to assess for incomplete resection or recurrence.

Results 21 submucosal lesions were resected by ESD between 2007 and 2013. 7 were oesophageal, 10 gastric and 4 duodenal. Sizes ranged from 10 to 35mm. Endoscopic clearance was achieved in all cases. Histology showed a wide range of diagnoses, mostly benign (table). There was 1 complication; a microperforation which was identified and clipped intraprocedurally, giving a complication rate of 4.7%. On follow up, there was 1 recurrence (recurrence rate 4.7%) which was managed endoscopically. 1 patient had surgery as the ESD specimen showed a synovial sarcoma. Endoscopic cure rate was 95.2%.

Conclusion ESD is a safe and novel, minimal access therapeutic technique which has the potential to transform management of submucosal tumours. Patients go from the uncertainty of having repeated endoscopies for an unknown diagnosis, to having it completely removed and cured in the vast majority, without the need for continuing endoscopies. In the remaining cases, ESD specimens provide an accurate histological diagnosis based on which definite management plans can be made.

Disclosure of Interest None Declared.