Endoscopic submucosal dissection for Barretts neoplasia in a Western setting: outcomes and lesion selection data from a large prospective series

Introduction

Endoscopic therapies for neoplasia in Barrett’s Oesophagus have become well established over recent years. Endoscopic Submucosal Dissection (ESD) offers the benefit of en bloc resection and accurate histological staging of lesions which is particularly advantageous in cancers and may eliminate the need for major surgery in selected patients. However, ESD is not widely available in the United Kingdom (UK) and it remains a high risk procedure in Western settings. Here we report our outcomes of ESD for Barrett’s neoplasia in a Western centre.

Methods

Retrospective review of prospectively maintained database.

Results

244 consecutive ESD were carried out in patients with Barrett’s neoplasia with visible lesions. The mean age was 69.8 years and 201 patients were male. Median Prague classification was C2M5. The median lesion size was 30mm (range 5–120mm). En bloc resection was achieved in 95.5% of cases. Bleeding, perforation and stricture occurred in 0.82%, 0.82% and 6.97% of cases respectively. These were all managed endoscopically with no patients requiring emergency surgery. R0 resection was achieved in 74% of cancer cases. Lesion morphology and histology are summarised in table 1. Paris classification IIa, Is and IIc were the most common morphology of cancers accounting for 88% of cancer cases, with only a small number of carefully selected IIb lesions demonstrating adenocarcinoma. In the cases of adenocarcinoma, 131 were intramucosal (pT1a) and 60 were submucosally invasive (pT1b). of the submucosally invasive cancers, 36 were SM1 and 24 >SM1. Following ESD, 93 patients underwent ablation of residual Barrett’s, 25 underwent further endoscopic resection, 21 entered endoscopic surveillance only and 25 patients underwent oesophagectomy. Of the patients who underwent oesophagectomy, 9 were for T1b >SM1 disease, 8 for poor prognostic factors (including poorly differentiated adenocarcinoma, lymphovascular invasion or signet ring cells), 4 for R1 resection at deep margin, 3 for patient choice and 2 for difficult recurrence.

Conclusions

Our data demonstrates that ESD by expert endoscopists for Barrett’s neoplasia is safe in a Western setting. The selection strategy for cases is effectively targeting cancers in patients based on endoscopic assessment, with almost 80% of procedures being carried out for early cancers. ESD for Barrett’s neoplasia was curative for the majority of our cases and helped identify those who would benefit from surgery.