Surgical and endoscopic management of high grade dysplasia and early oesophageal adenocarcinoma.

BACKGROUND: The introduction of endoscopic techniques has led to debate about optimal management of early oesophageal adenocarcinoma. The aim was to evaluate patient selection and outcomes for endoscopic or surgical treatment at a tertiary referral centre. METHODS: A prospectively collected database of consecutive patients staged with high-grade dysplasia (HGD) or T1 oesophageal adenocarcinoma treated with curative intent between 2005 and 2013 was undertaken. All patients were discussed at the multidisciplinary team meeting. Surgical treatment was by thoracoscopic assisted or standard/laparoscopic assisted Ivor Lewis oesophagectomy. Endoscopic treatment was a structured programme of endoscopic mucosal resection (EMR) and/or radiofrequency ablation (RFA). Outcomes included treatment variables, recurrence and complications. RESULTS: 83 patients treated; 50 with endoscopic therapy (EMR only-4, EMR then RFA-22, RFA only-24) and 38 by surgery (33 straight to surgery and 5 following EMR). Median age (67) and mean follow-up (21 months) were similar. HGD was more common in the endoscopic group (32/50, 64%, vs.3/33, 9%, p = 0.0001). Significant complications were more common following surgery (13/38, 34%, vs. 1/50, 2%, p = 0.0001). There were two in-hospital deaths following oesophagectomy (1 open, 1 thoracoscopic). Endoscopic treatment beyond 12 months for persisting HGD/intramucosal disease was required in 2 patients. Recurrence of HGD/invasive cancer was diagnosed in 2/36 (5.6%, T1a recurrence) of endoscopic and 1/38 (2.6%, T2N0 - subsequent hepatic metastases) surgical patients. CONCLUSION: A management algorithm including both endoscopic treatment and oesophagectomy provides optimal outcome for these patients. Due to additional morbidity of surgery, endoscopic treatment is appropriate first-line treatment.