Background: Surveillance programmes for oesophageal cancer in patients with Barrett's columnar-lined oesophagus have reported varying efficacies.
Aim: To review the effectiveness of an endoscopic surveillance programme for patients with Barrett's oesophagus in a UK general hospital.
Methods: Patients with Barrett's oesophagus (> or =3 cm histologically proven columnar-lined oesophagus) were identified from endoscopic and histological records and outcomes recorded when surveillance was performed with 2-yearly endoscopies and quadrantic biopsies at 3 cm intervals.
Results: One hundred and twenty-one patients (24%) entered the surveillance programme (mean age 60.2 years, 69.5% men, mean length of Barrett's mucosa 7.5 cm) of 505 patients identified with Barrett's oesophagus. Two hundred and five endoscopies were performed over a mean follow-up of 3.5 years. Five cases of high-grade dysplasia and two cases of adenocarcinoma were detected during the surveillance. One patient with high-grade dysplasia refused surgery and died of oesophageal carcinoma. The other six patients underwent oesophagectomy and five of the six resected specimens showed early (Tis to T2, N0) adenocarcinoma. All six patients remained well and tumour free at 24 months. No interval oesophageal cancers occurred.
Conclusion: This surveillance programme for classical Barrett's oesophagus was effective with six cancers being detected early and treated. The detection rate was 1/71 patient-years of surveillance. Endoscopic workload was not excessive and no interval cancers occurred.