Outcome of endoscopy surveillance for Barrett's oesophagus.

Background: Endoscopic surveillance of individuals with Barrett's oesophagus is undertaken to detect early stage oesophageal malignancy. The impact of a surveillance programme on endoscopy resources and disease detection is uncertain.

Methods: In 2004, we commenced a structured Barrett's oesophagus surveillance programme. The surveillance protocol specifies surveillance interval and number of oesophageal biopsies required according to previous endoscopy and biopsy findings. The first 3 years of surveillance were reviewed to assess programme adherence, impact on endoscopy resources and the incidence of high-grade dysplasia and adenocarcinoma in patients undergoing surveillance.

Results: Four hundred five patients were enrolled in the surveillance programme, and 776 patient years of endoscopy follow-up were analysed. Four-quadrant biopsies every 2 cm throughout the Barrett's oesophagus were performed in 89.8% of endoscopies. A total of 93.7% of patients had surveillance endoscopy performed at the appropriate time interval. Formalizing surveillance was followed by a decrease in the mean time interval for endoscopy surveillance from 16 months to 15 months, although the mode endoscopy surveillance interval lengthened from 1 year to 2 years. The mean number of biopsies per endoscopy increased from 5.9 to 7. In four patients, T1 stage oesophageal adenocarcinoma was identified, and in six patients, high-grade dysplasia was identified (combined incidence of adenocarcinoma/high-grade dysplasia 1 per 77.6 endoscopy years of follow-up).

Conclusions: Structured Barrett's surveillance detects malignant progression at an early stage, which provides opportunities for curative surgical or endoscopic intervention. Formalizing surveillance resulted in a high rate of adherence to agreed guidelines and rationalized the use of endoscopy resources without significantly increasing workload.