**5. Barrett's oesophagus surveillance**

As endoscopic recognition of dysplasia remains limited at present, a policy of regular endoscopic surveillance, in conjunction with a rigourous biopsy regimen, is recommeded. The frequency of surveillance endoscopy depends predominantly on the presence and degree of dysplasia identified, and also to a lesser extent on patient age, comorbidity and patient preference.

Several retrospective studies have demonstrated a survival benefit for patients with cancers detected by surveillance endoscopy rather than following symptom investigation. (Streitz et al., 1993; Peters et al., 1994; van Sandick et al., 1998; Fountoulakis et al., 2004; Corley et al., 2002) However, many other studies have failed to show this. (Wong et al., 2010) The Barrett's Oesophagus Surveillance Study (BOSS) is a multi-centre randomised control trial currently recruiting patients throughout the UK. In this study patients are randomised to either 'endoscopy at need' (no routine surveillance), or repeat OGD combined with BSG biopsy regimen every two years for a total of ten years. The study aims to define the objective value of endoscopic surveillance and the most appropriate surveillance protocol.

Despite the current lack of high level evidence, most clinicians elect to follow up patients with Barrett's oesophagus provided the pros and cons of surveillance have been fully discussed with the patient and they subsequently wish to proceed with surveillance endoscopy. Clearly, this approach is not suitable for all-comers, and management should therefore be individualised appropriately. For example, long-term follow-up of elderly patients who are unfit for intervention is not usually recommended.

Where surveillance is deemed appropriate BSG guidelines recommend 2-yearly endoscopy using a comprehensive biopsy protocol. (BSG Working Party, 2005) Quadrantic biopsies are recommended every 2cm of Barrett's oesophagus, with more targeted biopsies of any raised or suspicious looking areas (Box 1). Patients should have their reflux treated with a proton pump inhibitor so that the presence of oesophagitis does not complicate the histological identification of dysplasia. Patients with Barrett's oesophagus who are not enrolled in endoscopic surveillance should also receive proton pump inhibitor (PPI) therapy The effectiveness and dosage of PPI therapy with and without aspirin for the prevention of progression of Barrett's oesophagus is being addressed in the Aspirin Esomeprazole Chemotherapy Trial (ASPECT) in the united Kingdom.

For those patients undergoing routine endoscopic surveillance, the risk of developing adenocarcinoma is approximately 1% a year in the UK, and about half this figure in the US. (Jankowski et al., 2002) Critcially when dysplasia has been identified on endoscopy every effort must be made to ensure that the correct diagnosis has been reached and an appropriate management strategy should then be formulated.
