Article Text

Download PDFPDF

Endoscopic surveillance of patients with Barrett’s oesophagus
  1. H Barr
  1. Cranfield Postgraduate Medical School in Gloucestershire, Great Western Road, Gloucester GL1 3NN, UK;prof.barr{at}

Statistics from

I propose that a patient with a clinical diagnosis of Barrett’s oesophagus (BO) must have regularly endoscopic surveillance and protocol biopsy. BO is defined by endoscopically visible oesophageal columnar epithelium with intestinal metaplasia. The purpose of screening is detection of dysplastic change and early cancer, to allow early intervention and prevention of the suffering of symptomatic oesophageal adenocarcinoma and premature death.


The quandaries and uncertainties that occur in the minds of both the patient and their doctor when confronted with a diagnosis of BO illustrate the dilemmas of post modern evidence based medicine. Presentation of the facts leads to diverse interpretations. Surveillance can reveal ubiquitous human folly, best exposed by Swift’s wretched Struldbrugs decaying into immortality (Gullivier’s Travels, 1726). I will argue that Barrett’s surveillance prolongs life not death, relieving suffering. The current myopic approach to patients with BO exposes the “shoulder shrug” of restrictive reactive medicine.


Is the disease an “important” health problem?

To justify screening, the disease must meet criteria of “importance”, gauged by incidence, mortality, morbidity, and public perception. Over 25 years there has been a fivefold increase in oesophageal adenocarcinoma; the rate of increase exceeds that of any other cancer (8% per annum).1,2 Symptomatic adenocarcinoma is a lethal disease; 50% of patients have extensive locoregional or metastatic disease.3 Of those selected for resection, 73% have invasive tumours (>pT2), 60% have lymph node metastases, and 18% other metastases.4 Invasive cancer (>pT1) has a devastating biological predeterminism as 80% of patients have bone marrow micrometastatic disease.5 A median five year survival rate of 21% and a perioperative mortality of 7% demonstrate the impotence of current approaches with radical surgery and multimodal therapy.6,7

Is there a detectable preclinical phase?

There is now a clear causal relationship between symptomatic gastro-oesophageal reflux and oesophageal adenocarcinoma.8 Chronic reflux results in Barrett’s metaplastic change, and the route to carcinoma is a stepwise progression through dysplasia, invasive carcinoma, and metastatic disease, with adenocarcinomas invariably associated with Barrett’s mucosa.9,10,11 Progression is not inevitable and is related to the increasing length of the Barrett’s segment, obesity, socioeconomic status, and White males. Between 3% and 5% of patients with reflux symptoms undergoing endoscopy have BO with metaplasia compared with only 0.73% having endoscopy for all indications. Autopsy data demonstrate that the true incidence is 376 per 100 000.12 A more specific preinvasive phenotype can be recognised, with distinct morphological changes of dysplasia.13 The prevalence of dysplasia in long segment (>3 cm) is twice that of short segment Barrett’s which is four times that of oesophagogastric intestinal metaplasia.14 The time to dysplasia and carcinoma progression is significantly faster in patients with visible BO, and the only reliable method of dysplasia detection is an endoscopic biopsy protocol.15

Is treatment of the disease before it is symptomatic advantageous?

Detection of mucosal cancer is an end to surveillance and treatment with surgery or endoscopic ablation. However, high grade dysplasia can be managed by further surveillance, surgery, or mucosal ablation. Earlier stage disease is found in patients undergoing screening and is the major predictor of survival following surgery. Five year survival is 70% for mucosal cancer and 20% for invasive cancer. A non-randomised study reported significantly improved survival following surgical resection of surveyed patients compared with those with symptoms.16,17 The alternative strategy is to perform endoscopic ablation. A prospective study of 64 patients reported the complete eradication of high grade dysplasia and early type I, Iia, Iib, and IIc (<20 mm diameter) cancers.18 Over 45 patients treated by endoscopic ALA photodynamic therapy were free of high grade dysplasia (follow up 1–72 months).19 Effective reflux control may stabilise the metaplastic mucosa. A randomised double blind study has confirmed endoscopic regression of the metaplastic segment following proton pump inhibitor therapy.20

Is surveillance acceptable to the patient and health care provider?

Our society devotes considerable resource to the education of the individual about the importance of early detection of cancer, based on the assumption that the informed patient will seek help, participate in screening, and take some responsibility for cancer prevention. The message is confused if on detection of a premalignant condition we inform the patient to return when symptomatic, at which time treatment, if possible, is unlikely to cure and may be a monstrous medieval combination of mutilating surgery and toxic oncology.


Patients with Barrett’s oesophagus are the blameless victims of our deliberations. A patient with a colonic polyp with a 1 in 10 chance of malignant degeneration will have polyp clearance and screening. A similar risk in the metaplastic Barrett’s patient cannot be ignored. As proponents of surveillance we must convince our colleagues and heathcare providers that early detection and intervention in these patients is the only realistic method to impact on the disaster diagnosis of symptomatic oesophageal adenocarcinoma.


View Abstract

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.