Current UK system of Barrett's oesophagus monitoring 'not cost effective'

In collaboration with Adfero

The current system to diagnose and monitor Barrett's oesophagus in the UK may not be as effective as it could be. This is because people diagnosed using current criteria could be less likely to go on to develop oesophageal cancer than previously thought, according to a new study at Queens University Belfast.

Barrett's oesophagus occurs when abnormal cells start to develop on the inner lining of the foodpipe. Around 1 to 2 in every 100 people with Barrett's oesophagus will go on to develop oesophageal cancer.

Barrett's oesophagus is usually diagnosed in people who have a long history of burning indigestion - a sign of acid reflux. It is diagnosed by looking for the abnormal cells in the foodpipe, but there currently isn't worldwide agreement as to which changes count as 'risky' Barrett's oesophagus, and which should be excluded because they are unlikely to develop into cancer.

People who are diagnosed with Barrett's oesophagus are usually seen regularly by a doctor and have repeated endoscopies to spot the early signs of cancer. Many countries, including the USA, restrict the definition of Barrett's oesophagus that needs such monitoring to those cases containing abnormalities known as 'specialised intestinal metaplasia' (SIM). In the UK, people can be offered monitoring even if they do not have SIM, but have other abnormalities.

Previous studies have suggested that between 0.58 and 3 per cent of cases of Barrett’s oesophagus per year will progress to cancer (or to precancerous changes).

But the latest study, which followed 8,522 patients with Barrett's oesophagus in Northern Ireland for an average of seven years, found this combined rate to be lower, at 0.22 per cent.

In this new study, cases with and without SIM were included in the analysis, reflecting the UK definition of Barrett's oesophagus. However, when the researchers estimated the cancer risk for people with SIM only, the rate was only slightly lower than prior estimates - 0.38 per year.

This could mean that people without SIM are at relatively low risk of developing oesophageal cancer, and it could be more cost-effective to restrict monitoring to people with SIM only.

Writing in the Journal of the National Cancer Institute, the study authors suggested that the findings may have implications for the routine monitoring of people with Barrett's oesophagus.

They claimed: "Current recommendations for surveillance are based on higher estimates of cancer risk among patients with [Barrett's oesophagus] than were seen in this study and, therefore, they may not be justified."

In an accompanying editorial, Dr Douglas Corley, from Kaiser Permanente's research division in California, said that the study provides one of the first estimates of cancer incidence among a large multicentre population of people with Barrett's oesophagus.

He added that further research is needed to determine whether monitoring or treatment of patients with Barrett's oesophagus actually leads to a decrease in deaths from oesophageal cancer.

Dr Rebecca Fitzgerald, a Cancer Research UK oesophageal cancer expert, commented: "We need to be very cautious when comparing this overall rate of 0.22 per cent per year with previous publications, as the definitions for Barrett's oesophagus vary around the world. These variations can dramatically affect estimates.

"Nevertheless, this is a very informative study. One of the main challenges in the management of Barrett's oesophagus is separating people who will benefit most from surveillance from those who don't need it because they are less likely to develop oesophageal cancer.

"This research suggests that restricting surveillance to those patients with Barrett's oesophagus defined by the presence of so-called specialised intestinal metaplasia would bring the UK more in line with other international guidelines and would improve cost-effectiveness."


  • Bhat, S. et al. Risk of Malignant Progression in Barrett's Esophagus Patients: Results from a Large Population-Based Study. J Natl Cancer Inst DOI: 10.1093/jnci/djr203