Barrett's Oesophagus

Original article by Tom Leach | Last updated on 26/4/2015
Printer-friendly version

Introduction

Also known as Columnar Lined oesophagus (CLO)

  • This is a complication of long term oesophageal reflux.
  • It looks a bit like white pasty finger like projections coming up from the bottom of the oesophagus.
  • Strictly speaking it is defined as >3cm columnar epithelium present at the bottom of the oesophagus.
  • The mechanism by which it occurs is not fully understood. Some argue that it is a migration of the epithelium of the bottom 2cm of the oesophagus further up the lumen – i.e. a migration of essentially the gastric epithelium. Others would say that as both squamous and columnar epithelium are derived from the same stem cells, it is just perhaps a change in the intra-luminal conditions that determine what those stem cells develop into; more acidic conditions will cause a columnar epithelium to form. The second idea is more popular.
  • It is found in 10% of patients who have an endoscopy for reflux symptoms.
  • It is seen endoscopically as a pinker/redder epithelium as opposed to the whiter epithelium that lines the rest of the oesophagus. At first it may just extend up as thin bands of epithelium, but later it may encircle the entire lower oesophagus.
  • On microscopic examination, the cells resemble those of the stomach, however they secrete intestinal-like mucins, and as a result, Barrett’s oesophagus is actually a form of intestinal metaplasia.
Metaplasia – This usually occurs as a result of an environmental factor. It is when an existing epithelium is replaced by a different one; one that now has an ‘advantage’ due to the change in environmental conditions.
  • It is a major risk factor in oesophageal adenocarcinoma - there is a lifetime risk of 5-10%. Only about 5% of Barrett’s patients will actually ever get dysplasia. The risk for a Barrett’s patient to get ANY CANCER in their lifetime is:
  • No dysplasia – 2% (this is almost exactly the same as the general population)
  • Low-grade dysplasia – 30%
  • High-grade dysplasia – 50%
  • As such is defined as a ‘premalignant condition’, and patients who have it are at 100x times greater risk of cancer than the general population.(i.e. risk of oesophageal adenocarcinoma  goes from 4 in 100 000 to 400 in 100 000. This however, is still a relatively small risk overall (0.4%), and as such the patients overall risk of getting any cancer is still about 2-3%.
  • 95% of Barrett’s patients will die of another cause other than oesophageal cancer.
  • Most cases of CLO are not detected until cancer develops.
 

Epidemiology

  • Associated with men >50, particularly Caucasians.
  • Increased risk due to smoking (but not alcohol)
 

Surveillance

  • When you first find Barrett’s/if you suspect oesophageal carcinoma then you would biopsy the oesophagus. This has very strict guidelines, and you have to biopsy every quadrant (i.e. imagine the oesophagus as having a square lumen, and biopsy all four sides) for every 2cm of Barrett’s present. So if the patient has 6cm of Barrett’s, you have to do 12 biopsies. The idea behind this is so that you don’t miss any possible areas of high grade dysplasia, because macroscopically (ie as you look at the oesophagus down the endoscope) you cannot see a difference between areas of low grade and high grade dysplasia.
  • The evidence for surveillance is not conclusive. The current BSG (British society of Gastroenterologists) guidelines state that you can do surveillance if you want to, but you don’t have to. As a result, some trusts offer it and some don’t.
  • If you do offer surveillance, then you must be sure that if anything is found, that the patient is fit enough to undergo surgery. E.g. do you really want to offer surveillance to a 90 year old man?
  • Where it is offered, surveillance is currently recommended for those with low grade dysplasia about every 6-12 months, and for those without dysplasia every 2-3 years. Dysplasia takes about 2-5 years to develop.
  • Even with surveillance, it is not that likely you will find a carcinoma any sooner (hence the option not to survey). For example, you might have missed an area of high dysplasia last time round, and by the time you survey again it has developed beyond a treatable stage.
 

Treatment

  • There is no evidence that treatment with PPI’s or anti-reflux medicine leads to recession of Barrett’s.
  • However, treatment is basically that of GORD.