The article by van Laethem et al in this issue (see page 747) describes a well conducted study using the argon plasma coagulator (APC) to treat Barrett’s oesophagus in 31 patients. However, complete eradication of Barrett’s mucosa was only confirmed histologically in 19 (61%), the remainder having a few residual Barrett’s glands under the new squamous epithelium. Seventeen of those with apparent complete endoscopic and histological eradication at early follow up were biopsied again a year later, eight (47%) of whom had relapsing islands of Barrett’s metaplasia despite continuous treatment with omeprazole. These results lead to the same conclusion as that emerging from many recent studies on the endoscopic treatment of Barrett’s oesophagus using a variety of techniques.1-5It is possible to destroy specialised columnar epithelium in the oesophagus with a range of techniques and the treated areas heal with regeneration of squamous epithelium. However, there is a high risk of leaving some Barrett’s glands under the regenerated squamous epithelium. This raises two crucial questions. Firstly, does it matter if a few glands are left behind and, secondly, can any of the treatment options be modified to give reliable destruction of all the Barrett’s glands without any unacceptable complications.

The first question is tackled in the paper, although is very difficult to answer. Statistically, if …