Eosinophilic oesophagitis (EO) is essentially a new disease not generally recognised in the UK or many other western European countries until early this decade. It is characterised by dysphagia, usually intermittent with a wide range of symptom severity and disruption to quality of life. The mechanism underlying the symptom of dysphagia is often not clear in EO because the changes seen at endoscopy are very variable. In this issue of Gut (see page 1056), Mittal’s group from San Diego address the pathophysiology of dysphagia using endoscopic ultrasound, and their results are intriguing.1

Korsapati et al propose that the mechanism of dysphagia is through a discoordination of longitudinal muscle of the oesophagus.1 Standard manometry fails to show an abnormality of peristalsis in most patients with EO,2 and this is confirmed by Korsapati et al. They then use endoscopic ultrasound to measure the thickness (as a marker of muscle contraction) of the circular and longitudinal muscle of the oesophagus and observe its reaction to an edrophonium stress test. The combined thickness of the muscle and submucosal layers is significantly greater at rest in patients with EO compared with controls. During contractions, however, the thickness of the muscle in patients has a lower peak than in controls, indicating dysfunctional contraction. In patients with symptomatic EO suffering dysphagia, they observed that the longitudinal muscle fails to react normally, with both a weaker contraction and a time delay (asynchrony) during the edrophonium stress, whereas the circular muscle seems to be unaffected and works normally. This is the first observation of a dynamic effect of EO that contributes to the symptom of dysphagia.

Standard manometry may fail to show abnormalities because the condition is intermittent in many, or because the dynamics of a standard manometry test are insufficient or insensitive to detect the abnormality that is present. There is the potential that high resolution manometry or impedance testing in patients with EO may provide further clues to the pathophysiology of dysphagia in this disease. Alternative reasons for dysphagia are of course clear in those patients with stricture, but it is not clear how often fibrous stricture in EO or just a narrow calibre lumen with poor compliance contribute to the symptom of dysphagia. Other endoscopic abnormalities in EO include circular rings, sometimes quite fibrous in nature, and again in these patients the mechanism for dysphagia may be structural fibrosis rather than muscular discoordination. Radiological images of EO published in the radiology literature frequently show significant strictures, and these may change the nature of dysphagia from an intermittent muscular phenomenon to a constant obstructive rigidity.

The condition of EO, when recognised, usually results in a major disruption to quality of life and in extreme cases a threat to life from spontaneous perforation. The challenge in managing EO is to understand the pathophysiology in the individual patient and how the symptoms of dysphagia or chest pain are generated. No single approach is optimum, and individual consideration of a range of therapies is required.

Although the first case series of EO in adults were described in 1993 and 19942 3 and in children in 1995,4 little progress was made in understanding or even recognising the disease in adults until after 2003. Paediatric gastroenterologists embraced the condition early because they used biopsies routinely during endoscopy in young children and because children responded well to elimination or restriction diets. While systemic steroids and dilatation are helpful in some patients, their long-term value is limited by potentially serious side effects.

Since 2003 recognition of EO in adults increased from other centres and countries partly because of the description of its natural history by Straumann5 and partly because of new effective threatment of Montelukast tablets, or topical steroids.6 7 Montelukast is a leukotriene D4 receptor antagonist commonly used in bronchial asthma, which stabilises eosinophils preventing degranulation and, similarly in EO, it may reduce the effects of eosinophil degranulation on muscle discoordination in the oesophagus. In the past 5 years there has been a dramatic rise in the frequency of publications on EO or oesophageal eosinophilia, with >500 publications cited in PubMed. So far, publications have been restricted mostly to the western developed countries, but the author of this commentary is aware that the condition is diagnosed in North African and the Middle East and is not exclusive to the Caucasian race. There are now many thousands of reported patients with EO worldwide, and the disease is now considered to be common.

Concern over the accuracy, sensitivity and specificity of diagnostic criteria has resulted in the current guideline that the diagnosis should be based on a history of symptoms consistent with oesophageal dysfunction in patients who have not responded to proton pump inhibitors and in whom at least six biopsies (two from each of the upper, mid and lower oesophagus) have been taken within which at least one high power field demonstrates >15 intraepithelial eosinophils.10 These minimum diagnostic criteria are obviously not cast in stone but are important in helping to distinguish this condition from gastro-oesophageal reflux disease. Many patients with EO will have hundreds of eosinophils per high power field, with concentrations aggregated near the surface (eosinophilic microabscesses). Note that the endoscopic appearances, although quite characteristic, are variably expressed in patients and sometimes the endoscopy looks normal.8

There is a lot yet to learn about EO. What causes EO? Is it allergy? Is it reaction to luminal injury? Could it simply be a variant of reaction to refluxed gastric contents?9 The evidence from patients with poor response to proton pump inhibitors suggests that there is a distinct pattern of presentation that is not reflux related but a form of allergy.10 Why has it emerged in the past 20 years? Did it exist before this? Since it has become the most common cause of acute bolus obstruction10 and is now frequently diagnosed in many centers in the USA, the UK and other countries, it is quite possible that the coincidental existence of the two conditions gastro-oesophageal reflux disease and EO may be common.

The condition is frequently diagnosed in children and perhaps more often currently in the UK than in adults, but no epidemiological data are yet available. The natural history of EO is such that it is unlikely to disappear, and these children will need adult medical help in the future. The frequency of diagnosis in the UK is now much greater than the literature reports would imply (only one case series fully published6 and others mentioned in abstracts11 or letters), but the British Society of Gastroenterology has accepted a proposal by its Oesophageal Section to host a Disease Register of EO on its web site to be launched during 2009 which should help to identify the frequency of the condition in the UK and help to guide diagnosis and education about treatment options.

Future plans for research will depend on the collection of good epidemiological data. Meanwhile gastroenterologists need to consider that patients with predominant dysphagia but normal or inflamed endoscopic appearances who have a poor response to proton pump inhibitors may have EO. The distinct condition of EO will only be diagnosed if correctly biopsied based on the history and clinical suspicion. A successful diagnosis usually results in a good response to specific treatments.12 It is likely that every gastroenterologist will have seen such a patient and therefore if they have not made the diagnosis they may probably have missed it!