Approach

Eosinophilic oesophagitis (EoO) symptoms and signs are non-specific, and diagnosis is based on clinicopathological features. Clinically, there must be symptoms of oesophageal dysfunction. Histologically, there must be an oesophageal epithelial infiltrate of ≥15 eosinophils per high-power microscopy field (eos/hpf).[1][2][3] To account for variations in the area considered within a high-power field and the increasing use of digital optical microscopy, UK guidelines recommend using an updated peak eosinophil count of ≥15 per 0.3 mm².[4]

Competing causes of oesophageal eosinophilia must be excluded prior to confirming the diagnosis; a finding of increased eosinophils alone on biopsy is not itself diagnostic. While there are unusual causes of oesophageal eosinophilia (such as oesophageal Crohn's disease, connective tissue or autoimmune disorders, hypereosinophilic syndrome, a more diffuse eosinophilic gastroenteritis, parasitic infections, drug reactions, pill oesophagitis, and graft versus host disease affecting the oesophagus) these can usually be excluded by the history, physical examination, and standard blood tests. From a practical standpoint, the most common condition to evaluate for as contributing to oesophageal eosinophilia is gastro-oesophageal reflux disease (GORD).[80]

History

EoO has been reported in patients of all ages, but is more commonly seen in children and young adults, with the peak prevalence occurring between aged 30 years and 40 years.[26] Approximately two-thirds of patients are male, although the reasons for this are unknown.[27][28]​ Family history of EoO may be present.[77] Up to 80% of children and 60% of adults have been reported to have concomitant allergic conditions (e.g., asthma, atopic dermatitis, allergic rhinitis/sinusitis, or food allergies).[31][32][33][34][35][36]

In adolescents and adults, dysphagia is the hallmark feature.[1][2][28] It is often a long-standing symptom, and severity can range from food going down slowly, to transiently sticking, to sticking for a longer period of time requiring regurgitation, to frank food bolus impaction requiring urgent endoscopy to clear the bolus.[4][13][14][15]

When taking the history, it is not enough to simply ask if the patient has trouble swallowing. Many patients have coped with symptoms for years, have modified their eating to chew thoroughly, eat slowly, drink copious liquids, lubricate foods, or avoid foods that tend to get stuck. They may also not go out to eat for fear of having an episode of food sticking in public. These avoidance and modification behaviours minimise symptoms, and should be asked about specifically. In addition to dysphagia, patients may report heartburn (which can mimic symptoms of GORD) and chest discomfort.

In children, symptoms are non-specific. Infants may have failure to thrive, feeding intolerance, crying/irritability, and vomiting. Older children may report abdominal pain, heartburn (again, mimicking GORD), regurgitation, nausea, vomiting, chest pain, or food avoidance.[1][2][3] It may be unclear whether some of these symptoms (e.g., chest pain or vomiting) represent dysphagia or transient food impactions as younger children are not able to fully describe the sensation they are experiencing. UK guidelines recommend considering a diagnosis of EoO in all adult patients with dysphagia or food bolus obstruction and in children of all ages whose symptoms are consistent with their age.[4] Children under 6 are more likely to present with feeding difficulties, while older children present with abdominal pain, dysphagia, or food impaction.[31]

Physical examination

There are no specific findings on examination to suggest a diagnosis of EoO. However, the patient should be assessed for other potential conditions in the differential diagnosis: for example, evidence of atopy (e.g., wheezing due to asthma, eczema, or hives on the skin), uncontrolled acid reflux (e.g., dental erosion), connective tissue disorders (e.g., joint hyperflexibility, arthritis, skin tightening), signs of malnutrition, or other unexpected findings.

Initial investigations

Oesophagogastroduodenoscopy with biopsy is required to document increased oesophageal eosinophilia necessary to diagnose the condition. It should be ordered in all patients with suspected EoO. There are a number of characteristic endoscopic signs of EoO. While these are non-specific (and not formally part of the diagnostic criteria), they can be highly suggestive of EoO.[1][2][3] Diagnostic endoscopy should be performed on patients receiving no treatment (or when a proton pump inhibitor has been discontinued for 3-4 weeks, if possible).[81]

Endoscopic signs include:

  • Fixed oesophageal rings

  • Focal oesophageal strictures

  • Diffuse oesophageal narrowing

  • Oedema or congestion of the mucosa with loss of normal vascular markings

  • Linear furrows

  • White plaques or exudates (which correlate with the histological finding of eosinophilic microabscesses)

  • Crêpe-paper mucosa (a sign of mucosa fragility where the oesophageal mucosa tears from insufflation or passage of the scope).

These findings can be formally graded with the EoO Endoscopic Reference Score, which accounts for the five most typical EoO endoscopic features: oedema, rings, exudates, furrows, and strictures.[82]​ It is important to note that these findings are not required for the diagnosis of EoO. Endoscopy can appear normal in a number of patients (most commonly children), and biopsies should still be taken if this is the case.[4]

After the oesophagus is assessed visually, biopsies are obtained. Recommendations on how many biopsies should be taken vary from at least 2-4 in the US, to at least 6 in Europe, with 6 considered the gold standard.[1][4][83] Eosinophilic infiltrate in EoO is patchy, and an increasing number of biopsies maximises diagnostic sensitivity.[1][3][84][85]​​​[86][87][88]​​​ If a patient presents with food bolus obstruction, sufficient biopsies should be taken at the time of endoscopy to ensure a diagnosis of EoO is not missed - if the obstruction has spontaneously cleared or insufficient diagnostic biopsies have been obtained at the index endoscopy, elective endoscopy should be arranged before discharge.​[4] Recommendations in the UK also specify that biopsies be taken from targeted (visibly abnormal) and non-targeted areas.​[4]

On histological examination of the biopsy specimens, the convention is to quantify the peak eosinophil count in the most inflamed area of the oesophageal epithelium. Eosinophils are quantified per high-power microscopy field, and the count (eos/hpf) is reported. A count of ≥15 eos/hpf is required to diagnose EoO.[1][2]​​​[3]​ The standard size of a high-power field is 0.3 mm². To account for variation in the area considered to be a high-power field and the increasing use of digital optical microscopy, UK guidelines recommend using an updated peak eosinophil count of ≥15 per 0.3 mm².[4]​ There are also supporting histological features including basal zone hypertrophy, lamina propria fibrosis, surface layering of eosinophils, eosinophil degranulation, and eosinophil microabscesses. These, and several other features, have been developed into an EoO Histological Scoring System (HSS).[89] The UK guidelines recommend that concomitant histological features be included in the histological description, with the peak eosinophil count, to aid diagnosis of EoO.​[4]

While the elevated eosinophil counts and the associated endoscopic and histological findings are all suggestive of EoO, they are not specific, so the clinician must interpret the combined clinical and histological presentation to determine whether EoO is the correct diagnosis.

A full blood count (with differential) should be considered in all patients to assess for peripheral eosinophilia, which is only rarely seen with EoO, and suggests that the patient should be evaluated for other systemic causes.  [Figure caption and citation for the preceding image starts]: Endoscopic findings including white plaques, linear furrows, and oedemaFrom the collection of Dr Evan S. Dellon [Citation ends].com.bmj.content.model.Caption@7bece9a[Figure caption and citation for the preceding image starts]: Endoscopic findings including oesophageal rings, linear furrows, oedema, mild white plaques, and narrowingFrom the collection of Dr Evan S. Dellon [Citation ends].com.bmj.content.model.Caption@38efdb22[Figure caption and citation for the preceding image starts]: Oesophageal biopsy showing a diffuse eosinophilic epithelial infiltrate as well as basal cell hyperplasia and spongiosisFrom the collection of Dr Evan S. Dellon [Citation ends].com.bmj.content.model.Caption@10463c61

Emerging investigations

There are currently no non-invasive or minimally invasive techniques available in routine care, although they are being studied.[1][3]

Transnasal endoscopy

  • A less invasive method for sampling the oesophagus. In two small studies, it has been well-tolerated with only topical anaesthetic in both children and adults.[90][91]​​ It may have a role in monitoring disease activity after treatment.

Oesophageal string test

  • A minimally invasive oesophageal sampling technique where a capsule containing a string is swallowed, and the string that is attached to the capsule is held outside of the mouth. The capsule dissolves and the string remains in the oesophagus where it absorbs inflammatory factors related to EoO. After 1 hour, the string is removed and sent to a specialised laboratory where the inflammatory factors can be measured. Initial studies have shown that these factors correlate extremely well with the same factors measured in oesophageal biopsies, and further work indicates the oesophageal string test may provide a safe and minimally invasive way of disease monitoring.[92][93]​​

Cytosponge

  • A minimally invasive oesophageal sampling technique where a capsule containing a sphere-shaped cytology mesh (sponge) is swallowed, and an attached string retained outside of the mouth. After 5 minutes, the capsule dissolves, which releases the sponge, and the string is used to pull the sponge through the oesophagus where it collects a tissue sample before it is removed from the mouth. This sample can then be assessed for levels of oesophageal eosinophils in a standard pathology laboratory. A preliminary study showed excellent correlation between the eosinophil levels in the sponge sample and in matched endoscopic biopsy samples.[94] Additional studies are ongoing and indicate that cytosponge is a safe, well tolerated and accurate way to assess histological activity, but the findings need to be validated in larger studies.[95]​ 

Oesophageal physiological testing

  • Assessment of mucosal impedance, a proxy for oesophageal barrier function and integrity, has promise for monitoring EoO. Several studies have shown that mucosal impedance is lower in EoO than in other oesophageal conditions such as erosive and non-erosive reflux disease and achalasia, and that this low impedance is seen throughout the entire oesophagus.[64][66][96] A minimally invasive approach to measurement of mucosal impedance is not currently available.

  • UK guidelines recommend oesophageal physiological testing for patients with EoO who remain symptomatic even though they have received appropriate treatment, have no evidence of fibrostenotic disease at endoscopy and are in histological remission.[4]

EoO diagnostic panel (EDP)

  • A 94-gene expression panel that uses a single oesophageal biopsy to yield a summary gene expression score. This score has been shown to discriminate EoO cases from non-EoO controls (including patients with GORD) with a high degree of accuracy.[61]​​[97] Gene expression also appears to normalise with successful treatment. This test is currently being commercialised.

Use of this content is subject to our disclaimer