Eosinophilic oesophagitis is caused by a non-IgE-mediated immune response to antigens mainly, but not exclusively, from food. One of its key features is impairment of the oesophageal epithelial barrier allowing penetration of food allergens, which stimulate type 2 T helper cells promoted by thymic stromal lymphopoietin (TSLP). The release of proinflammatory cytokines, such as interleukin-13 (IL-13) and eotaxin-3 (CCL26) promotes recruitment of eosinophils to the oesophageal epithelium. Eosinophils degranulate and release toxic proteins, promoting inflammation and finally fibrosis through mediators such as transforming growth factor beta (TGF-β).

There is also evidence that EoE may run in families, as 5% to 7% of paediatric patients have first-degree relatives with EoE. Several predisposing genetic variants have been identified, some of which had been previously identified in other forms of atopy (the TSLP gene and the CAPN14 gene, encoding calpain 14) and could have a synergistic effect on other, more specific variants (IL4/KIF3A, interleukin 4/kinesin family member 3A) in the development of EoE. However, evidence from epidemiological studies supports that this familial clustering mainly results from exposure to the same environmental factors, while genetic factors play a much lesser role. Recent studies suggest that certain factors in early life, such as maternal fever, preterm birth, caesarean delivery or the use of antibiotics or acid-reducing drugs in childhood are associated with an increased risk of paediatric EoE.

The clinical presentation of EoE varies depending on the age of the child and their ability to express their symptoms. A high level of suspicion for EoE is required, as patients can develop adaptive dietary behaviours (Table 1).

In older children and adolescents, symptoms are more specific (Table 1). Food impaction may be the first symptom to develop, as solid food dysphagia may be intermittent.

Allergic comorbidities (atopic dermatitis, asthma, rhinoconjunctivitis, IgE-mediated food allergy) are more frequent in patients with EoE compared to the general population, and their presence should heighten suspicion of EoE.

The upper endoscopy is the initial diagnostic test to perform when EoE is suspected. It may reveal characteristic endoscopic findings, although these are not pathognomonic.

Abnormal findings may include signs of inflammation, such as oedema (with attenuation or loss of the vascular pattern), exudates or whitish plaques or linear furrows, or signs of fibrostenosis, such as concentric rings (trachealization) and narrowing. Fibrostenotic findings are more frequent in adolescents and adults, reflecting disease progression towards oesophageal remodelling. Another common finding is a friable mucosa with a crepe paper appearance or a very narrow oesophagus.7 The grading system known as the endoscopic reference score (EREFS) has standardised the description of endoscopic findings, although studies analysing the concordance of this score with histological findings have yielded contradictory results.8

The mucosa may appear normal in 10% to 30% of cases, so it is essential to obtain oesophageal biopsy samples in all patients with suspected EoE.

The gold-standard for diagnosis is evidence of elevated eosinophil counts in biopsy samples of oesophageal mucosa. Consensus guidelines have established a count of at least 15 eosinophils per high power field (eos/HPF) for diagnosis of EoE. Other characteristic histological findings are eosinophil sheets, eosinophil microabscesses, basal zone hyperplasia, intercellular oedema or spongiosis, eosinophil degranulation and lengthening of papillae in the squamous epithelium.

The eosinophilic infiltration of the oesophagus may be patchy, which calls for the collection of biopsy samples from both the proximal and distal oesophagus (at least 6 in total) to improve the diagnostic yield.5,6 At the time of diagnosis, the presence of eosinophilic infiltration in the stomach and duodenum must be ruled out by examination of biopsy samples.

Diagnosis of EoE requires fulfilment of all of the following criteria:

• (a)

  Symptoms of oesophageal dysfunction,

• (b)

  Eosinophilic oesophageal inflammation with ≥15 eos/HPF, with extension limited to the oesophagus, and

• (c)

  Ruling out other possible causes of oesophageal eosinophilia (Table 2).

  Table 2.

  Diseases associated with eosinophilic oesophagitis.

  • Gastritis, gastroenteritis or eosinophilic colitis with oesophageal involvement.

  • Gastroesophageal reflux disease.

  • Achalasia and other oesophageal motility disorders.

  • Hypereosinophilic syndrome.

  • Crohn disease with oesophageal involvement.

  • Infections (fungal, viral, parasitic).

  • Connective tissue diseases.

  • Autoimmune diseases and vasculitis.

  • Skin disorders with oesophageal involvement (e.g. penphigus).

  • Drug hypersensitivity reactions.

  • Drug-induced oesophagitis (e.g. by antibiotics).

  • Graft-versus-host disease.

Recent consensus guidelines have identified proton pump inhibitors (PPIs) as one possible treatment of EoE.5,6 Therefore, these agents are no longer considered to play a role in the diagnostic process, and the term PPI-responsive oesophageal eosinophilia is no longer used, as the patients it refers have the same clinical, endoscopic, histological and genetic characteristics of patients that do not respond to this therapy.

In addition to inhibiting the secretion of acid, PPIs have an anti-inflammatory effect on the oesophagus, reducing cytokine levels after Th2 activation. It has been demonstrated that PPIs block binding of the signal transducer and activator of transcription STAT6 to the eotaxin-3 promoter, decreasing the translation of the protein,10 restoring the integrity of the damaged oesophageal mucosa in these patients11 and reversing the inflammatory transcriptome.12

A systematic review found a 50% rate of histological remission following PPI therapy in children and adults, with no appreciable differences based on the PPI used.13 The first prospective study analysing the response to PPI therapy in the paediatric population (following treatment with esomeprazole at a dose of 1mg/kg twice a day for 8 weeks) found a 68.6% rate of remission and a clinical response rate of 80%.14 As for long-term response maintenance, the same group found a rate of histological and clinical remission of 70% in paediatric patients at 1 year of follow-up with administration of esomeprazole at a dose of 1mg/kg/day.15

Although the optimal dose changes depending on the drug (Table 3), in general, administration of 1 to 2mg/kg/day divided in 2 doses is recommended for induction therapy. High doses may be associated with a greater response rate. When it comes to maintenance, a dose of 1mg/kg/day has proven efficacious, although attempts should be made to taper it down to the minimum effective dose. At present, predicting the response to PPI therapy is not possible, although recent evidence suggests that it may be associated with specific cytochrome CYP2C19 and STAT6 polymorphisms.16

As for the long-term safety of PPI therapy, studies in adults have demonstrated that it is safe for use between 5 and 12 years,17 but there are few data on the paediatric population. Nevertheless, side effects tend to be mild, infrequent and transient and mainly consist of headache, diarrhoea and in some cases urticaria. There is evidence suggesting that prolonged use of PPIs may increase the risk of gastrointestinal and respiratory infections and iron deficiency.

Proton pump inhibitors tend to be less effective than swallowed steroids, but due to their easy administration and favourable safety profile they are a good option for treatment of paediatric EoE.

Dietary treatment is the sole approach that targets the cause of the disease. Its ultimate goal is to identify which food or foods trigger oesophageal inflammation and to eliminate them from the diet in the long term. The strategy includes the induction of clinical and histological remission with an initial diet that is more restrictive, followed by sequential reintroduction of eliminated foods with clinical and endoscopic evaluations to identify the culprit food or foods. The advantage of dietary treatment is that it can achieve long-term remission while avoiding the potential adverse effects of pharmacotherapy. However, it is important to consider that it may have a negative impact on quality of life,18 although to date no studies have been published with the primary objective of assessing quality of life throughout dietary treatment.

The viability of dietary treatment must be assessed in patients who already are on restrictive diets for management of IgE-mediated food allergy. Dietary treatment requires patients and families to be motivated, and adherence to it is poorer in the adolescent population.

Consists in avoiding all foods and substituting a nutritionally complete formula with free amino acids. It achieves histological remission in up to 90% of patients with EoE.19 Its poor palatability, which often requires the use of a nasogastric tube, low adherence to treatment, the considerable difficulty in reintroducing foods (which requires multiple endoscopies), the psychological and social impact and its high cost are factors that substantially restrict its prescription.

Th elemental diet is recommended for only short periods of time, in young children with EoE and an important nutritional impact and/or very severe symptoms that preclude a natural diet.

It consists in eliminating from the diet foods for which the patient has a positive allergy test result (skin prick, CAP-RAST and/or patch test). It is the dietary strategy that generally has the lowest documented efficacy, with only approximately 45% of patients responding to it, and a high variability in the results reported by different centres.19 The results of available allergy tests have exhibited very poor agreement with the foods identified with the method of elimination and reintroduction combined with histological evaluations,20 as reported in a document developed by members of the European Academy of Allergy and Clinical Immunology (EAACI).21 For this reason, food allergy test-guided elimination diets are not recommended for treatment of paediatric EoE.

In this approach, the foods most commonly associated with EoE are eliminated regardless of food allergy tests. A research group in the United States was the first to try an empiric diet with elimination of 6 food groups (cow's milk, egg, wheat, soy, fish-shellfish and nuts), achieving a histological remission rate of 74% (Table 4).22 These outcomes were replicated in studies conducted in Spain.19,23

The sequential reintroduction of each of the 6 eliminated food groups accompanied by an endoscopic and histological evaluation every 8 to 12 weeks allows the accurate identification of the foods that trigger EoE in each patient,24 but it requires imposing very restrictive diets and performance of many endoscopies. Furthermore, these studies have found that up to 75% of responders had only 1 or 2 trigger foods, most frequently cow's milk, followed by eggs or gluten and less frequently soy and legumes (Table 4). For this reason, the empiric six-food elimination diet is not currently recommended in the dietary management of EoE.

Other authors have used less restrictive diets, such as the four-food elimination diet (milk, wheat, egg and soy) with an efficacy of up to 64%.25,26 A recent study assessed the efficacy of a stepwise approach that started with the elimination of 2 foods (milk and gluten), then increased to 4 foods (milk, gluten, egg and legumes) if the patient did not respond, and finally 6 foods (further eliminating fish/shellfish and nuts), achieving remission rates in children of 40%, 57% and 76% in each respective step. Compared to the diet that eliminates all 6 food groups from the outset, this stepwise approach allowed reducing the number of endoscopies performed and the time to diagnosis by 20%. An additional advantage is that 90% of responders to an empiric diet with elimination of 2 or 4 foods only reacted to 1 or 2 food triggers. These patients are good candidates for long-term dietary therapy, unlike patients that needed to step up to the elimination of 6 foods, who reacted to multiple food triggers.27

Cow's milk is the food most frequently involved in paediatric EoE. Preliminary prospective studies evaluating the efficacy of the exclusive elimination of cow's milk have found a remission rate of 44% associated with a positive impact on quality of life.28

In the long term, a few studies have confirmed persistence of histological remission for at least 3 or 4 years without drugs, but also an attrition rate of approximately 50%.26,29

In sort, less restrictive empiric diets are the first line treatment for management of EoE, starting with the elimination of 2 foods and stepping up to elimination of 4 foods in patients that do not respond and remain motivated to identify the involved allergens.30

Elimination diets can affect the intake of macronutrients and micronutrients. Nutritional follow-up is necessary during the process of diagnosis, especially in case of the most restrictive diets and in patients with previous elimination of several foods due to IgE-mediated allergy. Supplementation with micronutrients and/or elemental formulas is recommended in these cases to achieve the recommended dietary allowance.31,32

There have been reports of cases of moderate-to-severe IgE-mediated allergic reactions following the reintroduction of previously eliminated foods. Although these reactions are infrequent, assessment by an allergist should be considered, in addition to the controlled and safe reintroduction of foods, especially in children at high risk of atopy and after prolonged elimination.33

Appendix B (online supplemental material) details the elimination diets for the most frequent food triggers, as well as strategies to replace the eliminated foods to maintain balanced nutrition.

Topical steroid administration is preferred over administration of systemic steroids. Swallowed fluticasone propionate has exhibited a similar efficacy to that of oral prednisone in a randomised controlled trial in children. However, systemic steroids were significantly associated with more severe adverse events.34 The use of systemic steroids should only be considered in emergency patients with severe dysphagia and significant weight loss or severe stenosis.

Several meta-analyses have confirmed the efficacy of topical steroid therapy, with variable remission rates averaging 68% for fluticasone and 76% for budesonide.35 The agents used most frequently in clinical trials are nebulised fluticasone (suspended for inhalation in a pressurised canister) and budesonide in a viscous solution. Viscous formulations achieve higher rates of histological remission compared to aerosol formulations because they remain in contact with the oesophageal mucosa for longer periods of time, especially the distal oesophagus.36

The steroids used in current clinical practice are the formulations developed for other diseases, such as asthma or rhinitis. The most widely used steroids are oral viscous budesonide and liquid fluticasone (Table 5). The first topical steroid specifically designed for EoE in adult patients has been recently authorised by the European Medicines Agency. It is budesonide prepared as orodispersible tablets containing a 1mg dose. The efficacy and safety of this formulation was proven in a randomised controlled trial in adults, which found histological remission rates of more than 90% in both the proximal and distal oesophagus.37 At present, this drug is marketed in several countries in Europe, but is not yet available in Spain.

The recommended induction dose depends on the age of the child (<10 years or ≥10 years) and the steroid used (Table 5).

In patients that initially respond to topical steroids, long-term treatment is effective in maintaining remission in a proportion of them. A prospective study in children that responded to nebulised fluticasone found that long-term administration of doses similar to those used in the induction phase achieved sustained remission in 63% of the patients at 2 years of follow-up.38 In patients that respond to induction therapy, maintenance therapy with half the initial dose is recommended (Table 5).

Treatment with topical steroids appears to be safe and not cause adverse events. Most clinical trials have not found differences compared to placebo in the incidence of adverse events with the exception of oesophageal candidiasis, observed in 5% to 26% of the patients,36,39 most frequently asymptomatic and a chance finding in the follow-up endoscopic examinations.

There is little evidence on the risk of adrenal suppression induced by long-term topical steroid therapy in children. Most of the data is from observational studies with small sample sizes.40 Until more information is available, it may be advisable to monitor cortisol levels to prevent adrenal insufficiency in children with EoE if they are given high doses of topical steroids for long periods of time or receive concomitant steroid therapy through other routes (oral, inhaled or nasal).5

In paediatric patients, stenosis is usually the product of inflammation and responds to anti-inflammatory therapy, so dilation is reserved for very select cases with severe oesophageal stenosis that persists after this treatment.

There is a dearth of data on the management of patients refractory to the three treatments described above. Combination therapy with PPIs and steroids may be useful in some cases, although there is little evidence on the subject. Different studies have analysed the efficacy of the monoclonal antibodies anti-IL 5 (mepolizumab and reslizumab), anti-IL13 and anti-IL 4 receptor α (dupilumab) in children and adults with EoE, and dupilumab has proven most efficacious. Biological therapy may be considered in cases of refractory EoE.