Eosinophilic Esophagitis: the newest esophageal inflammatory disease

Dan Atkins; Robert Kramer; Glenn T. Furuta; Kelley Capocelli; Mark Lovell


Nat Clin Pract Gastroenterol Hepatol. 2009;6(5):267-278. 

In This Article

Eosinophilic Esophagitis


EoE is a disease characterized by a variety of upper gastrointestinal symptoms that occur in association with dense esophageal eosinophilia ( Box 1 ); its symptoms and histopathology are unresponsive to pharmacological and surgical GERD treatments.[1] Alternative etiologies responsible for the eosinophilia must be ruled out before EoE can be diagnosed.[1,13] EoE has been reported in all age-groups and in all continents except Africa.[14,15] The prevalence of the disease is not certain, but is estimated to be 1-4 cases per 10,000 individuals.[16,17]

The natural history of the disease is as yet unknown;[18] two case reports suggest an association with Barrett esophagus, but this link has not been verified by other studies.[19,20,21] A number of names and acronyms have been used to describe this relatively new disease, including eosinophilic esophagitis, primary eosinophilic esophagitis, allergic eosinophilic esophagitis, idiopathic eosinophilic esophagitis and eosinophilic esophagitis (EoE or EE). The acronym EE has become deeply embedded in the allergy literature, but because of increasing confusion with erosive esophagitis (EE) within the gastroenterology community, the acronym EoE will be used in this Review.


Mechanisms of esophageal eosinophilic inflammation associated with EoE have been investigated in basic and translational studies. Mishra et al. conducted a series of studies using murine models of esophageal eosinophilia in which allergens or cytokines were used to initiate esophageal eosinophilia.[22,23,24,25]Aspergillus fumigatus, an ubiquitous airborne allergen, was used to sensitize mice, which developed epithelial eosinophilia that was isolated to the esophageal mucosa. Further studies using this model demonstrated that the eosinophilia and collagen deposition were interleukin 5 (IL-5)-dependent.[26] Translational studies have characterized the inflammatory profile of the mucosa revealing increased CD8+ lymphocytes, proinflammatory cytokines, the immunoglobulin E (IgE) receptor, mast cell infiltration, eosinophil degranulation and, in some cases, basophilia.[27,28,29,30,31,32,33,34,35,36,37]

In a landmark study, Blanchard et al. discovered an EoE transcriptome.[38] In this study, a genome-wide microarray analysis was performed on esophageal biopsies obtained from 13 children with EoE compared with five children with chronic esophagitis and six healthy controls. The most upregulated gene product was eotaxin-3 (also known as CCL26), a chemokine critical for eosinophil migration; a single-nucleotide polymorphism in this gene was associated with disease susceptibility. In a series of murine studies, mice deficient in the chemokine receptor CCR3 were shown to be protected from EoE. Other studies have shown increased expression of the gene encoding eotaxin-3 (CCL26) in the mucosa of animals with EoE.[39,40,41,42] Together, these findings support a role for eotaxin-3 (CCL26) in the pathogenesis of EoE.

The role of acid reflux in the pathogenesis of EoE is a matter of debate. While Sant'Anna et al. showed that pH monitoring of the distal esophagus revealed increased periods of alkalinization in children with EoE compared with that of healthy controls, Rosen et al. demonstrated that acid and nonacid reflux was not increased in children with EoE using pH-multichannel intraluminal impedance tracings.[43,44] The coming years will also bring new enlightenment as to the relationships between GERD and EoE.[45] There is probably clinical crossover between these two diseases in some patients that will likely be explained as the esophageal microenvironment becomes better defined.[4,9,46]

Association with Allergic Disease. Several lines of evidence support a role for allergic inflammation in the pathogenesis of EoE. The most obvious evidence for such involvement is the central role of the eosinophil. This cell is often considered synonymous with allergic disease because of its accumulation in sputum in asthma, in nasal secretions in allergic rhinitis and in the skin during flares of acute eczema.[47] The majority of patients with EoE have comorbid allergic disease such as food allergy, eczema, allergic rhinitis, or asthma accompanied by IgE-mediated sensitization to food and/or inhalant allergens.[1] An age-related inverse gradient for the presence of coexisiting allergic disease has previously been described in EoE; its frequency is highest in pediatric cohorts, where estimates reach 80%, and lowest in adult cohorts where estimates range from 40% to 70%.[1,48] Two studies have demonstrated a potential role for airborne allergens in EoE.[49,50] Histories of allergic disease in family members are noted in the majority of patients with EoE as exemplified by a report in which 73.5% of 103 patients with EoE had a positive family history of allergic disease.[16] The presence of other atopic diseases in patients with EoE affects peripheral eosinophil counts and total IgE levels making it difficult to ascribe changes in these parameters specifically to EoE.

As could be predicted by the intimate interaction between foods and the esophageal mucosa, food allergy often accompanies EoE and is considered contributory in a substantial proportion of patients.[51] Pediatric studies report coexistent food allergy in up to 73% of patients, whereas adult studies document food allergy in a more widely ranging proportion (25-82%) of patients.[49,52] Foods most often implicated in EoE-associated allergies include milk, egg, wheat, soy, peanuts, beans, rye and beef, although a variety of other foods may have a role in individual patients.[1] Allergic sensitization to more than one food is common in patients with EoE. When large panels of food allergy skin tests were applied to a cohort of 786 patients with EoE from aggregated case series data, the mean number of food allergies identified per patient varied from 2.7 ± 3.3 to 6 ± 4.2.[1] In addition to sensitivity to food allergens, sensitivity to seasonal pollen allergens is common in patients with EoE.[1] Isolated case reports have documented a correlation between pollen allergy skin test results and seasonal changes in EoE symptoms and esophageal eosinophil levels.[53,54]

Allergic reactions to foods are categorized as IgE-mediated, non-IgE-mediated or combined reactions according to the immune mechanism involved. In IgE-mediated food allergy, exposure of a genetically predisposed individual to an appropriate food results in the generation of allergen-specific IgE that occupies high affinity receptors for IgE on the surface of mast cells and basophils resulting in allergic sensitization.[55] Re-exposure of the individual to this food results in binding of adjacent cell surface allergen-specific IgE molecules, bridging of their high-affinity receptors, the explosive release of histamine, and the generation of newly formed mediators, some of which are chemotactic for eosinophils.[55] An important translational study demonstrated that EoE represents a TH2-type allergic response. In this report, the investigators demonstrated the presence of IgE-bearing mast cells within the epithelial cell layer of patients with EoE and identified potential allergens inducing the inflammatory response.[56] An IgE-mediated allergic reaction occurs within minutes to hours of allergen exposure.[55] By contrast, non-IgE-mediated allergic reactions typically orchestrated by TH2 lymphocytes are slower in onset, evolving over hours and days after food allergen exposure, resulting in eosinophil accumulation and symptoms localized to the gastrointestinal tract. This delay in symptom onset following food allergen exposure complicates accurate identification of offending foods in non-IgE mediated food allergy. EoE symptoms are often consistent with those seen in non-IgE mediated allergic reactions in that they are localized to the gastrointestinal tract and often delayed rather than immediate in onset.

EoE is considered to be a combined disorder in which both IgE-mediated and non-IgE-mediated immune mechanisms are involved. As a result, methods used to diagnose IgE-mediated food allergy, such as skin-prick testing and the measurement of serum food allergen-specific IgE levels are useful in identifying potential culprit foods, but will not detect those causing symptoms through a non-IgE mediated mechanism. Atopy patch testing (APT), where food allergen extracts are placed in constant contact with the skin under 12 mm aluminum chambers for 48 h then removed and the sites of exposure examined for a reaction at 72 h, has been proposed as a method for potentially identifying foods causing symptoms of EoE through a non-IgE mediated mechanism. Spergel and colleagues have reported substantial success as evidenced by biopsy-documented improvement in patients placed on elimination diets on the basis of information obtained through a combination of allergy skin-prick testing and APT.[48,57] The most common foods identified through patch testing by Spergel et al. in their cohort were milk, wheat, corn, beef, egg, potato, chicken, soy, barley, oat and rice.[58] In this cohort the combined negative predictive value of allergy skin-prick testing and APT for milk was considered unacceptably low, leading the authors to recommend consideration of milk elimination even when testing to milk was negative.[58] Despite the promising findings reported by Spergel and colleagues, some features have hindered wider use of these tests, including a lack of standardization of the procedure, and the time and expertise required to accurately perform such tests.[59]

Improvement of EoE following the use of food elimination diets emphasizes the significant causal role that food allergy has in most patients. Symptom resolution and normalization of esophageal biopsies has been achieved in more than 95% of children with EoE on elemental diets.[51] Difficulty in maintaining compliance to an elemental diet in adult patients, and the effect of such a diet on quality of life owing to the poor palatability and expense of these formulas has fostered the exploration of alternative dietary approaches. Elimination of the six foods most commonly involved in EoE (dairy products, egg, wheat, soy, peanuts, fish/shellfish) without the performance of allergy testing led to disease remission in approximately 74% of patients.[60] Food elimination diets based on the combined results of allergy skin-prick testing and APT are known to be effective for inducing remission in the majority of patients with EoE.[48,57,58] The marked improvement in symptoms noted upon removal of culprit foods from the diet underscores the effect that food-allergen exposure has in EoE.

A thorough history and assessment of allergic diseases is a vital part of the evaluation of a patient with suspected EoE because of the increased frequency of comorbid allergic disease, and the fact that allergens probably have a role in pathogenesis of this disease. Consultation with an allergist is indicated to aid in identifying, characterizing and treating comorbid allergic disease, in addition to identifying those allergens that are potential contributors to esophageal inflammation.[53] The allergist should aid in the development and monitoring of response to food elimination diets, as well as participating in the patient's long-term management.

Clinical Presentation

Significant variation exists in the clinical presentation of children and adults with EoE. Whether this variation is due to differences in the ability of these groups to enunciate discomfort or represents different disease stages is not yet certain. Regardless of age, if a patient develops GERD-like symptoms and is not responding to treatment, whether that be pharmacological or surgical, strong consideration should be given to a diagnosis of EoE.[1,61]

Children. Children with EoE present with a broad range of nonspecific symptoms that affect the upper gastrointestinal tract and respiratory tree.[62,63,64,65,66,67,68] In the largest pediatric study to date of 381 children, Liacouras et al. reported that the most common symptoms of EoE were divided into GERD-like symptoms and dysphagia. GERD-like symptoms predominated in this study (312 children) and included vomiting, regurgitation, abdominal pain, heartburn and water brash.[51] Of the remaining 69 children with dysphagia, 24 had radiological evidence of esophageal narrowing, but only one had an endoscopic stricture that required dilatation. These observations support the theory that symptoms related to EoE represent two different disease phenotypes.

Some patients with EoE experience acute symptoms due to intermittent esophageal spasm, probably related to smooth-muscle contraction.[69] Nurko and colleagues reported that of eight children with EoE and radiological evidence of Schatzki ring, none showed evidence of esophageal narrowing at the time of endoscopy.[70] Measures of esophageal contraction recorded by 24 h motility monitoring were reported to be disordered in 13 children with EoE compared with children with GERD and healthy controls.[71] By contrast, chronic symptoms of dysphagia might reflect longstanding esophageal remodeling that occurs secondary to chronic eosinophilic inflammation. This tenet is supported by the findings of isolated proximal and distal esophageal strictures, esophageal rings, long-segment narrowing of the esophagus, and esophageal fibrosis in patients with EoE.[72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92]

An important subgroup of children with EoE have feeding difficulties that include refusal to eat, chewing problems, choking on solids or liquids, or an inability to tolerate diet beyond certain textures or tastes.[93] Frequently these symptoms are overlooked in the context of a busy practice but they may represent one of the first clues to the diagnosis of EoE. Feeding difficulties may represent acute esophageal dysfunction with pain, or chronic learned behavior from longstanding disease. A family history of EoE, dysphagia or food impaction may be present in children with the disease; it is not unusual for new diagnoses within a family to be uncovered during the evaluation of a child.[94,95,96,97,98]

Adults. In contrast to children, most adults with EoE present in a stereotypical fashion with either food impaction or dysphagia.[1,27,99] Food impaction at initial presentation, and especially at recurrent presentations, should alert the clinician to a possible diagnosis of EoE.[27,80,83,84,100,101,102,103,104] Desai et al. found that EoE was the primary cause of food impaction in patients in a suburban private practice setting.[27] Of 31 adults presenting with acute food impaction, 17 showed clinicopathological features of EoE. In addition, a study from the Swiss Esophageal Esophagitis Databank reported that 35% of 251 adults with EoE presented with food impaction.[27,84,85,102,103,104,105,106,107] Of these patients, transmural perforation occurred in two patients who underwent rigid endoscopy and one patient experienced an esophageal rupture. In EoE, dysphagia can be longstanding, dating back to childhood, or acute in onset. In a study of 30 patients with EoE who were followed up for over 11 years, dysphagia was the predominant complaint and persisted and posed major problems in the patients' lifestyles.[83] As EoE has been increasingly recognized by clinicians, reports of acute and severe chest pain probably related to esophageal spasm have become associated with the disease (G. T. Furuta, personal communication).


Physical Examination. Physical examinations are important to rule out other causes for esophageal eosinophilia, such as IBD, and to identify co-morbid allergic diseases.[108] No pathognomonic findings on physical examination have been reported in EoE. White plaques commonly observed in the esophageal mucosa of patients with EoE have not been observed in the oropharynx.[109] Evidence of mucosal disruption is rarely observed.[110]

Identification of signs of allergic diseases, including allergic rhinitis, asthma and eczema, may be present and prompt further evaluation and treatment.

Laboratory Analysis. No single biomarker of eosinophilic inflammation provides support for or against the diagnosis of EoE.[111] If anemia, elevated erythrocyte sedimentation rate or serum levels of C-reactive protein, or markers of other systemic illness are identified, a search for an alternative diagnosis other than EoE should ensue. Peripheral eosinophilia is not a reliable sign for EoE as it is not always present.[1,112] When identified, it is difficult to differentiate whether peripheral eosinophilia occurs as a marker for EoE or other comorbid diseases.

Konnikoff et al. reported that the combination of peripheral eosinophilia with elevated serum levels of eotaxin-3 (CCL26) and eosinophil-derived neurotoxin correlated with increased esophageal eosinophil density and thus may provide a biomarker panel for monitoring EoE.[113] In their study of 47 children, these three biomarkers correlated with esophageal eosinophil density. Further validation of these studies will be important, because such markers could potentially eliminate the necessity of performing repeated endoscopies and biopsies in patients with suspected EoE.

Radiological Assessment. The use of contrast radiography serves two purposes in the evaluation of a patient with suspected EoE. Upper gastrointestinal radiography is useful to eliminate other causes of vomiting, such as anatomical malformations. In addition, barium studies can provide important information regarding the presence of esophageal stricture or long-segment narrowing.[79,80,114] Some of the first reports of EoE arose from the radiological literature that described single and multiple esophageal rings.[73,74,115] Communication between the gastroenterologist and the radiologist is important as esophageal strictures may be proximal and long-segment narrowing can be missed on radiography, as it can encompass much of the length of the esophagus. Schatzki rings are also occasionally visualized on radiography but may be absent at endoscopy showing that this finding represents a transient esophageal contraction.[71] Another study found Schatzki rings not to be associated with EoE.[116]

Endoscopic Features. A number of endoscopic features associated with EoE have been reported.[69,110] Loss of vascular pattern and longitudinal furrowing of the esophagus (Figure 1) probably represent epithelial edema.[117] White exudates (Figure 2) are the gross manifestation of eosinophilic pus and correlate with the histological finding of eosinophilic abscesses.[110,118,119,120] Concentric rings (Figure 3) may be transient, indicating esophageal contraction, or fixed, indicating tissue remodeling with stricture formation.[75,82,87,90,121,122,123,124,125] When esophageal pinch biopsies are obtained, the mucosa may feel rubbery at inflamed sites owing to fibrosis, thus making tissue samples more difficult to obtain compared with noninflamed sites.[110] Long-caliber esophageal narrowing or small-caliber esophagus is observed infrequently in EoE and may only be recognized after removal of the endoscope when longitudinal shearing and/or splitting of the mucosa or crepe paper esophagus is evident (Figure 4).[77,80,114] Perforation is unusual but has been reported.[88,91,126,127,128,129] Esophageal erosions and friability are rarely seen.[83]

Figure 1.

Longitudinal furrowing or ridging, loss of vascular pattern with a scant amount of white papular exudates in the midesophagus of a patient with eosinophilic esophagitis.

Figure 2.

Extensive, confluent patches of white exudate in the distal esophagus of a patient with eosinophilic esophagitis.

Figure 3.

Concentric rings with a white papular exudate in the proximal esophagus of a patients with eosinophilic esophagitis.

Figure 4.

Longitudinal shearing and/or splitting, termed crepe-paper esophagus in an adolescent patient with eosinophilic esophagitis.

Histological Features. EoE is characterized histologically by dense epithelial eosinophilia (Figures 5 and 6).[130] The limited size of endoscopic mucosal biopsies (3 mm in diameter or less) prevent characterization as to the depth of this inflammation, but ultrasonographic and limited surgical reports suggest deep involvement.[131,132,133] One study demonstrates the importance of obtaining multiple biopsies (five) to ensure accurate diagnosis.[134]

Figure 5.

Esophageal mucosa of a patient with eosinophilic esophagitis showing large numbers of eosinophils scattered throughout the epithelium in a diffuse pattern with microabscess formation on the superficial layer, and evidence of degranulation throughout (×40 magnification).

Figure 6.

Esophageal mucosa of a patient with eosinophilic esophagitis showing large numbers of eosinophils with evidence of superficial layering, basal zone hyperplasia and rete peg elongation (×40 magnification).

Diagnosis of eosinophilia is focused primarily on the peak number of eosinophils per high-power field (HPF) present in the epithelial surface; the most sensitive and specific threshold number of eosinophils (peak or mean) in the mucosal biopsy that correctly identifies patients with EoE continues to be a matter of discussion.[1,130] This issue is complicated by many factors, including variable size of individual HPFs examined, differences in the definition of the cellular morphology that constitutes an intact eosinophil, and the number of HPFs that should be assessed.[135] To address this issue, the North American Society of Pediatric Gastroenterology and Nutrition initiated the development of consensus recommendations to identify a diagnostic algorithm for EoE. Clinical expertise and data from the literature were reviewed at the First International Gastrointestinal Eosinophil Research Symposium (FIGERS) and the results of this symposium and review were recently published.[1,136] As a part of this process, ≥15 eosinophils per single HPF was chosen as the threshold number for the diagnosis of EoE with one important caveat: as discussed earlier, alternative etiologies, in particular GERD, must be ruled out as a cause for this inflammation.[11,137] In the future, additional pathological criteria and techniques will be identified to increase the specificity and sensitivity of diagnosis of EoE, and will prompt revisions to these recommendations.[138]

Epithelial surfaces affected by EoE may also demonstrate features such as superficial layering of eosinophils and microabscesses on the superficial layer (Figure 5);[139] these findings are the histological representations of the endoscopic feature termed white exudates or eosinophilic pus (Figure 2).[109,118]

Eosinophil degranulation occurs significantly more often in adult patients with EoE than in those with GERD.[27,140,141,142] Degranulation can be extensive and may represent the only identifiable footprint or ghost of eosinophils that resided in a microscopic field. In this regard, eosinophilic inflammation and degranulation are often patchy in EoE and offer new opportunities to add to the diagnostic algorithm. For example, Prothroe et al. reported the use of a novel eosinophil peroxidase antibody scoring system that used patterns of degranulation and eosinophil number as rubrics for analysis of mucosal biopsies from adults with EoE.[143] In this retrospective study, they were able to reliably discern EoE from GERD based solely on these histological features.

Other commonly identified histological features observed in children and adults with EoE include basal zone hyperplasia, rete peg elongation, intracellular edema, and dilated intercellular spaces (Figure 6).[130] In combination with other features of inflammation, these findings may also help to segregate EoE from other sources of esophageal inflammation. Finally, evidence shows that lamina propria fibrosis is present in mucosal biopsies obtained from children with EoE. Studies in this regard are limited by the fact that endoscopically obtained biopsies rarely contain enough lamina propria for examination and a validated esophageal fibrosis scoring system is lacking. Aceves et al. reported increased fibrosis in trichrome-stained samples from seven children with EoE compared with samples from children who had GERD.[144] With regard to the logical pathology and physiology of this process, the investigators identified increased expression of transforming growth factor β and increased evidence of its cognate signaling molecule phospho-SMAD in patients with fibrosis. Chehade et al. revealed similar findings upon comparing 21 children with EoE with six with GERD, and 17 healthy controls.[29] Interestingly, in both of these studies, some healthy controls showed evidence of fibrosis.


Clinicians use two very different end points for the treatment of EoE: symptom relief only (clinical remission) or symptomatic relief and histological remission (clinical and pathological remission).[145] The latter approach is complicated by the fact that the definition of histological remission varies between clinicians. While the esophagus typically does not contain resident eosinophils, no studies have determined a threshold number of eosinophils that may be tolerated, if any, in the esophageal epithelium of an EoE patient following treatment.

The rationale for seeking only clinical remission is that as limited data are available regarding the natural history of the disease,[146] the cost and morbidity associated with post-treatment endoscopy and the overall psychological effect of treatments, it is difficult to justify chronic treatment and repeated endoscopies.[145] However, many physicians seek to achieve clinical and pathological remission. A growing body of clinical experience has shown that patients may still have significant esophageal inflammation despite experiencing few or no symptoms.[147] In these circumstances, it is impossible to predict which patients will develop complications, such as esophageal remodeling with isolated or long-segment narrowing, and when they will develop complications. Thus, many clinicians feel that the threat of ongoing inflammation and subsequent esophageal remodeling consequences outweigh the risks of repeat endoscopic evaluation.[145]

As the exact incidence and breadth of esophageal complications in patients with EoE have not been determined, and the effects of chronic treatments and endoscopy have not yet been measured, each clinician will need to make a decision as to the approach or combination of approaches that should be used for the treatment of EoE.[148] Future studies will provide more guidance in this area of controversy.

Although the end point of treatment is a matter of debate, at least two safe and effective treatments have been identified ( Box 2 ). The reader is referred to a number of reviews on this topic for further details regarding the specifics of each treatment.[149,150,151] Regardless, evidence suggests that EoE is a chronic disease and without ongoing treatment, symptoms and inflammation will return or persist. No pharmacological maintenance treatment has yet been identified.[1]

Pharmacological Therapy. Two lines of evidence support the use of systemic and topical steroids for the treatment of children and adults with EoE. First, corticosteroids significantly reduce synthesis of eosinophil growth factors (IL-5, granulocyte macrophage colony-stimulating factor) and chemoattractants (eotaxin-3 [CCL26]) and induce eosinophil apoptosis.[152] Second, longstanding clinical experience has shown that eosinophilic inflammation in allergic and inflammatory responses is responsive to treatment with steroids.[153,154]

In this regard, Liacouras et al. reported the first successful use of corticosteroids (1-2 mg/kg daily for 2-4 weeks) in achieving clinical and pathological remission in 21 children with EoE.[153] When treatment was completed and steroids withdrawn, disease returned. Faubion et al. first reported on the use of topical corticosteroids for EoE in 1998.[155] In an effort to reduce children's systemic steroid exposure, this group used a metered dose inhaler to administer a topical preparation of budesonide or fluticasone to children with EoE. Four children received sprays of the metered dose inhaler in the mouth (without inhaling and without a spacer) 2-4 times daily for 4-8 weeks, and experienced remission of symptoms. Teitelbaum et al. demonstrated clinical and pathological success as well as a reduction of CD8+ lymphocytes in 11 children with EoE after 8 weeks of fluticasone treatment.[31] Konnikoff et al. performed the only double blind, placebo-controlled trial of treatment for EoE, in which 36 children were randomly allocated to receive fluticasone or placebo over a 3-month period.[156] Remission was achieved in 55% of fluticasone-treated children and 9% of placebo-treated children; the greatest effect of this treatment was shown in the proximal esophagus. Other studies have shown a similar response in adult patients with EoE treated with steroids, and that topical steroids can downregulate cytokines implicated in the pathogenesis of EoE, including eotaxin-3 (CCL26) and IL-5.[157,158,159] A 2007 study revealed the beneficial effect of this steroid treatment on the specific EoE inflammatory transcriptome.[160] In an effort to improve the delivery of steroids, Aceves et al. mixed budesonide with sucralose to make a viscous slurry.[161,162] Ingestion of this product led to clinical and pathological remission in 16 of the 20 children who were studied.

Side effects of steroid treatment include well-known complications, such as bone demineralization, cataracts, growth inhibition, diabetes, acne, and mood disturbances.[1,163] Oral and esophageal candida and herpes infections have been reported with the use of topical steroids in a limited number of patients with EoE.[31,164]

Therapeutic agents that are currently being investigated include anti-IL-5 monoclonal antibody.[165] As IL-5 is critical to the growth and development of eosinophils, IL-5 is increased in the esophageal mucosa of patients with EoE, and murine studies show the IL-5 dependency of EoE, anti-IL-5 therapy may benefit patients. One study has shown anti-tumor necrosis factor therapy to be ineffective for the treatment of EoE.[1,166]

Dietary Therapy. At least three lines of evidence support a role for nutritional treatment in EoE. First, animal models have demonstrated that eosinophilic gastrointestinal inflammation can be induced by food sensitization.[167] Second, a number of studies and clinical experiences link food allergies with allergic diseases, such as eczema, and eosinophilic gastrointestinal diseases, such as EoE.[168] Finally, use of an amino-acid-based diet formula and in some cases, dietary exclusions, leads to clinical and pathological remission of EoE.[51,65,169]

In 1995, Kelly et al. published the first study demonstrating the successful use of an elemental dietary formula for the treatment of EoE in children.[64] After receiving at least 8 weeks of an amino-acid-based dietary formula, 10 children achieved clinicopathological remission, and experienced a recurrence of symptoms upon the reintroduction of foods. Since then, other studies and clinical experience have shown similar results with the use of elemental dietary formulas. In an effort to provide a targeted approach to dietary therapy and to liberalize the diet of affected patients, removal of single or multiple foods has been attempted. Spergel et al. used skin-prick and skin-patch allergy testing to identify allergenic foods in 26 children with EoE.[48] In this study, 18 children had symptom resolution after food removal. Finally, to eliminate allergy testing all together, Kagalwalla et al. removed the six most common food allergens from the diets of 35 children with EoE and demonstrated that after 6 weeks of this diet, over 75% achieved clinical and pathological remission.[60] Similar results were reported by Gonsalves et al. in 2007.[170]

Dietary management can influence patients' lives in a number of ways. When eliminating any food product from the diet of any patient, especially that of a child, a dietician must be involved to insure that a properly balanced nutritional plan is provided.[171,172] The cost of special diets, especially those composed of elemental formulas, can be significant and reimbursements are often not provided. Diets can also affect psychosocial aspects (dating, birthday parties, school and professional activities) of patients' lives. These factors all need to be considered and discussed with families before dietary therapy is initiated to maintain a balance between sometimes difficult treatment plans and long-term treatment outcomes.[173]

Esophageal Narrowing. An unknown proportion of children and adults with EoE develop esophageal strictures or small-caliber esophagus.[174] It is important to remember, as mentioned earlier, that the radiographic appearance of a Schatzki ring or the endoscopic view of concentric esophageal narrowings may actually represent a transient esophageal contraction.[71] These findings may not require mechanical dilatation, but rather pharmacological or nutritional management.

If an esophageal stricture or long-segment narrowing of the esophagus is identified, two approaches, mechanical dilatation or pharmacological treatment with corticosteroids, have been used.[127,175,176,177] When strictures are untreatable with pharmacological therapy, dilatation with a bougienage may be necessary. It is not uncommon for the endoscope itself to accomplish dilatation merely with its initial passage. The advantages of a mechanical approach are immediate relief of the partial obstruction. In contrast to dilatation of a peptic stricture, esophageal dilatation should be performed gently and may require multiple sessions to achieve satisfactory results. The disadvantage of this therapy is the potential for esophageal perforation and pain severe enough to, in some cases, require hospitalization. Systemic steroids may be administered as first-line therapy in some patients with strictures, and mechanical dilatation performed subsequently if no response is observed. The advantage of this approach is that the steroid may reduce the acute inflammatory response. The efficacy of injecting esophageal strictures with corticosteroids has not been reported.