Clinical manifestations and diagnosis of eosinophilic esophagitis (EoE)

INTRODUCTION — Esophageal eosinophils were once considered to be a hallmark of gastroesophageal reflux disease [1]. However, it has become apparent that the esophagus, which is normally devoid of eosinophils, is an immunologically active organ that is capable of recruiting eosinophils in response to a variety of stimuli [2,3].

When the gastrointestinal eosinophilia is limited to the esophagus, is accompanied by characteristic symptoms, and other causes of esophageal eosinophilia have been ruled out, it is termed eosinophilic esophagitis. A panel of experts defined eosinophilic esophagitis as "a chronic, immune/antigen-mediated, esophageal disease characterized clinically by symptoms related to esophageal dysfunction and histologically by eosinophil-predominant inflammation" [4]. In the past, eosinophilic esophagitis was abbreviated as "EE," but because of confusion with erosive esophagitis, many prefer the abbreviation "EoE."

This topic will review the clinical manifestations and diagnosis of eosinophilic esophagitis in adults and children. The genetics, immunopathogenesis, and management of eosinophilic esophagitis; eosinophilic gastroenteritis; and other diseases with eosinophilic involvement of specific organs are discussed separately. (See "Eosinophilic esophagitis (EoE): Genetics and immunopathogenesis" and "Treatment of eosinophilic esophagitis (EoE)" and "Eosinophilic gastrointestinal diseases" and "Eosinophil biology and causes of eosinophilia", section on 'Disorders with eosinophilic involvement of specific organs'.)

EPIDEMIOLOGY

Eosinophilic esophagitis has been reported in several countries in North and South America, Europe, Asia, and Australia, but there have been no published reports from countries in Africa. A study of a pathology database from the United States found that the disease has been detected in most states that reported data [5]. The results of one survey suggested there may be regional variation, with a higher prevalence in northeastern states and lower prevalence in western states [6]. The diagnosis also appeared to be more common in urban as opposed to rural settings. In contrast, another study that evaluated a national pathology database, revealed the opposite patterns with an increased odds of EoE in the west and in rural settings [7]. Prevalence within the United States may also differ between climate zones with a higher prevalence in cold and arid zones as compared with the tropical zones [8]. Seasonal exacerbations of symptoms that have been described suggested a possible role of aeroallergens [9-11].

The first cases of probable eosinophilic esophagitis were reported in the late 1960s to 1970s [12-14]. Early reports (mostly from the 1990s) described patients with multiple esophageal rings, which were attributed to gastroesophageal reflux disease or a congenital origin [12,15-20]. The assumed association with gastroesophageal reflux disease (GERD) was based upon the observation that biopsies from patients with a ringed esophagus had basal zone hyperplasia, papillary lengthening, and intraepithelial eosinophils, findings that are seen in patients with documented reflux disease. However, careful review of these reports has raised questions about the association with GERD, since many of the patients did not respond to antisecretory therapy or have objective evidence of reflux on a 24-hour pH study [15,21]. In addition, esophageal dilation in some of these patients was associated with deep mucosal tears and esophageal perforation, complications also seen in patients with eosinophilic esophagitis who undergo dilation [16-18,20,22-26].

The incidence of eosinophilic esophagitis appears to be increasing [27-31]. Some of the increase may be due to increased recognition of the disorder, but increased recognition is unlikely to fully account for the rising incidence. One reason that increased recognition is unlikely to be the sole explanation is that gastrointestinal barium radiography has been practiced for many decades and it is likely that the characteristic ringed pattern in the esophagus would have been described earlier. While studies evaluating the prevalence of eosinophilic esophagitis in biopsy samples from patients undergoing endoscopy have not detected an increase since at least the late 1980s [32,33], population-based studies have supported an increasing incidence:

●A population-based study evaluated the incidence of eosinophilic esophagitis in Olmsted County Minnesota over 30 years [34]. The incidence increased significantly during the last three of the five-year intervals examined (from 0.35 per 100,000 population between 1991 and 1995 to 9.45 per 100,000 between 2001 and 2005). The prevalence was estimated to be 55 per 100,000 in 2006.

●Two population-based studies from Switzerland concluded that the incidence of eosinophilic esophagitis was increasing, and in one study, the incidence was approaching that of inflammatory bowel disease [28,35].

●A population-based study in children focused on 103 patients who had been identified from a single institution's pathology database in Hamilton County, Ohio [27]. The overall incidence per 10,000 population was estimated to be 1.28 in 2003, an increase from 0.91 in 2000. The authors noted that the most recent incidence rates exceed those of inflammatory bowel disease in children.

The majority of affected adults have been males in their 20s or 30s, although later presentations have been described [22,36-38]. In a series of 31 adults (24 males and 7 females), the mean age at diagnosis was 34 years (range 14 to 77 years) [36]. Symptoms (predominantly dysphagia) had been present for an average of 4.5 years prior to diagnosis. The male predominance may be related to variations in a gene located on the X-chromosome that have been associated with eosinophilic esophagitis. (See "Eosinophilic esophagitis (EoE): Genetics and immunopathogenesis", section on 'Overview of pathogenesis'.)

Among children, the disease is also more common in males (71 percent in the series described above) [27]. In another pediatric population-based study, significantly more of those with eosinophilic esophagitis were White children (84 percent compared with 73 percent of the surrounding community as a whole) [39]. Pediatric patients with eosinophilic esophagitis were also more likely to be male compared with the general population (76 versus 48 percent).

There are limited data on risk factors for eosinophilic esophagitis [40,41]. In a case-control study that included 115 patients with eosinophilic esophagitis and 225 controls, patients with eosinophilic esophagitis were significantly less likely to have ever smoked cigarettes or actively use NSAIDs (odds ratio [OR] 0.36 and 0.47, respectively) [41]. However, there was no significant difference in rates of smoking or NSAID exposure between cases with or without fibrostenosing disease or among patients with a post-treatment histologic response. Other potential risk factors include antibiotic exposure, acid suppressive medication use, and admission to neonatal intensive care unit during childhood, whereas exposure to breastmilk may be protective [42,43]. Exposure to Helicobacter pylori may also be associated with a reduced risk of eosinophilic esophagitis [44]. It is unclear if this apparent protective effect is due to H. pylori-induced immunomodulation [45].

Subtypes — Three pathogenically distinct subtypes of eosinophilic esophagitis have been described. In a cross-sectional study in which an eosinophilic esophagitis diagnostic panel of 96 molecular targets, in tandem with endoscopic and histological assessment, were used to evaluate adult and pediatric patients with active eosinophilic esophagitis, three distinct endotypes with unique features were identified [46]:

●EoEe1 – A mild subtype with normal-appearing esophagus, and mild histological, endoscopic, and molecular changes.

●EoEe2 – An inflammatory endotype with highest expression of inflammatory cytokines and steroid-responding genes and a steroid refractory phenotype.

●EoEe3 – A fibrostenotic endotype associated with a narrow-caliber esophagus, and characterized by the highest degree of endoscopic and histological severity and the lowest expression of epithelial differentiation genes.

Additional studies are needed to determine if endophenotyping patients with eosinophilic esophagitis can effectively guide management and improve patient outcomes. (See 'Laboratory tests' below.)

CLINICAL MANIFESTATIONS — The clinical manifestations of eosinophilic esophagitis vary with age [47]. Adults and teenagers frequently present with dysphagia and food impactions, whereas in younger children, symptoms often include feeding difficulties, gastroesophageal reflux symptoms, and abdominal pain.

Clinical manifestations in adults — Common clinical manifestations seen in adults include [5,22,26,34,48-75]:

●Dysphagia

●Food impaction

●Chest pain that is often centrally located and may not respond to antacids

●Gastroesophageal reflux disease-like symptoms/refractory heartburn

●Upper abdominal pain

Dysphagia to solid foods is the most common symptom [5,48-52]. Up to 15 percent of patients being evaluated for dysphagia with endoscopy are found to have eosinophilic esophagitis [50,52,76]. A history of food impaction is present in up to 54 percent of patients [26,53,54] and esophageal strictures have been noted in up to 31 percent of patients [34,55-60]. In one community-based series, 17 of 31 patients presenting with food impaction over a three-year period were diagnosed with eosinophilic esophagitis [53]. Psychosocial factors such as anxiety and hypervigilance may contribute to symptom reporting, particularly after food impaction [75].

Esophageal dysmotility may also be observed, suggesting possible eosinophil involvement of the muscular layers of the esophagus [61-63,77-79]. High-resolution endoscopic ultrasonography in affected children and adults has shown expansion of the esophageal wall and all individual tissue layers [64-66]. In a study including 109 patients (mean age 37 years) with a recent diagnosis of eosinophilic esophagitis who underwent high-resolution esophageal manometry, 17 patients (15 percent) had an obstructive motor disorder (eg, achalasia, esophagogastric junction outlet obstruction [EGJOO]) [79].

Eosinophilic esophagitis has been noted in 1 to 4 percent of patients with refractory reflux in prospective studies [51,80]. However, cost-effectiveness models suggest that evaluating for eosinophilic esophagitis in patients with refractory reflux is only cost-effective when the prevalence of eosinophilic esophagitis is greater than 8 percent [81].

Finally, spontaneous esophageal perforation (Boerhaave syndrome), esophageal perforation following endoscopy, and mucosal tears associated with endoscopy have been reported [22,25,26,67-69]. (See "Boerhaave syndrome: Effort rupture of the esophagus" and "Complications of endoscopic esophageal stricture dilation", section on 'Risk factors'.)

Clinical manifestations in children — Symptoms in children vary depending in part upon their age [27,82-87]. In one series, the most common presenting symptoms included [27]:

●Feeding dysfunction (median age 2.0 years)

●Vomiting (median age 8.1 years)

●Abdominal pain (median age 12.0 years)

●Dysphagia (median age 13.4 years)

●Food impaction (median age 16.8 years)

The possibility of disease progression was supported in a case-control study that suggested an increased rate of dysphagia (49 versus 6 percent) and food impaction (40 versus 3 percent) in children with esophageal eosinophilia who had been followed for an average of 15 years [88].

Feeding dysfunction continues to be defined, but is an increasingly recognized presentation of eosinophilic esophagitis [89]. It includes failure to develop normal eating patterns (eg, not advancing past liquids or soft foods) and adopting coping strategies (eg, refusing to eat solids after previously eating them, eating slowly, chewing excessively, drinking excessive liquids with meals).

Similar findings were described in a single center report involving a total of 381 children. The mean age was nine years and 66 percent were male. Patients most commonly presented with symptoms suggestive of gastroesophageal reflux (85 percent) or dysphagia (18 percent) [83]. The esophageal mucosa was abnormal by endoscopy in 68 percent, whereas in 32 percent it appeared normal despite severe histologic eosinophilia.

ASSOCIATIONS WITH OTHER DISORDERS — There is a strong association of eosinophilic esophagitis with allergic conditions such as food allergies, environmental allergies, asthma, and atopic dermatitis. It has been estimated that 28 to 86 percent of adults and 42 to 93 percent of children with eosinophilic esophagitis have another allergic disease [27,34,90-97]:

●In one series, 10 of 13 patients (77 percent) had a history of an allergic disorder defined as asthma, allergic rhinitis, urticaria, hay fever, atopic dermatitis, food allergy, or medicine allergy, and/or positive radioallergosorbent test (RAST) or positive allergic skin tests [92]. Twelve of 13 patients (92 percent) had an absolute peripheral eosinophilia, and 9 of 12 patients (75 percent) had concurrent eosinophilic gastroenteritis.

●In another report, 13 of 19 children (68 percent) had positive skin or RAST testing to a median of seven foods [90].

●In a third series, 18 of 23 adults (78 percent) had an atopic diathesis (most commonly allergic rhinitis) [91]. Most patients were sensitized to several environmental allergens.

An association with celiac disease has been reported in multiple studies, and in one series a response to a gluten-free diet was also noted [98,99]. In addition, an association with inflammatory bowel disease, chronic rhinosinusitis, connective tissue disorders, caustic injury, antibiotic exposure in infancy, herpes simplex virus esophagitis, and a Schatzki ring have also been described, although the strength of the associations is unclear [100-107].

DIAGNOSIS

Clinical suspicion and evaluation — The diagnosis of eosinophilic esophagitis is based upon symptoms, endoscopic appearance, and histological findings. Eosinophilic esophagitis should be suspected in patients with chronic symptoms of esophageal dysfunction (eg, dysphagia, food impaction, food refusal, abdominal pain, heartburn, regurgitation, chest pain, odynophagia). A history of atopic comorbidities (eg, asthma, atopic dermatitis, allergic rhinitis, or immediate food-type allergies) and a family history of eosinophilic esophagitis or dysphagia should raise the index of suspicion. A history of esophageal perforation or severe pain after dilation of a stricture should also raise suspicion of this disorder.

The diagnosis is established by upper endoscopy with esophageal biopsies and an evaluation to exclude other disorders that can cause esophageal eosinophilia (table 1). Radiographic and laboratory findings may support the diagnosis and help establish baseline esophageal luminal integrity, but are not required to establish the diagnosis. (See 'Differential diagnosis' below.)

Since the symptoms of eosinophilic esophagitis are not specific, the diagnosis may be missed. In one retrospective study that included 200 patients with symptomatic eosinophilic esophagitis, the median delay in diagnosis was six years (interquartile range 2 to 12 years) [108]. Increasing duration of delay in the diagnosis was associated with an increase in the prevalence of fibrotic features of eosinophilic esophagitis on biopsy and esophageal strictures.

Diagnostic criteria — The diagnosis of eosinophilic esophagitis requires all of the following [109]:

●Symptoms related to esophageal dysfunction.

●Eosinophil-predominant inflammation on esophageal biopsy, characteristically consisting of a peak value of ≥15 eosinophils per high power field (HPF) (or 60 eosinophils per mm²).

●Exclusion of other causes that may be responsible for or contributing to symptoms and esophageal eosinophilia (table 2).

Diagnostic criteria for eosinophilic esophagitis have evolved since eosinophilic esophagitis was first conceptually defined. In contrast to prior guidelines, persistence of mucosal eosinophilia in the esophagus after two months of treatment with a high-dose proton pump inhibitor (PPI) (eg, twice daily) is no longer a diagnostic criterion for eosinophilic esophagitis [4,109,110]. The rationale for exclusion of a PPI trial is that patients with clinical and histologic features compatible with eosinophilic esophagitis but who respond histologically to a PPI, (PPI-responsive esophageal eosinophilia [PPI-REE]), do not appear to be distinct from those with eosinophilic esophagitis [110-119]. Furthermore, patients with an initial response to PPIs may subsequently develop recurrent symptoms and eosinophilia consistent with eosinophilic esophagitis [120]. It is also recognized that GERD and eosinophilic esophagitis frequently co-exist and PPIs can improve esophageal eosinophilia by mechanisms independent of their effect on gastric acid secretion [121]. PPI-responsive esophageal eosinophilia is therefore considered a subset of eosinophilic esophagitis rather than a distinct disease [109,122].

The requirement of more than 15 eosinophils per HPF as a cutoff has not been extensively validated. One study supporting this cutoff confirmed endoscopic healing and symptomatic improvement when eosinophil counts decreased below this threshold [123]. Ongoing studies are helping to further clarify whether different cutoffs may be better at discriminating those with eosinophilic esophagitis versus other disorders. It is possible, for example, that patients with persistent esophageal eosinophilia after treatment with a PPI but who do not reach the 15 eosinophils per HPF threshold may still respond to treatment for eosinophilic esophagitis.

Endoscopy — A variety of morphologic features in the esophagus have been described in patients with eosinophilic esophagitis and a classification and grading scheme for these findings has been proposed [23,36,92,124-133]. A scoring system that relies on the assessment of Exudate, Rings, Edema, Furrows and Strictures has been developed and validated in children and adults and may be useful in clinical trials [133-137]. However, the assessment of endoscopic severity varies between endoscopists [138]. In addition, the endoscopic appearance alone has limited utility in the diagnosis of eosinophilic esophagitis. A 2012 meta-analysis that compared 4678 patients with eosinophilic esophagitis and 2742 controls estimated the frequency of the following endoscopic features [139]:

●Stacked circular rings ("feline" esophagus) (picture 1): 44 percent

●Strictures (particularly proximal strictures) (image 1): 21 percent

●Attenuation of the subepithelial vascular pattern: 41 percent

●Linear furrows (picture 2): 48 percent

●Whitish papules (representing eosinophil microabscesses) (picture 1): 27 percent

●Small caliber esophagus: 9 percent

Individual endoscopic features suggestive of eosinophilic esophagitis had low sensitivity ranging from 15 to 48 percent but high specificity ranging from 90 to 95 percent. Positive and negative predictive values ranged from 51 to 73 percent and 74 to 83 percent, respectively. Given the low sensitivity of endoscopic findings for eosinophilic esophagitis and variable positive predictive value, histology remains important in making a diagnosis of eosinophilic esophagitis, regardless of the endoscopic appearance. (See 'Histology' below.)

Complications associated with endoscopy in patients with eosinophilic esophagitis (even in the absence of esophageal dilation) include esophageal perforation and mucosal tears [25,67-69].

Histology — Esophageal biopsies from patients with eosinophilic esophagitis show an increased number of eosinophils. The vast majority of patients have at least 15 eosinophils per high power field (peak value) in at least one biopsy specimen [109]. Esophageal eosinophilia in the absence of clinical features is not sufficient to make a diagnosis of eosinophilic esophagitis.

During endoscopy, biopsies should be obtained from the distal esophagus as well as either the mid or proximal esophagus [140]. The sensitivity of biopsies for diagnosing eosinophilic esophagitis depends upon the number of biopsies obtained:

●In a report of 66 adults, the sensitivity was 100 percent after obtaining five biopsies compared with 55 percent with one biopsy [55].

●A second study found that the sensitivity for two, three, and six biopsies was 84, 97, and 100 percent, respectively [141].

●A third series evaluated biopsies obtained from the mid and distal esophagus in 102 consecutive patients with eosinophilic esophagitis [142]. The probability of one, four, five, and six biopsy fragments containing >15 eosinophils per high power field was 0.63, 0.98, 0.99, and >0.99, respectively.

We suggest that two to four biopsies be obtained from the distal esophagus, as well as another two to four from the mid or proximal esophagus.

Biopsy specimens should be fixed in formalin or paraformaldehyde rather than Bouin's fixative since formalin is more effective at preserving the integrity of eosinophils. Immunohistochemical studies have demonstrated that the number of eosinophils and amount of degranulation are underestimated by standard (hematoxylin and eosin) staining, although the clinical relevance for making a diagnosis is unclear [143].

As noted above, a threshold of 15 eosinophils per high power field is generally required for the diagnosis (picture 3) [144]. Approaches to improved diagnostic accuracy (and response to treatment) continue to be evaluated. One approach involves a scoring system (the EoE Histology Scoring system [EoE HSS]) that evaluates certain histologic features such basal zone hyperplasia [145]. The EoE HSS can discriminate between treated and untreated patients and provides a comprehensive assessment of esophageal mucosa.

Other histologic findings suggestive of eosinophilic esophagitis include [23,111,144,146-154]:

●Eosinophil microabscesses

●Superficial layering of eosinophils

●Sheets of eosinophils

●Extracellular eosinophil granules

●Subepithelial and lamina propria fibrosis and inflammation

●Basal cell hyperplasia

●Papillary lengthening

●Increased numbers of mast cells, B cells, and IgE-bearing cells

The association between histologic findings and symptoms is incompletely understood. Some studies suggest that histologic findings correlate with symptom severity [5,112,155,156], but discordant data have also been published [50,57,82,157]. One of the most detailed studies found that symptoms were only modestly predictive of histologic or endoscopic findings [158].

Biopsies of the gastric antrum and duodenum should also be obtained in patients with symptoms suggestive of eosinophilic gastroenteritis (eg, abdominal pain, nausea, vomiting, diarrhea, weight loss, ascites), visible mucosal abnormalities, or when there is a high index of suspicion [109]. Establishing the diagnosis of eosinophilic gastroenteritis rather than/in addition to eosinophilic esophagitis is important since it will influence treatment [110]. More controversial is the presence of incidental gastric eosinophils in patients without compatible symptoms. At least one report found that such patients had a similar clinical presentation and response to treatment as patients with eosinophilic esophagitis who did not have gastric eosinophilia [159]. (See "Eosinophilic gastrointestinal diseases".)

Radiology — Barium studies are not sensitive for diagnosing eosinophilic esophagitis, but can help characterize anatomic abnormalities and provide information on the length and diameter of strictures [160,161]. Findings described in patients with eosinophilic esophagitis undergoing barium studies include strictures and a ringed esophagus (picture 4) [132]. In addition, other causes for symptoms can be ruled out (eg, malrotation as a cause for vomiting). In addition, barium studies can help to assess luminal narrowing that is not evident at endoscopy [162,163]. They can be especially helpful in children for especially when combined with the use of a barium-coated pill [163].

Laboratory tests — Approximately 50 to 60 percent of patients with eosinophilic esophagitis will have elevated serum IgE levels (>114,000 units/L) [91,96]. Peripheral eosinophilia is seen in 40 to 50 percent of patients but is generally mild [56,84,164,165]. In some reports, the peripheral blood eosinophil level decreased with topical glucocorticoid therapy [166].

Genetic and molecular markers of disease activity continue to be studied [167]. A 96-gene EoE diagnostic panel, for example, has been developed based upon analysis of esophageal biopsies [168]. This diagnostic panel, which appears to be able to differentiate EoE from control individuals, including those with GERD, may also be able to differentiate patients with active and inactive disease and identify glucocorticoid exposure. Genetic testing for the EoE molecular transcriptome is commercially available but its role in clinical management is still evolving [46]. (See 'Subtypes' above.) Measurement of eosinophil progenitor cells correlate with histopathology and thus may provide a method of monitoring disease activity [169].

Evaluation for allergies — Because of the strong association of eosinophilic esophagitis with allergies, we suggest that patients with eosinophilic esophagitis undergo evaluation by an allergist. Children are often treated with dietary therapy, and uncontrolled data suggest that the information gained from allergy testing may help guide therapy. In addition, dietary therapy is sometimes used in the treatment of adults who are willing to make dietary modifications. Allergy testing may also help with the management of concomitant atopic disease, which is common in patients with eosinophilic esophagitis. (See "Treatment of eosinophilic esophagitis (EoE)", section on 'Dietary therapy' and "Overview of skin testing for IgE-mediated allergic disease" and "Diagnostic evaluation of IgE-mediated food allergy".)

Other diagnostic tests — Other diagnostic tests that have been evaluated but that are not routinely used include functional lumen imaging probe [170], endoscopic ultrasound [171,172], impedance planimetry to measure esophageal pressures and distensibility [173-175], mucosal impedance contour analysis that evaluates esophageal mucosal integrity [176], esophageal manometry [63,177,178], and endoscopic confocal laser microscopy [179,180]. In addition, an esophageal string test has been developed as a tool to measure esophageal inflammation in patients with esophagitis. In one study that included 41 children, measurement of eosinophil-derived protein biomarkers distinguished between children with active eosinophilic esophagitis, treated eosinophilic esophagitis in remission, GERD, and those with a normal esophagus [181,182].

The Cytosponge has been developed for use in diagnosis and monitoring of EoE. This test captures esophageal cells that can then be assessed for levels of eosinophil derived neurotoxin (EDN). In several studies of adult patients, the levels of EDN was highly correlated with mucosal eosinophilia [183-185].

Findings on standard esophageal manometry in most patients with eosinophilic esophagitis are nonspecific [63,177,178].

DEFINING DISEASE SEVERITY — For patients with an established diagnosis of eosinophilic esophagitis (EoE), a scoring system to assess disease severity has been developed by a panel of experts in gastroenterology and allergy immunology [186]. The following clinical features are assigned a point value ranging from 1 to 15 (see 'Diagnosis' above):

●Symptoms and associated complications – Symptom frequency, food impaction, hospitalization

●Endoscopic features – Edema, furrows, exudates, rings, strictures

●Histology – Eosinophil burden per high power field

The sum of the values determines the Index of Severity for Eosinophilic Esophagitis (I-SEE; mild [1 to 6 points], moderate [7 to 14 points] or severe [15 points or higher]). The severity index can be assessed at initial diagnosis and then at subsequent visits. Further studies that validate and refine the I-SEE are awaited before using it routinely in clinical practice.

DIFFERENTIAL DIAGNOSIS — The differential diagnosis includes a variety of conditions that can cause morphologic or histologic findings that resemble eosinophilic esophagitis

Other causes of esophageal eosinophilia — Esophageal eosinophilia is the finding of eosinophils in the squamous epithelium of the esophagus [110]. Esophageal eosinophilia can be seen in association with multiple conditions. These include gastroesophageal reflux disease (GERD), recurrent vomiting due to other causes, parasitic and fungal infections, congenital rings, Crohn disease, periarteritis, allergic vasculitis, drug injury, connective tissue diseases, bullous pemphigoid, pemphigoid vegetans, graft-versus-host disease, achalasia, drug hypersensitivity, celiac disease, vasculitis, carcinoma, and a number of causes of peripheral eosinophilia in which the esophagus (along with other organ systems) may become involved (table 2). (See "Eosinophil biology and causes of eosinophilia", section on 'Disorders with eosinophilic involvement of specific organs'.)

There are occasional patients who have characteristic findings of esophageal rings but do not have esophageal eosinophilia. In addition to considering the diagnoses above, it is important to ensure that adequate biopsies were obtained. In our clinical experience, some of these patients have responded to dietary elimination therapy used to treat eosinophilic esophagitis. Whether such patients have eosinophilia in deeper layers of the esophagus not sampled by standard biopsies is unknown.

Distinction from GERD — The most common consideration in the differential diagnosis of eosinophilic esophagitis is GERD. Case reports have described children with a dense eosinophilic infiltrate that resolved after treatment with a proton pump inhibitor (PPI) alone [187], suggesting that the histologic findings may have been due to GERD:

●In a series of 36 children with ≥15 eosinophils/HPF, 14 (39 percent) responded histologically to high-dose PPIs alone [111].

●In a study of 712 patients with upper gastrointestinal symptoms undergoing endoscopy, 35 (5 percent) had ≥15 eosinophils/HPF on biopsies obtained from the upper-middle esophagus [188]. Twenty-six patients (75 percent) had a clinicopathologic remission on treatment with a PPI, including half of the patients with a typical eosinophilic esophagitis phenotype. Based upon these findings, the authors concluded that using histologic criteria alone to diagnose eosinophilic esophagitis may lead to an overestimation of the prevalence of the disorder.

Histologic features suggestive of eosinophilic esophagitis rather than GERD include:

●Large numbers of intraepithelial eosinophils on histologic examination [150,189-191]. In two reports, the presence of more than 20 eosinophils/HPF was typically associated with non-acid-related causes of esophagitis [191] and patients with eosinophilic esophagitis had significantly more eosinophils than patients who responded to therapy for GERD (28 to 31 versus 5 per HPF overall, and 19 to 32 versus 1 per HPF with biopsies from the proximal esophagus) [189]. Patients with eosinophilic esophagitis are also more likely to have ≥15 eosinophils/HPF in three or more biopsies taken at different levels [111].

●Other histologic findings favoring eosinophilic esophagitis include proximal esophageal involvement, subepithelial and lamina propria fibrosis [149], eosinophilic abscesses [84,190], more severe basal cell hyperplasia [192], activated mucosal mast cells/increased epithelial tryptase density [151,154,193,194], and degranulating eosinophils [56].

●Assessment of eotaxin-3 and major basic protein levels in esophageal biopsy specimens (by immunohistochemistry or real-time polymerase chain reaction [PCR]) has been suggested to help differentiate GERD from eosinophilic esophagitis, but further studies are needed [150,195,196]. (See "Eosinophilic esophagitis (EoE): Genetics and immunopathogenesis", section on 'Role of the immune system and environmental factors'.)

Prediction models have also been developed to identify patients with eosinophilic esophagitis and distinguish them from patients with GERD [56,190]. One such model that incorporated eight clinical and endoscopic features (younger age; male sex; presence of dysphagia and food allergies; presence of esophageal rings, furrows, and plaques; and lack of a hiatal hernia) predicted eosinophilic esophagitis with an accuracy, sensitivity, and specificity of 84, 97, and 92 percent, respectively [197].

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Eosinophilic esophagitis".)

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

In addition, manufacturers of elemental formulas have template letters for medical necessity of amino acid diets available on their websites.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)

●Basics topics (see "Patient education: Upper endoscopy (The Basics)" and "Patient education: Eosinophilic esophagitis (The Basics)")

●Beyond the Basics topic (see "Patient education: Upper endoscopy (Beyond the Basics)")

SUMMARY AND RECOMMENDATIONS

●Clinical presentation – Eosinophilic esophagitis should be considered in adults with a history of food impaction, with persistent dysphagia, or with gastroesophageal reflux disease that fails to respond to medical therapy. In children, symptoms that may be associated with eosinophilic esophagitis vary by age and include feeding disorders, vomiting, abdominal pain, dysphagia, and food impaction.

In particular, the diagnosis should be considered in young males or boys, and in those with a history of food or environmental allergies, asthma, or atopy. A history of esophageal perforation or severe pain after dilation of a stricture should also raise suspicion of this disorder. (See 'Clinical manifestations' above and 'Clinical suspicion and evaluation' above.)

●Diagnosis – Making a diagnosis of eosinophilic esophagitis requires the presence of both symptoms and histologic findings. In addition, other disorders that can cause esophageal eosinophilia, should be ruled out. (See 'Diagnostic criteria' above and 'Differential diagnosis' above.)

The diagnosis of eosinophilic esophagitis no longer requires a trial of proton pump inhibitor treatment.

In patients suspected of having eosinophilic esophagitis, the first diagnostic test is typically an upper endoscopy with esophageal biopsies. We suggest that, at a minimum, two to four biopsies be obtained from the distal esophagus, as well as another two to four from the mid or proximal esophagus. (See 'Diagnosis' above.)

●Endoscopic features – A variety of morphologic features in the esophagus have been described in patients with eosinophilic esophagitis. Endoscopic findings include (see 'Endoscopy' above):

•Stacked circular rings ("feline" esophagus) (picture 1)

•Strictures (particularly proximal strictures) (image 1)

•Attenuation of the subepithelial vascular pattern

•Linear furrowing that may extend the entire length of the esophagus (picture 2)

•Whitish papules (representing eosinophil microabscesses) (picture 1)

•Small caliber esophagus

●Histologic features – Histologic findings suggestive of eosinophilic esophagitis include (see 'Histology' above):

•Peak eosinophil count of ≥15 eosinophils per high power field (picture 3)

•Eosinophil microabscesses

•Superficial layering of eosinophils

•Sheets of eosinophils

•Extracellular eosinophil granules

•Subepithelial and lamina propria fibrosis and inflammation

•Basal cell hyperplasia

•Papillary lengthening

●Evaluation for allergies – Because of the strong association of eosinophilic esophagitis with allergies, we suggest that patients with eosinophilic esophagitis undergo evaluation by an allergist. The results of the evaluation may have treatment implications (eg, elimination diets). (See "Allergy testing in eosinophilic esophagitis" and "Dietary management of eosinophilic esophagitis".)

ACKNOWLEDGMENT — The UpToDate editorial staff thank Dr. Glenn Furuta, MD, for his contributions as author to prior versions of this topic review.