Eosinophilic esophagitis (EoE) is an increasingly recognized immune antigen-mediated esophageal disease found in both children and adults. It is defined as a clinicopathologic disease characterized by symptoms of esophageal dysfunction accompanied by an eosinophil-predominant esophageal inflammation that occurs in the absence of other causes of esophageal eosinophilia. Classic symptoms in adults include dysphagia to solids and food bolus impaction but a variety of other symptoms are also encountered. Despite the increasing awareness of EoE among practicing physicians, a long delay from onset of symptoms to diagnosis remains a problem in this disease.
Key points
- •
Eosinophilic esophagitis (EoE) has protean symptoms in adults.
- •
Diagnostic delay in EoE is common and is associated with endoscopic features of fibrosis and remodeling.
- •
No endoscopic or clinical parameters differentiate EoE from PPI-responsive esophageal eosinophilia.
- •
The sensitivity of endoscopic findings of EoE is problematic, emphasizing the importance of obtaining biopsies in individuals in whom EoE is a diagnostic consideration.
- •
Patients with EoE have concerns about the impact of the disease on quality of life.
Introduction
Eosinophilic esophagitis (EoE) is an increasingly recognized immune antigen-mediated esophageal disease found in both children and adults. It is defined as a clinicopathologic entity characterized by symptoms of esophageal dysfunction accompanied by an eosinophil-predominant esophageal inflammation that occurs in the absence of other causes of esophageal eosinophilia. As such, symptoms are central to the diagnosis of EoE. Classic symptoms of EoE in adults include dysphagia to solids and food bolus impaction, but a variety of other symptoms are also encountered, some of which are uncommon and not widely recognized. Despite the increasing awareness of EoE among practicing physicians, a long delay from onset of symptoms to diagnosis remains an important clinical problem, in part because of altered eating behaviors on the part of patients that may result in a delay in seeking medical attention. Furthermore, this diagnostic delay seems to be associated with complications related to ongoing esophageal remodeling. A variety of endoscopic features are associated with EoE but the performance characteristics of these features are problematic. EoE needs to be distinguished from proton pump inhibitor (PPI)–responsive esophageal eosinophilia (PPI-REE), but there are currently no endoscopic or clinical features that allow that distinction. This article summarizes current knowledge of the clinical and endoscopic presentation of EoE in adults.
Introduction
Eosinophilic esophagitis (EoE) is an increasingly recognized immune antigen-mediated esophageal disease found in both children and adults. It is defined as a clinicopathologic entity characterized by symptoms of esophageal dysfunction accompanied by an eosinophil-predominant esophageal inflammation that occurs in the absence of other causes of esophageal eosinophilia. As such, symptoms are central to the diagnosis of EoE. Classic symptoms of EoE in adults include dysphagia to solids and food bolus impaction, but a variety of other symptoms are also encountered, some of which are uncommon and not widely recognized. Despite the increasing awareness of EoE among practicing physicians, a long delay from onset of symptoms to diagnosis remains an important clinical problem, in part because of altered eating behaviors on the part of patients that may result in a delay in seeking medical attention. Furthermore, this diagnostic delay seems to be associated with complications related to ongoing esophageal remodeling. A variety of endoscopic features are associated with EoE but the performance characteristics of these features are problematic. EoE needs to be distinguished from proton pump inhibitor (PPI)–responsive esophageal eosinophilia (PPI-REE), but there are currently no endoscopic or clinical features that allow that distinction. This article summarizes current knowledge of the clinical and endoscopic presentation of EoE in adults.
Dysphagia
Dysphagia is the most common symptom in adults with EoE. Dysphagia may be either chronic or intermittent and is seen in 25% to 100% of adult patients with EoE. Two recent studies provide perhaps the best estimate of the magnitude of this problem using modern consensus definitions of EoE that exclude PPI-REE. A Swiss EoE database study of 200 patients with EoE found dysphagia to be present in 95% of patients before the diagnosis of EoE. The median diagnostic delay from onset of symptoms to initial diagnosis of EoE was 6 years. Dellon and colleagues took the issue of dysphagia a step further and examined the relationship between the presence of dysphagia and the endoscopic phenotype of EoE in the University of North Carolina EoE database. Dysphagia was present in 83% of the adult cohort. From the inflammatory to the mixed inflammatory/fibrostenotic to the fibrostenotic endoscopic phenotypes (described later), dysphagia increased from 36% to 77% to 92% of the patients respectively. Symptoms in each of these phenotypes were similarly present before the diagnosis of EoE for a mean of 5.3, 8.3, and 8.4 years respectively. Both of these studies highlight the importance of the diagnostic delay that is so common in EoE as well as the cardinal symptom of dysphagia. What accounts for this diagnostic delay? Although initially part of the problem was likely the lack of physician awareness and recognition, it also may be related to patient reporting of symptoms. Symptoms of dysphagia may be minimized by patients because of a variety of adaptive behaviors to lessen this symptom, including eating more slowly, using liquid chasers with a meal, and chewing food more carefully. Thus, elucidating these behavioral modifications is crucial in the clinical assessment of these patients.
How common is EoE in patients undergoing endoscopy for unexplained dysphagia? This question is difficult to answer because 2 studies that addressed this question did not use current consensus criteria for the diagnosis of EoE, thereby allowing an answer only to the issue of esophageal eosinophilia and not EoE. Esophageal eosinophilia was found in 15% to 16% of patients undergoing endoscopy for unexplained dysphagia. Esophageal eosinophilia was encountered in 5% to 10% of these patients with dysphagia despite an endoscopically normal-appearing esophagus.
What accounts for dysphagia in EoE? Several different mechanisms have been proposed, with the most important being esophageal remodeling. This process of progressive fibrosis may lead to fixed rings, strictures, decreased esophageal distensibility, and narrow-caliber esophagus. Dysphagia may also be caused by dysmotility related to alterations in neuromuscular function of the esophagus, leading to dysfunction of longitudinal muscle contraction. A wide spectrum of esophageal motility abnormalities have also been described in EoE; notably weak peristalsis and failed peristalsis. However, the relationship between these abnormalities and dysphagia remains unclear. Overall, it seems that esophageal remodeling is likely the key driver of the symptom of dysphagia.
Food bolus impaction
Food bolus impaction is another common clinical manifestation of EoE in adults. It can either precede the diagnosis of EoE or be an ongoing manifestation of the disease. The symptom is characterized by either a sense of food sticking for a period of time before eventual passage or sticking to the point of requiring endoscopic removal. Food bolus impaction warranting endoscopic removal is encountered in 33% to 54% of adult patients with EoE. A study of 251 patients in the Swiss EoE database, diagnosed by consensus criteria, found that 87 patients (35%) had one or more food bolus impactions that required endoscopic removal. Studies of unselected consecutive patients presenting with food bolus impaction in the community and in tertiary care settings find EoE by consensus criteria in 12% to 42% of patients. However, this may underestimate the magnitude of the problem because many patients with a food bolus impaction do not have esophageal biopsies obtained at the time of bolus removal or are not on PPI therapy. When considering esophageal foreign body impactions more broadly, patients with EoE are more likely to be younger, white men with a food bolus impaction than patients without known EoE. Food bolus impaction is more likely to be found in the fibrostenotic and mixed endoscopic phenotypes of EoE compared with the inflammatory phenotype. It is also more associated with lower esophageal distensibility than in patients who only have solid food dysphagia. It is clear from the studies to date that multiple episodes of food bolus impaction are characteristic of EoE in adults.
Chest pain
Unexplained chest pain unrelated to swallowing may be encountered in a subset of patients with EoE. The large natural history studies of Dellon and Schoepfer and colleagues found chest pain in 13% and 36% of patients respectively. It is unclear whether this symptom is caused by underlying inflammation, gastroesophageal reflux disease (GERD), acid hypersensitivity, or some other cause.
Gastroesophageal reflux–type symptoms
The classic GERD symptoms of heartburn and acid regurgitation have been described in 7% to 100% of patients with EoE. EoE is found in approximately 1% to 9% of patients with GERD symptoms refractory to PPI therapy. Given the different treatment implications of the two diseases, it is important to differentiate the clinical presentation of the two entities as well as that of PPI-REE, as described later. Clinical features that tend to distinguish EoE from GERD include male gender, younger age, atopy, dysphagia, and prior food impaction, although there is clear overlap between the two entities.
PPI-responsive esophageal eosinophilia
The entity of PPI-REE has recently been described. PPI-REE is defined as symptoms of esophageal dysfunction, with many patients having endoscopic features of EoE, with esophageal eosinophilia that responds to PPI therapy. PPI-REE is viewed as being separate from the esophageal eosinophilia seen in the setting of erosive esophagitis and typical GERD symptoms. The cause of this entity is poorly understood at present and it may represent an atypical presentation of GERD, a variant of allergic EoE that responds to PPI therapy, or a separate entity. Experimental models suggest that PPIs reduce cytokine-stimulated eotaxin-3 expression, a mechanism that may explain a reduction in eosinophils independent of reduced acid secretion.
At present there are no clinical, endoscopic, histologic, or pH testing criteria that allow the distinction of PPI-REE from EoE. The Chapel Hill group recently examined consecutive patients referred for endoscopy, of whom 66 were found to have esophageal eosinophilia and dysphagia when naive to PPI therapy. PPI-REE was found in 24 of these patients (36%), whereas the other 40 were classified as having EoE. On univariate analysis, the patients with PPI-REE were older, and were even more male predominant and less likely to have endoscopic features of EoE such as rings, narrowing, or furrowing than patients with EoE. However, none of these features reliably separated the two groups after multivariable analysis. Furthermore, there was no difference in the absolute number of eosinophils before PPI therapy between the two groups. Given the treatment implications inherent in the two entities, an increased understanding of the significance of this entity is eagerly awaited.
Perforation and intramural dissection
Patients with previously undiagnosed EoE may present acutely with spontaneous esophageal rupture (Boerhaave syndrome). Esophageal perforation typically occurs in the setting of a food bolus impaction followed by induced vomiting. Symptoms suggesting perforation include chest and epigastric pain after vomiting. To date, a limited number of patients have presented in this manner, often in the setting of prior unrecognized esophageal symptoms. Another complication seen as a presentation of EoE is esophageal dissection between the mucosal and submucosal layers. Patients with dissection may have hematemesis in addition to chest pain and dysphagia. Although both of these complications may precede the diagnosis of EoE, they may also result from endoscopy and instrumentation, including food bolus removal and dilation.
Allergic diathesis
Because EoE is a chronic immune/antigen-mediated disease, features of atopy including asthma, allergic rhinitis, atopic dermatitis, food allergies, and environmental allergies are commonly seen in adults with EoE. The overall magnitude of the problem is difficult to ascertain given the evolving definition of EoE and early studies that did not require exclusion of GERD or PPI-REE. However, several recent studies using consensus-based definitions for EoE describe atopy in 60% to 96% of adult patients. In addition, oral allergy syndrome, characterized by oropharyngeal pruritus with or without angioedema of the lip, tongue, palate, and posterior oropharynx after ingestion of uncooked fruits and vegetables, may be encountered in these patients, although the magnitude of this problem remains unclear.
Aeroallergens have also been implicated in EoE, because several studies have found a seasonal predilection for the diagnosis of EoE coinciding with peak pollen season, extending from the spring to early fall. However, given that atopy is not a universal finding, it is unclear what the presence or absence of atopy means in terms of disease presentation, treatment, and natural history.
Miscellaneous symptoms
The clinical presentation of adults is different from that encountered in children. Nevertheless, a variety of clinical findings more characteristic of pediatric EoE may also be seen in adults, including abdominal pain, nausea, vomiting, and failure to thrive. The frequency of these symptoms is difficult to estimate, but practitioners should be aware that nonspecific symptoms such as these may be the only clinical manifestation of EoE in selected adults.
Endoscopic appearance
Patients with EoE have a wide variety of endoscopic findings. These findings include concentric rings (fixed or transient) ( Fig. 1 ), longitudinal furrows ( Figs. 2 and 3 ), white exudates (plaques) ( Fig. 4 ), loss of vascularity (also referred to as edema or mucosal pallor) ( Figs. 5 and 6 ), strictures, narrow-caliber esophagus (see Fig. 1 ), crepe-paper esophagus (mucosal fragility or laceration with passage of an adult diagnostic endoscope), and even a normal-appearing esophagus.
