21
Eosinophilic esophagitis

Calies Menard-Katcher, Glenn T. Furuta, and Robert E. Kramer

Introduction

Over the last decade, eosinophilic esophagitis (EoE) has emerged as one of the most common, if not the most common, causes of food impaction in children and adults. The diagnosis of EoE depends on the presence of symptoms and an abnormal esophageal biopsy that contains greater than 15 eosinophils per high‐power field (HPF). Other causes for symptoms and esophageal eosinophilia need to be considered and ruled out prior to assigning the diagnosis of EoE. To date, no peripheral biomarker has been identified that could replace mucosal eosinophilia as a diagnostic benchmark. While there are several efforts to introduce less invasive methods of monitoring of mucosal inflammation once EoE is diagnosed, esophageal endoscopy continues to be essential to the diagnosis and management of EoE.

Endoscopy is not only helpful in its ability to obtain biopsies but also serves in several other ways. First, it helps to exclude other causes for presenting symptoms. Second, assessment of the esophageal mucosa by means of a validated endoscopic assessment score (EREFS) helps in determining the likelihood of identifying inflammation on biopsy and of response to treatment [1,2]. Endoscopy can also be a necessary method to retrieve foreign bodies and dislodge impacted food. Finally, it may be used to perform dilation in patients with EoE‐related stricture(s) or long segment narrowing. This chapter will review each of these topics.

Mucosal biopsy procurement

Biopsies are obtained for three reasons in patients with EoE. At the initial presentation, they are obtained to verify or exclude the presence of mucosal eosinophilia. Biopsies are procured from the proximal and distal esophagus and two studies in adults and children suggest that three biopsies from each site raises the probability of establishing the diagnosis to over 95% [3,4].

Biopsies have also been obtained to monitor disease activity. Several studies provide supporting evidence that reducing inflammation may reflect better outcomes [5] and, to date, using symptoms reported in the clinical setting as a primary assessment of treatment response does not appear to reliably correspond to presence or absence of mucosal eosinophilia. Since the reduction in mucosal eosinophilia may reflect better outcomes, surveillance endoscopic and histologic assessment has been deemed critical in therapeutic studies as well as clinical practice. Follow‐up endoscopy for assessment of mucosal response to treatment is generally performed at approximately 8–12 weeks, though further studies will hopefully provide insight into the optimal timing of follow‐up endoscopic assessment.

Finally, if patients with EoE develop unexpected symptoms while on treatment, endoscopy with biopsy assessment can help to assess for potential opportunistic infection, such as candida or herpes esophagitis.

Assessment of esophageal gross findings

To date, mucosal eosinophil enumeration has served as a biomarker for disease activity and therapeutic success. Since the esophageal biopsy represents less than 1% of the total surface area and counting eosinophils can be fraught with intraobserver variability, alternative methods have been sought. One of these focuses on the gross appearance of the esophageal mucosa and encompasses endoscopic features presently associated with EoE. The Eosinophilic Esophagitis Endoscopic Reference Score (EREFS) documents the presence or absence and severity of mucosal edema, rings, exudate, furrows, and stricture [1] (Figures ). While EREFS has not replaced the utility and specificity of histology assessment in either the clinical setting or in therapeutic trials, at least two studies have shown that its results can be used reliably to assess disease activity. Separate studies in adults and later in children and adolescents determined that the EREFS classification system identified patients with EoE with an AUC of 0.93 [2,6].

Photos depict exudate which is the whitish coating on the esophageal surface. Exudate represents eosinophilic purulent material. — **Figure 21.1** Exudate – whitish coating on the esophageal surface. Exudate represents eosinophilic purulent material.

Photos depict circumferential rings along the length of the esophagus. — **Figure 21.2** Circumferential rings along the length of the esophagus. This finding is representative of chronic remodeling.

A common language describing the endoscopic appearance allows clinicians and researchers to provide a more global assessment of changes seen in response to treatment. It also allows one to describe the possible progression of fibrostenotic complications that occur with chronic inflammation such as corrugated rings, stricture, and narrowing. This becomes important as we attempt to monitor patients over time in their response to treatment.

Photos depict mucosal edema and linear furrows. These finding are representative of mucosal edema with loss of vascular pattern.

Photos depict longitudinal rent. This split can occur with the passage of the endoscope or dilator and represents fragility of the esophageal mucosa. — **Figure 21.4** Longitudinal rent. This split can occur with the passage of the endoscope or dilator and represents fragility of the esophageal mucosa. It is also termed crepe paper esophagus or fragile esophageal mucosa.

Therapeutic uses for endoscopy

Since patients with EoE can develop the complications of food impaction (Figure 21.5) and esophageal stricture, endoscopy is a necessary tool in both circumstances. Food impaction occurs in 33–55% of children and adults with EoE and often is the presenting feature [7]. Methods to remove food range from using single or multiple devices, including snare, net retriever, tripod grasper, rat tooth forceps, biopsy forceps, and suction. Due to the frequent need for multiple passes of the endoscope to fully remove an impaction, use of an overtube can be considered in children large enough to accommodate them, potentially to minimize the trauma of repeated esophageal intubation. Suction using a transparent suction cap secured to the end of the endoscope has been shown to be effective and may reduce procedure time compared to other pull removal techniques [8]. Newer devices that combine a modified snare and a suction cap have been developed specifically to aid in the removal of impacted food. Often, difficult impactions require the use of multiple tools as they are rarely removed as a single piece. While gentle pressure to “push” the impaction into the stomach has been reported, extreme caution should be exercised as it often unknown if there may be a more distal stricture and longitudinal tearing of the mucosa due to crepe paper esophagus may occur.

Photos depict food impaction present in the esophageal lumen. — **Figure 21.5** Food impaction present in the esophageal lumen. Mucosa is edematous.

Timing of this procedure is acute if there is drooling or other evidence of complete esophageal obstruction that puts the patient at risk for aspiration. For this reason, it should be done with endotracheal intubation. Even if the patient is able to manage their own secretions, removal of the impacted food bolus should be performed within 24 hours from the onset of symptoms to avoid tissue necrosis and decrease the risk of perforation during the procedure.

Focal strictures or long segment narrowings occur in a subset of children and many adults with EoE. Strictures may be isolated and focal but can also occur in a discontinuous or long segment fashion. The strictures or narrowing that occur in EoE can be detected by endoscopy but often require a high index of suspicion, complete esophageal insufflation and experience with EoE‐related strictures. Studies in both children and adults demonstrate that narrowing can be missed in up to 55% of patients if endoscopy alone is used as a diagnostic tool, compared to barium esophagram and endoscopy together [9,10]. If a patient has solid food dysphagia, particularly when it persists after initiating EoE‐directed treatment, performance of a barium esophagram, often with a coated pill, can be helpful in planning endoscopic interventions.

The method of performing dilation of the EoE‐related narrowing can be different from that used for peptic or caustic strictures. More frequently, EoE‐related strictures are long segment or diffuse, making them more amenable to bougie dilation with either Maloney or wire‐guided Savary dilators. When focal strictures exist, balloon dilation is a reasonable approach and has the benefit of offering direct visualization during dilation as well as directing all the force radially. Bougie dilation directs some of the dilating force tangentially, which has lead to some concern that perforation risk is increased with this method, though there has not been clear evidence to support this contention. A balloon pull‐through technique has also been described for adults in the management of EoE narrowing [11].

Complications include bleeding and esophageal perforation but several studies in adults and children have found these complications to be rare and no more frequent than in esophageal dilations for other underlying etiologies [12,13]. A systemic review of dilation in EoE and a metaanalysis comparing dilation method found that perforation from esophageal dilation in EoE is rare, and no evidence of a significant difference in perforation risk related to dilator type has been found [14]. Postoperative chest pain, however, is expected and can be preemptively treated with over‐the‐counter pain medications [12].

Dilation can improve dysphagia when used in the appropriate patient but it should not be viewed as an alternative to medication or dietary management of EoE – strategies that are directed at targeting the chronic inflammation. When inflammation is controlled, patients require fewer dilations to achieve a similar improvement in esophageal diameter [15]. When patients have fibrostenotic features and require dilation, more than half of patients – both adolescents and adults – will require repeated dilation in their management, often within a year of initial dilation [13,16].

Future alternative devices for mucosal assessment

While endoscopy will likely remain a critical and necessary part of EoE diagnosis, several approaches have been developed with the goal of either minimizing the invasive nature of repeat endoscopy used in monitoring response to treatment or providing an alternate global assessment of the esophagus in EoE.

Unsedated transnasal endoscopy is currently being used to sample the mucosa for histology [17]. This allows the patient to avoid anesthesia or sedation which may minimize cost and/or the time our patients need to take for procedures. Even less invasive are approaches that assess the esophageal mucosa without the need for sedation or biopsy, including the Esophageal String Test [18], Cytosponge^TM [19], and confocal microscopy [20]. These approaches ideally will allow assessment of histological response to treatment without the use of endoscopy.

The EndoFlip® (Medtronic), a novel device used during endoscopy to identify the distensibility of the esophagus, may provide an outcome measure that more closely associates with patient outcomes. An adult study identified differences in patients with EoE compared to normal and it may be able to predict the likelihood of having a food impaction [21]. Differences in distensibility have also been seen in the pediatric EoE population and appear to be associated with clinical fibrostenotic features [22]. Future studies will be needed to show how these new tools will complement or be incorporated into current management approaches in EoE.

Acknowledgments

This work was supported by NIH 1K24DK100303 (Furuta) and Consortium for Gastrointestinal Eosinophilic Researchers (CEGIR). CEGIR (U54 AI117804) is part of the Rare Diseases Clinical Research Network (RDCRN), an initiative of the Office of Rare Diseases Research (ORDR), NCATS, and is funded through collaboration between NIAID, NIDDK, and NCATS and the support patient advocacy groups APFED, CURED and EFC (Furuta). NIH K23DK109263 (Menard‐Katcher).