Optimizing the endoscopic examination in eosinophilic esophagitis

With its increasing incidence and prevalence, providers will encounter eosinophilic esophagitis (EoE) more commonly in clinical practice. In the endoscopy suite, in particular, the suspicion for this condition should be high. Prior studies have shown that ~6% of all patients undergoing upper endoscopy for any reason, >15% having endoscopy for dysphagia, and >50% requiring endoscopy in the setting of a food impaction, will be diagnosed with EoE.¹ In this context, a careful esophageal exam is critical. For EoE, this will help to not only optimize diagnosis, but allow potential therapeutics such as dilation in patients with esophageal strictures or narrowing, and set the stage for monitoring treatment response and assessing outcomes.

Assessment of endoscopic features of EoE

In other esophageal conditions, it is routine practice to assess findings with a classification system. The LA Grade classification for erosive esophagitis and the Prague classification for Barrett’s Esophagus are well known to practitioners. Similarly, the EoE Endoscopic Reference Score (EREFS) should be used for patients with suspected or known EoE.² This was developed as a way to quantify the major endoscopic features of EoE, including edema, rings, exudates, furrows, and strictures (the first letter of these findings also comprise the acronym for the system). Edema is graded as absent (grade 0) or present (1; decreased or absent vascularity). Rings are graded as absent (0), mild (1; subtle ridges), moderate (2; distinct rings), or severe (3; standard scope will not pass). Exudates are graded as absent (0), mild (1; ≤10% of the mucosal surface area), or severe (2; >10% of the surface). Furrows are graded as absent (0), mild (1; present without depth), or severe (2; depth noted). Stricture is graded as absent (0) or present (1), and if possible the minimum diameter of the esophagus is recorded. Narrowing and crepe-paper mucosa are not formally part of the score. The score ranges from 0-9, with higher scores indicating more severe endoscopic disease activity, and in clinical practice each element should reflect the overall worst area in the esophagus.

Since its introduction, inter- and intra-observer agreement of EREFS have been excellent,^{2, 3} the findings discriminate EoE from other conditions with high levels of accuracy (though it is possible to have a normal-appearing esophagus in EoE),^{4, 5} higher scores have been associated with adverse outcomes, the endoscopic appearance is a major determinant of how providers assess disease activity, and EREFS has been responsive to treatment in a number of clinical trials (scores improve with active treatment and remain relatively unchanged with placebo). As such, it is the ideal system to use for endoscopic findings in EoE. On a practical note, when recording the EREFS findings, it is helpful to abbreviate the second “E” as “Ex”, so in this way edema and exudates are not confused (Figures 1 and 2).

Figure 1.

Suboptimal endoscopic views in EoE. (A) Lack of insufflation and inadequate washing of saliva obscures exudates, rings, edema, and furrows. (B) Lack of insufflation. (C) View obscured by blood from scope passage across an unappreciated stricture. (D and E) Lack of washing saliva and debris in the distal esophageal area precludes evaluation of possible strictures. (F) Optimal endoscopic view, with EREFS of E1 R2 Ex0 F1 S1 (11mm diameter).

Figure 2.

Challenge assessing for strictures and narrowing. (A) This first view of the proximal esophagus is not fully insufflated. (B) This is the preferred view with full insufflation, and now all findings are visible with EREFS of E1 R1 Ex1 F2 S13. However, the degree of narrowing or stricture diameter can be difficult to assess. (C) Balloon dilation is performed with the subsequent “dilation effect” shown here after the 15mm balloon, which helps to confirm the initial diameter of 13mm.

Approaching the endoscopic exam in EoE

Despite the ready availability of EREFS as a tool, there are still a number of challenges to its wide implementation and, more importantly, optimal and accurate use, in clinical practice. This largely centers on best practices for examination of the esophagus in EoE. First, it is important to fully examine the esophagus on insertion, before the entire upper exam is complete, because advancing the scope (or performing therapeutics like dilation) can rub off the exudates and falsely lower the score for this feature. It is reasonable to intubate the esophagus and advance to the stomach to make sure there is not retained food or significant fluid that could impact the safety of the exam, but after this, withdraw back to the esophagus for the exam, with gentle washing and suctioning of the mucosa to clear off saliva, mucous, blood, and any other debris. Then, the esophagus must be fully insufflated to accurately assess the features and record the EREFS score. If this is not done, it is difficult, if not impossible, to determine the extent of edema, depth of furrows, and severity or rings. Full insufflation is also important to distinguish between felinization (which is the formation of transient rings due to esophageal foreshortening and/or contraction of the longitudinal muscles, and which is not considered as EoE-related or scored in EREFS) and fixed rings (which are scored for EREFS). Failure of any of these steps leads to a poor esophageal exam (Figures 1A–1E) and inaccurate characterization of disease severity. In contrast, a fully insufflated and clean esophagus allows for clear assessment (Figure 1F). The key thing is to take time during the exam – it can take several minutes to fully assess the esophagus in EoE.

Assessing for strictures and narrowing – have a high level of suspicion

The most challenging parts of the endoscopic exam in EoE are determining whether strictures or esophageal narrowing are present and estimating the esophageal caliber. It is simple to detect these findings when they are very severe and the standard adult scope does not pass. Of note, if there is a severe stricture or area of narrowing and resistance is encountered, do not push blindly as this could lead to a complication; best practice is to switch out to a smaller diameter scope. However, studies have shown that endoscopists do not reliably detect less severe strictures on visual exam alone.^{6, 7} With that knowledge comes the opportunity to improve practice. In EoE, strictures can occur at any location in the esophagus, can be focal or multi-focal, or there can be diffuse narrowing of the esophagus.⁸ An endoscopist should be on the alert for subtle findings. Sometimes strictures can “hide” at the gastroesophageal junction or at the level of the upper esophageal stricture (UES), areas that can be difficult to fully insufflate. Similarly, narrow caliber in the proximal esophagus can be missed if the scope is inserted too rapidly and the exam begins 25 cm or more from the incisors. The tactile sensation of passing the scope can also provide a clue to strictures or narrowing. Is there any mild resistance? This could be a stricture. Is there the feeling of “speed bumps”, or a plication of folds on advancing the scope? This could indicate rings and narrowing. Sometimes the extent of narrowing is only noted after esophageal dilation or if crepe-paper is noted after scope passage. With incomplete insufflation (Figure 2A) the presence of rings is not clear, but with full insufflation (Figure 2B) an area of mild rings and narrowing is notable. After balloon dilation to 15mm (Figure 2C) a good “dilation effect is seen, and the caliber noted in Figure 2B is estimated to be 13mm, a size that is hard to determine on visual inspection alone. In patients with ongoing dysphagia despite treatment, assessing for potential subtle strictures like this is particularly important. The approach to dilation is beyond the scope of this article, as is using a tool such as the functional lumen imaging probe (FLIP), but this may have utility in cases where it is critical to assess esophageal compliance and caliber.

Obtaining esophageal biopsies in EoE

A basic but critical aspect of the endoscopic exam in EoE is obtaining esophageal biopsies for the histologic assessment used for diagnosis and treatment monitoring. The approach to biopsies is based on the understanding that eosinophilic infiltration in EoE is patchy, that the biopsy yield is increased by targeting specific endoscopic findings of EoE (particularly exudates and furrows), and that a higher number of biopsies increases the diagnostic yield.^{9, 10} Therefore, current guidelines recommend at least 6 biopsies from at least two different esophageal locations. One common approach is to take 4 fragments distally and 4 fragments proximally, again targeting the areas where there are findings, and avoiding the so-called sub-UES area, an area of several centimeters where the esophagus often appears normal in EoE and biopsies tend to lack inflammation. Supplemental Figure 1 demonstrates the “turn and suck” approach to targeted esophageal biopsies in EoE. As with scope passage, there also is a tactile feel when obtaining esophageal biopsies. The “pull sign”, a sense of requiring increased force to remove the tissue during a biopsy, has been reported to be highly specific for EoE.

Take home points

The endoscopic exam in EoE should be an area of focus given that EoE will be commonly encountered in the procedures unit. It is important to do a careful esophageal exam, with full insufflation, washing of debris, and sufficient time to fully assess for all findings. The presence or absence of features should be assessed and recorded with the EoE Endoscopic Reference Score (EREFS) at each endoscopy to quantify the endoscopic severity of disease activity. At the same time, a careful assessment for signs of fibrostenosis – strictures and narrowing – should be made, and dilation performed if clinically indicated. Appropriate technique for obtaining esophageal biopsies should also be used, with multiple biopsies targeting active features of EoE from several locations in the esophagus. These techniques will allow for an optimal exam, an increased diagnostic yield, and accurate monitoring of endoscopic features of EoE after treatment and during long-term follow-up.

Supplementary Material

1

Supplemental Figure 1. Approach to esophageal biopsies in EoE. (A) The biopsy forceps are deployed, in this case in the proximal esophageal in an area of clearly visible EoE findings. (B) The forceps are opened, pulled back against the scope, and using the “turn and suck” technique an area of furrows is targeted for the biopsy. (C) The forceps are closed and advanced slightly to confirm the targeted area has been captured, and then the biopsy is obtained. (D) The area of the biopsy, targeted across the furrow to increase the sensitivity of detecting eosinophilic inflammation.