The Role of Esophageal Physiologic Tests in Eosinophilic Esophagitis

Since its first description by Straumann in Switzerland and Attwood in the USA, eosinophilic esophagitis (EoE) has become one of the most common causes of esophageal symptoms, mainly dysphagia and food impactions in adults. It has a negative impact on health-related quality of life and is a burden to health care systems [1]. Studies from Switzerland and the USA have shaped our understanding of EoE’s natural history leading to the development of conceptual models for disease progression from an inflammatory phenotype to a combined inflammatory-fibrostenotic one [2, 3]. Its pathophysiology involves the interplay of genetic susceptibility, early-life exposures, and environmental factors, mainly food allergens leading to a T-helper 2-mediated immune response with eosinophil recruitment into the esophagus [4].

Uncontrolled inflammation and long diagnostic delay are associated with fibrotic remodeling of the esophageal wall [2, 5‒7], epithelial mesenchymal transition, subepithelial fibrosis [8], angiogenesis, smooth muscle hypertrophy, and a marked inflammatory infiltration of the myenteric plexus [9, 10]. The full-thickness fibro-inflammatory cascade makes EoE a transmural esophageal disease, ultimately leading to strictures with food impactions [2, 5, 7, 11, 12] as well as dysmotility [10, 12, 13] with impaired bolus clearance and dysphagia.

Although dysmotility seems to be mainly secondary to the biomechanical remodeling-dependent muscular dysfunction with asynchrony of circular and longitudinal muscle contraction during swallows [14, 15], there is evidence that eosinophil degranulation products lead to smooth muscle hypertrophy promoting fibroblast to myofibroblast differentiation [8], activate the muscarinic acetylcholine receptors with subsequent smooth muscle reactivity [16, 17], alter smooth muscle contractility [4], cause myenteric axonal necrosis and impairment of neurotransmitters delivery to the esophagus [16, 18, 19]. Moreover, there is an increasing body of evidence for the role of mast cells in EoE [4, 16, 20, 21] which promote muscle cell differentiation in the muscularis propria into a more contractile phenotype. In addition, their degranulation products (tryptase, histamine, serotonin) activate smooth muscle contraction, decrease lower esophageal sphincter relaxation, and may participate in neuron damage in the myenteric plexus [22].

The correlation between EoE-related symptoms on one hand and endoscopic and histologic activity on the other hand is only modest [23] which indicates a major role for esophageal remodeling processes that are not fully captured by validated instruments to evaluate clinical, endoscopic, and histologic activity. Complementary esophageal physiological testing allows for a comprehensive understanding of the full-thickness structural and functional esophageal alterations of EoE, highlighting the key role of dysmotility, impaired bolus clearance, esophageal hypervigilance, and symptom-specific anxiety [24] in symptom generation. Physiological esophageal tests have the potential to fill this gap in EoE. In this review, we present the findings and the clinical utility of esophageal functional tests in EoE which are summarized in Table 1.

High-resolution manometry (HRM) is the gold standard for the evaluation of esophageal primary peristalsis and lower esophageal sphincter integrity and function through the representation of acquired data as pressure topography plots. The globally adopted Chicago Classification version 4.0^© provides standardized interpretation parameters and diagnostic criteria for HRM. Although esophageal mechanical obstruction is key in the development of dysphagia in EoE, up to one-third of patients present symptoms of esophageal dysfunction despite a normal endoscopy, unveiling the importance of motor dysfunction secondary to the fibrotic esophageal wall remodeling [16, 23] with an added component of reactive contractile response encompassing spasms and hypercontractility secondary to strictures. The prevalence of motor dysfunction in EoE increases with longer disease duration [25, 26].

Although dysmotility in EoE spans the entire motility spectrum [27‒30], normal motility is the most common finding, followed by ineffective motility in approximately one-third of patients [31]. Achalasia (subtypes I, II, III) and esophagogastric junction (EGJ) outflow obstruction may be present too and may need targeted management [32]. Even though HRM mainly evaluates the circular muscle function through primary peristalsis [15], it may also reflect longitudinal muscle contractions responsible for esophageal axial shortening. In normal esophageal physiology, there is a synchronization of circular and longitudinal muscle contraction during peristalsis to facilitate the propulsion of the bolus along the esophagus [33]. In EoE, there is a lack of coordination between these muscle layers [5] with a selective longitudinal layer impairment due to fibrosis precluding esophageal shortening along that axis [15]. This phenomenon is followed by abnormal longitudinal muscle relaxation that may be reflected as stenosis on imaging [15] and endoscopy. An ensuing panesophageal pressurization (PEP) pattern is seen on HRM characterized by simultaneous esophageal pressurization exceeding 30 mm Hg, extending from the upper esophageal sphincter to the EGJ and impairing bolus transit. PEP is one of the most frequent abnormalities in EoE [27, 29, 31] and is seen in 36% of patients [27], reflecting the reduced wall compliance and dysregulated wall mechanical properties secondary to fibrosis. PEP has been shown to be associated with previous episodes of food impaction [29].

One other pressurization pattern that is important to bolus transit is intrabolus pressure (IBP) which normally increases with increasing bolus volume and viscosity. IBP is the compartmentalized pressure encountered by a bolus when transiting through the esophagus [34] and reflects the resistance to bolus flow which is increased in EoE [27, 35], especially in the fibrostenotic subtype compared to the inflammatory one [35]. Elevated IBP is linked to dysphagia [36], and an IBP above 16 mm Hg seems to predict a fibrostenotic phenotype with a 70–75% sensitivity and specificity but cannot predict the location of the stricture [35]. Abnormal pressurization patterns reflect impaired bolus transit secondary to the fibrotic transformation of the esophageal wall, impairing normal biomechanics. IBP as well as esophageal motor abnormalities parallels disease duration [25, 26, 35]. Motility abnormalities can resolve in up to 86% of patients after 8 weeks of topical steroids except for intrabolus pressurization [30], corroborating the hypothesis that eosinophils might cause reversible esophageal motility abnormalities [37].

For the time being, HRM does not predict EoE’s phenotypes which may be either due to its scarce use or to the heterogeneity of protocols applied in research trials until now. However, it plays a central role in patients who remain symptomatic despite a well-controlled inflammation by unraveling esophageal motor abnormalities that may need specific therapies [31]. Moreover, HRM may raise the clinical suspicion of EoE in patients with previously negative endoscopy in whom a repeated endoscopy with esophageal biopsies is necessary to achieve a final diagnosis of EoE. Therefore, HRM may help to unmask a missed EoE diagnosis as well as esophageal strictures in established EoE patients in cases of high IBP.

Functional lumen imaging probe (FLIP) is the new kid in the block for the evaluation of the esophagus and is already integrated in several guidelines and expert consensuses [38]. It uses high-resolution impedance planimetry technology during volume-controlled distension of an infinitely compliant bag filled with a conducting solution. The Northwestern group transformed high-resolution impedance planimetry into FLIP topography plots [13, 39] with a color-coded display of diameter, axial length, and time. Its advantage over HRM does not solely lie in its ability to objectively determine the biomechanical properties of the esophageal wall [40] and evaluate the contractile response to distension (secondary peristalsis), but it also reflects the function of the esophageal longitudinal muscle [15, 41]. The detailed assessment of the biomechanical properties of the esophageal wall and the EGJ provided by FLIP with a greater accuracy and objectivity than endoscopy, reflects the impact of inflammation and fibrotic remodeling on bolus transit and motility in EoE [35], and emphasizes the importance of FLIP in EoE [40, 42].

When compared to healthy esophagus subjects, EoE patients more frequently present with an abnormal secondary peristalsis (33% of patients) that is correlated with the fibrostenotic rather than the inflammatory features of the disease [26]. Moreover, EoE patients display reduced esophageal cross-sectional area (<225 mm², especially <125 mm²) and compliance (<450 mm³/mm Hg), defined as the change in volume as a function of intraluminal pressure as well as a reduced distensibility plateau (DP) (<17 mm, especially 12.5 mm) [41, 43‒45], defined as the narrowest, fixed diameter that does not expand despite increasing pressure [26]. These parameters are more robust predictors of the risk for food impaction and the need for dilation than endoscopy [43] and eosinophilia [40]. Within EoE, DP is lower in patients with prior food impactions compared to those with dysphagia alone [40]. Reduced distensibility correlates with the endoscopic severity of rings, a marker of esophageal remodeling, but not with eosinophilia [40, 41], exudates, or furrows [43] which are markers of inflammation.

The application of FLIP has further consolidated the concept of EoE being a progressive fibrostenotic disease [2, 46] since the reduced distensibility and strictures formation are proportional to the duration of disease and to the diagnostic delay [47]. The Northwestern group went a step further and developed a physiomechanical classification of EoE based on FLIP panometry [44] with the ability to predict histologic response rates to proton pump inhibitors (PPIs). The application of this classification opens new horizons in guiding management, potentially leading to the upfront avoidance of therapies such as PPIs that would fail in patients with spastic-reactive fibrostenosis or nonreactive fibrostenosis, probably due to the greater symptom duration (∼20 years) and the more severe ring subscores (EREFS grade 2–3) compared to patients with a “normal” physiomechanical classification. Research focusing on the standardization of the FLIP protocol is ongoing and is the cornerstone of the future of this technology [48].

Esophageal distensibility measured by FLIP is modifiable with medical and dietary therapy, and DP improvement [49, 50] correlates with an improvement in endoscopic rings and is a better indicator of symptomatic improvement than eosinophilia, suggesting that symptom improvement in EoE is more related to the improvement in the biomechanical properties of the esophageal wall [50]. Hence, FLIP will likely be a key player in the field of EoE not only by an early prediction of therapeutic response, guiding management, and monitoring therapeutic response but also by filling the gap between inflammatory and symptom improvement.

In a recent study, 40% of EoE patients showed a diffusely narrowed esophagus when using a standardized barium esophagram (BE) [51]. By using a standardized approach to calculation of the maximal esophageal diameter and the minimal diameter, 55% of EoE patients had abnormal esophageal caliber, reflecting the impact of remodeling on the esophageal wall [52]. Esophageal diameter evaluated by BE increases over time in EoE patients in remission, independent of dilation and medical therapy according to a recent study with a follow-up of 12 years [53]. Hence, BE is an objective test that fills the gap of the suboptimal detection of esophageal luminal dimensions on endoscopy which has poor sensitivity (14.7–25%) and only modest specificity (79.2%) for detection of a narrowed esophagus even at a cutoff diameter of maximal esophageal diameter ≤15 mm compared with BE [54]. Furthermore, BE is a cheap and readily available test with a greater accuracy for the detection of strictures and narrow-caliber esophagus as well as for the location of strictures than endoscopy [54]. Its sensitivity is further increased with the use of a 12.5-mm barium tablet [55, 56].

The relationship between gastroesophageal reflux disease (GERD) and EoE is complex. They are not considered mutually exclusive disorders [57] and can coexist, but the directionality of the link between them remains unknown [58]. PPIs are a mainstay in the medical armamentarium of EoE [58]. They decrease esophageal eosinophilia by both, an anti-inflammatory effect [58‒60] and by restoring the integrity of the epithelial barrier with the reduction of esophageal exposure to acid reflux, hence reducing transepithelial allergen influx to the mucosa and submucosa [61]. Reflux-like symptoms in EoE may be either due to concomitant GERD or to physiological reflux in a hypersensitive eosinophil-inflamed esophageal mucosa with the association of motility and clearance impairments.

The diagnosis of GERD is based on the updated Lyon 2.0 consensus [62] which defines specific and validated endoscopic, pH-impedance and wireless-pH criteria. In the field of EoE, pH testing may be used in patients in histologic remission who present reflux symptoms. It can be done either off or on PPI depending on the absence or the presence of endoscopic proof of GERD, respectively. pH-impedance monitoring allows for the assessment of flow of air and liquids as well as their directionality through the esophagus by measuring the changes in esophageal conductivity or impedance, expressed in Ohms. Low baseline impedance or mean nocturnal baseline impedance reflects an impairment in the esophageal mucosal barrier integrity and has now been integrated in the GERD diagnosis algorithm [62]. Physiologically, there is a caudal decrease of baseline impedance throughout the esophagus with a proximal to distal gradient but not in EoE [63]. In EoE, the dilation of the intercellular space and the reduction in the expression in intercellular junction proteins and adhesion molecules [64‒66] increase the permeability of the mucosa, decreasing its transepithelial resistance and baseline impedance throughout the esophagus [63, 67]. Another novel pH-impedance metric is the post-reflux swallow-induced peristaltic wave (PSPW) index which reflects the physiological chemical clearance of reflux linked to a vagally mediated esophago-salivary reflex eliciting primary peristalsis and delivering bicarbonate and epidermal growth factor to the esophagus to increase the pH and heal the mucosa.

Although some previous studies have shown that pH monitoring is not helpful to accurately predict histologic and symptomatic response to therapy in EoE, recent studies using pH-impedance [68, 69] proved it as a predicting tool of therapeutic response to PPIs. EoE patients have an abnormal reflux burden with a higher number of reflux episodes compared to controls [68]. The low mean nocturnal baseline impedance gradient between the mid and distal esophagus off PPI and the low PSPW index on PPI are present in PPI-refractory patients, suggesting a role for reflux in the pathogenesis of EoE in genetically susceptible patients, leading to the inflammatory eosinophilic cascade [61].

As previously mentioned, impedance measures the conductivity of the epithelium to low-intensity electrical current and reflects the dilation of the intercellular spaces between esophageal epithelial cells (spongiosis) secondary to chronic esophageal disorders such as GERD and EoE [70]. The endoscope-guided direct mucosal impedance (MI) test uses a through-the-scope probe formed of impedance electrodes mounted on a balloon allowing for direct sensor contact onto the epithelium [70]. This technology can distinguish normal esophageal mucosa from inflammatory disorders such as EoE and GERD [70]. In GERD, the typical MI pattern shows a gradual decrease in impedance from the proximal to the distal esophagus, while in active EoE, MI is consistently low all along the esophagus [70, 71]. MI may alleviate costs by reducing the need for esophageal biopsies since there is an inverse correlation between MI and both eosinophilia and dilation of the intercellular spaces [67, 72]. The typical EoE MI pattern is capable of predicting the diagnosis of EoE with a sensitivity of 100%, a specificity of 96%, a positive predictive value of 96%, and a negative predictive value of 100% [73, 74]. Improvement in MI values correlates well with successful EoE treatment (PPI, steroids, diet) [70, 72].

Since current diagnosis and follow-up in EoE require serial endoscopies with biopsies, there is a growing interest in less invasive tests for the upfront prediction of therapeutic response and follow-up. The exact place of esophageal functional testing in EoE is yet to be determined taking into account the mounting evidence of its utility in clinical practice and its ability to help streamline upfront effective therapy, reducing the interval to histologic remission as well as the need for repetitive endoscopies and biopsies. Esophageal physiological testing can segregate patients into different phenotypes characterized by various therapeutic responses. Physiological testing is indicated in symptomatic patients despite histologic remission and an esophageal caliber >16 mm [75], so the biomechanical properties of the esophageal wall as well as esophageal hypervigilance contributing to persistent symptoms can be unmasked [76]. Finally, the lack of correlation between esophageal distensibility and mucosal eosinophil density and the better correlation between the former and symptoms suggest the future need to target both of these variables to improve outcomes [40].

J.W.E. has no relevant financial, professional, or personal relationships to disclose. E.S. has received consulting fees and/or speaker fees and/or research funding from Dr. Falk Pharma, GSK, Celgene/Receptos/BMS, and Regeneron-Sanofi. She is an employee of Tillotts Pharma AG. A.M.S. has received consulting fees and/or speaker fees and/or research funding from Adare/Ellodi, AstraZeneca, Dr. Falk Pharma, Gossamer Bio, GSK, Celgene/Receptos/BMS, and Regeneron-Sanofi.

Work supported by a grant from the Swiss National Science Foundation (32003B_204751/1 to A.M.S.) and the Swiss EoE foundation (to A.M.S.).

Jeanine Wakim El-Khoury and Alain M. Schoepfer contributed to study concept and design, acquisition of data, analysis and interpretation of data, drafting of the manuscript, and critical revision of the manuscript for important intellectual content. Ekaterina Safroneeva contributed to study concept and design, acquisition of data, drafting of the manuscript, and critical revision of the manuscript for important intellectual content.