Investigating Airway Symptoms in Eosinophilic Esophagitis (EoE)

Authors: Angela Huynh, Amanda Adsett, Amy Callaghan, Hien Q. Huynh, Hamdy El‐Hakim, André Isaac, Daniela M. Isaac

Abstract

Objectives

Diagnostic delay of eosinophilic esophagitis (EoE) is significant, and efforts are needed to raise awareness and reduce prolonged symptoms and complications. Airway and swallowing symptoms often overlap, especially in children, complicating diagnosis and increasing delays. This study aims to report the range of airway presentations in pediatric EoE patients and identify independent predictors of an EoE diagnosis in patients with aerodigestive symptoms.

Methods

We conducted a retrospective case–control study of pediatric patients < 7 years with histologically confirmed EoE managed in an aerodigestive clinic at a tertiary pediatric center. Variables included demographics, atopic comorbidities, risk factors, airway and dysphagia symptoms, and adaptive eating behaviors. EoE cases were age‐matched to controls with aerodigestive issues and no EoE. Multivariate regression analysis identified independent predictors of EoE.

Results

Fifty EoE patients were included, along with 50 controls. The mean age was 2.4 years, with 70% male. Common EoE symptoms included solid food dysphagia (62%), food pocketing (62%), and self‐reported food allergies (72%). Airway symptoms included chronic cough (38%) and choking (32%). Multivariate analysis identified food allergy (OR 15.7) and food pocketing (OR 11.81) as independent predictors, while dysphagia with liquids (OR 0.058) and recurrent pneumonia (OR 0.022) were negative predictors.

Conclusion

Airway symptoms in pediatric EoE can overlap with other conditions, complicating diagnosis. Key features such as adaptive eating behavior and food allergies should raise suspicion for EoE.

Level of Evidence

Introduction

Eosinophilic esophagitis (EoE) is a chronic immune‐mediated disease characterized by the infiltration of eosinophils into the esophageal epithelium, resulting in clinical symptoms of esophageal dysfunction [1]. This condition can manifest at any age, and the clinical presentations in children with EoE vary based on the age of onset. It has been observed that EoE is more common in males, and there is an increased risk of developing the condition among first‐degree family members [2, 3]. Younger children typically experience symptoms such as feeding difficulties, vomiting, gastroesophageal reflux, food impaction, and abdominal pain [4]. The variations in clinical onset and presentation arise from a combination of different factors, as infants and children express their symptoms differently compared to adolescents and adults. In addition, they are more likely to have less fibrosis and strictures in their esophagus in the inflammatory stage of their disease.

EoE is closely linked to various atopic conditions including concurrent asthma, allergic rhinitis, atopic dermatitis, and IgE‐mediated food allergies [5, 6]. Diagnosis of EoE relies on symptoms, endoscopic observations, and histological findings. The differential diagnosis for EoE is broad and includes GERD, parasitic causes, Crohn's disease, allergic vasculitis, connective tissue disease, and other disorders associated with esophageal eosinophilia [7]. The confirmation of the diagnosis entails performing an upper endoscopy with esophageal biopsies while excluding other disorders that may cause esophageal eosinophilia. On histology, EoE is typically defined histologically by eosinophil‐predominant inflammation, with a peak value of greater than 15 eosinophils per high‐power field or 60 eosinophils per mm^2^ [8, 9].

EoE often goes undiagnosed or unrecognized for significant periods due to the invasive nature of investigations and the nonspecific nature of symptoms, particularly in early childhood [10, 11, 12]. There have been previous reports on associations with EoE and airway clinical manifestations, including subglottic stenosis, chronic cough, and atypical croup [13, 14, 15]. In some cases, these airway symptoms can occur with EoE in the absence of more typical esophageal symptoms, making the diagnosis difficult in these populations.

The objective of this study was to investigate airway symptoms in patients with and without EoE and to determine independent predictors of an eventual EoE diagnosis in patients with both airway and swallowing complaints. We conducted a retrospective chart review to better understand the clinical presentations of pediatric patients with esophageal dysfunction with a specific emphasis on the presence of airway symptoms and to determine the likelihood of an EoE diagnosis.

Methods and Materials

Study Design

A retrospective case–control study was performed of pediatric patients < 7 years who presented with a combination of airway and swallowing symptoms and who underwent endoscopic esophageal mucosal biopsies at the Stollery Children's Hospital, Edmonton, Canada, a tertiary pediatric referral center. This study was approved by the University of Alberta Ethics Research Board (Pro00134229). Participants were former and existing patients within the Divisions of Pediatric Otolaryngology and Pediatric Gastroenterology who were seen from January 2019 to January 2023. Clinical data were extracted from the electronic medical record system.

EoE patients who underwent upper aerodigestive endoscopy were identified through the diagnostic codes and histological findings. The EoE patients were age‐matched to control patients who also underwent upper aerodigestive endoscopy with biopsies to investigate an airway and swallowing related problem but did not have histological findings of EoE. All patients underwent esophagoscopy (EGD) with mucosal biopsies and were < 7 years of age. The EoE patients who underwent EGD with biopsies were diagnosed at the first GI endoscopy. Patients were excluded from the study if they had undergone previous esophageal surgery and if no follow‐up information was available.

Study Outcomes

The primary outcome of the study was to compare the nature of aerodigestive symptoms and airway diagnoses in patients with EoE with those without EoE. Airway symptoms were defined as whether a patient experienced chronic cough, hoarseness, recurrent croup, subglottic stenosis, choking, stridor, chest congestion, recurrent pneumonia, or cyanotic spells at the time of consultation. Secondary outcomes included time to diagnosis of EoE after onset of symptoms and airway factors associated with an EoE diagnosis. Atypical croup was defined as documented croup diagnosis in a child less than 6 months of age or older than 6 years, recurrence greater than three times per year, prolonged episodes lasting greater than 1 week, or episodes requiring intubation or ICU admission. The presence of a croup diagnosis was limited by available documentation and usually included the presence of typical barky cough and biphasic stridor. A cough was considered chronic if parents/caregivers reported the duration of the cough 4 weeks or longer. Subglottic stenosis included any grade of stenosis that was diagnosed on bronchoscopy at any age. Patients were considered to have food allergies if reported by the parents/caregivers at the time of initial consultation. Diagnostic delay was defined as the time between initial presentation to a medical provider and final diagnosis.

Statistical Methods

Descriptive statistics were used to characterize the study patients. Univariate analysis of the medical and demographic characteristics of the participants was tested for statistical significance using the Chi‐square and Fisher exact tests for categorical variables and the Mann–Whitney U test for continuous variables. Multivariate analysis was conducted using binary logistic regression to determine independent predictors of an EoE diagnosis. p‐values were reported as two‐tailed, and p < 0.05 was considered statistically significant. SPSS software, version 29 (IBM) was used for univariate and multivariate analyses.

Results

A total of 74 pediatric patients with histologically confirmed EoE with aerodigestive symptoms were identified during the study period. Of these, 50 were less than 7 years at symptom onset and were included in the analysis. Fifty age‐matched control patients with aerodigestive symptoms who presented with symptoms and were deemed to require upper GI endoscopy and esophageal biopsies, which excluded EoE, were subsequently identified (Table 1). The mean age of EoE patients was 2.4 years, and the mean age of the control population was 2.6 years. EoE was diagnosed 49 months after caregivers of patients sought medical attention. In the EoE group, 26 (52%) had reflux symptoms, 19 (38%) were taking bronchodilators for presumed asthma, and 36 (72%) had a history of a food allergy. Of the patients with an eventual EoE diagnosis, 27 were seen initially by a GI physician, whereas 23 were initially referred to otolaryngology, 5 of whom had previous airway endoscopy prior to their eventual EoE diagnosis.

The most common airway presentations in patients with EoE were chronic cough (n = 19; 38%) and caregiver‐perceived choking with feeding (n = 16; 32%) (summarized in Table 2). Within the Control group, there was an increased variety of airway symptoms compared to the EoE group, the most common of which were chronic cough (n = 39; 78%), choking (n = 34; 68%), chest congestion (n = 26, 52%), and stridor (n = 18, 36%) (Figure 1). All of the EoE patients presented concurrently with GI complaints, with solid food dysphagia (n = 31, 62%) and adaptive eating behaviors such as food pocketing (n = 31, 62%) being the most common (Table 3). Univariate analysis demonstrated that several symptoms and diagnoses were more common in the EoE group, including asthma (42% vs. 16%, p = 0.004), food allergy (72% vs. 12%, p < 0.001), and food pocketing (62% vs. 24%, p < 0.001). A history of hoarseness, stridor, and cyanosis was all more common in the non‐EoE group (Table 2). In contrast, a history of food impaction, atypical recurrent croup, solid food dysphagia, emesis, and abdominal pain were all not statistically significantly different between the two groups (Tables 2 and 3).

The results of the binary logistic regression are summarized in Table 4. This identified food allergy (OR 15.73; 95% CI 3.27–75.63) and food pocketing (OR 11.81; 95% CI 1.57–88.95) as the factors most strongly associated with EoE diagnosis. In contrast, patients demonstrating symptoms of liquid dysphagia (OR 0.058, 95% CI 0.009–0.358) or recurrent pneumonia (OR 0.022, 95% CI 0.002–0.312) were less likely to be diagnosed with EoE.

Discussion

Early diagnosis and treatment are paramount in managing EoE, as it is a chronic inflammatory disease that can progress to esophageal fibrosis and stenosis [16]. EoE presentation has included a spectrum of airway symptoms such as wheezing, asthma, and nasal congestion—all of which can be secondary to type 2 inflammation. Based on literature, there is a high rate of reported airway symptoms and an increase in otolaryngological surgeries in EoE patients [17]. Therefore, there is a need to investigate whether there is a particular phenotype that indicates an EoE diagnosis when patients present with both airway and GI complaints.

The literature has reported an association between EoE and extraesophageal symptoms including cough, recurrent croup, and throat clearing. As many as 60% of patients diagnosed with EoE have associated respiratory symptoms [13]. There is variability between studies with regards to airway symptoms reported. Studies have observed cough in 7%–46%, recurrent croup in 4%–21%, hoarseness in 3%–38%, throat clearing in 30%, and dysphonia in 7% [14, 18, 19]. For example, Cooper et al. [15] reported that 6.25% of children with croup and cough were found to have > 15 eosinophils per high‐power field on esophageal biopsy. The current study found a slightly higher rate of atypical croup in patients with EoE (12%, Table 2). Recognizing that EoE can present with airway symptoms will help increase awareness in other subspecialties such as otolaryngology and pulmonology to consider EoE in the differential diagnosis. There was also a high prevalence of chronic cough in both the Control and EoE groups. Given the frequency of this symptom in patients presenting to GI and otolaryngology as well as in primary care, clinicians should be aware of EoE in the differential diagnosis of a child with chronic cough.

In keeping with previous EoE knowledge, our study participants had a male predominance (3.9:1 male to female ratio) [20, 21]. Food impaction and environmental allergies are symptoms often considered strong predictors for EoE in the setting of dysphagia. However, our data did not identify these as strongly associated with EoE when compared to non‐EoE patients who presented with concurrent airway and swallowing symptoms. We found that the most prevalent GI complaints from EoE patients who also presented with extraesophageal symptoms were adaptive eating behaviors such as food pocketing and self‐reported food allergies. Additionally, patients presenting with symptoms of recurrent pneumonia or liquid dysphagia were less likely to be diagnosed with EoE. In patients who initially present undifferentiated with both digestive and extraesophageal symptoms, this knowledge can help direct the most pertinent investigations.

The mean time to diagnosis was 49 months in EoE patients with aerodigestive symptoms, suggesting a diagnostic delay similar to those with EoE without aerodigestive symptoms. This is in keeping with previous studies that state a delay of EoE diagnosis in children of up to 1.2–3.5 years, and in certain studies, up to 6 years. In the EoE patient group, 76.7% (n = 57) were seen first by a pediatric gastroenterologist and 23.3% (n = 17) by a pediatric otolaryngologist. Therefore, in patients who present with both airway symptoms and dysphagia, knowledge of the symptoms that are most associated with EoE may help shorten the time to diagnosis.

This study had several limitations. This is a retrospective, single‐center study which can limit the generalizability of the results. With this study design, it was not possible to determine causal relationships between EoE and the airway diagnoses studied, beyond simple associations. Depending on whether the patients were initially seen by gastroenterology, otolaryngology, or pulmonology, the assessment may have been focused on different systems, and thus different symptoms may have been elucidated. Other adaptive eating behaviors, such as slow eating, washing food down with water, or food lubrication other than food pocketing, may be more commonly seen in this group and may not have been asked. The study was underpowered to examine rare outcomes such as subglottic stenosis. Despite the young average age in the study, the diagnostic delay was long, suggesting that many patients/parents began reporting symptoms very early in life. It is not possible to determine to what degree all of the reported symptoms were related to EoE, particularly for patients who presented with symptoms as infants; however, it is still interesting to note that patients with an eventual EoE diagnosis often report or present with dysphagia and airway symptoms very early in life.

Conclusion

Airway presentations are common and varied in the pediatric EoE population and can be difficult to differentiate from other aerodigestive conditions. A history of food pocketing and food allergy is a strong predictor of an EoE diagnosis, whereas food impaction may not be pathognomonic among dysphagia patients. Liquid dysphagia and recurrent pneumonia are less likely to indicate an EoE diagnosis in this population. Heightened awareness of these factors can help clinicians when investigating patients with upper aerodigestive tract symptoms.

Conflicts of Interest

The authors declare no conflicts of interest.