Received: 11 February 2020 | Revised: 5 June 2020 | Accepted: 19 June 2020

DOI: 10.1111/all.14495

ORIGINAL ARTICLE
Food Allergy and Gastrointestinal Disease

Food-induced immediate response of the esophagus—A newly

identified syndrome in patients with eosinophilic esophagitis

Luc Biedermann1 | Mark Holbreich2 | Dan Atkins3 | Mirna Chehade4 |

5 6 7 7
Evan S. Dellon | Glenn T. Furuta | Ikuo Hirano | Nirmala Gonsalves |
Thomas Greuter | Sandeep Gupta8 | David A. Katzka9 | Willemijn De Rooij10 |
1

Ekaterina Safroneeva11 | Alain Schoepfer12 | Philipp Schreiner1 | Dagmar Simon13 |

14,15 10 10
Hans Uwe Simon | Marijn Warners | Albert-Jan Bredenoord | Alex Straumann1

1
Department of Gastroenterology and
Hepatology, University Hospital Zurich, Abstract
Zurich, Switzerland
2
Allergy and Asthma Consultant, Background: Dysphagia is the main symptom of adult eosinophilic esophagitis (EoE).
Indianapolis, IN, USA
3 We describe a novel syndrome, referred to as “food-induced immediate response of
Children's Hospital Colorado Anschutz
Medical Campus Aurora, Aurora, CO, USA the esophagus” (FIRE), observed in EoE patients.
Methods: Food-induced immediate response of the esophagus is an unpleasant/pain-
4
Mount Sinai Center for Eosinophilic
Disorders, Icahn School of Medicine at
Mount Sinai, New York, NY, USA
ful sensation, unrelated to dysphagia, occurring immediately after esophageal contact
5
Divison of Gastroenterology and with specific foods. Eosinophilic esophagitis experts were surveyed to estimate the
Hepatology, University of North Carolina prevalence of FIRE, characterize symptoms, and identify food triggers. We also sur-
School of Medicine, Chapel Hill, NC, USA
6 veyed a large group of EoE patients enrolled in the Swiss EoE Cohort Study for FIRE.
Digestive Health Institute, Children's
Hospital Colorado, Gastrointestinal Results: Response rates were 82% (47/57) for the expert and 65% (239/368) for the
Eosinophilic Diseases Program, University
patient survey, respectively. Almost, 90% of EoE experts had observed the FIRE
of Colorado School of Medicine, Aurora,
CO, USA symptom complex in their patients. Forty percent of EoE patients reported experi-
7
Feinberg School of Medicine, Nothwestern encing FIRE, more commonly in patients who developed EoE symptoms at a younger
University, Chicago, IL, USA
8 age (mean age of 46.4 years vs 54.1 years without FIRE; P < .01) and in those with
Department of Pediatrics, University of
Illinois, Peoria, IL, USA high allergic comorbidity. Food-induced immediate response of the esophagus symp-
9
Division of Gastroenterology and toms included narrowing, burning, choking, and pressure in the esophagus appearing
Hepatology, Mayo Clinic, Rochester, MN,
USA
within 5 minutes of ingesting a provoking food that lasted less than 2 hours. Symptom
10
Department of Gastroenterology, severity rated a median 7 points on a visual analogue scale from 1 to 10. Fresh fruits/
Academic Medical Center, Amsterdam, The vegetables and wine were the most frequent triggers. Endoscopic food removal was
Netherlands
11 significantly more commonly reported in male patients with vs without FIRE (44.3%
Institute of Social and Preventive
Medicine, University of Bern, Bern, vs 27.6%; P = .03).
Switzerland
12
Conclusions: Food-induced immediate response of the esophagus is a novel syn-
Department of Gastroenterology and
Hepatology, Centre Hospitalier Universitaire drome frequently reported in EoE patients, characterized by an intense, unpleasant/
Vaudois and University of Lausanne, painful sensation occurring rapidly and reproducibly in 40% of surveyed EoE patients
Lausanne, Switzerland
13 after esophageal contact with specific foods.
Department of Dermatology, University
Hospital Inselspital Bern, Bern, Switzerland

Luc Biedermann, Mark Holbreich, Albert-Jan Bredenoord and Alex Straumann contributed equally.

© 2020 EAACI and John Wiley and Sons A/S. Published by John Wiley and Sons Ltd.

Allergy. 2020;00:1–9.  wileyonlinelibrary.com/journal/all | 1

2 | BIEDERMANN et al.

14
Institute of Pharmacology, University of
KEYWORDS
Bern, Bern, Switzerland
15 eosinophilic esophagitis (EoE), clinical symptoms, immediate response, food-induced
Department of Clinical Immunology and
Allergology, Sechenov University, Moscow, immediate response of the esophagus (FIRE), oral allergy syndrome
Russia

Correspondence
Luc Biedermann, Department of
Gastroenterology, University Hospital
Zurich, Raemistrasse 100, 8006 Zuerich,
Switzerland.
Email [email protected]

GR APHIC AL ABSTR ACT

This study describes food-induced immediate response of the esophagus (FIRE), a novel syndrome frequently reported in eosinophilic
esophagitis (EoE) patients. Eosinophilic esophagitis expert physicians' and patients' surveys were performed to characterize FIRE. A
burning and choking unpleasant or painful sensation with rapid occurrence upon ingestion of trigger food is typical of FIRE, occurring in a
considerable fraction of EoE patients unrelated from classical solid food dysphagia.

1 | I NTRO D U C TI O N abdominal pain, likely due to the inability of children to precisely

articulate complaints.3-5
Eosinophilic esophagitis (EoE) is a chronic inflammatory dis- Over the past few years, our increasing clinical experience
ease, defined clinically by symptoms of esophageal dysfunction identified EoE patients reporting symptoms that were different
and histologically by eosinophil-predominant inflammation.1,2 than manifestations of altered esophageal transport, such as dys-
The clinical portion of this definition relies purely on patient-re- phagia and food impaction. Patients reported a previously unde-
ported outcomes (PROs), whereas the histologic portion involves scribed esophageal symptom complex consisting of discomfort
the quantitative determination of eosinophilic infiltration which occurring exclusively and rapidly after esophageal contact with
serves as a biological marker, although pathological findings in ad- specific foods, such as certain fruits or nuts. Due to the intense
dition to tissue eosinophilia are important. The inclusion of PRO's and often frightening symptoms, patients typically have avoided
in the disease definition, and the fact that dysphagia is often an any further exposures to the inciting food(s). Although the de-
undisclosed complaint, illustrates the importance of physician scriptions had several features in common with food pollen allergy
awareness of this disease in order to avoid unnecessary diagnos- syndrome (FPAS; formerly referred to as oral allergy syndrome),
tic delay. the site of symptoms was unequivocally the esophagus and ret-
The main function of the esophagus is the transport of nutrients rosternal region rather than the oropharynx, clearly separating
from the oral cavity to the stomach. Based on functional studies, these patients' reports from FPAS. 6-8 Of note, most patients also
in particular high-resolution manometry, we know that this volun- explicitly distinguished this symptomatology from either solid
tarily initiated act lasts only a few seconds under healthy conditions. food dysphagia or reflux symptoms.
Solid food dysphagia up to food impaction is the hallmark-symptom The purposes of this study were (a) to evaluate among a large
in adults with EoE. Thus, EoE patients are often asymptomatic ex- group of EoE experts in Europe and the USA whether such a phe-
cept when they ingest solid food. In children, the spectrum of symp- nomenon was independently observed in individual clinical practices
toms is broader including food refusal in addition to chest and/or across physicians; (b) to characterize this novel EoE-related symptom
BIEDERMANN et al. | 3

complex using the perspective of physicians; (c) to receive a compre- gain more clarity regarding the initial findings, a patient survey was
hensive description of this syndrome from the perspective of a large developed that assessed the following items: (a) whether patients
number of EoE patients; and (d) to roughly estimate the prevalence experienced immediate reactions in their esophagus upon ingestion
of this syndrome among the EoE population. of specific foods or beverages; and (b) comprehensive descriptions
of their reactions including the triggering food categories. Patients
were informed that they should disregard the well-known EoE-
2 | M E TH O DS associated solid food dysphagia and symptoms of gastroesophageal
reflux disease (patients' questionnaire; Table S2).
2.1 | Definitions and design of the study The questionnaires were distributed among all patients, that had
previously given their written informed consent for inclusion into
Increasing clinical recognition of atypical symptoms in EoE patients led the Swiss EoE Cohort Study and who were treated at the Swiss EoE
to a face to face meeting on May 7, 2017 of pediatric and adult gastro- Clinics in Zurich and Olten, Switzerland. In total, 368 patients with
enterologists and allergists. Participants reached a consensus to per- previously confirmed diagnosis of EoE according to established cri-
form a survey involving physicians experienced in the treatment of EoE teria4 were invited to participate either via mail or in the outpatient
patients and patients affected with EoE, to assess the phenomenon clinics by a sub-group of authors (PS, LB, TG, and AS). The study
“food-induced immediate response of the esophagus” (FIRE) from the was supported by the Swiss National Science Foundation (SNF grant
experts' and patients' perspective. Physicians experienced clinically 32473B_160115) and was approved by the local ethics committee
and scientifically in the field of EoE were considered as EoE experts. (EKNZ 2015-388). All data were anonymized.
Clinical experiences of the group were gathered and consensus
opinion identified that FIRE was defined as:
2.4 | Statistical analyses
1. symptoms of an unpleasant or painful retrosternal sensation,
2. symptoms occurring immediately and reproducibly after contact All collected data were anonymized and entered into a database.
of the esophageal surface with a particular food or drink, Results of quantitative data are presented as median plus inter-
3. symptoms appearing separately from and unrelated to the well- quartile ranges (IQRs) and range for non-normally distributed data
characterized swallowing difficulties (ie, dysphagia or food impac- or mean ± SD and range for data with normal distribution, whereas
tion) in EoE. categorical data are summarized as the percentage of the group
total. The Mann-Whitney test chi-square test and t test were used
Exclusionary symptoms included a delayed passage or even in- to analyze potential differences between groups for nonparametric
terruption of the swallowing process specifically occurring when and parametric quantitative data, respectively, and for difference
swallowing solid, dry, or fibrous foods (ie, typical solid food dyspha- among categorical variables. A P < .05 was considered statistically
gia), and any symptoms consistent with gastroesophageal reflux dis- significant. All statistical analyses were performed using Prism (ver-
ease were deliberately not included in the reporting of FIRE. sion 7; GraphPad Software).

2.2 | Physicians' survey 3 | R E S U LT S

The investigators developed a physician-oriented questionnaire 3.1 | Physicians' survey

with the purpose of determining (a) whether other providers who
cared for EoE patients encountered FIRE in their EoE patients; and The survey was sent to 57 physicians and 47 responded (response
(b) to obtain detailed information regarding the frequency, timing, rate 82.5%), consisting of 25 gastroenterologists, 21 allergists, and
clinical manifestation, and triggers of these reactions. In addition, one epidemiologist, all of whom were familiar with EoE. Physicians
defined patient characteristics were recorded. (physicians' question- treated adult, pediatric, or both patient groups (40.4%, 38.3%, and
naire; Table S1). This questionnaire was distributed to EoE experts 21.3%, respectively). Almost, 90% of these EoE experts declared
in Europe and the USA, selected by the group of authors based on that a FIRE at least “may” exist in patients with EoE according to a
previous scientific and clinical contributions and recognized exper- Likert scale from 1 to 5, with “may exist”,3 “is likely to exist”,4 and
tise in the field. “absolutely does exist”5 in 30.4%, 30.4%, and 26.1%, respectively.
In contrast, only a minority stated that FIRE is “unlikely to exist” (2;
8.7%) or “absolutely does not exist” (1; 4.4%). The majority of ex-
2.3 | Patients' survey perts estimated the prevalence of FIRE symptoms in the EoE popula-
tion between <5% and 20%. (Figure 1).
Results of the physician survey were discussed by the investiga- Food-induced immediate response of the esophagus was most
tors at a second face to face meeting on June 4, 2018. In order to frequently characterized as burning, painful, and narrowing with a
4 | BIEDERMANN et al.

sensation of heat and pressure on the chest by physician (Figure S1).

Milk products, wheat, nuts, and fruits were considered to be the
most frequent offending food products (Figure S2 illustrating
whether physicians considered candidate food items as typically
FIRE-provoking).
About a third of experts considered certain beverages in addi-
tion to solid food to also trigger FIRE in EoE patients (35.5%), while
another third did not identify any role of beverages in FIRE (32.3%).
Physicians who attributed a role of certain beverages in inducing
FIRE symptoms, wine, beer and milk (29.4%, 23.5%, and 23.5%, re-
spectively), rated these beverages as rather typical or very typical in
inducing FIRE symptoms) (Figure S3).
The onset of FIRE symptoms after food ingestion ranged from
F I G U R E 1 Estimated prevalence of food-induced immediate
response of the esophagus (FIRE) symptoms within the eosinophilic 1 to 60 minutes and 76.7% of patients had onset within 5 minutes.
esophagitis (EoE) population. The percentage of EoE-patients The typical duration of an episode of FIRE ranged from <10 minutes
affected by FIRE as estimated by EoE experts (percentage of EoE to 6 hours. The duration was <30 minutes in 72.4% and <2 hours in
expert physicians on y-axis) is depicted (unsure refers to the answer 96.5% of patients (Figure 2).
“I can't decide this according to my current experience”)
A majority of 48.3% of EoE experts felt that typical EoE pa-
tients with FIRE did not differ in terms baseline clinical charac-
teristics from their counterparts without FIRE, with however a
large fraction of undecided responses (41.4%). In contrast 10.4%
thought that patients with vs those without FIRE differed in terms
of one or more epidemiological features. Likewise, 58.1% think
of FIRE as a chronic/recurrent/reproducible phenomenon in EoE
patients and only 12.9% considered it a temporary phenomenon,
resolving over time (29.3% could not decide based on their current
clinical experience). While FIRE is felt to be seen more often in pa-
tients with active EoE by 35.5%, 22.6% of physicians also consid-
ered FIRE a phenomenon independent of EoE activity (Figure S4).

3.2 | Patient survey

In total, 239 of 368 patients responded to the questionnaire (re-

sponse rate 64.9%). Table 1 provides an overview of baseline char-
acteristics of the patient cohort.
The prevalence of FIRE among all patients was 39.7% (95/239) (ie,
positive response to the key question, Nr6, from our questionnaire;

TA B L E 1 Demographic and disease-specific characteristics of

the eosinophilic esophagitis patients included in the survey

Patient number, n 239

Sex, male/female, n (%) 168/71 (70.3%/29.7%)
Current age; y, mean (range, SD) 51 (20-89, 14.4)
Age at symptom onset; y, median (range, 31 (6-79,20-44)
IQR)
F I G U R E 2 Typical duration of food-induced immediate response
Age at diagnosis; y, mean (range, SD) 40.2 (6-85; 14.5)
of the esophagus (FIRE) symptoms and latency. The estimated
typical interval between ingestion of FIRE-triggering agent and Diagnostic delay; y, median (range, IQR) 5 (0-52; 1-12)
occurrence of symptoms (A) and the estimated typical duration of Presence of allergic diseases, n (%) 139 (58.2%)
FIRE symptoms (B) as estimated by eosinophilic esophagitis (EoE)-
History of bolus obstruction, n (%) 81 (33.9%)
experts (percentage of EoE expert physicians on y-axis) are shown
(h., hours; min., minutes; sec., seconds) Abbreviation: IQR, interquartile range.
BIEDERMANN et al. | 5

Table S1: “Did you ever experience an unpleasant or even painful sensa- endoscopically. There was a trend toward a higher percentage of
tion in the mouth, throat or esophagus (behind the breastbone), which is patients in need of endoscopic food bolus removal was higher in pa-
different from, and unrelated to, the well-known swallowing difficulties, tients with vs without FIRE (40% vs 29.9%; P = .09). In contrast, tak-
immediately after the ingestion of a particular food or drink? (We are ing only male patients into consideration, we observed a significant
not referring to the typical EoE symptoms with a slight choking or even association of FIRE with a prior history of endoscopic food bolus re-
blockade in the swallowing process specifically occurring when swallow- moval (44.3% vs 27.6% in male patients without FIRE, P = .03; 30.4%
ing solid, dry, or fibrous foods). There was no significant difference vs 34.8% in female patients, n.s.; Figure S4).
in FIRE prevalence in male vs female patients (41.7% vs 35.2%). Of
note, patients with FIRE were significantly younger than their coun-
terparts without FIRE (mean age of 46.4 years vs 54.1 years; P < .01). 3.2.3 | Characteristics of FIRE and
Likewise, we observed a younger age of EoE symptom onset (mean associated symptoms
age at symptom onset of 28 years vs 35.6 years, standard deviation
12.6 years vs 15.4 years; P < .01) as well as diagnosis of EoE (mean The median intensity of FIRE symptoms on a visual analogue scale
age at diagnosis of 35.1 years vs 43.3 years, standard deviation from 1 to 10 with 1 as the lowest and 10 as the highest intensity was
13.9 years vs 14 years, P < .01). Interestingly, the presence of FIRE 7 (IQR 5-8; range 2-10). The distribution of patient-reported FIRE
was associated with a significant increase in diagnostic delay in pa- intensity is depicted in Figure 4.
tients with vs without FIRE (8.2 years vs 7.5 years, P < .01; Figure 3). Although there was a considerably wide distribution of pa-
tient-reported FIRE intensity, the majority of patients reported an
intensity as high as 8 out of 10, leading to the overall result that 50%
3.2.1 | Allergic diseases and FIRE of patients reported high intensity scores between 7 and 10.
The latency, that is, the interval between ingestion of the FIRE-
The prevalence of allergic diseases was 58.2% (139/239). We observed triggering food or beverage and the onset of symptoms, was less
significantly more allergic diseases in general and allergic rhino-con- than 5 minutes in the majority of EoE patients experiencing FIRE
junctivitis (ARC) as well as FPAS, in FIRE patients compared to patients symptoms (Figure 5A). The duration of FIRE symptoms was reported
without FIRE, (allergies overall 69.2% vs 53.2%, P = .02; ARC 51.6% vs to be shorter than 120 minutes in the majority of patients (Figure 5B).
33.3%, P < .001; FPAS 26.3% vs 11.1%, P < .01). In contrast, prevalence The most frequent triggering foods and beverages are depicted in
of allergic asthma and food allergies was similar in the two groups. Figure 6A. The characteristic features of FIRE most commonly re-
ported by patients were a narrowing and burning sensation with
a feeling of pressure and a choking sensation (Figure 6B). Anxiety,
3.2.2 | History of food bolus impaction requiring was also a typical symptom of FIRE, according to a third of patients.
endoscopic intervention and FIRE Further, several patients reported the experienced sensation in FIRE
as resembling a cramp or prick. These patient-reported characteris-
One-third of the entire cohort (81/239; 33.9%) reported hav- tics were in line with the results from the free-text question (no. 16),
ing suffered from food bolus impaction that had to be removed with the aforementioned attributes most frequently mentioned and
in addition “panic”, “scraping”, “squeezing”, or “itching”.
While the majority of patients did not experience an increase in
dysphagia during the few days immediately after experiencing FIRE
(68.4%), almost one third (29.5%) had more difficulties swallowing in
the subsequent days (don't know/ not sure 1.1%).
Food-induced immediate response of the esophagus was re-
ported to occur only when EoE was active in 28.4% of patients,
whereas 38.9% of patients considered FIRE to occur independent
of EoE activity (FIRE only when EoE is inactive: 3.2%; don't know/
not sure: 28.4%).
Regarding the temporal occurrence of FIRE in relation to the first
manifestations of EoE, 31.6%, 24.2%, and 15.8% of patients stated,
that FIRE had occurred prior, after or simultaneous to the first EoE
symptoms, respectively (don't know/not sure: 24.2%). The intensity
of FIRE symptoms over time remained roughly stable in 34.7% of
F I G U R E 3 Age and diagnostic delay in food-induced immediate
patients, while worsening and improvement was reported by 25.3%
response of the esophagus (FIRE) vs no-FIRE patients. Current age,
and 13.7%, respectively (don't know/not sure 13.7%). In 12.7% of
age at symptom onset and at diagnosis, as well as diagnostic delay
according to presence or absence of FIRE are depicted (error bars patients, FIRE symptoms had completely resolved after a median of
represent standard deviation). 4 years.
6 | BIEDERMANN et al.

include pain as well as those suggestive of esophageal constriction,

whereas in PFAS itching represents a prominent symptom. In addi-
tion, the patient-reported symptoms suggest that FIRE occurs in the
absence of the typical symptoms of PFAS. Whether the pathogen-
esis of FIRE symptoms are represented within the context of EoE or
are unrelated is not certain but recognition of this syndrome may
provide initial clues toward establishing the diagnosis of EoE.
Based on multiple clinical observations made by EoE-experienced
physicians in the United States and in Europe, and based on a com-
prehensive assessment of its features from expert and patient per-
spectives, we can provide a fairly clear picture of FIRE, identified by
the following most important characteristics:

F I G U R E 4 Intensity of food-induced immediate response of

the esophagus (FIRE) symptoms. The reported intensity of FIRE • First, FIRE is characterized by pronounced retrosternal symptoms
symptoms according to a visual analogue scale (VAS) from 1 to 10 strictly linked to the ingestion of a specific food or beverage, usu-
with the number of patients (y-axis) is depicted ally appearing shortly after the exposure with a latency of less
than 5 minutes with limited duration, as most patients perceive
symptoms for less than 30 minutes.
• Second, despite the rather limited temporal impact of the symp-
toms, FIRE has a considerably high symptom intensity, graded by
at least half of the patients as 7 or more on a 1-10 visual analog
scale. Likewise, and in addition to the intensity of the symptoms,
the impact on the patient is exemplified by the character of as-
sociated symptoms including perceptions of anxiety as well as
chest pains or burning sensations in about 30%-40% of patients.
As a consequence, patients usually try to carefully avoid further
exposures.
• Third, FIRE appears to be associated with distinctive baseline
characteristics of the underlying EoE, in particular a younger age
at symptom onset and at diagnosis, a high incidence of comorbid
atopic disease and a high prevalence in men with previous food
bolus impaction requiring endoscopic removal.

Our data show that FIRE has a surprisingly high prevalence

among adult EoE patients. EoE expert estimates reach 20%, whereas
patient reports reach close to 40%. However, it is uncertain whether
this syndrome is strictly linked to EoE or if it affects patients without
F I G U R E 5 Duration and latency of food-induced immediate
EoE. To date, no corroborative reports have been published.
response of the esophagus (FIRE). Temporal sequence of FIRE
with latency (A) of symptoms and duration (B) upon ingestion of The detailed clinical description of this phenomenon suggests
provoking food items that under the conditions associated with EoE contact of specific
foods with the esophageal surface can rapidly and reproducibly
evoke impressive symptoms. Of note, today the mechanisms lead-
4 | D I S CU S S I O N ing to this phenomenon remain undetermined. It is even possible
that more than one mechanism may be involved. The definition of
This report describes a newly recognized, clinically defined syndrome this symptom complex is therefore currently formulated without
associated with EoE. Based on our findings and the esophageal origin requiring determination of the pathogenesis. Interestingly, a small
of symptoms in these EoE patients, we propose the acronym FIRE: study reported an immediate (and also a delayed) response in the
Food-induced Immediate Response of the Esophagus. This syn- esophagus upon the injection of six different allergen extracts in
drome encompasses esophageal symptoms occurring rapidly after patients with EoE but not in control subjects.9 In addition, some sim-
contact of the esophageal surface with a specific food. Like pollen ilarities with PFAS and the high comorbidity with allergic diseases
food allergy syndrome (PFAS), FIRE is triggered by fruits and vegeta- are indicators that, in at least a portion of the patients, an immedi-
bles but FIRE symptoms, like EoE, originate in the esophagus instead ate allergic response of the esophageal mucosa likely induces the
of the oro-pharyngeal cavity. In contrast to PFAS, FIRE symptoms symptoms of FIRE. A chemical irritation of the inflamed mucosa and
BIEDERMANN et al. | 7

the well-known hypersensitivity of the esophagus in EoE10 are other identified triggers of FIRE, such as for instance with milk, which was
potential explanations for FIRE in some patients. A more detailed however more frequently identified as a FIRE trigger by physicians
understanding of the pathogenic mechanisms underlying this im- as opposed to patients. It remains unclear, if triggers of FIRE might
mediate local reaction has the potential to improve our overall un- sustain or even cause EoE. In this respect, the temporal relationship
derstanding of immune reactions in the esophagus and perhaps the with FIRE symptoms antecedent to EoE diagnosis in about a third
genesis of symptoms in EoE. of patients is noteworthy also in view of the discussion of a poten-
Results of several studies treating EoE with protein-free ele- tial causal relationship of oral immunotherapy in IgE-mediated food
mental diets have consistently shown that EoE is triggered by food allergies and subsequent de novo EoE.14 In case of such a potential
proteins in over 90% of the patients.11 The identification and elimi- relationship of FIRE triggers and EoE, FIRE could aid in identifying
nation of causative foods offers the option of a drug-free treatment. foods that should be eliminated from a given patient's diet, although
Of note, EoE is not simply an immunoglobulin E (IgE)-mediated food because of the severity of their FIRE symptoms, most patients avidly
allergy.12,13 Unfortunately, the search for causative foods is still a avoid them anyway.
challenge, because IgE-based tests—for example, skin prick tests, Our work has several strengths and limitations. The confir-
determination of food-specific serum IgEs—do not reliably identify mation of this newly observed phenomenon by a large group of
triggering foods in EoE. If and how many patients with FIRE have physicians with significant experience in the field of EoE in several
positive skin tests to the foods that trigger their FIRE remains to European countries and the USA represents a strength. Further,
be determined. Whether foods and beverages triggering FIRE are our findings are based on a comprehensive survey of both EoE
causing EoE as well remains to be clarified. Of note, there was only experts and patients affected by EoE. Answers to the question-
a small overlap between food items associated with EoE with the naires from physicians and patients were anonymized and most

F I G U R E 6 A and B, Pareto chart of

most the frequent triggers (A, multiple
answers possible) and characteristic
features (B, multiple answers possible)
of food-induced immediate response of
the esophagus (FIRE). Patients had to
report in a free-text field those foods
or beverages they considered to induce
FIRE symptoms. Several items not fitting
in one of the categories are reported
as “miscellaneous”, including mustard,
chestnut, meat, chocolate, rice, and
bread (A). Patient were to report from a
given selection of characteristic features
with multiple selections possible (from
1—if only one feature to 12—if all given
features would apply, B).Cumulative
frequency of all given answers is depicted
by the orange line and represented on the
rights y-axis.)
8 | BIEDERMANN et al.

importantly, patients responded to the questions without input personal fees from Sanofi-Aventis AG, personal fees from Calypso
from the physicians treating their EoE, which would have invari- Biotech AG outside the submitted work. Dr Holbreich, Dr Atkins, Dr
ably introduced the potential of bias or (involuntary) suggestion Dellon, Dr de Rooij, Dr Safroneeva, Dr Schoepfer, Dr D. Simon, Dr HU
toward any desired responses. Finally, the high response rates to Simon, Dr Warners and Dr Straumann have nothing to disclose. Dr
physician and patient questionnaires considerably minimized the Chehade reports grants from NIH (R01-AI140133, U54-AI117804),
risk of response bias. grants from PCORI, grants from APFED/AAAAI, grants and personal
Certainly, our study has several limitations. We cannot rule out fees from Regeneron, grants and personal fees from Shire, grants
the possibility that physicians and patients participating in the sur- and personal fees from Allakos, personal fees from Adare, personal
veys were prone to suggestion due to the simple description of the fees from Nutricia, personal fees from Medscape, personal fees from
possibility of such an immediate response upon food ingestion at Annenberg Center for Health Sciences, outside the submitted work.
the beginning of the questionnaire. However, we deliberately de- Dr Furuta reports other from EnteroTrack, grants from Holoclara,
signed the introductory wording of the questionnaires to be as short grants from National Institutes of Health, other from Shire/Takeda,
and neutral as possible. Evidently and as applies to every question- outside the submitted work. Dr Hirano reports personal fees and
naire study, our investigation remains on a completely descriptive other from Shire/Takeda, personal fees and other from Regeneron/
level. Our study did neither objectify endoscopic/histologic disease Sanofi, personal fees and other from Adare, personal fees and other
activity nor concomitant medication. We also did not perform any from Allakos, personal fees and other from Receptos/Celgene,
longitudinal analyses regarding intensity of FIRE symptoms during personal fees and other from AstraZeneca, personal fees from
induction treatment for EoE. We therefore cannot determine, Gossamer, personal fees from Arena, personal fees from Esocap,
whether disease activity or lack of response to initiated treatments personal fees from Lilly, outside the submitted work. Dr Gonsalves
for EoE impact on FIRE symptoms. According to the limited available reports personal fees from Allakos Inc, personal fees from Up To
information from the questionnaires, FIRE does not appear to be Date, outside the submitted work. Dr Greuter reports grants from
strongly related to persistent esophageal inflammation or refractory Novartis Research Foundation, personal fees from Falk Pharma,
disease. Physicians’ and patients' responses were rather congruent grants from Vifor, outside the submitted work. Dr Katzka reports
although reflecting some uncertainty, with roughly a third in favor grants from Shire, outside the submitted work. Dr Schreiner reports
of an association of disease activity and FIRE, whereas the major- personal fees from Pfizer, personal fees from Takeda, personal fees
ity of responders were either uncertain or considered FIRE to be a from Janssen-Cilag, outside the submitted work. Dr Bredenoord re-
phenomenon independent of disease activity of underlying EoE (the ports grants from Nutricia, grants from Norgine, grants from SST,
latter reported by almost 40% of patients). By design of the study, grants from Bayer, personal fees from Laborie, personal fees from
the classification of patients’ symptoms was not physician based but Arena, personal fees from Esocap, personal fees from Medtronic,
patient reported. Therefore, we cannot absolutely rule out, that an personal fees from Dr Falk Pharma, personal fees from Calypso,
erroneous overlap with PFAS occurred in a few patients. However, personal fees from Reckett Benkiser, personal fees from Regeneron,
among our patients with FIRE, none reported an exclusive localiza- personal fees from AstraZeneca, other from SST, outside the sub-
tion of symptoms in mouth/tongue and/or throat; that is, all patients mitted work.
reported symptoms in their esophagus (behind the breastbone) and/
or chest and/or upper abdomen. Furthermore, due to the design of AU T H O R C O N T R I B U T I O N
this study we cannot determine, whether there are differences in AS and MH initiated the study and organized group meetings. All
FIRE triggers between the United States or Europe. authors participated on group meetings prior to design of physi-
In conclusion, this is the first systematic report on FIRE, a novel cians' questionnaires. LB, MW, AS, and AJB created the physicians'
clinically defined syndrome characterized by an intense, localized, questionnaire, MW and LB performed the analysis. All authors ana-
esophagus-attributed sensation, occurring in a large portion of EoE lyzed the physicians' questionnaire and provided critical input for
patients rapidly after, and linked to, the ingestion of a specific food the patients’ questionnaire. LB, PS, TG, AS, ES, and Asch created
or beverage. Despite some similarities, FIRE symptoms are different the patients’ questionnaire. LB analyzed the patient data. LB and AS
from symptoms of PFAS. Further investigation of the phenomenon, drafted the manuscript. All authors gave critical feedback at the time
FIRE will ultimately advance our current understanding of the overall of analyzes of the patient data and drafting manuscript. All authors
pathogenesis of EoE and better define the overlap of EoE with other critically read the manuscript and edited the draft version. All au-
atopic diseases. thors approved the final manuscript.

AC K N OW L E D G M E N T S ORCID
We acknowledge all patients and colleagues participating in this study. Luc Biedermann https://orcid.org/0000-0003-0824-4125
Mirna Chehade https://orcid.org/0000-0002-7177-2791
C O N FL I C T O F I N T E R E S T Evan S. Dellon https://orcid.org/0000-0003-1167-1101
Dr Biedermann reports personal fees from Dr Falk Pharma, personal Dagmar Simon https://orcid.org/0000-0001-8965-9407
fees from Vifor AG Switzerland, personal fees from Esocap AG, Hans Uwe Simon https://orcid.org/0000-0002-9404-7736
BIEDERMANN et al. | 9

REFERENCES 11. Arias A, González-Cervera J, Tenias JM, Lucendo AJ. Efficacy of

dietary interventions for inducing histologic remission in patients
1. Attwood SE, Smyrk TC, Demeester TR, Jones JB. Esophageal eosin-
with eosinophilic esophagitis: a systematic review and meta-analy-
ophilia with dysphagia. A distinct clinicopathologic syndrome. Dig
sis. Gastroenterology. 2014;146:1639-1648.
Dis Sci. 1993;38:109-116.
12. Simon D, Cianferoni A, Spergel JM, et al. Eosinophilic esophagitis is
2. Straumann A, Spichtin HP, Bernoulli R, Loosli J, Vogtlin J. Idiopathic
characterized by a non-IgE-mediated food hypersensitivity. Allergy.
eosinophilic esophagitis: a frequently overlooked disease with typ-
2016;71:611-620.
ical clinical aspects and discrete endoscopic findings. Schweiz Med
13. Clayton F, Fang JC, Gleich GJ, et al. Eosinophilic esophagitis in adults
Wochenschr. 1994;124:1419-1429.
is associated with IgG4 and not mediated by IgE. Gastroenterology.
3. Furuta GT, Liacouras CA, Collins MH, et al. Eosinophilic esophagitis in chil-
2014;147:602-609.
dren and adults: a systematic review and consensus recommendations
14. Lucendo AJ, Arias Á, González-Cervera J, Olalla JM, Molina-Infante
for diagnosis and treatment. Gastroenterology 2007;133:1342-1363.
J. Dual response to dietary/topical steroid and proton pump inhibi-
4. Liacouras CA, Furuta GT, Hirano I, et al. Eosinophilic esophagitis:
tor therapy in adult patients with eosinophilic esophagitis. J Allergy
updated consensus recommendations for children and adults. J
Clin Immunol. 2016;137:931-934.e2.
Allergy Clin Immunol. 2011;128:3-20.e6; quiz 21-2.
5. Straumann A, Aceves SS, Blanchard C, et al. Pediatric and adult eosino-
philic esophagitis: similarities and differences. Allergy. 2012;67:477-490.
6. Tuft L, Blumstein GI. Studies in food allergy: II. Sensitization to fresh S U P P O R T I N G I N FO R M AT I O N
fruits: clinical and experimental observations. J Allergy Clin Immunol. Additional supporting information may be found online in the
1942;13:574-582. Supporting Information section.
7. Amlot PL, Kemeny DM, Zachary C, Parkes P, Lessof MH. Oral al-
lergy syndrome (OAS): symptoms of IgE-mediated hypersensitivity
to foods. Clin Exp Allergy. 1987;17:33-42.
8. Saunders S, Platt MP. Oral allergy syndrome. Curr Opin Otolaryngol How to cite this article: Biedermann L, Holbreich M, Atkins
Head Neck Surg. 2015;23:230-234. D, et al. Food-induced immediate response of the
9. Warners MJ, Terreehorst I, van den Wijngaard RM, et al. Abnormal re- esophagus—A newly identified syndrome in patients with
sponses to local esophageal food allergen injections in adult patients
eosinophilic esophagitis. Allergy. 2020;00:1–9. https://doi.
with eosinophilic esophagitis. Gastroenterology. 2018;154:57-60.e2.
10. Krarup AL, Villadsen GE, Mejlgaard E, Olesen SS, Drewes AM, org/10.1111/all.14495
Funch-Jensen P. Acid hypersensitivity in patients with eosinophilic
oesophagitis. Scand J Gastroenterol. 2010;45:273-281.