- Email: [email protected]
Efficacy of a 4-Food Elimination Diet for Children With Eosinophilic Esophagitis
Accepted Manuscript Efficacy of a 4-Food Elimination Diet for Children With Eosinophilic Esophagitis A.F. Kagalwalla, J.B. Wechsler, K. Amsden, S. Schwartz, M. Makhija, A. Olive, C.M. Davis, M. Manuel-Rubio, S. Marcus, M. Sulkowski, K. Johnson, J.N. Ross, M.E. Riffle, M. Groetch, H. Melin-Aldana, D. Schady, H. Palac, K.-Y.A. Kim, B.K. Wershil, M.H. Collins, M. Chehade
PII: DOI: Reference:
S1542-3565(17)30689-4 10.1016/j.cgh.2017.05.048 YJCGH 55281
To appear in: Clinical Gastroenterology and Hepatology Accepted Date: 30 May 2017 Please cite this article as: Kagalwalla A, Wechsler J, Amsden K, Schwartz S, Makhija M, Olive A, Davis C, Manuel-Rubio M, Marcus S, Sulkowski M, Johnson K, Ross J, Riffle M, Groetch M, Melin-Aldana H, Schady D, Palac H, Kim K-Y, Wershil B, Collins M, Chehade M, Efficacy of a 4-Food Elimination Diet for Children With Eosinophilic Esophagitis, Clinical Gastroenterology and Hepatology (2017), doi: 10.1016/ j.cgh.2017.05.048. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Efficacy of a 4-Food Elimination Diet for Children With Eosinophilic Esophagitis
RI PT
Kagalwalla AF,1,2,3 Wechsler JB,1,2 Amsden K,2 Schwartz S,2 Makhija M,1,4 Olive A,5 Davis CM,6 Manuel-Rubio M,2 Marcus S,7 Sulkowski, M, 2 Johnson K,2 Ross JN,2 Riffle ME,8 Groetch M,8 Melin-Aldana H,1,9 Schady D,10 Palac H,11 Kim K-YA,11Wershil BK,1,2 Collins MH,12 Chehade M8 1. Northwestern University Feinberg School of Medicine, Chicago, Illinois
SC
2. Eosinophilic Gastrointestinal Diseases Program, Division of Gastroenterology, Hepatology, and Nutrition, Department of Pediatrics, Ann & Robert H. Lurie Children's Hospital of Chicago, Chicago, Illinois
M AN U
3. John H Stroger Hospital of Cook County, Chicago, Illinois
4. Division of Allergy & Clinical Immunology, Department of Pediatrics, Ann & Robert H. Lurie Children's Hospital of Chicago, Chicago, Illinois 5. Section of Gastroenterology, Hepatology and Nutrition, Department of Pediatrics, Baylor College of Medicine, Texas Children's Hospital, Houston, Texas
TE D
6. Section of Immunology, Allergy and Rheumatology, Department of Pediatrics, Baylor College of Medicine, Texas Children's Hospital, Houston, Texas 7. GI Care for Kids, Atlanta, Georgia
8. Mount Sinai Center for Eosinophilic Disorders, Icahn School of Medicine at Mount Sinai, New York City, New York
EP
9. Department of Pathology and Laboratory Medicine, Ann & Lurie Children's Hospital of Chicago, Chicago, Illinois
AC C
10. Department of Pathology, Baylor College of Medicine, Texas Children's Hospital, Houston, Texas 11. Department of Preventive Medicine, Feinberg School of Medicine, Chicago, Illinois 12. Department of Pathology and Laboratory Medicine, Cincinnati Children’s Hospital Medical Center and University of Cincinnati College of Medicine, Cincinnati, Ohio
Short title: Empiric elimination diet for treating eosinophilic esophagitis
ACCEPTED MANUSCRIPT
Grant Support: This work was supported by the Buckeye Foundation, American Partnership for Eosinophilic Disorders (APFED) and Campaign Urging Research for Eosinophilic Disease (CURED).
RI PT
Disclosures: There are no potential conflicts of interest for any of the authors pertaining to this study. Author contributions (all have reviewed and approved the final draft of the manuscript) Kagalwalla: Study conception and design; data acquisition; analysis and interpretation; manuscript drafting; critical revision; funding acquisition.
SC
Wechsler: Data acquisition; data analysis; data interpretation; manuscript drafting; critical revision
M AN U
Amsden: Data acquisition; data interpretation; critical revision
Schwartz: Data acquisition; data interpretation; critical revision
Makhija: Allergy data acquisition; data interpretation; critical revision Olive: Data acquisition; data interpretation; critical revision
TE D
Davis: Data acquisition; data interpretation; critical revision
Manuel-Rubio: Data acquisition; data interpretation; critical revision Marcus: Data acquisition; data interpretation; critical revision
EP
Sulkowski: Data acquisition; critical revision
Johnson: Data acquisition; data interpretation; critical revision
AC C
Ross: Data acquisition; data interpretation; critical revision Riffle: Data acquisition; critical revision Groetch: Data acquisition; critical revision Schady: Data acquisition; critical revision Melin-Aldana: Data interpretation; critical revision Palac: Data interpretation; statistical analysis Kim: Data interpretation; statistical analysis; critical revision
ACCEPTED MANUSCRIPT
Wershil: Data acquisition; data interpretation; critical revision Collins: Data acquisition; data interpretation; critical revision
AC C
EP
TE D
M AN U
SC
RI PT
Chehade: Data acquisition; data interpretation; critical revision
ACCEPTED MANUSCRIPT
Corresponding Author: Amir F Kagalwalla, MBBS
Phone: 312-227-4200 Fax: 312-227-9645
AC C
EP
TE D
M AN U
E-mail: [email protected]
SC
Chicago, Illinois 60611-2991
RI PT
Ann & Robert H. Lurie Children's Hospital of Chicago
ACCEPTED MANUSCRIPT
Abstract Background & Aims: A 6-food elimination diet induces remission in most children and adults
RI PT
with eosinophilic esophagitis (EoE). The effectiveness of empiric elimination of only 4 foods has not been studied in children. We performed a prospective observational outcome study in children with EoE treated with dietary exclusion of cow's milk, wheat, egg, and soy. The
SC
objective was to assess the clinical, endoscopic, and histologic efficacy of this treatment in EoE.
Methods: We recruited children (1-18 years old, diagnosed per consensus guidelines) from 4
M AN U
medical centers. Study participants (N=78) were given a proton pump inhibitor twice daily and underwent a baseline esophagogastroduodenoscopy. Subjects were instructed on dietary exclusion of cow's milk, wheat, egg, and soy. Clinical, endoscopic, and histologic assessments were made after 8 weeks. Responders had single foods reintroduced for 8 weeks, with repeat
TE D
endoscopy to assess for recurrence of active disease. The primary endpoint was histologic remission (fewer than 15 eosinophils per high-powered field). Secondary endpoints included symptom and endoscopic improvements and identification of foods associated with active
EP
histologic disease.
AC C
Results: After 8-weeks on 4-food elimination diet, 50 subjects were in histologic remission (64%). The subjects' mean baseline clinical symptoms score was 4.5 which decreased to 2.3 after 8 weeks of 4-food elimination diet (p<0.001). The mean endoscopic baseline score was 2.1 which decreased to 1.3 (p<0.001). After food reintroduction, the most common food triggers that induced histological inflammation were cow's milk (85%), egg (35%), wheat (33%), and soy (19%). One food trigger that induced recurrence of esophageal inflammation was identified in 62% of patients and cow's milk-induced EoE was present in 88% of these patients.
ACCEPTED MANUSCRIPT
Conclusions: In a prospective study of children with EoE, 8 weeks of 4-food elimination diet induced clinical, endoscopic, and histological remission in more than 60% of children with EoE. While less restrictive than 6-food elimination diet, 4-food elimination diet was nearly as
AC C
EP
TE D
M AN U
SC
Keywords: SFED; esophagus; inflammation; trial
RI PT
effective, and can be recommended as a treatment for children with EoE.
ACCEPTED MANUSCRIPT
Introduction Eosinophilic esophagitis (EoE) is a chronic, immune-mediated inflammatory disease of the
RI PT
esophagus characterized clinically by symptoms of esophageal dysfunction and histologically by eosinophil-predominant inflammation.1,2 Kelly et al. demonstrated clinical and histological remission in children with isolated esophageal eosinophilia when treated with an exclusive
SC
amino-acid based formulation.3 Reintroduction of incriminating foods resulted in recurrence, thereby establishing that EoE is triggered by food antigen. Six food elimination diet (SFED)
M AN U
excluding cow's milk, wheat, egg, soy, peanut/tree nut, and fish/shell-fish demonstrated clinical and histological remission in 74% of children.4 Other retrospective pediatric and prospective and retrospective adult studies showed remission in 58-81% of children and adults treated with SFED.5-10 Cow's milk, wheat, egg, and soy were identified as the four foods most likely causing inflammation in EoE.5,6,11 A recently published four food group elimination diet study in Spanish
TE D
subjects demonstrated remission in 54% adults. 12 We investigated whether excluding fewer foods is effective in inducing clinical, endoscopic, and
AC C
EP
histological remission. We also wanted to identify the frequency of specific EoE trigger foods.
Materials and Methods: The institutional Review Board (IRB) approved this study at each of the following participating institutions: Lurie Children's Hospital of Chicago, Chicago Illinois, Mount Sinai Medical Center, New York NY, Texas Children's Hospital, Houston Texas, and GI Care for Kids, Atlanta Georgia. Parents or legal guardians provided written informed consent for participation. All
ACCEPTED MANUSCRIPT
investigators followed the same protocol detailing the study's design, inclusion criteria, endoscopic grading, allergy testing, and treatment endpoints. A manual of operations and 4-FED elimination diet patient handouts were developed and shared by all the centers. Subjects were
RI PT
recruited from outpatient gastroenterology clinics at the participating institutions. EoE cases, aged 1-18 years, were diagnosed per consensus guidelines,1 that included 1) symptoms of esophageal dysfunction, and 2) the presence of > 15 eosinophils per high power field (eos/hpf) in
SC
esophageal biopsies. Subjects were treated twice daily Proton Pump Inhibitor (PPI) for 8 weeks followed by baseline diagnostic esophagogastroduodenoscopy (EGD), and only those who
M AN U
demonstrated persistent esophageal eosinophilia were included in the study. Subjects previously treated with topical steroids underwent a baseline EGD off steroids for 3 months and were included on demonstrating esophageal eosinophilia. After diagnostic EGD, all subjects continued PPI once daily for the remainder of the study to treat co-morbid GERD and limit PPI as a
TE D
confounder between patients throughout the study. All EGD were performed under general anesthesia. Other medications for the treatment of asthma and allergic rhinitis including nasal steroids were continued unchanged throughout the study. Patients on systemic or topical steroids
EP
or those with eosinophilic gastroenteritis and eosinophilic colitis were excluded from the study,
AC C
as were patients with celiac disease, and inflammatory bowel disease. Study Endpoints
The primary study treatment end point was histological remission with eosinophil count <15 eos/hpf. Secondary endpoints included symptomatic and endoscopic improvement and identification of specific food triggers. Allergy Testing
ACCEPTED MANUSCRIPT
Food specific serum specific IgE (sIgE) ImmunoCAP testing and skin prick testing (SPT) to cow's milk, wheat, egg, and soy was performed. Positive tests were defined as sIgE > 0.35 KU/L
RI PT
and SPT > 3 mm maximal wheal diameter. Study Design
A dietitian instructed both parents and patients on proper food elimination, food label reading,
SC
appropriate food substitution, and avoiding cross contact with eliminated foods per protocol as shown in Supplementary Tables 1a/b. The dietitian's contact information was provided to
M AN U
address concerns about implementing the elimination diet. The dietitian also reviewed a threeday pretreatment diet log to assess the pretreatment diet in order to provide guidance about food substitutions when the four foods were excluded. Identification of contamination resulted in removing the contamination and delaying the post-treatment EGD by an additional 4 weeks.
TE D
Four foods were eliminated and endoscopic biopsies was performed eight weeks after the elimination diet to assess histological response. In responders defined as those who achieved <15 eos/hpf, eliminated foods were reintroduced. Single foods were introduced sequentially in
EP
the same order starting with soy, then egg, followed by wheat, and finally milk, with EGD performed at least eight weeks after every new food reintroduction. Immediate hypersensitivity-
AC C
inducing food was not introduced, moving instead to the next food in the protocol. If the reintroduced food did not result in the recurrence of eosinophilia (< 15 eos/hpf) it was kept in the diet and the next food was reintroduced. A trigger food was identified by the recurrence of inflammation (> 15 eos/hpf) with food reintroduction. The identified trigger food was again removed from the diet. Eight weeks after eliminating the trigger food repeat normal (<15
ACCEPTED MANUSCRIPT
eos/hpf) baseline EGD (defined as washout period) was performed. The next food was introduced after the washout EGD, and this study design is illustrated in Figure 1.
RI PT
Demographics and atopic medical history was collected at the time of recruitment. The presence of IgE-mediated food allergy was determined from history of immediate reaction to specific foods confirmed with food-specific allergy testing (IgE or SPT).
SC
Symptom Data Collection
Parents and subjects completed non-validated 17-question symptom instrument at the time of
M AN U
study entry (baseline) and at the time of each subsequent EGD. The 17 symptoms in the instrument are listed in Table 1. Each symptom was scored absent (0) or present (+1). The total symptom score was determined by adding the individual patient symptom scores.
TE D
Endoscopic Findings
Endoscopic features, including edema, rings, white plaques, linear furrows, and strictures were scored absent (0) or present (+1). The total (composite) endoscopic score was derived by adding
EP
the component scores for individual findings. The total scores ranged from 0 to 5.
AC C
Histological Analysis
Four biopsy samples for histological assessment were obtained from two different levels of the esophagus. Eosinophils were reported as peak eosinophil counts/hpf from the most dense areas of the esophagus. Presence of basal cell hyperplasia was also described. Gastric and duodenal biopsies were also obtained. Pathologists reviewed the biopsies at each of the participating institutions, and their assessment was used to define remission or identify trigger foods. The study's central pathologist (MHC) provided oversight by reviewing random biopsies from
ACCEPTED MANUSCRIPT
participating institutions. There was no inter-observer variability, and inter-observer consistency was identified between the participating pathologists and the central pathologist. Eosinophil count was assessed at high power magnification X 400 HPF (0.23mm2). The cutoff value of 15
RI PT
eos/hpf was used to evaluate remission. Anthropometric Measurements
SC
Z scores for weight, height, and BMI at baseline and each subsequent EGD were determined using the SAS program for CDC 2000 growth charts (2 to < 20 years).
M AN U
Statistical analysis
Changes in peak eos/hpf before and after 4-FED were evaluated using a two-sided paired t-test. Differences in baseline characteristics between treatment responders and non-responders were
TE D
assessed using a two-sample t-test, the Wilcoxon rank-sum test or the chi-square/Fisher's exact tests. Association between demographic and clinical characteristics of interest and treatment response was tested with logistic regression. The area under the curve (AUC) of the receiver
EP
operating characteristics (ROC) curve was calculated to assess the predictive value for the final model. Pair wise comparisons for each pair of geographic regions explored for significant
AC C
omnibus test using two-sample t-tests, Wilcoxon rank sum tests, chi-square, or Fisher's exact tests as appropriate using Bonferroni corrections. Sensitivity, specificity, negative predictive value (NPV), and positive predictive value (PPV) were calculated for each food to evaluate the validity of SPT and serum sIgE allergy test compared to 4-FED identified trigger foods. Changes in standardized height, weight, and BMI z-scores was assessed between responders and nonresponders using two-sample t-tests. All statistical analyses were performed using SAS, version
ACCEPTED MANUSCRIPT
9.4 (SAS Institute, Cary, NC). Unless otherwise specified, alpha =0.05 was used to determine
RI PT
statistical significance.
Results
SC
Patients
Between 2011 and 2016, 96 children consented to participate and of these 18 were excluded: 5
M AN U
left prior to starting the diet, 8 found the diet too difficult, 1 dropped out for insurance reasons, 1 was lost to follow up and 3 did not meet the inclusion criteria (Figure 2). All 78 subjects (67% male, mean age 9.01, 83% white, 90% atopic) who met inclusion criteria were incident cases, except for 3 who had been previously treated with topical steroids, and are the basis of this study
TE D
Table 1.
Histological, Clinical and Endoscopic Responses
EP
Histological remission was achieved in 50/78 (64%) subjects. The peak eosinophil count in the proximal esophagus decreased from 43+32 and 3+3 per hpf and in the distal esophagus
AC C
decreased from 55+34 to 4+4 per hpf with 4-FED treatment (Table 3). Basal cell hyperplasia resolved in 36% of the responders (p<0.001) and persisted in all the non-responders. Food contamination was not identified in any patients prior to the scheduled post four food elimination EGD.
All symptoms resolved in 36% of respondents and symptom score decreased in 91%. The mean baseline symptom score decreased from 4.5 to 2.3 in responders (p< 0.001). Table 2 shows specific symptoms that resolved in a significant number of 4-FED treatment responders:
ACCEPTED MANUSCRIPT
abdominal pain (p<0.05), poor appetite (p =0.01), vomiting (p=0.01), food impaction (p<0.05), choking/gagging (p=0.01), regurgitation (p=0.01), pocket/spit out food (p=0.01). Interestingly non-responders had resolution of chest pain (p<0.05) and atypical symptoms such as early satiety
RI PT
(p<0.05), and nausea (p=0.5).
The mean endoscopic baseline score decreased from 2.1 to 1.3 (p< 0.001) in the responders. The
SC
changes in specific endoscopic findings are shown in Table 2. Exudates resolved in 96% respondents (p < 0.001). Edema resolved in 66% (p <0.001) and furrows in 62% (p < 0.001).
M AN U
Rings persisted in all subjects who had them at baseline.
Mean weight and BMI z-scores in responders decreased 0.2 + 0.43 and 0.31+-0.56 respectively from baseline values. On completion of food reintroduction process weight and BMI z-scores showed increase from baseline scores.
TE D
Baseline characteristics of patients showed no differences between responders and nonresponders regarding demographics, symptoms, endoscopic findings, eosinophilic density in biopsies but there were more female responders (p = 0.003), more asthmatics responders (p =
EP
0.041), higher proportion with family history of allergy in non-responders (p = 0.002) and higher proportion of non-responders with food specific IgE sensitization to wheat (p=0.010), egg (p
AC C
=0.001) and soy (p = 0.007). At baseline all included subjects consumed all four foods except 5 subjects with anaphylactic food allergies who avoided milk (1), egg (3) and/or soy (1). These foods were not reintroduced during the reintroduction phase. Forty-seven subjects reintroduced between 1-4 foods, and the distribution of the trigger foods in these subjects is shown in Figure 3a. Twenty-five of the 47 subjects were challenged with all foods except foods avoided at baseline (egg 2 and soy 1). Distribution of specific food triggers in
ACCEPTED MANUSCRIPT
these 25 subjects is shown in Figure 3b. In this group of 25 the distribution of the number of identified food triggers is shown in figure 3c. From responders who introduced all foods single food trigger was identified in 16 patients and included 14 (88%) to cow's milk, one each to soy
RI PT
(6%) and wheat (6%). Two patients reintroduced all 4 foods without reactivity.
Since histologic, symptom, endoscopic scores decreased after treatment the correlation between
SC
histology, symptoms and endoscopic and weight z-scores was analyzed at multiple time points that included baseline, after 4-FED, and after each food reintroduction as shown in Figure 4 a, b,
M AN U
c, and d.
The utility of specific sIgE and SPT testing to identify EoE food triggers was assessed in treatment responders. Sensitivity, specificity, NPV and PPV were calculated to cow's milk, wheat egg and soy, and the results suggest that serum sIgE and SPT are not useful in identifying
TE D
EoE food triggers, shown in Supplementary Table 2. More responders had post treatment EGD performed during the pollen season than the non-responders (70% vs 50% p =0.09). Thus pollen season did not influence the responses.
EP
Family history of food allergy (OR=0.21, 95% CI=[0.05, 0.90]) and positive serum sIgE test to any of the four foods (OR=0.15, 95% CI=[0.03, 0.90]) were significantly associated with
AC C
treatment non-response. Gender (OR=1.64; 95% CI=[0.31, 8.73]) and asthma (OR=4.89; 95% CI=[0.88, 27.23]) were not statistically associated with treatment response in the multivariable model. The AUC of the final model which included asthma, family history of food allergy, and positive serum sIgE to any of the four foods was 0.80 (95% CI, 0.67-0.93) shown in Supplementary Figure 1.
ACCEPTED MANUSCRIPT
Demographic, clinical, allergy testing, endoscopic, and histological characteristics in the three different regions is shown in Supplementary Table 3. A higher proportion of subjects from the
history of eczema (p=0.014) compared to the Midwest or South.
SC
Discussion
RI PT
East Coast had IgE-mediated food allergy (P=0.027), exudates (p=0.001) and a strong family
M AN U
This study demonstrates that exclusion of milk, wheat, egg, and soy for eight weeks induces clinical, endoscopic, and histological remission in a majority of children with EoE. This is the largest prospective, multi-centered outcome study to date evaluating the efficacy of an empiric elimination diet for the treatment of EoE in children or adults. It is additionally the first
TE D
prospective elimination diet study conducted in children with EoE.
Cow’s milk was the most common food trigger identified in 85% of patients, followed by egg (in 35% of patients), wheat (in 33% of patients), and soy (in 19% of patients) with the frequency of
EP
trigger foods that is identical to the SFED study.11 The number of subjects in whom milk was a food trigger is even higher in our study than previously reported11, thereby making cow’s milk
AC C
elimination the most significant component of any elimination diet therapy and providing the basis for milk only elimination diet approach. Thirty-six percent of responders had more than one food trigger, and this finding supports single food reintroduction process to accurately identify food triggers.
Symptom scores decreased in 91% of the histologic responders, with complete symptom resolution in 36% of the histological remitters. Symptom response, however, is an unreliable marker of mucosal healing, as has been previously documented, and histology remains the only
ACCEPTED MANUSCRIPT
reliable modality to assess treatment response.4,7,8,13 One or more abnormal endoscopic findings persisted in more than 50% of histologic responders, demonstrating, as previously reported that
RI PT
endoscopic findings are not a reliable biomarker for histologic disease activity.14 Elimination of multiple foods from the diet increases the risk for malnutrition and nutritional deficiencies.15,16 Nutritional guidance regarding proper food substitution, vitamin and mineral
SC
replacement to prevent malnutrition and nutritional deficiencies was provided at the outset. In spite of this, responders demonstrated temporary weight loss with subsequent catch up weight
M AN U
gain once reintroduction of foods was completed.
SFED is first line non-pharmacologic treatment for EoE in children and adults.2 Excluding 6-8 foods simultaneously, the long reintroduction process and, costs associated with lost work/school, and of multiple EGD, are limitations of this approach. The present study by being
TE D
less restrictive, requiring less time to complete food reintroduction and fewer EGD is likely to be accepted by patients and their families.
Female gender was identified as one of the statistically significant predictors of response in
EP
our subjects. This observation differs from adults, where a higher likelihood of response was reported in males.7 History of asthma was another positive predictor of response. Family history
AC C
of food allergies was a predictor of poor response as was the presence of food sensitization to 4FED foods based on serum sIgE testing with abnormal serum sIgE levels predicting lower response. This last finding has also been previously reported by Erwin et al,17 and supports the concept that EoE patients with IgE-mediated food allergy represent a unique EoE phenotype, as we have previously reported.18 Food sensitization appears to be a marker for a more broadly food-allergic phenotype which does not respond to 4-FED.
ACCEPTED MANUSCRIPT
The large cohort size and the multi-center design with children recruited from four large metropolitan areas in three different geographic regions of the country is the major strength of this study and reinforces the external validity of our results. Other strengths of this study include
RI PT
single food reintroduction and washout period when the trigger food is identified. This added step removes the likelihood of recurrence of inflammation being ascribed to factors other than
relative incidence of each antigen-triggering inflammation.
SC
true allergy to the trigger food. The identical order of food reintroduction removes bias about the
M AN U
Limitations of this study include non-randomization and the absence of a control group. There is potential selection bias by recruiting nonconsecutive patients and families motivated in elimination diet although, this mirrors clinical practice. A non-validated symptom instrument was used since at the time this study's inception a validated symptom instrument was not available and this is another potential limitation of this study. Another potential limitation of this
TE D
study is the applicability of this dietary approach to children in countries with dietary habits that are different and unique from those in United States.
EP
Twenty-four subjects who responded 4-FED dropped out before completing the reintroduction process identifying all possible trigger foods. This is emblematic of the difficulties of even
AC C
temporarily excluding four foods and undergoing multiple EGD. It highlights the fact that dietary therapy is not suited for every patient. Careful patient selection, based on detailed and realistic explanation of the constraints of the diet, the number of EGD required, and the length of time to complete the reintroduction process are essential for the success of this approach. In summary, this prospective multicenter empiric 4-FED diet study achieved clinical remission in 9 out of 10 histological respondents. Histological remission was achieved in more than six out
ACCEPTED MANUSCRIPT
of 10 children, which is nearly identical to remission with SFED but with a diet that is less stringent. Subsequent systematic food reintroduction identified cow’s milk as the most frequently identified trigger followed by wheat, egg, and soy. Thus, four-food empiric elimination diet has
RI PT
nearly comparable efficacy to SFED but is less restrictive and requires less time to complete the reintroduction. Our results provide evidence that 4-FED can be offered to children in preference to SFED and as an alternative to topical corticosteroid refractory patients for the treatment of
AC C
EP
TE D
M AN U
SC
EoE.
ACCEPTED MANUSCRIPT
References Furuta GT, Katzka DA. Eosinophilic Esophagitis. N Engl J Med 2015;373:1640-8.
2.
Liacouras CA, Furuta GT, Hirano I, et al. Eosinophilic esophagitis: updated consensus recommendations for children and adults. J Allergy Clin Immunol 2011;128:3-20 e6; quiz 21-2.
3.
Kelly KJ, Lazenby AJ, Rowe PC, et al. Eosinophilic esophagitis attributed to gastroesophageal reflux: improvement with an amino acid-based formula. Gastroenterology 1995;109:1503-12.
4.
Kagalwalla AF, Sentongo TA, Ritz S, et al. Effect of six-food elimination diet on clinical and histologic outcomes in eosinophilic esophagitis. Clin Gastroenterol Hepatol 2006;4:1097-102.
5.
Henderson CJ, Abonia JP, King EC, et al. Comparative dietary therapy effectiveness in remission of pediatric eosinophilic esophagitis. J Allergy Clin Immunol 2012;129:15708.
6.
Spergel JM, Brown-Whitehorn TF, Cianferoni A, et al. Identification of causative foods in children with eosinophilic esophagitis treated with an elimination diet. J Allergy Clin Immunol 2012;130:461-7 e5.
7.
Gonsalves N, Yang GY, Doerfler B, et al. Elimination diet effectively treats eosinophilic esophagitis in adults; food reintroduction identifies causative factors. Gastroenterology 2012;142:1451-9 e1; quiz e14-5.
8.
Lucendo AJ, Arias A, Gonzalez-Cervera J, et al. Empiric 6-food elimination diet induced and maintained prolonged remission in patients with adult eosinophilic esophagitis: a prospective study on the food cause of the disease. J Allergy Clin Immunol 2013;131:797-804.
9.
Wolf WA, Jerath MR, Sperry SL, et al. Dietary elimination therapy is an effective option for adults with eosinophilic esophagitis. Clin Gastroenterol Hepatol 2014;12:1272-9.
10.
Rodriguez-Sanchez J, Gomez Torrijos E, Lopez Viedma B, et al. Efficacy of IgE-targeted vs empiric six-food elimination diets for adult eosinophilic oesophagitis. Allergy 2014;69:936-42.
12.
13.
SC
M AN U
TE D
EP
AC C
11.
RI PT
1.
Kagalwalla AF, Shah A, Li BU, et al. Identification of specific foods responsible for inflammation in children with eosinophilic esophagitis successfully treated with empiric elimination diet. J Pediatr Gastroenterol Nutr 2011;53:145-9.
Molina-Infante J, Arias A, Barrio J, et al. Four-food group elimination diet for adult eosinophilic esophagitis: A prospective multicenter study. J Allergy Clin Immunol 2014;134:1093-9 e1. Safroneeva E, Straumann A, Coslovsky M, et al. Symptoms Have Modest Accuracy in Detecting Endoscopic and Histologic Remission in Adults With Eosinophilic sophagitis. Gastroenterology 2016;150:581-590 e4.
ACCEPTED MANUSCRIPT
van Rhijn BD, Verheij J, Smout AJ, et al. The Endoscopic Reference Score shows modest accuracy to predict histologic remission in adult patients with eosinophilic esophagitis. Neurogastroenterol Motil 2016;28:1714-1722.
15.
Christie L, Hine RJ, Parker JG, et al. Food allergies in children affect nutrient intake and growth. J Am Diet Assoc 2002;102:1648-51.
16.
Liu T, Howard RM, Mancini AJ, et al. Kwashiorkor in the United States: fad diets, perceived and true milk allergy, and nutritional ignorance. Arch Dermatol 2001;137:6306.
17.
Erwin EA, Kruszewski PG, Russo JM, et al. IgE antibodies and response to cow's milk elimination diet in pediatric eosinophilic esophagitis. J Allergy Clin Immunol 2016;138:625-628 e2.
18.
Pelz BJ, Wechsler JB, Amsden K, et al. IgE-associated food allergy alters the presentation of paediatric eosinophilic esophagitis. Clin Exp Allergy 2016;46:1431-1440.
AC C
EP
TE D
M AN U
SC
RI PT
14.
ACCEPTED MANUSCRIPT
Table & Figure Legend Table 1. Baseline demographic, clinical, and histologic characteristics of all subjects who underwent 4-FED.
RI PT
Table 2. Histologic, endoscopic and symptomatic response to 4-FED in responders and nonresponders
SC
Figure 1. Diagram outlining the order of food reintroduction in responders Single foods reintroduced every 8 weeks in 4-FED responders starting with soy, egg, wheat and milk. Inflammation inducing trigger foods removed followed by a normal baseline EGD demonstrating remission of inflammation before the next food reintroduction.
M AN U
Figure 2. Diagram of patient flow Ninety-six patients were screened for the study; 18 were excluded and 78 underwent upper endoscopy with biopsies after 4-FED. 50 demonstrated histologic remission. Figure 3. Food triggers identified in subjects reintroducing between 1-4 foods in the diet and in those who reintroduced all foods back. a) Food triggers in patients who reintroduced 1-4 foods,
TE D
b) Food triggers in 25 patients who completed reintroduction of all foods. Included: 3 subjects with anaphylaxis, to egg (2) and soy (1) who did not reintroduce these foods in their diet. c) Distribution by number of food triggers in 25 who reintroduced all foods. Figure 4. Comparison of histology with symptom response, endoscopic findings and weight z-scores.
EP
A) Eosinophil counts at 6 different time points: baseline, post 4-FED, after introduction of soy, wheat, egg and milk.
AC C
B) Symptom scores at 6 different time points: baseline, post 4-FED and after introduction of soy, wheat, egg and milk. C) Endoscopic scores at 6 different time points: baseline, post 4-FED, and after soy, wheat, egg and milk. D) Weight z-scores at 6 different time points: baseline, post 4-FED after soy, wheat, egg and milk.
ACCEPTED MANUSCRIPT
Table 1. N (%)
9.01 ± 4.88
EP
AC C
NonResponders n=28 (36%)
9.21 ± 4.9
8.66 ± 4.92
29 (58%) 21 (42%) 39 (78%)
23 (82.14%) 5 (17.86%) 26 (92.86%)
SC
52 (66.67%) 26 (33.33%) 65 (83.33%)
p-valuea
0.6376 0.0300* 0.1194
39 (50%) 33 (42.31%) 32 (41.03%) 27 (34.62%) 24 (30.77%) 25 (32.05%) 21 (26.92%) 20 (25.64%) 18 (23.08%) 19 (24.36%) 18 (23.08%) 18 (23.08%) 18 (23.08%) 10 (12.82%) 12 (15.38%) 8 (10.26%) 8 (10.26%)
25 (50%) 21 (42%) 21 (42%) 15 (30%) 15 (30%) 16 (32%) 12 (24%) 12 (24%) 12 (24%) 14 (28%) 9 (18%) 12 (24%) 13 (26%) 5 (10%) 6 (12%) 4 (8%) 5 (10%)
14 (50%) 12 (42.86%) 11 (39.29%) 12 (42.86%) 9 (32.14%) 9 (32.14%) 9 (32.14%) 8 (28.57%) 6 (21.43%) 5 (17.86%) 9 (32.14%) 6 (21.43%) 5 (17.86%) 5 (17.86%) 6 (21.43%) 4 (14.29%) 3 (10.71%)
0.9999 0.9414 0.8152 0.2522 0.8441 0.9897 0.4367 0.6574 0.7960 0.3168 0.1550 0.7960 0.4129 0.4814 0.3319 0.4478 0.9999
32 (41.03%) 24 (31.17%) 22 (28.21%) 14 (17.95%) 46 (58.97%)
20 (40%) 12 (24.49%) 18 (36%) 10 (20%) 29 (58%)
12 (42.86%) 12 (42.86%) 4 (14.29%) 4 (14.29%) 17 (60.71%)
0.8056 0.0942 0.0409* 0.5281 0.8152
55 (71.43%) 40 (51.95%) 41 (53.25%) 34 (44.16%) 31 (40.26%) 17 (22.08%)
32 (65.31%) 22 (44.9%) 26 (53.06%) 15 (30.61%) 18 (36.73%) 9 (18.37%)
23 (82.14%) 18 (64.29%) 15 (53.57%) 19 (67.86%) 13 (46.43%) 8 (28.57%)
0.1157 0.1014 0.9656 0.0015* 0.4041 0.2990
M AN U
TE D
Age (years), mean ± SD Gender Male Female White Presenting symptoms Abdominal Pain Slow Eating Early Satiety Dysphagia Vomiting Poor Appetite Feeding Difficulties Regurgitation Poor Weight Gain Food Impaction Nausea Pocketing food/Spitting after chewing Gagging/Choking Heartburn Chest Pain Odynophagia Nocturnal awakening due to symptoms Atopy Eczema IgE-mediated food allergy (n=77) Asthma Allergic conjunctivitis Allergic rhinitis Family History (n=77) Rhinitis Asthma GERD Food Allergy Eczema Hiatal Hernia
Responders n=50 (64%)
RI PT
Patients completing the 4-FED n=78
Variable
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
EoE 11 (14.29%) 6 (12.24%) 5 (17.86%) Stricture 9 (11.69%) 6 (12.24%) 3 (10.71%) Celiac 4 (5.19%) 4 (8.16%) 0 (0%) Recurrent Vomiting 4 (5.19%) 1 (2.04%) 3 (10.71%) Common food allergens on skin prick test Egg (n=54) 18 (33.33%) 10 (27.03%) 8 (47.06%) Milk (n=55) 13 (23.64%) 9 (24.32%) 4 (22.22%) Soy (n=54) 21 (38.89%) 12 (33.33%) 9 (50%) Wheat (n=53) 11 (20.75%) 8 (22.22%) 3 (17.65%) Food sensitization on serum specific IgE test Egg (n=53) 24 (45.28%) 11 (29.73%) 13 (81.25%) Milk (n=56) 31 (55.36%) 18 (47.37%) 13 (72.22%) Soy (n=55) 21 (38.18%) 10 (26.32%) 11 (64.71%) Wheat (n=53) 24 (45.28%) 13 (34.21%) 11 (73.33%) Endoscopy visual findings Furrows (n=76) 56 (73.68%) 34 (70.83%) 22 (78.57%) Edema (n=75) 53 (70.67%) 32 (66.67%) 21 (77.78%) Exudates (n=76) 41 (53.95%) 24 (50%) 17 (60.71%) Rings (n=76) 9 (11.84%) 5 (10.42%) 4 (14.29%) Strictures (n=76) 1 (1.32%) 0 (0%) 1 (3.57%) Baseline peak eosinophil count, median (IQR) 60 (40-100) 52.5 (40-85) 80 (48-100) Post-treatment peak eosinophil count, median (IQR) 8 (3-40) 5 (1-8) 51 (35.5-90) a P-values < .05 were considered statistically significant. Chi-square and Fisher's exact tests were used to assess differences in response groups for categorical variables. A two-sample t-test was used to compare age between groups and the nonparametric Wilcoxon rank-sum test was used to compare baseline peak EOS and post-treatment EOS.
0.5157 0.9999 0.2903 0.1337 0.1470 0.9999 0.2363 0.9999 0.0005* 0.0806 0.0068* 0.0100* 0.4599 0.3104 0.3660 0.7178 0.3684 0.1211 <.0001*
ACCEPTED MANUSCRIPT Table 2
Furrow Edema Exudate Rings Stricture Symptoms
AC C
EP
TE D
Abdominal pain Slow eating Early satiety Poor appetite Dysphagia Vomiting Food impaction Gagging/choking Regurgitation Poor weight gain Pockets/spits after chewing Feeding difficulties Nausea Chest pain Heartburn Nocturnal awakening Odynophagia
* * *
*
* * * * * *
*
RI PT
Visual Findings
* * *
SC
Peak Eos/hpf Distal Eos/hpf Proximal Eos/hpf
Non-responders (n=28) Pre Post mean (sd) mean (sd) p-value 74.4 (36.4) 62.1 (31.6) 0.15 58.3 (37.6) 51.5 (30.1) 0.48 63.7 (36.7) 44.8 (33.1) 0.03 count (%) count (%) p-value 22 (79%) 23 (82%) 1.00 21 (75%) 20 (74%) 1.00 17 (61%) 17 (61%) 1.00 4 (14%) 4 (14%) 1.00 1 (4%) 0 (0%) count (%) count (%) p-value 14 (50%) 11 (39%) 0.58 12 (43%) 6 (21%) 0.08 11 (39%) 5 (18%) 0.04 9 (32%) 5 (18%) 0.34 12 (43%) 9 (32%) 0.58 9 (32%) 6 (21%) 0.55 5 (18%) 3 (11%) 0.68 5 (18%) 4 (14%) 1.00 8 (29%) 2 (7%) 0.08 6 (21%) 2 (7%) 0.13 6 (21%) 2 (7%) 0.13 9 (32%) 5 (18%) 0.29 9 (32%) 2 (7%) 0.05 6 (21%) 0 (0%) 0.04 5 (18%) 3 (11%) 0.68 3 (11%) 1 (4%) 0.48 4 (14%) 1 (4%) 0.37
M AN U
Histology
Responders (n=50) Pre Post mean (sd) mean (sd) p-value 62.0 (33.7) 5.0 (4.1) <0.001 55.9 (33.9) 4.1 (3.9) <0.001 43.9 (32.4) 2.6 (3.3) <0.001 count (%) count (%) p-value 34 (71%) 17 (35%) <0.001 32 (67%) 12 (25%) <0.001 24 (50%) 1 (2%) <0.001 5 (10%) 6 (12%) 1.00 0 (0%) 0 (0%) count (%) count (%) p-value 25 (50%) 15 (30%) 0.03 21 (42%) 18 (36%) 0.55 21 (42%) 13 (26%) 0.06 16 (32%) 5 (10%) 0.01 15 (30%) 11 (22%) 0.45 15 (30%) 4 (8%) 0.01 14 (28%) 5 (10%) 0.04 13 (26%) 3 (6%) 0.01 12 (24%) 2 (4%) 0.01 12 (24%) 5 (10%) 0.07 12 (24%) 1 (2%) 0.01 12 (24%) 5 (10%) 0.12 9 (18%) 7 (14%) 0.77 6 (12%) 4 (8%) 0.68 5 (10%) 5 (10%) 1.00 5 (10%) 1 (2%) 0.13 4 (8%) 5 (10%) 1.00
*
* *
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
Supplementary Table 1a handout provided to families
Four Food Elimination Diet
•
• • •
• •
RI PT
•
SC
•
M AN U
•
This diet is free of milk, soy, egg, and wheat. Any or all of these restricted foods on the elimination diet may be an ingredient in many processed foods. They may be also used as preservatives or flavors. There is no substitute to careful reading of food labels in order to make certain that none of the four food allergens is contained in a given food. Ingredients in a given food can change without warning. So read labels carefully on all foods every time you shop and even for foods that you have bought before. Avoid buying foods from bulk bins there is highly likelihood of mislabeling the foods. There is also a high probability of cross contamination of foods sold in bulk bins. Foods served in a restaurant or fast food establishments may be cross contaminated with above restricted foods, and many foods contain hidden sources of antigens and thus caution should be exercised when eating out. Cross-contamination occurs when one food comes into contact with another food & their proteins mix. Each food then contains small amounts of the other food that we often can’t see. Even a trace amount of food from cross-contamination can cause an allergic reaction. When cooking at home, cook the allergy-free meal first, cover, and cook/prepare the rest of the meal. Make sure utensils and cookware are clean prior to use (thoroughly washed with soap and water). Make sure you wash your hands when preparing foods. When calling food manufacturers regarding possible cross-contamination, ask about dedicated lines vs. shared equipment. Shared equipment = cross contamination, avoid.
TE D
• •
Vitamins
Always check labels for allowed ingredients with vitamins
EP
Phlexy-Vits - www.shsna.com Kirkman® Children’s Multivitamin and Mineral Hypoallergenic Capsule - www.kirkmanlabs.com Tums® Calcium Supplement (does contain cornstarch for corn allergies) Nano VM (has a hypoallergenic vitamin for 1-3years, and 4-8 years) - www.solacenutrition.com Freeda Vitamins Vitalets - www.freedavitamins.com Hero Nutritionals®: Yummy Bears Multivitamin & minerals - www.heronutritionals.com Carlson for Kids vitamins: Chewable Vitamins and Minerals - www.carlsonlabs.com Nature’s Plus: Animal Parade Vitamins - www.naturesplus.com
AC C
• • • • • • • •
Elemental Formulas Can be ordered through your local pharmacy/home health care agency with a prescription or directly through manufacturer.
ACCEPTED MANUSCRIPT
Diet free of Milk, Wheat, Egg & Soy **always check labels, manufacturers can change ingredients** Foods Allowed Enriched rice milk, hemp, coconut, quinoa, cashew, almond, pea or oat milk In recipes: substitute fruit or vegetable juice, homemade soup stock, potato water Instead of butter on toast, use pure jelly, jam, honey or herb flavored olive oil Allowed margarines (milk/soy free*) Cheese: rice/pea protein based Yogurt: coconut milk and almond yogurt *may contain soy oil/soy lecithin Grains & flours Amaranth & amaranth flour Barley & barley flour Buckwheat & buckwheat flour Chickpea or garbanzo flour Millet & millet flour Oats & oat flour Potato starch & flour Quinoa & quinoa flour Rice & rice flour Rye flour Sago flour Tapioca starch & flour Wild rice & wild rice flour Breads & baked goods – check freezer section for ready made breads with allowed ingredients Baked goods & specialty baking mixes containing allowed foods from specialty shops. (Rice, brown rice, millet or tapioca) Homemade baked goods with allowed foods Pure corn tortillas, wheat free tortillas Crackers and snacks Plain potato chips, corn chips, rice chips Pure rye crisp crackers Rice cakes, rice crackers Plain popcorn Cereals Cream of Rice Oatmeal and oat bran (plain) Puffed rice Rolled Oats/grits Corn/buckwheat based cereal Homemade granola with allowed grains Pasta Brown rice pasta wild rice pasta corn pasta Rice noodles & pasta bean pasta quínoa pasta All plain, fresh, & frozen vegetables and their juices except: soy bean sprouts & mixed sprouts. Tomato sauce: check labels for milk, soy or wheat All plain, fresh & frozen fruits & their juices (avoid pie filling fruits, may contain wheat) All fresh or frozen, plain meat or poultry Avoid all deli meats, luncheon meats with allowed ingredients All, check breading, sauces and marinades for allowed ingredients Ener-G® Egg Replacer / egg substitute recipes (see handout on cooking guidelines) All plain bean or legumes and bean or legume dishes prepared with allowed foods, except soy. All, check nuts for wheat flour; avoid nuts from bulk bins All, including sunflower, sesame, cumin, poppy, pumpkin, flax, caraway, anise Oils: olive, soybean oil (not cold pressed, expeller pressed, or extruded oil), canola, sunflower,
AC C
EP
TE D
M AN U
Breads & cereals
SC
RI PT
Type of Food Milk & Dairy
Vegetables
Fruit Meat & poultry Fish & Shellfish Eggs Legumes Nuts Seeds Fats & oils
ACCEPTED MANUSCRIPT
Spices & herbs
safflower, coconut, corn or peanut Meat drippings and poultry fat Homemade gravy made with allowed foods Tahini Lard All pure fresh or dried herbs & spices
AC C
EP
TE D
M AN U
SC
RI PT
The information included in this document is for informational purposes only and is not intended to substitute in any way for medical education, training, treatment, advice, or diagnosis by a healthcare professional. A qualified healthcare professional should always be consulted before making any healthcare-related decision
ACCEPTED MANUSCRIPT
Supplementary Table 1b provided to families at the time 4-FED was discussed
Shopping Guide (4-FED)**ALWAYS CHECK LABELS AS INGREDIENTS CAN CHANGE**
Nut butters: - Sunflower butter - Peanuts & peanut butter - Tree nuts/nut butters
Seeds: -
Flax Sesame Pumpkin Sunflower Poppyseed Chia
Tahini (SoyFree)
Cereals:
Bratwurst,
hotdogs,sausage (meat/turkey)
Lean Meats:
AC C
- Oatmeal/Oats (wheat free) - Puffed rice - Wheat Free Granola - Cream of Rice - Brown rice cereal - Rice - Corn - Quinoa - Buckwheat
EP
(Wheat,soy,egg,milk free) Brown or white rice Corn Cornmeal/polenta Corn Tortilla Corn Chips Millet & millet flour Quinoa Teff Peas Potatoes (w/skin) Sweet potatoes (w/skin) Plantains Pasta: Corn Rice Lentil Legume Quinoa pasta Yucca
-
(canned, fresh or frozen)
Artichoke hearts Asparagus Beets Bok choy Broccoli Brussels sprouts Cabbage Carrots Cauliflower Celery Collard greens Cucumber Daikon Eggplant Endive Green beans Green onions Hearts of palm Jicama slices Kale Kohlrabi Leeks Mushrooms Mustard greens Okra Onions Pea pods Peppers (all varieties) Radishes Romaine Salsa Salad greens Scallions Spaghetti sauce Spinach Summer squash Swiss chard Tomato Tomato sauce Tomatillos Grape tomatoes Turnips Water chestnuts Watercress Zucchini
Poultry Pork Lamb Veal Beef Poultry
Deli Meat (packaged) (not from deli counter)
Apples Applesauce (no
Dairy Substitutes Fortified Milks: - Almond - Coconut - Hemp - Oat - Rice - Cashew - Quinoa (under 3 years old: recommend elemental formulas to replace milk in diet)
RI PT
Bagels/breads
- Azuki - Black - Garbanzo - Kidney - Pinto - White Bean/lentil soups Hummus Fish Shellfish
Fruits -unlimited
(canned, fresh, or frozen)
sugar added) Apricots Bananas Blackberries Blueberries Cantaloupe Cherries Grapefruit Grapes Honeydew Kiwi Mandarin oranges Mango Nectarine Orange Papaya Persimmons Peach Pear Pineapple Plum Pomegranate Prunes Raspberries Strawberries Tangerines Watermelon Individual fruit cups (in own juice) Dried fruit
SC
Canned beans:
amaranth flour
Vegetables - unlimited
TE D
Amaranth and
Proteins
M AN U
Whole Grains
Freeze Dried
Vegetables/Fruits
Margarines/spreads Yogurts: - Almond - Coconut
Cheese: - Rice based - Pea based
ACCEPTED MANUSCRIPT
Sweets/Treats: Allowed Carob Chips,
(chicken nuggets, fish nuggets, tator tots, French fries, macaroni and cheese)
Allergy free pizzas
fresh or dried –
McCormick Brand spices
Lemon/lime juice Mustard Fats/Oil: - Canola - Grape seed - Olive - Safflower - Sesame - Avocado - coconut Vinegars Beverages: - La Croix™ - Juice spritzers - Tea, hot/iced - Diet/regular pop - Gatorade - Vitamin water - 100% fruit juice (8floz/day limit)
TE D
Wheat free Breads
cookies, cereal bars 100% fruit sorbet 100% all fruit bars Hard candy Plain potato chips Rice crackers (plain) Rice cakes (plain) Popcorn (air popped or stovetop),plain Potato (sweet potato) chips, plain Corn chips Wheat free pretzel chips, pretzels
Avocado Garlic Herbs and spices,
RI PT
Allergy free kids foods
Miscellaneous
SC
Snacks/ Desserts
M AN U
Frozen Foods
AC C
EP
Sweeteners: - Sugar - Equal ® - Splenda - Honey - Jam/jelly Tupperware/baggies - Sports bottle w/straw
ACCEPTED MANUSCRIPT
Supplementary Table 2.
PPV 0.75 (0.33-1) 0.25 (0-0.67) 0.60 (0.17-1) 0.18 (0-0.41) 0.75 (0.45-1) 0.38 (0.04-0.71) 0.75 (0.33-1) 0.25 (0-0.55)
RI PT
NPV 0.14 (0-0.33) 0.76 (0.56-0.97) 0.77 (0.6-0.95) 0.81 (0.64-0.98) 0.10 (0-0.29) 0.73 (0.51-0.96) 0.70 (0.51-0.88) 0.84 (0.7-0.98)
Sensitivity 0.20 (0-0.4) 0.20 (0-0.55) 0.38 (0.04-0.71) 0.33 (0-0.71) 0.40 (0.15-0.65) 0.43 (0.06-0.8) 0.30 (0.02-0.58) 0.33 (0-0.71)
Specificity 0.67 (0.13-1) 0.81 (0.62-1) 0.89 (0.76-1) 0.65 (0.47-0.84) 0.33 (0-0.87) 0.69 (0.46-0.91) 0.94 (0.83-1) 0.78 (0.62-0.93)
SC
18 32 27 21 18 33 27 23
M AN U
N
TE D
Serum sIgE
p(95% CI)
EP
Skin Prick
4-FED Trigger Food Milk Wheat Egg Soy Milk Wheat Egg Soy
AC C
Allergy Test
ACCEPTED MANUSCRIPT
Supplementary Table 3. N (%) Midwest n=52 (66.7%)
Age (years), mean ± SD Gender Male Female White Presenting symptoms Abdominal Pain Slow Eating Early Satiety Dysphagia Vomiting Poor Appetite Feeding Difficulties Regurgitation Poor Weight Gain Food Impaction Nausea Pocketing food/Spitting after chewing Gagging/Choking Heartburn Chest Pain Odynophagia Nocturnal awakening due to symptoms Atopy Eczema IgE-mediated food allergy (n=77) Asthma Allergic conjunctivitis Allergic rhinitis Family History (n=77) Rhinitis Asthma GERD Food Allergy Eczema Hiatal Hernia EoE
9.11 ± 4.93
8.79 ± 4.87
11 (73.33%) 4 (26.67%) 15 (100%)
6 (54.55%) 5 (45.45%) 10 (90.91%)
p-value
0.9700 0.5954 0.0717
29 (55.77%) 21 (40.38%) 20 (38.46%) 18 (34.62%) 17 (32.69%) 17 (32.69%) 14 (26.92%) 13 (25%) 9 (17.31%) 13 (25%) 11 (21.15%) 13 (25%) 15 (28.85%) 8 (15.38%) 11 (21.15%) 3 (5.77%) 7 (13.46%)
4 (26.67%) 9 (60%) 8 (53.33%) 7 (46.67%) 3 (20%) 5 (33.33%) 5 (33.33%) 4 (26.67%) 6 (40%) 4 (26.67%) 5 (33.33%) 3 (20%) 1 (6.67%) 1 (6.67%) 1 (6.67%) 3 (20%) 1 (6.67%)
6 (54.55%) 3 (27.27%) 4 (36.36%) 2 (18.18%) 4 (36.36%) 3 (27.27%) 2 (18.18%) 3 (27.27%) 3 (27.27%) 2 (18.18%) 2 (18.18%) 2 (18.18%) 2 (18.18%) 1 (9.09%) 0 (0%) 2 (18.18%) 0 (0%)
0.1320 0.2208 0.5545 0.3206 0.6647 0.9999 0.7466 0.9999 0.1574 0.9999 0.6156 0.9999 0.2380 0.8772 0.1950 0.1144 0.5461
18 (34.62%) 12 (23.08%) 16 (30.77%) 9 (17.31%) 31 (59.62%)
10 (66.67%) 9 (60%) 1 (6.67%) 5 (33.33%) 10 (66.67%)
4 (36.36%) 3 (30%) 5 (45.45%) 0 (0%) 5 (45.45%)
0.0797 0.0269* 0.0669 0.1102 0.5470
38 (74.51%) 26 (50.98%) 27 (52.94%) 21 (41.18%) 16 (31.37%) 11 (21.57%) 6 (11.76%)
13 (86.67%) 8 (53.33%) 7 (46.67%) 8 (53.33%) 11 (73.33%) 2 (13.33%) 2 (13.33%)
4 (36.36%) 6 (54.55%) 7 (63.64%) 5 (45.45%) 4 (36.36%) 4 (36.36%) 3 (27.27%)
0.0187* 0.9703 0.6908 0.7035 0.0138* 0.4341 0.4231
M AN U
TE D
EP
AC C
South n=11 (14.1%)
8.84 ± 5.07
SC
35 (67.31%) 17 (32.69%) 40 (76.92%)
East Coast n=15 (19.2%)
RI PT
Variable
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Stricture 7 (13.73%) 2 (13.33%) 0 (0%) 0.5765 Celiac 1 (1.96%) 0 (0%) 3 (27.27%) 0.0173* Recurrent Vomiting 2 (3.92%) 1 (6.67%) 1 (9.09%) 0.4152 Common food allergens on skin prick test Egg (n=54) 9 (25%) 4 (44.44%) 5 (55.56%) 0.1617 Milk (n=55) 9 (24.32%) 2 (22.22%) 2 (22.22%) 0.9999 Soy (n=54) 12 (34.29%) 4 (44.44%) 5 (50%) 0.6559 Wheat (n=53) 6 (17.14%) 2 (22.22%) 3 (33.33%) 0.4772 Food sensitization on serum sIgE test Egg (n=53) 14 (40%) 8 (66.67%) 2 (33.33%) 0.2392 Milk (n=56) 19 (51.35%) 7 (58.33%) 5 (71.43%) 0.6176 Soy (n=55) 11 (31.43%) 7 (53.85%) 3 (42.86%) 0.3661 Wheat (n=53) 15 (42.86%) 6 (50%) 3 (50%) 0.9175 Endoscopy findings Furrows (n=76) 38 (73.08%) 13 (92.86%) 5 (50%) 0.0606 Edema (n=75) 36 (69.23%) 12 (92.31%) 5 (50%) 0.0881 Exudates (n=76) 26 (50%) 13 (92.86%) 2 (20%) 0.0012* Rings (n=76) 9 (17.31%) 0 (0%) 0 (0%) 0.1391 Strictures (n=76) 1 (1.92%) 0 (0%) 0 (0%) 0.9999 Baseline peak eosinophil count, median (IQR) 70 (42.5-100) 65 (29-100) 40 (25-60) 0.0347* Post-treatment peak eosinophil count, median (IQR) 8 (2-22.5) 35 (7-85) 6 (0-52) 0.0728 a P-values < .05 were considered statistically significant. Chi-square and Fisher's exact tests were used to assess differences by region for categorical variables. A two-sample t-test was used to compare age between regions and the nonparametric Wilcoxon rank-sum test was used to compare baseline peak EOS and post-treatment EOS.
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
Supplementary Material Four Food Elimination Diet for Eosinophilic Esophagitis in Children
RI PT
Amir F Kagalwalla; Joshua B Wechsler; Katie Amsden; Sally Schwartz; Melanie M Makhija; Anthony Olive; Carla M Davis; Maria Manuel-Rubio; Seth Marcus; Maureen Sulkowski; Kristin Johnson; Jessica N Ross; Mary Ellen Riffle; Marion Groetch; Hector Melin-Aldana; Deborah A Schady; Hannah Palac; Kwang-Youn A Kim; Barry K Wershil; Margaret H Collins; Mirna Chehade
SC
Supplementary Table 1 a & b. Elimination diet handouts provided to families prior to starting the 4-FED Resource lists of allergy-free food companies, elemental formula information, allergy-free vitamin information, support websites, allowed foods, menu planning, recipes, and shopping guides as well as trouble shooting tips for travelling, school and eating out.
M AN U
Supplementary Table 2. Final multivariable logistic regression model for treatment response based on history of asthma, family history of food allergy, positive serum sIgE and gender
Supplementary Table 3. Demographic, clinical, allergy testing, endoscopic, and histological characteristics compared in the three different regions.
AC C
EP
TE D
Supplementary Figure 1. ROC curve identifying factors associated with response with an AUC = 0.8 ROC curve for final multivariable model identifed asthma, family history of food allergy and positive serum sIgE test to any of the four foods as significantly associated with treatment non-response. The AUC of the final model was 0.8.
PII: DOI: Reference:
S1542-3565(17)30689-4 10.1016/j.cgh.2017.05.048 YJCGH 55281
To appear in: Clinical Gastroenterology and Hepatology Accepted Date: 30 May 2017 Please cite this article as: Kagalwalla A, Wechsler J, Amsden K, Schwartz S, Makhija M, Olive A, Davis C, Manuel-Rubio M, Marcus S, Sulkowski M, Johnson K, Ross J, Riffle M, Groetch M, Melin-Aldana H, Schady D, Palac H, Kim K-Y, Wershil B, Collins M, Chehade M, Efficacy of a 4-Food Elimination Diet for Children With Eosinophilic Esophagitis, Clinical Gastroenterology and Hepatology (2017), doi: 10.1016/ j.cgh.2017.05.048. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Efficacy of a 4-Food Elimination Diet for Children With Eosinophilic Esophagitis
RI PT
Kagalwalla AF,1,2,3 Wechsler JB,1,2 Amsden K,2 Schwartz S,2 Makhija M,1,4 Olive A,5 Davis CM,6 Manuel-Rubio M,2 Marcus S,7 Sulkowski, M, 2 Johnson K,2 Ross JN,2 Riffle ME,8 Groetch M,8 Melin-Aldana H,1,9 Schady D,10 Palac H,11 Kim K-YA,11Wershil BK,1,2 Collins MH,12 Chehade M8 1. Northwestern University Feinberg School of Medicine, Chicago, Illinois
SC
2. Eosinophilic Gastrointestinal Diseases Program, Division of Gastroenterology, Hepatology, and Nutrition, Department of Pediatrics, Ann & Robert H. Lurie Children's Hospital of Chicago, Chicago, Illinois
M AN U
3. John H Stroger Hospital of Cook County, Chicago, Illinois
4. Division of Allergy & Clinical Immunology, Department of Pediatrics, Ann & Robert H. Lurie Children's Hospital of Chicago, Chicago, Illinois 5. Section of Gastroenterology, Hepatology and Nutrition, Department of Pediatrics, Baylor College of Medicine, Texas Children's Hospital, Houston, Texas
TE D
6. Section of Immunology, Allergy and Rheumatology, Department of Pediatrics, Baylor College of Medicine, Texas Children's Hospital, Houston, Texas 7. GI Care for Kids, Atlanta, Georgia
8. Mount Sinai Center for Eosinophilic Disorders, Icahn School of Medicine at Mount Sinai, New York City, New York
EP
9. Department of Pathology and Laboratory Medicine, Ann & Lurie Children's Hospital of Chicago, Chicago, Illinois
AC C
10. Department of Pathology, Baylor College of Medicine, Texas Children's Hospital, Houston, Texas 11. Department of Preventive Medicine, Feinberg School of Medicine, Chicago, Illinois 12. Department of Pathology and Laboratory Medicine, Cincinnati Children’s Hospital Medical Center and University of Cincinnati College of Medicine, Cincinnati, Ohio
Short title: Empiric elimination diet for treating eosinophilic esophagitis
ACCEPTED MANUSCRIPT
Grant Support: This work was supported by the Buckeye Foundation, American Partnership for Eosinophilic Disorders (APFED) and Campaign Urging Research for Eosinophilic Disease (CURED).
RI PT
Disclosures: There are no potential conflicts of interest for any of the authors pertaining to this study. Author contributions (all have reviewed and approved the final draft of the manuscript) Kagalwalla: Study conception and design; data acquisition; analysis and interpretation; manuscript drafting; critical revision; funding acquisition.
SC
Wechsler: Data acquisition; data analysis; data interpretation; manuscript drafting; critical revision
M AN U
Amsden: Data acquisition; data interpretation; critical revision
Schwartz: Data acquisition; data interpretation; critical revision
Makhija: Allergy data acquisition; data interpretation; critical revision Olive: Data acquisition; data interpretation; critical revision
TE D
Davis: Data acquisition; data interpretation; critical revision
Manuel-Rubio: Data acquisition; data interpretation; critical revision Marcus: Data acquisition; data interpretation; critical revision
EP
Sulkowski: Data acquisition; critical revision
Johnson: Data acquisition; data interpretation; critical revision
AC C
Ross: Data acquisition; data interpretation; critical revision Riffle: Data acquisition; critical revision Groetch: Data acquisition; critical revision Schady: Data acquisition; critical revision Melin-Aldana: Data interpretation; critical revision Palac: Data interpretation; statistical analysis Kim: Data interpretation; statistical analysis; critical revision
ACCEPTED MANUSCRIPT
Wershil: Data acquisition; data interpretation; critical revision Collins: Data acquisition; data interpretation; critical revision
AC C
EP
TE D
M AN U
SC
RI PT
Chehade: Data acquisition; data interpretation; critical revision
ACCEPTED MANUSCRIPT
Corresponding Author: Amir F Kagalwalla, MBBS
Phone: 312-227-4200 Fax: 312-227-9645
AC C
EP
TE D
M AN U
E-mail: [email protected]
SC
Chicago, Illinois 60611-2991
RI PT
Ann & Robert H. Lurie Children's Hospital of Chicago
ACCEPTED MANUSCRIPT
Abstract Background & Aims: A 6-food elimination diet induces remission in most children and adults
RI PT
with eosinophilic esophagitis (EoE). The effectiveness of empiric elimination of only 4 foods has not been studied in children. We performed a prospective observational outcome study in children with EoE treated with dietary exclusion of cow's milk, wheat, egg, and soy. The
SC
objective was to assess the clinical, endoscopic, and histologic efficacy of this treatment in EoE.
Methods: We recruited children (1-18 years old, diagnosed per consensus guidelines) from 4
M AN U
medical centers. Study participants (N=78) were given a proton pump inhibitor twice daily and underwent a baseline esophagogastroduodenoscopy. Subjects were instructed on dietary exclusion of cow's milk, wheat, egg, and soy. Clinical, endoscopic, and histologic assessments were made after 8 weeks. Responders had single foods reintroduced for 8 weeks, with repeat
TE D
endoscopy to assess for recurrence of active disease. The primary endpoint was histologic remission (fewer than 15 eosinophils per high-powered field). Secondary endpoints included symptom and endoscopic improvements and identification of foods associated with active
EP
histologic disease.
AC C
Results: After 8-weeks on 4-food elimination diet, 50 subjects were in histologic remission (64%). The subjects' mean baseline clinical symptoms score was 4.5 which decreased to 2.3 after 8 weeks of 4-food elimination diet (p<0.001). The mean endoscopic baseline score was 2.1 which decreased to 1.3 (p<0.001). After food reintroduction, the most common food triggers that induced histological inflammation were cow's milk (85%), egg (35%), wheat (33%), and soy (19%). One food trigger that induced recurrence of esophageal inflammation was identified in 62% of patients and cow's milk-induced EoE was present in 88% of these patients.
ACCEPTED MANUSCRIPT
Conclusions: In a prospective study of children with EoE, 8 weeks of 4-food elimination diet induced clinical, endoscopic, and histological remission in more than 60% of children with EoE. While less restrictive than 6-food elimination diet, 4-food elimination diet was nearly as
AC C
EP
TE D
M AN U
SC
Keywords: SFED; esophagus; inflammation; trial
RI PT
effective, and can be recommended as a treatment for children with EoE.
ACCEPTED MANUSCRIPT
Introduction Eosinophilic esophagitis (EoE) is a chronic, immune-mediated inflammatory disease of the
RI PT
esophagus characterized clinically by symptoms of esophageal dysfunction and histologically by eosinophil-predominant inflammation.1,2 Kelly et al. demonstrated clinical and histological remission in children with isolated esophageal eosinophilia when treated with an exclusive
SC
amino-acid based formulation.3 Reintroduction of incriminating foods resulted in recurrence, thereby establishing that EoE is triggered by food antigen. Six food elimination diet (SFED)
M AN U
excluding cow's milk, wheat, egg, soy, peanut/tree nut, and fish/shell-fish demonstrated clinical and histological remission in 74% of children.4 Other retrospective pediatric and prospective and retrospective adult studies showed remission in 58-81% of children and adults treated with SFED.5-10 Cow's milk, wheat, egg, and soy were identified as the four foods most likely causing inflammation in EoE.5,6,11 A recently published four food group elimination diet study in Spanish
TE D
subjects demonstrated remission in 54% adults. 12 We investigated whether excluding fewer foods is effective in inducing clinical, endoscopic, and
AC C
EP
histological remission. We also wanted to identify the frequency of specific EoE trigger foods.
Materials and Methods: The institutional Review Board (IRB) approved this study at each of the following participating institutions: Lurie Children's Hospital of Chicago, Chicago Illinois, Mount Sinai Medical Center, New York NY, Texas Children's Hospital, Houston Texas, and GI Care for Kids, Atlanta Georgia. Parents or legal guardians provided written informed consent for participation. All
ACCEPTED MANUSCRIPT
investigators followed the same protocol detailing the study's design, inclusion criteria, endoscopic grading, allergy testing, and treatment endpoints. A manual of operations and 4-FED elimination diet patient handouts were developed and shared by all the centers. Subjects were
RI PT
recruited from outpatient gastroenterology clinics at the participating institutions. EoE cases, aged 1-18 years, were diagnosed per consensus guidelines,1 that included 1) symptoms of esophageal dysfunction, and 2) the presence of > 15 eosinophils per high power field (eos/hpf) in
SC
esophageal biopsies. Subjects were treated twice daily Proton Pump Inhibitor (PPI) for 8 weeks followed by baseline diagnostic esophagogastroduodenoscopy (EGD), and only those who
M AN U
demonstrated persistent esophageal eosinophilia were included in the study. Subjects previously treated with topical steroids underwent a baseline EGD off steroids for 3 months and were included on demonstrating esophageal eosinophilia. After diagnostic EGD, all subjects continued PPI once daily for the remainder of the study to treat co-morbid GERD and limit PPI as a
TE D
confounder between patients throughout the study. All EGD were performed under general anesthesia. Other medications for the treatment of asthma and allergic rhinitis including nasal steroids were continued unchanged throughout the study. Patients on systemic or topical steroids
EP
or those with eosinophilic gastroenteritis and eosinophilic colitis were excluded from the study,
AC C
as were patients with celiac disease, and inflammatory bowel disease. Study Endpoints
The primary study treatment end point was histological remission with eosinophil count <15 eos/hpf. Secondary endpoints included symptomatic and endoscopic improvement and identification of specific food triggers. Allergy Testing
ACCEPTED MANUSCRIPT
Food specific serum specific IgE (sIgE) ImmunoCAP testing and skin prick testing (SPT) to cow's milk, wheat, egg, and soy was performed. Positive tests were defined as sIgE > 0.35 KU/L
RI PT
and SPT > 3 mm maximal wheal diameter. Study Design
A dietitian instructed both parents and patients on proper food elimination, food label reading,
SC
appropriate food substitution, and avoiding cross contact with eliminated foods per protocol as shown in Supplementary Tables 1a/b. The dietitian's contact information was provided to
M AN U
address concerns about implementing the elimination diet. The dietitian also reviewed a threeday pretreatment diet log to assess the pretreatment diet in order to provide guidance about food substitutions when the four foods were excluded. Identification of contamination resulted in removing the contamination and delaying the post-treatment EGD by an additional 4 weeks.
TE D
Four foods were eliminated and endoscopic biopsies was performed eight weeks after the elimination diet to assess histological response. In responders defined as those who achieved <15 eos/hpf, eliminated foods were reintroduced. Single foods were introduced sequentially in
EP
the same order starting with soy, then egg, followed by wheat, and finally milk, with EGD performed at least eight weeks after every new food reintroduction. Immediate hypersensitivity-
AC C
inducing food was not introduced, moving instead to the next food in the protocol. If the reintroduced food did not result in the recurrence of eosinophilia (< 15 eos/hpf) it was kept in the diet and the next food was reintroduced. A trigger food was identified by the recurrence of inflammation (> 15 eos/hpf) with food reintroduction. The identified trigger food was again removed from the diet. Eight weeks after eliminating the trigger food repeat normal (<15
ACCEPTED MANUSCRIPT
eos/hpf) baseline EGD (defined as washout period) was performed. The next food was introduced after the washout EGD, and this study design is illustrated in Figure 1.
RI PT
Demographics and atopic medical history was collected at the time of recruitment. The presence of IgE-mediated food allergy was determined from history of immediate reaction to specific foods confirmed with food-specific allergy testing (IgE or SPT).
SC
Symptom Data Collection
Parents and subjects completed non-validated 17-question symptom instrument at the time of
M AN U
study entry (baseline) and at the time of each subsequent EGD. The 17 symptoms in the instrument are listed in Table 1. Each symptom was scored absent (0) or present (+1). The total symptom score was determined by adding the individual patient symptom scores.
TE D
Endoscopic Findings
Endoscopic features, including edema, rings, white plaques, linear furrows, and strictures were scored absent (0) or present (+1). The total (composite) endoscopic score was derived by adding
EP
the component scores for individual findings. The total scores ranged from 0 to 5.
AC C
Histological Analysis
Four biopsy samples for histological assessment were obtained from two different levels of the esophagus. Eosinophils were reported as peak eosinophil counts/hpf from the most dense areas of the esophagus. Presence of basal cell hyperplasia was also described. Gastric and duodenal biopsies were also obtained. Pathologists reviewed the biopsies at each of the participating institutions, and their assessment was used to define remission or identify trigger foods. The study's central pathologist (MHC) provided oversight by reviewing random biopsies from
ACCEPTED MANUSCRIPT
participating institutions. There was no inter-observer variability, and inter-observer consistency was identified between the participating pathologists and the central pathologist. Eosinophil count was assessed at high power magnification X 400 HPF (0.23mm2). The cutoff value of 15
RI PT
eos/hpf was used to evaluate remission. Anthropometric Measurements
SC
Z scores for weight, height, and BMI at baseline and each subsequent EGD were determined using the SAS program for CDC 2000 growth charts (2 to < 20 years).
M AN U
Statistical analysis
Changes in peak eos/hpf before and after 4-FED were evaluated using a two-sided paired t-test. Differences in baseline characteristics between treatment responders and non-responders were
TE D
assessed using a two-sample t-test, the Wilcoxon rank-sum test or the chi-square/Fisher's exact tests. Association between demographic and clinical characteristics of interest and treatment response was tested with logistic regression. The area under the curve (AUC) of the receiver
EP
operating characteristics (ROC) curve was calculated to assess the predictive value for the final model. Pair wise comparisons for each pair of geographic regions explored for significant
AC C
omnibus test using two-sample t-tests, Wilcoxon rank sum tests, chi-square, or Fisher's exact tests as appropriate using Bonferroni corrections. Sensitivity, specificity, negative predictive value (NPV), and positive predictive value (PPV) were calculated for each food to evaluate the validity of SPT and serum sIgE allergy test compared to 4-FED identified trigger foods. Changes in standardized height, weight, and BMI z-scores was assessed between responders and nonresponders using two-sample t-tests. All statistical analyses were performed using SAS, version
ACCEPTED MANUSCRIPT
9.4 (SAS Institute, Cary, NC). Unless otherwise specified, alpha =0.05 was used to determine
RI PT
statistical significance.
Results
SC
Patients
Between 2011 and 2016, 96 children consented to participate and of these 18 were excluded: 5
M AN U
left prior to starting the diet, 8 found the diet too difficult, 1 dropped out for insurance reasons, 1 was lost to follow up and 3 did not meet the inclusion criteria (Figure 2). All 78 subjects (67% male, mean age 9.01, 83% white, 90% atopic) who met inclusion criteria were incident cases, except for 3 who had been previously treated with topical steroids, and are the basis of this study
TE D
Table 1.
Histological, Clinical and Endoscopic Responses
EP
Histological remission was achieved in 50/78 (64%) subjects. The peak eosinophil count in the proximal esophagus decreased from 43+32 and 3+3 per hpf and in the distal esophagus
AC C
decreased from 55+34 to 4+4 per hpf with 4-FED treatment (Table 3). Basal cell hyperplasia resolved in 36% of the responders (p<0.001) and persisted in all the non-responders. Food contamination was not identified in any patients prior to the scheduled post four food elimination EGD.
All symptoms resolved in 36% of respondents and symptom score decreased in 91%. The mean baseline symptom score decreased from 4.5 to 2.3 in responders (p< 0.001). Table 2 shows specific symptoms that resolved in a significant number of 4-FED treatment responders:
ACCEPTED MANUSCRIPT
abdominal pain (p<0.05), poor appetite (p =0.01), vomiting (p=0.01), food impaction (p<0.05), choking/gagging (p=0.01), regurgitation (p=0.01), pocket/spit out food (p=0.01). Interestingly non-responders had resolution of chest pain (p<0.05) and atypical symptoms such as early satiety
RI PT
(p<0.05), and nausea (p=0.5).
The mean endoscopic baseline score decreased from 2.1 to 1.3 (p< 0.001) in the responders. The
SC
changes in specific endoscopic findings are shown in Table 2. Exudates resolved in 96% respondents (p < 0.001). Edema resolved in 66% (p <0.001) and furrows in 62% (p < 0.001).
M AN U
Rings persisted in all subjects who had them at baseline.
Mean weight and BMI z-scores in responders decreased 0.2 + 0.43 and 0.31+-0.56 respectively from baseline values. On completion of food reintroduction process weight and BMI z-scores showed increase from baseline scores.
TE D
Baseline characteristics of patients showed no differences between responders and nonresponders regarding demographics, symptoms, endoscopic findings, eosinophilic density in biopsies but there were more female responders (p = 0.003), more asthmatics responders (p =
EP
0.041), higher proportion with family history of allergy in non-responders (p = 0.002) and higher proportion of non-responders with food specific IgE sensitization to wheat (p=0.010), egg (p
AC C
=0.001) and soy (p = 0.007). At baseline all included subjects consumed all four foods except 5 subjects with anaphylactic food allergies who avoided milk (1), egg (3) and/or soy (1). These foods were not reintroduced during the reintroduction phase. Forty-seven subjects reintroduced between 1-4 foods, and the distribution of the trigger foods in these subjects is shown in Figure 3a. Twenty-five of the 47 subjects were challenged with all foods except foods avoided at baseline (egg 2 and soy 1). Distribution of specific food triggers in
ACCEPTED MANUSCRIPT
these 25 subjects is shown in Figure 3b. In this group of 25 the distribution of the number of identified food triggers is shown in figure 3c. From responders who introduced all foods single food trigger was identified in 16 patients and included 14 (88%) to cow's milk, one each to soy
RI PT
(6%) and wheat (6%). Two patients reintroduced all 4 foods without reactivity.
Since histologic, symptom, endoscopic scores decreased after treatment the correlation between
SC
histology, symptoms and endoscopic and weight z-scores was analyzed at multiple time points that included baseline, after 4-FED, and after each food reintroduction as shown in Figure 4 a, b,
M AN U
c, and d.
The utility of specific sIgE and SPT testing to identify EoE food triggers was assessed in treatment responders. Sensitivity, specificity, NPV and PPV were calculated to cow's milk, wheat egg and soy, and the results suggest that serum sIgE and SPT are not useful in identifying
TE D
EoE food triggers, shown in Supplementary Table 2. More responders had post treatment EGD performed during the pollen season than the non-responders (70% vs 50% p =0.09). Thus pollen season did not influence the responses.
EP
Family history of food allergy (OR=0.21, 95% CI=[0.05, 0.90]) and positive serum sIgE test to any of the four foods (OR=0.15, 95% CI=[0.03, 0.90]) were significantly associated with
AC C
treatment non-response. Gender (OR=1.64; 95% CI=[0.31, 8.73]) and asthma (OR=4.89; 95% CI=[0.88, 27.23]) were not statistically associated with treatment response in the multivariable model. The AUC of the final model which included asthma, family history of food allergy, and positive serum sIgE to any of the four foods was 0.80 (95% CI, 0.67-0.93) shown in Supplementary Figure 1.
ACCEPTED MANUSCRIPT
Demographic, clinical, allergy testing, endoscopic, and histological characteristics in the three different regions is shown in Supplementary Table 3. A higher proportion of subjects from the
history of eczema (p=0.014) compared to the Midwest or South.
SC
Discussion
RI PT
East Coast had IgE-mediated food allergy (P=0.027), exudates (p=0.001) and a strong family
M AN U
This study demonstrates that exclusion of milk, wheat, egg, and soy for eight weeks induces clinical, endoscopic, and histological remission in a majority of children with EoE. This is the largest prospective, multi-centered outcome study to date evaluating the efficacy of an empiric elimination diet for the treatment of EoE in children or adults. It is additionally the first
TE D
prospective elimination diet study conducted in children with EoE.
Cow’s milk was the most common food trigger identified in 85% of patients, followed by egg (in 35% of patients), wheat (in 33% of patients), and soy (in 19% of patients) with the frequency of
EP
trigger foods that is identical to the SFED study.11 The number of subjects in whom milk was a food trigger is even higher in our study than previously reported11, thereby making cow’s milk
AC C
elimination the most significant component of any elimination diet therapy and providing the basis for milk only elimination diet approach. Thirty-six percent of responders had more than one food trigger, and this finding supports single food reintroduction process to accurately identify food triggers.
Symptom scores decreased in 91% of the histologic responders, with complete symptom resolution in 36% of the histological remitters. Symptom response, however, is an unreliable marker of mucosal healing, as has been previously documented, and histology remains the only
ACCEPTED MANUSCRIPT
reliable modality to assess treatment response.4,7,8,13 One or more abnormal endoscopic findings persisted in more than 50% of histologic responders, demonstrating, as previously reported that
RI PT
endoscopic findings are not a reliable biomarker for histologic disease activity.14 Elimination of multiple foods from the diet increases the risk for malnutrition and nutritional deficiencies.15,16 Nutritional guidance regarding proper food substitution, vitamin and mineral
SC
replacement to prevent malnutrition and nutritional deficiencies was provided at the outset. In spite of this, responders demonstrated temporary weight loss with subsequent catch up weight
M AN U
gain once reintroduction of foods was completed.
SFED is first line non-pharmacologic treatment for EoE in children and adults.2 Excluding 6-8 foods simultaneously, the long reintroduction process and, costs associated with lost work/school, and of multiple EGD, are limitations of this approach. The present study by being
TE D
less restrictive, requiring less time to complete food reintroduction and fewer EGD is likely to be accepted by patients and their families.
Female gender was identified as one of the statistically significant predictors of response in
EP
our subjects. This observation differs from adults, where a higher likelihood of response was reported in males.7 History of asthma was another positive predictor of response. Family history
AC C
of food allergies was a predictor of poor response as was the presence of food sensitization to 4FED foods based on serum sIgE testing with abnormal serum sIgE levels predicting lower response. This last finding has also been previously reported by Erwin et al,17 and supports the concept that EoE patients with IgE-mediated food allergy represent a unique EoE phenotype, as we have previously reported.18 Food sensitization appears to be a marker for a more broadly food-allergic phenotype which does not respond to 4-FED.
ACCEPTED MANUSCRIPT
The large cohort size and the multi-center design with children recruited from four large metropolitan areas in three different geographic regions of the country is the major strength of this study and reinforces the external validity of our results. Other strengths of this study include
RI PT
single food reintroduction and washout period when the trigger food is identified. This added step removes the likelihood of recurrence of inflammation being ascribed to factors other than
relative incidence of each antigen-triggering inflammation.
SC
true allergy to the trigger food. The identical order of food reintroduction removes bias about the
M AN U
Limitations of this study include non-randomization and the absence of a control group. There is potential selection bias by recruiting nonconsecutive patients and families motivated in elimination diet although, this mirrors clinical practice. A non-validated symptom instrument was used since at the time this study's inception a validated symptom instrument was not available and this is another potential limitation of this study. Another potential limitation of this
TE D
study is the applicability of this dietary approach to children in countries with dietary habits that are different and unique from those in United States.
EP
Twenty-four subjects who responded 4-FED dropped out before completing the reintroduction process identifying all possible trigger foods. This is emblematic of the difficulties of even
AC C
temporarily excluding four foods and undergoing multiple EGD. It highlights the fact that dietary therapy is not suited for every patient. Careful patient selection, based on detailed and realistic explanation of the constraints of the diet, the number of EGD required, and the length of time to complete the reintroduction process are essential for the success of this approach. In summary, this prospective multicenter empiric 4-FED diet study achieved clinical remission in 9 out of 10 histological respondents. Histological remission was achieved in more than six out
ACCEPTED MANUSCRIPT
of 10 children, which is nearly identical to remission with SFED but with a diet that is less stringent. Subsequent systematic food reintroduction identified cow’s milk as the most frequently identified trigger followed by wheat, egg, and soy. Thus, four-food empiric elimination diet has
RI PT
nearly comparable efficacy to SFED but is less restrictive and requires less time to complete the reintroduction. Our results provide evidence that 4-FED can be offered to children in preference to SFED and as an alternative to topical corticosteroid refractory patients for the treatment of
AC C
EP
TE D
M AN U
SC
EoE.
ACCEPTED MANUSCRIPT
References Furuta GT, Katzka DA. Eosinophilic Esophagitis. N Engl J Med 2015;373:1640-8.
2.
Liacouras CA, Furuta GT, Hirano I, et al. Eosinophilic esophagitis: updated consensus recommendations for children and adults. J Allergy Clin Immunol 2011;128:3-20 e6; quiz 21-2.
3.
Kelly KJ, Lazenby AJ, Rowe PC, et al. Eosinophilic esophagitis attributed to gastroesophageal reflux: improvement with an amino acid-based formula. Gastroenterology 1995;109:1503-12.
4.
Kagalwalla AF, Sentongo TA, Ritz S, et al. Effect of six-food elimination diet on clinical and histologic outcomes in eosinophilic esophagitis. Clin Gastroenterol Hepatol 2006;4:1097-102.
5.
Henderson CJ, Abonia JP, King EC, et al. Comparative dietary therapy effectiveness in remission of pediatric eosinophilic esophagitis. J Allergy Clin Immunol 2012;129:15708.
6.
Spergel JM, Brown-Whitehorn TF, Cianferoni A, et al. Identification of causative foods in children with eosinophilic esophagitis treated with an elimination diet. J Allergy Clin Immunol 2012;130:461-7 e5.
7.
Gonsalves N, Yang GY, Doerfler B, et al. Elimination diet effectively treats eosinophilic esophagitis in adults; food reintroduction identifies causative factors. Gastroenterology 2012;142:1451-9 e1; quiz e14-5.
8.
Lucendo AJ, Arias A, Gonzalez-Cervera J, et al. Empiric 6-food elimination diet induced and maintained prolonged remission in patients with adult eosinophilic esophagitis: a prospective study on the food cause of the disease. J Allergy Clin Immunol 2013;131:797-804.
9.
Wolf WA, Jerath MR, Sperry SL, et al. Dietary elimination therapy is an effective option for adults with eosinophilic esophagitis. Clin Gastroenterol Hepatol 2014;12:1272-9.
10.
Rodriguez-Sanchez J, Gomez Torrijos E, Lopez Viedma B, et al. Efficacy of IgE-targeted vs empiric six-food elimination diets for adult eosinophilic oesophagitis. Allergy 2014;69:936-42.
12.
13.
SC
M AN U
TE D
EP
AC C
11.
RI PT
1.
Kagalwalla AF, Shah A, Li BU, et al. Identification of specific foods responsible for inflammation in children with eosinophilic esophagitis successfully treated with empiric elimination diet. J Pediatr Gastroenterol Nutr 2011;53:145-9.
Molina-Infante J, Arias A, Barrio J, et al. Four-food group elimination diet for adult eosinophilic esophagitis: A prospective multicenter study. J Allergy Clin Immunol 2014;134:1093-9 e1. Safroneeva E, Straumann A, Coslovsky M, et al. Symptoms Have Modest Accuracy in Detecting Endoscopic and Histologic Remission in Adults With Eosinophilic sophagitis. Gastroenterology 2016;150:581-590 e4.
ACCEPTED MANUSCRIPT
van Rhijn BD, Verheij J, Smout AJ, et al. The Endoscopic Reference Score shows modest accuracy to predict histologic remission in adult patients with eosinophilic esophagitis. Neurogastroenterol Motil 2016;28:1714-1722.
15.
Christie L, Hine RJ, Parker JG, et al. Food allergies in children affect nutrient intake and growth. J Am Diet Assoc 2002;102:1648-51.
16.
Liu T, Howard RM, Mancini AJ, et al. Kwashiorkor in the United States: fad diets, perceived and true milk allergy, and nutritional ignorance. Arch Dermatol 2001;137:6306.
17.
Erwin EA, Kruszewski PG, Russo JM, et al. IgE antibodies and response to cow's milk elimination diet in pediatric eosinophilic esophagitis. J Allergy Clin Immunol 2016;138:625-628 e2.
18.
Pelz BJ, Wechsler JB, Amsden K, et al. IgE-associated food allergy alters the presentation of paediatric eosinophilic esophagitis. Clin Exp Allergy 2016;46:1431-1440.
AC C
EP
TE D
M AN U
SC
RI PT
14.
ACCEPTED MANUSCRIPT
Table & Figure Legend Table 1. Baseline demographic, clinical, and histologic characteristics of all subjects who underwent 4-FED.
RI PT
Table 2. Histologic, endoscopic and symptomatic response to 4-FED in responders and nonresponders
SC
Figure 1. Diagram outlining the order of food reintroduction in responders Single foods reintroduced every 8 weeks in 4-FED responders starting with soy, egg, wheat and milk. Inflammation inducing trigger foods removed followed by a normal baseline EGD demonstrating remission of inflammation before the next food reintroduction.
M AN U
Figure 2. Diagram of patient flow Ninety-six patients were screened for the study; 18 were excluded and 78 underwent upper endoscopy with biopsies after 4-FED. 50 demonstrated histologic remission. Figure 3. Food triggers identified in subjects reintroducing between 1-4 foods in the diet and in those who reintroduced all foods back. a) Food triggers in patients who reintroduced 1-4 foods,
TE D
b) Food triggers in 25 patients who completed reintroduction of all foods. Included: 3 subjects with anaphylaxis, to egg (2) and soy (1) who did not reintroduce these foods in their diet. c) Distribution by number of food triggers in 25 who reintroduced all foods. Figure 4. Comparison of histology with symptom response, endoscopic findings and weight z-scores.
EP
A) Eosinophil counts at 6 different time points: baseline, post 4-FED, after introduction of soy, wheat, egg and milk.
AC C
B) Symptom scores at 6 different time points: baseline, post 4-FED and after introduction of soy, wheat, egg and milk. C) Endoscopic scores at 6 different time points: baseline, post 4-FED, and after soy, wheat, egg and milk. D) Weight z-scores at 6 different time points: baseline, post 4-FED after soy, wheat, egg and milk.
ACCEPTED MANUSCRIPT
Table 1. N (%)
9.01 ± 4.88
EP
AC C
NonResponders n=28 (36%)
9.21 ± 4.9
8.66 ± 4.92
29 (58%) 21 (42%) 39 (78%)
23 (82.14%) 5 (17.86%) 26 (92.86%)
SC
52 (66.67%) 26 (33.33%) 65 (83.33%)
p-valuea
0.6376 0.0300* 0.1194
39 (50%) 33 (42.31%) 32 (41.03%) 27 (34.62%) 24 (30.77%) 25 (32.05%) 21 (26.92%) 20 (25.64%) 18 (23.08%) 19 (24.36%) 18 (23.08%) 18 (23.08%) 18 (23.08%) 10 (12.82%) 12 (15.38%) 8 (10.26%) 8 (10.26%)
25 (50%) 21 (42%) 21 (42%) 15 (30%) 15 (30%) 16 (32%) 12 (24%) 12 (24%) 12 (24%) 14 (28%) 9 (18%) 12 (24%) 13 (26%) 5 (10%) 6 (12%) 4 (8%) 5 (10%)
14 (50%) 12 (42.86%) 11 (39.29%) 12 (42.86%) 9 (32.14%) 9 (32.14%) 9 (32.14%) 8 (28.57%) 6 (21.43%) 5 (17.86%) 9 (32.14%) 6 (21.43%) 5 (17.86%) 5 (17.86%) 6 (21.43%) 4 (14.29%) 3 (10.71%)
0.9999 0.9414 0.8152 0.2522 0.8441 0.9897 0.4367 0.6574 0.7960 0.3168 0.1550 0.7960 0.4129 0.4814 0.3319 0.4478 0.9999
32 (41.03%) 24 (31.17%) 22 (28.21%) 14 (17.95%) 46 (58.97%)
20 (40%) 12 (24.49%) 18 (36%) 10 (20%) 29 (58%)
12 (42.86%) 12 (42.86%) 4 (14.29%) 4 (14.29%) 17 (60.71%)
0.8056 0.0942 0.0409* 0.5281 0.8152
55 (71.43%) 40 (51.95%) 41 (53.25%) 34 (44.16%) 31 (40.26%) 17 (22.08%)
32 (65.31%) 22 (44.9%) 26 (53.06%) 15 (30.61%) 18 (36.73%) 9 (18.37%)
23 (82.14%) 18 (64.29%) 15 (53.57%) 19 (67.86%) 13 (46.43%) 8 (28.57%)
0.1157 0.1014 0.9656 0.0015* 0.4041 0.2990
M AN U
TE D
Age (years), mean ± SD Gender Male Female White Presenting symptoms Abdominal Pain Slow Eating Early Satiety Dysphagia Vomiting Poor Appetite Feeding Difficulties Regurgitation Poor Weight Gain Food Impaction Nausea Pocketing food/Spitting after chewing Gagging/Choking Heartburn Chest Pain Odynophagia Nocturnal awakening due to symptoms Atopy Eczema IgE-mediated food allergy (n=77) Asthma Allergic conjunctivitis Allergic rhinitis Family History (n=77) Rhinitis Asthma GERD Food Allergy Eczema Hiatal Hernia
Responders n=50 (64%)
RI PT
Patients completing the 4-FED n=78
Variable
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
EoE 11 (14.29%) 6 (12.24%) 5 (17.86%) Stricture 9 (11.69%) 6 (12.24%) 3 (10.71%) Celiac 4 (5.19%) 4 (8.16%) 0 (0%) Recurrent Vomiting 4 (5.19%) 1 (2.04%) 3 (10.71%) Common food allergens on skin prick test Egg (n=54) 18 (33.33%) 10 (27.03%) 8 (47.06%) Milk (n=55) 13 (23.64%) 9 (24.32%) 4 (22.22%) Soy (n=54) 21 (38.89%) 12 (33.33%) 9 (50%) Wheat (n=53) 11 (20.75%) 8 (22.22%) 3 (17.65%) Food sensitization on serum specific IgE test Egg (n=53) 24 (45.28%) 11 (29.73%) 13 (81.25%) Milk (n=56) 31 (55.36%) 18 (47.37%) 13 (72.22%) Soy (n=55) 21 (38.18%) 10 (26.32%) 11 (64.71%) Wheat (n=53) 24 (45.28%) 13 (34.21%) 11 (73.33%) Endoscopy visual findings Furrows (n=76) 56 (73.68%) 34 (70.83%) 22 (78.57%) Edema (n=75) 53 (70.67%) 32 (66.67%) 21 (77.78%) Exudates (n=76) 41 (53.95%) 24 (50%) 17 (60.71%) Rings (n=76) 9 (11.84%) 5 (10.42%) 4 (14.29%) Strictures (n=76) 1 (1.32%) 0 (0%) 1 (3.57%) Baseline peak eosinophil count, median (IQR) 60 (40-100) 52.5 (40-85) 80 (48-100) Post-treatment peak eosinophil count, median (IQR) 8 (3-40) 5 (1-8) 51 (35.5-90) a P-values < .05 were considered statistically significant. Chi-square and Fisher's exact tests were used to assess differences in response groups for categorical variables. A two-sample t-test was used to compare age between groups and the nonparametric Wilcoxon rank-sum test was used to compare baseline peak EOS and post-treatment EOS.
0.5157 0.9999 0.2903 0.1337 0.1470 0.9999 0.2363 0.9999 0.0005* 0.0806 0.0068* 0.0100* 0.4599 0.3104 0.3660 0.7178 0.3684 0.1211 <.0001*
ACCEPTED MANUSCRIPT Table 2
Furrow Edema Exudate Rings Stricture Symptoms
AC C
EP
TE D
Abdominal pain Slow eating Early satiety Poor appetite Dysphagia Vomiting Food impaction Gagging/choking Regurgitation Poor weight gain Pockets/spits after chewing Feeding difficulties Nausea Chest pain Heartburn Nocturnal awakening Odynophagia
* * *
*
* * * * * *
*
RI PT
Visual Findings
* * *
SC
Peak Eos/hpf Distal Eos/hpf Proximal Eos/hpf
Non-responders (n=28) Pre Post mean (sd) mean (sd) p-value 74.4 (36.4) 62.1 (31.6) 0.15 58.3 (37.6) 51.5 (30.1) 0.48 63.7 (36.7) 44.8 (33.1) 0.03 count (%) count (%) p-value 22 (79%) 23 (82%) 1.00 21 (75%) 20 (74%) 1.00 17 (61%) 17 (61%) 1.00 4 (14%) 4 (14%) 1.00 1 (4%) 0 (0%) count (%) count (%) p-value 14 (50%) 11 (39%) 0.58 12 (43%) 6 (21%) 0.08 11 (39%) 5 (18%) 0.04 9 (32%) 5 (18%) 0.34 12 (43%) 9 (32%) 0.58 9 (32%) 6 (21%) 0.55 5 (18%) 3 (11%) 0.68 5 (18%) 4 (14%) 1.00 8 (29%) 2 (7%) 0.08 6 (21%) 2 (7%) 0.13 6 (21%) 2 (7%) 0.13 9 (32%) 5 (18%) 0.29 9 (32%) 2 (7%) 0.05 6 (21%) 0 (0%) 0.04 5 (18%) 3 (11%) 0.68 3 (11%) 1 (4%) 0.48 4 (14%) 1 (4%) 0.37
M AN U
Histology
Responders (n=50) Pre Post mean (sd) mean (sd) p-value 62.0 (33.7) 5.0 (4.1) <0.001 55.9 (33.9) 4.1 (3.9) <0.001 43.9 (32.4) 2.6 (3.3) <0.001 count (%) count (%) p-value 34 (71%) 17 (35%) <0.001 32 (67%) 12 (25%) <0.001 24 (50%) 1 (2%) <0.001 5 (10%) 6 (12%) 1.00 0 (0%) 0 (0%) count (%) count (%) p-value 25 (50%) 15 (30%) 0.03 21 (42%) 18 (36%) 0.55 21 (42%) 13 (26%) 0.06 16 (32%) 5 (10%) 0.01 15 (30%) 11 (22%) 0.45 15 (30%) 4 (8%) 0.01 14 (28%) 5 (10%) 0.04 13 (26%) 3 (6%) 0.01 12 (24%) 2 (4%) 0.01 12 (24%) 5 (10%) 0.07 12 (24%) 1 (2%) 0.01 12 (24%) 5 (10%) 0.12 9 (18%) 7 (14%) 0.77 6 (12%) 4 (8%) 0.68 5 (10%) 5 (10%) 1.00 5 (10%) 1 (2%) 0.13 4 (8%) 5 (10%) 1.00
*
* *
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
Supplementary Table 1a handout provided to families
Four Food Elimination Diet
•
• • •
• •
RI PT
•
SC
•
M AN U
•
This diet is free of milk, soy, egg, and wheat. Any or all of these restricted foods on the elimination diet may be an ingredient in many processed foods. They may be also used as preservatives or flavors. There is no substitute to careful reading of food labels in order to make certain that none of the four food allergens is contained in a given food. Ingredients in a given food can change without warning. So read labels carefully on all foods every time you shop and even for foods that you have bought before. Avoid buying foods from bulk bins there is highly likelihood of mislabeling the foods. There is also a high probability of cross contamination of foods sold in bulk bins. Foods served in a restaurant or fast food establishments may be cross contaminated with above restricted foods, and many foods contain hidden sources of antigens and thus caution should be exercised when eating out. Cross-contamination occurs when one food comes into contact with another food & their proteins mix. Each food then contains small amounts of the other food that we often can’t see. Even a trace amount of food from cross-contamination can cause an allergic reaction. When cooking at home, cook the allergy-free meal first, cover, and cook/prepare the rest of the meal. Make sure utensils and cookware are clean prior to use (thoroughly washed with soap and water). Make sure you wash your hands when preparing foods. When calling food manufacturers regarding possible cross-contamination, ask about dedicated lines vs. shared equipment. Shared equipment = cross contamination, avoid.
TE D
• •
Vitamins
Always check labels for allowed ingredients with vitamins
EP
Phlexy-Vits - www.shsna.com Kirkman® Children’s Multivitamin and Mineral Hypoallergenic Capsule - www.kirkmanlabs.com Tums® Calcium Supplement (does contain cornstarch for corn allergies) Nano VM (has a hypoallergenic vitamin for 1-3years, and 4-8 years) - www.solacenutrition.com Freeda Vitamins Vitalets - www.freedavitamins.com Hero Nutritionals®: Yummy Bears Multivitamin & minerals - www.heronutritionals.com Carlson for Kids vitamins: Chewable Vitamins and Minerals - www.carlsonlabs.com Nature’s Plus: Animal Parade Vitamins - www.naturesplus.com
AC C
• • • • • • • •
Elemental Formulas Can be ordered through your local pharmacy/home health care agency with a prescription or directly through manufacturer.
ACCEPTED MANUSCRIPT
Diet free of Milk, Wheat, Egg & Soy **always check labels, manufacturers can change ingredients** Foods Allowed Enriched rice milk, hemp, coconut, quinoa, cashew, almond, pea or oat milk In recipes: substitute fruit or vegetable juice, homemade soup stock, potato water Instead of butter on toast, use pure jelly, jam, honey or herb flavored olive oil Allowed margarines (milk/soy free*) Cheese: rice/pea protein based Yogurt: coconut milk and almond yogurt *may contain soy oil/soy lecithin Grains & flours Amaranth & amaranth flour Barley & barley flour Buckwheat & buckwheat flour Chickpea or garbanzo flour Millet & millet flour Oats & oat flour Potato starch & flour Quinoa & quinoa flour Rice & rice flour Rye flour Sago flour Tapioca starch & flour Wild rice & wild rice flour Breads & baked goods – check freezer section for ready made breads with allowed ingredients Baked goods & specialty baking mixes containing allowed foods from specialty shops. (Rice, brown rice, millet or tapioca) Homemade baked goods with allowed foods Pure corn tortillas, wheat free tortillas Crackers and snacks Plain potato chips, corn chips, rice chips Pure rye crisp crackers Rice cakes, rice crackers Plain popcorn Cereals Cream of Rice Oatmeal and oat bran (plain) Puffed rice Rolled Oats/grits Corn/buckwheat based cereal Homemade granola with allowed grains Pasta Brown rice pasta wild rice pasta corn pasta Rice noodles & pasta bean pasta quínoa pasta All plain, fresh, & frozen vegetables and their juices except: soy bean sprouts & mixed sprouts. Tomato sauce: check labels for milk, soy or wheat All plain, fresh & frozen fruits & their juices (avoid pie filling fruits, may contain wheat) All fresh or frozen, plain meat or poultry Avoid all deli meats, luncheon meats with allowed ingredients All, check breading, sauces and marinades for allowed ingredients Ener-G® Egg Replacer / egg substitute recipes (see handout on cooking guidelines) All plain bean or legumes and bean or legume dishes prepared with allowed foods, except soy. All, check nuts for wheat flour; avoid nuts from bulk bins All, including sunflower, sesame, cumin, poppy, pumpkin, flax, caraway, anise Oils: olive, soybean oil (not cold pressed, expeller pressed, or extruded oil), canola, sunflower,
AC C
EP
TE D
M AN U
Breads & cereals
SC
RI PT
Type of Food Milk & Dairy
Vegetables
Fruit Meat & poultry Fish & Shellfish Eggs Legumes Nuts Seeds Fats & oils
ACCEPTED MANUSCRIPT
Spices & herbs
safflower, coconut, corn or peanut Meat drippings and poultry fat Homemade gravy made with allowed foods Tahini Lard All pure fresh or dried herbs & spices
AC C
EP
TE D
M AN U
SC
RI PT
The information included in this document is for informational purposes only and is not intended to substitute in any way for medical education, training, treatment, advice, or diagnosis by a healthcare professional. A qualified healthcare professional should always be consulted before making any healthcare-related decision
ACCEPTED MANUSCRIPT
Supplementary Table 1b provided to families at the time 4-FED was discussed
Shopping Guide (4-FED)**ALWAYS CHECK LABELS AS INGREDIENTS CAN CHANGE**
Nut butters: - Sunflower butter - Peanuts & peanut butter - Tree nuts/nut butters
Seeds: -
Flax Sesame Pumpkin Sunflower Poppyseed Chia
Tahini (SoyFree)
Cereals:
Bratwurst,
hotdogs,sausage (meat/turkey)
Lean Meats:
AC C
- Oatmeal/Oats (wheat free) - Puffed rice - Wheat Free Granola - Cream of Rice - Brown rice cereal - Rice - Corn - Quinoa - Buckwheat
EP
(Wheat,soy,egg,milk free) Brown or white rice Corn Cornmeal/polenta Corn Tortilla Corn Chips Millet & millet flour Quinoa Teff Peas Potatoes (w/skin) Sweet potatoes (w/skin) Plantains Pasta: Corn Rice Lentil Legume Quinoa pasta Yucca
-
(canned, fresh or frozen)
Artichoke hearts Asparagus Beets Bok choy Broccoli Brussels sprouts Cabbage Carrots Cauliflower Celery Collard greens Cucumber Daikon Eggplant Endive Green beans Green onions Hearts of palm Jicama slices Kale Kohlrabi Leeks Mushrooms Mustard greens Okra Onions Pea pods Peppers (all varieties) Radishes Romaine Salsa Salad greens Scallions Spaghetti sauce Spinach Summer squash Swiss chard Tomato Tomato sauce Tomatillos Grape tomatoes Turnips Water chestnuts Watercress Zucchini
Poultry Pork Lamb Veal Beef Poultry
Deli Meat (packaged) (not from deli counter)
Apples Applesauce (no
Dairy Substitutes Fortified Milks: - Almond - Coconut - Hemp - Oat - Rice - Cashew - Quinoa (under 3 years old: recommend elemental formulas to replace milk in diet)
RI PT
Bagels/breads
- Azuki - Black - Garbanzo - Kidney - Pinto - White Bean/lentil soups Hummus Fish Shellfish
Fruits -unlimited
(canned, fresh, or frozen)
sugar added) Apricots Bananas Blackberries Blueberries Cantaloupe Cherries Grapefruit Grapes Honeydew Kiwi Mandarin oranges Mango Nectarine Orange Papaya Persimmons Peach Pear Pineapple Plum Pomegranate Prunes Raspberries Strawberries Tangerines Watermelon Individual fruit cups (in own juice) Dried fruit
SC
Canned beans:
amaranth flour
Vegetables - unlimited
TE D
Amaranth and
Proteins
M AN U
Whole Grains
Freeze Dried
Vegetables/Fruits
Margarines/spreads Yogurts: - Almond - Coconut
Cheese: - Rice based - Pea based
ACCEPTED MANUSCRIPT
Sweets/Treats: Allowed Carob Chips,
(chicken nuggets, fish nuggets, tator tots, French fries, macaroni and cheese)
Allergy free pizzas
fresh or dried –
McCormick Brand spices
Lemon/lime juice Mustard Fats/Oil: - Canola - Grape seed - Olive - Safflower - Sesame - Avocado - coconut Vinegars Beverages: - La Croix™ - Juice spritzers - Tea, hot/iced - Diet/regular pop - Gatorade - Vitamin water - 100% fruit juice (8floz/day limit)
TE D
Wheat free Breads
cookies, cereal bars 100% fruit sorbet 100% all fruit bars Hard candy Plain potato chips Rice crackers (plain) Rice cakes (plain) Popcorn (air popped or stovetop),plain Potato (sweet potato) chips, plain Corn chips Wheat free pretzel chips, pretzels
Avocado Garlic Herbs and spices,
RI PT
Allergy free kids foods
Miscellaneous
SC
Snacks/ Desserts
M AN U
Frozen Foods
AC C
EP
Sweeteners: - Sugar - Equal ® - Splenda - Honey - Jam/jelly Tupperware/baggies - Sports bottle w/straw
ACCEPTED MANUSCRIPT
Supplementary Table 2.
PPV 0.75 (0.33-1) 0.25 (0-0.67) 0.60 (0.17-1) 0.18 (0-0.41) 0.75 (0.45-1) 0.38 (0.04-0.71) 0.75 (0.33-1) 0.25 (0-0.55)
RI PT
NPV 0.14 (0-0.33) 0.76 (0.56-0.97) 0.77 (0.6-0.95) 0.81 (0.64-0.98) 0.10 (0-0.29) 0.73 (0.51-0.96) 0.70 (0.51-0.88) 0.84 (0.7-0.98)
Sensitivity 0.20 (0-0.4) 0.20 (0-0.55) 0.38 (0.04-0.71) 0.33 (0-0.71) 0.40 (0.15-0.65) 0.43 (0.06-0.8) 0.30 (0.02-0.58) 0.33 (0-0.71)
Specificity 0.67 (0.13-1) 0.81 (0.62-1) 0.89 (0.76-1) 0.65 (0.47-0.84) 0.33 (0-0.87) 0.69 (0.46-0.91) 0.94 (0.83-1) 0.78 (0.62-0.93)
SC
18 32 27 21 18 33 27 23
M AN U
N
TE D
Serum sIgE
p(95% CI)
EP
Skin Prick
4-FED Trigger Food Milk Wheat Egg Soy Milk Wheat Egg Soy
AC C
Allergy Test
ACCEPTED MANUSCRIPT
Supplementary Table 3. N (%) Midwest n=52 (66.7%)
Age (years), mean ± SD Gender Male Female White Presenting symptoms Abdominal Pain Slow Eating Early Satiety Dysphagia Vomiting Poor Appetite Feeding Difficulties Regurgitation Poor Weight Gain Food Impaction Nausea Pocketing food/Spitting after chewing Gagging/Choking Heartburn Chest Pain Odynophagia Nocturnal awakening due to symptoms Atopy Eczema IgE-mediated food allergy (n=77) Asthma Allergic conjunctivitis Allergic rhinitis Family History (n=77) Rhinitis Asthma GERD Food Allergy Eczema Hiatal Hernia EoE
9.11 ± 4.93
8.79 ± 4.87
11 (73.33%) 4 (26.67%) 15 (100%)
6 (54.55%) 5 (45.45%) 10 (90.91%)
p-value
0.9700 0.5954 0.0717
29 (55.77%) 21 (40.38%) 20 (38.46%) 18 (34.62%) 17 (32.69%) 17 (32.69%) 14 (26.92%) 13 (25%) 9 (17.31%) 13 (25%) 11 (21.15%) 13 (25%) 15 (28.85%) 8 (15.38%) 11 (21.15%) 3 (5.77%) 7 (13.46%)
4 (26.67%) 9 (60%) 8 (53.33%) 7 (46.67%) 3 (20%) 5 (33.33%) 5 (33.33%) 4 (26.67%) 6 (40%) 4 (26.67%) 5 (33.33%) 3 (20%) 1 (6.67%) 1 (6.67%) 1 (6.67%) 3 (20%) 1 (6.67%)
6 (54.55%) 3 (27.27%) 4 (36.36%) 2 (18.18%) 4 (36.36%) 3 (27.27%) 2 (18.18%) 3 (27.27%) 3 (27.27%) 2 (18.18%) 2 (18.18%) 2 (18.18%) 2 (18.18%) 1 (9.09%) 0 (0%) 2 (18.18%) 0 (0%)
0.1320 0.2208 0.5545 0.3206 0.6647 0.9999 0.7466 0.9999 0.1574 0.9999 0.6156 0.9999 0.2380 0.8772 0.1950 0.1144 0.5461
18 (34.62%) 12 (23.08%) 16 (30.77%) 9 (17.31%) 31 (59.62%)
10 (66.67%) 9 (60%) 1 (6.67%) 5 (33.33%) 10 (66.67%)
4 (36.36%) 3 (30%) 5 (45.45%) 0 (0%) 5 (45.45%)
0.0797 0.0269* 0.0669 0.1102 0.5470
38 (74.51%) 26 (50.98%) 27 (52.94%) 21 (41.18%) 16 (31.37%) 11 (21.57%) 6 (11.76%)
13 (86.67%) 8 (53.33%) 7 (46.67%) 8 (53.33%) 11 (73.33%) 2 (13.33%) 2 (13.33%)
4 (36.36%) 6 (54.55%) 7 (63.64%) 5 (45.45%) 4 (36.36%) 4 (36.36%) 3 (27.27%)
0.0187* 0.9703 0.6908 0.7035 0.0138* 0.4341 0.4231
M AN U
TE D
EP
AC C
South n=11 (14.1%)
8.84 ± 5.07
SC
35 (67.31%) 17 (32.69%) 40 (76.92%)
East Coast n=15 (19.2%)
RI PT
Variable
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Stricture 7 (13.73%) 2 (13.33%) 0 (0%) 0.5765 Celiac 1 (1.96%) 0 (0%) 3 (27.27%) 0.0173* Recurrent Vomiting 2 (3.92%) 1 (6.67%) 1 (9.09%) 0.4152 Common food allergens on skin prick test Egg (n=54) 9 (25%) 4 (44.44%) 5 (55.56%) 0.1617 Milk (n=55) 9 (24.32%) 2 (22.22%) 2 (22.22%) 0.9999 Soy (n=54) 12 (34.29%) 4 (44.44%) 5 (50%) 0.6559 Wheat (n=53) 6 (17.14%) 2 (22.22%) 3 (33.33%) 0.4772 Food sensitization on serum sIgE test Egg (n=53) 14 (40%) 8 (66.67%) 2 (33.33%) 0.2392 Milk (n=56) 19 (51.35%) 7 (58.33%) 5 (71.43%) 0.6176 Soy (n=55) 11 (31.43%) 7 (53.85%) 3 (42.86%) 0.3661 Wheat (n=53) 15 (42.86%) 6 (50%) 3 (50%) 0.9175 Endoscopy findings Furrows (n=76) 38 (73.08%) 13 (92.86%) 5 (50%) 0.0606 Edema (n=75) 36 (69.23%) 12 (92.31%) 5 (50%) 0.0881 Exudates (n=76) 26 (50%) 13 (92.86%) 2 (20%) 0.0012* Rings (n=76) 9 (17.31%) 0 (0%) 0 (0%) 0.1391 Strictures (n=76) 1 (1.92%) 0 (0%) 0 (0%) 0.9999 Baseline peak eosinophil count, median (IQR) 70 (42.5-100) 65 (29-100) 40 (25-60) 0.0347* Post-treatment peak eosinophil count, median (IQR) 8 (2-22.5) 35 (7-85) 6 (0-52) 0.0728 a P-values < .05 were considered statistically significant. Chi-square and Fisher's exact tests were used to assess differences by region for categorical variables. A two-sample t-test was used to compare age between regions and the nonparametric Wilcoxon rank-sum test was used to compare baseline peak EOS and post-treatment EOS.
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
Supplementary Material Four Food Elimination Diet for Eosinophilic Esophagitis in Children
RI PT
Amir F Kagalwalla; Joshua B Wechsler; Katie Amsden; Sally Schwartz; Melanie M Makhija; Anthony Olive; Carla M Davis; Maria Manuel-Rubio; Seth Marcus; Maureen Sulkowski; Kristin Johnson; Jessica N Ross; Mary Ellen Riffle; Marion Groetch; Hector Melin-Aldana; Deborah A Schady; Hannah Palac; Kwang-Youn A Kim; Barry K Wershil; Margaret H Collins; Mirna Chehade
SC
Supplementary Table 1 a & b. Elimination diet handouts provided to families prior to starting the 4-FED Resource lists of allergy-free food companies, elemental formula information, allergy-free vitamin information, support websites, allowed foods, menu planning, recipes, and shopping guides as well as trouble shooting tips for travelling, school and eating out.
M AN U
Supplementary Table 2. Final multivariable logistic regression model for treatment response based on history of asthma, family history of food allergy, positive serum sIgE and gender
Supplementary Table 3. Demographic, clinical, allergy testing, endoscopic, and histological characteristics compared in the three different regions.
AC C
EP
TE D
Supplementary Figure 1. ROC curve identifying factors associated with response with an AUC = 0.8 ROC curve for final multivariable model identifed asthma, family history of food allergy and positive serum sIgE test to any of the four foods as significantly associated with treatment non-response. The AUC of the final model was 0.8.