Eosinophilic esophagitis (EoE) is an immune-mediated chronic inflammatory disorder of the esophagus. EoE is thought to be triggered, in most subjects, by the ingestion of food antigen(s) or, in some cases, via inhaled aeroallergens (1–5). In a seminal pediatric study published in 1995, Kelly et al (6) first demonstrated that clinical symptoms and esophageal eosinophilic inflammation in children with EoE were reversed by the exclusion of all of the intact proteins from the diet and by the substitution with an amino acid–based elemental diet. The authors went on to identify specific food antigens including cow's milk, soy, wheat, egg, and peanuts as proteins that induced esophageal inflammation in their cohort (6). These observations form the basis of most dietary approaches offered today to treat EoE.
Presently, the treatment goals of EoE include resolution of clinical symptoms, maintenance of remission and prevention of disease relapse, prevention of complications such as fibrosis and strictures by maintaining histological remission, prevention of adverse iatrogenic reactions such as nutritional deficiencies from dietary treatment or toxicity related to drugs, and maintenance of quality of life.
A number of different treatment options, including topical steroids, elemental diet, directed elimination diet, and empiric 6-food elimination diet, are available to treat children with EoE. In most instances, treatment is individualized and the selection of the specific treatment modality is based on a discussion of the outcomes of different therapies, the likelihood of acceptance of treatment by the child, the family's financial resources, the lifestyle of the child, and the suitability of the treatment to the family. Because multiple modalities to treat EoE exist, identifying the one that is simplest, has the least potential for adverse effects, and at the same time is easy to implement is needed.
A dietary approach is based on the hypothesis that food antigen(s) trigger eosinophilic inflammation and clinical and histological remission can be induced and then maintained by identifying and avoiding only the specific causative food antigen(s) (7). It is believed that prolonged and continuous elimination of causative food antigen(s) induces long-term remission. All of the presently accepted dietary treatment options initially require elimination of multiple food antigens to demonstrate clinical and histological remission followed by sequential reintroduction of single foods to identify causative food antigens (2,4–10). Each single food reintroduction is punctuated with endoscopic esophageal biopsies to exclude disease recurrence before the next food is reintroduced (5). This process is difficult and costly because it requires multiple, invasive endoscopies and introduces the possibility of iatrogenic effects such as nutrient deficiency resulting from simultaneous multiple food eliminations. In addition, patient compliance may be compromised, both intentionally and unintentionally, because many families report the elimination process to be both confusing and frustrating. The objective of the present study was to explore the histological response in children who had only 1 common food antigen (cow's milk) eliminated from their diet.
Children and adolescents with chronic gastrointestinal symptoms, including vomiting, malnutrition, feeding aversion, abdominal pain, dysphagia, and food impaction, who were refractory to either at least 6 weeks of proton pump inhibitor (PPI) therapy or had a normal 24-hour pH probe study were diagnosed as having EoE based on the presence of >15 eosinophils per high-power field (eos/hpf) in esophageal biopsies. Patients with esophageal eosinophilia (>15 eos/hpf) who were neither pretreated with PPI nor underwent 24-hour pH probe study were also included if they had symptoms of esophageal dysfunction and their histological disease remitted (esophageal eosinophil count <15 eos/hpf) with milk elimination therapy as the only intervention as described in the 2011 consensus guidelines (11,12).
Allergic and Nonallergic Phenotypes
Patients were characterized as either allergic or nonallergic based on the presence of reactive airway disease, allergic rhinitis, eczema, environmental, or known food allergies.
Our primary study endpoint was remission with esophageal peak eosinophil count ≤15 eos/hpf. Remission was further categorized as complete remission defined as a peak eosinophil count ≤1 eos/hpf on follow-up endoscopy or significant improvement defined as a peak eosinophil count of 2 to 15 eos/hpf. Treatment failure was defined as peak eosinophil counts >15 eos/hpf.
Resolution of clinical symptoms was the secondary treatment endpoint. Symptom response was obtained by patient/parent-reported symptoms following a treatment period of milk elimination. Response was sought specifically to the persistence, improvement, or resolution of the initial symptom(s).
Biopsies and Histology
Biopsies were obtained from the distal and mid esophagus. Four-quadrant biopsy samples from the distal and 4 biopsy samples from the mid-esophagus for a total of 8 esophageal biopsies as well as gastric and duodenal biopsies were obtained during initial and each subsequent endoscopy. All 8 esophageal tissue samples were individually assessed for eosinophil density and presence of basal cell hyperplasia. All of the fields were evaluated under ×400 magnification and eosinophils were counted from areas with the highest density in each of the 8 biopsies. The highest (peak) value was recorded based on the review of all 8 tissue samples and accepted for our analysis. All of the biopsies were reviewed for uniformity by a single board-certified pathologist (H.M.A.) experienced with evaluating biopsies from children with EoE.
Study Design and Participants
This retrospective study examined consecutive children with EoE seen either at Ann & Robert H. Lurie Children's Memorial Hospital of Chicago or at a suburban pediatric subspecialty practice setting (S.N.) between January 1, 2006 and December 31, 2011. Patients from both practice sites had endoscopic and histological evaluation of the biopsies performed at Lurie Children's. Children with EoE were offered a number of different options including topical swallowed steroids (fluticasone or oral viscous budesonide), elemental diet, allergy-tested elimination diet, empiric 6-food elimination, and only milk elimination diet. Of these, only patients with EoE who were treated with the elimination of cow's milk from the diet and had repeat upper endoscopy with esophageal biopsies after exclusion of cow's milk from the diet were included in the study. Response for the purpose of the present study was defined as ≤15 eos/hpf following the exclusive milk elimination diet. Furthermore, within this group of respondents, complete remission was defined as a peak eosinophil count ≤1 eos/hpf and significant remission was defined as a peak eosinophil count of 2 to 15 eos/hpf.
A registered dietitian (S.R.) initially met in person to instruct families and regularly counseled parent(s)/guardian(s) during the single-food elimination treatment period (in person, as well as responding to telephone and e-mail inquiries). Specific written instructions were also given to the families regarding food cross-contamination by carefully reading labels and offering a balanced age-appropriate diet after eliminating cow's milk–containing foods from the diet. The dietitian also counseled the families throughout the food elimination process, ensuring adherence as well as adequacy of energy intake and food substitutions to meet all the protein/energy needs required for growth. Elemental or soy-based formula was supplemented for children whose diet was predominantly milk based to make up for the calories lost by milk elimination, while still allowing all of the major foods in their diet other than milk.
Patients who were prescribed oral, nasal, airway, or swallowed steroids after the diagnostic endoscopy and concurrently along with the cow's milk–elimination treatment phase were excluded. Children who did not undergo repeat upper endoscopy after eliminating cow's milk from their diet were also excluded.
The study was approved by the institutional review board at Lurie Children's.
Statistical Methods and Analysis
The Wilcoxon signed-rank test was used to analyze the differences in pretreatment and posttreatment histology. The Kruskal-Wallis test was used to test the differences among means of the pretreatment eosinophil counts of the 3 remission groups. The Mann-Whitney test was used to test the differences between the mean ages of the remission group and the treatment failure group and between the mean ages of the treatment remission groups. Significance was defined as 2-tailed P value of <0.05 throughout. All analyses were conducted using SPSS 12.0 for Windows (SPSS Inc, Chicago, IL).
From January 1, 2006 to December 31, 2011, we identified 161 children who met the diagnostic criteria for EoE, received treatment, and underwent subsequent repeat upper endoscopy at Lurie Children's or the suburban pediatric subspecialty practice. Of the 161 children, 99 (61%) were treated with empiric elimination diets, 37 (23%) with topical steroids, 13 (8%) with a combination of elimination diet and steroids, and 12 (7%) with elemental diet. The remission rate of milk only elimination is 65% (95% CI 42%–88%) similar to that of other treatment options within our clinical practice. Table 1 shows the remission with different treatment modalities. Of the 99 children with EoE treated with elimination diet, 17 excluded only cow's milk from their diet and underwent subsequent repeat upper endoscopy at least 6 weeks after this single-food elimination and these subjects are the basis of this report. Fourteen of 17 were pretreated with PPI before the diagnostic endoscopy. One patient had a normal 24-hour pH probe study. The last 2 patients were neither pretreated with PPI nor underwent a 24-hour pH probe study but were included based on their clinical presentation and resolution of both clinical symptoms and the esophageal eosinophilia after only excluding cow's-milk protein from their diet.
The mean age for the cohort of 17 (71% male) patients was 5.5 ± 3.2 (range 1–12) years. Fourteen (82%) patients were white; details are shown in Table 2.
Vomiting and dysphagia were the most common symptoms in our cohort followed by regurgitation, coughing, and choking. Clinical symptoms improved or resolved in all 17 patients. The individual symptoms of the patients and their posttreatment responses are shown in Table 3.
Cow's milk accounted for >30% of calories in 14 of 17 patients before elimination of milk from their diet. The milk intake in both those in remission and those who failed the treatment was identical, before the study. Four children required supplemental elemental formula to enhance their caloric intake and this ranged from 25% to 75%. All 4 patients also continued to ingest other major foods, except milk. All of the children except 2 maintained or gained weight during the period of milk elimination.
Eleven of 17 underwent skin prick testing or radioallergosorbent test and 3 patients had atopy patch testing, as shown in Table 4. Four of 11 allergy-tested patients were positive to different foods and only 1 of the 11 was positive to milk and this was based on atopy patch testing. Overall, 15 of 17 (88%) were characterized as having a history of atopy based on history of reactive airway disease, eczema, allergic rhinitis, or allergy testing. The patient profile including demographics and atopy are shown in Table 2.
The treatment response in individual subjects, atopic history, and individual pretreatment, posttreatment eosinophil counts, and the presence or absence of basal cell hyperplasia for each of the 3 response groups are shown in Table 4. Seven of 17 (41%) children demonstrated complete histological remission; 4 of 17 (24%) exhibited significant histological remission. Six children failed treatment (35%). Of the 6 children who failed treatment, the esophageal eosinophil count decreased between 33% and 89% in 4 children, indicating that cow's milk was one of the incriminating food antigens, but possibly additional food(s) or environmental allergens were responsible for the eosinophilic inflammation. The posttreatment mean eosinophil counts for each of the 3 response groups are compared in Table 5. The differences between the mean pretreatment eosinophil counts and ages between those in remission and those who failed treatment are shown in Figure 1. The mean pretreatment eosinophil load is lower for those in remission compared with those who failed treatment; the values were 76 ± 40 and 113 ± 80, respectively, for the 2 groups. The mean age of the children with histological remission was significantly lower 4.2 ± 2.2 compared with 8.0 ± 3.6 (P < 0.05) for those who failed treatment. The individual pretreatment and posttreatment eosinophil values for the entire group are demonstrated in Figure 2.
There were no treatment-related complications. There were also no procedure-related complications, including perforation, bleeding, or infection in any of the patients.
The present dietary treatments for EoE initially require elimination of multiple foods to induce clinical and histological remission. This approach, even for a short period of time, is problematic and places a significant burden on the child and the family's lifestyle; it affects compliance and family resources. In the present study, clinical and histological remission was achieved with elimination of only cow's-milk protein from the diet of children with EoE with only 1 additional upper endoscopy and without disrupting the lifestyle of the child and the family. These findings, if they can be substantiated by larger prospective studies, may have long-term implications because continuing single-food exclusion may serve as maintenance therapy to prevent histological recurrence and complications such as tissue remodeling and strictures.
The present study illustrates that, in a small subset of patients, isolated cow's milk elimination is an effective dietary therapy for EoE; however, the retrospective nature and the small number of subjects in the present study preclude drawing broader inference that isolated cow's milk elimination alone will be an effective dietary modality for managing children with EoE. Larger prospective studies, like the one that is underway in our institution, are needed to validate this hypothesis that cow's milk elimination alone may be an effective therapy in children with EoE.
A major methodical flaw of the present study relates to the absence of a formal tool to assess changes in symptoms. Symptoms in all of our subjects improved. Two of 6 children failed treatment and demonstrated no change in the eosinophilic inflammation and yet had resolution of their symptoms, which is not easily explained. This failure to use a validated clinical instrument or patient-related outcomes tool remains a major problem for clinical trials in subjects with EoE and is even more problematic in children in whom cognitive abilities vary with age.
The goal of any therapy in EoE is to prevent adverse outcomes and long-term complications. Several pediatric studies have shown the presence of subepithelial fibrosis in excess of 50% in children with EoE; additionally, subepithelial fibrosis was demonstrated in all of the children with food impaction and dysphagia (13,14). Unbridled eosinophilic esophageal inflammation can lead to esophageal remodeling, subepithelial fibrosis, and possibly esophageal stricture in some patients. Esophageal strictures were identified in 27% of children being evaluated for food impaction (15). Aceves et al (16) have shown that fibrosis may be reversed with steroids in some cases. We have shown that epithelial mesenchymal transition that contributes to fibrosis reverses with a decrease in eosinophilic load. The reversal of epithelial mesenchymal transition directly correlated with a decrease in eosinophil count irrespective of whether patients were treated with swallowed steroids, elemental diet, or 6-food elimination diet (17,18). Based on these findings, it would appear reasonable to infer that reversing or interrupting unbridled eosinophilic inflammation, leading to resolution of fibrosis, offers the prospect of preventing complications including structural and functional impairment of the esophagus in EoE.
Topical or swallowed corticosteroid therapy is the most frequently used therapeutic option to treat EoE (19). Several prospective studies have demonstrated that swallowed corticosteroids are highly efficacious in inducing clinical and histological remission with success rates ranging from 50% to 94% of subjects (20,21). The potential side effects of corticosteroids include opportunistic infections such as Candida esophagitis. The single most significant limitation of corticosteroids is the recurrence of the esophageal inflammation once the steroids are discontinued. This is also true of elimination diets in which the eosinophilic esophageal inflammation recurs once the incriminating food protein is reintroduced. Corticosteroid resistance has also been described (22). Presently, there are no data addressing the benefit and risks associated with long-term continuous administration of induction dose corticosteroids in maintaining remission. The efficacy of low-dose corticosteroids to maintain remission is unproven. In a recent study by Straumann et al (23), 50% of adults with EoE treated with low-dose maintenance corticosteroids for 50 weeks achieved either partial or complete histological remission. The eosinophil load in the maintenance group increased from 0.4 to 31.8 eos/hpf. Additional long-term multicenter studies are needed to validate the findings of this small study and establish that this eosinophil load is sufficiently low enough to prevent fibrosis.
Exclusive elemental diet is presently the most effective therapy because it has been shown to induce remission in up to 95% of patients. More significantly, an elemental diet, of all the other dietary approaches, is the most likely to induce complete histological remission in the highest number of patients (10). Compliance with this approach often requires placement of tubes (nasogastric or gastrostomy). Costs related to purchasing expensive formula are also an impediment for some patients. The food reintroduction process is prolonged, and the number of upper endoscopies required as all foods are reintroduced is much higher than those required with other elimination diets.
Both directed and empiric elimination diets have demonstrated identical remission rates of 77% and 74%, respectively (2,4). Both therapies require elimination of multiple foods initially to establish remission. The food reintroduction process after remission is established is long and patients report the process to be difficult. This is often the reason patients discontinue food reintroductions and drop out of therapy (7). Multiple endoscopies necessary to confirm disease activity are also unacceptable to many patients and parents and are another reason for discontinuing treatment.
Our present study eliminating a single food addresses many of the drawbacks associated with other elimination diets. Of the 17 children with EoE who participated in trials with cow's milk only elimination, both clinical and complete or significant histological remission was induced in 11 children. This was achieved with only 1 additional upper endoscopy. The clinical and histological improvement demonstrated with exclusion of cow's milk from the diets of patients supports the concept that milk is a major offending antigen in this condition and in fact, in some patients, may be the only offending antigen.
Children in remission on milk only elimination were younger in age and tended to have lower pretreatment eosinophil counts than those who failed treatment. The children who were in remission on this treatment were younger in age than those children who have remission on topical steroids or treatment with elimination diet and steroids. Although the small sample size in our study precludes the power to detect a statistical difference in the pretreatment eosinophil load between the 2 response groups, our findings do support the conclusion that the milk elimination diet is more likely to be successful in inducing remission in younger children. Larger prospective studies are also needed to validate the findings of the present study that younger age is a predictor of higher response to this treatment.
This is the first study demonstrating that removal of a single food alone for the treatment of EoE induces clinical and histological remission in children. Younger children with EoE for whom elimination of multiple foods may not be feasible, and patients in whom milk intake is excessively disproportional to their dietary needs may be suitable candidates for elimination of only cow's milk from their diet. Additional large prospective studies are needed to validate and better define the role of single-food elimination for the treatment of EoE in children.
The authors thank Ms Katie Neighbors for invaluable advice with the manuscript and Mr Gang Zhang for assistance with the statistical section of the manuscript.
1. Fox VL, Nurko S, Furuta GT. Eosinophilic esophagitis: it's not just kid's stuff. Gastrointest Endosc
2. Spergel JM, Andrews T, Brown-Whitehorn TF, et al. Treatment of eosinophilic esophagitis with specific food elimination diet directed by a combination of skin prick and patch tests. Ann Allergy Asthma Immunol
3. Spergel JM. Eosinophilic esophagitis in adults and children: evidence for a food allergy component in many patients. Curr Opin Allergy Clin Immunol
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
5. Gonsalves N, Yang GY, Doerfler B, et al. Elimination diet effectively treats eosinophilic esophagitis in adults; food reintroduction identifies causative factors. Gastroenterology
6. Kelly KJ, Lazenby AJ, Rowe PC, et al. Eosinophilic esophagitis attributed to gastroesophageal reflux: improvement with an amino acid-based formula. Gastroenterology
7. 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
8. Markowitz JE, Spergel JM, Ruchelli E, et al. Elemental diet is an effective treatment for eosinophilic esophagitis in children and adolescents. Am J Gastroenterol
9. Kagalwalla AF, Shah A, Ritz S, et al. Cow's milk protein-induced eosinophilic esophagitis in a child with gluten-sensitive enteropathy. J Pediatr Gastroenterol Nutr
10. Liacouras CA, Spergel JM, Ruchelli E, et al. Eosinophilic esophagitis: a 10-year experience in 381 children. Clin Gastroenterol Hepatol
11. Liacouras CA, Furuta GT, Hirano I, et al. Eosinophilic esophagitis: updated consensus recommendations for children and adults. J Allergy Clin Immun
12. Blanchard C, Stucke EM, Rodriquez-Jiminez B, et al. A striking local esophageal cytokine expression profile in eosinophilic esophagitis. J Allergy Clin Immum
13. Chehade M, Sampson HA, Morotti RA, et al. Esophageal subepithelial fibrosis in children with eosinophilic esophagitis. J Pediatr Gastroenterol Nutr
14. Aceves SS, Newbury RO, Dohil R, et al. Esophageal remodeling in pediatric eosinophilic esophagitis. J Allergy Clin Immunol
15. Hurtado CW, Furuta GT, Kramer RE. Etiology of esophageal food impactions in children. J Pediatr Gastroenterol Nutr
16. Aceves SS, Newbury RO, Chen D, et al. Resolution of remodeling in eosinophilic esophagitis correlates with epithelial response to topical corticosteroids. Allergy
17. Rea B, Akhtar N, Jacques K, et al. Epithelial mesenchymal transition induced esophageal remodeling in eosinophilic esophagitis reverses with treatment. Gastroenterology
18. Kagalwalla AF, Akhtar N, Woodruff SA, et al. Eosinophilic esophagitis: epithelial mesenchymal transition contributes to esophageal remodeling and reverses with treatment. J Allergy Clin Immunol
19. Spergel JM, Book WM, Mays E, et al. Variation in prevalence, diagnostic criteria, and initial management options for eosinophilic gastrointestinal diseases in the United States. J Pediatr Gastroenterol Nutr
20. Konikoff MR, Noel RJ, Blanchard C, et al. A randomized, double-blind, placebo-controlled trial of fluticasone propionate for pediatric eosinophilic esophagitis. Gastroenterology
21. Schaefer ET, Fitzgerald JF, Molleston JP, et al. Comparison of oral prednisone and topical fluticasone in the treatment of eosinophilic esophagitis: a randomized trial in children. Clin Gastroenterol Hepatol
22. Caldwell JM, Blanchard C, Collins MH, et al. Glucocorticoid-regulated genes in eosinophilic esophagitis: a role for FKBP51. J Allergy Clin Immunol
23. Straumann A, Conus S, Degen L, et al. Long-term budesonide maintenance treatment is partially effective for patients with eosinophilic esophagitis. Clin Gastroenterol Hepatol