Eosinophilic gastrointestinal diseases (EGIDs) are a diverse group of intestinal diseases involving the infiltration of eosinophils into the bowel wall. EGID can present with a variety of clinical conditions, which are largely dependent on the location of eosinophils in the intestinal wall. We describe the first reported pediatric cases of EGID presenting with symptomatic partial bowel obstruction from intestinal masses due to isolated focal mural involvement. Both patients subsequently responded favorably to therapy with exclusive elemental nutrition in the first case and exclusive elemental nutrition with steroids in the second case. These cases reveal the wide-ranging clinical manifestations of EGID, expand on the differential diagnosis of focal intestinal masses, and provide guidance on the evaluation of ambiguous cases.

A 13-year-old girl with asthma and atopic dermatitis had multiple outside hospital admissions for acute-on-chronic abdominal pain and emesis. A nontender, ill-defined mobile mass was palpable along the left lower quadrant on examination. Outside hospital endoscopy revealed a 1- to 2-cm duodenal bulge proximal to the ampulla, and the histology was positive for Helicobacter pylori with normal duodenal and colonic biopsies. Computed tomography scan revealed gastric pylorus and duodenal bulb with severe mucosal thickening and adjacent lymphadenopathy concerning for malignancy versus inflammatory process. She was transferred to our institution and ultimately discharged on triple therapy with outpatient follow-up. Outpatient magnetic resonance enterography (MRE) 3 months later demonstrated circumferential mural thickening of the antrum, pylorus, and duodenum (Fig 1C), leading to admission for esophagogastroduodenoscopy, which revealed corresponding luminal narrowing of the proximal duodenum, protruding duodenal ulcer and thickened duodenal folds, as well as gastric nodularity (Fig 1A). Endoscopic ultrasound on the same day revealed thickening of the antropyloric duodenum with apparent infiltration beyond the serosa suggestive of a possible inflammatory reaction. Endoscopic biopsies revealed prominent eosinophilia with degranulation and peak eosinophil counts of >100 per high-power field (HPF) in the stomach and 97 per HPF in the duodenum (Fig 1B).

FIGURE 1

A, Endoscopic image before diagnosis revealing a duodenal mass (left) and after treatment revealing improved mucosal thickening (right). B, Histology before diagnosis revealing prominent eosinophilia with degranulation (left) and after treatment without increased eosinophils (right). C, MRE before diagnosis with circumferential mural thickening (left) and after treatment revealing reduced mural thickening (right).

FIGURE 1

A, Endoscopic image before diagnosis revealing a duodenal mass (left) and after treatment revealing improved mucosal thickening (right). B, Histology before diagnosis revealing prominent eosinophilia with degranulation (left) and after treatment without increased eosinophils (right). C, MRE before diagnosis with circumferential mural thickening (left) and after treatment revealing reduced mural thickening (right).

Close modal

Given the concern for mucosa-associated lymphoid tissue lymphoma, a full-thickness biopsy was indicated; however, it was deferred because of concern for inflammatory bowel disease (IBD) and risk of dehiscence. Laboratory values, including normal hemoglobin and platelets, erythrocyte sedimentation rate of 18 mm/hour, undetectable C-reactive protein, calprotectin of 20 μg/g, and negative IBD panel, did not support an IBD diagnosis. Serology was negative for celiac disease, hepatitis, and HIV. An eosinophilic gastrointestinal disease (EGID) diagnosis was favored given high tissue eosinophilia, peak absolute eosinophil count of 1400 k/UL with 15% eosinophils, elevated immunoglobulin E (IgE) of 4582 kU/L, and high titers of IgE specific to sesame 25 kUA/L, wheat 14.3 kUA/L, milk 14 kUA/L, corn 9.2 kUA/L, soy 8.8 kUA/L, peanut 6.6 kUA/L, and egg 4.8 kUA/L. Although positron emission tomography scan was negative for diffuse disease, focal mucosa-associated lymphoid tissue lymphoma remained a concern without full-thickness biopsy. We initiated a 2-week trial of exclusive elemental nutrition (EEN) therapy for presumed EGID, after which she showed clinical remission with resolution of symptoms. Repeat endoscopy and biopsies revealed improved mucosal thickening and ulcer resolution, and no eosinophils were seen on histology (Fig 1 A and B). Enteric-coated steroids were added to EEN because malignancy was no longer a concern. By the 3-month follow-up, she reintroduced hypoallergenic foods, and endoscopic biopsy and MRE revealed disease remission (Fig 1C). The patient was placed on a steroid taper plan, initially on 40 mg of prednisone for 3 months, then decreased by 5 mg weekly. She was lost to follow-up thereafter.

A 13-year-old girl with a history of chronic pancreatitis status post sphincterotomy for common bile duct dilation required multiple hospitalizations over 1 year for persistent symptoms of emesis, decreased oral intake, abdominal pain, and constipation. On esophagogastroduodenoscopy, she was found to have a pyloric antral mass, duodenal ulcer, and stricture of the second portion of the duodenum (Fig 2A). Histology revealed increased lamina propria eosinophils with a peak count of 85 per HPF with degranulation and epithelial encroachment. Fine-needle aspiration and biopsy of the mass, performed because of concern for malignancy, revealed fibroconnective tissue. During this time, she was found to have severe iron deficiency anemia requiring transfusion, a 15-lb weight loss (BMI 17.6), and an IBD panel with perinuclear anti-neutrophil cytoplasmic antibodies >100 U. Abdominal MRI revealed gastric wall thickening and posterior wall transmural inflammation with ulceration. Repeat endoscopy with endoscopic ultrasound revealed the duodenal ulcer occupying 80% of the lumen, thickened prepyloric wall and perigastric lymph nodes. Lymph node biopsy did not reveal a malignancy. She was discharged with intact protein formula supplementation and a soft mechanical diet.

FIGURE 2

A, Endoscopic image before diagnosis revealing pyloric antral mass (left) and improved after treatment (right). B, Histology before diagnosis revealing increased eosinophils and neutrophils in the lamina propria (left) and after treatment without increased eosinophils (right). C, MRE before diagnosis revealing circumferential wall thickening (left); the other panel is blank because MRE was not obtained after treatment (right).

FIGURE 2

A, Endoscopic image before diagnosis revealing pyloric antral mass (left) and improved after treatment (right). B, Histology before diagnosis revealing increased eosinophils and neutrophils in the lamina propria (left) and after treatment without increased eosinophils (right). C, MRE before diagnosis revealing circumferential wall thickening (left); the other panel is blank because MRE was not obtained after treatment (right).

Close modal

Outpatient follow-up was poor, and she languished without a definitive diagnosis, returning 9 months later with epigastric pain, emesis, decreased oral intake, weight loss, and constipation. An upper endoscopy found the gastric antral submucosal mass now causing partial pyloric obstruction. Histology revealed increased eosinophils and neutrophils in the lamina propria (Fig 2B). MRE revealed circumferential wall thickening of the antrum, pylorus, and first portion of the duodenum (Fig 2C). Again, we considered IBD, focal lymphoma, and EGID, as well as other oncologic processes. Full-thickness biopsy results and a positron emission tomography scan result were negative. Although her hemoglobin was 8.5 g/dL, platelets 530 K/μL, and p-ANCA >100 U, other laboratory results were reassuring, including an erythrocyte sedimentation rate of 23 mm/hour, C-reactive protein 0.9 mg/dL, calprotectin 120 μg/g, and negative Hemoccult and celiac panel results. She was diagnosed with EGID, given her tissue eosinophilia with peak absolute eosinophil count 630 k/UL, percent eosinophils 8%, elevated total IgE 2808 kU/L, and high titers of IgE specific to wheat 14.3 kUA/L, egg 11.8 kUA/L, peanut 5.4 kUA/L, milk 3.1 kUA/L, and soy 2.1 kUA/L. Given her limited tolerance to EEN orally, a gastrostomy tube was placed. She was discharged with steroids and EEN therapy. The patient received a steroid taper plan, initially on 40 mg of prednisone for 3 months and then decreased by 5 mg weekly. Two months later, she had improved clinically with endoscopy demonstrating ulcer healing and improved antral mass, and biopsies revealing minimal eosinophilic involvement (1 eosinophil per HPF) (Fig 2 A and B). Her latest biopsy, 15 months after diagnosis, revealed disease remission on elemental formula as monotherapy. Because of clinical remission and favorable endoscopy, a control MRE was not performed.

EGIDs describe a diverse group of diseases characterized by eosinophil-rich inflammation of the bowel wall in the absence of known causes of eosinophilia.1  Clinical manifestations depend on the location and depth of eosinophil infiltration and include failure to thrive, abdominal pain, emesis, and diarrhea. In EGID, eosinophils are thought to aggregate in the mucosa, although epithelial infiltration is found in eosinophilic gastritis.2,3  These are the first reported pediatric cases of focal mural involvement as the isolated presenting feature of EGID. Only 1 similar case has been reported in the literature in a 65-year-old woman who presented with a nonmalignant gastric mass demonstrating transmural eosinophilic infiltration.4 

Eosinophils are physiologically present in relatively small numbers in the intestinal mucosa. Eosinophil presence is considered pathologic if it exceeds double the normal number of eosinophils in a particular part of the gastrointestinal tract. For example, the normal number of eosinophils in the duodenum is 26 per HPF. If a multiple of that number is found in the duodenum, as in 53 per HPF, it can be considered a pathologic process.5,6  Our patients had lamina propria eosinophils as high as >100 per HPF, leading to symptomatic partial intestinal obstruction and confirming the diagnosis of focal mural EGID.

These presentations mimic many conditions, including IBD and focal lymphomas, and warrant a heightened suspicion of EGID so as not to delay diagnosis.4,7  Without full-thickness biopsy to rule out lymphoma in the first case, we withheld steroid therapy until disease regression with EEN provided diagnostic confirmation. In the second case, we initiated systemic steroids with EEN, as oncologic processes were refuted by full-thickness biopsy. We also found that anticipating long-term EEN therapy with gastrostomy tube placement increased adherence and improved nutrition.

Dr Manrriquez was involved in conceptualizing the case report, analyzing and interpreting data, and drafting the case report; Dr Alharbi was involved in acquiring all figures, and analyzing and interpreting data; Dr Braskett was involved in acquiring, analyzing, and interpreting data; Dr Bhardwaj was involved in conceptualizing the case report and analyzing and interpreting data; and all authors were involved in revising the case report, approved the final version of the case report, approved the final manuscript as submitted, and agree to be accountable for all aspects of the work.

FUNDING: No external funding.

EEN

exclusive elemental nutrition

EGID

eosinophilic gastrointestinal disease

HPF

high-power field

IBD

inflammatory bowel disease

IgE

immunoglobulin E

MRE

magnetic resonance enterography

1
Egan
M
,
Furuta
GT
.
Eosinophilic gastrointestinal diseases beyond eosinophilic esophagitis
.
Ann Allergy Asthma Immunol
.
2018
;
121
(
2
):
162
167
2
Jawairia
M
,
Shahzad
G
,
Mustacchia
P
.
Eosinophilic gastrointestinal diseases: review and update
.
ISRN Gastroenterol
.
2012
;
2012
:
463689
3
Hurrell
JM
,
Genta
RM
,
Melton
SD
.
Histopathologic diagnosis of eosinophilic conditions in the gastrointestinal tract
.
Adv Anat Pathol
.
2011
;
18
(
5
):
335
348
4
Holroyd
DJ
,
Banerjee
S
,
Chaudhary
KS
,
Reshamwalla
S
,
Vijay
V
,
Ward
MW
.
Transmural eosinophilic gastritis with gastric outlet obstruction: case report and review of the literature
.
Ann R Coll Surg Engl
.
2010
;
92
(
4
):
W18
W20
5
Collins
MH
.
Histopathologic features of eosinophilic esophagitis and eosinophilic gastrointestinal diseases
.
Gastroenterol Clin North Am
.
2014
;
43
(
2
):
257
268
6
Collins
MH
,
Capocelli
K
,
Yang
GY
.
Eosinophilic gastrointestinal disorders pathology
.
Front Med (Lausanne)
.
2018
;
4
:
261
7
Saka
R
,
Sasaki
T
,
Matsuda
I
, et al
.
Chronic ileocolic intussusception due to transmural infiltration of diffuse large B cell lymphoma in a 14-year-old boy: a case report
.
Springerplus
.
2015
;
4
:
366

Competing Interests

POTENTIAL CONFLICT OF INTEREST: The authors have indicated they have no potential conflicts of interest to disclose.

FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.