new england journal
n engl j med
data, and knowledge about risk adjustment to make
analogous public reporting feasible for most other
areas of medicine.
Thomas H. Lee, M.D.
Partners Healthcare System
Boston, MA 02199
Gregg S. Meyer, M.D.
Massachusetts General Physicians Organization
Boston, MA 02114
Troyen A. Brennan, M.D., J.D., M.P.H.
Brigham and Women’s Hospital
Boston, MA 02115
to the editor: Eosinophilic esophagitis is an
emerging disease worldwide, as documented by
recent case series from Switzerland, Australia, Italy,
Spain, Japan, England, and the United States.1,2
Eosinophilic esophagitis mimics gastroesopha-
geal reflux disease and may result in narrowing
and stricture of the esophagus.1-3 This disease is
differentiated from reflux esophagitis on the ba-
sis of the magnitude of mucosal eosinophilia and
a lack of response to acid suppression.4 We report
findings from a population-based demographic
study of the pediatric population with eosinophilic
esophagitis residing in the vicinity of our medical
center (Hamilton County, Ohio), a region with a
single pediatric gastroenterology and pathology
Cases of eosinophilic esophagitis were system-
atically identified from our institution’s pathology
database; the criteria were the presence of epithe-
lial proliferative changes (e.g., thickening of the
basal epithelial layer and elongation of the papil-
lae), a minimum of 24 eosinophils per high-power
field (¬400) in the distal esophagus, and the ab-
sence of eosinophilia in any other intestinal seg-
ment. Data from the 2000 Census for the popu-
lation 0 to 19 years of age were used to calculate
During the years 1991 through 2003, there were
924 possible cases of eosinophilic esophagitis, of
which 315 were subsequently found to meet the di-
agnostic criteria. Of the 315 cases, only 2.8 percent
had been identified before 2000. A total of 103 pa-
tients with conditions that met the histologic crite-
ria for the disease resided within Hamilton Coun-
ty at the time of diagnosis. More than 70 percent
of these 103 patients had coexisting eosinophilic
involvement in the proximal esophagus. Approxi-
mately 71 percent were male, with a mean (±SD)
age of 10.5±5.4 years. The patients presented with
the typical symptoms and atopic history that have
been described previously (Tables 1 and 2).1-4 How-
ever, our demographic analysis revealed a strong
familial pattern (Table 2), including three sibling
pairs; the mother of one of the pairs of siblings was
also given a diagnosis of eosinophilic esophagitis
at our institution. The familial cases of eosinophil-
* Patients may have had more than one symptom, but only the most prominent
symptom is included here. The median age varied significantly according to
the primary symptom (P<0.001 by the Kruskal–Wallis test).
Table 1. Presenting Symptoms among 103 Pediatric Patients with Eosinophilic
(Interquartile Range)No. (%)
Feeding disorder 2.0 (1.2–6.2)14 (13.6)
Vomiting 8.1 (3.5–12.3)27 (26.2)
Abdominal pain12.0 (9.6–15.2)27 (26.2)
Dysphagia13.4 (10.0–16.7)28 (27.2)
Food impaction16.8 (13.7–19.6)7 (6.8)
* There was a history of a positive response to skin-prick
test or to a radioallergosorbent test, or an anaphylactic
response to a specific dietary antigen.
†The diagnosis of eosinophilic esophagitis was made
at our institution for patients with a family history of the
disease (i.e., in a first-degree relative).
Table 2. History of Atopy in the 103 Pediatric Patients.
Possible food allergy*
Family history of atopic disease
Family history of eosinophilic esophagitis†
Family history of esophageal dilatation
The New England Journal of Medicine
Downloaded from nejm.org on September 13, 2015. For personal use only. No other uses without permission.
Copyright © 2004 Massachusetts Medical Society. All rights reserved.