Food Allergy in Infants With Atopic

Dermatitis: Limitations of Food-Specific

IgE Measurements
Jonathan M. Spergel, MD, PhDa, Mark Boguniewicz, MDb, Lynda Schneider, MDc, Jon M. Hanifin, MDd, Amy S. Paller, MDe,
Lawrence F. Eichenfield, MDf

abstract Children with atopic dermatitis (AD) have a higher risk for
BACKGROUND AND OBJECTIVES:
development of food allergies. The objective of this study was to examine incidence of food
allergy development in infants with AD and the predictive value of food-antigen–specific
immunoglobulin E measurements.
METHODS:This trial examined the long-term safety and efficacy of pimecrolimus cream 1% in
.1000 infants (3–18 months) with mild-to-severe AD without a history of food allergy. Food
allergy development was followed throughout a 36-month randomized double-blind phase
followed by an open-label (OL) phase up to 33 months. Additionally, sIgE for cow’s milk, egg
white, peanut, wheat, seafood mix, and soybean was measured by ImmunoCAP at baseline,
end of the double-blind phase, and end of OL phase.
RESULTS: Bythe end of the OL phase, 15.9% of infants with AD developed at least 1 food allergy;
allergy to peanut was most common (6.6%), followed by cow’s milk (4.3%) and egg white
(3.9%). Seafood, soybean, and wheat allergies were rare. Levels of sIgE for milk, egg, and
peanut increased with severity of AD, as determined by Investigator’s Global Assessment
score. We assigned sIgE decision points for the 6 foods and tested their ability to predict
definite food allergy in this population. Positive predictive values for published and newly
developed sIgE decision points were low (,0.6 for all values tested).
CONCLUSIONS: Ina large cohort of infants at risk for development of food allergy, sIgE levels were
not clinically useful for predicting food allergy development.

a
WHAT’S KNOWN ON THIS SUBJECT: Food The Children’s Hospital of Philadelphia, Perelman School of Medicine at University of Pennsylvania, Philadelphia,
Pennsylvania; bNational Jewish Health and University of Colorado School of Medicine, Denver, Colorado; cBoston
allergies are often thought to be a common Children’s Hospital, Harvard Medical School, Boston, Massachusetts; dOregon Health and Science University,
trigger in atopic dermatitis (AD). Serum Portland, Oregon; eNorthwestern University Feinberg School of Medicine, Chicago, Illinois; and fRady Children’s
Hospital, San Diego, University of California San Diego, San Diego, California
immunoglobulin E is frequently used to assess
food sensitization and clinical allergy, but few This trial has been registered at www.clinicaltrials.gov (identifier NCT00124709).
studies have assessed it longitudinally in infants www.pediatrics.org/cgi/doi/10.1542/peds.2015-1444
and children with AD. DOI: 10.1542/peds.2015-1444
WHAT THIS STUDY ADDS: In a large infant Accepted for publication Aug 31, 2015
population with mild to moderate AD, 15.9% of Address correspondence to Jonathan M. Spergel, MD, PhD, The Children’s Hospital of Philadelphia,
patients developed food allergy during the study. Perelman School of Medicine at University of Pennsylvania, 3550 Market St, Philadelphia, PA
19104–4399. E-mail: [email protected]
Serum immunoglobulin E decision points had
PEDIATRICS (ISSN Numbers: Print, 0031-4005; Online, 1098-4275).
low positive predictive values, indicating that
they are of limited use in this population. Copyright © 2015 by the American Academy of Pediatrics

Downloaded from www.aappublications.org/news at Indonesia:AAP Sponsored on November 15, 2020

ARTICLE PEDIATRICS Volume 136, number 6, December 2015
Atopic dermatitis (AD) is an contribute to the development of predict clinical allergy not if a food
inflammatory skin disorder that food allergies.14 Screening for food was causing AD in a population of
commonly presents in childhood. This allergy relies on measures of .1000 infants with AD but no history
disorder is associated with many antigenic sensitization, such as skin of food allergy.
comorbid conditions leading to prick tests and in vitro assays that
impaired overall health and increased measure food-antigen–specific
METHODS
health care utilization.1 Children with immunoglobulin E (sIgE). However,
AD are more likely to develop other .50% of patients who are Patients
atopic conditions, including food sensitized to a particular food based
This analysis was performed by
allergy, than children who have no on a positive screening test may
using data from SAM, a dual-phase
history of the disease.2–4 Previous not react to it in a food challenge,
study designed to explore the long-
clinical studies have documented highlighting the fact that
term safety and efficacy of 1%
greatly varying estimates of the rate sensitization does not indicate
pimecrolimus cream in infants with
of IgE-mediated food allergy in AD the presence of a clinical food
AD. To be eligible for the study,
patients, ranging from 15% to 40%. allergy.5,8,15,16 Because food
patients had to be 3 to 18 months
The most commonly cited range is challenges are time-consuming and
of age with a diagnosis of AD, as
30% to 40%, all in tertiary care potentially dangerous, it could be defined by the American Academy of
centers, reflecting more severely advantageous to determine the
Dermatology Consensus Conference
affected patients.5–8 In a recent study serum sIgE concentration that
criteria,21 for no more than
of 4453 infants in the HealthNuts predicts clinical food allergy. A
3 months before enrollment. Patients
cohort from Australia, the authors number of sIgE decision points have
also had to have at least mild
found that by 12 months of age, been developed from subject cohorts
disease activity, as defined by an
infants with eczema were 11 times of varying ages and with known or
Investigator’s Global Assessment
suspected food allergy (reviewed in
more likely to develop peanut allergy (IGA) score $2 at the start of the
and 5.8 times more likely to develop Sicherer and Sampson17). Recent study. Study participants were
egg allergy, compared with infants work by Fleischer and colleagues required have a parent or sibling
without eczema.4 These estimates are found that allergy testing has a high with a history of AD, allergic rhinitis,
also likely influenced by varying false-positive rate. In their study, allergic conjunctivitis, or asthma but
definitions of allergy, by study negative food challenges occurred
to have no atopic conditions other
methodologies, and by subject for 89% of 364 challenges in 125
than AD. Patients who had received
demographics. children evaluated for AD at a
treatment within 7 days before the
referral center.18 In the Australian
Because AD and food allergy often first application of study medication
HealthNuts cohort of 5276 infants, with topical or systemic agents
occur together, a persistent question egg sIgE $1.7 kUA/L and peanut
has been whether allergic reactions to known or suspected to affect AD
sIgE $34 kUA/L were associated were excluded from participation.
food contribute to AD signs and with 95% positive predictive values Topical corticosteroid use was
symptoms or clinical findings. for challenge-proven food allergy permitted before randomization.
Urticaria caused by food allergy has but this was not demonstrated
been implicated in exacerbating AD, for other foods.19 However, the Study Design
as histamine release, pruritus, and predictive value of specific IgE in In total, 36 clinics in the United
resultant scratching can exacerbate patients with AD is unclear. Most States participated in the SAM study,
existing skin lesions.9,10 Removal of recently, in the Learning Early About which was divided into 2 phases.
the allergenic food from the diet can Peanut allergy study, Du Toit and The first phase was a 36-month,
lead to resolution of AD in selected colleagues used egg allergy and AD randomized, double-blind (DB),
cases.9 However, formation of new as risk factors for developing peanut vehicle-controlled phase
eczematous lesions after a food allergy. In this study, they found (randomized 1:1 placebo cream
challenge is uncommon,9,11 and that 3.4% of the total high-risk versus 1% pimecrolimus cream) and
parental suspicions of food allergy population developed peanut the second was an open-label (OL)
have been shown to decrease when allergy. Interestingly, sIgE to peanut phase in which qualified patients
skin symptoms are controlled with was not predictive of peanut received 1% pimecrolimus cream
proper medication.12,13 allergy.20 (active drug) for up to 33 months
The sensitization pattern of IgE is In this prespecified analysis of or the patient’s sixth birthday,
influenced by maternal and patients enrolled in the Study of the whichever occurred sooner (see
environmental factors during the Atopic March (SAM), we examined Supplemental Information for
first year of life, which may the ability of sIgE concentrations to treatment plans).

Downloaded from www.aappublications.org/news at Indonesia:AAP Sponsored on November 15, 2020

PEDIATRICS Volume 136, number 6, December 2015 e1531
Development of allergies, including Total IgE and sIgE for cow’s milk, regression models.25 Performance
food allergies, was monitored peanut, wheat, seafood mix (codfish, characteristics of investigational
throughout the study as detailed in shrimp, tuna, salmon, and blue sIgE decision points were also
this article. Trial protocols were mussel), egg white, and soybean were calculated for each food allergy. The
approved by the independent ethics determined for all patients from investigational sIgE points were
committee or institutional review blood obtained by venipuncture determined for the ideal sensitivity
board of each study center and during visits 2, 14, and 20 and specificity by using logistic
written informed consent was (at weeks 1, 158, and 303, regression analysis. In addition, a
obtained from caregivers. respectively) using the ImmunoCAP receiver operating characteristic
assay (Phadia, Portage, MI). (ROC) analysis of definite food allergy
Assessment of AD Severity and for sIgE values was performed and
IgE-Mediated Food Allergy Statistical Methods the null hypothesis of whether the
The limit of detection was 0.1 kU/L area under the ROC curve was 0.5
Enrolled patients were evaluated
and lower limit of quantification was tested.26
for AD severity at each visit.
Investigators used the IGA scale, was 0.35 kU/L. sIgE decision points
in which scores represented the were selected by using published RESULTS
following: 0, clear; 1, almost clear; 90% positive predictive values in
older children (14 kU/L, 15 kU/L, and Patients and Development of
2, mild disease; 3, moderate disease;
7 kU/L for peanut, cow’s milk, and Allergies
4, severe disease; and 5, very severe
disease. In addition, Total Body egg white, respectively),23 and novel A total of 1091 patients were
Surface Affected (scale of 0% to decision points selected by the randomized; 1087 received at least
100%) and Eczema Area and investigators of 5 kU/L for peanut, 1 dose of study medication and were
5 kU/L for cow’s milk, and 2 kU/L for included in the safety population,
Severity Index (EASI)22 were
egg white for assessment based on and 1065 had at least 1 postbaseline
calculated at each visit.
optimal potential predictive values. efficacy measurement and were
Diagnosis of food allergy in enrolled For seafood, wheat, and soybean, a included in the ITT population.
patients used the criteria of decision point of 0.35 kU/L, the lower Patients were infants with a
Thompson and Hanifin,13 in which limit of quantification for the mean age of 7.3 months, were
clinical symptoms were assigned to a ImmunoCAP test, was used for data predominantly white, and most had
point system and the sum was used analysis. mild to moderate AD (92% with an
to assign definite (.25 points), IGA score of 2 or 3; Table 1). By the
probable (16–25 points), or possible Statistical Analysis end of the OL phase, 15.9% of
(5–15 points) food allergy.13 Major The safety population included all patients had developed a food
criteria (15 points each) included the randomized patients who were
presence of lip or face swelling, dispensed trial medication. The TABLE 1 Demographic Characteristics of
urticaria, nausea, vomiting, wheezing, intent-to-treat (ITT) population Subjects at Baseline
or respiratory distress after food included all randomized patients No. of subjects 1087
ingestion. Minor criteria (5 points who were dispensed trial medication
each) were repeated reaction on Gender, n (%)
and from whom at least 1 Boys 675 (62.1)
exposure to the same food, reaction postbaseline efficacy measurement Girls 412 (37.9)
happening within 30 minutes of was obtained. All statistical analyses Age, mo, mean (SD) 7.3 (3.9)
ingesting food, and reaction after were performed by using SAS, Race, n (%)
ingestion of milk, egg, soybean, White 748 (68.8)
versions 8.2 and 9.1.3 (SAS Institute,
African-American 146 (13.4)
wheat, peanut, or fish/seafood. Inc, Cary, NC). All statistical tests Other 193 (17.8)
Possible, probable, or definite food were conducted against a 2-sided Atopic Dermatitis Severity
allergies that occurred during the alternative hypothesis, using a IGA score,a n (%)
study were reported as adverse significance level of .05. The Mild 563 (51.8)
events and the suspected food Moderate 440 (40.5)
percentages of patients with food
Severe 78 (7.2)
as well as the grading category allergy were compared between the Very Severe 6 (0.6)
recorded. Definite food allergy 2 treatment groups using the TBSA, mean (SD)b 17.6 (15.7)
diagnoses were also reported on a Cochran-Mantel-Haenszel test, EASI score, mean (SD)c 6.5 (6.0)
nonskin atopic symptom case report adjusting for center and gender.24 For Baseline demographics at enrollment (safety population).
form, together with any exclusion each food allergy, the relationship TBSA, total body surface area affected.
a IGA, criteria for study participation, IGA $ 2.
diets prescribed by the treating between sIgE and food allergy was b TBSA scale 0% to 100%.

physician. investigated by using logistic c EASI scale 0–72.

Downloaded from www.aappublications.org/news at Indonesia:AAP Sponsored on November 15, 2020

e1532 SPERGEL et al
allergy. The mean 6 SD time to
first diagnosis of food allergy was
439 6 372 days and the median was
292 days. The most common food
allergies among those enrolled,
defined by definite adverse effects
described previously,13 were to
peanut, cow’s milk, and egg white,
occurring in 6.6%, 4.3%, and 3.9% of
the ITT population, respectively. In
contrast, only 0.4% of patients
developed allergy to seafood, 0.3%
to wheat, and 0.4% to soybean by
study visit 20, at the end of the OL
phase. The percentage of food
allergy decreased over time with the
exception of fish consistent with
the known natural history of food
allergy (Supplemental Table 5).
The magnitude of food allergy
development was similar to that of
other atopic conditions that occurred
during the study: 10.7% of patients
developed asthma, 14.1% developed
allergic conjunctivitis, and 22.4%
developed allergic rhinitis by the end
of the OL phase.

Relationship Between Food Allergy,

AD Severity, and AD Treatment
Patients’ AD was classified at
baseline according to IGA criteria.
The percentage of patients who
developed 1 or more food allergies
by the end of the study increased
with increasing IGA of AD severity at
baseline (Fig 1A). The relative
proportion of patients with 1, 2, and
3 or more food allergies within these
IGA groups did not change with
increasing severity; allergy to 1 food
continued to be most common
regardless of IGA score (n = 111
[81.6%] of 136 patients with allergy
to any food), whereas allergy to 3 or
4 foods was found in only 5 (3.7%)
of 136 patients with food allergy.
Total serum IgE and sIgE for milk,
egg, and peanut measured at the end
of the OL phase (Fig 1B and C) were
also increased in patients with
increasing baseline IGA scores. FIGURE 1
AD Severity (by baseline IGA score) and food hypersensitivity at the end of the OL phase (visit 20). A,
One secondary objective of the SAM
Frequency of patients with 1 or more food allergies grouped by baseline IGA score; B, Mean total IgE;
was to assess the development of and C, selected allergen-specific IgE measurements by baseline IGA score. Error bars represent
food allergy by patients in the standard error of the mean.

Downloaded from www.aappublications.org/news at Indonesia:AAP Sponsored on November 15, 2020

PEDIATRICS Volume 136, number 6, December 2015 e1533
pimecrolimus- and vehicle- TABLE 3 Logistic Regression Analysis for Peanut, Cow’s Milk, and Egg Allergy
treatment groups. There were no Food Explanatory Variable Odds Ratio (95% CI) Pa
significant differences between Peanut IgE at baseline (, 5 and $ 5 kU/L) b
5.118 (2.8–9.354) ,.0001
treatment groups in the percentage IGA group at baseline (2 and .2)c 1.699 (0.959–3.010) .0694
of patients with food allergy at the Treatment (control and pimecrolimus)d 1.491 (0.856–2.599) .1582
end of the DB phase (16.1% in the Age group at baseline (,12 mo and $ 12 mo)e 1.113 (0.523–2.366) .7817
pimecrolimus group and 13.7% in IgE at baseline (, 14 and $ 14 kU/L)b 3.507 (1.665–7.387) .0010
IGA group at baseline (2 and .2)c 1.882 (1.071–3.308) .0279
the vehicle group; P = .1196). Treatment (control and pimecrolimus)d 1.523 (0.880–2.635) .1324
Likewise, no significant difference in Age group at baseline (,12 mo and $ 12 mo)e 1.036 (0.492–2.181) .9265
rates between groups was seen at Cow’s milk IgE at baseline (,5 and $ 5 kU/L)b 11.694 (5.426–25.202) ,.0001
the end of the OL phase (17.3% IGA group at baseline (2 and .2)c 1.525 (0.735–3.167) .2573
of patients derived from the Treatment (control and pimecrolimus)d 2.410 (1.144–5.075) .0206
IgE at baseline (,15 and $ 15 kU/L)b 11.269 (4.587–27.687) ,.0001
pimecrolimus group and 14.5% IGA group at baseline (2 and .2)c 1.718 (0.839–3.517) .2573
from the vehicle group; P = .0718), Treatment (control and pimecrolimus)d 2.344 (1.121–4.899) .0206
during which time all patients Egg IgE at baseline (,2 and $ 2 kU/L)b 5.514 (2.852–10.660) ,.0001
received 1% pimecrolimus . IGA group at baseline (2 and .2)c 1.859 (0.927–3.730) .0827
Treatment (control and pimecrolimus)d 1.853 (0.943–3.643) .0735
Age group at baseline (,12 mo and $ 12 mo)e 0.728 (0.249–2.129) .5626
Relationship Between sIgE and Food IgE at baseline (,7 and $ 7 kU/L)b 2.077 (0.910–4.745) .0827
Allergy IGA group at baseline (2 and .2)c 2.290 (1.158–4.528) .0172
Treatment (control and pimecrolimus)d 1.846 (0.950–3.588) .0704
Analyses were performed to examine
Age group at baseline (,12 mo and $ 12 mo)e 0.601 (0.209–1.727) .3445
the relationship between measured
Logistic regression analysis for peanut, cow’s milk, and egg (ITT population, n = 1065). CI, confidence interval.
sIgE and clinical food allergy. A a From a Wald test.

logistic regression model for b IgE value at baseline classified into 2 groups: ,5 kU/L (reference category) and $5kU/L.
c IGA at baseline classified into 2 groups: IGA at baseline = 2 (reference category) and IGA at baseline $2.
development of clinical allergy to d Treatment during DB phase with 2 groups: control (reference category) and pimecrolimus.
each food (binary response) was e Age group at baseline classified into 2 groups: ,12 mo (reference category) and $12 mo of age. Age group is not

fitted by using the natural logarithm included in the regression analysis for cow’s milk because there were no patients with definite milk allergy $12 mo of
of sIgE at baseline as the continuous age.

explanatory variable. Baseline sIgE

values for cow’s milk, peanut, egg
white, and seafood mix were statistically significant predictors treatment group, and age group at
associated with a statistically of clinical allergy to these foods, baseline as explanatory variables
increased risk of developing allergies perhaps in part due to low numbers (Table 3). Odds ratios for the
to these foods (Table 2). Higher levels of patients with these allergies development of milk, egg, and
of AD severity (by IGA) were also (Table 2). It has also been shown that peanut allergy ranged from 2 to .11
predictive for development of food soy and wheat allergens perform for patients with sIgE at baseline
allergy (Table 3). poorly in in vitro testing. Based on above the cutoffs versus those with
the logistic regression model, the sIgE at baseline below cutoffs. In
In contrast, wheat and soybean estimated probabilities of developing addition, baseline IGA score was a
baseline sIgE levels were not definite food allergy by the end of statistically significant risk factor for
the OL phase (visit 20) by sIgE developing peanut and egg allergy.
decision level are shown for each of Comparison of pimecrolimus group
TABLE 2 Logistic Regression Analysis for the 6 foods in Fig 2. Even if a sIgE versus the vehicle group was
Food Allergy
cutoff at the upper limit of associated with milk allergy with an
Food Odds Ratio (95% CI) Pa quantification for the ImmunoCAP odds ratio of .2.
Cow’s milk 1.876 (1.606–2.192) ,.001 test (100 kU/L) was selected, the
Egg white 1.443 (1.258–1.655) ,.001
For the sIgE test to be clinically
estimated probability of patients useful, it would need to have high
Peanut 1.584 (1.410–1.779) ,.001
developing clinical food allergy did predictive values. Therefore,
Seafood mix 2.782 (1.493–5.195) .0013
Wheat 1.332 (0.765–2.318) .3109 not reach 0.90 for any food. In fact, performance characteristics were also
Soybean 1.730 (0.995–3.006) .0520 this probability did not reach 0.50 for determined for sIgE decision points
Food allergy as reported at study visits 3–20 (end of OL). most foods (Fig 2) indicating that by using sIgE measured at baseline or
Logistic regression analysis for reported food allergy screening sIgE was not useful. at the end of the DB phase (Table 4).
with log sIgE at baseline as the explanatory variable (ITT
population, n = 1065). CI, confidence interval. We examined the development of Negative predictive values (NPVs) for
a From a Wald test. The probability of having definite
food allergy by exploring risk factors baseline sIgE were high for all
food allergy was estimated from a simple logistic model
P – 1/(exp[–a – bx] + 1), where x is defined as the natural including baseline sIgE levels, IGA decision points tested (range of
logarithmic sIgE value at baseline. score (AD severity) at baseline, 0.94–1.00) and positive predictive

Downloaded from www.aappublications.org/news at Indonesia:AAP Sponsored on November 15, 2020

e1534 SPERGEL et al
FIGURE 2
The probability of a patient with definite allergy to food is estimated by a simple logistic model P = 1/[exp(2a 2 bx) + 1], where x = ln (sigE at baseline)
and the estimated values of a and b are as follows: (A) a = 22.7833, b = 0.6293; (B) a = 22.9095, b = 0.3666; (C) a = 22.4700, b = 0.4599; (D) a = 25.2812,
b= 0.2866; (E) a = 23.3837, b = 1.0242; (F) a = 24.9241, b = 0.5479.

values (PPVs) were low, with all phase, when patients had aged 3 A range of sIgE cutoff values was
values #0.3. Using these same sIgE years, PPVs were increased relative to also used to create ROC curves for
decision points at the end of the DB baseline, but values remained ,0.6. each food and to calculate the area
under the ROC (Supplemental
Figure 3).
TABLE 4 Performance Characteristics of sIgE for Food Allergy
Food IgE Cutoff, kU/L Sensitivity Specificity PPV NPV
DISCUSSION
Baseline
Peanut 5 0.36 0.91 0.22 0.95 In the SAM study, 15.9% of a
14 0.19 0.95 0.20 0.94 population of .1000 infants with
Cow’s milk 5 0.38 0.96 0.26 0.97 AD developed 1 or more
15 0.24 0.98 0.30 0.97 IgE-mediated food allergies. Unlike
Egg white 2 0.54 0.85 0.14 0.98
7 0.20 0.92 0.10 0.96
most studies of food allergy in
Seafood mix 0.35 0.50 0.98 0.11 1.00 children with AD, this study
Wheat 0.35 0.33 0.86 0.01 1.00 enrolled patients of all AD severities
Soybean 0.35 0.67 0.88 0.02 1.00 who had no history of food
End of DB phase allergy at baseline and followed
Peanut 5 0.62 0.89 0.43 0.95
14 0.49 0.93 0.48 0.93
them prospectively for allergy
Cow’s milk 5 0.42 0.97 0.41 0.97 development. The development of
15 0.35 0.99 0.56 0.97 food allergy during the SAM study
Egg white 2 0.45 0.86 0.16 0.96 was lower than the range reported
7 0.17 0.93 0.13 0.95 in many studies.1,5–8,19,20, This is
Seafood mix 0.35 1.00 0.90 0.06 1.00
Wheat 0.35 0.50 0.72 0.01 1.00
most likely because the study
Soybean 0.35 0.67 0.77 0.02 1.00 enrolled infants predominantly with
Performance characteristics of sIgE for predicting definite food allergy (ITT population, n = 1065) as reported at study mild to moderate AD (92%), in
visits 3–20 (end of OL). contrast to many previous studies

Downloaded from www.aappublications.org/news at Indonesia:AAP Sponsored on November 15, 2020

PEDIATRICS Volume 136, number 6, December 2015 e1535
whose subjects mostly had is much lower but agrees closely sIgE decision points, both
moderate to severe disease. with reported rates for the general published values and the novel
The rate of food allergy in our population: 0.1% for fish or shellfish, decision points used in this study,
cohort of children is similar to data 0.4% for wheat, and 0.4% for had high NPVs, in particular for
derived from the 2007 National soybean.30 peanut, egg white, and cow’s milk.
Survey of Children’s Health, a Thus, patients with mild AD with
Although this study used careful
prospective questionnaire-based sIgE levels below these cutoffs
determination of clinically relevant
would be unlikely to have or
study of 91 642 children aged food allergy based on history, we
develop these specific allergies, and
0–17 years. In this study, food did not confirm diagnoses with
would not benefit from food
allergy in the past 12 months was food challenges. sIgE testing was
challenges or elimination diets.
reported in 15.1% of children with performed only for common food
Similarly, elevated sIgE, as defined
AD, of whom 67% had mild and allergens, leaving the possibility
by the decision points tested, had
26% moderate AD.1,27 that reactions to less common
very low PPVs for food allergy, both
One limitation in our study is that food allergens were missed. In
for sIgE values at baseline and at
very young children (,4 months of addition, food challenges were
the end of the DB phase. PPVs were
age) were not enrolled, as this group not completed on all patients,
particularly low for seafood, wheat,
might have higher risk factors for suggesting possible false-positive
and soybean; other studies have
development of food allergy.4 An testing leading to a higher rate.
found sIgE to be nonpredictive for
IGA score $2 at baseline was However, the criteria were similar
these foods. 2,19,32,38 Thus, despite
predictive for development of food to what has been used in studies
an increased likelihood of allergy
allergy and, as expected, the sponsored by the National
development with increasing sIgE
percentage of food allergies Institutes of Health. 32 shown for cow’s milk, egg, and
increased with patients’ baseline AD Although food challenges are peanut, our data do not support the
severity. This ranged from 8% for time-consuming and potentially use of sIgE testing for the diagnosis
those with mild AD at baseline to dangerous for patients, they of food allergy in subjects without
24% for those with severe baseline represent the only definitive test for a history of reaction to that food.
AD. Food allergy in patients with AD food allergy. In practice, sIgE Consistent with our data, work
may result from food protein testing is widely used to screen from the National Jewish Hospital
absorption and sensitization for potential food allergy in showed that 89% of challenges
through damaged skin.28,29 This patients with AD. Maternal and in patients with AD who were
finding adds to the body of work environmental factors have been avoiding foods based on sIgE were
that supports the “atopic march,” the shown to influence the sensitization negative, indicating a high false-
hypothesis that AD pathology pattern of IgE during the first year positive rate for food allergy.18 In
predisposes individuals to the of life. This early sensitization is addition, the recent examination of
development of other atopic suspected to contribute to the development of peanut allergy in
diseases and comorbid conditions development of food allergies in the Learning Early About Peanut
associated with impaired overall at-risk infants.4,18,33 Our allergy study did not find that sIgE
health and increased health care assessment relied on strong was predictive for development of
utilization.1,28,29 historical indicators.13 Used peanut allergy. 20 Therefore, we
As in other studies, allergies to properly, sIgE tests then can be used recommend that children with
peanut, cow’s milk, and egg white to confirm or negate a diagnosis persistent AD in spite of optimized
were the most common, and were based on history.2 sIgE values at management should be screened
seen in a higher percentage of study which the probability of allergy for food allergy, as suggested by the
participants than in the general reaches $90% have been developed National Institutes of Health
population (3.9% to 6.6% of patients based on various patient cohorts; guidelines on food allergy.2
in this study had peanut allergy most of these cohorts have a strong
versus a reported 1.0% to 2.5% of history of suspected food allergy,
all North American children).2,30 and it is unclear whether such CONCLUSIONS
Estimated rates of allergy to wheat decision points are useful for Among infants with AD and a family
and soybean vary widely; in clinical prophylactic use in high-risk history of atopy, 15.9% developed a
studies of children and adolescents populations.6,19,32,34 The decision food allergy during a period of
with AD, rates as high as 30% have points have varied based on age and .3 years. Although peanut, cow’s
been reported.6,20,31 The 0.3% to patient history, and some cohorts milk, and egg white allergies were
0.4% prevalence found in this study do not reach the 90% level.34–37 common in this patient population,

Downloaded from www.aappublications.org/news at Indonesia:AAP Sponsored on November 15, 2020

e1536 SPERGEL et al
allergies to seafood, soybean, and food-elimination diets on the basis
wheat were quite rare, even in this of sIgE levels alone.
ABBREVIATIONS
higher risk population. Current AD: atopic dermatitis
US guidelines recommend ACKNOWLEDGMENTS DB: double-blind
consideration of food allergy EASI: Eczema Area and Severity
The authors acknowledge the
evaluation if there is persistent AD Index
investigators who participated in
in spite of optimal management IGA: Investigator’s Global
the SAM (please see online
and/or if there is a reliable history Assessment
Supplemental Information for full
of an immediate reaction after IgE: immunoglobulin E
listing). In addition, they thank
ingestion of a specific food.2 The ITT: intent to treat
Shangbin Liu, PhD, of Premier
results of this study of food allergy NPV: negative predictive value
Research Group Limited for
OL: open label
development in .1000 infants with statistical programming and
PPV: positive predictive value
AD do not support the use of sIgE analysis. Editorial assistance was
ROC: receiver operating
testing for these infants as a provided by Amy Troy, PhD, of
characteristic
diagnostic substitute for food Oxford PharmaGenesis, Inc, and was
SAM: Study of the Atopic March
challenge and should discourage funded by Novartis Pharmaceuticals
sIgE: specific immunoglobulin E
pediatricians from prescribing Corporation.

FINANCIAL DISCLOSURE: Dr Boguniewicz has consulted for Valeant. The other authors have indicated they have no financial relationships relevant to this article to
disclose.
FUNDING: Novartis Pharmaceutical Company funded the Study of Atopic March Trial.
POTENTIAL CONFLICT OF INTEREST: The authors have indicated they have no potential conflicts of interest to disclose.

REFERENCES
1. Silverberg JI, Simpson EL. Association children with atopic dermatitis. Dermatol Ther (Heidelb). 2006;19(2):
between severe eczema in children and Pediatrics. 1998;101(3). Available at: 97–103
multiple comorbid conditions and www.pediatrics.org/cgi/content/full/101/
13. Thompson MM, Hanifin JM. Effective
increased healthcare utilization. Pediatr 3/E8
therapy of childhood atopic dermatitis
Allergy Immunol. 2013;24(5):476–486
7. Sampson HA. Role of immediate food allays food allergy concerns. J Am Acad
2. Boyce JA, Assa’ad A, Burks AW, et al; hypersensitivity in the pathogenesis of Dermatol. 2005;53(2 suppl 2):S214–S219
NIAID-Sponsored Expert Panel. Guidelines atopic dermatitis. J Allergy Clin Immunol.
14. Depner M, Ege MJ, Genuneit J, et al;
for the diagnosis and management of 1983;71(5):473–480
PASTURE Study Group. Atopic
food allergy in the United States: report
of the NIAID-sponsored expert panel. J 8. Sampson HA, McCaskill CC. Food sensitization in the first year of life.
Allergy Clin Immunol. 2010;126(suppl 6): hypersensitivity and atopic dermatitis: J Allergy Clin Immunol. 2013;131(3):
S1–S58 evaluation of 113 patients. J Pediatr. 781–788
1985;107(5):669–675
3. Eichenfield LF, Hanifin JM, Beck LA, et al. 15. Perry TT, Matsui EC, Kay Conover-Walker
Atopic dermatitis and asthma: parallels 9. Breuer K, Heratizadeh A, Wulf A, et al. M, Wood RA. The relationship of allergen-
in the evolution of treatment. Pediatrics. Late eczematous reactions to food in specific IgE levels and oral food
2003;111(3):608–616 children with atopic dermatitis. Clin Exp challenge outcome. J Allergy Clin
Allergy. 2004;34(5):817–824 Immunol. 2004;114(1):144–149
4. Martin PE, Eckert JK, Koplin JJ, et al;
HealthNuts Study Investigators. Which 10. Burks W. Skin manifestations of food 16. Noh G, Ahn HS, Cho NY, Lee S, Oh JW. The
infants with eczema are at risk of food allergy. Pediatrics. 2003;111(6 pt 3): clinical significance of food specific
allergy? Results from a population-based 1617–1624 IgE/IgG4 in food specific atopic
cohort. Clin Exp Allergy. 2015;45(1): dermatitis. Pediatr Allergy Immunol.
11. Niggemann B, Sielaff B, Beyer K, Binder
255–264 2007;18(1):63–70
C, Wahn U. Outcome of double-blind,
5. Burks AW, Mallory SB, Williams LW, placebo-controlled food challenge tests 17. Sicherer SH, Sampson HA. Food allergy:
Shirrell MA. Atopic dermatitis: clinical in 107 children with atopic dermatitis. epidemiology, pathogenesis, diagnosis,
relevance of food hypersensitivity Clin Exp Allergy. 1999;29(1):91–96 and treatment. J Allergy Clin Immunol.
reactions. J Pediatr. 1988;113(3):447–451 2014;133(2):291–307, quiz 308
12. Rowlands D, Tofte SJ, Hanifin JM. Does
6. Eigenmann PA, Sicherer SH, Borkowski food allergy cause atopic dermatitis? 18. Fleischer DM, Bock SA, Spears GC, Wilson
TA, Cohen BA, Sampson HA. Prevalence of Food challenge testing to dissociate CG, Miyazawa NK, Gleason MC, et al. Oral
IgE-mediated food allergy among eczematous from immediate reactions. food challenges in children with a

Downloaded from www.aappublications.org/news at Indonesia:AAP Sponsored on November 15, 2020

PEDIATRICS Volume 136, number 6, December 2015 e1537
 diagnosis of food allergy. J Pediatr. 2011; 25. Bewick V, Cheek L, Ball J. Statistics Natural course of sensitization to cow’s
158(4):578–583.e1 review 14: Logistic regression. Crit Care. milk and hen’s egg in childhood atopic
2005;9(1):112–118 dermatitis: ETAC study group. Clin Exp
19. Peters RL, Allen KJ, Dharmage SC, et al;
Allergy. 2002;32(1):70–73
HealthNuts Study. Skin prick test 26. Bewick V, Cheek L, Ball J. Statistics review
responses and allergen-specific IgE 13: receiver operating characteristic 34. Boyano Martínez T, García-Ara C,
levels as predictors of peanut, egg, curves. Crit Care. 2004;8(6):508–512 Díaz-Pena JM, Muñoz FM, García
and sesame allergy in infants. J 27. Wahn U, Warner J, Simons FE, et al; Sánchez G, Esteban MM. Validity of
Allergy Clin Immunol. 2013;132(4): EPAAC Study Group. IgE antibody specific IgE antibodies in children with
874–880 responses in young children with atopic egg allergy. Clin Exp Allergy. 2001;
dermatitis. Pediatr Allergy Immunol. 31(9):1464–1469
20. Du Toit G, Roberts G, Sayre PH, et al; LEAP
Study Team. Randomized trial of peanut 2008;19(4):332–336 35. Celik-Bilgili S, Mehl A, Verstege A, et al.
consumption in infants at risk for peanut 28. Ker J, Hartert TV. The atopic march: The predictive value of specific
allergy. N Engl J Med. 2015;372(9): what’s the evidence? Ann Allergy Asthma immunoglobulin E levels in serum
803–813 Immunol. 2009;103(4):282–289 for the outcome of oral food challenges.
Clin Exp Allergy. 2005;35(3):268–273
21. Eichenfield LF, Hanifin JM, Luger TA, 29. Spergel JM. From atopic dermatitis to
Stevens SR, Pride HB. Consensus asthma: the atopic march. Ann Allergy 36. García-Ara C, Boyano-Martínez T, Díaz-Pena
conference on pediatric atopic Asthma Immunol. 2010;105(2):99–106; JM, Martín-Muñoz F, Reche-Frutos M,
dermatitis. J Am Acad Dermatol. 2003; quiz 107–109, 117 Martín-Esteban M. Specific IgE levels in the
49(6):1088–1095 diagnosis of immediate hypersensitivity to
30. Sicherer SH, Sampson HA. Food allergy. cows’ milk protein in the infant. J Allergy
22. Hanifin JM, Thurston M, Omoto M, Cherill J Allergy Clin Immunol. 2010;125(2 suppl Clin Immunol. 2001;107(1):185–190
R, Tofte SJ, Graeber M; EASI Evaluator 2):S116–S125
Group. The eczema area and severity 37. Sicherer SH, Sampson HA. Cow’s milk
31. Lo YC, Yang YH, Chiang BL. Food-specific protein-specific IgE concentrations in two
index (EASI): assessment of reliability in immunoglobulin E among children with age groups of milk-allergic children and
atopic dermatitis. Exp Dermatol. 2001; atopic dermatitis: a retrospective study. in children achieving clinical tolerance.
10(1):11–18 J Microbiol Immunol Infect. 2005;38(5): Clin Exp Allergy. 1999;29(4):507–512
23. Sampson HA. Utility of food-specific IgE 338–342
38. García BE, Gamboa PM, Asturias JA, et al;
concentrations in predicting 32. Sicherer SH, Wood RA, Vickery BP, et al. Clinical Immunology Committee; Spanish
symptomatic food allergy. J Allergy Clin The natural history of egg allergy in an Society of Allergology and Clinical
Immunol. 2001;107(5):891–896 observational cohort. J Allergy Clin
Immunology. Guidelines on the clinical
24. Stokes M, Davis C, Koch G. Categorical Immunol. 2014;133(2):492–499
usefulness of determination of specific
Data Analysis Using SAS System. 2nd ed. 33. Wolkerstorfer A, Wahn U, Kjellman NI, immunoglobulin E to foods. J Investig
Cary, NC: SAS Institute, Inc.; 2000 Diepgen TL, De Longueville M, Oranje AP. Allergol Clin Immunol. 2009;19(6):423–432

Downloaded from www.aappublications.org/news at Indonesia:AAP Sponsored on November 15, 2020

e1538 SPERGEL et al
 Food Allergy in Infants With Atopic Dermatitis: Limitations of Food-Specific
IgE Measurements
Jonathan M. Spergel, Mark Boguniewicz, Lynda Schneider, Jon M. Hanifin, Amy S.
Paller and Lawrence F. Eichenfield
Pediatrics 2015;136;e1530
DOI: 10.1542/peds.2015-1444 originally published online November 23, 2015;

Updated Information & including high resolution figures, can be found at:
Services http://pediatrics.aappublications.org/content/136/6/e1530
References This article cites 35 articles, 3 of which you can access for free at:
http://pediatrics.aappublications.org/content/136/6/e1530#BIBL
Subspecialty Collections This article, along with others on similar topics, appears in the
following collection(s):
Allergy/Immunology
http://www.aappublications.org/cgi/collection/allergy:immunology_s
ub
Permissions & Licensing Information about reproducing this article in parts (figures, tables) or
in its entirety can be found online at:
http://www.aappublications.org/site/misc/Permissions.xhtml
Reprints Information about ordering reprints can be found online:
http://www.aappublications.org/site/misc/reprints.xhtml

Downloaded from www.aappublications.org/news at Indonesia:AAP Sponsored on November 15, 2020

 Food Allergy in Infants With Atopic Dermatitis: Limitations of Food-Specific
IgE Measurements
Jonathan M. Spergel, Mark Boguniewicz, Lynda Schneider, Jon M. Hanifin, Amy S.
Paller and Lawrence F. Eichenfield
Pediatrics 2015;136;e1530
DOI: 10.1542/peds.2015-1444 originally published online November 23, 2015;

The online version of this article, along with updated information and services, is
located on the World Wide Web at:
http://pediatrics.aappublications.org/content/136/6/e1530

Data Supplement at:

http://pediatrics.aappublications.org/content/suppl/2015/11/18/peds.2015-1444.DCSupplemental

Pediatrics is the official journal of the American Academy of Pediatrics. A monthly publication, it
has been published continuously since 1948. Pediatrics is owned, published, and trademarked by
the American Academy of Pediatrics, 345 Park Avenue, Itasca, Illinois, 60143. Copyright © 2015
by the American Academy of Pediatrics. All rights reserved. Print ISSN: 1073-0397.

Downloaded from www.aappublications.org/news at Indonesia:AAP Sponsored on November 15, 2020