Original Article

The Association of the Delayed Introduction of

Cow’s Milk with IgE-Mediated Cow’s Milk Allergies
Yutaro Onizawa, MDa, Emiko Noguchi, MD, PhDb, Masafumi Okada, MD, PhDc, Ryo Sumazaki, MD, PhDd, and
Daisuke Hayashi, MDa Ibaraki, Japan

What is already known about this topic? A few studies have suggested that the early introduction of cow’s milk formula
might be protective against IgE-mediated cow’s milk allergies. However, these studies have limited data about the factors
affecting parental choices regarding feeding patterns.

What does this article add to our knowledge? In this case-control study, the delayed introduction of cow’s milk formula
was an independent risk factor for an IgE-mediated cow’s milk allergy. Most parents chose exclusive or almost exclusive
breastfeeding because of reasons other than allergy prevention.

How does this study impact current management guidelines? In contrast to the current guidelines, our result suggests
that regular consumption of cow’s milk formula might play an important role in preventing IgE-mediated cow’s milk
allergies.

BACKGROUND: Although exclusive breastfeeding at least 4 to
6 months has been recommended to prevent IgE-mediated cow’s
milk allergy (IgE-CMA), early introduction of food allergens has
received a lot of attention in recent years for the prevention of
food allergies.
OBJECTIVES: We aimed to determine whether IgE-CMA is
associated with a feeding pattern in early infancy.
METHODS: In a case-control study, we retrospectively
compared the patient background, past history of atopic
dermatitis, bronchial asthma, family history of allergic diseases,
feeding patterns in early infancy, and the reason for choosing
early infancy feeding patterns of patients with IgE-CMA with
age- and sex-matched healthy controls using a questionnaire
completed by their mothers. To minimize the influence of con-
founders, we also compared patients with IgE-CMA with those
with IgE-mediated egg allergy (IgE-EA).
RESULTS: A total of 51 patients with IgE-CMA were compared
with 102 controls (1:2 matching) and 32 unmatched patients
a
Department of Pediatrics, Ryugasaki Saiseikai Hospital, Ibaraki, Japan
b
Department of Medical Genetics, Faculty of Medicine, University of Tsukuba,
c
Ibaraki, Japan
Department of Epidemiology, Faculty of Medicine, University of Tsukuba, Ibaraki,
Japan
d
Department of Child Health, Faculty of Medicine, University of Tsukuba, Ibaraki,
Japan
This study was supported by Management Expenses Grants of University of
Tsukuba.
Conflicts of interest: The authors declare that they have no relevant conflicts.
Received for publication July 17, 2015; revised December 29, 2015; accepted for
publication January 21, 2016.
Corresponding author: Emiko Noguchi, MD, PhD, Department of Medical Genetics,
University of Tsukuba, 1-1-1 Tennoudai, Tsukuba, Ibaraki 305-8575, Japan.
E-mail: enoguchi@md.tsukuba.ac.jp.
2213-2198
Ó 2016 American Academy of Allergy, Asthma & Immunology
http://dx.doi.org/10.1016/j.jaip.2016.01.012
with IgE-EA. In a multivariable logistic regression analysis, the
adjusted odds ratio of delayed (started more than 1 month after
birth) or no regular cow’s milk formula (less than once daily)
was 23.74 (95% CI, 5.39-104.52) comparing the CMA group
with the Control group, and 10.16 (95% CI, 2.48-41.64)
comparing the CMA group with the EA group. Only 3 (6.5%), 2
(4.8%), and 3 (14.3%) mothers in the CMA group, the Control
group, and the EA group chose “To prevent allergic disease” as a
reason for choosing exclusive or almost exclusive breastfeeding in
the first month of life, respectively.
CONCLUSIONS: The early introduction of cow’s milk formula
is associated with lower incidence of IgE-CMA. Ó 2016
American Academy of Allergy, Asthma & Immunology (J Allergy
Clin Immunol Pract 2016;4:481-8)
Key words: Food allergy; IgE-mediated cow’s milk allergy;
Breast milk; Allergy prevention
Cow’s milk allergy (CMA) is the second most common food
allergy in the United States1 and Japan2 and can be potentially
fatal.3 CMA was shown to affect between 0.5% and 4.9% of
children, although the prevalence varies because of different
diagnostic criteria, study designs, methodologies, and the ethnic
background of participants.4-9 Among CMA-affected children,
48% to 73% are considered to have IgE-mediated CMA (IgE-
CMA).4,9,10 The outgrowth rate of IgE-CMA also differs among
studies. Earlier studies have demonstrated good prognoses in
patients with CMA. The resolution rate of CMA was reported to
be 78% at 6 years of age in a study by Bishop et al in 1984,11
92% at 5 years of age in a study by Host et al in 1985,4 and
85% at 8 years of age in a study by Saarinen et al in 1994.10
However, studies from the 2000s presented worse prognoses.
A large study including 807 subjects by Skripak et al12 reported a
resolution rate of 19% at 4 years of age and 42% at 8 years of age
from 1994 to 2007. Likewise, other studies demonstrated a poor
prognosis, including a 53% resolution rate at 5 years of age in a

481
482 ONIZAWA ET AL
Abbreviations used
AD- Atopic dermatitis
AAP- American Academy of Pediatrics
ATS-DLD- American Thoracic Society-Division of Lung Diseases
BA- Bronchial asthma
CM- Cow’s milk
CMA- Cow’s milk allergy
EA- Egg allergy
IgE-CMA- IgE-mediated cow’s milk allergy
IgE-EA- IgE-mediated egg allergy
ISAAC- International Study of Asthma and Allergies in
Childhood
OR- Odds ratio
study by Wood et al13 in 2009 and 57% at 5 years of age in a
study by Elizur et al14 in 2004-2006.
To prevent food allergies and IgE-CMA in early childhood,
exclusive breastfeeding for at least 4 to 6 months was first rec-
ommended by the Section of Paediatrics of the European
Academy of Allergology and Clinical Immunology and the
American Academy of Pediatrics (AAP) in the early 2000s.15,16
The delayed introduction of potentially allergenic foods in
high-risk infants (eg, dairy products delayed until 1 year; eggs
until 2 years; and peanuts, nuts, and fish until 3 years of age) was
also recommended by the AAP in 2000. In 2008, the AAP issued
a new guideline that there is no current convincing evidence that
delaying the introduction of these foods beyond 4 to 6 months
has a significant protective effect on the development of allergic
disease, although solid foods should not be introduced before
4 to 6 months of age.15
Recently, the early introduction of food allergens has received
much attention in the prevention of food allergies. In 2008, Du
Toit et al17 compared the prevalence of peanut allergies between
Jewish school children in the United Kingdom and those in
Israel and reported that the prevalence of peanut allergies in the
United Kingdom was 10 times higher than that in Israel (1.85%
vs 0.17% for the prevalence of peanut allergies, respectively) and
that the early consumption of peanuts in the first years of life is
common in Israel but not in the United Kingdom. Subsequently,
they conducted a randomized, controlled, open-label trial to
determine whether the early introduction of dietary peanuts
could serve as an effective strategy for the prevention of peanut
allergies and found that the avoidance of peanuts in early infancy
was associated with the development of peanut allergies (13.7%
in the avoidance group vs 1.9% in the consumption group for
the prevalence of peanut allergies, respectively).18 In addition to
peanuts, the early introduction of eggs,19 cereal grains,20 and
fish21 was also associated with lower rates of food allergies in a
cross-sectional study and a birth cohort study.
Previous studies have suggested the efficacy of the early
introduction of cow’s milk (CM) to prevent IgE-CMA. In a large
birth cohort study of 13,019 infants, Katz et al9 reported that
only 0.05% of the infants who started on regular CM formula
within the first 14 days developed IgE-CMA, whereas 1.75% of
the infants who started formula between the ages of 105 and 194
days developed IgE-CMA. The other birth cohort study of 6209
infants showed that breastfeeding during the first 2 months at
home, whether exclusively or combined with infrequent exposure
to small amounts of CM formula, was an independent risk factor
for IgE-CMA.22 CM can be introduced soon after birth in some
J ALLERGY CLIN IMMUNOL PRACT
MAY/JUNE 2016
babies because most infant formula is made with CM, although
other foods, such as eggs and peanuts, are, in most cases,
introduced after the baby starts solid food at 4 to 6 months of
age. Therefore, the factors that affect the feeding patterns of CM
formula can be modifiable risk and/or protective factors for
developing a milk allergy. The aim of this study was to clarify
whether the early introduction of CM formula was associated
with IgE-CMA in Japan and to elucidate factors affecting the
timing of the introduction of CM formula to babies.
METHODS
Study design and population
This research and study protocol was approved by the ethical
committee of Ryugasaki Saiseikai Hospital and the University of
Tsukuba Hospital. The patients and controls were recruited at
Ryugasaki Saiseikai Hospital from November 2014 to February
2015. Information regarding past histories of allergic diseases, family
histories of allergic diseases, and feeding patterns in early infancy was
retrospectively collected using a questionnaire after parents gave
informed consent to participate in the study.
The CMA group included patients with IgE-CMA who visited
Ryugasaki Saiseikai Hospital between October 2013 and February
2015. IgE-CMA was defined as having immediate-type allergic
reactions within 2 hours after dairy intake (eg, urticaria, wheezing,
vomiting, and diarrhea), positive CM-specific IgE (
0.7 kUA/L),
and a diagnosis by a board-certified allergy specialist (D.H.). The
Control group consisted of age- and sex-matched children with 1:2
matching who visited the hospital between November 2014 and
February 2015 for a common cold or routine immunization. The
inclusion criterion of control subjects was not having a history of
CMA or other food allergies (Control). We also recruited patients
with egg allergy (EA group) with IgE-mediated egg allergies (IgE-
EA) without a history of milk-related food allergies who visited the
hospital in the same time periods. IgE-EA was defined as having
both immediate-type allergic reaction, positive egg-specific IgE
(
0.7 kU/L), and a diagnosis by the specialist (D.H.). Children
were excluded from the study if they (i) were younger than 1 year old
at the enrollment, (ii) were a preterm birth (gestational age < 36
weeks), (iii) were of low birth weight (<2500 g), (iv) had congenital
anomalies, or (v) had non-IgE-CMA.
The questionnaires were completed by parents at the hospital
visit. Seven patients in the CMA group who had no appointment
between November 2014 and February 2015 completed the ques-
tionnaires by a telephone interview administered by the pediatrician
(Y.O.).
Feeding patterns
The questionnaire included the following information about
feeding patterns in early infancy.
1. Any use of CM formula at the maternity hospital.
2. The timing of the introduction of nonregular CM formula.
3. The timing of the introduction of regular CM formula.
4. Feeding patterns in the first month of life and the reason for
choosing these patterns.
5. The timing of discontinuation of CM formula and the reason for
discontinuation.
Regular CM formula was defined as the consumption of CM
formula at least once daily (regardless of the amount). Early regular
continuous CM formula was defined as regular consumption of CM
J ALLERGY CLIN IMMUNOL PRACT ONIZAWA ET AL 483
VOLUME 4, NUMBER 3

formula starting within the first month of life and continuing until 6 CMA patients in Control patients EA patients
months or the onset of IgE-CMA. Ryugasaki Saiseikai approached approached
Hospital (n = 57) (n = 104) (n = 33)
Feeding patterns in the first month of life were categorized as
follows: Excluded (n = 3)
LBWI (n = 1)
1. Exclusive breastfeeding: no CM formula use. <1 year old (n = 2)
2. Almost exclusive breastfeeding: CM formula use less than once Declined to Declined to
Approached participate participate
daily. (n = 54) (n = 2) (n = 1)
3. Mixed 1: CM formula use at least once daily with a predomi-
nance of breast milk. Unable to contact
4. Mixed 2: CM formula use at least once daily with a predomi- (n = 3)

nance of CM.
5. Exclusive CM: no breast milk use. Analyzed Analyzed Analyzed
(n = 51) (n = 102) (n = 32)

Definition of atopic dermatitis and bronchial asthma
Any history of atopic dermatitis (AD) and bronchial asthma (BA)
was assessed by the questionnaire based on the International Study
of Asthma and Allergies in Childhood (ISAAC) core questionnaire23
and the Japanese Guideline for Atopic Dermatitis 2014.24 We asked
2 questions regarding AD: “Has your child ever had an itchy rash
that came and went for at least 6 months? (or 2 months if 1 year
old)” and “Has this itchy rash at any time affected any of the
following places: the folds of the elbows, behind the knees, in front
of the ankles, under the buttocks, or around the neck, ears, or eyes?”
Subjects who responded positively to both questions were regarded
as having AD. We also asked a question regarding BA based on the
American Thoracic Society-Division of Lung Diseases (ATS-DLD)
Epidemiology Questionnaire,25 and those who responded positively
to the following question were regarded as having BA: “Has your
child experienced 2 or more attacks of wheezing without febrile
symptoms?” Information on physician-diagnosed AD and physician-
diagnosed BA was similarly collected.
Statistical analysis
Statistical analysis was performed with SPSS 21.0 software (IBM,
Armonk, NY). We analyzed quantitative variables using t-tests and
Mann-Whitney U-tests. Normally distributed data were analyzed
using t-tests, and data not normally distributed were analyzed using
Mann-Whitney U-tests. Fisher’s exact test was used to evaluate the
categorical variables. A multivariable logistic regression analysis was
performed to identify independent risk factors for IgE-CMA. The
presence or absence of CMA was treated as the dependent variable, and
the risk factors (parental history of allergic diseases [AD, BA, food
allergy, and perennial allergic rhinitis], age at delivery, a past history of
physician-diagnosed AD and BA, and feeding patterns in early infancy
[“Delayed or no regular CM formula” or “No early regular continuous
CM formula”]) were treated as the independent variable, and all of
them were simultaneously included in the model. All reported P values
were 2-tailed. The level of statistical significance was set at P < .05.
RESULTS
Study population
Among the 57 patients with IgE-CMA who were treated in
Ryugasaki Saiseikai Hospital, 1 patient was excluded for low
birth weight and 2 patients for age at enrollment (<1 year old)
(Figure 1). Of the remaining 54 patients, 3 had no appointment
during the study period and were not able to be contacted by
telephone. A total of 51 patients completed the questionnaires,
and 7 of them completed the questionnaire via telephone
interviews. Twenty-six (51.0%) patients in the CMA group had
undergone food challenge for the diagnosis. Although 20
CMA group Control group EA group
FIGURE 1. A flow chart of the study design. CMA, cow’s milk
allergy; EA, egg allergy; LBWI, low birth weight infant.
(39.2%) patients in the CMA group did not undergo food
challenge, they experienced immediate-type allergic reactions
more than once. The remaining 5 (9.8%) patients in the CMA
group experienced allergic symptoms only once; however, 2 of
them experienced anaphylaxis after dairy intake (wheezing or
vomiting in addition to urticaria).
In the Control group, the parents of 104 children who visited
the hospital for a common cold or routine immunization were
initially recruited, and 2 declined to participate in the study. The
remaining 102 age- and sex-matched children (1:2 matching)
completed the questionnaires.
In the EA group, 33 patients with IgE-EA who were treated in
Ryugasaki Saiseikai Hospital were recruited to fill out the ques-
tionnaires and 1 declined to participate in the study. A total of 32
patients completed the questionnaires. Nineteen (59.4%)
patients in the EA group had been diagnosed on the ground of
multiple episodes of immediate-type allergic reactions, and
9 patients (28.1%) had undergone food challenge for diagnosis.
The remaining 4 (12.5%) patients experienced allergic symp-
toms once before enrollment.
Risk factors of IgE-CMA
The characteristics of patients and controls are shown in
Table I. Sex, gestational age, and birth weight were not signifi-
cantly different between groups (P > .05). Maternal age at de-
livery, paternal asthma, paternal rhinitis, maternal asthma,
maternal food allergy, and pet ownership were significantly
different between the CMA group and the Control group
(P < .05). Only paternal rhinitis and maternal asthma were
significantly different between the CMA group and the EA group
(P < .05). The frequency of patients with AD and BA was
significantly higher in the CMA group than in the Control group
(P < .001 for AD and P ¼ .012 for BA).
Table II and Figure E1 (available in this article’s Online
Repository at www.jaci-inpractice.org) show the feeding patterns
in early infancy. CM formula use at the maternity hospital was
not significantly different between groups (P > .05). CM
formula use after discharge from the maternity hospital through
the first month of life was less frequent in the CMA group than
in the Control group (P < .001). In addition, the rate of the
introduction of regular CM formula within the first month of life
484 ONIZAWA ET AL J ALLERGY CLIN IMMUNOL PRACT
MAY/JUNE 2016

TABLE I. Characteristics of the patients and control subjects

CMA (n [ 51) Control (n [ 102) P value EA (n [ 32) P value

Male, n (%) 37 (72.5) 74 (72.5) e 22 (68.8) .805

Age (mo), mean (SD) 63.7 (36.0) 64.7 (36.4) .807 56.4 (37.9) .244
Gestational age (wk), mean (SD) 38.8 (1.3) 39.0 (1.2) .449 38.8 (1.5) .882
Birth weight (g), mean (SD) 3176.1 (346.1) 3082.8 (286.2) .191 3119.6 (348.0) .549
Paternal age at delivery (y), mean (SD) 34.1 (5.7) 33.0 (6.3) .300 32.8 (5.1) .276
Maternal age at delivery (y), mean (SD) 32.4 (5.3) 30.4 (5.3) .036 31.7 (4.8) .582
Maternal BMI 20.4 (2.3) 21.1 (3.5) .226 20.0 (1.8) .578
Only child, n (%) 12 (23.5) 28 (27.4) .698 7 (21.9) 1.000
Paternal asthma, n (%) 8 (15.7) 4 (3.9) .021 4 (12.5) .759
AD, n (%) 7 (13.7) 6 (5.8) .126 3 (9.4) .734
Food allergy, n (%) 2 (3.9) 3 (2.9) 1.000 4 (12.5) .199
Rhinitis*, n (%) 19 (37.3) 19 (18.6) .017 6 (15.6) .047
Maternal asthma, n (%) 13 (25.5) 6 (5.9) .001 2 (6.3) .039
AD, n (%) 8 (15.7) 7 (6.9) .093 5 (15.6) 1.000
Food allergy, n (%) 8 (15.7) 3 (2.9) .007 4 (12.5) .759
Rhinitis*, n (%) 18 (35.3) 27 (26.5) .266 6 (18.6) .138
Pet ownership, n (%) 9 (17.6) 35 (34.3) .038 8 (25.0) .577
Maternal smoking, n (%) 13 (25.5) 30 (29.4) .704 12 (37.5) .326
CM- or egg-specific IgE†, geometric mean (range) 9.40 (0.72 to >100) NA e 8.33 (0.80 to >100) e
Physician-diagnosed AD, n (%) 23 (45.1) 6 (5.9) <.001 14 (43.8) 1.000
ADz, n (%) 34 (66.7) 18 (17.6) <.001 24 (75.0) .470
Physician-diagnosed BA, n (%) 19 (37.3) 12 (11.8) <.001 8 (25.0) .337
BAz, n (%) 21 (41.2) 21 (20.6) .012 9 (28.1) .251

AD, Atopic dermatitis; BA, bronchial asthma; BMI, body mass index; CM, cow’s milk; CMA, cow’s milk allergy; EA, egg allergy; NA, not available; SD, standard deviation.
Paternal and maternal ages at delivery were compared using t-tests.
Age, gestational age, birth weight, and maternal BMI were compared using Mann-Whitney U-tests.
Fisher’s exact test was used to evaluate the rest of the categorical variables.
Bold values indicate significant differences (P < .05).
*Allergic perennial rhinitis only.
†Values > 100 were treated as 100 to calculate the geometric mean.
zDefined in the “Methods” section.

TABLE II. Comparison of feeding patterns between the CMA group and the Control group and between the CMA group and the EA group
CMA Control EA
(n [ 51) (n [ 102) P value (n [ 32) P value

Supplementary CM formulain maternity hospital, n (%) 31 (60.8) 73 (70.6) .273 21 (65.6) .816
Exclusive breastfeeding*, n (%) 35 (68.6) 24 (23.5) <.001 14 (43.8) .038
Early regular CM formula†, n (%) 6 (11.8) 60 (58.8) <.001 14 (43.8) .001
Early regular continuous CM formulaz, n (%) 2 (3.9) 52 (51.0) <.001 11 (34.4) <.001
CM, Cow’s milk; CMA, cow’s milk allergy; EA, egg allergy.
Fisher’s exact test was used to evaluate the categorical variables.
Bold values indicate significant differences (P < .05).
*After discharge from maternity hospital to the first month of life.
†Started consumption of CM formula on a regular basis (at least once daily) within the first month of life.
zStarted regular consumption of CM formula within the first month of life and continued until 6 mo or until the onset of CMA.

was significantly lower in the CMA group (P < .001). Early
regular continuous CM formula was very rare in the CMA group
compared with the Control and EA groups (3.9% in the CMA
group, 51.0% in the Control group, and 34.4% in the EA
group) (P < .001).
We further analyzed factors associated with CMA using
multivariable logistic regression analysis (Table III). Controlling
for allergic symptoms, parental age at delivery and family history
of allergic diseases, the adjusted odds ratio (OR) of delayed or no
regular CM formula was 23.74 (95% confidence interval [CI],
5.39-104.52), and that of no early regular continuous CM for-
mula was 92.76 (95% CI, 9.05-951.04) in a comparison of the
CMA group and the Control group. The results were similar if
ISAAC or ATS-DLD was used to define AD and BA (data not
shown). In a comparison of the CMA group and the EA group,
the adjusted OR of delayed or no regular CM formula was 10.16
(95% CI, 2.48-41.64), and that of no early regular continuous
CM formula was 21.58 (95% CI, 3.33-139.95).
CM formula use at each month of age in each group is shown in
Figure E2 (available in this article’s Online Repository at www.jaci-
J ALLERGY CLIN IMMUNOL PRACT ONIZAWA ET AL 485
VOLUME 4, NUMBER 3

TABLE III. Multivariable logistic regression analysis for IgE-CMA according to feeding patterns adjusted by variables of allergic
symptoms, parental age at delivery, and family history of allergy
CMA vs Control
cOR 95% CI P value aOR 95% CI P value

Delayed* or no regular CM formula 10.71 4.19-27.39 <.001 23.74 5.39-104.52 <.001

No early regular continuous† CM formula 25.48 5.88-110.40 <.001 92.76 9.05-951.04 <.001
CMA vs EA
cOR 95% CI P value aOR 95% CI P value

Delayed* or no regular CM formula 5.83 1.94-17.55 .001 10.16 2.48-41.64 .001

No early regular continuous† CM formula 12.83 2.61-63.00 <.001 21.58 3.33-139.95 .001
aOR, Adjusted odds ratio; CI, confidence interval; CM, cow’s milk; IgE-CMA, IgE-mediated cow’s milk allergy; cOR, crude odds ratio; EA, egg allergy.
Each feeding pattern was adjusted for parental age at delivery, asthma, atopic dermatitis, food allergy, allergic rhinitis, prior history of physician-diagnosed asthma, and
physician-diagnosed atopic dermatitis (aOR).
*Defined as “Did not start regular CM formula within the first month of life.”
†Defined as “Started regular CM formula within the first month of life and continued until 6 mo of age or until the onset of CMA.”

n.s.
100
CMA
Number of patients (%)

90
80 n.s.
70 Control
60
50 EA
40
30 n.s.
20 n.s.
10
0
No need to use Better for To prevent Other
CM formula child care allergic disease

FIGURE 2. The reasons for choosing “exclusive breastfeeding” or “almost exclusive breastfeeding” in the first month of life. Two mothers
in the CMA group and 1 mother in the EA group chose “other.” All mothers in the CMA group answered that their baby drank neither CM
formula nor breast milk from a bottle. One mother in the EA group answered that her baby drank more by direct breastfeeding. Multiple
answers were allowed. CM, Cow’s milk; CMA, cow’s milk allergy; EA, egg allergy; n.s., not significant.

inpractice.org). Thirty-five (68.6%) patients in the CMA group,
90 (88.2%) subjects in the Control group, and 24 (75.0%) pa-
tients in the EA group responded that they had used CM formula
within the first month of life, including the maternity hospital stay
(P ¼ .007 between the CMA group and Control group, and
P ¼ .623 between the CMA group and EA group). Fewer infants
received CM formula in the CMA group at 2 months of age than in
the other groups (P  .001 compared with the Control group and
P ¼ .011 compared with the EA group).
The age of onset of CMA symptoms is shown in Figure E3
(available in this article’s Online Repository at www.jaci-
inpractice.org). All except 1 patient experienced symptoms of
CMA after the first month of life. Of 51 patients, 16 (31.4%) in
the CMA group responded that they had never used CM for-
mula until the onset of IgE-CMA, including the maternity
hospital stay.
Reasons for choosing a feeding pattern
We next analyzed the reason for choosing the feeding pattern in
each group (Figure 2). The majority of each group chose “no need
to use CM formula” or “better for child care” as a reason for
choosing exclusive or almost exclusive breastfeeding in the first
month of life. Only 3 (6.7%), 2 (4.8%), and 3 (16.7%) in the
CMA group, the Control group, and the EA group chose “to
prevent allergic diseases,” respectively (P ¼ 1.000 in the CMA
group vs the Control group, P ¼ .341 in the CMA group vs the EA
group).
We also analyzed the reasons for discontinuing regular CM
formula by 6 months of age, even though they started regular
CM formula within the first month of life. The frequency of
discontinuation of CM formula by 6 months of age was 6/6
(100%), 9/60 (15.0%), and 2/14 (14.3%) in the CMA group,
Control group, and EA group, respectively. All subjects other
than 2 patients in the CMA group and 1 patient in the Control
group answered “no need to use CM formula.” Two patients in
the CMA group answered “suspected CMA symptoms” as a
reason, of which 1 patient developed an immediate-type allergic
reaction to CM and the other was diagnosed with infantile AD
associated with IgE-CMA and shortly afterward developed an
immediate-type allergic reaction. One patient in the Control
group answered “started solid foods” as a reason for dis-
continuing CM.
Worst-case analysis
Four patients in the CMA group reported that they experi-
enced symptoms suspected to be CMA in the maternity hospital
486 ONIZAWA ET AL
(eg, vomiting and/or diarrhea). All of them breastfed because
they felt “no need to use CM formula” (not because of
“suspected CMA symptoms”). We conducted a worst-case
analysis in which all 4 patients were considered to have under-
gone early regular continuous CM formula, although it was
uncertain whether CMA caused their symptoms because the
appropriate diagnoses were not made by physicians. In this case,
no early regular continuous CM formula was also an indepen-
dent risk factor for IgE-CMA (adjusted OR 15.88 [95% CI,
3.84-65.74] in the CMA group vs the Control group, 9.32
[95% CI, 1.96-44.35] in the CMA group vs the EA group).
DISCUSSION
Our results support the hypothesis that early, regular, and
continuous consumption of CM formula within the first month
of life prevents IgE-CMA, which is consistent with other studies
regarding the prevention of CM,9,22 peanut,17,18 egg,19 cereal
grain,20 and fish allergies.21
Ryugasaki city is a commuter town in the Ibaraki prefecture of
Japan and has a population of 79,280. Ryugasaki Saiseikai
Hospital is a core hospital in this region. Although our study was
not a population-based study, the majority of patients with food
allergies in this region visit the hospital. We invited all patients
with CMA and those with EA who had an allergy appointment
or who visited the hospital for a common cold or routine
immunization during the study period to participate.
In this study, the diagnosis of IgE-CMA or IgE-EA was
confirmed by a board certified allergy specialist (D.H.) based on
positive allergen-specific IgE and immediate-type allergic
reactions after the ingestion of CM or eggs, and some of them
underwent oral challenge tests. In the Control group, past his-
tories and medical records were carefully examined to exclude the
possibility of food allergy by the pediatrician (Y.O.). Although
food challenges were not performed in all of patients in the
study, most patients experienced symptoms more than once, and
oral food challenge was performed to confirm the diagnosis in
inconclusive cases. When we excluded the patients who experi-
enced allergic reactions once (5 in the CMA group and 4 in the
EA group) from the analysis, no early regular CM formula
remained an independent risk factor for CMA (Table EI available
in this article’s Online Repository at www.jaci-inpractice.org).
In this study, we enrolled the EA group in addition to the
Control group, which consisted of individuals without a history
of food allergies to minimize the influence of confounders such as
AD, BA, and a family history of allergic diseases. Epicutaneous
sensitization has been reported to have an important role in the
development of food allergies, and the presence of AD is the
main skin-related risk factor for food sensitization in young
infants.26,27 The prevalence of AD was not significantly different
between the CMA group and the EA group in our study.
We examined not only feeding patterns but also the reason for
choosing feeding patterns because parents with a history of
allergic diseases might be more likely to choose breastfeeding for
their babies than those without allergic diseases. Consistent with
previous studies,28,29 the parents of patients with IgE-CMA had
significantly higher rates of allergic diseases than the parents of
the controls (Table I). However, most parents of patients with
IgE-CMA did not feed CM to their children because they felt no
need to use CM formula, not for the future prevention of allergic
diseases (Figure 2). We examined the age of onset of CMA
J ALLERGY CLIN IMMUNOL PRACT
MAY/JUNE 2016
symptoms (Figure E3, available in this article’s Online
Repository at www.jaci-inpractice.org), and most cases experi-
enced allergic reactions after the first month of life. All 4 cases in
the worst-case analysis discontinued CM because they felt no
need to use CM formula, not for the prevention of CMA. To our
knowledge, this is the first report to investigate the reason for
choosing a feeding pattern in terms of CMA prevention.
Exclusive breastfeeding has been advocated to prevent IgE-
CMA.15,16,30 This discrepancy may be due to the amount and
frequency of food protein allergen intake. In antigen sensitization
and oral tolerance, it has been shown that the amount and the
frequency of allergen intake were associated with sensitization
and immune tolerance. In animal studies, the oral intake of small
amounts of allergen stimulated IgE production, whereas larger
allergen doses suppressed specific IgE production.31,32 In a
prospective observational study, breastfeeding exclusively or
combined with infrequent exposure to small amounts of CM
during the first 2 months of life was associated with an increased
risk of developing IgE-CMA when CM was used in a maternity
hospital.6 It has also been reported that the long-term elimina-
tion of CM induced IgE-CMA without previous problems after
CM intake.33,34 These results suggest that sufficient amounts
and the continuous intake of CM are important for the pre-
vention of IgE-CMA. In this study, the percentage of infants
who tried CM formula within the first month of life was not
significantly different between the CMA group and the EA group
(P ¼ .623), whereas the number of children with regular con-
sumption of CM formula was significantly less in the CMA
group than in the EA group. This indicates the possibility that
small amounts and infrequent exposure to CM might lead to
allergic reactions.
To determine how long we should continue regular CM, we
defined “early regular continuous CM formula” as the regular
consumption of CM formula started within the first month of
life and continued until 6 months or until the onset of IgE-
CMA, and early regular continuous CM formula was associ-
ated with IgE-CMA, with an OR of 21.58 to 92.76. Our present
results, combined with previous studies that suggest the risk of
long-term discontinuation of CM, demonstrate that the regular
consumption of CM should be continued until 6 months, at
which time the initiation of solid foods is recommended in
Japan.35
The early introduction of other solid foods has been reported
to be beneficial to prevent food allergies, such as those to pea-
nuts,17,18 eggs,19 cereal grains,20 and fish,21 whereas others have
reported that the introduction of solid foods earlier than
4 months of age increases the prevalence of food allergies.36 CM
is usually introduced at an earlier age than solid foods, and
sensitization to CM may be induced earlier than sensitization to
solid foods. Therefore, immune tolerance can be promoted
before the onset of CMA, which has been reported at an average
age of 2.8 to 3.5 months.6,37
There are a few limitations in our study. First, the possibility
of recall bias cannot be excluded because of the retrospective
study design, that is, a case-control study using a questionnaire.
In this study, all data were collected by the same procedure and
the same questionnaire regardless of the study group, except for 7
patients in the IgE-CMA group whose information was collected
by telephone interview. However, even though the information
of these 7 patients was excluded, the association remains statis-
tically significant (data not shown). The exclusive breastfeeding
J ALLERGY CLIN IMMUNOL PRACT
VOLUME 4, NUMBER 3
rate at 4 weeks postpartum has been reported to be 28.9% to
42.4% in Japan.38-40 Parents of the CMA group in this study
chose exclusive breastfeeding at a significantly higher rate in
comparison with these data. Moreover, no parents in the CMA
group chose a formula-dominated feeding pattern, in contrast to
other groups (Figure E1, available in this article’s Online Re-
pository at www.jaci-inpractice.org).
Second, we did not investigate the amount of CM formula
consumed in each group because of the retrospective study
design. Although our data suggest that the regular consumption
of CM formula prevents IgE-CMA, further prospective studies
are needed to clarify the minimum daily amount of CM formula
required to prevent IgE-CMA.
In 2012, the AAP reaffirmed its recommendation of exclusive
breastfeeding for approximately 6 months and the continuation
of breastfeeding for 1 year or longer.41 The Japanese Ministry of
Health, Labor and Welfare35 and the Japan Pediatric Society42
recommend breastfeeding, although a concrete determination
of the period or method (exclusive or partial) has yet to be made.
Thus, exclusive formula feeding has decreased every year in
Japan.35 Our results should not be considered to contradict the
efficacy of breastfeeding. Many benefits of breastfeeding have
been well documented by numerous studies, for example, for the
prevention of respiratory infection43 and sudden infant death
syndrome.44 On the other hand, various mixed feeding methods
have not been assessed in these studies. Although the regular
consumption of CM formula is preferred to lower the risk of IgE-
CMA, as shown in Figure E1 (available in this article’s Online
Repository at www.jaci-inpractice.org), mixed feeding with a
predominance of breast milk was also associated with a lower
incidence of IgE-CMA.
CONCLUSIONS
In conclusion, our study supports the hypothesis that regular
exposure to CM (at least once daily) started within the first
month of life prevents IgE-CMA. A large prospective study will
be needed to review the benefits and adverse effects of the regular
consumption of small amounts of CM.
REFERENCES
1. Gupta RS, Springston EE, Warrier MR, Smith B, Kumar R, Pongracic J, et al.
The prevalence, severity, and distribution of childhood food allergy in the
United States. Pediatrics 2011;128:e9-17.
2. Akiyama H, Imai T, Ebisawa M. Japan food allergen labeling regulation—
history and evaluation. Adv Food Nutr Res 2011;62:139-71.
3. Pumphrey RS, Gowland MH. Further fatal allergic reactions to food in the
United Kingdom, 1999-2006. J Allergy Clin Immunol 2007;119:1018-9.
4. Host A, Halken S. A prospective study of cow milk allergy in Danish infants
during the first 3 years of life. Clinical course in relation to clinical and
immunological type of hypersensitivity reaction. Allergy 1990;45:587-96.
5. Schrander JJ, van den Bogart JP, Forget PP, Schrander-Stumpel CT,
Kuijten RH, Kester AD. Cow’s milk protein intolerance in infants under 1 year
of age: a prospective epidemiological study. Eur J Pediatr 1993;152:640-4.
6. Saarinen KM, Juntunen-Backman K, Järvenpää AL, Kuitunen P, Lope L,
Renlund M, et al. Supplementary feeding in maternity hospitals and the risk of
cow’s milk allergy: a prospective study of 6209 infants. J Allergy Clin Immunol
1999;104:457-61.
7. Venter C, Pereira B, Grundy J, Clayton CB, Roberts G, Higgins B, et al.
Incidence of parentally reported and clinically diagnosed food hypersensitivity
in the first year of life. J Allergy Clin Immunol 2006;117:1118-24.
8. Kvenshagen B, Halvorsen R, Jacobsen M. Adverse reactions to milk in infants.
Acta Paediatr 2008;97:196-200.
9. Katz Y, Rajuan N, Goldberg MR, Eisenberg E, Heyman E, Cohen A, et al. Early
exposure to cow’s milk protein is protective against IgE-mediated cow’s milk
protein allergy. J Allergy Clin Immunol 2010;126:77-82.e1.
ONIZAWA ET AL 487
10. Saarinen KM, Pelkonen AS, Makela MJ, Savilahti E. Clinical course and
prognosis of cow’s milk allergy are dependent on milk-specific IgE status.
J Allergy Clin Immunol 2005;116:869-75.
11. Bishop JM, Hill DJ, Hosking CS. Natural history of cow milk allergy: clinical
outcome. J Pediatr 1990;116:862-7.
12. Skripak JM, Matsui EC, Mudd K, Wood RA. The natural history of IgE-
mediated cow’s milk allergy. J Allergy Clin Immunol 2007;120:1172-7.
13. Wood RA, Sicherer SH, Vickery BP, Jones SM, Liu AH, Fleischer DM, et al.
The natural history of milk allergy in an observational cohort. J Allergy Clin
Immunol 2013;131:805-12.
14. Elizur A, Rajuan N, Goldberg MR, Leshno M, Cohen A, Katz Y. Natural course
and risk factors for persistence of IgE-mediated cow’s milk allergy. J Pediatr
2012;161:482-487.e1.
15. Greer FR, Sicherer SH, Burks AW, American Academy of Pediatrics Com-
mittee on Nutrition; American Academy of Pediatrics Section on Allergy and
Immunology. Effects of early nutritional interventions on the development of
atopic disease in infants and children: the role of maternal dietary restriction,
breastfeeding, timing of introduction of complementary foods, and hydrolyzed
formulas. Pediatrics 2008;121:183-91.
16. Muraro A, Dreborg S, Halken S, Høst A, Niggemann B, Aalberse R, et al.
Dietary prevention of allergic diseases in infants and small children: Part III.
Critical review of published peer-reviewed observational and interventional
studies and final recommendations. Pediatr Allergy Immunol 2004;15:291-307.
17. Du Toit G, Katz Y, Sasieni P, Mesher D, Maleki SJ, Fisher HR, et al. Early
consumption of peanuts in infancy is associated with a low prevalence of peanut
allergy. J Allergy Clin Immunol 2008;122:984-91.
18. Du Toit G, Roberts G, Sayre PH, Bahnson HT, Radulovic S, Santos AF, et al.
Randomized trial of peanut consumption in infants at risk for peanut allergy.
N Engl J Med 2015;372:803-13.
19. Koplin JJ, Osborne NJ, Wake M, Martin PE, Gurrin LC, Robinson MN, et al.
Can early introduction of egg prevent egg allergy in infants? A population-based
study. J Allergy Clin Immunol 2010;126:807-13.
20. Poole JA, Barriga K, Leung DY, Hoffman M, Eisenbarth GS, Rewers M, et al.
Timing of initial exposure to cereal grains and the risk of wheat allergy. Pedi-
atrics 2006;117:2175-82.
21. Kull I, Bergstrom A, Lilja G, Pershagen G, Wickman M. Fish consumption
during the first year of life and development of allergic diseases during child-
hood. Allergy 2006;61:1009-15.
22. Saarinen KM, Savilahti E. Infant feeding patterns affect the subsequent
immunological features in cow’s milk allergy. Clin Exp Allergy 2000;30:400-6.
23. Williams H, Robertson C, Stewart A, Aït-Khaled N, Anabwani G, Anderson R,
et al. Worldwide variations in the prevalence of symptoms of atopic eczema in
the International Study of Asthma and Allergies in Childhood. J Allergy Clin
Immunol 1999;103:125-38.
24. Katayama I, Kohno Y, Akiyama K, Aihara M, Kondo N, Saeki H, et al.
Japanese Guideline for Atopic Dermatitis 2014. Allergol Int 2014;63:377-98.
25. Nishima S, Chisaka H, Fujiwara T, Furusho K, Hayashi S, Hiraba K, et al.
Surveys on the prevalence of pediatric bronchial asthma in Japan: a comparison
between the 1982, 1992, and 2002 surveys conducted in the same region using
the same methodology. Allergol Int 2009;58:37-53.
26. Flohr C, Perkin M, Logan K, Marrs T, Radulovic S, Campbell LE, et al. Atopic
dermatitis and disease severity are the main risk factors for food sensitization in
exclusively breastfed infants. J Invest Dermatol 2014;134:345-50.
27. Dharmage SC, Lowe AJ, Matheson MC, Burgess JA, Allen KJ, Abramson MJ.
Atopic dermatitis and the atopic march revisited. Allergy 2014;69:17-27.
28. Tsai HJ, Kumar R, Pongracic J, Liu X, Story R, Yu Y, et al. Familial aggre-
gation of food allergy and sensitization to food allergens: a family-based study.
Clin Exp Allergy 2009;39:101-9.
29. Pyrhonen K, Hiltunen L, Kaila M, Nayha S, Laara E. Heredity of food allergies
in an unselected child population: an epidemiological survey from Finland.
Pediatr Allergy Immunol 2011;22:e124-32.
30. Lucas A, Brooke OG, Morley R, Cole TJ, Bamford MF. Early diet of preterm
infants and development of allergic or atopic disease: randomised prospective
study. BMJ 1990;300:837-40.
31. Jarrett EE. Perinatal influences on IgE responses. Lancet 1984;2:797-9.
32. Fritsche R, Pahud JJ, Pecquet S, Pfeifer A. Induction of systemic immunologic
tolerance to beta-lactoglobulin by oral administration of a whey protein hy-
drolysate. J Allergy Clin Immunol 1997;100:266-73.
33. Flinterman AE, Knulst AC, Meijer Y, Bruijnzeel-Koomen CA, Pasmans SG.
Acute allergic reactions in children with AEDS after prolonged cow’s milk
elimination diets. Allergy 2006;61:370-4.
34. Al Dhaheri W, Diksic D, Ben-Shoshan M. IgE-mediated cow milk allergy and
infantile colic: diagnostic and management challenges. BMJ Case Rep 2013.
http://dx.doi.org/10.1136/bcr-2012-007182.
488 ONIZAWA ET AL
35. Ministry of Health, Labour and Welfare (in Japanese). Available from: http://www.
mhlw.go.jp/shingi/2007/03/dl/s0314-17c.pdf; 2007. Accessed June 21, 2015.
36. Grimshaw KE, Maskell J, Oliver EM, Morris RC, Foote KD, Mills EN, et al.
Introduction of complementary foods and the relationship to food allergy. Pe-
diatrics 2013;132:e1529-38.
37. Santos A, Dias A, Pinheiro JA. Predictive factors for the persistence of cow’s
milk allergy. Pediatr Allergy Immunol 2010;21:1127-34.
38. Otsuka K, Dennis CL, Tatsuoka H, Jimba M. The relationship between
breastfeeding self-efficacy and perceived insufficient milk among Japanese
mothers. J Obstet Gynecol Neonatal Nurs 2008;37:546-55.
39. Ebina S, Kashiwakura I. Relationship between feeding modes and infant weight
gain in the first month of life. Exp Ther Med 2013;5:28-32.
J ALLERGY CLIN IMMUNOL PRACT
MAY/JUNE 2016
40. Ministry of Health, Labour and Welfare (in Japanese). Available from: http://
www.mhlw.go.jp/shingi/2007/03/dl/s0314-17b-1.pdf; 2007. Accessed June 21,
2015.
41. Breastfeeding and the use of human milk. Pediatrics 2012;129:e827-41.
42. Pediatricians and promotion of breast-feeding in infancy (in Japanese). J Jpn
Pediatr Soc 2011;115:1363-89.
43. Chantry CJ, Howard CR, Auinger P. Full breastfeeding duration and associated
decrease in respiratory tract infection in US children. Pediatrics 2006;117:
425-32.
44. Hauck FR, Thompson JM, Tanabe KO, Moon RY, Vennemann MM. Breast-
feeding and reduced risk of sudden infant death syndrome: a meta-analysis.
Pediatrics 2011;128:103-10.
J ALLERGY CLIN IMMUNOL PRACT ONIZAWA ET AL 488.e1
VOLUME 4, NUMBER 3

70 40

CMA symptoms (n)

Number of patients (%)

60 CMA

Age of onset of
50
30
Control
40 20
EA
30
20 10
10
0
0
< 1 month 1-6 months > 6 months
Exclusive BF Almost Mixed 1 Mixed 2 Exclusive CM
exclusive BF
FIGURE E3. Age of onset of CMA symptoms. CMA, Cow’s milk
FIGURE E1. Feeding patterns in the first month of life (after allergy.
discharge from maternity hospital). BF, Breastfeeding; CM, cow’s
milk; CMA, cow’s milk allergy; EA, egg allergy.

100
CM formula use (%)

80
CMA
60 Control
40 EA
20
0
0 1 2 3 4 5 months

FIGURE E2. Using rate of CM formula at each month of age. CM,

Cow’s milk; CMA, cow’s milk allergy; EA, egg allergy.
488.e2 ONIZAWA ET AL J ALLERGY CLIN IMMUNOL PRACT
MAY/JUNE 2016

TABLE E1. Multivariable logistic regression analysis for IgE-CMA according to feeding patterns adjusted by variables of allergic
symptoms, parental age at delivery and family history of allergy (when confined to patients who had experienced allergic symptoms
more than 2 times or undergone oral food challenge)
CMA vs Control
cOR 95% CI P value aOR 95% CI P value

Delayed* or no regular CM formula 9.52 3.70-24.49 <.001 20.28 4.72-87.05 <.001

No early regular continuous† CM formula 22.88 5.27-99.44 <.001 70.17 7.41-664.75 <.001
CMA vs EA
cOR 95% CI P value aOR 95% CI P value

Delayed* or no regular CM formula 5.12 1.64-15.99 .005 9.11 2.11-39.45 .003

No early regular continuous† CM formula 10.66 2.10-54.03 .002 18.37 2.65-127.23 .003
aOR, Adjusted odds ratio; CI, confidence interval; IgE-CMA, IgE-mediated cow’s milk allergy; CM, cow’s milk; cOR, crude odds ratio; EA, egg allergy.
Each feeding pattern was adjusted for parental age at delivery, asthma, atopic dermatitis, food allergy, allergic rhinitis, prior history of physician-diagnosed asthma, and
physician-diagnosed atopic dermatitis (aOR).
*Defined as “Did not start regular CM formula within the first month of life.”
†Defined as “Started regular CM formula within the first month of life and continued until 6 mo of age or until the onset of CMA.”