Cow's milk protein allergy in children: a practical guide
on behalf of EWGPAG
Dipartimento dell'Eta Evolutiva, Clinica Pediatrica Universita di Parma
A joint study group on cow's milk allergy was convened by the Emilia-Romagna Working Group for Paediatric Allergy and by the Emilia-Romagna Working Group for Paediatric Gastroenterology to focus best practice for diagnosis, management and follow-up of cow's milk allergy in children and to offer a common approach for allergologists, gastroenterologists, general paediatricians and primary care physicians. The report prepared by the study group was discussed by members of Working Groups who met three times in Italy. This guide is the result of a consensus reached in the following areas. Cow's milk allergy should be suspected in children who have immediate symptoms such as acute urticaria/angioedema, wheezing, rhinitis, dry cough, vomiting, laryngeal edema, acute asthma with severe respiratory distress, anaphylaxis. Late reactions due to cow's milk allergy are atopic dermatitis, chronic diarrhoea, blood in the stools, iron deficiency anaemia, gastroesophageal reflux disease, constipation, chronic vomiting, colic, poor growth (food refusal), enterocolitis syndrome, protein-losing enteropathy with hypoalbuminemia, eosinophilic oesophagogastroenteropathy. An overview of acceptable means for diagnosis is included. According to symptoms and infant diet, three different algorithms for diagnosis and follow-up have been suggested.
Cows milk protein allergy (CMPA) affects from 2 to 6%
of children, with the highest prevalence during the first
year of age . About 50% of children have been shown
to resolve CMPA within the first year of age, 80-90%
within their fifth year [2,3]. The rate of parent-reported
CMPA is about 4 times higher than the real one in
children . So, many children are referred for suspected
CMPA based on parent perception, symptoms such as
cutaneous eruption, insomnia, persistent nasal
obstruction, sebhorreic dermatitis or positive results to
unorthodox investigations. Moreover, parents often put
their children on unnecessary diet without an adequate
medical and dietary supervision. These inappropriate
dietary restrictions may provoke nutritional unbalances,
especially in the first year of age. Therefore, an accurate
diagnosis of CMPA is important in order to avoid not
only the risk of rickets, decreased bone mineralization
, anaemia, poor growth and hypoalbuminemia, but
also that of immediate clinical reactions or severe
chronic gastroenteropathy leading to malabsorption.
Recently, three guidelines [6-8] reporting different
approaches to the infant with CMPA have been
In view of these considerations, a study group with
expert representatives of Emilia-Romagna Working
Group for Paediatric Allergy and of that for Paediatric
Gastroenterology (EWGPGA), was constituted. As
mmembers of the expert panel, the authors were
assigned to review practice with regard to diagnosis,
management and follow-up of CMPA for both
community and hospital paediatrician in order to share the
same approach towards the child. The document
prepared by the study group was based on existing
recommendations, clinical experience and evidence from the
literature. The report was discussed and received input
by the members (see participant list in
acknowledgments) of EWGPGA which included clinicians
experienced in paediatric allergy, paediatric gastroenterology
and general paediatricians, in three meetings held in
November 2008, February 2009 and March 2009 and a
consensus was reached. According to the symptoms and
the type of infant diet, three different algorithms for
diagnosis and follow-up have been suggested. These
approaches refer to the child in the first year of age.
Recommendations for older children have been briefly
Cows milk protein allergy: when should we
A positive atopic familiar history is common in children
with suspected CMPA . The diagnosis of CMPA is
based on a detailed history of symptoms (Fig. 1), skin
prick test and serum specific IgE to cows milk protein,
elimination diet and oral food challenge. Clinical
manifestations due to CMPA [6-14] can be divided into
IgEmediated immediate clinical reactions (onset of the
symptoms within the 30 minutes after the ingestion of
cows milk) and non IgE-mediated delayed reactions
(hours-days after food ingestion), most affecting the skin
and the gastrointestinal system. However, immediate
and delayed reactions can be associated in atopic
eczema and in eosinophilic oesophageal gastroenteritis
The negative predictive value of skin prick test/specific
IgE for immediate reaction is excellent (>95%) ,
however a small number of these patients can have
clinical reaction. Therefore, despite negative IgE tests if
there is a strong suspicion of CMPA, an oral food
challenge is necessary to confirm the absence of clinical
allergy. On the other hand, about 60% of children with
positive IgE tests have CMPA [15,16]. Prick by prick
test with cows milk substitutes may be considered.
Oral food challenge, open or blind, remains the gold
standard to definitely ascertain children with food
allergy when the diagnosis is unclear . OFC should
be performed under medical supervision in a setting
with emergency facilities, especially in case of positive
skin prick test or serum specific IgE to cows milk and
in infants at risk of an immediate reaction.
Cows milk substitutes
About 10% of children with CMPA react to extensively
hydrolyzed formula (eHF) . In comparison with eHF,
soy formula (SF) provokes more frequently reactions in
children with CMPA aged less than 6 months  but
not in older children. SF mainly induces gastrointestinal
Amino acid formula (AAF) is non allergenic . Its
use is limited by the high cost and bad taste.
Rice is allergenic and is often involved in the onset of
enterocolitis syndrome in Australian infants .
Contrasting data have been reported on the effect of protein
content on growth . In Italian children, rice formula
has been shown to be tolerated by children with CMPA
. Larger long-term studies are warranted to clarify
Figure 1 Immediate and late onset reactions in children with cows milk protein allergy.
the use of rice formula in infants with CMPA. Rice
formula may be a choice in selected cases taking into
consideration the taste and the cost.
Home-made meals may be a dietary option after 4
months of age.
Mammalian milks are not nutritionally adequate.
Goats milk commonly provokes clinical reactions in
more than 90% of children with CMPA , donkeys
milk in about 15% [24,25] and has a high cost.
A child fed with cows milk formula with
mildmoderate symptoms (Fig. 2)
In infants with immediate symptoms (vomiting, acute
hives, angioedema, wheezing, rhinitis, dry cough) or late
symptoms (moderate/severe atopic dermatitis, diarrhoea,
blood in the stools, iron deficiency anaemia,
gastroesophageal reflux disease (GORD), constipation) a CMPA
can be suspected [6-8,10-14]. Other causes are to be
considered for patients unresponsive to treatment.
Infant colic (more than 3 hours of crying a day, 3 days
for more than 3 weeks) is not unanimously considered
as a consequence of CMPA. The paediatrician has to
consider the opportunity of a cows milk free diet in the
most troublesome cases [26,27]. Mild immediate
reactions may be of difficult interpretation because they
can be the result of causes different from CMPA.
However, if these symptoms are strongly related to cows
milk ingestion, we recommend to eliminate cows milk
and follow the algorithm for severe reactions (Fig. 2).
Regarding delayed onset gastrointestinal symptoms,
other pathologies (i.e. infections) should be excluded
before investigating allergic sensitization.
In mild atopic dermatitis, investigations for CMPA are
not necessary in the absence of a clear relation between
cows milk intake and onset of symptoms.
When a CMPA is suspected, infants should go on a
24 week diet without cows milk protein. Four weeks
should be considered for chronic gastrointestinal
symptoms. Infants should be fed with eHF or SF in children
aged more than 6 months and without gastrointestinal
If the symptoms improve on a restrict diet, an OFC to
cows milk is necessary to definitely ascertain the
diagnosis. If the oral food challenge is positive, the child
must follow the elimination diet and can be
re-challenged after 6 months (a shorter period for GORD) and
in any case, after 9-12 months of age. If the oral food
challenge is negative, a free-diet can be followed.
Children less than 1 year fed with regular cows milk formula with suspected mild-moderate CMPA
Elimination diet for 2-4 week (4 weeks for gastrointestinal symptoms):
Extensively hydrolyzed formula
Soy formula if >6 months of age (without gastrointestinal symptoms)
Oral food challenge (consider to perform
challenge test in a clinical setting in case
of positive specific IgE and/or SPT)
Avoid cows milk for at least 6 months
and until 9-12 months of age (assess a
shorter period in GORD).
Figure 2 Algorithm for children < 1 year fed with cows milk formula and mild-moderate symptoms.
When there is strong suspicion of IgE-mediated
reactions, in infants who do not respond to a diet with eHF or
SF an attempt may be made with a 14-days diet with AAF.
Cows milk substitutes are used in children aged less
than 12 months. In older children with CMPA, eHF or
AAF are not usually necessary because an adequate diet
is easily accessible.
A child fed with cows milk formula with severe
symptoms (Fig. 3)
Immediate severe symptoms are considered laryngeal
edema, acute asthma with severe respiratory difficulty,
anaphylaxis. The following are delayed onset severe
symptoms: chronic diarrhoea or chronic vomiting with
poor growth, intestinal bleeding with iron deficiency
anaemia, protein losing enteropathy with
hypoalbuminemia, eosinophilic gastroenteropathy confirmed by biopsy
If any of these immediate symptoms are observed as a
consequence of suspected CMPA, infants should follow
a cows milk free diet. As substitutes, SF (if older than 6
months of age) or eHF or AAF can be used. eHF and
SF should be started under medical supervision because
of possible clinical reactions. If an AAF is adopted, it
may be administered for 2 weeks and then the infant
may be switched to SF or eHF.
In children with late severe gastrointestinal symptoms
with poor growth, anaemia or hypoalbuminemia or
eosinophilic oesophagogastroenteropathy, it is
recommended to start an elimination diet with AAF and then
switched with eHF. The effect of the diet should check
out within 10 days for enterocolitis syndrome, 1-3
weeks for enteropathy and 6 weeks for eosinophilic
In children with anaphylaxis and concordant positive
IgE tests or severe gastrointestinal reactions, oral food
challenge is not necessary for diagnosis. The oral food
challenge for tolerance acquisition should be performed
not before 6-12 months after the last reaction. Children
have to eliminate cows milk until 12 months of age, but
in those with enterocolitis syndrome until 2-3 years of
Children with any severe symptoms should be referred
to a specialized centre.
eHF or AAF is used in children aged less than 12
months and in older children with severe gastrointestinal
Figure 3 Algorithm for children <1 year fed with cows milk formula and severe symptoms. In child less than1 year of age, infant formula
is not compulsory.
symptoms. In children > 12 months with anaphylaxis,
cows milk substitutes are not always nutritionally
A breast-fed infant with a suspected CMPA (Fig. 4)
In exclusively breast-fed infants, suspected symptoms
to the cows milk proteins are almost always non
IgEmediated as atopic dermatitis, vomiting, diarrhoea,
blood in the stools, GORD, colic .
A maternal diet without cows milk is not
recommended for mild symptoms.
There is no evidence that a maternal diet without egg
and cows milk in infants with bloody stools
(proctocolitis) is of value [30,31].
In infants with moderate-severe symptoms, cows milk
protein, eggs and other foods should be eliminated by
the mothers diet only if history suggests an unequivocal
reaction. Moreover, the infant should be referred to a
specialized centre. The maternal elimination diet has to
be followed for 4 weeks. If there is no improvement the
diet should be stopped. If symptoms improved, its
recommended that the mother ingested large amounts
of cows milk for one week. If symptoms occurred, the
mother will continue the diet with supplemental intake
of calcium. The infant can be weaned as recommended
for healthy children, but cows milk should be avoided
until 9-12 months of age and for at least 6 months from
the beginning of the diet. If the volume of breast milk is
insufficient, eHF or SF formula (if > 6 months) should
If after the reintroduction of cows milk in mothers
diet symptoms do not occur, the excluded foods can be
introduced one by one in the diet.
The diagnosis of CMPA is based on oral food challenge
that follows a 2-4 week elimination diet.
A diagnostic oral food challenge is unnecessary in
immediate reactions or late gastrointestinal reactions
with anaemia, poor growth or hypoalbuminemia if
the causative role of cows milk is clear. Children can
be challenged after 6-12 months from the reaction and
Breast-fed infants with suspected reactions to cows milk: atopic dermatitis, vomiting,
diarrhoea, stool blood, GORD, poor growth, infantile colic.
Clinical evaluation, family history
Maternal diet without
cows milk for 2-4 weeks.
Figure 4 Algorithm for breast-fed infants with suspected non-IgE mediated reactions to cows milk protein.
not before 12-24 months of age according to the
Diets must be nutritionally balanced. In children with
CMPA, a supplementation with calcium must be
Diet is not requested in children with mild atopic
dermatitis and negative history for cows milk reactions.
SF should not be used in infants < 6 months of age
with allergic symptoms and in those with late
Children with gastrointestinal reactions and anaemia,
poor growth or hypoalbuminemia should be given AAF
and then switched to eHF.
eHF or AAF is used in children aged less than 12
months and in older children with severe
gastrointestinal symptoms. In children > 12 months with
anaphylaxis, cows milk substitutes are not always nutritionally
List of abbreviations
CMPA: cows milk protein allergy; EWGPAG:
EmiliaRomagna Working Group for Paediatric Allergy and of
that for Paediatric Gastroenterology; eHF: extensively
hydrolyzed formula; SF: soy; AAF: amino acid formula;
GORD: gastroesophageal reflux disease.
Physicians and specialists of Emilia-Romagna Emilia-Romagna Working Group
for Paediatric Allergy and for Paediatric Gastroenterology (EWGPGA) who
contributed to this guide.
Paediatric gastroenterology: Patrizia Alvisi (U.O. di Pediatria Ospedale
Maggiore, Bologna, Italy), Sergio Amarri (U.O. di Pediatria, Reggio Emilia,
Italy), Paolo Baldassarri (U.O. di Pediatria, Forl,), Sandra Brusa (U.O. di
Pediatria, Imola, Italy), Marisa Calacoci (Pediatra Libero Professionista, Ferrara,
Italy), Iliana Cecchini (U.O. di Pediatria, Cesena, Italy), Sara Denti (U.O. di
Pediatria, Carpi, Italy), Annarita Di Biase (U.O. di Pediatria, Modena, Italy),
Cristina Host (U.O. di Pediatria, Ferrara, Italy), Andrea Lambertini (U.O. di
Pediatria Ospedale Maggiore, Bologna, Italy), Angelo Miano (Pediatra Libero
Professionista, Cesena, Italy), Annamaria Metri (U.O. di Pediatria, Faenza, Italy),
Marco Occari (U.O. di Pediatria, Mantova, Italy), Renzo Pini (U.O. di Pediatria,
Rimini, Italy), Marina Rossidoria (Pediatra Libero Professionista, Bologna, Italy).
Paediatric allergy: Andrea Valenti (U.O. di Pediatria, Lugo, Italy), Monica Vallini
(Pediatra Libera scelta, Bologna, Italy, Italy), Iole Venturi (U.O. di Pediatria,
Ravenna, Italy), Laura Viola (U.O. di Pediatria, Rimini, Italy). Ermanno Baldo (U.
O.di Pediatria, Rovereto, Italy), Mauro Bandini (U.O. di Pediatria, Ravenna,
Italy), Filippo Bernardi (Clinica Pediatrica Ospedale S. Orsola, Bologna, Italy),
Adriana Borghi (U.O. Pediatria, Carpi, Italy), Paolo Bottau (U.O. di Pediatria,
Imola, Italy), Lucetta Capra (U.O. di Pediatria Azienda
OspedalieraUniversitaria S. Anna, Ferrara, Italy), Giovanni Cavagni (Ospedale Pediatrico
Bambin Ges, Roma, Italy), Matteo Corchia (Clinica Pediatrica, Parma, Italy),
Danilo Dalpozzo (U.O.di Pediatria, Imola, Italy), Leonardo Loroni (U.O. di
Pediatria, Ravenna, Italy), Laura Giovannini (Pediatra Libero Professionista,
Lugo, Italy), Massimo Masi (Dipartimento Salute della donna, del bambino e
delladolescente, Policlinico S. Orsola-Malpighi, Bologna, Italy), Giuseppe
Menna (ISS Istituto Sicurezza Sociale, U.O.C.Pediatria-Ospedale della
Repubblica di San Marino, Italy), Patrizia Preti (U.O. di Pediatria Ospedale
Maggiore, Bologna, Italy), Giampaolo Ricci (Dipartimento Salute della donna,
del bambino e delladolescente, Policlinico S. Orsola-Malpighi, Bologna, Italy),
Giuseppe Timoncini (U.O. di Pediatria, Forl, Italy), Loretta Biserna (U.O. di
Pediatria, Ravenna, Italy), Elena Zamuner (U.O. di Pediatria, Ravenna, Italy).
1Dipartimento dellEt Evolutiva, Clinica Pediatrica Universit di Parma, Parma,
Italy. 2UO di Pediatria, AUSL Imola, Imola, Italy. 3Dipartimento Salute della
donna, del bambino e delladolescente Policlinico S Orsola-Malpighi, Clinica
Pediatrica, Bologna, Italy. 4Dipartimento Emergenza ed Accettazione
diagnostica, UO di Pediatria, Fidenza, Italy. 5Pediatria, AUSL di Ravenna, Italy.
6UO Pediatria, AUSL di Cesena, Italy.
CC, FB, BB, LC, MM, PP conceived the design of the study and participated
in its coordination. They prepared the draft of the manuscript and revised it.
All authors read and approved the final manuscript.