Jornal de Pediatria - Vol. 78, Nº3, 2002 197
0021-7557/02/78-03/197
Jornal de Pediatria
Copyright © 2002 by Sociedade Brasileira de Pediatria
ORIGINAL ARTICLE
Contamination of expressed human milk
by mycelial fungi
Franz R. Novak,1 João Aprígio Guerra de Almeida,2 Manoel J.S. Santos,3 Bodo Wanke4
Abstract
Objective: to characterize the genera of mycelial fungi detected in expressed human milk received at
the human milk bank of Instituto Fernandes Figueira after home collection.
Methods: we studied 821 expressed human milk samples randomly obtained from flasks filled by the
donors at home. The possible presence of molds, yeasts and mesophilic microorganisms was investigated.
A total of 48 strains of mycelial fungi were isolated from the human milk bank samples and identified
through standard laboratory techniques.
Results: microbiological analysis revealed the occurrence of molds and yeasts in 43 samples (5.2%),
with counts reaching 103 CFU/ml. The following microorganisms were identified: Aspergillus niger group
(6.3%), Aspergillus sp. (4.2%), Paecilomyces sp. (12.6%), Penicillium sp. (60.4%), Rhizopus sp. (2.0%),
and Syncephalastrum sp. (14.5%). Four samples showed the presence of more than one mycelial fungus
type.
Conclusions: the presence of molds and yeasts in human milk manually expressed at home suggests
that the hygiene conditions of the collection site may contaminate milk. Thus, when hospitalized priature
babies receive the raw product, it is very important to observe the collection, storage and transport
conditions in order to avoid the presence and consequences of contaminants increase.
J Pediatr (Rio J) 2002; 78 (3): 197-201:human milk, human milk bank, mycelial fungi, mycotoxins,
aflatoxins.
Introduction
The human milk bank is a specialized center where the
breastfeeding practice is encouraged and where collection,
processing and quality control of colostrum, transition milk
and mature milk are carried out for later use (according to
the physician’s or nutritionist’s recommendations) by infants
who depend on breastmilk for survival.1
1. Ph.D. Microbiology, Universidade Federal do Rio de Janeiro; Professor,
Graduate Course on Women and Children’s Health, Instituto Fernandes
- IFF / Fundação Oswaldo Cruz; Member, Milk Bank team, IFF.
2. Ph.D., Public Health, Instituto Fernandes Figueira - IFF / Fundação
Oswaldo Cruz; Professor, Graduate Course on Women and Children’s
Health, Instituto Fernandes - IFF / Fundação Oswaldo Cruz; Chief, Milk
Bank team, IFF.
3. Graduate student. Micobiology Laboratory of the Research Center Hospital
Evandro Chagas. Fundação Oswaldo Cruz.
4, PhD. Chief of the Micobiology Laboratory of the Research Center Hospital
Evandro Chagas. Fundação Oswaldo Cruz.
Manuscript received Nov 01 2001. Accepted for publication Apr 10 2002.
197
198 Jornal de Pediatria - Vol. 78, Nº3, 2002
At present, Brazil has the world’s largest network of
human milk banks, totaling 151 units in 22 states. In the year
2000 alone, more than 90,000 liters of human milk were
collected, benefiting more than 80,000 preterm and lowweight infants. At the time, approximately 60,000 donors
were registered at the bank.2
In technological terms, human milk is an unstructured
foodstuff, since it offers no physical protection against
infectious microorganisms. Although human milk is the
ideal food for infants, with all necessary nutrients (with
appropriate quality and quantity), it can be an excellent
culture medium for different kinds of microorganisms when
its protection factors are no longer present.3 Mother’s milk
obtained from healthy donors, submitted to a strict hygiene
control, is free from pathogens. The presence of pathogens
is associated with external contamination factors.3
There is scientific evidence of the efficacy and safety of
human milk pasteurization in the elimination of pathogenic
agents.4-6 However, the presence of mycelial fungi in
expressed human milk can be a problem when the milk is to
be used in natura, in case of mother’s milk given directly to
the infant, as occurs at neonatal ICUs.
Fungi are a diversified group of organisms with
remarkable ecological and economic importance.
Approximately 70,000 fungal species have already been
identified, but their total number is estimated at 1.5 million.7
These fungal species are of relevant importance since they
are the primary decomposers in all terrestrial ecosystems,
form important symbiotic associations with vascular plants
(mycorrhiza), include most plant pathogens, provide genetic
systems for molecular biologists and are essential to
industrial biotechnology.7
However, these species of microorganisms are
undesirable in foods, since they can produce a wide variety
of enzymes that cause deterioration. On top of that, several
fungi can produce toxic metabolites during their
reproduction. These metabolites are generally called
mycotoxins and, when eaten with foods, they cause harmful
biological alterations, which can range from allergies to
carcinogenesis.8
Since molds are largely present in nature, their occurrence
in expressed human milk can be a sign of external
contamination. In addition to deterioration, these
microorganisms can produce mycotoxins.3
Part of the microbiological criterion used worldwide for
screening milk-based foods consists in counting molds and
yeasts.10 Since the available literature does not contain
enough information on fungal microbiota of technological
interest for the preservation of expressed human milk, the
present study aims at characterizing the genera of mycelial
fungi found in samples of expressed human milk, thus
contributing towards filling this gap in literature.
Contamination of expressed human milk... - Novak FR et alii
Methods
The study was carried out at the Human Milk Bank of
Instituto Fernandes Figueira (BLH/IFF) with the aim of
describing the occurrence of mycelial fungi in 821 individual
samples of expressed human milk (EHM), obtained from
20% of the registered donors. The convenience samples
were randomly obtained from the flasks received from each
donor, in 262 different locations in the city of Rio de
Janeiro, between October 1998 and January 2000.
The donors were asked to follow the health and hygiene
protocol established by the Ministry of Health for the
milking procedure(11). After milking, the milk was stored
in flasks previously sterilized at 121°C for 15 minutes and
kept at –6°C to –10°C, for at most 5 days, in the ice-box of
the refrigerator or in the freezer at the donor’s house. The
flasks were transported in isothermal boxes, which contained
a thermometer for the measurement of minimal and maximum
temperatures.
At the Human Milk Bank of Instituto Fernandes Figueira
(BLH/IFF), the flasks were thawed and samples were taken
for the screening of molds, yeasts and mesophilic
microorganisms. Since collection was not closely followed,
information such as time, lactation period, or whether the
milk was obtained at the beginning or at the end of the
milking procedure, was not taken into consideration.
The microbiological analyses were carried out at the
laboratory of food control of Instituto Fernandes Figueira,
and the identification of the strains of mycelial fungi was
made at the Laboratory of Mycology of the Hospital Evandro
Chagas Research Center (CPqHEC). Both laboratories
belong to Fundação Oswaldo Cruz, Rio de Janeiro, Brazil.
The screening of molds and yeasts was carried out
according to the method described by Marvin (1976).
Samples of 1.0 ml of milk and of its decimal dilutions were
plated in duplicate on potato dextrose agar (Merck) using
the pour plate technique. After medium solidification, the
plates were incubated at 25°C for five days. The colonies
were counted and the results were expressed in colonyforming units per ml (CFU/ml).
Mesophilic bacteria were screened according to the
method described in the Compendium of Methods for the
Microbiological Examination of Foods.12 Samples of 1.0
ml of milk and of its decimal dilutions were selected and
plated in duplicate on standard agar (Merck) using the pour
plate technique. After medium solidification, the plates
were incubated at 35°C for 48 hours. The colonies were
counted and the results were expressed in CFU /ml. Mesophil
counts were performed since mesophils presented a direct
relationship with the total population of microorganisms
and could be somewhat proportionate to the population of
mycelial fungi, if both presented the same source of
contamination.12
Mycelial fungi were initially isolated from the plates
that presented mold and yeast growth. Typical colonies of
Contamination of expressed human milk... - Novak FR et alii
Jornal de Pediatria - Vol. 78, Nº3, 2002 199
mycelial fungi were selected, plated on potato dextrose
agar, and incubated at 25°C for 3-5 days. The pure cultures
that were isolated were maintained in tubes with the same
inclined agar up to identification.
we consider that donors were not exposed to any kind of
risks.
The identification of mycelial fungi was made by taking
the colonies from the primary isolation medium, inoculating
them onto Sabouraud agar and keeping them at room
temperature. After growth, the colonies with at least 10mm
in diameter were used in the preparation of slides for direct
microscopy. The colonies were stained with Aman’s
lactophenol and/or cotton blue, and were bleached with
NaOH at 10% so that the morphological structures of the
fungi could be observed. The fragments were examined in
a microscope and the morphologies were compared with
those described in specialized literature.13
Results
The sociocultural profile of the donors was as follows:
nurturing mothers from different sociocultural background
aged between 13 and 45 years, of which 2.1% were more
than 40 years old, and 13.7% were aged between 13 and 19
years. With respect to the level of education, 334 (40.7%)
donors had primary education, 254 (30.9%) had finished
their secondary education, and 233 (28.4%) had a university
degree.
Of the 821 analyzed samples of expressed human milk,
43 (5.2%) were contaminated with 48 strains of mycelial
fungi. Three samples presented two strains and one sample,
three strains, with different macroscopic aspects.
The results (Figure 1) show the presence of molds and
yeasts, with counts up to 103 CFU/ml, in addition to the
presence of the following microorganisms: Aspergillus
niger group (6.3%), Aspergillus sp. (4.2%), Paecilomyces
sp. (12.6%), Penicillium sp. (60.4%), Rhizopus sp. (2.0%)
and Syncephalastrum sp. (14.5%).
By studying the correlation between mold and yeast
counts and the total populations represented by mesophil
counts, we found a correlation coefficient (r) of 0.55, with
a 95%CI: 0.32 < R < 0.72, which does not show a strong
correlation between these two groups.
The statistical analysis involved two steps: frequency
distribution analysis (for assessing the sociocultural profile
of the donors) and the correlation test between the counts of
mycelial fungi and mesophilic microorganisms, by using
the System Analysis Statistical (SAS) software version
6.11, by SAS Institute Inc., Cary, NC, USA.
The study was based on the regulatory guidelines for
human research (CNS 196/96 resolution).14 After approval
by the Research Ethics Committee of Instituto Fernandes
Figueira (IFF), the samples of expressed human milk were
collected, with no change to the milking routine at the IFF
milk bank. The collected samples were used only for
microbial counts and for the isolation of strains of mycelial
fungi.
The data on the sociocultural profile of the donors were
directly obtained from the electronic file at the Human Milk
Bank of Instituto Fernandes Figueira (BLH-IFF). There
was no need to view the medical history of the donors in
order to get additional information, or establish personal
contact with them, thus preserving their identity. Therefore,
Discussion
Expressed human milk, even when obtained from healthy
women, following all hygiene procedures, may present
normal-occurring contaminants. However, the presence of
mycelial fungi of different species suggests that the health
Figure 1 - Count of molds and yeasts and mesophilic microorganisms found in the human
milk samples
200 Jornal de Pediatria - Vol. 78, Nº3, 2002
and hygiene care was not strictly followed during the
milking procedure.
The results of the present study show the presence of
molds and yeasts in only 5.2% of the samples, with counts
up to 103 CFU/ml, suggesting that most donors follow the
recommendations for collection and storage of expressed
milk.
The fungal spores present on the foods handled by the
donors are probably the source of the fungi detected in
expressed human milk, since they are very similar to those
found by other researchers in different foods.
A study on the presence of microorganisms in table
butters sold in the city of São Paulo revealed that 94% of the
samples contained the following fungi: Cladosporium (18%),
Penicillium (12%), Geotrichum (8%), Aspergillus (6%),
and Trichoderma (6%).15
Alexandre et al. (1996) investigated 71 samples of
dehydrated fruits sold in Santiago do Chile and observed
that 45 (73.8%) presented fungal growth, which usually
included Aspergillus, Rhizopus, Penicillium and Mucor.
Taniwaki et al. (1989) analyzed three apple cultivars
produced in the state of São Paulo and isolated the following
fungi: Cladosporium sp., Phoma sp., Fusarium sp.,
Trichoderma sp., Alternaria sp., Phompsia sp.
andPenicillium sp., in addition to other molds that could not
be identified.
Of 30 samples of cassava flour bought from different
stores in Niterói, state of Rio de Janeiro, Kraemer et al.
(1998) isolated fungi of the following genera: Aspergillus
(36.5%), Penicillium (18.2%), Rhizopus (10.5%),
Paecilomyces (7.1%), Mucor (5,4%), Neurospora (3.1%),
Cladosporium (2.3%), Aureobasidium (1.4%),
Syncephalastrum (1.1%), Metarrhizium (0.8%),
Trichoderma (0.3%), Trichosporon (0.3%) and Humicola
(0.3%).
Costa et al. (1986) detected molds and yeasts in 11.7%
of 2,533 samples of cow’s milk obtained from 32 properties
in 18 towns in the state of São Paulo.
In the present study, the milking procedure was carried
out at home by the donors themselves. We can infer that the
hygiene of the donors or of the collection site also contributed
to the presence of such contaminants. There is no doubt that
the environment plays an essential role in the quality of
expressed human milk, as widely discussed by Assis in
1981.20 The analysis of 1,143 samples of expressed human
milk carried out at the Human Milk Bank of Mother-Child
Institute of the State of Pernambuco (Instituto MaternoInfantil de Pernambuco) revealed a better microbiological
quality of the milk collected at the human milk bank in
relation to the milk collected at the rooming-in facilities and
at the donor’s home.21
Almeida, in 1986,3 observed molds and yeasts in 69.4%
of expressed human milk samples collected at Instituto
Contamination of expressed human milk... - Novak FR et alii
Fernandes Figueira, with counts of up to 106 CFU/ml.
Collection techniques were modified and counts were
repeated, and the incidence of these microorganisms was
reduced to 16.7%, whereas counts totaled less than 3.0 x
102 CFU/ml.
The presence of fungi in expressed human milk can be
directly related to cultural habits, showing that efficient
hand-washing is not a widespread practice in our country. 22
Most skin infections caused by fungi are associated with
dermatophytes, including different species of the genera
Microsporum, Trichophyton and Epidermophyton, which
cause several clinical manifestations known as
dermatophytosis or tineas.23 These genera of fungi were not
found in the present study.
The data obtained here show the importance of handwashing for donors who handle foods immediately before
the collection of expressed human milk, since fungal spores
found on foods can be passed from the hands of the donors
on to the milk. This is based on the fact that the fungi
identified in the present study are compatible with those
found in various foods.
Even though the protocols handed over by human milk
banks try to instruct donors properly, we should not forget
how valuable such information is when directly given by the
professionals involved, as these professionals can show
slight details that are crucial for the successful collection of
expressed human milk.
If expressed human milk is pasteurized, sensitive mycelial
fungi will be eliminated; however, when the milk is supplied
in natura to hospitalized preterm infants, the conditions of
collection, storage and transportation should be checked so
that the presence and consequences of the multiplication of
these contaminants can be avoided.
References
1. Almeida JAG. Amamentação: um híbrido natureza cultura. Rio
de Janeiro: Editora Fiocruz; 1999.
2. Ministério da Saúde. Informe Saúde. Ano 4. Nº 69. Brasília:
Ministério da Saúde; 2000.
3. Almeida JAG. Qualidade do leite humano coletado e processado
em bancos de leite [dissertação]. Viçosa: Universidade Federal
de Viçosa; 1986.
4. Wright KC, Feeney AM. The bacteriological screening of donated
human milk: laboratory experience of British Paediatric
Association’s published guidelines. J Infect 1998; 36:23-7.
5. Lepri L, Del Bubba M, Maggini R, Donzelli GP, Galvan P. Effect
of pasteurization and storage on some components of pooled
human milk. J Chromatogr B Biomed Sci 1997; 704:1-10.
6. Baum JD. Donor breast milk. Acta Paediatr Scand 1982;
299(Suppl): 51-7.
7. Li S, Marquardt RR, Abramson D. Immunochemical detection of
molds: a review. J Food Prot 2000; 63:281-91.
Contamination of expressed human milk... - Novak FR et alii
8. Corrêa B, Galhardo M, Costa EO, Sabino M. Distribution of
molds and aflatoxins in dairy cattle feeds and raw milk. Rev
Microbiol 1997; 28:279-83.
9. Baldissera MA, Santurio JM, Canto SH, Pranke PH, Almeida
CAA, Schimidt C. Aflatoxin, ochratoxin A and zearalenone in
animal feedstuffs in south Brazil. Part II. Rev Inst Adolfo Lutz
1993; 53:5-10.
10. ICMSF - International Comission on Microbiological
Specifications for Foods. New York: Academic Press; 1978.
11. Ministério da Saúde. Normas gerais para bancos de leite humano.
2nd ed. Brasília: Ministério da Saúde; 1998.
12. Marvin LS. Compendium of methods for the microbioloical
examination of foods. Washington: American Public Health
Association; 1976.
13. Figueiras MJ, Gené J. Atlas of Clinical Fungi. Centralbureau
voor Schimmelcultures, Baarn and Delft, The Netherlands/Spain:
Universitat Rovira i VirgiliReus; 1995.
14. Ministério da Saúde. Diretrizes e normas regulamentadoras de
pesquisas envolvendo seres humano – Resolução 196/96 do
Conselho Nacional de Saúde. Rio de Janeiro: Editora Fiocruz;
1998.
15. Paula CR, Gambale W, Iaria ST, Reis Filho SA. Ocorrência de
fungos em manteigas oferecidas ao consumo público no Município
de São Paulo. Rev Microbiol 1988;19:317-20.
16. Alexandre SM, Ibáñez HV, Thompson ML. Hongos filamentosos
contaminantes de superficie de diferentes frutas deshidratadas de
venta libre en Santiago de Chile. Rev Chil Infectol 1996; 13:210-5.
17. Taniwaki MH, Bleinroth EW, De Martin ZJ. Patulin producing
moulds in apple and industrialized juice. Colet Inst Tecnol
Alimentos 1989;19:42-9.
Jornal de Pediatria - Vol. 78, Nº3, 2002 201
18. Kraemer FB, Stussi JSP. Evaluation of samples of manioc flour
(Manihot utilissima). Hig Aliment 1998; 12:38-40.
19. Costa EO, Coutinho SD, Castilho W, Teixeira CM, Gambale W,
Gandra CRP, et al. Bacterial etiology of bovine mastitis in the
State of São Paulo. Rev Microbiol 1986;17:107-12.
20. Assis MAA. Estudos sobre a preservação de colostro humano
para Bancos de Leite [dissertação]. Campinas: UNICAMP;
1981.
21. Serva VB, Albuquerque MRG, Rolim EG. Evaluation of
microbiologic quality of the human milk from human milk bank.
Revista do IMIP 1991;5:30-3.
22. Novak FR. Ocorrência de Staphylococcus aureus resistente a
meticilina em leite humano ordenhado [tese]. Rio de Janeiro:
Universidade Federal do Rio de Janeiro; 1999.
23. Lima EO, Chaves LM, Oliveira NMC. Isolation of dermatophytes
of tineas in the region of João Pessoa beaches. Ciencia, Cultura
e Saude 1994;13:55-8.
Correspondence:
Dr. Franz Reis Novak
Banco de Leite Humano - Instituto Fernandes Figueira
Av. Rui Barbosa, 716 - Flamengo
CEP 22250-020 – Rio de Janeiro, RJ, Brazil
Telephone/Fax: + 55 21 2553.5669
E-mail: [email protected]