Microbiological quality of human milk from a Brazilian milk bank



Álvaro B Serafini; Maria Cláudia D P B André; Márcia A V Rodrigues; André Kipnis; Cynthia O Carvalho; Maria Raquel H Campos;Érica C Monteiro; Fábia Martins; Thiago F N Jubé

Departamento de Microbiologia, Imunologia, Parasitologia e Patologia do Instituto de Patologia Tropical e Saúde Pública da Universidade Federal de Goiás. Goiânia, GO, Brazil





OBJECTIVE: The objectives of the present study were to determine the prevalence of potentially pathogenic microorganisms that indicate the hygienic and sanitary conditions of human milk samples collected at a Human Milk Bank.
METHODS: Three hundred and thirty eight (338) samples of human milk collected from a milk bank in a maternity in the municipality of Goiânia, in the state of Goias, Brazil were submitted to microbiological analysis. The latter were plated on McConkey agar according to the type of bacteria. Among the total number of samples collected, 194 consisted of raw milk and the remaining 144 were pasteurized milk.
RESULTS: The presence of Staphylococcus spp., Streptococcus spp., yeasts and molds, and Enterobacteriaceae was verified in the raw milk samples. Staphylococcus aureus were isolated in 10 (5.2%) samples, Staphylococcus epidermidis in 28 (14.4%) samples, Streptococcus spp. in three (1.6%) samples, yeasts and molds in 43 (22.2%) and Enterobacteriaceae in 49 (25.3%) samples. In a hundred and forty four (144) samples which underwent thermal treatment Staphylococcus aureus was detected in five (3.5%) samples, Staphylococcus epidermidis in 15 (10.4%), Staphylococcus lugdenensis in two (1.4%), Streptococcus spp. in four (2.8%), yeasts and molds in 37 (25.7%), and Enterobacteriaceae in nine (6.3%).
CONCLUSIONS: Analysis indicated a high degree of contamination in raw human milk, and as for the pasteurized milk, despite elimination of the great majority of potentially pathogenic microorganisms, the percentage of yeasts and molds was higher than in raw milk, demonstrating that a lower degree of initial contamination would be necessary for pasteurization to be an efficient means of microbiological control.

Keywords: Milk banks. Milk, human. Quality control. 




Exogenous infections, originating from the hospital environment, deserve greater attention from public health professionals than endogenous ones, for human beings are the sources of the agents that provoke them. The former include hospital employees, patients and visitors as well as hospital instruments and equipment.5 Since it is very difficult to implement measures to control hospital infections in all sectors of the hospital simultaneously, efforts should be directed primarily towards higher risk sectors, such as nurseries, nutritional services, surgical wards, intensive care units and centers of sterilization.

Among these sectors, particular attention should be given to the nursery, due to cross-contamination or to the transmission of pathogenic or potentially pathogenic microorganisms by means of milk stored in the human milk bank, which may be considered a potential microbiological risk factor. It is widely acknowledged that for breastfed infants, the first six months can be the healthiest period in their lifetime. Human milk fulfills all of the baby’s nutritional and hygienic requirements.25 However, premature newborns do not have the strength to suckle their mother’s breasts and must be fed by other means. Furthermore, due to physiological or emotional problems, some mothers are unable to produce milk. On the other hand, milk derived from other animals may cause allergies. For these as well as for other reasons, many infants are fed milk from human milk banks (HMB), supplied by voluntary donors who produce a surplus.

The microbiological quality of expressed milk distributed by these milk banks is a public health issue, for the children who will consume this product have low resistance to neonatal infections. 1,2,13,21 The most important issue concerning HMBs is the bacteriological control of donated milk,11 for consumption of contaminated human milk may be the cause of neonatal diseases.23

It is important to obtain more epidemiological data on bacterial contamination of human milk and the development of educational work with the mothers, nurses, nurses’ aides, nutritionists, pediatricians, and intensive care physicians, must be carried out so as to make them aware of the risks involved in the preparation and consumption of human milk.21 The purpose of this study is to investigate the prevalence of microorganisms in milk being offered by a human milk bank so as to subsidize the competent authorities conducting programs for the prevention of neonate infections.



Three hundred and thirty eight (338) samples of human milk were collected from a milk bank in a maternity in the municipality of Goiânia, in the state of Goias. Among the total number of samples collected, 194 consisted of raw milk and the remaining 144 were pasteurized milk. These samples were packaged and sent immediately to the Laboratório de Microbiologia de Alimentos do Departamento de Microbiologia do Instituto de Patologia Tropical e Saúde Pública de Goiânia (Laboratory of Food Microbiology of the Department of Microbiology of Tropical Pathology and Public Health of Goiania), where they were submitted to microbiological analysis.

Due to the quantity of samples and the fact that the material in question was produced by secretion, procedures adopted for microbiological analysis were those indicated by Koneman et al14 (1997) with samples being plated initially on blood agar and McConkey’s agar and, according to the type of bacteria isolated, posterior identification was undergone in appropriate mediums. Bacteria pertaining to the Stapphylococci gender were tested for their capacity to produce coagulase and resistance to novobiocin. As to the Streptococci, based on hemolysis, tests of resistance to bacitracin, optochin and other biochemical tests were conducted. In order to identify gram-negatives, screening was undertaken in triple sugar iron (TSI) agar and identification was obtained by means of biochemical tests. Molds and yeast were identified by means of a technique indicated by the American Public Health Association25, which consisted in plating the samples on a 10% potato dextrose agar, acidified by tartaric acid until the pH was 3.5, and incubating it at 25o C±1 for 5 to 7 days.



Potentially pathogenic and/or commensal organisms were identified in 70.4% (136 /194) of the raw milk samples and 50.7% (73/144) of the samples of pasteurized milk (see Table 1).



Among the samples of raw milk, Streptococci of the viridans group were isolated in three (2.2%); 10 samples (7.35%) presented colonies of Staphylococcus aureus; S. epidermidis were isolated in 28 (20.59%) samples; and 49 (36.0%) samples were contaminated by Enterobacteriaceae. Furthermore, 43 (31.6%) samples of raw milk were contaminated by molds and yeast (see Table 2).



Microorganisms were identified in seventy-three of the 144 samples of pasteurized milk from the HMB. Among these, Staphylococcus lugdenensis, Streptococcus pyogenes, and Streptococcus of the viridans group were isolated in two (2.7%) samples; Staphylococcus aureus in five (6.9%) samples; S. epidermidis in 15 (20.6%) and Enterobacteriaceaewere detected in 12.3% of the samples. Furthermore, 37 (50.7%) samples tested positive for molds and yeasts (see Table 2).



The presence of high levels of contaminants in raw human milk entails in a reduction of its biological benefits because its nutrients are being utilized by the microflora present in the milk, thus diminishing its protective immunological qualities.2 These situations, in the majority of cases, lead to the classification of this product as improper for consumption, taking into account the vulnerability of the recipient infants. Furthermore, the greater the quantity and diversity of microorganisms present in the milk, the less efficient will the outcome of the process of pasteurization be. It may be noted that there is a great gap in the criteria for selection of milk for pasteurization.3

The prevalence of Staphylococcus aureus in the oropharynx of human beings varies from 35to 40% and in the mouth and saliva its prevalence varies from 10 to 35%. However, it’s presence in human milk may be interpreted as being due to contamination from a secondary source, such as the skin and nasal fossa, or unsatisfactory hygienic or sanitary conditions of the utensils employed in its manipulation. The greatest concern with respect to its presence is the occurrence of strains which produce toxins that are resistant to pasteurization.3

Pereira et al (1995) reported the presence of Staphylococcus in al the samples of maternal milk collected from 19 women with symptoms of mastitis. Among the 19 samples, 8 synthesized detectable quantities of enterotoxins, being that in some cases, it was observed that they also produced the toxin of the toxic chock syndrome.

As to the raw milk, the results presented in this study are similar to those reported by Caroll et al7 (1979) who observed S. aureus in 13 (6.2%) of the 207 samples of expressed human milk. Other researchers found coagulase positive Staphylococcus in different proportions of human milk samples submitted to analysis: 28.1% (Nikodemuz16), 29% (Almeida et al3), 51.7% (Almeida et al2) and furthermore, among 7,570 samples, bacterial growth was observed in 230, being that 40% of these were S. aureus.8

Wyatt and Mata26 (1969), isolated enterobacterias in 18% of the 51 samples of colostrum they submitted to analysis, revealing the low level of personal hygiene and sanitary conditions of the study population, living in the small village of Santa Maria Cauqui in Guatemala.

Other authors found E. coli in 8.5% of the 59 samples of maternal milk22, in 2% of 44 samples9, indicating faecal contamination of the milk, and enterobacterias in 15 (7%) samples. The present results reveal the presence of enterobacterias in 25.3% of the samples of raw milk analyzed and in 6.2% of the samples of pasteurized milk. Therefore, they presented inferior hygienic conditions than the studies cited above.

Among the enterobacterias, coliforms have been singled out as particularly important in bacteriological control of HMBs, since their presence may indicate faecal contamination, even if it originates from an indirect source and doesn’t necessarily imply in the isolation of E. coli.2 Research has indicated that contamination by coliform microorganisms may originate from the environment, as observed in a study in which this group of bacteria were isolated in 38.4% of a total of 472 samples17, and among a total 5,710 samples of human milk, coliforms were detected in 1,139 (19.9%)17 and in a total of 837 samples, 71 (8.48%) were contaminated with coliform microorganisms, whereas only three of these were E. coli.18 In this study, coliforms were present in 21.1% of the samples of raw milk and in 5.6% of the pasteurized milk samples, indicating that the efficiency of pasteurization may be compromised by an initially high count of microbes.

The presence of molds and yeast in foods may also indicate contamination originating from the environment or resulting from handling in inadequate hygienic and sanitary conditions.15 Almeida4 (1986) found a 69.4% prevalence of fungus and yeast in samples obtained after employing the usual methods of hygiene (soap and water) on the mammary gland. Other researchers, however, observed that 230 (6.5%) of a total of 7,570 samples analyzed contained molds and yeast.8 The results of this study, in which molds and yeast were detected in 22% of the samples of raw milk, are inferior to those observed by other authors. However, the percentages of molds and yeast detected in the pasteurized milk (25.7%) are a matter of concern, for they indicate a possible environmental contamination occurring after the process of pasteurization or that this process may be ineffective in itself. The presence of pathogenic yeast in pasteurized human expressed milk suggests that this could be a source of infection to neonates during breastfeeding. Quality control by means of yeast counts in expressed human milk may be a good indicator of problems related to hygiene, storage or transportation.21

Inadequate pasteurization not only presumably endangers the beneficial properties of expressed human milk, but may also increase the susceptibility to subsequent contamination.6

The RDC ( Resolução da Diretoria Coletiva) [Resolution of the Collective Directory] Resolution no 12 , instituted by the Ministry of Health14 in January, 2001, establishes criteria for the microbiological control of human milk for the first time in Brazil. According to this resolution, the quantity of viable mesophilic aerobic microorganisms permitted is up to 100 CFU/mL, coliforms at 35oC and Staphylococcus coagulase-positive absent in 1 mL and Salmonella sp absent in 25 mL. According to this resolution, 4.9% of the samples analyzed in the present investigation were improper for human consumption, for they did not present the analytical parameters established in Annex 1 of the aforementioned Resolution. Taking into consideration the results obtained, the authors believe that efforts should be made to improve the microbiological control of expressed human milk, including the milk which is going to be pasteurized. In this sense, more rigorous measures for monitoring the quality of human milk are indispensable so as to guarantee safe feeding for neonates.



1. Almeida JAG. Brasil perde 180 milhões de litros de leite materno no Rio. Folha de São Paulo 12 de maio de 1999; Caderno Cotidiano: 3-5.        

2. Almeida JAG, Novak FR, Almeida CHG, Serva VB. Avaliação parcial da flora microbiana do leite humano ordenhado no IMIP. Rev Inst Mat Inf Pernambuco 1989;3:13-6.        

3. Almeida JAG, Novak FR, Silva IS. Estudo da ocorrência de Staphylococcus aureus em amostras de leite humano odenhado. In: I Congresso Brasileiro de Bancos de Leite Humano; 8-12 de julho de 1998. Brasília (DF); 1998. p. 10.        

4. Almeida JAG. Qualidade do leite humano coletado e processado em bancos de leite humano [Tese de mestrado]. Viçosa: Universidade Federal de Viçosa; 1986.        

5. Bittencourt BB, Melo CG. A infecção hospitalar e o serviço de nutrição. Rev Paul Hosp 1972;20:28-31.        

6. Bjorkstén B, Burman LG, Chatev P, Fredrikzon B, Gothefors L, Hernell O. Collecting and banking human milk: to heat or not to heat? BMJ 1980;281:765-9.        

7. Carroll L, Davies DP, Osman M, McNeish A. Bacteriological criteria for feeding raw breast-milk to babies on neonatal units. Lancet 1979;2:732-3.        

8. Colaço W, Serva VB, Lira CS. Perfil microbiológico do leite humano ordenhado, distribuído no banco de leite humano do IMIP, no período de julho/95 a dezembro/99. In: XII Encontro Nacional de Analista de Alimentos; 04 a 08 de novembro de 2001. Maceió; 2001. p. 213.        

9. Eidelman AI, Szilagyi G. Patterns of bacterial colonization of human milk. Obstetrics & Gynecology 1979;53:550-2.        

10. Herceg RJ, Peterson LR. Normal flora in health and disease. In: Mandell, Douglas, Bennilt, editors. Principles and pratice of infectious disease. 3rd ed. New York: Churchill Levingstone; 1990. p. 6-14.        

11. Ikonem RS, Miettinem A, Groonos P. Bacteriological quality control in a human milk bank. Klin Pädiatr 1982;194:295-7.        

12. Koneman EW, Allen SD, Janda WM, Schreckenberger PC, Winn Jr WC. Color atlas and textbook of diagnostic microbiology. 5th ed. Philadelphia: Lippincott; 1997.        

13. Law B, Urias B, Lertzman J, Robson D, Romance L. Is ingestion of milk-associated bacteria by premature infants fed raw human milk controlled by routine bacteriologic screening? J Clin Microbiol 1989;27:1560-6.        

14. Ministério da Saúde. Resolução no 12/2001 da Agência Nacional de Vigilância Sanitária – ANVISA. Regulamento Técnico Sobre Padrões Microbiológicos Para Alimentos. Diário Oficial da União. Brasília (DF); 10 de janeiro de 2001.        

15. Mislivec PB, Beuchat LR, Cousin MA. Yeasts and molds. In: Vanderzant C, Splittstoesser DF, editors. Compendium of methods for the microbiological examination of foods. 3rd ed. Washington (DC): American Public Health Association; 1992. p. 239-49.        

16. Nikodemuz I. Microflora in human milk sample. Nahrung 1986;30:901-6.        

17. Novak FR, Almeida JAG, Almeida CHG. Detecção e identificação de coliformes em leite humano ordenhado. Rev Inst Mat Inf Pernambuco 1989;3:17-9.        

18. Novak FR, Almeida JAG, Asensi MD, Moraes BA, Rodrigues DP. Resistência antimicrobiana de coliformes isolados de leite humano ordenhado. Cad Saúde Pública 2001;17:317-21.        

19. Pereira ML, Santos EJ, Sellos I, Bergdoll M. Staphylococci in breast milk from women without mastites. Rev Microbiol 1995;26:117-20.        

20. Picollo RC, Pimentel EP, Fávero LM, Rizzo MA. Surto de salmonelose ocorrido em cantina escolar, no município de São Paulo, em 1991. Rev Hig Alimentar 1992;6:28-30.        

21. Rosa CA, Novak FR, Almeida JAG, Hagler LC, Hagler AN. Yeasts from human milk collected in Rio de Janeiro, Brazil. Rev Microbiol 1990;21:361-3.        

22. Szollosy E, Marjai E, Lanatos J. Bacterial contamination and sparing heat treatment of mother's milk. Acta Microbiol Acad Sci Hungarian 1974;21:319-25.        

23. Tyson J, Edwards W, Rosenfeld A, Beer A. Collection methods and contamination of bank milk. Arch Dis Child 1982;57:396-8.        

24. Unicef - Fundo das Nações Unidas para a Infância. Importância do leite materno na infância. Rev Hig Alimentar 1984;3:172-3.        

25. Vanderzant C, Splittstoesser DF, editors. Compendium of methods for the microbiological examination of foods. 3rd ed. Washington (DC): American Public Health Association; 1992.        

26. Wyatt RG, Mata LJ. Bacteria in colostrum and milk of Guatemalan Indian Women J Pediatr 1969;91:102-7.        



Correspondence to
Álvaro Bisol Serafini
Instituto de Patologia Tropical e Saúde Pública
Rua Delenda Rezende de Melo, s/n Setor Universitário
74605-050 Goiânia, GO, Brazil
E-mail: abisol@iptsp.ufg.br

Received on 23/1/2002. Reviewed on 27/6/2003. Approved on 3/7/2003.
Financiado pela Fundação de Apoio à Pesquisa/FUNAPE/UFG (Processo n 66.357).

Faculdade de Saúde Pública da Universidade de São Paulo São Paulo - SP - Brazil
E-mail: revsp@org.usp.br