Infectivity of cysts of the ME-49 Toxoplasma gondii strain in bovine milk and homemade cheese

Infecciosidade de cistos de Toxoplasma gondii ME-49 em leite bovino e queijo caseiro

Abstracts

OBJECTIVE: Analyze the infectivity and storage resistance of cysts of the ME-49 strain of Toxoplasma gondii in artificially infected bovine milk and homemade fresh cheese. METHODS: Pasteurized bovine milk was infected with 10 cysts/ml of the ME-49 strain of T.gondii and inoculated in different groups of mice, immediately or after storage at 4ºC for 5, 10 and 20 days. Homemade fresh cheese was prepared with artificially infected milk, and also tested in groups of mice, using the same storage process. Infection was identified by the presence of cysts in the brain or serological testing in challenged mice after 5 weeks, confirmed by Western Blot and histology. RESULTS: The infectivity of cysts of the ME-49 strain of T.gondii was maintained in the milk even after storage for 20 days at refrigerator temperatures. Cysts were also able to survive the production process of homemade fresh cheese and storage for a period of 10 days in the same conditions. CONCLUSIONS: These data demonstrated that milk and dairy products could be an important source of T.gondii in human contamination, reinforcing the importance of milk pasteurization before any processing or ingestion.

Toxoplasmose; Milk; Cheese; Food contamination; Food hygiene; Food preservation; Toxoplasmosis, animal; Mice; Enzyme-linked immunosorbent assay; Antibodies; Toxoplasma gondii


OBJETIVO: Analisar a infecciosidade e a resistência de cistos de T. gondii em leite e queijo fresco caseiro, pela infecção artificial de leite bovino. MÉTODOS: O leite bovino pasteurizado foi infectado artificialmente com 10 cistos/ml de T.gondii cepa ME49 e inoculado em grupos de camundongos, imediatamente ou após ser estocado por 5, 10 e 20 dias a 4ºC. Preparou-se queijo fresco caseiro com leite infectado, sendo testado em grupos de camundongos, utilizando a mesma conservação. A infecção foi detectada pela presença de cistos no cérebro dos camundongos desafiados ou testes sorológicos após cinco semanas, também confirmada por Western Blotting e histologia. RESULTADOS: A infecciosidade dos cistos da cepa ME49 de T.gondii foi mantida mesmo quando armazenado no leite até 20 dias de conservação em condições de refrigeração a 4ºC. Os cistos resistiram ao processo de fabricação do queijo e eram infectantes após um período de 10 dias nas mesmas condições. CONCLUSÕES: Os achados mostraram que o leite e seus derivados podem ser uma importante fonte de contaminação humana pelo T.gondii, reforçando a importância da pasteurização do leite antes de qualquer processamento ou ingestão.

Toxoplasmose; Leite; Queijo; Contaminação de alimentos; Higiene dos alimentos; Conservação de alimentos; Toxoplasma; Toxoplasmose animal; Camundongos; ELISA; Anticorpos antiprotozoários; Toxoplasma gondii


Infectivity of cysts of the ME-49 Toxoplasma gondii strain in bovine milk and homemade cheese* * Supported by CNPq (Process n. 143364/1998-7), Fapesp (Process n. 98/1681-0, 98/13323-0, 96/5875-8 & 99/04926-6), Fundação do Desenvolvimento Administrativo (Fundap) and Laboratório de Investigação Médica do Hospital das Clínicas Faculdade Medicina USP-49.

Infecciosidade de cistos de Toxoplasma gondii ME-49 em leite bovino e queijo caseiro

RM Hiramoto, M Mayrbaurl-Borges, AJ Galisteo Jr, LR Meireles, MS Macre and HF Andrade Jr

Laboratório de Protozoologia, Instituto de Medicina Tropical de São Paulo, Faculdade de Medicina da Universidade de São Paulo. São Paulo, SP, Brasil

KEYWORDS
Toxoplasmose, transmission.# Milk, microbiology.# Cheese, microbiology.# Food contamination.# Food hygiene, methods. Food preservation. Toxoplasmosis, animal, chemically induced. Mice, blood. Enzyme-linked immunosorbent assay. Antibodies, protozoan. ¾ Toxoplasma gondii. ABSTRACT

OBJECTIVE:

Analyze the infectivity and storage resistance of cysts of the ME-49 strain of Toxoplasma gondii in artificially infected bovine milk and homemade fresh cheese.

METHODS:

Pasteurized bovine milk was infected with 10 cysts/ml of the ME-49 strain of T.gondii and inoculated in different groups of mice, immediately or after storage at 4oC for 5, 10 and 20 days. Homemade fresh cheese was prepared with artificially infected milk, and also tested in groups of mice, using the same storage process. Infection was identified by the presence of cysts in the brain or serological testing in challenged mice after 5 weeks, confirmed by Western Blot and histology.

RESULTS:

The infectivity of cysts of the ME-49 strain of T.gondii was maintained in the milk even after storage for 20 days at refrigerator temperatures. Cysts were also able to survive the production process of homemade fresh cheese and storage for a period of 10 days in the same conditions.

CONCLUSIONS:

These data demonstrated that milk and dairy products could be an important source of T.gondii in human contamination, reinforcing the importance of milk pasteurization before any processing or ingestion.

DESCRITORES
Toxoplasmose, transmissão.# Leite, microbiologia.# Queijo, microbiologia. Contaminação de alimentos.# Higiene dos alimentos, métodos. Conservação de alimentos. Toxoplasma, patogenicidade. Toxoplasmose animal, induzido quimicamente. Camundongos, sangue. ELISA. Anticorpos antiprotozoários. ¾ Toxoplasma gondii. RESUMO

OBJETIVO:

Analisar a infecciosidade e a resistência de cistos de T. gondii em leite e queijo fresco caseiro, pela infecção artificial de leite bovino.

MÉTODOS:

O leite bovino pasteurizado foi infectado artificialmente com 10 cistos/ml de T.gondii cepa ME49 e inoculado em grupos de camundongos, imediatamente ou após ser estocado por 5, 10 e 20 dias a 4oC. Preparou-se queijo fresco caseiro com leite infectado, sendo testado em grupos de camundongos, utilizando a mesma conservação. A infecção foi detectada pela presença de cistos no cérebro dos camundongos desafiados ou testes sorológicos após cinco semanas, também confirmada por Western Blotting e histologia.

RESULTADOS:

A infecciosidade dos cistos da cepa ME49 de T.gondii foi mantida mesmo quando armazenado no leite até 20 dias de conservação em condições de refrigeração a 4oC. Os cistos resistiram ao processo de fabricação do queijo e eram infectantes após um período de 10 dias nas mesmas condições.

CONCLUSÕES:

Os achados mostraram que o leite e seus derivados podem ser uma importante fonte de contaminação humana pelo T.gondii, reforçando a importância da pasteurização do leite antes de qualquer processamento ou ingestão.

INTRODUCTION

Toxoplasma gondii is an obligatory intracellular Apicomplexa parasite, with a complex life cycle, that causes toxoplasmosis in humans, and domesticated and wild animals. Infection is usually asymptomatic in people with a normal immune function and occasionally results in eye involvement. Toxoplasmosis can cause severe disease in fetuses of acutely infected pregnant woman, immunocompromised individuals (Aids) and therapeutically immunosuppressed patients, as cancer or transplant recipients.2 In Brazil, T.gondii infects around 60% of the adult population.13 Toxoplasmosis is acquired by ingesting food and water contaminated with oocysts from feces of infected cats or by the ingestion of raw or undercooked meat, containing tissue cysts, of infected intermediary host.

Milk was implicated as a source of infection of T.gondii in several reports. Toxoplasmosis was described in breast fed child whose mother had recently acquired toxoplasmosis1 and ingestion of unpasteurized goat milk was considered a possible source of T.gondii infection in children of rural areas,3,17,20 showing the importance of milk in the spread of T.gondii infection. Experimentally, there are reports of chronically infected lactating mice excreting cysts in their milk.16

This transmission was attributed both to tachyzoites in the milk and suckling trauma but also to tissue cyst excretion16 due to specific exocytic lipid secretion in the mammary cell that allows the secretion of lipidic droplets with the luminal upper layer of the cell cytoplasm,5 described in some works of cyst isolation in milk from acutely infected goats.6 Despite of those few reports, studies were conducted to verify the infectivity of T.gondii cysts in milk, without any description of their infectivity in fresh homemade cheese, a common protein source in rural areas. The present study assesses the infectivity of cysts of the ME-49 strain of T.gondii in artificially infected bovine milk and derived fresh homemade cheese.

METHODS

Strains, material and animals

The ME-49 strain of T.gondii was cryopreserved in liquid nitrogen or by successive passages of cysts from infected brains, homogenized in phosphate buffered saline (PBS) and inoculated intraperitoneal (i.p.) in C57Bl/6j mice. RH strains have been maintained routinely in the Laboratório de Protozoologia in the past 15 years by cryopreservation or by i.p. passage in mice. All plastic ware, reagents and conjugates were purchased from one manufacturer and the solutions were prepared with high-quality water. Isogenic C57Bl/6j mice were obtained from São Paulo University's colony (Centro de Bioterismo/Faculdade de Medicina/USP) and kept in sterilized cages and absorbent media with food and water "ad libitum". The management of these animals before or during the experiments followed the Principles of Laboratory Animal Care14 and the Principles of Ethics in Animal Experimentation (COBEA).4

Experimental infection

Pasteurized bovine whole milk found in the market was infected with cysts of the ME-49 strain from brains of previously infected C57Bl/6j mice. The contaminated milk was administrated orally to groups of eight mice, fresh or stored at 4oC for 5, 10 and 20 days. A dose of 12 cyst/mouse was used. Cysts kept and stored under the same conditions in sterile PBS were used as controls, with mice follow-up after 5 weeks.

Homemade fresh cheese was produced with manual technique, using products available in the market for Brazilian farmers. The pasteurized milk was contaminated with cysts of T.gondii of the ME-49 strain (12 cysts/1000 ml milk) in microcentrifuge tubes. After that, 1ml of commercial clot was added per tube and incubated for 30 min at 37oC. After centrifuging at 800 g/5 min, the supernatant was discarded, the cheese washed out with saline solution by centrifugation, removed from tubes and stored in a refrigerator. The cheese was offered individually to groups of four C57Bl/6j mice, fresh or stored at 4oC for 5, 10 and 20 days. Ingestion was carefully observed for each mouse, usually occurring in the first half-hour. Mortality, morbidity and serological evidence of infection in C57Bl/6j mice were monitored by 40 days after the ingestion.

Mice blood samples

Blood samples (5 ml) were collected from the mice tails in standardized filter papers dried and stored at ¾20oC until use. Before use, the soluble extract containing antibodies was obtained by extraction with 100 ml PBS during 18 hours at 4oC, with gentle agitation. The soluble extract was recovered by centrifugation and considered a 1/20 dilution of blood.

T.gondii antigen preparation and ELISA

Tachyzoites of the RH strain were harvested from the peritoneal cavity of previously infected mice in phosphate buffered saline (PBS), recovered by centrifugation, washed, counted and submitted to sonication for several periods of 4 cycles/30 seconds in an ice bath, until all parasites were destroyed, as detected by phase contrast microscopy. The solution was isotonized by 0.3 M NaCl addition (v/v) and cleared by centrifugation at 10,000 g for 3 min. The protein content of the supernatant was determined and aliquots were maintained at ¾70oC until use as purified antigen. Ninety-six high binding plates (Multiwell Plate/polystyrene) were coated with 50 ml/well of a solution containing 10 mg/ml of the antigen extract in carbonate-buffered 0.1 M, pH 9.0, overnight at 4oC. The plates were then washed 5 times with PBS containing 0.02% Tween 20 (PBS-T), and free protein binding sites blocked for 1 hour at 37oC with PBS containing 2% fat-free lyophilized milk in humid chamber. Serum dilutions, 100 ml, were added to each well and the plates were incubated for 1 hour at 37oC. After additional washing with PBS-T, bound IgG were detected by incubation for 1 hour with peroxidase-conjugated anti-mouse immunoglobulin G at 37oC, followed by 5 washings with PBS-T. The bound conjugates were developed with 100 ml per well of orto¾phenylenediamine dihydrochloride 1 mg/ml, H2O2 0.03% in phosphate-citrate buffer 0.2 M, pH 5.0, for 30 min, stopped with 25 ml per well of HCl 4N, and the optical density (OD) read at 492 nm in an ELISA reader.

Western blot

Specific antigen reactive antibodies were detected on Sodium dodecyl sulfate ¾ polyacrylamide gel electrophoresis (SDS-PAGE) isolated antigens. Purified parasites, RH strain from peritoneal exudates of previously infected mice (108/ml) were sonicated to disruption in PBS and immediately centrifuged for 10,000 g/3 min. The supernatant, 400 mg protein/ml, was collected and submitted to SDS-PAGE in a 12,5% separating gel, with subsequent eletrophoretic transfer to nitrocellulose membranes. The membranes were blocked with a 5% fat-free lyophilized milk in PBS-T, cut in 3 mm strips and stored dried at ¾20oC until use. Diluted serums were reacted with T.gondii antigen strips re-hydrated with PBS-T, for 18 hours at 4oC. After careful washing with PBS-T, the antibodies were detected by sequential incubation with peroxidase-conjugated anti-mouse IgG, 60 min, with subsequent PBS washings, and developed with 4-cloro-1-naphtol 0.6 mg/ml and hydrogen peroxide 0.03% for 15 min, with constant agitation. After development, the strips were dried and scanned for documentation. Standard molecular weight protein markers were stained with Amido Black B.

Histology

All mice were killed under ethyl ether anesthesia 5 weeks after the challenge. Brain, heart, lung, liver and spleen were removed and fixed in 100 volumes of 10% formaldehyde in Sorensen buffer,with once solution replacement after 1 hour. The organs were included in paraffin, and a 7 mm thick section was obtained and stained with hematoxylin-eosin. When necessary, representatives fields were photographed in a Zeiss Axiophot microscope with planapochromatic optics.

Data analysis

A mouse was considered infected when there were T.gondii cysts identifiable in the squash preparation of central nervous system tissue, or when a positive serology was found in ELISA, confirmed by Western blot of specific bands. Comparison between frequencies was performed using Epi Info 6.01 statistical software, using 1-tail Fisher's exact test. Significance was considered when the probability of equality between frequencies was less than 0.05.

RESULTS

T.gondii cysts infectivity in milk

Mice mortality induced by 12 cysts in milk, related to storage, can be seen in (Table 1). Deaths due to infection in most animals occurred by neurological symptoms after 20-30 days of infection, with a decaying pattern with longer storage periods. Interestingly, when compared with PBS stored cysts, the cysts maintained in milk induce a more severe disease, specially when compared in 10 days storage time (p=0.038, 1-tailed Fisher exact test).

Serology of animals, both from survivors or before death in all animals, tested tail blood samples using ELISA as described, could be seen in Table 2. All animals were infected up to five days of storage, regardless of medium, but cysts stored in milk had a prolonged survival, infecting all animals after 20 days of storage (p=0.038, 1-tailed Fisher exact test), when compared to PBS stored cysts in the same period. The positive serology was confirmed with Western Blot, with clearly identifiable bands in similar control infected mice (data not shown). The histological picture of one surviving mouse, sacrificed in the 40th day, show an intense encephalitic process with several T.gondii cysts dispersed in brain with foci of inflammatory cells.

T.gondii infectivity in homemade cheese

Only one animal died in the group that ingested fresh cheese immediately after it was made. All other groups showed no deaths during the experiment. After 40 days, all mice groups were killed, followed by search of cysts of T.gondii in the brain, and serological testing. T.gondii cysts were found in the brain of all mice from the groups that had received homemade cheese fresh or stored during 5 or 10 days at 4oC, a result also confirmed by specific serology, as seen in Table 3. Mice that received cheese stored for 20 days at 4oC didn't show either brain cysts or seroconversion. The histological feature of a mouse infected with cheese stored for 10 days, confirmed by serology, presented fewer inflammatory foci and T.gondii cysts, despite evidence of encephalitis, suggesting a less intense disease. The infection was similar in both forms of contamination, as observed by the detection of tissue cysts, production of specific antibodies with recognition of the same protein bands, without any selective effect of the milk or cheese on the agent, varying only in parasite.

DISCUSSION

These data show clearly that milk and homemade cheese could keep up the infectivity of T.gondii cysts for periods up to 10 days for homemade cheese and up to 20 days for milk, under regular refrigeration conditions. Transmission of human toxoplasmosis occurs mainly through the ingestion of food containing cysts of T.gondii, found in sheep, pigs, cow, hen and goats.6-8 Toxoplasmosis outbreak has been mainly related to the ingestion of undercook meat,1 but few outbreaks could be attributed to milk ingestion despite T.gondii was isolated in the milk of naturally infected cows.19 The ingestion of raw milk has been considered as a possible source of the transmission of T.gondii,18 and parasites were isolated, by mouse inoculation, in goat milk.3 Goat milk is usually produced by small farms producers under poor conditions without pasteurization, only frozen, with low safety for contaminations,15 but most milk sold in urban areas is pasteurized, a process that eliminates the T.gondii infectivity.10

These data also show that milk promoted higher infectivity and mortality, revealing probably better preservation or efficiency of T.gondii cysts when compared to PBS as diluent solution, even in the first periods of storage. This fact could be ascribed to nutrients in the milk that contributed for maintaining the cysts viable. Another explanation is related to the first steps of host invasion. The bradyzoites, the form found in the cysts, are resistant to the gastric digestion, a method for differentiating the tachyzoites.12 Milk proteins can generate a competitive effect in the gastric juice, resulting in better infectivity of bradyzoites, both by reducing the acid content and allowing a more restrict pepsin digestion, as demonstrated in infected milk artificially exposed to gastric conditions.16

Homemade cheese was also a good vehicle for T.gondii cysts. The home processing of cheese involves bacterial acidification and coagulation of the casein, but this should be a mild process to maintain the palatability of the cheese. The clot could retain the agent during washing but the entire process reduces the viability of the cysts, as demonstrated by the cessation of transmission after 10 days of storage in the refrigerator. Despite its infectivity, the disease was quite less intense in mice that ingested experimentally infected cheese, a fact that could explain the absence of related deaths on this form of transmission that could induce a less symptomatic disease in humans.

The cheese contamination could occur in two ways, first the exogenous one, associated with low hygiene conditions, allowing the contact of milk with oocysts from cats. Endogenous contamination of the milk in the mammary gland is facilitated by cellular exocytosis of milk secretion. During pre-lactation, relatively stable mammary cells could harbor T.gondii cysts. Those silent cysts could be secreted from the mammary gland cells by exocytosis, coated by host cell membranes, similar to milk fat globules secretion,5 allowing milk contamination inside the gland. This contamination is worsened due to smaller concentration of proteolytic enzymes in the intestine of children and suckling animals, increasing the survival of T.gondii forms.9

Toxoplasmosis transmission by milk or fresh cheese, unpasteurized or inadequately processed, can be one of the important ways of contamination for this agent, as they areimportant food source in rural areas. Milk and cheese are one of the main sources of proteins for children, and cheese is a cheap and easy way of getting food proteins. Any reference about the viability of the cysts of T.gondii in fresh cheese was not found, just in milk, mainly of ovine.6 The agent is destroyed after pasteurization when appropriately exposed to larger temperatures than 70ºC for 10 min, but pasteurization is sometimes not appropriate or the milk is processed in large volumes with some areas without adequate temperature, a frequent problem.11 Cheese production after boiling is a time consuming and expensive process, especially in small farms, despite its efficiency in the elimination of this contaminant.

The present study showed that T.gondii cysts could survive in the milk and fresh derived cheese, which could be considered a potential source of T.gondii transmission, especially in rural areas where there's no pasteurization.

ACKNOWLEDGMENTS

To Ricardo T. Gazzinelli of the Fiocruz-MG by supplying of T.gondii ME-49 strain; to Roselaine Pereira Alvin Cardoso for technical assistance in serological assays and antigen production (Laboratório de Protozoologia); to Cleiton Alves for technical assistance in histology preparation (Laboratório de Dermatologia), both of the Instituto de Medicina Tropical da Universidade de São Paulo.

Correspondence to:

Heitor Franco de Andrade Junior

Lab. Protozoologia do Instituto de Medicina Tropical ¾ USP

Av. Dr. Enéas de Carvalho Aguiar, 470

05403-000 São Paulo, SP, Brazil

E-mail: hfandrad@usp.br

The publication of this article was subsidied by Fapesp (Process n. 01/01661-3).

Submitted on 14/7/2000. Reviewed on 9/10/2000. Approved on 1/12/2000.

  • 1. Bonametti AM, Passos JN, Koga de Silva EM, Macedo ZS. Probable transmission of acute toxoplasmosis through breast feeding. J Trop Pediatr 1997;43:116.
  • 2. Beaman MH, McCabe RE, Wong SY, Remington HS. Toxoplasma gondii In: Mandell GL, Bennett JE, Dolin R. Principles and practice of infectious disease New York: Churchill Livingstone; 1995. p. 2455-75.
  • 3. Chiari CA, Neves DP. Toxoplasmose humana adquirida através da ingestăo de leite de cabra. Mem Inst Oswaldo Cruz 1984;79:337-40.
  • 4
    [COBEA] Colégio Brasileiro de Experimentação Animal. Princípios éticos na experimentação animal. Available from: URL: http://www.meusite.com.br/COBEA/etica.htm
  • 5. Deyrup-Olsen I, Luchtel DL. Secretion of mucous granules and other membrane-bound structures: a look beyond exocytosis. Int Rev Cytol 1998;183:95-141.
  • 6. Dubey JP. Persistence of encysted Toxoplasma gondii in caprine livers and public health significance of toxoplasmosis in goats. J Am Vet Med Assoc 1980;177:1203-7.
  • 7. Dubey JP, Sharma SP. Parasitemia and tissue infection in sheep fed Toxoplasma gondii oocysts. J Parasitol 1980;66:111-4.
  • 8. Dubey JP, Murrell KD, Fayer R, Schab GA. Distribution of Toxoplasma gondii tissue cysts in commercial cuts of pork. J Am Vet Med Assoc 1986;188:1035-7.
  • 9. Dubey JP. Toxoplasmosis: an overview. Southeast Asian J Trop Med Public Health 1991;22(Suppl):88-119.
  • 10. Dubey JP. Strategies to reduce transmission of Toxoplasma gondii to animals and humans. Vet Parasitol 1996;64:65-70.
  • 11. Fahey T, Morgan D, Gunneburg C, Adak GK, Majid F, Kaczmarski E. An outbreak of Campylobacter jejuni enteritis associated with failed milk pasteurization. J Infect 1995;31:137-43.
  • 12. Gross U, Bohne W, Soęte M, Dubremetz JF. Developmental differentiation between Tachyzoites and Bradyzoites of Toxoplasma gondii Parasitol Today 1996;12:30-3.
  • 13. Guimarăes ACS, Kawarabay M, Borges MM, Tolezano JE, Andrade Jr HF. Regional variation in toxoplasmosis seronegativity in the Săo Paulo metropolitan region. Rev Inst Med Trop Săo Paulo 1993;35:479-83.
  • 14. Institute of Laboratory Animal Resources. Commission on Life. Guide for the care and use of laboratory animals Washington (DC): National Academy Press; 1996.
  • 15. Little CL, DeLouvois SJ. Health risks associated with unpasteurized goats' and ewes' milk on retail sale in English and Wales. A PHLS Dairy Products Working Group Study. Epidemiol Infect 1999;122:403-8.
  • 16. Pettersen EK. Transmission of toxoplasmosis via milk from lactating mice. Acta Pathol Microbiol Immunol Scand 1984;92:175-6.
  • 17. Riemann HP, Meyer ME, Theis JH, Kelso G, Behymer DE. Toxoplasmosis in an infant fed unpasteurized goat milk. J Pediatr 1975;87:573-6.
  • 18. Sacks JJ, Roberto RR, Brooks NF. Toxoplasmosis infection associated with raw goat's milk. JAMA 1982;248:1728-32.
  • 19. Sanger VL, Chamberlain DM, Chamberlain KW, Cole CR, Farrell RL. Toxoplasmosis. V. Isolation of Toxoplasma from cattle. JAMA 1953;123:87-91.
  • 20. Skinner LJ, Timperley AC, Wightman D, Chatterton JM, Ho-Yen DO. Simultaneous diagnosis of toxoplasmosis in goats and goatowner's family. Scand J Infect Dis 1990;22:359-61.

  • *
    Supported by CNPq (Process n. 143364/1998-7), Fapesp (Process n. 98/1681-0, 98/13323-0, 96/5875-8 & 99/04926-6), Fundação do Desenvolvimento Administrativo (Fundap) and Laboratório de Investigação Médica do Hospital das Clínicas Faculdade Medicina USP-49.

Publication Dates

  • Publication in this collection
    07 Aug 2001
  • Date of issue
    Apr 2001

History

  • Received
    14 July 2000
  • Reviewed
    09 Oct 2000
  • Accepted
    01 Dec 2000
Faculdade de Saúde Pública da Universidade de São Paulo Avenida Dr. Arnaldo, 715, 01246-904 São Paulo SP Brazil, Tel./Fax: +55 11 3061-7985 - São Paulo - SP - Brazil
E-mail: revsp@usp.br