Abstract
The globalization of food supply impacts patterns of foodborne disease outbreaks worldwide, and consumers are having increased concern about microbiological food safety. In this sense, the assessment of epidemiological data of foodborne diseases in different countries has not only local impact, but it can also be of general interest, especially in the case of major global producers and exporters of several agricultural food products, such as Brazil. In this review, the most common agents of foodborne illnesses registered in Brazil will be presented, compiled mainly from official databases made available to the public. In addition, some representative examples of studies on foodborne bacterial pathogens commonly found in Brazilian foods are provided.
Introduction
B
The agencies responsible for food control in Brazil are the Ministry of Agriculture, Livestock and Food Supply and the Ministry of Health, through the National Health Surveillance Agency (ANVISA). In recent decades, these two agencies made a major stride in establishing food standards and technical regulations in order to develop foreign trade and guarantee food safety (Salay et al., 2001; WHO, 2012; Brazil, 2001b).
Technical departments of the Ministry of Agriculture, Livestock and Food Supply are in charge of controlling the quality and safety of agricultural products at the production level, and created several monitoring programs, such as (1) AGROFIT—safe utilization of pesticides for plant protection purposes; (2) SIF—Federal Inspection System; (3) SISBOV—origin certification for beef and buffalo meat; (4) SISLEG—Federal agricultural legislative system; (5) PNCRC—control of residues and contaminants in animal products; (6) PNRP—pathogen reduction; (7) PNSA—poultry sanitation; and (8) PNCEBT—control and eradication of animal brucellosis and tuberculosis (Brazil 2009a).
ANVISA coordinates, supervises, and controls activities regarding registration, information, inspection, risk control, and rulemaking to assure health surveillance over food, beverages, water, ingredients, packages, technologies, contamination limits, and veterinary residues (Brazil, 2009b).
Following a global trend, Brazilian consumers have expressed increased concerns about risks derived not only from foodborne pathogens, but also from artificial chemical preservatives used to control them (Schuenzel and Harrison, 2002; Rodríguez et al., 2003; Parada et al., 2007; Castellano et al., 2008). Brazilians have a relatively healthy diet, and eating outside the home is becoming more common, with an increasing demand for convenient ready-to-eat (RTE) and minimally processed foods (Canada, 2010; Sant'Ana et al., 2012).
The most relevant agents causing foodborne illnesses in Brazil are highlighted in this review, based on data available from public official databases (Brazil, 2008a,b; Brazil, 2011). Also, examples of studies on occurrence of pathogenic bacteria in important food commodities consumed in Brazil are provided. With this review, we intend to contribute to improving food safety at local and global levels.
Epidemiology of Main Foodborne Diseases in Brazil
Data regarding foodborne diseases in Brazil are scarce and most reported foodborne outbreaks occurred in the South and Southeast regions, the most populated areas in the country (Brazil, 2011). The occurrence of outbreaks in other regions is not well known. Only a few Brazilian Federative States, mainly from the South and Southeast regions, have a structured foodborne diseases surveillance system and report regularly to the local or national health authorities (van Amson, 2006; Brazil, 2011).
From 2000 to 2011, 8663 foodborne disease outbreaks were registered, affecting 163,425 persons and causing 112 deaths (Brazil, 2011). Official data on epidemiology of foodborne diseases suggest that since 2008, there has been a decreasing trend in the number of outbreaks, despite the constant increase of the population (Brazil, 2011; IBGE, 2012).
Foods most commonly implicated in outbreaks were mixed meals, eggs and egg-related products, desserts, water, raw and processed beef, milk and dairy products, and chicken and pork (Brazil, 2011). Most outbreaks were caused by bacteria, mainly Salmonella spp., followed by Staphylococcus spp., Bacillus cereus, Clostridium perfringens, Shigella spp., and Clostridium botulinum (Araujo et al., 2002; Brazil, 2011). However, for approximately 50% of the registered outbreaks, the etiological agent of the disease could not be determined (Brazil, 2008a; Brazil, 2011).
In the present review, a list of the most relevant foodborne pathogens was compiled based on (1) data published in 2008 and 2011 by the Department of Health, which is considered the official database of foodborne diseases in Brazil (Brazil, 2008a,b; Brazil, 2011), and on (2) the scientific literature available in PubMed and Web of Science. Prevalence or outbreak investigations included in the review were conducted in Brazil and published mainly from 2000 to 2012 (August) in English or Portuguese. The pathogens included in the survey are those listed by the Brazilian official database as commonly implicated in foodborne outbreaks in Brazil. The initial search in Web of Science and PubMed retrieved 52 possibly eligible studies using the following search terms in Title/Abstract: Food AND Brazil AND Staphylococcus OR Salmonella OR Bacillus cereus OR Clostridium perfringens OR botulism OR Shigella. Abstracts of the 52 eligible studies were analyzed and 37 of them were excluded because (1) the full article was not available, (2) results were inconclusive, (3) data were related to molecular or serological characterization of strains obtained in previous studies, and (4) the same paper was detected multiple times in different databases. Therefore, 15 studies were considered eligible and are compiled in Table 1.
Results were compiled on August 2012. This is not intended to be a systematic review.
A, Occurrence studies; and B, outbreak investigation.
Brazilian Federative States: Amazonas (AM), Bahia (BA), Ceará (CE), Goiás (GO), São Paulo (SP), Pernambuco (PE), Rio Grande do Sul (RS), Minas Gerais (MG), Rio de Janeiro (RJ), Maranhão (MA), Mato Grosso (MT), Mato Grosso do Sul (MS), Pará (PA), Paraná (PR) and Rio Grande do Norte (RN).
n.i., not informed; CFU, colony-forming units; MPN, most probable number; SEA, Staphylococcal enterotoxins A; SEB, Staphylococcal enterotoxins B; SEC, Staphylococcal enterotoxins C; SED, Staphylococcal enterotoxins D.
A literature survey on data regarding Listeria monocytogenes was also done, but in Brazil, there have been no reported cases of listeriosis linked to the consumption of contaminated foods (Souza et al., 2008), and in the indexed literature, there are only two reports of outbreaks of listeriosis in Brazil. The first one was reported by Landgraf et al. in 1999 and involved five neonates in São Paulo city, with meningitis due to Listeria, while the second outbreak was recently reported by Martins et al. (2010) to occur among elderly hospitalized patients, in the city of Rio de Janeiro. For both outbreaks, no source of infection was determined, but Martins et al. (2010) hypothesized that the bacterium was likely acquired via contaminated foods prepared in the hospital kitchen.
The registration of listeriosis cases is not mandatory in the National Notifiable Diseases Surveillance System (Brito et al., 2008), but a few compiled data on occurrence and characterization are found, such as the paper published by Hofer et al. (2000) on the species and serovars of Listeria genus from different sources (1971–1997), including foods.
Following a global trend, in Brazil Salmonella Enteritidis (SE) has been associated with human foodborne infections caused by the ingestion of contaminated foods of animal origin, mainly undercooked poultry meat and eggs (Vaz et al., 2010). Geimba et al. (2004) affirmed that although Brazil is the second-largest producer of poultry meat in the world, few studies have been conducted in Salmonella isolates from this country.
Nunes et al. (2003) demonstrated that S. enteritidis phage type (PT) 4 was the type most commonly found among the SE Brazilian isolates from 1995 to 1997 (healthy and diseased chicken, outbreaks of human gastroenteritis related to the consumption of egg products, poultry meat, pipped embryos of broiler chicks, meat meal and the rearing environment, and diverse food products [i.e., cheese, mayonnaise, cake, and bacon]). Those authors concluded that the results obtained were in accordance with the worldwide trends in distribution patterns for different SE phage types. However, as stated by Kottwitz et al. (2012), it is important to note that since 2002, a decline of outbreaks caused by SE PT4 has been observed, probably as a result of the control measures adopted in the poultry production chain, which may have favored the introduction of other phage types such as the PT9 (Kottwitz et al., 2012).
Mürmann et al. (2011) pointed out that although chicken and eggs are the foods commonly associated with salmonellosis outbreaks, infections due to pork consumption also have been reported, mainly in regions where it is an important part of the human diet, as in the southern region of Brazil. Salmonella enterica was detected in 24.4% of the pork sausages samples analyzed in the city of Porto Alegre, southern Brazil (Mürmann et al., 2009) and thus, undercooked pork sausage may also represent a significant health risk for consumers.
Staphylococcus aureus is the second leading cause of bacterial foodborne disease in Brazil, usually linked to inadequate food handling (Rapini et al., 2005). Coagulase-positive staphylococci produce heat-stable enterotoxins, with emetic activity. It is the main cause of food poisoning that occurs after ingestion of foods contaminated with S. aureus by improper handling and subsequent storage at elevated temperatures (Soares et al., 2012). In this sense, recently Soares et al. (2012) interviewed 166 food handlers via a questionnaire to determine levels of knowledge, attitudes, and practices in food safety. Also, the presence of coagulase-positive staphylococci species was analyzed on the hands of food handlers in municipal schools of Bahia, northeast Brazil. The authors revealed that despite a high prevalence of training and positive attitudes about food safety, 53.3% of the samples taken from hands presented coagulase-positive staphylococci.
Colombari et al. (2007) reported details of an outbreak caused by S. aureus in 1998, which affected approximately 180 people attending a benefit luncheon at a public school in São Paulo State. High counts of S. aureus were observed in a vegetable salad prepared with mayonnaise sauce, broiled chicken, and pasta with tomato sauce. Isolates from oropharyngeal secretions of food handlers and from the salad presented the same phage-type profile and resistance to antibiotics, belonged to the same RAPD cluster, and produced enterotoxin A, indicating improper handling during manufacture.
According to the official database of the National Secretary of Health Surveillance (SVE) of the Ministry of Health, Bacillus cereus is also an important etiological agent of bacterial foodborne diseases in Brazil. However, little is known about the prevalence and characteristics of this pathogen in Brazilian foods (Chaves et al., 2011). Aragon-Alegro et al. (2008) reported that counts of B. cereus in ready-to-eat meals, spices, dairy products, starches, flours, and other foods varied from 102 colony-forming units (CFU)/g or /mL to 104 CFU/g or mL of food. All tested isolates were positive for at least one enterotoxin gene, reinforcing the risk to the consumer. Similar results were obtained by Chaves et al. (2011), who evaluated the genetic diversity, antimicrobial resistance, and toxigenic profiles of B. cereus strains isolated from various food items from the southwest region of Brazil. A high prevalence of toxin-encoding genes in the isolates was observed. Another study indicated that genes encoding enterotoxins were also widely spread among B. cereus and B. thuringiensis strains isolated from ground roasted coffee sold commercially in Brazil (Chaves et al., 2012).
According to the World Health Organization (WHO, 2005), Shigella is considered one of the most important cause of bloody diarrhea worldwide, linked to the consumption of contaminated food or drinking water. However, as previously discussed, in Brazil there are few reports on foodborne shigellosis, probably because there is no current regulation that requires the investigation of Shigella in water or foods (De Paula et al., 2010).
De Paula et al. (2010) investigated the antimicrobial resistance and molecular patterns of Shigella strains responsible for foodborne outbreaks that occurred in Rio Grande do Sul state from 2003 to 2007. Outbreaks were investigated by the state government Surveillance Service, and Shigella strains (n=152) were isolated from foods and fecal samples of patients: 71.1% of the isolates were S. flexneri, 21.5% S. sonnei, and 0.7% S. dysenteriae. However, no specification about the kind of foods analyzed was provided.
Also, little is known about Vibrio spp. in Brazilian foods. Only a few reports on foodborne diseases caused by Vibrio spp. were reported, despite the high temperature of the seawater and the widespread occurrence of Vibrio spp. in the marine environment (Leal et al., 2008). Araujo et al. (2007) reported the first case of severe infection caused by V. vulnificus, which occurred in Brazil in 1994. The patient was an 86-year-old man, who ingested seafood (mussels and octopus) at a seashore in São Paulo state, and he presented with vomiting, diarrhea, decreased intake of liquids, poor diuresis, and fever. Leal et al. (2008) described the investigation of a foodborne outbreak caused by V. parahaemolyticus that occurred in the northeast region of Brazil (Alagoas, Ceará and Pernambuco states). V. parahaemolyticus O3:K6 and O3:KUT were isolated from the stools of patients, and isolates were positive for virulence factors thermolabile haemolysin and thermostable direct haemolysin. No isolates were obtained from foods, but clinical and epidemiological investigations indicated that this was a foodborne outbreak. Costa-Sobrinho et al. (2010) conducted a study for estimation of the risks associated with consumption of oysters (Crassostrea brasiliana) from one important producing area located in São Paulo state coastal region. V. parahaemolyticus was detected in 99.2% (122/123) of the samples, with population densities varying between 0.78 and 5.04 log MPN/g, but only one isolate was Kanagawa positive and presented the tdh gene.
With regard to Clostridium genus, despite the considerable number of outbreaks caused by Clostridium perfringens, there are few reports in indexed literature from Brazil on these bacteria related to food safety (Schocken Iturrino et al., 1986; Tortora and Zebral, 1988).
Botulism is a foodborne disease of mandatory notification in Brazil. The Reference Center for Botulism (CR BOT-CVE/SP), responsible for recording botulism cases and outbreaks that occurred in São Paulo state, confirmed 20 episodes of foodborne botulism between 1997 and 2010, with 5 deaths (CVE, 2010). Eduardo et al. (2007) reported on cases of botulism type A and B in São Paulo city due to the consumption of commercial pie made of chicken, hearts of palm, and peas. Moreover, data collected from all Brazilian regions (1999–2008) made available by the Federal Health Department (Brazil, 2008a) revealed 37 confirmed foodborne outbreaks of botulism associated with meats and home-canned vegetables. This year, a botulism outbreak (seven cases) was reported by the government of the state of Santa Catarina (south region), and it was caused by eating a commercial brand of mortadella, a kind of deli meat (Santa Catarina, 2012).
Prevalence of Foodborne Pathogens in Selected Brazilian Food Commodities
Retail meat
According the Association of Brazilian Beef Exporters (ABIEC), Brazil's cattle herd is estimated at 209 million. Per year, 5 billion poultry, 40 million bovine, and 30 million swine are slaughtered in the country. The productivity rates of Brazilian beef are constantly increasing and production costs are among the lowest in the world, conferring great competitiveness (ABIEC, 2012).
Bovine meat is a suitable environment for growth of many microorganisms, including pathogens. According to general literature data, Salmonella spp., Escherichia coli O157:H7, and L. monocytogenes are the main pathogens found in meats, but Campylobacter spp., S. aureus, C. botulinum, and C. perfringens may also be important (ICMSF, 2005; Aymerich et al., 2008; Sofos, 2008).
In Brazil, studies conducted with raw meat at retail level (ground meat and/or meat cuts) indicated that prevalence of Salmonella spp. is around 10%–20% (Almeida et al., 2002; Sigarini et al., 2006). Aragon-Alegro et al. (2005) and Barros et al. (2007) observed that L. monocytogenes was present in 48.3% and 17.6% of the tested meat products, respectively, depending on the type of meat product.
Studies on E. coli O157:H7 and other Shiga toxin–producing E. coli (STEC) in raw meat and meat products in Brazil indicated that their prevalence is low. Rigobelo et al. (2008) observed that 1.4% of beef carcasses tested from an abattoir in São Paulo State were positive for STEC. Moreover, no E. coli O157 was detected in hamburgers produced by eight manufacturers in Brazilian southern states (Silveira et al., 1999).
It is believed that cattle in Brazil are less frequently infected by STEC than in other countries, such as the United States or Mexico (Arthur et al., 2002; Varela-Hernandez et al., 2007), but more recent data indicate that this may not be true for all STEC serotypes (Oliveira et al., 2008).
According to surveys conducted by Irino et al. (2005) with fecal samples of bovine cattle in São Paulo state, the isolation rate of STEC may range from 3.8% to 84.6% depending on the farm analyzed. Similarly, Oliveira et al. (2008) reported that the occurrence of STEC in healthy beef and dairy cattle and goats reared in Minas Gerais State ranged from 17.5% to 57.5%. Despite detection of STEC in Brazilian cattle, the occurrence of human illnesses caused by these bacteria is low. Only sporadic cases of gastrointestinal diseases and hemolytic uremic syndrome caused by non-O157 STEC strains have been reported (Guth et al., 2002; Irino et al., 2002; De Toni et al., 2009).
Poultry
Poultry meat is an important constituent of the daily diet of Brazilians. Its cost is low if compared to other animal protein sources, and the consumption has increased significantly in recent years (Gonzalez, 2000; Capita et al., 2007; Chiarini et al., 2009). In addition, poultry meat plays an important role in the Brazilian agribusiness and international trade.
The main microbiological hazards in poultry meat are Salmonella and Campylobacter (ICMSF, 2005; Arsenault et al., 2007). Many studies, carried out in different Brazilian regions, indicate that the overall prevalence of Salmonella in chicken carcasses varies between 5.4% and 46.6% (Fuzihara et al., 2000; Matheus et al., 2003; Reiter et al., 2007; Asensi et al., 2009; Duarte et al., 2009). It is common sense that there are no control measures efficient to completely eliminate Salmonella spp. from poultry carcasses for commercial use (Ristori et al., 2008). Hence, labeling of raw poultry with clear guidance for proper use, preparation and storage of the product is mandatory (Brazil, 2001a). In another food safety action, ANVISA launched in 2006 a national program to monitor the prevalence and antimicrobial resistance of enterococci and Salmonella in frozen poultry carcasses (PREBAF), which included verification of compliance with the labeling legislation (Brazil, 2006). Results indicated that 88% of the sampled carcasses were properly labelled and 3% were positive for Salmonella spp. Among these, S. Enteritidis was the most common serotype, confirmed in 50% of the Salmonella positive samples. Enterococcus spp. was present in 99% of the samples, and E. faecalis was the most common species (Brazil, 2008b).
Cheeses
Brazil is also a great producer of milk and several types of cheeses, largely consumed in the country. According to the United States Department of Agriculture (USDA, 2012), Brazil is the third largest cheese producer in the world, after the European Union and the United States. The most popular cheese in Brazil is Minas cheese, a fresh nonripened cheese obtained by enzymatic coagulation of pasteurized bovine milk with rennet or other coagulating enzymes. This cheese has high water activity, pH 6.6–6.8, low salt content (1.4%–1.6%), and no preservatives are added, so that the shelf-life scarcely exceeds 20 days under refrigeration (Gonzalez et al., 2000; Moraes et al., 2009; Alves et al., 2011). Minas cheese is the most consumed cheese in the country (39%), followed by mozzarella (18%), prato (14%), and other types of cheeses (ricotta, provolone, and camembert) (29%) (Planzer et al., 2009). Production of cheeses with raw milk for commercial purposes is prohibited in Brazil, unless they are submitted to a maturation period greater than 60 days (Brazil, 1997; Moraes et al., 2009). However, homemade cheeses and informal sales of noninspected dairy products are common in some regions due to cultural and economic factors. For instance, Minas Gerais state permits commercialization of artisanal Minas cheese produced with raw milk, provided that good manufacturing practices are followed (Minas Gerais, 2002).
Moraes et al. (2009) were not able to detect L. monocytogenes and Salmonella spp. in Minas cheese samples collected from several noninspected commercial establishments, whereas coagulase-positive staphylococci were detected in 30.9% of the samples. Silva et al. (2003) conducted a study in Minas Frescal cheese processing facilities, monitoring the critical control points in the processing and the industrial environment, and detected Listeria spp. in 6.0% of the samples (4% L. innocua, 1% L. monocytogenes, and 1% L. grayi). A study by Brito et al. (2008) revealed that 11% of Minas cheese samples were positive for L. monocytogenes. Barancelli et al. (2011) studied the incidence of L. monocytogenes in three cheese manufacturing plants from the northeastern region of São Paulo (2008–2009) and the pathogen was found in samples from two plants, at percentages of 13.3% (n=128) and 9.6% (n=114), with the prevalence of isolates of the serotype 4b.
Sangaletti et al. (2009) reported that coagulase-positive Staphylococcus and Salmonella were not detected in samples of Minas cheese during 30 days of storage at 4°C. In another study carried out by Gonzalez et al. (2000), E. coli strains belonging to EPEC serogroups O127 and O128 were present in 11.3% of the soft white cheese samples.
Ready-to-eat and minimally processed foods
The global consumer trend is consumption of meals that are healthy and easy to prepare, with less time dedicated to cooking at home (Canada, 2010). Following this trend, Brazilians are increasing the demand for minimally processed products and RTE meals. It is widely known that such products can be vehicles for the transmission of bacterial, parasitic, and viral pathogens capable of causing human illness. It can be even speculated that modified atmosphere used in RTE foods may favor the survival of pathogens requiring reduced oxygen concentration for growth (Beuchat, 1996; Beuchat, 2002; Gomes and De Martinis, 2004). Fröder et al. (2007) and Sant'Ana et al. (2011) studied minimally processed vegetables sold in São Paulo city and observed that Salmonella was present in 3% and 0.4% of the samples, respectively. The prevalence of Listeria species in minimally processed vegetables commercialized in the city of Ribeirão Preto, São Paulo state was 2.5% for L. innocua and 1.2% for L. monocytogenes (Oliveira et al., 2010). Sant'Ana et al. (2012) observed that 3.1% of the samples of RTE vegetables marketed in São Paulo were positive for the pathogen, with counts between 1.0×101 and 2.6×102 CFU/g.
Conclusion
Great progress has been achieved by Brazil in the last few decades in programs to guarantee food safety from farm to fork, in a concerted effort of agricultural and health agencies. However, as in many other countries, there is still a subnotification of foodborne bacterial diseases and lack of more complete official epidemiological data. Also, quantitative data on microbiological hazards in foods are needed if risk assessments programs are to be implemented.
Footnotes
Disclosure Statement
No competing financial interests exist.
