2011 Journal o f Food Protection, Vol. 76, No. 12, 2013, Pages 2011-2017 doi: 10.4315/0362-028X.JFP-13-164 Copyright © , International Association for Food Protection
Antimicrobial Susceptibility and Serovars of Salm onella Circulating in Commercial Poultry Carcasses and Poultry Products in Brazil RENATA G. COSTA,12 * MARCIA L. FESTIVO,1 MAIARA S. ARAUJO,1 ELIANE M. F. R EIS,1 NORMA S. LAZARO ,1 a n d DALIA P. RODRIGUES' 'National Reference Laboratory fo r Bacterial Enteroinfections, Oswaldo Cruz Institute, Oswaldo Cruz Foundation, Avenida Brasil, 4365-CEP 21040-900, Rio de Janeiro, Brazil; and tech n o lo g y Institute, Federal Rural University o f Rio de Janeiro, Brazil BR 465-CEP 23890-000, Km 7, Seroptdica, Rio de Janeiro, Brazil MS 13-164: Received 25 April 2013/Accepted 23 August 2013
ABSTRACT Antimicrobial resistance was evaluated for 12,582 strains of Salmonella isolated by public and private laboratories from commercial poultry carcasses and poultry products from different regions in Brazil between 2007 and 2011. These isolates were submitted unsolicited to the National Reference Laboratory for Bacterial Enteroinfections, Oswaldo Cruz Institute, Oswaldo Cruz Foundation, Rio de Janeiro, Brazil, for conclusive antigenic characterization. The prevalent Salmonella enterica serogroups were 0 :4 (B) (28.9%), 0 :9 (D l) (22.3%), 0:21 (L) (15.9%), and 0 :7 (C l) (14.3%); the most prevalent serovars were Enteritidis, Minnesota, Typhimurium, Schwarzengrund, and Mbandaka. The antimicrobial susceptibility profiles of 1,234 strains indicated that 54.5% were resistant to up to eight antimicrobial drugs. Multidrug resistance (three or more classes) was observed in 16.4% o f the strains, with 190 distinct patterns. The results showed increased resistance to ampicillin (12.4 to 18.9%), tetracycline ( ~ 15.2 to =18.9% ), and gentamicin (f7.0 to =9.6% ) during the study period. Decreased resistance to nitrofurans (61.9 to 9.2%), quinolones (44.4 to 15.5%), and folate inhibitors (11.7 to 7.2%) was observed. Resistance to third-generation cephalosporins was detected in 5.5% of the strains throughout the study period; resistance to fluoroquinolones has been observed in 0.3% of the strains since 2009. These findings highlight the importance o f surveillance along the food chain for detection of altered patterns among foodbome zoonotic bacteria that are important to public health.
The broiler industry is a major segment of the agro industrial complex in Brazil, which occupies third place in production and first place in exportation worldwide (31). Currently, the National Avian Health Program provides the sanitary and surveillance guidelines for epidemic prevention and sanitary conditions required for industry operations. These guidelines help to control and eradicate major diseases in poultry, including those caused by Salmonella serovars that are relevant to the health of humans and animals (20). The control of dissemination of diseases is difficult because of the emergence of new serovars and the reemergence of others. The increase in numbers of cases of Salmonella infections and significant changes in serovar prevalences over the last decades in Brazil have been associated with poultry and human infections (9, 17, 21). Salmonella serovars Enteritidis and Agona were the most commonly detected in poultry in the past 20 yr and were among the top five associated with human infections (10, 14). More recently, Salmonella serovars Minnesota and Typhimurium have become the most frequently detected serovars in chickens, whereas Salmonella serovar Enteritidis * Author for correspondence. Tel: (5521) 2562 1646; Fax: (5521) 2562 1651; E-mail: [email protected].
remains the most common cause of human infections (9,17, 21, 26, 34). In addition to increased Salmonella prevalences in the last decades, increased antimicrobial resistance has been observed, due to the indiscriminate use of drugs in humans, agriculture, and production of food of animal origin (1,4). Selective pressure leads microorganisms to develop and/ or acquire a variety of genetic elements that are exchanged between bacteria that live in the same habitat, e.g., enteric bacteria in the gastrointestinal tract of humans and animals, and that have characteristics that can persist for a long period. The genus Salmonella can play an important role as a recipient or donor of resistance genes, which is relevant in spreading these elements through the food chain (4,17). In response, the European Union has increasingly restricted the use of antimicrobials; and the United States has expanded its program to monitor antimicrobial resis tance and has established criteria for the rational use of antimicrobials (22). Brazil has established prudent practices for the use of additive antimicrobials; the use of tetracycline and sulfonamides in animal feed was forbidden through normative instructions in 1998 (18), and these were extended to include chloramphenicol and nitrofurantoin in 2003 (19). Salmonella resistance to antimicrobials varies in
2012
J. Food Pint., Vol. 76, No. 12
COSTA ET AL.
frequency and extent, based on the use of antibiotics in humans and animals and on ecological differences in the epidemiology of Salmonella infections (1, 3). Because Salmonella is a pathogen of economic importance and poses a risk to human health, the present study evaluated the frequencies and antimicrobial resistance profiles of Salmo nella serovars in commercial poultry carcasses and poultry products in Brazil. MATERIALS AND METHODS Salmonella strains. Between 2007 and 2011, 12,582 strains of Salmonella were evaluated. These strains were isolated in public and private laboratories from commercial poultry carcasses and poultry by-products from different regions of Brazil. The isolates were sent unsolicited to the National Reference Laboratory for Bacterial Enteroinfections (LRNEB), Oswaldo Cruz Foundation, Rio de Janeiro, for conclusive antigenic characterization in nutrient agar medium, or similarly, at room temperature. The institution of origin, identification number, and source of isolation of the strains were included in the LRNEB database. Antigenic characterization. After reisolation and confirma tion of biochemical profiles, the strains were serotyped through the identification of surface antigens with somatic antisera, and flagella antigens with flagellar antisera, following the Kauffmann-White scheme. Antigenic characterization was tested by the slide agglutination technique with poly and monovalent antisera, somatic and flagellar, prepared at the Laboratory of Enterobacteria at the Oswaldo Cruz Institute, Oswaldo Cruz Foundation. Identification of specific serovars was performed and presented according to the criteria reported by Grimont and Weill (11) and Guibourdenche et al. (12). Antimicrobial susceptibility test. Following routine proto col at the LRNEB, approximately 10% of the strains were randomly selected for antimicrobial susceptibility monitoring regardless of serovars. These strains were analyzed by the disc diffusion method (CLSI, annually updated), and results were interpreted according to the latest CLSI edition (5). Antimicrobial drugs (Oxoid, Basingstoke, UK) representative of the following antimicrobial classes were used: beta-lactams (ampicillin [AMP], cephalothin [CEP], cefoxitin [FOX], ceftria xone [CRO], ceftiofur [TIO], ceftazidime [CAZ], and imipenem [IPM]); phenicols (chloramphenicol [CHL]); tetracyclines (tetra cycline [TCY]); quinolones (nalidixic acid [NAL] and ciproflox acin [CIP]); aminoglycosides (gentamicin [GEN]); folate inhibitors (trimethoprim-sulfamethoxazole [SXT]); and nitrofurans (nitrofu rantoin [NIT]). The selection criteria included elective drugs for the control and therapy of bacterial enteric infections, those used as animal growth promoters, and those used in surveillance of antimicrobial resistance according to WFIO guidelines. The following standard strains were used to assure quality control and reliability in the results: Escherichia coli ATCC 25922, E. coli ATCC 35218, Pseudomonas aeruginosa ATCC 27853, Enterococcus faecalis ATCC 29212, and Staphylococcus aureus ATCC 25923.
RESULTS Identified Salmonella serovars. The serogroups with the highest frequencies were 0:4 (B) (28.9%), 0:9 (Dl) (22.3%), 0:21 (L) (15.9%), and 0:7 (Cl) (14.3%); 0:4 (B) was the most prevalent between 2008 and 2011 (Table 1).
TABLE 1. Distribution o f Salmonella serogroups ( “O” groups) isolated from poultry between 2007 and 2011 Percentage
“ O” groups 0 :4 (B) 0:7 (C,) 0:8 (C2-C3) 0 :9 (D,) 0:3,10 (E0 0:1,3,19 (E4) 0:11 (F) 0:13 (G) 0:16 (I) 0:17 (J) 0:18 (K) 0:21 (L) 0:28 (M) 0:30 (N) 0:35 (0) 0:40 (R) 0:47 (X) Rough form Total no. of strains
2007
2008
2009
2010
2011
17.7 8.9 4.7 50.3 2.7 1.3 0 0.6 0.2 0 1 11.4 0 0.1 0.2 0.5 0 0.4
28.4 18.4 4.6 24.2 5.5 1.0 0 3.7 0 0 1.2 9.9 0.1 0.1 0.3 0.3 0 2.2
39.2 11.1 6.1 10.4 4.9 2.6 0.1 1.3 0.1 0 0.6 22.3 0 0.1 0.1 0 0.1 1
32.3 15.5 6.9 11.3 5.5 2.4 0 1.4 0.2 0.2 0.8 22.9 0 0.1 0.2 0 0 0.3
26.7 17.7 10.7 15.4 3.4 6.9 0 4.2 0.2 0 0.4 13.1 0 0 0.1 0 0.1 1.1
2,190
777
3,683
3,764
2,168
During the study period, 61 distinct serovars were identi-
Tied; Enteritidis, Minnesota, Typhimurium, Schwarzengrund, and Mbandaka were the most prevalent serovars, accounting for 53.4% of all characterized strains. In particular, Salmonella serovar Enteritidis appeared in increased numbers in 2007 and 2008. Conversely, increased prevalences of other serotypes were observed during this period (Table 2). All serovars were identified in all geographical regions of Brazil during the study period, with the exception of Salmonella serovar Typhimurium, which was not detected in the northeastern region. Among the 20 most prevalent serovars, Salmonella serovar Heidelberg was identified exclusively in the southern region between 2007 and 2009, and in the southern and midwestem regions between 2010 and 2011. No differences were observed in the distribution of prevalences from serogroups and serovars between whole carcasses and by-products (data not shown). Antimicrobial resistance of Salmonella isolates. Antimicrobial susceptibility was determined for 1,234 strains; 32% were identified as susceptible, 13.5% as intermediary, and 54.5% as resistant to at least one anti microbial agent. Between 2007 and 2011, a progressive increase was observed in aminoglycoside resistance and a slightly less progressive increase in antifolate ( o
Antimicrobial Susceptibility and Serovars of Salm onella Circulating in Commercial Poultry Carcasses and Poultry Products in Brazil RENATA G. COSTA,12 * MARCIA L. FESTIVO,1 MAIARA S. ARAUJO,1 ELIANE M. F. R EIS,1 NORMA S. LAZARO ,1 a n d DALIA P. RODRIGUES' 'National Reference Laboratory fo r Bacterial Enteroinfections, Oswaldo Cruz Institute, Oswaldo Cruz Foundation, Avenida Brasil, 4365-CEP 21040-900, Rio de Janeiro, Brazil; and tech n o lo g y Institute, Federal Rural University o f Rio de Janeiro, Brazil BR 465-CEP 23890-000, Km 7, Seroptdica, Rio de Janeiro, Brazil MS 13-164: Received 25 April 2013/Accepted 23 August 2013
ABSTRACT Antimicrobial resistance was evaluated for 12,582 strains of Salmonella isolated by public and private laboratories from commercial poultry carcasses and poultry products from different regions in Brazil between 2007 and 2011. These isolates were submitted unsolicited to the National Reference Laboratory for Bacterial Enteroinfections, Oswaldo Cruz Institute, Oswaldo Cruz Foundation, Rio de Janeiro, Brazil, for conclusive antigenic characterization. The prevalent Salmonella enterica serogroups were 0 :4 (B) (28.9%), 0 :9 (D l) (22.3%), 0:21 (L) (15.9%), and 0 :7 (C l) (14.3%); the most prevalent serovars were Enteritidis, Minnesota, Typhimurium, Schwarzengrund, and Mbandaka. The antimicrobial susceptibility profiles of 1,234 strains indicated that 54.5% were resistant to up to eight antimicrobial drugs. Multidrug resistance (three or more classes) was observed in 16.4% o f the strains, with 190 distinct patterns. The results showed increased resistance to ampicillin (12.4 to 18.9%), tetracycline ( ~ 15.2 to =18.9% ), and gentamicin (f7.0 to =9.6% ) during the study period. Decreased resistance to nitrofurans (61.9 to 9.2%), quinolones (44.4 to 15.5%), and folate inhibitors (11.7 to 7.2%) was observed. Resistance to third-generation cephalosporins was detected in 5.5% of the strains throughout the study period; resistance to fluoroquinolones has been observed in 0.3% of the strains since 2009. These findings highlight the importance o f surveillance along the food chain for detection of altered patterns among foodbome zoonotic bacteria that are important to public health.
The broiler industry is a major segment of the agro industrial complex in Brazil, which occupies third place in production and first place in exportation worldwide (31). Currently, the National Avian Health Program provides the sanitary and surveillance guidelines for epidemic prevention and sanitary conditions required for industry operations. These guidelines help to control and eradicate major diseases in poultry, including those caused by Salmonella serovars that are relevant to the health of humans and animals (20). The control of dissemination of diseases is difficult because of the emergence of new serovars and the reemergence of others. The increase in numbers of cases of Salmonella infections and significant changes in serovar prevalences over the last decades in Brazil have been associated with poultry and human infections (9, 17, 21). Salmonella serovars Enteritidis and Agona were the most commonly detected in poultry in the past 20 yr and were among the top five associated with human infections (10, 14). More recently, Salmonella serovars Minnesota and Typhimurium have become the most frequently detected serovars in chickens, whereas Salmonella serovar Enteritidis * Author for correspondence. Tel: (5521) 2562 1646; Fax: (5521) 2562 1651; E-mail: [email protected].
remains the most common cause of human infections (9,17, 21, 26, 34). In addition to increased Salmonella prevalences in the last decades, increased antimicrobial resistance has been observed, due to the indiscriminate use of drugs in humans, agriculture, and production of food of animal origin (1,4). Selective pressure leads microorganisms to develop and/ or acquire a variety of genetic elements that are exchanged between bacteria that live in the same habitat, e.g., enteric bacteria in the gastrointestinal tract of humans and animals, and that have characteristics that can persist for a long period. The genus Salmonella can play an important role as a recipient or donor of resistance genes, which is relevant in spreading these elements through the food chain (4,17). In response, the European Union has increasingly restricted the use of antimicrobials; and the United States has expanded its program to monitor antimicrobial resis tance and has established criteria for the rational use of antimicrobials (22). Brazil has established prudent practices for the use of additive antimicrobials; the use of tetracycline and sulfonamides in animal feed was forbidden through normative instructions in 1998 (18), and these were extended to include chloramphenicol and nitrofurantoin in 2003 (19). Salmonella resistance to antimicrobials varies in
2012
J. Food Pint., Vol. 76, No. 12
COSTA ET AL.
frequency and extent, based on the use of antibiotics in humans and animals and on ecological differences in the epidemiology of Salmonella infections (1, 3). Because Salmonella is a pathogen of economic importance and poses a risk to human health, the present study evaluated the frequencies and antimicrobial resistance profiles of Salmo nella serovars in commercial poultry carcasses and poultry products in Brazil. MATERIALS AND METHODS Salmonella strains. Between 2007 and 2011, 12,582 strains of Salmonella were evaluated. These strains were isolated in public and private laboratories from commercial poultry carcasses and poultry by-products from different regions of Brazil. The isolates were sent unsolicited to the National Reference Laboratory for Bacterial Enteroinfections (LRNEB), Oswaldo Cruz Foundation, Rio de Janeiro, for conclusive antigenic characterization in nutrient agar medium, or similarly, at room temperature. The institution of origin, identification number, and source of isolation of the strains were included in the LRNEB database. Antigenic characterization. After reisolation and confirma tion of biochemical profiles, the strains were serotyped through the identification of surface antigens with somatic antisera, and flagella antigens with flagellar antisera, following the Kauffmann-White scheme. Antigenic characterization was tested by the slide agglutination technique with poly and monovalent antisera, somatic and flagellar, prepared at the Laboratory of Enterobacteria at the Oswaldo Cruz Institute, Oswaldo Cruz Foundation. Identification of specific serovars was performed and presented according to the criteria reported by Grimont and Weill (11) and Guibourdenche et al. (12). Antimicrobial susceptibility test. Following routine proto col at the LRNEB, approximately 10% of the strains were randomly selected for antimicrobial susceptibility monitoring regardless of serovars. These strains were analyzed by the disc diffusion method (CLSI, annually updated), and results were interpreted according to the latest CLSI edition (5). Antimicrobial drugs (Oxoid, Basingstoke, UK) representative of the following antimicrobial classes were used: beta-lactams (ampicillin [AMP], cephalothin [CEP], cefoxitin [FOX], ceftria xone [CRO], ceftiofur [TIO], ceftazidime [CAZ], and imipenem [IPM]); phenicols (chloramphenicol [CHL]); tetracyclines (tetra cycline [TCY]); quinolones (nalidixic acid [NAL] and ciproflox acin [CIP]); aminoglycosides (gentamicin [GEN]); folate inhibitors (trimethoprim-sulfamethoxazole [SXT]); and nitrofurans (nitrofu rantoin [NIT]). The selection criteria included elective drugs for the control and therapy of bacterial enteric infections, those used as animal growth promoters, and those used in surveillance of antimicrobial resistance according to WFIO guidelines. The following standard strains were used to assure quality control and reliability in the results: Escherichia coli ATCC 25922, E. coli ATCC 35218, Pseudomonas aeruginosa ATCC 27853, Enterococcus faecalis ATCC 29212, and Staphylococcus aureus ATCC 25923.
RESULTS Identified Salmonella serovars. The serogroups with the highest frequencies were 0:4 (B) (28.9%), 0:9 (Dl) (22.3%), 0:21 (L) (15.9%), and 0:7 (Cl) (14.3%); 0:4 (B) was the most prevalent between 2008 and 2011 (Table 1).
TABLE 1. Distribution o f Salmonella serogroups ( “O” groups) isolated from poultry between 2007 and 2011 Percentage
“ O” groups 0 :4 (B) 0:7 (C,) 0:8 (C2-C3) 0 :9 (D,) 0:3,10 (E0 0:1,3,19 (E4) 0:11 (F) 0:13 (G) 0:16 (I) 0:17 (J) 0:18 (K) 0:21 (L) 0:28 (M) 0:30 (N) 0:35 (0) 0:40 (R) 0:47 (X) Rough form Total no. of strains
2007
2008
2009
2010
2011
17.7 8.9 4.7 50.3 2.7 1.3 0 0.6 0.2 0 1 11.4 0 0.1 0.2 0.5 0 0.4
28.4 18.4 4.6 24.2 5.5 1.0 0 3.7 0 0 1.2 9.9 0.1 0.1 0.3 0.3 0 2.2
39.2 11.1 6.1 10.4 4.9 2.6 0.1 1.3 0.1 0 0.6 22.3 0 0.1 0.1 0 0.1 1
32.3 15.5 6.9 11.3 5.5 2.4 0 1.4 0.2 0.2 0.8 22.9 0 0.1 0.2 0 0 0.3
26.7 17.7 10.7 15.4 3.4 6.9 0 4.2 0.2 0 0.4 13.1 0 0 0.1 0 0.1 1.1
2,190
777
3,683
3,764
2,168
During the study period, 61 distinct serovars were identi-
Tied; Enteritidis, Minnesota, Typhimurium, Schwarzengrund, and Mbandaka were the most prevalent serovars, accounting for 53.4% of all characterized strains. In particular, Salmonella serovar Enteritidis appeared in increased numbers in 2007 and 2008. Conversely, increased prevalences of other serotypes were observed during this period (Table 2). All serovars were identified in all geographical regions of Brazil during the study period, with the exception of Salmonella serovar Typhimurium, which was not detected in the northeastern region. Among the 20 most prevalent serovars, Salmonella serovar Heidelberg was identified exclusively in the southern region between 2007 and 2009, and in the southern and midwestem regions between 2010 and 2011. No differences were observed in the distribution of prevalences from serogroups and serovars between whole carcasses and by-products (data not shown). Antimicrobial resistance of Salmonella isolates. Antimicrobial susceptibility was determined for 1,234 strains; 32% were identified as susceptible, 13.5% as intermediary, and 54.5% as resistant to at least one anti microbial agent. Between 2007 and 2011, a progressive increase was observed in aminoglycoside resistance and a slightly less progressive increase in antifolate ( o
