In Memoriam Dr. Lucie Dutil, friend and colleague at the Canadian Integrated Program for Antimicrobial Resistance Surveillance. Table of Contents   I. Introduction ................................................................................................................................................ 1 A. Executive Report ..................................................................................................................................... 1 B. NARMS Program ..................................................................................................................................... 1 C. NARMS Components .............................................................................................................................. 2 D. Links to Additional Information ................................................................................................................. 2 II. Summary of the 2010 NARMS Executive Report .................................................................................. 4 III. Methods ................................................................................................................................................... 10 A. Sampling Methodology .......................................................................................................................... 10 B. Antimicrobial Susceptibility Testing Methods ......................................................................................... 11 C. Breakpoints ............................................................................................................................................ 12 D. Reporting Methods ................................................................................................................................ 15 IV. Non-Typhoidal Salmonella Data ........................................................................................................... 17 A. Non-Typhoidal Salmonella Isolates Tested ........................................................................................... 18 B. Isolation of Non-Typhoidal Salmonella from Retail Meats ...................................................................... 19 C. Non-Typhoidal Salmonella Serotypes .................................................................................................... 13 D. Antimicrobial Susceptibility among all Non-Typhoidal Salmonella .......................................................... 23 E. Antimicrobial Susceptibility among Salmonella serotype Enteritidis ...................................................... 48 F. Antimicrobial Susceptibility among Salmonella serotype Typhimurium ................................................. 55 G. Antimicrobial Susceptibility among Salmonella serotype Newport ......................................................... 62 H. Antimicrobial Susceptibility among Salmonella serotype I 4,[5],12:i:- ..................................................... 69 I. Antimicrobial Susceptibility among Salmonella serotype Heidelberg ..................................................... 76 V. Campylobacter Data ............................................................................................................................... 83 A. Campylobacter jejuni and Campylobacter Isolates Tested ................................................................... 83 B. Isolation of Campylobacter from Retail Poultry ..................................................................................... 84 C. Campylobacter Species ........................................................................................................................ 85 D. Antimicrobial Susceptibility among Campylobacter jejuni ...................................................................... 86 E. Antimicrobial Susceptibility among Campylobacter coli ......................................................................... 91 F. Multidrug Resistance among Campylobacter Species ........................................................................... 96 VI. Escherichia coli Data ............................................................................................................................. 98 A. E. coli Isolates Tested ............................................................................................................................ 98 B. Isolation of E. coli from Retail Meats ...................................................................................................... 99 C. Antimicrobial Susceptibility among E. coli ............................................................................................ 100 Appendices ................................................................................................................................................ 107 Appendix A. Concentration Ranges Used for Susceptibility Testing ........................................................... 107 Appendix B. Antimicrobial Agents and Antimicrobial Susceptibility Testing Methods .................................. 109 Appendix C. Impact of New Breakpoints for Ciprofloxacin for Salmonella ................................................... 111 I. Introduction A. Executive Report This report summarizes, in an integrated format, the National Antimicrobial Resistance Monitoring System data on Salmonella (non-typhoidal) and Campylobacter recovered in 2010 from human clinical cases, retail meats, and food animals at federally inspected slaughter and processing plants. In addition, the report includes susceptibility data for Escherichia coli recovered from retail meats and chicken carcasses in 2010. Summary data from prior years are also included. Suggested Citation: FDA. National Antimicrobial Resistance Monitoring System – Enteric Bacteria (NARMS): 2010 Executive Report. Rockville, MD: U.S. Department of Health and Human Services, Food and Drug Administration, 2012. B. NARMS Program The National Antimicrobial Resistance Monitoring System – Enteric Bacteria (NARMS) is a national public health surveillance system in the United States, which tracks changes in the susceptibility of certain enteric bacteria to antimicrobial agents of human and veterinary medical importance. The NARMS program was established in 1996 as a collaboration among three federal agencies: the U.S. Food and Drug Administration (FDA), the Centers for Disease Control and Prevention (CDC), and the U.S. Department of Agriculture (USDA). NARMS monitors antimicrobial susceptibility among enteric bacteria from humans, retail meats, and food animals. Monitoring is conducted for several enteric pathogens, including Salmonella and Campylobacter. Generic Escherichia coli (E. coli) and Enterococcus are also tested due to their ubiquitous presence in animals, foods, and humans and their potential to serve as reservoirs of antimicrobial resistance genes for bacterial pathogens. Shigella, typhoidal Salmonella and Vibrio are tested from humans only. In addition to monitoring antimicrobial susceptibility, NARMS conducts epidemiologic and microbiologic research studies. Some studies examine risk factors and clinical outcomes of infections with specific bacterial serotypes or subsets of bacteria that exhibit particular resistance patterns. Other studies focus on understanding the genetic mechanisms of antimicrobial resistance in enteric bacteria and the mechanisms that permit the transfer of resistance between bacteria, on improving methods for isolation and typing, and on developing new methods for antimicrobial susceptibility testing. Additionally, NARMS examines Salmonella and Campylobacter strains for genetic relatedness using pulsed-field gel electrophoresis (PFGE). PFGE patterns are entered into CDC’s PulseNet database or USDA’s VetNet database. PulseNet and VetNet are national molecular subtyping networks for foodborne and zoonotic disease surveillance. The following are the primary objectives of NARMS:  To monitor trends in antimicrobial resistance among enteric bacteria from humans, retail meats, and animals;  To disseminate timely information on antimicrobial resistance to promote interventions that reduce resistance among foodborne bacteria;  To conduct research to better understand the emergence, persistence, and spread of antimicrobial resistance; and  To provide data that assist the FDA in making decisions related to the approval of safe and effective antimicrobial drugs for animals. 1 C. NARMS Components The NARMS program has three components, which are briefly described below. 1. Human Component The human component of NARMS was launched in 1996 within the framework of CDC’s Emerging Infections Program and the Foodborne Diseases Active Surveillance Network (FoodNet). Initially, it included non-Typhi Salmonella and Escherichia coli O157 isolates from 14 state and local health departments. Surveillance later expanded to include additional bacteria and testing sites. In 1999, testing of Salmonella serotype Typhi and Shigella was added. By 2003, NARMS conducted nationwide surveillance for Salmonella, Shigella, and E. coli O157 from humans. Testing of Campylobacter from humans began in 5 FoodNet sites in 1997 and expanded to all 10 FoodNet sites by 2003. In 2009, NARMS began testing Vibrio species other than V. cholerae from all 50 states. Antimicrobial susceptibility testing of NARMS human isolates was performed at CDC’s laboratories in the National Center for Emerging and Zoonotic Infectious Diseases in Atlanta, Georgia. 2. Retail Meat Component The retail meat component of NARMS was launched in 2002, following a 15-month pilot study in Iowa. Retail meat surveillance was conducted through an ongoing collaboration among FDA’s Center for Veterinary Medicine (CVM), CDC, and state departments of public health.1 Participating sites purchased chicken breasts, ground turkey, ground beef, and pork chops at retail stores and cultured them for Salmonella and Campylobacter. 2 Three or four sites also cultured retail meats for E. coli and Enterococcus.3 Isolates were sent to CVM’s Office of Research in Laurel, Maryland for species and serotype confirmation, antimicrobial susceptibility testing, and genetic analysis. 3. Animal Component The animal component of NARMS began in 1997 with monitoring of Salmonella, and later expanded to include Campylobacter (1998), E. coli (2000), and Enterococcus (2003) isolated from chicken carcasses. This report includes data for Campylobacter and E. coli from chicken carcass rinsates and data for Salmonella from carcass rinsates (chicken), carcass swabs (turkey, cattle and swine), and ground products (chicken, turkey, and beef). Isolates were recovered from samples obtained at federally inspected slaughter and processing plants. Antimicrobial susceptibility testing for the animal component of NARMS was conducted at the USDA’s Agricultural Research Service (ARS) Bacterial Epidemiology and Antimicrobial Resistance Research Unit at the Russell Research Center in Athens, Georgia. D. Links to Additional Information Additional information about NARMS, including comprehensive annual reports for each NARMS component and culture methodology, can be found on the FDA, CDC, and USDA websites listed below. The FDA website also includes NARMS Executive Reports. 1 Most of the sites were participating FoodNet sites. In 2008, the Pennsylvania Department of Health joined the NARMS retail meat surveillance program, testing only Salmonella that year, then added Campylobacter in 2009. 2 Beginning in 2008, all FoodNet sites tested for Campylobacter in retail poultry only. 3 From 2002 through 2006 and in 2010, four sites cultured retail meats for E. coli and Enterococcus and from 2007- 2009, three sites cultured retail meats for E. coli and Enterococcus. 2 FDA: http://www.fda.gov/AnimalVeterinary/SafetyHealth/AntimicrobialResistance/ NationalAntimicrobialResistanceMonitoringSystem/default.htm CDC: http://www.cdc.gov/narms USDA: http://www.ars.usda.gov/saa/bear/narms Information about the Foodborne Diseases Active Surveillance Network (FoodNet) can be found on the following CDC website: http://www.cdc.gov/foodnet/ 3 II. Summary of the NARMS 2010 Executive Report Highlights from results of testing of Salmonella and Campylobacter strains isolated from humans, retail meats, and food animals in 2010 are described below. For more information, including a description of changes in surveillance methods over time, refer to other sections of this report and individual agency NARMS 2010 reports for human, retail meat, and food animal isolates. Non-Typhoidal Salmonella A total of 3,947 non-typhoidal Salmonella isolates were tested, consisting of 2,474 from humans, 400 from retail meats, and 1,073 from healthy food animals at slaughter. Among retail meats, Salmonella was isolated from 15% of ground turkey samples, 13% of chicken breast samples, 1.5% of pork chop samples, and 0.5% of ground beef samples. Some key findings about serotype distribution and antimicrobial resistance are described below. Serotype Frequencies Some of the most common serotypes among isolates from humans were also common among food isolates, particularly isolates from poultry sources. • Among isolates from humans, Enteritidis (21%), Typhimurium (15%), and Newport (12%) remained the three most common serotypes, followed by Javiana (7.2%), I 4,[5],12:i:- (3.1%), and Heidelberg (2.5%). • The four most common serotypes among retail chicken breast isolates were Typhimurium (46%), Enteritidis (16%), Heidelberg (12%), and Kentucky (12%); the four most common serotypes among isolates from chickens at slaughter were Kentucky (43%), Enteritidis (27%), Typhimurium (10%), and Heidelberg (4.4%). • Hadar and Saintpaul remained among the top three serotypes recovered from retail ground turkey (10% and 24% of retail ground turkey isolates, respectively) and from turkeys at slaughter (20% and 14% of turkey isolates, respectively). Serotypes Heidelberg and IIIa 18:z4,z23:- increased in prevalence from 2009, and were two of the top four serotypes isolated from retail ground turkey (8.4% and 11% of all serotypes, respectively) and turkeys at slaughter (9.3% and 7.3%, respectively). • Montevideo (25%) and Dublin (17%) remained the predominant serotypes among isolates from cattle at slaughter. The proportion of cattle isolates that are serotype Dublin has steadily increased since 1997. This serotype is associated with invasive salmonellosis in humans (Jones et al., 2008). Quinolones Resistance to nalidixic acid is correlated with decreased susceptibility to fluoroquinolones (Crump et al., 2003), a class of drugs important for treating complicated Salmonella infections in humans (Pegues and Miller, 2010). In the United States, fluoroquinolones are approved for the 4 treatment of certain respiratory infections in swine and cattle (Animal Drugs @ FDA), but these agents are not currently approved for use in poultry. • Nalidixic acid resistance has remained <3% from all sources since 2004. In 2010, 2.0% of human Salmonella isolates were resistant to nalidixic acid. Enteritidis was the most common serotype (55%) among the nalidixic acid-resistant isolates from humans. Among the serotype Enteritidis isolates tested from humans, 5.2% were resistant. Enteritidis isolates were rarely resistant to the other agents tested. • Nalidixic acid resistance was rare in retail meat and animal isolates. It was found only in seven isolates from cattle (2.8%), one isolate from turkey (0.7%), and one from retail ground turkey (0.5%) in 2010. Of the seven cattle isolates, six were serotype Dublin. Both the turkey and retail ground turkey isolates were serotype Albert. Cephems Ceftriaxone, an extended-spectrum cephalosporin, is important for treating complicated Salmonella infections in humans (Pegues and Miller, 2010). Ceftiofur, a closely related antimicrobial agent, is licensed for use in food animal production (Animal Drugs @ FDA). • Among all isolates from humans, ceftriaxone resistance has been relatively stable since 2004 (<4%). In 2010, 2.8% of isolates were resistant. Newport (31%), Typhimurium (26%), and Heidelberg (21%) were the most common serotypes among the ceftriaxone-resistant isolates. • Ceftriaxone resistance was found in 35% of retail chicken breast isolates in 2010, after rising from 16% in 2007 to 38% in 2009. Typhimurium (81%) was the predominant serotype among these ceftriaxone-resistant isolates. Among retail ground turkey isolates, ceftriaxone resistance rose from 5.7% in 2009 to 16% in 2010, the highest since testing began in 2002. • Among isolates from food animals at slaughter, resistance to ceftriaxone was 22% among isolates from cattle, 15% among isolates from turkeys, and 12% among isolates from chickens. Only 2 isolates (1.8%) from swine were ceftriaxone resistant. Resistance in isolates from cattle and turkeys was the highest observed since testing began in 1997. Dublin was the predominant serotype (55%) among ceftriaxone-resistant isolates from cattle. Heidelberg (22%), Brandenburg (13%), and Schwarzengrund (13%) were the most common serotypes among ceftriaxone-resistant turkey isolates. Kentucky (55%), Typhimurium (24%) and Heidelberg (12%) were the most common serotypes among the ceftriaxone-resistant chicken isolates. • Among serotype Newport isolates from humans, ceftriaxone resistance declined to 7.2% after peaking at 26% in 2001. A decline was also observed among cattle isolates; resistance declined to 60% (3/5) in 2010 after peaking at 82% (22/27) in 2005. • Among serotype Typhimurium isolates from humans, ceftriaxone resistance declined from 6.5% in 2009 to 4.9% in 2010. Among retail chicken breast isolates, resistance steadily rose from 44% in 2007 to 61% in 2010. • Human infections with serotype Heidelberg are associated with invasive disease (Jones et al., 2008). Although the percentage of serotype Heidelberg among all Salmonella isolated from humans and poultry has declined, ceftriaxone resistance has increased. Among human strains, ceftriaxone resistance increased from 8.0% in 2008 to 21% in 2009 and 24% in 2010. 5 Among retail chicken breast isolates, resistance rose from 17% in 2008 to 32% in 2009 then declined to 24% in 2010. Among isolates from chickens at slaughter, ceftriaxone resistance increased from 8.5% in 2008 to 18% in 2009 and rose to 32% in 2010. Resistance in isolates from retail ground turkey and turkeys at slaughter increased from 3.5% and 13% in 2008 to 10% and 33% in 2009, and rose to 24% and 36% in 2010, respectively. No Resistance Detected • In 2010, 85% of isolates from humans had no resistance to any antimicrobial agents tested, an increase from 74% in 1999. • Among isolates from retail meats and food animals at slaughter, the percentage that had no resistance to any antimicrobial agents tested was highest in bovine sources (61% in cattle and 57% in retail ground beef) and lowest in turkeys (25% in turkeys at slaughter and 31% in retail ground turkey). Multidrug Resistance • Among isolates from humans, resistance to ≥3 antimicrobial classes was 9.1%, the lowest since 1996. However, among serotype Heidelberg isolates, resistance to ≥3 antimicrobial classes increased from 17% in 2007 to 28% in 2008 and 34% in 2010. Similarly, among serotype I 4,[5],12:i:- isolates, resistance to ≥3 antimicrobial classes increased from 5.5% in 2007 to 13% in 2009 and 22% in 2010. Typhimurium (44%) was the most common serotype among isolates resistant to ≥3 classes. • Among retail chicken breast isolates, resistance to ≥3 antimicrobial classes declined to 43%, after a steady increase from 24% in 2006 to 49% in 2009. Typhimurium (81%) was the predominant serotype among isolates with resistance to ≥3 classes. • Among retail ground turkey isolates, resistance to ≥3 antimicrobial classes was found in 34%, a decline from the peak resistance of 52% in 2008. Similarly, the percentage of Heidelberg isolates resistant to ≥3 classes declined to 65% from the peak of 83% in 2008. Heidelberg (16%), Saintpaul (16%), and I 4,[5],12:r:- (13%) were the most common serotypes among isolates resistant to ≥3 classes. • Among isolates from turkeys at slaughter, resistance to ≥3 antimicrobial classes increased from 30% in 2008 to 37%. Hadar (16%), Heidelberg (14%), Saintpaul (11%) and Schwarzengrund (11%) were the most common serotypes among isolates resistant to ≥3 classes. • Among isolates from cattle at slaughter, resistance to ≥3 antimicrobial classes increased from 22% in 2007 to 28%. Dublin was the predominant serotype (55%) among isolates with resistance ≥3 antimicrobial classes. An important multidrug resistance (MDR) pattern in Salmonella is resistance to ampicillin, chloramphenicol, streptomycin, sulfonamides, and tetracycline (ACSSuT). This pattern is associated with invasive disease in humans (Varma et al., 2005a; Varma et al, 2005b). Another important MDR pattern linked to severe illness in humans is resistance to ACSSuT with 6 additional resistance to amoxicillin-clavulanic acid and ceftriaxone (ACSSuTAuCx) (Gupta, 2003).  Among isolates from humans, resistance to at least ACSSuT declined to 4.3%, the lowest since testing began in 1996. Typhimurium (64%) and Newport (21%) were the most common serotypes among isolates with this resistance pattern. ACSSuT resistance was detected in 19% of Typhimurium and 7.2% of Newport isolates.  Among isolates from retail poultry and poultry at slaughter, ACSSuT resistance remained low (<5%). In 2010, 7.3% of isolates from swine at slaughter were resistant to ACSSuT, a decline from the peak resistance of 13% in 2009.  Among isolates from cattle at slaughter, resistance to at least ACSSuT was 19%. Dublin (52%) and Typhimurium (15%) were the most common serotypes among cattle isolates with this resistance pattern. ACSSuT resistance was 59% among Dublin and 47% among Typhimurium isolates.  In 2010, 1.3% of isolates from humans were resistant to at least ACSSuTAuCx. Newport (67%) and Typhimurium (21%) were the most common serotypes among isolates with this resistance pattern. Resistance was 7.2% among Newport isolates and 1.9% among Typhimurium isolates.  Resistance to ACSSuTAuCx among isolates from retail poultry, poultry at slaughter, and swine remained low (<4.5%), as it has been since testing began.  ACSSuTAuCx resistance among cattle isolates from slaughter was 16%, an increase from 9.5% in 2009. Serotype Dublin accounted for 55% of isolates with this pattern. Fifty-four percent of serotype Dublin isolates from cattle were resistant to at least ACSSuTAuCx.  Campylobacter A total of 2,136 Campylobacter isolates were tested, including 1,310 from humans, 518 from retail meats (505 from chicken breasts and 13 from ground turkey) and 308 from chickens at slaughter. Poultry are a major source of human C. jejuni infections. All sources except retail ground turkey yielded higher proportions of C. jejuni than C. coli. The distribution of these predominant species varied by source: 88% C. jejuni and 9% C. coli among isolates from humans; 70% C. jejuni and 29% C. coli among retail chicken breast isolates; 39% C. jejuni and 54% C. coli among retail ground turkey isolates; 68% C. jejuni and 32% C. coli among chicken isolates. Resistance to three important antimicrobial classes highlighted below was mostly higher among C. coli than among C. jejuni isolates from all sources. Some key antimicrobial resistance findings are described below.     Macrolides   The macrolides, erythromycin and azithromycin are important antimicrobial agents for the treatment of severe campylobacteriosis in humans (Allos and Blaser, 2010). Macrolides are also authorized for use in food-producing animals (Animal Drugs @ FDA).  Erythromycin resistance in C. jejuni isolated from humans, retail chicken breasts, and chickens at slaughter has remained below 4.0% since testing began. 7