Prevalence and Antimicrobial Resistance of Enterococcus Species Isolated from Retail Meats

Transcription

1 APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Dec. 2003, p Vol. 69, No /03/$ DOI: /AEM Prevlence nd Antimicrobil Resistnce of Enterococcus Species Isolted from Retil Mets Joshu R. Hyes, 1,2 Lind L. English, 2 Peggy J. Crter, 2 Terry Proescholdt, 2 Kyung Y. Lee, 2 Dvid D. Wgner, 2 nd Dvid G. White 2 * Deprtment of Cell Biology nd Moleculr Genetics, University of Mrylnd, College Prk, Mrylnd 20742, 1 nd Center for Veterinry Medicine, U.S. Food nd Drug Administrtion, Lurel, Mrylnd Received 16 My 2003/Accepted 10 September 2003 From Mrch 2001 to June 2002, totl of 981 smples of retil rw mets (chicken, turkey, pork, nd beef) were rndomly obtined from 263 grocery stores in Iow nd cultured for the presence of Enterococcus spp. A totl of 1,357 enterococcl isoltes were recovered from the smples, with contmintion rtes rnging from 97% of pork smples to 100% of ground beef smples. Enterococcus fecium ws the predominnt species recovered (61%), followed by E. feclis (29%), nd E. hire (5.7%). E. fecium ws the predominnt species recovered from ground turkey (60%), ground beef (65%), nd chicken brest (79%), while E. feclis ws the predominnt species recovered from pork chops (54%). The incidence of resistnce to mny production nd therpeutic ntimicrobils differed mong enterococci recovered from retil met smples. Resistnce to quinupristin-dlfopristin, humn nlogue of the production drug virginimycin, ws observed in 54, 27, 9, nd 18% of E. fecium isoltes from turkey, chicken, pork, nd beef smples, respectively. No resistnce to linezolid or vncomycin ws observed, but high-level gentmicin resistnce ws observed in 4% of enterococci, the mjority of which were recovered from poultry retil mets. Results indicte tht Enterococcus spp. commonly contminte retil mets nd tht dissimilrities in ntimicrobil resistnce ptterns mong enterococci recovered from different met types my reflect the use of pproved ntimicrobil gents in ech food niml production clss. Protection of the food supply includes considertions of the microbiologicl qulity nd sfety of commodities vilble for public consumption. While these concerns often ddress specific pthogenic microorgnisms tht present n immedite risk to public helth, there is growing interest in commensl components of the flor ssocited with food-producing nimls tht my lso impct consumers. Species of the genus Enterococcus comprise lrge proportion of the utochthonous microflor ssocited with the gstrointestinl trcts of nimls nd re frequently responsible for significnt morbidity nd mortlity in predisposed humns (27). Enterococci re common components of the microflorl community of mmmls, birds, insects, nd reptiles nd re commonly found in soil, on plnts, nd in wter. These orgnisms re prticulrly chllenging to eliminte becuse of their bility to dpt to environmentl stresses. Thus, it is not surprising tht ntimicrobil-resistnt vrints of enterococci hve been recovered from mets, diry products, nd redy-to-et foods nd hve even been found within probiotic formultions (29). In the clinicl environment, enterococci cn persist for long periods of time on surfces nd cn redily be trnsferred mong the ptient popultion, mny of whom my be prone to coloniztion (46). Enterococci, prticulrly Enterococcus feclis nd E. fecium, present serious chllenges to the control of ntimicrobil resistnce s they re the third leding cuse of nosocomil * Corresponding uthor. Miling ddress: Center for Veterinry Medicine, U.S. Food nd Drug Administrtion, Lurel, MD Phone: (301) Fx: (301) E-mil: dwhite@cvm.fd.gov. infections in intensive cre units in the United Sttes (18). Additionlly, infections cused by other Enterococcus species (E. durns, E. vium, E. rffinosus, E. gllinrum, nd E. csseliflvus) occsionlly occur nd wrrnt ttention (44). Perhps more importntly, enterococci re dept in cquiring nd trnsferring elements tht confer resistnce to ntimicrobils. In ddition, they re known to be intrinsiclly resistnt to severl ntibiotics. As result, therpeutic options for tretment of enterococcl infections re incresingly limited (44). In 1980 the reported development of nd subsequent increse in resistnce to the glycopeptide vncomycin mong clinicl isoltes of Enterococcus spp. ws followed by flurry of reserch into new ntimicrobils for lterntive therpy. Ironiclly, the 1999 Food nd Drug Administrtion pprovl of the streptogrmin quinupristin-dlfopristin (Q-D; Synercid) to tret vncomycin-resistnt E. fecium infections cme fter more thn 20 yers of widespred use of the streptogrmin nlogue virginimycin in niml production. This hs revived concerns tht use of ntimicrobils in food niml production might compromise the efficcy of relted drugs in humn clinicl medicine through selection of resistnt popultions nd their subsequent trnsfer through the food supply (30). Enterococci of food-borne origin hve not been conclusively identified s direct cuses of clinicl infections; however, the consumption of met crrying ntibiotic-resistnt bcteril popultions is possible route of trnsfer nd could result in either coloniztion or trnsfer of resistnce determinnts to host-dpted strins. Dt on the prevlence of ntimicrobilresistnt enterococci from retil food re unfortuntely sprse in the United Sttes nd re urgently needed for scientific ssessments of the reltive risks of using ntimicrobils in 7153

2 7154 HAYES ET AL. APPL. ENVIRON. MICROBIOL. niml husbndry. The dt reported here re the results of pilot surveillnce project undertken in Iow to determine the prevlence nd ntimicrobil resistnce profiles of enterococci in retil mets. MATERIALS AND METHODS Smple collection. Between Mrch 2001 nd June 2002, 981 pckges of retil turkey, chicken, pork, nd beef were purchsed from 263 seprte grocery stores round Iow. Turkey nd beef smples were predominntly ground products, while smples of pork nd chicken were predominntly whole cuts. Grocery loctions in Iow (supermrkets or superstores) were drwn from two dtbses, the Chin Stores Grocery Guide (Chin Store Guide, Tmp, Fl.) nd the Single Unit Grocery Guide. These guide dtbses were filtered by sles volume to eliminte most of the nongrocery convenience-type stores. This list ws inspected, nd the obvious helth food nd convenience stores were eliminted. Field personnel smpled one pckge ech of turkey, chicken, pork, nd beef from six different supermrket stores on weekly bsis. Retil met smples were seled in plstic bg, lbeled with unique identifying number, nd plced into cooler with ice pcks. Field personnel trnsported the food specimens to Food nd Drug Administrtion-Center for Veterinry Medicine lbortories in Lurel, Md., within 48 h of collection. Smple processing nd isoltion of enterococci. Two hundred twenty-five milliliters of Enterococcosel enrichment broth (BBL Microbiology Systems, Cockeysville, Md.) ws dded to 25 g of septiclly weighed ground smple in stomcher bg. Bgs were stomched with Stomcher 400 circultor (Sewrd, Inc., London, Englnd) t 230 rpm for 2 min. Whole cuts were dded to sterile Whirl-Pk bg (Nsco, Fort Atkinson, Wis.), nd t lest 225 ml of Enterococcosel broth ws dded to cover the met smple. Bgs were plced on n Innov 2100 pltform shker (New Brunswick Scientific, Edison, N.J.) nd shken t 200 rpm for 15 min, followed by the septic removl of the whole cut. Enrichment broths were then closed nd incubted in wter bth t 45 C nd evluted t 24 nd 48 h for blckening of the culture broth. When blckening ws observed, 10- l loop ws used to strek the surfce of n Enterococcosel gr plte, which ws incubted t 35 C for 24 2 h. If no growth or no blckening ws observed in the enrichment broth fter 48 h of incubtion, the culture ws deemed negtive nd discrded. From ech Enterococcosel gr plte, up to three colonies of distinctive morphology were streked for isoltion onto blood gr pltes. Identifiction of enterococci. Presumptive enterococci were identified on the bsis of esculin hydrolysis, Grm stin, ctlse rection, nd pyrrolidonyl rylmidse test results (BBL). Hemolytic rection nd pigmenttion were lso recorded. Use of the Enterococcus AccuProbe culture identifiction kit (Gen- Probe, Inc., Sn Diego, Clif.) ws reserved for isoltes tht were mbiguously identified. The VITEK (biomérieux, Inc., Hzelwood, Mo.) microbil identifiction system ws routinely used to distinguish Enterococcus species. Supplementry testing included rbinose nd sucrose utiliztion (Sigm-Aldrich, St. Louis, Mo.), s well s ssys for motility nd methyl- -D-glucopyrnosidse production (17). Isoltes were frozen t 70 C in brucell broth with 20% glycerol. Antimicrobil susceptibility testing of enterococci. Antibiogrms for ech of the enterococcl isoltes were determined with the Sensititre ntimicrobil susceptibility testing system for 17 ntimicrobils (Trek Dignostic Systems, Inc., Westlke, Ohio). The ntimicrobils nd tested rnges were s follows: bcitrcin, 8 to 128 IU/ml); chlormphenicol, 2 to 32 g/ml; ciprofloxcin, 0.12 to 4 g/ml; erythromycin nd linezolid, 0.5 to 8 g/ml; bmbermycin (Flvomycin), lincomycin, Q-D, nd slinomycin, 1 to 32 g/ml; nitrofurntoin, 2 to 128 g/ml; penicillin, 0.5 to 16 g/ml; tetrcycline, 4 to 32 g/ml; tylosin, 0.25 to 32 g/ml; vncomycin, 0.5 to 32 g/ml; gentmicin nd knmycin, 128 to 1,024 g/ml; streptomycin, 512 to 2,048 g/ml. Microtiter pltes contining the tested ntimicrobils with finl inoculum concentrtion of pproximtely CFU/ml were incubted t 37 C for 24 1 h in mbient ir. E. feclis strins ATCC nd ATCC were used s qulity control orgnisms. The pltes were removed nd red mnully for growth to score the MIC determintions using the following NCCLS brekpoints: chlormphenicol nd vncomycin, 32 g/ ml; erythromycin nd linezolid, g/ml; penicillin nd tetrcycline, 16 g/ml; Q-D nd ciprofloxcin, 4 g/ml; nitrofurntoin, 128 g/ml; gentmicin, 500 g/ml; streptomycin, 1,000 g/ml (45). Non-NCCLS resistnce brekpoints for bcitrcin ( 64 IU/ml), tylosin ( g/ml), bmbermycin ( g/ml), nd slinomycin ( g/ml) hve been used elsewhere (3, 4, 45), while no brekpoint for lincomycin hs been estblished. A brekpoint of 500 g/ml ws used for knmycin. Enterococcl ntibiogrms recovered from different isoltes from TABLE 1. Prevlence of Enterococcus spp. mong retil met products from Iow Met clss No. smpled No. positive % Positive Turkey Chicken Pork Beef All mets the sme retil met smple tht differed by less thn 2 dilutions for one or more ntimicrobil MICs were considered duplictes, nd only single isolte ws included for further nlysis. Chi-squre nlysis ws performed using commercil sttisticl nlysis softwre (SAS Institute, Cry, N.C.) to determine significnt differences in resistnce rtes mong met types s well s between popultions E. fecium nd E. feclis. RESULTS Isoltion nd identifiction of enterococcl species. Enterococci were observed to be ubiquitous mong retil met products collected from Iow, with the recovery of enterococci from 99% of 981 smples cultured (Tble 1). Only 13 isoltes were not identified to species. Resistnce profiles were estblished for ll 1,511 isoltes except for 1 tht did not grow in Mueller- Hinton broth. The collection ws reduced to 1,357 unique isoltes fter the removl of isoltes of the sme species with nondistinct susceptibility ptterns from the sme met smple. Among ll met clsses, E. fecium (61%) ws the most frequently encountered species, followed by E. feclis (29%), E. hire (5.7%), E. csseliflvus (2.1%), E. durns (1.2%), E. gllinrum (0.7%), nd E. vium (0.1%), lthough differences in species prevlence vried by met commodity (Tble 2). Notbly, E. fecium ws the predominnt species recovered from turkey, beef, nd chicken met, while E. feclis ccounted for the mjority of isoltes from pork. The predominnce of E. fecium reltive to E. feclis ws gretest mong enterococci isolted from chicken (5:1), followed by beef (4:1) nd turkey (2:1). E. csseliflvus nd E. gllinrum were isolted more frequently from turkey thn from other met clsses, while E. durns ws recovered more frequently from pork nd beef smples. Interestingly, E. hire ws more often recovered from beef thn from the other mets nlyzed. Antimicrobil resistnce of E. fecium nd E. feclis isoltes. To ssess the differences tht might exist mong Enterococcus spp. isolted from different met products, the ntimicrobil resistnce profiles of the comprtively lrge popultions of E. fecium (n 825) nd E. feclis (n 388) were exmined (Tble 3). The distributions of bcitrcin MICs for E. fecium nd E. feclis were shifted to the upper rnge tested, with MICs for the mjority of E. fecium isoltes from turkey, chicken, nd beef nd E. feclis isoltes from turkey nd chicken exceeding the upper limit ( 128 g/ml). Resistnce to chlormphenicol ws seen t very low level ( 1%) cross the popultions of E. fecium recovered, while resistnt subpopultion of E. feclis ws observed only mong popultions isolted from pork. Resistnce to ciprofloxcin ws observed t higher frequency mong E. fecium isoltes thn mong E. feclis isoltes, with the gretest prevlence mong E. fecium isoltes recovered from turkey nd chicken

3 VOL. 69, 2003 ANTIBIOTIC-RESISTANT ENTEROCOCCI FROM RETAIL MEATS 7155 TABLE 2. Reltive prevlence of Enterococcus spp. by retil met clss Species No. of Enterococcus sp., isoltes (% of met clss isoltes) in: Turkey Chicken Pork Beef All mets E. vium 1 (0.3) (0.1) E. csseliflvus 21 (5.9) 3 (1.0) 2 (0.7) 3 (0.8) 29 (2.1) E. durns 0 1 (0.3) 7 (2.3) 8 (2.0) 16 (1.2) E. feclis 110 (31) 51 (16) 161 (54) 66 (17) 388 (29) E. fecium 213 (60) 245 (79) 114 (38) 254 (65) 826 (61) E. gllinrum 6 (1.7) 0 1 (0.3) 2 (0.5) 9 (0.7) E. hire 3 (0.8) 10 (3.2) 10 (3.4) 54 (14) 77 (5.7) Unidentified 3 (0.8) 1 (0.3) 3 (1.0) 4 (1.0) 11 (0.8) Totl ,357 (41 nd 22%, respectively; P 0.01). The rnges of MICs of ciprofloxcin for E. fecium isoltes were more widely distributed thn those for E. feclis. The distributions of MICs of the glycolipid ntimicrobil bmbermycin were reltively consistent mong both species nd did not pper to vry mong retil met commodities. MICs were consistently higher mong E. fecium isoltes (MIC t which 50% of isoltes were inhibited [MIC 50 ] 32 g/ml) thn mong E. feclis isoltes (MIC 50 2 g/ml), which my reflect species-specific intrinsic resistnce to or tolernce of this ntimicrobil (P 0.01). This is contrsted with the MIC distributions for the ionophore slinomycin nd the mcrolides erythromycin nd tylosin, which were elevted for both enterococcl species isolted from turkey nd chicken met (P 0.01). Differences between species in the rnge of lincomycin MICs were similrly observed: clustered distribution t the upper level of tested concentrtions for E. feclis isoltes nd greter rnge mong E. fecium isoltes. Resistnce to nitrofurntoin ws observed in one-hlf of ll E. fecium isoltes, while it ws observed mong only 5.5% of E. feclis isoltes from turkey. E. fecium isoltes were lso more often resistnt to penicillin (P 0.01), with the highest rtes from turkey nd chicken sources (P 0.01). Tetrcycline resistnce ws observed more frequently mong E. feclis isoltes (P 0.01), with the highest prevlence mong both E. fecium nd E. feclis isoltes from turkey, followed by those from pork, chicken, nd beef. Resistnce to vncomycin or linezolid ws not observed mong E. fecium or E. feclis isoltes, but MICs for 48% of ll E. fecium isoltes were distributed 1 dilution wy from clinicl resistnce to linezolid (MIC 4 g/ml). Over 94% of ll E. feclis isoltes were resistnt to the streptogrmin Q-D, likely due to the purported intrinsic resistnce of this species to this ntimicrobil. Resistnce to this streptogrmin ws highest mong E. fecium isolted from turkey (54%), followed by chicken (27%), beef (18%), nd pork (9%). It is notble tht the distribution of MICs of Q-D for E. fecium of poultry origin reveled tht the vlues were bimodlly distributed nd ccounted for 76% of ll resistnt E. fecium isoltes. Antimicrobil resistnce profiles of other Enterococcus spp. Among the less frequently recovered enterococcl species, decresed susceptibility to bmbermycin ws observed mong ll species, with some vribility mong E. csseliflvus nd E. gllinrum popultions (Tble 4). Erythromycin resistnce ws observed in between 0 nd 44% of the less frequently isolted enterococcl species. No striking differences mong the MICs for these popultions of bcitrcin nd slinomycin were observed lthough less vribility in bcitrcin MICs ws observed mong E. csseliflvus isoltes. Resistnce to tetrcycline ws frequent, with over 70% of ll isoltes displying resistnce, while resistnce to nitrofurntoin ws less common. No resistnce to linezolid or vncomycin ws observed; however, resistnce to Q-D mong 100, 41, 33, nd 14% of E. vium, E. csseliflvus, E. gllinrum, nd E. hire isoltes, respectively, ws observed. No resistnce to Q-D mong E. durns isoltes ws observed. Similr to wht ws observed for E. fecium nd E. feclis, 74% of these other species tht were Q-D resistnt were of poultry origin. High-level minoglycoside resistnce mong Enterococcus spp. Resistnce to high-level minoglycosides ws prevlent cross ll species recovered (Tble 5). Aside from the single isolte of E. vium tht ws resistnt, the observed frequency of resistnce to ny of the three tested minoglycosides ws highest mong isoltes of E. csseliflvus (86%), followed by those of E. fecium (58%), E. gllinrum (56%), E. durns (38%), E. feclis (17%), nd E. hire (12%). The ptterns of susceptibility to high-level minoglycosides were interesting in tht resistnce to knmycin ws the most prevlent, followed by resistnce to streptomycin nd resistnce to gentmicin. Upon closer exmintion of high-level minoglycoside resistnce mong E. feclis nd E. fecium isoltes, the resistnce frequencies for both popultions were highest for those tht originted from poultry met, with rtes of 27, 33, 11, nd 5% for E. feclis nd 74, 62, 41, nd 47% for E. fecium isoltes from turkey, chicken, pork, nd beef, respectively (P 0.01). Specificlly, high-level gentmicin resistnce ws observed more frequently mong isoltes from poultry sources. DISCUSSION This work describes the distribution of enterococci mong retil met products from the Iow nd estblishes bseline for ntimicrobil resistnce mong isolted Enterococcus spp. to ntimicrobils of humn nd veterinry importnce. Although we did not ttempt to quntitte the enterococcl popultion within smples from Iow in this study, the demonstrtion of ner omnipresence of enterococci is likely reflective of sizble popultion mong the norml nturl microflor of retil met products. This is consistent with isoltion rtes of 82 to 86% from chickens reported from previous study of

4 7156 HAYES ET AL. APPL. ENVIRON. MICROBIOL. TABLE 3. Antimicrobil resistnce profiles of E. fecium nd E. feclis isoltes from retil mets Antimicrobil nd met clss Resistnce E. fecium (n 825 c ) E. feclis (n 388 g ) brekpoint,b MIC 50 MIC 90 Rnge % Resistnt MIC 50 MIC 90 Rnge % Resistnt Bcitrcin 64 Turkey d,e d,e Chicken Pork Beef Bmbermycin Turkey e e Chicken Pork Beef Chlormphenicol 16 Turkey Chicken Pork Beef Ciprofloxcin 4 Turkey d,e e Chicken Pork Beef Erythromycin Turkey d d Chicken Pork Beef Lincomycin NA f Turkey NA NA Chicken NA NA Pork NA NA Beef NA NA Linezolid Turkey Chicken Pork Beef Nitrofurntoin 128 Turkey e e Chicken Pork Beef Penicillin Turkey d,e e Chicken Pork Beef Q-D 4 Turkey d,e e Chicken Pork Beef Slinomycin Turkey Chicken Pork Beef Continued on fcing pge

5 VOL. 69, 2003 ANTIBIOTIC-RESISTANT ENTEROCOCCI FROM RETAIL MEATS 7157 TABLE 3 Continued Antimicrobil nd met clss Resistnce E. fecium (n 825 c ) E. feclis (n 388 g ) brekpoint,b MIC 50 MIC 90 Rnge % Resistnt MIC 50 MIC 90 Rnge % Resistnt Tetrcycline Turkey d,e d,e Chicken Pork Beef Tylosin Turkey Chicken Pork Beef Expressed in microgrms per milliliter except for bcitrcin, for which the units re interntionl units (IU) per milliliter. b Resistnce brekpoints were those provided by NCCLS for chlormphenicol, ciprofloxcin, erythromycin, linezolid, nitrofurntoin, penicillin, Q-D, nd tetrcycline (47) nd those suggested for the bmbermycin nd slinomycin (3) nd tylosin nd bcitrcin (4). c 213, 244, 114, nd 254 isoltes from turkey, chicken, pork, nd beef, respectively. d Denotes sttisticlly significnt differences mong isoltes from the different met types in resistnce to the indicted ntimicrobil (P 0.01). e Denotes sttisticlly significnt differences between E. fecium nd E. feclis isoltes in resistnce to the indicted ntimicrobil (P 0.01). f NA, not pplicble (no estblished NCCLS brekpoint). g 110, 51, 161, nd 66 isoltes from turkey, chicken, pork, nd beef, respectively. wider geogrphicl re (43). Indeed, studies of cooked poultry met suggest tht enterococci do not constitute the lrgest bcteril popultion on such products (9). While no study hs previously determined the reltive proportions of Enterococcus spp. from multiple met types in the United Sttes, E. feclis hs been observed more frequently mong limited number of frozen chicken smples from Michign (53). The predominnce of E. feclis on retil pork products is consistent with studies of the enterococcl microflor of pork crcsses t U.S. processing fcilities (40), lthough the influence of culturl methodology on the recovered popultion of enterococci is importnt (15). Comprtively decresed susceptibility mong E. fecium isoltes, compred to E. feclis isoltes, to the glycolipid bmbermycin hs been previously scribed to intrinsic resistnce differences between the two species (16, 24, 25) lthough reduced tolernce mong E. fecium isoltes from unexposed environments suggests otherwise (4). The prevlence of chlormphenicol resistnce hs been reported more often mong E. feclis isoltes thn mong E. fecium isoltes from production environments (2, 58) nd rw met products brod (28, 38, 49), while rtes of resistnce mong E. fecium isoltes to ciprofloxcin, erythromycin, nitrofurntoin, penicillin, nd tetrcycline re trditionlly higher (26, 28, 47 49). The observtion of decresed susceptibility of E. fecium isoltes, compred to E. feclis isoltes, to slinomycin seen in this study, especilly those of poultry origin, is consistent with previous ionophore susceptibility results from production environments of Denmrk (3) but differs from results for isoltes of broiler origin from Jpn (58) nd Belgium (16). Similrly, the decresed reltive susceptibility of E. fecium of poultry origin to bcitrcin is most similr to the distributions of MICs for enterococci from of chicken nd swine from Denmrk, Finlnd, nd Norwy (4) but differs from those for enterococci from Belgium (16). The frequencies of resistnce to high-level minoglycosides mong the more cliniclly relevnt Enterococcus spp. from food niml production environments, especilly mong E. feclis isoltes, re often reported (22); however, the incresed prevlence of gentmicin resistnce mong E. feclis nd E. fecium isoltes from poultry met seen in this study is inconsistent with the observtions of enterococci from different production environments from Denmrk nd Belgium (2, 16). While dt from comprble sources re few, these geogrphicl differences likely reflect differences in ntimicrobil use in food niml production prctices. Surveillnce of enterococci from food sources for resistnce to the oxzolidinone linezolid hs not been reported previously. Resistnce mong isoltes of E. fecium tht re resis- TABLE 4. MIC rnge nd resistnce profiles of Enterococcus spp. other thn E. feclis nd E. fecium from retil met for selected ntimicrobils Species (n) MIC rnge b (% of isoltes resistnt) of: BAC BMB ERY LNZ NIT Q-D SAL TET E. vium (1) 128 (100) 32 (100) 2 (0) 2 (0) 128 (100) 8 (100) 1 (0) 32 (100) E. csseliflvus (29) (100) 2 32 (97) 0.5 (31) 2 4 (0) (55) 1 32 (41) 1 8 (0) 4 32 (79) E. durns (16) 128 (69) (100) (0) 2 4 (0) (38) 1 2 (0) 1 8 (0) 4 32 (75) E. gllinrum (9) 128 (78) 1 32 (67) 0.5 (44) 1 4 (0) (22) 1 32 (33) 1 4 (0) 4 32 (89) E. hire (77) 128 (22) (100) 0.5 (17) (0) (10) 1 16 (14) 1 8 (0) 4 32 (71) BAC, bcitrcin; ERY, erythromycin; BMB, bmbermycin; LNZ, linezolid; NIT, nitrofurntoin; SAL, slinomycin; TET, tetrcycline. Resistnce brekpoints for Enterococcus spp. were those used by NCCLS (47) for erythromycin ( g/ml), linezolid ( g/ml), nitrofurntoin ( 128 g/ml), Q-D ( 4 g/ml), nd tetrcycline ( g/ml), with the exception of g/ml for the bmbcrmycin nd slinomycin (3) nd 64 IU/ml for bcitrcin (4). b Expressed in interntionl units per milliliter for bcitrcin nd in microgrms per milliliter for ll other ntimicrobils.

6 7158 HAYES ET AL. APPL. ENVIRON. MICROBIOL. TABLE 5. Frequency of high-level minoglycoside resistnce of Enterococcus spp. from retil met Species (n) nd met clss No. (%) of isoltes resistnt to : HLK HLS HLG E. vium (1) 1 (100) 0 0 E. csseliflvus (29) 25 (86) 12 (41) 3 (10) E. durns (16) 6 (38) 0 0 E. feclis (388) 45 (12) b,c 44 (11) b,c 26 (6.7) c Turkey 24 (53) 21 (48) 13 (50) Chicken 12 (27) 8 (18) 9 (35) Pork 8 (18) 12 (27) 4 (15) Beef 1 (2.2) 3 (6.8) 0 E. fecium (825) 414 (50) b,c 175 (21) b,c 26 (3.2) c Turkey 137 (33) 84 (48) 15 (58) Chicken 114 (28) 63 (36) 10 (38) Pork 47 (11) 18 (10) 0 Beef 116 (28) 10 (5.7) 1 (3.8) E. gllinrum (9) 5 (56) 3 (33) 2 (22) E. hire (77) 8 (10) 5 (6.5) 0 Unidentified (11) 7 (64) 1 (9.1) 2 (18) Totl 511 (38) 240 (18) 59 (4.4) Resistnce brekpoints for Enterococcus spp. were 500 g/ml for high-level knmycin (HLK), 1,000 g/ml for high-level streptomycin (HLS), nd 500 g/ml for high-level gentmicin (HLG). b Denotes sttisticlly significnt differences mong isoltes from the different met types in resistnce to the indicted ntimicrobil (P 0.01). c Denotes sttisticlly significnt differences between E. fecium nd E. feclis isoltes in resistnce to the indicted ntimicrobil (P 0.01). tnt to mny ntimicrobils hs been observed (10; R. D. Gonzles, P. C. Schreckenberger, M. B. Grhm, S. Kelkr, K. DenBesten, nd J. P. Quinn, Letter, Lncet 357:1179, 2001) nd t lest in one cse without prior exposure (10). Additionlly, isoltes hve been observed to develop resistnce during the course of tretment (6, 34) nd exhibit cross-resistnce to other oxzolidinones (34). The incresed MICs for E. fecium suggest tht the development of clinicl resistnce mong isoltes of this species my not be difficult dpttion following incresed clinicl usge of this ntimicrobil in humn clinicl medicine. Resistnce to Q-D mong food niml production environments in the United Sttes is not surprising, given the use of the nlogue virginimycin since 1974 (32, 57). The higher frequency of Q-D resistnce mong E. fecium isoltes from turkey thn from chicken might be relted to the different periods of time tht the flocks re exposed to ntimicrobils prior to slughter (32). The resistnce rte of 3% of E. fecium isoltes from rw chicken smples reported from n erlier surveillnce study in the United Sttes using comprble nonselective enrichment methods (43) is much lower thn the 26% observed mong smples from this study. E. feclis isoltes hve been shown to be intrinsiclly resistnt to streptogrmins (51); however, the recent observtion of trnsferble resistnce my lend some significnce to the resistnce seen mong other species in this study (50). While the griculturl usge of ntimicrobils tht hve nlogues in humn medicine is mtter of incresing public concern, resistnce mong E. fecium isoltes from clinicl environments hs been shown to be higher thn resistnce mong those from the community (23), which my or my not reflect similr selection in the clinicl environment. The bsence of vncomycin-resistnt enterococci (VRE) from domestic retil mets in this study is consistent with previous observtions (21, 39, 53; Y. Ike, K. Tnimoto, Y. Ozw, T. Nomur, S. Fujimoto, nd H. Tomit, Letter, Lncet 353:1854, 1999) nd reflects the bsence of isoltion of VRE from both processing (12) nd food niml production environments (31, 35, 53, 57) in the United Sttes. In contrst, VRE re frequently isolted from retil met products (11, 36 38, 42, 49, 55, 56) from Europen countries s result of selection of resistnt popultions by the use of the glycopeptide voprcin in food niml production environments (1, 4, 5, 8, 16, 24, 52, 54). The persistence of VRE on frms tht hve discontinued the use of voprcin for growth promotion illustrtes the impct posed by ntimicrobil usge in food niml production environments (7, 13, 14, 33, 41). It is cler tht resistnt enterococci recovered from rw met products reflect this use of ntimicrobils, but the extent to which these popultions pose risk to the consumer nd the efficcy of therpeutic ntimicrobils to tret disese is unknown. The recent observtions of vncomycin resistnce elements of enterococcl origin in U.S. clinicl isoltes of Stphylococcus ureus suggest tht lterntive therpies, such s linezolid nd Q-D, should be more frequently employed (19, 20). As result, resistnt popultions of enterococci tht my hve entered the humn microflor through the consumption of contminted retil met products my be mplified s result of the inevitble increse in selective pressure in the clinicl environment. Although existing evidence does not suggest tht enterococci of food-borne origin be regrded s bcteril pthogens, they could serve s potentil reservoirs of virulence nd ntimicrobil resistnce genes for host-dpted strins. Our observtions suggest tht Enterococcus spp. commonly contminte retil met products nd tht differences observed in ntimicrobil susceptibility phenotypes my reflect the extent of use of ntimicrobils in specific food niml production environments. Therefore, effective control strtegies imed t reducing enterococcl contmintion of retil mets my become more significnt in the future, with incresing recognition of these bcteri s humn opportunistic pthogens. ACKNOWLEDGMENTS We re grteful to S. Altekruse for contributions to study design; B. A. Cook, S. L. Ayers, S. Hubert, nd S. A. Gines for ssistnce with smple processing; S. M. Bodeis for dt review; nd L. D. Youngmn nd R. D. Wlker for mnuscript review. This project ws supported in prt by n ppointment of J. R. Hyes to the Reserch Fellowship Progrm t the Center for Veterinry Medicine dministered by the Ok Ridge Associted Universities through contrct with the U.S. Food nd Drug Administrtion. REFERENCES 1. Arestrup, F. M Occurrence of glycopeptide resistnce mong Enterococcus fecium isoltes from conventionl nd ecologicl poultry frms. Microb. Drug Resist. 1: Arestrup, F. M., Y. Agerso, P. Gerner-Smidt, M. Mdsen, nd L. B. Jensen Comprison of ntimicrobil resistnce phenotypes nd resistnce genes in Enterococcus feclis nd Enterococcus fecium from humns in the community, broilers, nd pigs in Denmrk. Dign. Microbiol. Infect. Dis. 37: Arestrup, F. M., F. Bger, N. E. Jensen, M. Mdsen, A. Meyling, nd H. C. Wegener Surveillnce of ntimicrobil resistnce in bcteri isolted from food nimls to ntimicrobil growth promoters nd relted therpeutic gents in Denmrk. APMIS 106: Arestrup, F. M., H. Kruse, E. Tst, A. M. Hmmerum, nd L. B. Jensen Associtions between the use of ntimicrobil gents for growth pro-

7 VOL. 69, 2003 ANTIBIOTIC-RESISTANT ENTEROCOCCI FROM RETAIL MEATS 7159 motion nd the occurrence of resistnce mong Enterococcus fecium from broilers nd pigs in Denmrk, Finlnd, nd Norwy. Microb. Drug Resist. 6: Arestrup, F. M., A. M. Seyfrth, H. D. Emborg, K. Pedersen, R. S. Hendriksen, nd F. Bger Effect of bolishment of the use of ntimicrobil gents for growth promotion on occurrence of ntimicrobil resistnce in fecl enterococci from food nimls in Denmrk. Antimicrob. Agents Chemother. 45: Aucklnd, C., L. Tere, F. Cooke, M. E. Kufmnn, M. Wrner, G. Jones, K. Bmford, H. Ayles, nd A. P. Johnson Linezolid-resistnt enterococci: report of the first isoltes in the United Kingdom. J. Antimicrob. Chemother. 50: Bger, F., F. M. Arestrup, M. Mdsen, nd H. C. Wegener Glycopeptide resistnce in Enterococcus fecium from broilers nd pigs following discontinued use of voprcin. Microb. Drug Resist. 5: Bger, F., M. Mdsen, J. Christensen, nd F. M. Arestrup Avoprcin used s growth promoter is ssocited with the occurrence of vncomycinresistnt Enterococcus fecium on Dnish poultry nd pig frms. Prev. Vet. Med. 31: Brkt, R. K., M. W. Griffiths, nd L. J. Hrris Isoltion nd chrcteriztion of Crnobcterium, Lctococcus, nd Enterococcus spp. from cooked, modified tmosphere pckged, refrigerted, poultry met. Int. J. Food Microbiol. 62: Bsustoglu, A., H. Aydogn, C. Beyn, A. Ylcin, nd S. Unl First glycopeptide-resistnt Enterococcus fecium isolte from blood culture in Ankr, Turkey. Emerg. Infect. Dis. 7: Btes, J., J. Z. Jordens, nd D. T. Griffiths Frm nimls s puttive reservoir for vncomycin-resistnt enterococcl infection in mn. J. Antimicrob. Chemother. 34: Bodnruk, P. W., P. J. Krkr, nd R. B. Tompkin Absence of high-level vncomycin resistnce in enterococci isolted from met-processing fcilities. Emerg. Infect. Dis. 7: Borgen, K., G. S. Simonsen, A. Sundsfjord, Y. Wsteson, O. Olsvik, nd H. Kruse Continuing high prevlence of VnA-type vncomycin-resistnt enterococci on Norwegin poultry frms three yers fter voprcin ws bnned. J. Appl. Microbiol. 89: Borgen, K., M. Sorum, H. Kruse, nd Y. Wsteson Persistence of vncomycin-resistnt enterococci (VRE) on Norwegin broiler frms. FEMS Microbiol. Lett. 191: Butye, P., L. A. Devriese, nd F. Hesebrouck Comprison of direct nd enrichment methods for the selective isoltion of vncomycin-resistnt enterococci from feces of pigs nd poultry. Microb. Drug Resist. 5: Butye, P., L. A. Devriese, nd F. Hesebrouck Differences in ntibiotic resistnce ptterns of Enterococcus feclis nd Enterococcus fecium strins isolted from frm nd pet nimls. Antimicrob. Agents Chemother. 45: Crvlho, M. D. G. S., L. M. Teixeir, nd R. R. Fcklm Use of tests for cidifiction of methyl- -D-glucopyrnoside nd susceptibility to efrotomycin for differentition of strins of Enterococcus nd some relted gener. J. Clin. Microbiol. 36: Centers for Disese Control nd Prevention Ntionl Nosocomil Infections Surveillnce (NNIS) system report, dt summry from Jnury 1992 June 2001, issued August Am. J. Infect. Control 29: Centers for Disese Control nd Prevention Stphylococcus ureus resistnt to vncomycin United Sttes, Morb. Mortl. Wkly. Rep. 51: Centers for Disese Control nd Prevention Vncomycin-resistnt Stphylococcus ureus Pennsylvni, Morb. Mortl. Wkly. Rep. 51: Coque, T. M., J. F. Tomyko, S. C. Ricke, P. C. Okhyusen, nd B. E. Murry Vncomycin-resistnt enterococci from nosocomil, community, nd niml sources in the United Sttes. Antimicrob. Agents Chemother. 40: Donbedin, S. M., L. A. Thl, E. Hershberger, M. B. Perri, J. W. Chow, P. Brtlett, R. Jones, K. Joyce, S. Rossiter, K. Gy, J. Johnson, C. Mckinson, E. DeBess, J. Mdden, F. Angulo, nd M. J. Zervos Moleculr chrcteriztion of gentmicin-resistnt enterococci in the United Sttes: evidence of spred from nimls to humns through food. J. Clin. Microbiol. 41: Duh, R. W., K. V. Singh, K. Mlthum, nd B. E. Murry In vitro ctivity of 19 ntimicrobil gents ginst enterococci from helthy subjects nd hospitlized ptients nd use of n ce gene probe from Enterococcus feclis for species identifiction. Microb. Drug Resist. 7: Dutt, G. N., nd L. A. Devriese Susceptibility of fecl streptococci of poultry origin to nine growth-promoting gents. Appl. Environ. Microbiol. 44: Dutt, G. N., nd L. A. Devriese Observtions on the in vitro sensitivity nd resistnce of Grm positive intestinl bcteri of frm nimls to growth promoting ntimicrobil gents. J. Appl. Bcteriol. 56: Eliopoulos, G. M Antibiotic resistnce in Enterococcus species: n updte. Curr. Clin. Top. Infect. Dis. 16: Frnz, C. M., W. H. Holzpfel, nd M. E. Stiles Enterococci t the crossrods of food sfety? Int. J. Food Microbiol. 47: Frei, A., D. Goldenberger, nd M. Teuber Antimicrobil susceptibility of intestinl bcteri from Swiss poultry flocks before the bn of ntimicrobil growth promoters. Syst. Appl. Microbiol. 24: Girff, G Enterococci from foods. FEMS Microbiol. Rev. 26: Gorbch, S. L Antimicrobil use in niml feed time to stop. N. Engl. J. Med. 345: Hrwood, V. J., M. Brownell, W. Perusek, nd J. E. Whitlock Vncomycin-resistnt Enterococcus spp. isolted from wstewter nd chicken feces in the United Sttes. Appl. Environ. Microbiol. 67: Hyes, J. R., A. C. McIntosh, S. Qiyumi, J. A. Johnson, L. L. English, L. E. Crr, D. D. Wgner, nd S. W. Joseph High-frequency recovery of quinupristin-dlfopristin-resistnt Enterococcus fecium isoltes from the poultry production environment. J. Clin. Microbiol. 39: Heuer, O. E., K. Pedersen, J. S. Andersen, nd M. Mdsen Vncomycin-resistnt enterococci (VRE) in broiler flocks 5 yers fter the voprcin bn. Microb. Drug Resist. 8: Johnson, A. P., L. Tysll, M. V. Stockdle, N. Woodford, M. E. Kufmnn, M. Wrner, D. M. Livermore, F. Asboth, nd F. J. Allerberger Emerging linezolid-resistnt Enterococcus feclis nd Enterococcus fecium isolted from two Austrin ptients in the sme intensive cre unit. Eur. J. Clin. Microbiol. Infect. Dis. 21: Joseph, S. W., J. R. Hyes, L. L. English, L. E. Crr, nd D. D. Wgner Implictions of multiple ntimicrobil-resistnt enterococci ssocited with the poultry environment. Food Addit. Contm. 18: Klre, I., D. Bdstubner, C. Konstbel, G. Bohme, H. Clus, nd W. Witte Decresed incidence of VnA-type vncomycin-resistnt enterococci isolted from poultry met nd from fecl smples of humns in the community fter discontinution of voprcin usge in niml husbndry. Microb. Drug Resist. 5: Klre, I., H. Heier, H. Clus, G. Bohme, S. Mrin, G. Seltmnn, R. Hkenbeck, V. Antnssov, nd W. Witte Enterococcus fecium strins with vna-medited high-level glycopeptide resistnce isolted from niml foodstuffs nd fecl smples of humns in the community. Microb. Drug Resist. 1: Klein, G., A. Pck, nd G. Reuter Antibiotic resistnce ptterns of enterococci nd occurrence of vncomycin-resistnt enterococci in rw minced beef nd pork in Germny. Appl. Environ. Microbiol. 64: Knudtson, L. M., nd P. A. Hrtmn Antibiotic resistnce mong enterococcl isoltes from environmentl nd clinicl sources. J. Food Prot. 56: Knudtson, L. M., nd P. A. Hrtmn Enterococci in pork processing. J. Food Prot. 56: Kruse, H., B. K. Johnsen, L. M. Rorvik, nd G. Schller The use of voprcin s growth promoter nd the occurrence of vncomycin-resistnt Enterococcus species in Norwegin poultry nd swine production. Microb. Drug Resist. 5: Lemck, R., nd M. Bulte Occurrence of the vncomycin-resistnt genes vna, vnb, vnc1, vnc2 nd vnc3 in Enterococcus strins isolted from poultry nd pork. Int. J. Food Microbiol. 60: McDonld, L. C., S. Rossiter, C. Mckinson, Y. Y. Wng, S. Johnson, M. Sullivn, R. Sokolow, E. DeBess, L. Gilbert, J. A. Benson, B. Hill, nd F. J. Angulo Quinupristin-dlfopristin-resistnt Enterococcus fecium on chicken nd in humn stool specimens. N. Engl. J. Med. 345: Murry, B. E The life nd times of the Enterococcus. Clin. Microbiol. Rev. 3: Ntionl Committee for Clinicl Lbortory Stndrds Performnce stndrds for ntimicrobil susceptibility testing. Twelfth informtionl supplement, M100-S12. Ntionl Committee for Clinicl Lbortory Stndrds, Wyne, P. 46. O Connell, N. H., nd H. Humphreys Intensive cre unit design nd environmentl fctors in the cquisition of infection. J. Hosp. Infect. 45: Ppprskevs, J., A. Vtopoulos, P. T. Tssios, A. Avlmi, N. J. Legkis, nd V. Klpothki Diversity mong high-level minoglycoside-resistnt enterococci. J. Antimicrob. Chemother. 45: Pesce, A., E. A. Debbi, M. Toni, nd G. C. Schito Antibiotic resistnce of clinicl isoltes of Enterococcus in Itly. Clin. Infect. Dis. 15: Quednu, M., S. Ahrne, A. C. Petersson, nd G. Molin Antibioticresistnt strins of Enterococcus isolted from Swedish nd Dnish retiled chicken nd pork. J. Appl. Microbiol. 84: Simjee, S., D. G. White, D. D. Wgner, J. Meng, S. Qiyumi, S. Zho, nd P. F. McDermott Identifiction of vt(e) in Enterococcus feclis isoltes from retil poultry nd its trnsferbility to Enterococcus fecium. Antimicrob. Agents Chemother. 46: Singh, K. V., G. M. Weinstock, nd B. E. Murry An Enterococcus feclis ABC homologue (Ls) is required for the resistnce of this species to clindmycin nd quinupristin-dlfopristin. Antimicrob. Agents Chemother. 46: Stobberingh, E., B. A. vn den, N. London, C. Driessen, J. Top, nd R. Willems Enterococci with glycopeptide resistnce in turkeys, turkey

8 7160 HAYES ET AL. APPL. ENVIRON. MICROBIOL. frmers, turkey slughterers, nd (sub)urbn residents in the south of The Netherlnds: evidence for trnsmission of vncomycin resistnce from nimls to humns? Antimicrob. Agents Chemother. 43: Thl, L. A., J. W. Chow, R. Mhyni, H. Bonill, M. B. Perri, S. A. Donbedin, J. Silvermn, S. Tber, nd M. J. Zervos Chrcteriztion of ntimicrobil resistnce in enterococci of niml origin. Antimicrob. Agents Chemother. 39: vn den Bogrd, A. E., L. B. Jensen, nd E. E. Stobberingh Vncomycin-resistnt enterococci in turkeys nd frmers. N. Engl. J. Med. 337: vn den Brk, N., A. vn Belkum, M. vn Keulen, J. Vliegenthrt, H. A. Verbrugh, nd H. P. Endtz Moleculr chrcteriztion of vncomycinresistnt enterococci from hospitlized ptients nd poultry products in The Netherlnds. J. Clin. Microbiol. 36: Wegener, H. C., M. Mdsen, N. Nielsen, nd F. M. Arestrup Isoltion of vncomycin resistnt Enterococcus fecium from food. Int. J. Food Microbiol. 35: Welton, L. A., L. A. Thl, M. B. Perri, S. Donbedin, J. McMhon, J. W. Chow, nd M. J. Zervos Antimicrobil resistnce in enterococci isolted from turkey flocks fed virginimycin. Antimicrob. Agents Chemother. 42: Yoshimur, H., M. Ishimru, Y. S. Endoh, nd A. Kojim Antimicrobil susceptibilities of enterococci isolted from feces of broiler nd lyer chickens. Lett. Appl. Microbiol. 31: