Archives of Medical Research 32 (2001) 159–163

0188-4409/01 $–see front matter. Copyright © 2001 IMSS. Published by Elsevier Science Inc.PII S0188-4409(01)00265-X

ORIGINAL ARTICLE

Antimicrobial Resistance from Enterococci in a Pediatric Hospital. Plasmids in

Enterococcus faecalis

Isolates with High-Level Gentamicinand Streptomycin Resistance

Guadalupe Miranda,* Linda Lee,** Cindy Kelly,** Fortino Solórzano,*** Blanca Leaños,*Onofre Muñoz**** and Jan Evans Patterson**

*Unidad de Investigación en Epidemiología Hospitalaria, Hospital de Pediatría, Centro Médico Nacional Siglo XXI, Instituto Mexicano del Seguro Social (IMSS), Mexico City, Mexico

**Department of Medicine, Infectious Diseases, University of Texas Health Science Center at San Antonio (UTHSCSA), San Antonio, TX, USA***Departamento de Infectología, Hospital de Pediatría, Centro Medico Nacional Siglo XXI, IMSS, Mexico City, Mexico

****Coordinación de Investigación Médica, Centro Médico Nacional Siglo XXI, IMSS, Mexico City, Mexico

Received for publication March 1, 2000; accepted December 11, 2000 (00/042).

Background. Enterococcus

spp. is an important nosocomial and community-acquiredpathogen. Recent studies have documented the increasing importance of this pathogen inchildren, particularly in the hospital setting. Our objective in this study was to report thefrequency of antimicrobial resistance in enterococci and to determine the characteristics ofhigh-level gentamicin resistance (HLGR) plasmids in

Enterococcus faecalis

clinical iso-lates.

Methods.

Two hundred eighty-nine enterococcal isolates were collected during an 18-month period from a tertiary-care pediatric hospital in Mexico City. Isolates werescreened for antibiotic resistance, including HLGR. High-level, gentamicin-resistant

E.faecalis

strains were selected for pulsed-field electrophoresis (PFGE) typing and plasmidanalysis. Transferability of resistance markers was carried out using filter matings.

Results.

Seventy-six percent of isolates were

E. faecalis,

10% were

E. avium,

5.2%

E.faecium

, 5.2%

E. raffinossus,

1.38%

E. malodoratus,

0.6%

E. hirae,

and 0.6%

E. cas-seliflavus.

Antimicrobial resistance was ampicillin and penicillin 29%, imipenem 17%,and vancomycin 3%, HLGR 5%. The following 15 high-level, gentamicin-resistant iso-lates were identified: six

E. faecalis;

four

E. avium;

three

E. faecium,

and two

E. cas-seliflavus.

Five of the six

E. faecalis

isolates were different by PFGE and transferred gen-tamicin and streptomycin resistance on filter membranes. Transfer frequencies ranged

from 8.2

3

10

2

4

to 6.92

3

10

2

5

transconjugants/recipient cell. The plasmid content of do-nors and transconjugants were homogeneous (one plasmid of 47 kb).

Conclusions.

In this pediatric hospital, antimicrobial resistance in

Enterococcus

spp. iscommon. Frequency of high-level, gentamicin-resistant strains is low. Mechanism ofHLGR appears to be due to a single plasmid dissemination. © 2001 IMSS. Published byElsevier Science Inc.

Key Words:

Enterococci, Pediatric patients, High-level gentamicin resistance, HLGR.

Introduction

Enterococci are significant nosocomial pathogens, capablealso of causing a variety of community-acquired infections(1,2). Since 1994, several studies have documented the in-creasing importance of this pathogen in hospitalized children(3). Enterococci are intrinsically resistant to different com-

Address reprint requests to: Guadalupe Miranda, M.D., Unidad deInvestigación en Epidemiología Hospitalaria, Hospital de Pediatría, CentroMédico Nacional Siglo XXI, IMSS, Av. Cuauhtémoc 330, Col. Doctores,06720 México, D.F., Mexico. Tel.: (

1

525) 627-6900, exts. 3322, 3330;FAX: (

1

525) 627-6949; E-mail:guadalumiranda@infosel.net.mx

160

Miranda et al./ Archives of Medical Research 32 (2001) 159–163

monly used antimicrobial agents (4–6). Their relative resis-tance to penicillin and other beta-lactam agents, includingcephalosporins, is related to unique enterococcal penicillin-binding proteins that allow synthesis of the cell wall (7–9).

High-level resistance to gentamicin (HLGR) and otheraminoglycosides (4,6,10,11), which eliminates synergistickilling of enterococci, was first reported in 1979 (12). Thishigh-level resistance is mediated by the following aminogly-coside-modifying enzymes: phosphotransferases [APH(2

0

),APH(3

9

]; nucleotidyltransferases [ANT(6), ANT(4

9

)], andacetyltransferases [AA(6

9

)]. The corresponding genes arelocated on transferable plasmids (12–14), with the excep-tion of the chromosomally encoded AAC(6

9

) of

E. faecium.

In the presence of HLGR, the recommendation is to not useaminoglycoside in combination with a beta-lactam agent forthe treatment of severe infections. Glycopeptides (vanco-mycin and teicoplanin) are useful alternatives for the treat-ment of infections caused by high-level, gentamicin-resis-tant enterococci, but resistance to vancomycin appeared in1988. High-level penicillin resistance has also emerged andis on the increase. Strains resistant to most available antibi-otics are now being isolated from infected children andadults (15–17). New agents for the treatment of seriousGram-positive bacterial infections have been under devel-opment since 1998; among them, quinupristin/dalfopristin(streptogramin agent) and linezolid (oxazolidinone) havebeen approved by the Federal and Drug Administration(FDA) for clinical use in the U.S. (18).

Different patterns of resistance have been reported frommany countries. The frequency has not been well studied inMexico, particularly in the pediatric population.

The objective of this study was to report the frequency ofantimicrobial resistances in enterococci and to determine thepresence and characteristics of HLGR plasmids in high-level,gentamicin-resistant

Enterococcus faecalis

clinical isolates.

Materials and Methods

All enterococcal isolates were collected from a tertiary-carepediatric hospital in Mexico City during an 18-month pe-riod. Isolates were identified as enterococci by using bile es-culin agar and growth in 6.5% NaCl. Species identificationwas carried out using the automated system Vitek (Bio-Merieux, Hazelwood, MO, USA) and the conventional testscheme described by Facklam (19). A total of 289 isolateswas collected: 38% were isolated from urine, 30% fromcatheter tips, 13.8% from surgical wounds, 11.8% fromblood, 3.1% from peritoneal fluid, 2.1% from cerebrospinalfluid, and 3.3% from others. Susceptibility testing was per-formed with agar dilution method on Mueller-Hinton agaraccording to the standards of the National Committee forClinical Laboratory Standards (NCCLS) (20) with serialtwofold dilutions of the following antimicrobials: ampicil-lin, penicillin, and imipenem (0.125–64

m

g/mL); chloram-phenicol (0.5

2

64

m

g/mL), and vancomycin (0.25 –64

m

g/

mL). Bacterial suspensions were adjusted to 0.5 McFarlandturbidity standard and 0.002 mL were delivered with aSteers replicator. The inoculum was allowed to absorb intothe agar prior to aerobic incubation at 35

8

C for 18 h. Sus-ceptibility breakpoints (ampicillin, penicillin, and imipenem16

m

g/mL, vancomycin 32

m

g/mL, chloramphenicol 32

m

g/mL) followed NCCLS (20) recommendations for entero-cocci. The reference strain

E. faecalis

ATCC 29212 wasused as control. Mueller-Hinton agar containing gentamicin500

m

g/mL and streptomycin 2,000

m

g/mL was used toscreen for HLGR. Isolates detected as aminoglycoside resis-tant by agar dilution were confirmed by broth tube dilutionusing brain-heart infusion broth with concentrations of gen-tamicin (500 and 1,000

m

g/mL) and streptomycin (1,000

m

g/mL).

E. faecalis

strains with a gentamicin minimal inhibitoryconcentration (MIC) of 500

m

g/mL were considered ashigh-level, gentamicin-resistant and were selected for fur-ther study. To corroborate that all isolates were different,typing by pulsed-field gel electrophoresis (PFGE) of ge-nomic DNA was performed according to the procedure ofMurray et al. (21). DNA digested with

Sma

I (Roche Diag-nosis, Mannheim, Germany) and restriction fragments wereseparated using a contour-clamped homogeneous field elec-trophoresis system (CHEF-DRIII, BioRad, Hercules, CA,USA). Electrophoresis was carried out at 13

8

C for 21 h at200 V with switch times ramped from 5–35 sec. The gelswere stained with ethidium bromide and photographed un-der ultraviolet light. Genomic patterns that were different bymore than seven fragments were considered unrelated, ac-cording to criteria established by Tenover et al. (22). Ifidentical patterns were found, only one isolate of each pat-tern was included.

Transferability of resistance markers was done using fil-ter matings as described by Forbes and Schaberg (12,23)with the following modifications (24):

Enterococcus faeca-lis

JH2-2 (rifampin- and fusidic acid-resistant) was used asthe susceptible plasmid-free recipient strain. Incubationswere performed at 37

8

C for 24 h. Donors and transconju-gants were tested for erythromycin, tetracycline, streptomy-cin, gentamicin, chloramphenicol, and for beta-hemolysisby using 5% horse blood agar plate. Transfer frequencieswere expressed as the number of transconjugants per recipi-ent cell present at the time of plating on selective media.

We isolated the plasmid DNA by the procedure ofAnderson and McKay (25). The molecular size of the puri-fied plasmid DNA was determined (23) in transconjugantsor donors by agarose gel electrophoresis of the undigestedplasmid DNA. The plasmid DNA was digested with

Hind

III and

Eco

RI.

Results

A total of 289 enterococcal isolates were collected and ana-lyzed. Most isolates corresponded to

E. faecalis

(76.1%).

Miranda et al. / Archives of Medical Research 32 (2001) 159–163

161

The remaining isolates were

E. avium

10%,

E. faecium

5.2%,

E. raffinosus

5.2%,

E. hirae

1.4%,

E. malodoratus

1.4%, and

E. casseliflavus

0.6%. Resistance rates to the an-timicrobials were ampicillin and penicillin 29%, imipenem17%, and vancomycin 3%. No isolate had intermediate re-sistance to vancomycin. Resistance rate according to differ-ent species is shown in Table 1; minimal inhibitory concen-trations are shown in Table 2.

During the study period, 15/289 (5.1%) high-level, gen-tamicin-resistant isolates were identified. Source of isolatesincluded urinary tract (nine), surgical wound (one), blood(one), cerebrospinal fluid (one), and catheter tip (three). Ofthe 15 isolates with HLGR, six were

E. faecalis,

four

E.avium

, three

E. faecium,

and two,

E. casseliflavus.

The six isolates selected for further evaluation withPFGE and filter matings were

E. faecalis.

All were resis-tant to high levels of streptomycin (

.

1,000

m

g/mL) andgentamicin (

.

500

m

g/mL, and

.

1,000

m

g/mL); nonewere resistant to vancomycin. One of the six strains was

b

-hemolytic.Of the six high-level, gentamicin-resistant

E. faecalis,

we found five different genomic patterns by PFGE; two iso-lates were identical (Figure 1). The five isolates with a dis-tinct PFGE pattern were included in the transfer studies.

Altogether, the five

E. faecalis

isolates transferred gen-tamicin and streptomycin resistance on filter membranes.Transfer frequencies ranged from 3.3

3

10

2

4

to 6.9

3

10

2

5

transconjugants/recipient cell. All isolates were susceptibleto erythromycin and resistant to tetracycline and chloram-phenicol. The five

E. faecalis

transconjugants and donorscontained only one plasmid (47 kb).

Eco

RI and

Hind

III re-

striction-digestion patterns of plasmid DNA were identical;the latter is shown in Figure 2. When comparing antibioticresistance markers (donor and transconjugant), agarosegel electrophoresis, and transfer properties on filter mem-branes, the plasmid content of donors and transconjugantswas homogeneous.

Discussion

There are few reports in Mexico concerning the species andantimicrobial susceptibility of

Enterococcus

spp. Some dif-ferences have been observed between a recent report froman adult tertiary-care center (26) and our pediatric hospital;we had a low frequency of

E. faecium

(5.2 vs. 15%), resis-tance to beta-lactams was very high (29 vs. 14%), and high-level gentamicin and streptomycin resistance was lower (5vs. 12% and 5 vs. 47%, respectively).

Table 1.

Antimicrobial resistance among 289 strains of

Enterococcus

spp.

Antimicrobial

E. faecalis

(

n

5

220)

E. avium

(

n

5

29)

E. faecium

(

n

5

15)

E. raffinosus

(

n

5

15)Other

a

(

n

5

10)

Ampicillin 13.1% 24% 46.6% 53.3% 28%Penicillin 7.2% 24% 53.3% 80% 36%Imipenem 12.7% 25.5% 53.3% 46.6% 16%Vancomycin 1.3% 13.7% 13.3% 0% 0%HLGR

b

2.7% 13.7% 20% 0% 20%

a

Other

5

E. hirae

(4)

, E. malodoratus

(4)

, E. casseliflavus

(2);

b

HLGR

5

high-level gentamicin resistance.

Table 2.

Minimum inhibitory concentrations (MIC) of 289

Enterococcus

isolates to four antimicrobial agents

AntimicrobialSusceptibility

breakpoint (mg/L)

E. faecalis

(220)

1MIC50 11MIC90

E. avium(29)

E. faecium(15)

E. raffinosus (15)

E. hirae(4)

E. malodoratus(4) and E. casseliflavus (2)

Ampicillin $16 1 32 1 32 16 .32 8 .32 1 1 1 .32Penicillin $16 0.5 2 1 8 1 .32 8 .32 1 1 1 8Imipenem $4 2 16 2 16 1 .32 4 .32 0.5 1 2 2Vancomycin $32 0.5 2 0.5 .32 0.5 4 .0.5 2 1 2 ,0.5 ,0.5

1MIC50 and 11MIC90 5 minimum inhibitory concentrations for 50 and 90% of the isolates.

Figure 1. DNA genomic patterns of E. faecalis digested by Sma I: lane 1,standard lambda ladder; lanes 2, 3, 6, and 7, different genomic patterns,and lanes 4 and 5, identical isolates.

162 Miranda et al./ Archives of Medical Research 32 (2001) 159–163

High-level, gentamicin-resistant isolates are endemic insome regions of the U.S., Thailand, Italy, Chile, Japan, andthe U.K. (1). Isolates with HLGR manifest high-level resis-tance to netilmycin, kanamycin, tobramycin, and amikacin,and infections caused by these microorganisms are difficultto treat (27). Gentamicin resistance appears on a variety ofdifferent conjugative and nonconjugative plasmids in E.faecalis (13,14). One study revealed that gentamicin-resis-tant plasmids from diverse geographic areas in the U.S.were heterogeneous (24), and in another study a commonplasmid was found in 15 isolates with HLGR; similar PFGEtyping suggested clonal dissemination, although there wasno direct evidence for patient-to-patient spread of thesestrains (28). In our study, the plasmids were homogeneousand all transfer gentamicin and streptomycin resistance tothe transconjugants, suggesting the transmission of a singleplasmid. E. faecalis isolates in this report were different byPFGE, eliminating genomic relatedness of the strains.

The rapid emergence of resistant enterococci, especiallyin the hospital setting, can be attributed to the fact that theseorganisms are nosocomial pathogens; thus, they are exposedto multiple antibiotics in hospitals throughout the world. In-trahospital spread may explain the progressive increase ofresistant enterococci (29); interhospital spread has also oc-curred between geographically related and non-related hos-pitals (2,30). The frequency of high-level aminoglycoside

resistance is also increasing; during 1997 the SENTRY An-timicrobial Surveillance Program reported high-level ami-noglycoside resistance rates of 48–66.7% in bacterial patho-gens causing bloodstream infections in the U.S., Canada,and Latin America (31,32). Detection of endemic high-level, gentamicin-resistant enterococci could be an evidenceof cross-transmission (33).

In this study, we found low vancomycin resistance fre-quency (3%); however, glycopeptide resistance has beenpredominant in the U.S., with 20% of vancomycin-resistantnosocomial strains of Enterococcus spp. (31). This problemhas somewhat diverted attention from aminoglycoside resis-tance. A susceptibility surveillance system to detect vanco-mycin resistance as well as high-level aminoglycoside resis-tance is necessary to modify antimicrobial-use policies inhospitals. Nosocomial spread of resistant strains may be re-duced by reinforcing proper antimicrobial use and strict in-fection control measures.

AcknowledgmentsThis project was sponsored in part by the collaborative programbetween the University of Texas Health Science Center at San An-tonio (UTHSCSA) and the Coordinación de Investigación Médica,Instituto Mexicano del Seguro Social (IMSS), Mexico City.

References1. Moellering RC Jr. Emergence of Enterococcus as a significant patho-

gen. Clin Infect Dis 1992;14:1175.2. Patterson JE, Sweeney AH, Simms M, Carley N, Mangi R, Sabetta J,

Lyons RW. An analysis of 110 serious enterococcal infections. Epide-miology, antibiotic susceptibility and outcome. Medicine 1995;74:191.

3. Christie C, Hammond J, Reising S, Patterson JE. Clinical and molecu-lar epidemiology of enterococcal bacteremia in a pediatric teachinghospital. J Pediatr 1994;125:392.

4. Gordon S, Swenson JM, Hill BC, Pigott NE, Facklam RR, Cooksey R,Thornsberry C, Enterococcal Study Group, Jarvis WR, Tenover FC.Antimicrobial susceptibility patterns of common and unusual speciesof enterococci causing infections in the United States. J Clin Microbiol1992;30:2373.

5. Hoffman SA, Moellering RC Jr. The enterococcus “putting the bug inour ears”. Ann Intern Med 1987;106:757.

6. Murray BE. The life and times of the Enterococcus. Clin Microb Rev1990;3:46.

7. Moellering RC Jr, Weinberg AN. Studies of antibiotic synergismagainst enterococci. II. Effect of various antibiotics on the uptake of 14C-labelled streptomycin by enterococci. J Clin Invest 1971;50:2580.

8. Moellering RC Jr, Wennersterns C, Weinberg AN. Synergy of penicil-lin and gentamicin against enterococci. J Infect Dis 1971;124(Suppl124):5207.

9. Standiford HD, De Maine JB, Kirby M. Antibiotic synergism of en-terococci. Relation to inhibitory concentrations. Arch Intern Med1970;126:255.

10. Morris JG, Shay DK, Hebden JN, McCarter RJ Jr, Perdue BE, JarvisWR, Johnson JA, Dowling TC, Polish LB, Schwalbe RS. Enterococciresistant to multiple antimicrobial agents including vancomycin. AnnIntern Med 1995;123:250.

11. Watanakunakorn CH. Rapid increase in the prevalence of high-levelaminoglycoside resistance among enterococci isolated from blood cul-tures 1989–1991. J Antimicrob Chemother 1992;30:289.

Figure 2. Agarose gel electrophoresis of Hind III of purified digested plas-mid DNA, lanes 2–5, identical restriction profile. Marker DNA 1 kb wasrun on the first and last lanes.

Miranda et al. / Archives of Medical Research 32 (2001) 159–163 163

12. Horodniceanu T, Bougueleret L, El-Solh N, Beith G, Delbos F. High-level, plasmid-borne resistance to gentamicin in Streptococcus faecalissubsp. zymogenes. Antimicrob Agents Chemother 1979;16:686.

13. Forbes BA, Schaberg DR. Transfer of resistance plasmids from Sta-phylococcus epidermidis to Staphylococcus aureus: evidence for con-jugative exchange of resistance. J Bacteriol 1983;153:627.

14. Sham DF, Gilmore MS. High-level resistance among enterococci. DevBiol Stand 1995;85:99.

15. Green M. Vancomycin resistant enterococci: impact and managementin pediatrics. Adv Pediatr Infect Dis 1997;13:257.

16. Rice LB, Shlaes DM. Vancomycin resistance in the enterococcus. Rel-evance in pediatrics. Pediatr Clin North Am 1995;42:601.

17. Tucci V, Haran HA, Isenberg HD. Epidemiology and control of vanco-mycin-resistant enterococci in an adult and children’s hospital. Am JInfect Control 1997;25:371.

18. Plouffe JF. Emerging therapies for serious Gram-positive bacterial in-fections: a focus on linezolid. Clin Infect Dis 2000;31(Suppl 4):S144.

19. Facklam RD, Sham DF, Teixeria LM. Enterococcus. In: Murray PR,Baron EJ, Pfaller MA, Tenover FC, Yolken RH, editors. Manual of clinicalmicrobiology. 7th ed. Washington, D.C., USA: ASM Press;1999. p. 297.

20. National Committee for Clinical Laboratory Standards. Methods fordilution antimicrobial susceptibility tests for bacteria that grow aerobi-cally; Approved standard—5th ed. Villanova, PA, USA: NCCLS doc-ument M7-A5;2000. p. 27.

21. Murray BE, Shing KV, Heat JH, Sharma BR, Weinstock GM. Com-parison of genomic DNAs of different enterococcal isolates using re-striction endonucleases with infrequent recognition sites. J Clin Micro-biol 1990;28:2059.

22. Tenover FC, Arbeit RD, Goering RV, Mickelson PA, Murray BE,Persing DH, Swaminathan R. Interpreting chromosomal DNA restric-tion patterns produced by pulsed-field gel electrophoresis. Criteria forbacterial strain typing. J Clin Microb 1995;33:2233.

23. Schaberg DR, Tompkins LS, Falkow S. Use of agarose gel electro-phoresis of plasmid deoxyribonucleic acid to fingerprint Gram-nega-tive bacilli. J Clin Microbiol 1981;13:1105.

24. Patterson JE, Masecar BK, Kauffman CA, Schaberg DR, HierholzerWJ, Zervos MJ. Gentamicin resistance plasmids of enterococci from di-verse geographic areas are heterogeneous. J Infect Dis 1988;158:212.

25. Anderson DG, McKay LL. Simple and rapid method for isolating largeplasmid DNA from lactic streptococci. Appl Environ Microbiol1983;46:549.

26. Sifuentes-Osornio J, Ponce-de-León A, Muñoz-Trejo T, Villalobos-Zapata Y, Ontiveros-Rodríguez C, Gómez-Roldán C. Antimicrobialsusceptibility patterns and high-level gentamicin resistance among en-terococci isolated in a Mexican tertiary care center. Rev Invest Clin1996;48:91.

27. Jones RN, Sader HD, Erwin ME, Anderson S. Emerging multiple resis-tant enterococci among clinical isolates. I. Prevalence data from a 97medical center surveillance study in the United States. EnterococcusStudy Group. Diagn Microbiol Infect Dis 1995;21:85.

28. Antalek MD, Mylotte JM, Lesse AJ, Sellick JA Jr. Clinical and molec-ular epidemiology of Enterococcus faecalis bacteremia, with specialreference to strains with high-level resistance to gentamicin. Clin In-fect Dis 1995;20:103.

29. Shay DK, Maloney SA, Montecalvo M, Banerjee S, Wormser GP, Ar-dunio MJ, Bland LA, Jarvis WR. Epidemiology and mortality risk ofvancomycin-resistant enterococcal bloodstream infections. J Infect Dis1995;172:993.

30. Tomayko JF, Murray BE. Analysis of Enterococcus faecalis isolatesfrom intercontinental sources by multilocus enzyme electrophoresisand PFGE. J Clin Microbiol 1995;33:2903.

31. Pfaller MA, Jones RN, Doern GV, Kugler K, and the SENTRY Parti-cipants Group. Bacterial pathogens isolated from patients with blood-stream infection: frequencies of occurrence and antimicrobial suscepti-bility patterns from the SENTRY Antimicrobial Surveillance Program(United States and Canada, 1997). Antimicrob Agent Chemother1998;42:1762.

32. Sader HS, Pfaller MA, Jones RN, Doern GV, Gales AC, Winokur PL,Kugler KC. Bacterial pathogens isolated from patients with blood-stream infections in Latin America, 1997: frequency of occurrence andantimicrobial susceptibility patterns from the SENTRY AntimicrobialSurveillance Program. Braz J Infect Dis 1999;3:97.

33. van den Braak N, van Belkum A, Kreft D, te Witt R, Verbrug HA,Endtz H. The prevalence and clonal expansion of high-level gentami-cin-resistant enterococci isolated from blood cultures in a Dutch uni-versity hospital. J Antimicrob Chemother 1999;44:795.