Emergence of Vancomycin-Intermediate Staphylococcus aureus and S. sciuri , Greece

Journal of Antimicrobial Chemotherapy (2002) 50, 383–391DOI: 10.1093/jac/dkf142

383. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

© 2002 The British Society for Antimicrobial Chemotherapy

Emergence of vancomycin-intermediate Staphylococcus aureus in a Belgian hospital: microbiological and clinical features

Olivier Denis1*, Claire Nonhoff1, Baudouin Byl2, Christiane Knoop3, Sophie Bobin-Dubreux4 and

Marc J. Struelens1

1Service de Microbiologie, 2Unité d’Epidémiologie et d’Hygiène Hospitalière and 3Service de Pneumologie, Hôpital Erasme, Université Libre de Bruxelles, 808 route de Lennik, 1070 Brussels, Belgium;

4Hôpital Edouard Herriot, Université Claude Bernard, Lyon, France

Received 13 May 2002; accepted 10 June 2002

In 1999, all clinical Staphylococcus aureus isolates from patients admitted to a BelgianUniversity hospital were tested for decreased vancomycin susceptibility. Three vancomycin-intermediate Staphylococcus aureus (VISA) and four hetero-VISA strains were detected among2145 isolates tested. They emerged from strains that belonged to locally endemic methicillin-resistant S. aureus (MRSA) genotypes in three patients who had received repeated courses ofvancomycin therapy. A cystic fibrosis patient with MRSA/VISA-associated broncho-pulmonaryexacerbation was successfully treated by continuous vancomycin infusion plus fusidic acidfollowed by oral minocycline–fusidic acid. Two other patients had VISA recovered from speci-mens of undetermined clinical significance. Emergence of VISA variants of endemic MRSAstrains in Belgium warrants active microbiological surveillance and careful monitoring ofvancomycin therapy. Therapy with high-dose vancomycin administered by continuous infusionin combination with other antimicrobials may be a therapeutic option worth investigating forVISA infection.

Keywords: Staphylococcus aureus, VISA, Belgium, vancomycin

Introduction

Staphylococcus aureus is a major pathogen causing a divers-ity of infections including bacteraemia, pneumonia, skin, softtissue and osteo-articular infections.1 In the past two decades,the prevalence of methicillin-resistant S. aureus (MRSA)among both nosocomial and community-acquired infectionshas increased throughout the world.2,3 In Belgium, S. aureusbacteraemia was due to MRSA in 22% of cases in 1999.3

Many MRSA strains are co-resistant to other antibioticclasses, including macrolides, aminoglycosides and fluoro-quinolones. Until recently, glycopeptides were the last uni-formly effective drugs available for treatment of MRSAinfections.

Since 1997, infections caused by MRSA strains with inter-mediate susceptibility to vancomycin (VISA) (MIC 8–16 mg/L)

have been reported from Japan, France, the United States,Korea and Germany.4–10 These strains were recovered frompatients who failed therapy with vancomycin that had oftenbeen used for prolonged periods of time. Other strains, namedhetero-VISA, appear to be borderline susceptible to vanco-mycin (MIC 2–4 mg/L) but exhibit low-level subpopulations(10–6 cells) able to grow at vancomycin concentrations of4–8 mg/L.11 Such strains have been described in Europe, Asiaand Brazil.12–20 Hetero-VISA strains could represent first-stepmutants that are precursors of VISA strains in patients receiv-ing prolonged courses of vancomycin. Both VISA and hetero-VISA strains belong to a restricted range of epidemic MRSAgenotypes.10–15,17–20

We performed a prospective study to assess the incidenceand characteristics of VISA strains recovered from patients

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

*Corresponding author. Tel: +32-2-555-45-18; Fax: +32-2-555-31-10; E-mail: [email protected]

by guest on October 2, 2015

http://jac.oxfordjournals.org/D

ownloaded from

http://jac.oxfordjournals.org/

O. Denis et al.

384

admitted to a Belgian University hospital and we report theclinical course of VISA infections.

Materials and methods

Definitions

VISA was defined as an S. aureus strain with: (i) reproduciblegrowth on brain–heart infusion (BHI) agar containing 6 mg/Lvancomycin; (ii) a vancomycin MIC > 4 mg/L by Etest onMueller–Hinton (MH) agar and by broth microdilution; (iii) apopulation analysis profile similar to homogeneous VISAcontrol strains Mu50 and HIP5827.4,6

Hetero-VISA was defined as an S. aureus strain with: (i) avancomycin MIC ≤ 4 mg/L by Etest on MH agar or by brothmicrodilution; (ii) a population analysis profile similar tohetero-VISA strain Mu3 with detectable subpopulation grow-ing at 4 mg/L vancomycin.11

Bacterial strains

Between 1 January 1999 and 31 December 1999, all clinicalisolates (n = 2145) of S. aureus from patients admitted toErasme University hospital, an 858 bed tertiary care teachinghospital, were studied. Identification of S. aureus was per-formed by coagulase test with human plasma. MRSA strainswere confirmed by multiplex PCR for mecA and nuc genes asdescribed previously.21

Antimicrobial susceptibility testing

All isolates were tested by a vancomycin screen agarmethod.22 Briefly, 10 µL of 0.5 McFarland suspension of thefirst pure subculture was inoculated on to BHI agar supple-mented with 6 mg/L vancomycin (BBL, Becton Dickinson,USA). Isolates showing growth after 24 or 48 h on thismedium were subcultured on BHI agar supplemented with1 mg/L vancomycin before determination of the MIC ofvancomycin.

The MIC of vancomycin was determined by a brothmicrodilution method according to NCCLS guidelines and bytwo Etest methods (AB Biodisk, Solna, Sweden). In the firstprotocol, 3 mL of McFarland 0.5 suspension was flooded onto MH agar and incubated for 24 h at 35°C.23 In the second,BHI agar was inoculated with 100 µL of a McFarland 2 sus-pension and incubated for 48 h at 35°C.24 MICs were inter-preted according to NCCLS breakpoints.

Strains with vancomycin MICs > 4 mg/L were character-ized by population analysis.11 Briefly, 100 µL of an overnightsuspension (McFarland 2) was spread on to agar plates sup-plemented with 0, 2, 4 and 8 mg/L vancomycin. The numberof colonies was counted after incubation for 48 h at 35°C.

Each assay included the following control strains: vanco-mycin-susceptible S. aureus strain ATCC 29213, hetero-VISA strain Mu3 and VISA strains Mu50 and HIP5827.4,6,11

Susceptibility to other antimicrobials was determined bya disc diffusion method (Rosco, NeoSensitab, Taastrup,Denmark) using interpretative inhibitory zone size accordingto NCCLS breakpoints. The antibiotics tested included peni-cillin G (P), rifampicin (R), erythromycin (E), clindamycin(Cl), minocycline (M), doxycycline (D), fusidic acid (F),gentamicin (G), amikacin (A), ciprofloxacin (C), co-trimox-azole (SXT) and mupirocin (M). Oxacillin (O) was tested onoxacillin screen agar (6 mg/L) according to NCCLS recom-mendations. Oxacillin MICs were determined by Etest onMH agar supplemented with 2% NaCl incubated for 24 h at35°C.

VISA strains were examined by multiplex PCR for thepresence of vanA, vanB and vanC-1/2/3 genes coding forvancomycin resistance in enterococci.25

Electron microscopy

All cultures were grown to an OD600 of 0.6 in BHI broth priorto processing for electron microscopic examination, asdescribed by Mani et al.26 The strains were cultured at 37°C in10 mL of BHI broth in two series of tubes containing for eachstrain either no antibiotic or 2 mg/L vancomycin. Cells wereharvested in the post-exponential growth phase (18–24 h→OD600 of 0.6) and fixed in 2% glutaraldehyde in 0.05 Mcacodylate buffer (pH 7.2)–MgSO4 0.05% for 1.5 h, thenrinsed twice in cacodylate buffer 0.1 M. They were treatedwith 1% osmium tetroxide in 0.2 M cacodylate buffer for 2 hat 4°C. Cells were dehydrated with graded concentrations ofethanol and embedded in Epon A+B. Ultra-thin sections werestained with uranyl acetate and lead citrate, then examinedwith a transmission electron microscope (TEM) (JEOL1200MX) at different magnifications and photographed.

Pulsed-field gel electrophoresis (PFGE)

Macrorestriction (SmaI) and PFGE analysis were performedas described previously.21 SmaI patterns were normalized andcompared using the Dice coefficient and UPGMA clusteringmethod with BioNumerics software version 1.0 (AppliedMaths, Ghent, Belgium). Patterns differing by one to six DNAfragments were considered as subtypes (designated by a lettersuffix) and those distinguished by seven or more DNAfragments were labelled as distinct types (designated bynumeral).21

Infection control and epidemiological investigation

The isolation procedures for patients colonized with MRSAfollowed Belgian national guidelines.27 Briefly, carriers were



ownloaded from


Emergence of VISA in a Belgian hospital

385

placed in single rooms. Healthcare workers wore disposablegloves and gowns before entering the patient’s room. In addi-tion, they wore face masks for aerosol-generating procedures.Hand disinfection with hydro-alcoholic solution was under-taken after glove removal. In addition, patients with VISAcolonization/infection were cared for by dedicated nursingstaff. Compliance with the isolation procedures was strictlymonitored by infection control personnel. Patients and health-care workers in contact with patients colonized with VISAwere screened for colonization to assess potential transmis-sion of VISA. Swabs obtained from anterior nares and handswere inoculated on to sheep blood agar and into MRSAselective broth medium containing colistin (32 mg/L), oxacil-lin (2 mg/L) and 2.5% NaCl. S. aureus isolates were identifiedby the coagulase test and tested for susceptibility to oxacillinand vancomycin by agar screen methods.

Results

Vancomycin susceptibility

Of 2145 isolates tested, of which 881 (41.1%) were oxacillinresistant, 146 (4.9%) S. aureus (of which 107 were MRSA)isolated from 106 patients grew on vancomycin screen agarafter 24 or 48 h (Figure 1). By the heavy inoculum Etestprocedure on BHI agar, 73 (50%) of these strains showeddecreased susceptibility to vancomycin (MIC 6–16 mg/L).By Etest with a standard inoculum on MH agar, only fourstrains showed decreased susceptibility to vancomycin (MIC6–8 mg/L). By the heavy inoculum Etest method on BHI agar,the distribution of MICs for the 146 strains showed a four-foldgreater modal MIC value (8 versus 2 mg/L; P < 0.001) thanthat determined by standard inoculum on MH agar. By brothmicrodilution, four strains showed intermediate susceptibilityto vancomycin (MIC 8–16 mg/L) (Table 1). No VISA strainscarried van resistance genes.

Population analysis indicated that only the three strainswith intermediate MIC by Etest MH agar and microdilution(P1V44, P2V136, P3V156) had a vancomycin resistance pro-file similar to homogeneous VISA reference strains Mu50and HIP5827 (Table 1 and Figure 2). Four additional isolates,P1V69, P4V144, P5V133 and P6V6419, had a populationprofile overlapping with heterogeneous VISA referencestrain Mu3 (Table 1 and Figure 2).

Proportion of VISA isolates

The proportion of hetero-VISA strains was 0.1% of S. aureusand 0.4% of MRSA strains, whereas the proportion of VISAstrains was 0.1% of S. aureus and 0.3% of MRSA strains.

Antimicrobial susceptibility of VISA and hetero-VISA strains

All strains with reduced susceptibility to vancomycin posses-sed the mecA gene and exhibited multiple resistance to otherantimicrobials (Table 1). Hetero-VISA and two VISA strains,P2V136 and P3V156, were susceptible to minocycline, doxy-cycline, fusidic acid and co-trimoxazole (Table 1). VISAP1V44 strain was susceptible to clindamycin, gentamicin,minocycline, doxycycline, fusidic acid, co-trimoxazole andrifampicin. This isolate was borderline susceptible to oxacil-lin (MIC 2 mg/L) but possessed the mecA gene by PCR. ThisVISA strain derived from a high-level oxacillin-resistantMRSA (MIC > 256 mg/L) progenitor strain P39575 isolatedpreviously from patient 1.

PFGE analysis

By PFGE, VISA and hetero-VISA strains belonged to eitherof the two multidrug-resistant MRSA types A and D, both ofwhich are endemic in our hospital. These PFGE types weredistinct from VISA strains from Japan and the USA (Mu50and HIP5827). VISA isolate P1V44 and hetero-VISA isolateP1V69 recovered from patient 1 in 1999 were indistinguish-able by PFGE from MRSA strain P39575, a fully vancomycin-susceptible strain isolated during a previous hospitalization inDecember 1998.

TEM analysis

Three MRSA strains isolated consecutively from patient 1(vancomycin-susceptible MRSA strain P39575, hetero-VISA strain P1V69 and VISA strain P1V44) were examinedby TEM (Figure 3). MRSA strain P39575 and hetero-VISAstrain P1V69 showed normal cell wall thickness but anuneven cell surface, irregular cell shape and abnormal septa-tion (Figure 3c and d). These abnormalities were more pro-nounced in hetero-VISA strain P1V69. Strains P1V69 andFigure 1. Vancomycin susceptibility tests of 2145 S. aureus isolates.



ownloaded from


O. Denis et al.

386

P39575 showed increased cell wall thickness when grown in2 mg/L vancomycin. P1V44 strain showed a thickened cellwall (Figure 3b), irregular cell shape and abnormal septationin contrast to the thin and regular cell wall morphology ofS. aureus ATCC 29213 strain (Figure 3a). When grown in thepresence of vancomycin (2 mg/L), the abnormalities of strainP1V44 were slightly enhanced, whereas strain ATCC 29213showed no morphological change (not shown).

Case reports of VISA infection

Patient 1. An 18-year-old woman with cystic fibrosis,colonized with MRSA since 1993, presented three successivepulmonary exacerbations attributed to MRSA (strain P39575)and Pseudomonas aeruginosa during the winter 1998–1999.Each episode required hospital admission and intravenous(iv) treatment with a combination of vancomycin, ceftazid-

Table 1. Microbiological characteristics of VISA and hetero-VISA isolates, Erasme Hospital, 1999

Abbreviations: P, penicillin; E, erythromycin; Cl, clindamycin; A, amikacin; G, gentamicin; C, ciprofloxacin; R, rifampicin; F, fusidic acid; M, mupirocin.

MIC (mg/L)

vancomycin

oxacillinEtestPatient/strain

Month of isolation Origin

Etest MH agar

EtestBHI agar

broth microdilution

Population profile

Resistance pattern

PFGE pattern

1/P1V44 08/99 sputum 8 16 16 2 VISA PECAR A1a1/P1V69 09/99 sputum 3 6 4 >256 h-VISA PECARF A1a

2/P2V136 12/99 decubitus sore 8 16 4 >256 VISA PEClCGAM D4b

3/P3V156 03/99 broncheal swab 6 16 8 >256 VISA PEClCGARM D4a

4/P4V144 12/99 wound 3 6 8 >256 h-VISA PEClCGAM D4c

5/P5V133 12/99 sputum 6 8 4 >256 h-VISA PEClCGAM D4d6/P6V6419 02/99 sputum 3 6 8 >256 h-VISA PEClCGAM D4e

Figure 2. Population analysis of Belgian VISA and hetero-VISA strains and reference strains.



ownloaded from



387

ime and tobramycin. In September 1999, she was hospitalizedfor pulmonary exacerbation. Sputum showed Gram-positivecocci and culture grew P. aeruginosa and S. aureus (strainP1V44) that was susceptible to oxacillin and intermediatelysusceptible to vancomycin (Table 1 and Figure 1). Shereceived a 10 day course of iv oxacillin in combination withceftazidime and tobramycin. She was discharged after only amoderate improvement. When she was readmitted 2 weekslater for a new deterioration, sputum culture grew an MRSAstrain with a subpopulation hetero-resistant to vancomycin(strain P1V69). Intermittent vancomycin (2 g/day) wasadministered for 12 days in combination with anti-pseudo-monal treatment. After a transient improvement, she was againreadmitted a few days later. She received high doses (3 g/day)of vancomycin by continuous infusion, in combination withceftazidime and tobramycin for another 21 days. Two weeksafter discharge, she was again readmitted for the fourth timein 3 months. As a hetero-VISA strain was the only organismisolated again from a sputum sample that showed numerousGram-positive cocci on direct examination, the relapse wasattributed to persisting infection with S. aureus of decreasedvancomycin susceptibility. High doses of vancomycin(3 g/day) were administered by continuous infusion in com-bination with fusidic acid (1.5 g/day) for 11 days. Serum levelsof vancomycin ranged between 14.3 and 33.8 mg/L with amedian of 27.8 mg/L. The iv treatment was followed by oraltreatment with minocycline (200 mg/day) and fusidic acid(1.5 g/day) for 3 months. Her clinical progression was goodand she remained exacerbation-free for 8 months thereafter.During the next 24 months, only vancomycin-susceptibleMRSA strains were isolated from sputum.

Patient 2. A 54-year-old woman was admitted to the intensivecare unit (ICU) in November 1999, for multiple trauma with

hepatic and pulmonary contusions. On day 11, she developeda primary MRSA bacteraemia. She was given intermittentvancomycin for 11 days adapted to serum levels, whichwere all in the expected range (trough, 5–10 mg/L; peak 30–40 mg/L). She improved but fever and bacteraemia recurredon day 22. Rifampicin was added to intermittent vancomycinfor 10 more days. Because of the emergence of rifampicinresistance, rifampicin was then discontinued and replaced byiv fusidic acid (1.5 g/day). During this treatment (day 47), shepresented with a second recurrence of bacteraemia due to anMRSA strain that was susceptible to vancomycin. Concomit-antly, a VISA strain (P2V136) was isolated from a decubitusulcer. From day 56 to 112, she was given continuous infusionof vancomycin (2–3 g/day) in combination with fusidic acid(1.5 g/day). During continuous infusion, serum levels ofvancomycin ranged between 24.2 and 51.2 mg/L, with amedian of 33.3 mg/L. She presented on day 233 with a newepisode of bacteraemia due to an MRSA strain that was sus-ceptible to vancomycin (MIC 1 mg/L) attributed to pulmon-ary infection. Sepsis resolved rapidly under treatment withcontinuous vancomycin infusion combined with fusidic acid.She died suddenly on day 243 of a non-infectious cause.

Patient 3. A 46-year-old man underwent a liver transplant foralcoholic cirrhosis in January 1999. The post-operativecourse was complicated by splenic haemorrhage, acute renalfailure and graft rejection. On day 14, he developed a deep-seated surgical infection and bacteraemia due to MRSA.This strain was susceptible to vancomycin (MIC 3 mg/L).Intermittent vancomycin was administered from day 14 to21 and dosing was adapted to serum levels. His clinicalstate deteriorated and surgical drainage was performed onday 20. Treatment was altered thereafter to continuous ivinfusion of vancomycin (1.5 g/day) combined with rifampicin(900 mg/day). During continuous infusion, serum levels ofvancomycin ranged between 18.5 and 54.6 mg/L with amedian of 38 mg/L. The patient developed multiple organfailure and died 33 days after transplantation. Post-mortemexamination disclosed generalized cytomegalovirus infectionand lung abscess (4 cm diameter). A VISA strain (P3V156)was isolated from a post-mortem bronchial swab.

Epidemiological investigation and infection control intervention

Screenings were performed in the ICU and pulmonary wardafter patients 1 and 2 were identified. No VISA or hetero-VISA were detected among patients from those wards (n = 30)and healthcare workers (n = 69). Infection control proceduresas described above were applied to patients 1 and 2 duringtheir entire hospital stay. No transmission to patients orhealthcare workers occurred during their hospital stay.

Figure 3. Transmission electron micrographs of (a) S. aureus ATCC29213, (b) VISA strain P1V44, (c) hetero-VISA strains P1V69 and(d) MRSA P39575. Magnification ×60 000.



ownloaded from


O. Denis et al.

388

Discussion

Well-documented infections caused by VISA strains havebeen reported from Asia and the USA.4,6–9 To our knowledge,our study reports the first clinical descriptions of cases ofwell-documented VISA infection from Europe with a rigor-ous set of criteria. The cases reported previously from Europewere associated with VISA strains that were not confirmed bypopulation analysis or electron microscopy.5,10 Previous re-ports of VISA infection described patients who had receivedprolonged or repeated courses of vancomycin for MRSAinfection, often administered for several months.4,6–8 In manyof these patients, venous administration was irregular due torenal failure and, in some cases, vancomycin underdosingappeared as a contributing factor. In most cases, vancomycinfailed to cure the infection and, in four cases, shift to a com-bination of other anti-staphylococcal drugs was reported to besuccessful. However, the majority of these patients had manyrisk factors for treatment failure, including infection of in-dwelling devices, undrained abscesses, and underlying patho-logy such as cancer and end-stage renal failure. Likewise, theshift to a non-glycopeptide was concomitant with surgicaldrainage or removal of the infected indwelling device,therefore making interpretation of the vancomycin failureuncertain. In the patients studied here, the VISA strains alsoemerged following vancomycin therapy. In our study, onlypatient 1 was considered infected by a VISA strain. Thiscystic fibrosis patient had hetero-VISA and VISA variants ofMRSA that were isolated repeatedly from her sputum sam-ples. As in previous reports of vancomycin failure for VISAinfection, she showed only a transient response to intermittentor continuous high-dose vancomycin therapy. However, herpulmonary infection with VISA was treated successfully withhigh-dose continuous vancomycin in combination with fusidicacid, followed by suppressive oral treatment with a combina-tion of minocycline and fusidic acid. To our knowledge, this isthe first documented attempt to treat VISA infection withcontinuous infusion of high-dose vancomycin to achieve ablood level above MICs. However, as this therapy wasadministrated in combination with other anti-staphylococcalantibiotics it was not possible to conclude with certainty thatthe high-dose vancomycin was clinically effective. Furtherinvestigation of this therapeutic approach may be useful.

In contrast to the rarity of reported cases of VISAinfection, the proportion of S. aureus with heterogeneousexpression of reduced susceptibility to vancomycin rangeswidely between surveys worldwide, including those fromthe European region: Spain 56%, France 20%, Germany2–14%, Italy 1%, Japan 5–26%, Brazil 4%, England 0–16%,Greece 1%.10,11,13–15,17,18,20,24,28,29 The results from these stud-ies are difficult to compare because most investigations used aretrospective study design and many restricted their analysesto a limited number of strains, usually from selected groups of

patients. More importantly, definitions of VISA and methodsfor MIC determination of vancomycin differ markedlybetween investigators. For example, hetero-VISA strains arevariously defined by different authors as S. aureus strains thatexhibit a cell subpopulation growing at either 4 or 8 mg/Lvancomycin.11,13,24,30 Using more restricted criteria and in anunselected population, we observed a low incidence of 0.3%VISA and 0.4% hetero-VISA among the patients in ourhospital colonized or infected by MRSA.

The low-level resistance of VISA strains to vancomycin(MIC 8 mg/L) makes them difficult to detect by standardsusceptibility test methods.23 A reversible vancomycin resist-ance phenotype observed with some VISA strains may com-pound this difficulty.31 Tenover et al.23 have shown that thedisc diffusion method is inadequate to detect VISA strains. Inour study, we used BHI agar supplemented with 6 mg/Lvancomycin for VISA screening as recommended by theNCCLS, but with prolonged incubation (48 h) instead of therecommended 24 h.22 Using this modified technique, vanco-mycin agar screening had a low specificity in our study, since95% of strains that grew after 48 h on screen agar were notconfirmed as VISA. Other authors have reported artefactualgrowth of occasional vancomycin-susceptible S. aureus iso-lates on a vancomycin agar screen when using media thatwere prepared in-house.23,32 Our finding of low specificityusing this screening method is in contrast to the report ofWalsh et al.,33 who found it specific. The Etest method using aheavy inoculum on BHI agar with prolonged incubation alsoproved to be of low specificity in the present study (90% offalse positive). This was also reported by Aucken et al.29 but itis again at variance with the findings of Walsh et al.33 Inaddition to technical differences, these discrepancies betweenstudies could be explained by differences in study design. Ourstudy looked at consecutive S. aureus clinical isolates,whereas Walsh et al.33 examined a selected collection ofstrains referred for reduced susceptibility to glycopeptides.Hetero-VISA strains are even more difficult to detect thanVISA strains. They do not grow reproducibly on vancomycinscreen agar and appear to be susceptible by Etest on MH agarwith a standard inoculum and by microdilution MIC testing.The population analysis profile is at present the only reliableconfirmation method but is too time consuming for routineuse.

Many VISA and hetero-VISA strains are related to epi-demic gentamicin-resistant MRSA clones that are dissemin-ated in hospitals in Europe, Brazil and Japan.10–15,17,18,24,34 TheVISA and hetero-VISA isolates in the present study belongedto two different MRSA types that are endemic in our hospitaland unrelated to the US and Japanese VISA strains. One ofthese genotypes has been widely disseminated in westernEurope since the mid-1980s and has been reported in morethan 80% of Belgian hospitals.21 Outbreaks of S. aureus withreduced susceptibility to glycopeptides have been described



ownloaded from



389

in French hospitals among patients who had not receivedglycopeptides.30,35 It appears that these strains are hetero-VISA. In our study, the three patients with VISA infection orcolonization showed no space–time clustering. According tolocal policy, the patients were already cared for using MRSAisolation precautions, including placement in a private roomand use by healthcare personnel of gowns, gloves andmasks.27 When the VISA strains were identified, nursing staffwere increased in number for the care of VISA carriers toconsolidate compliance with the isolation precautionsaccording to the published recommendations.36

The mechanism of decreased susceptibility of S. aureus toglycopeptides is not yet well understood. Like other VISAstrains described previously, the VISA strains studied herelacked the vanA and vanB genes that are responsible forglycopeptide resistance in enterococci. The strains that weexamined by TEM showed the same abnormality of cell wallstructure that has been described in other strains.4,6–7,9,37

Strains Mu50 and Mu3 have an accelerated cell wall turnoverwith an increased proportion of glutamine non-amidatedmuropeptides and overexpression of penicillin-binding pro-teins 2 and 2′.38–40 These abnormalities suggest increasedproduction of D-Ala-D-Ala residues acting as false targets,which trap the antibiotic away from its lethal target site of cellwall synthesis adjacent to the membrane.38 Moreover, allVISA strains have a loss of detectable PBP4 activity, whichleads to a decrease in cell wall cross-linking.41 Interestingly,in one of the VISA clinical isolates detected in our study,strain P1V44, the decreased susceptibility to vancomycin wasaccompanied by a parallel loss of expression of oxacillinresistance. This phenomenon has been reported previously tooccur in vivo by Sieradzki et al.42 It was reproduced in vitro byselection of a mutant VISA strain from an MRSA strainexposed to increasing vancomycin concentrations.

In conclusion, our study describes the first cases of infec-tion and colonization with VISA and hetero-VISA strains inBelgium. These VISA strains emerged during vancomycintherapy for infections due to MRSA strains that are endemicin Belgian hospitals. Although these strains appear to be rare,they were associated with therapeutic difficulties, as reportedpreviously. Therefore, it is desirable to improve the methodsof routine detection of VISA strains to ensure their epidemio-logical surveillance. The optimal management of patientsinfected with VISA strains is not well defined. We concurwith the recommendation that isolation precautions used forcontrol of MRSA nosocomial transmission should be strictlyenforced. Therapy with high-dose vancomycin administeredby continuous infusion in combination with other anti-staphylococcal drugs may be a therapeutic option worthinvestigating in the case of VISA infection, but the removal ofany foreign body or undrained collections should always bethe first and foremost intervention.

Acknowledgements

We thank Keiichi Hiramatsu and Fred C. Tenover for kindlyproviding strains Mu3, Mu50 and HIP5827, Ariane Deplanoand Sylviane Rottiers for their expert technical assistance,and Frédérique Jacobs and Jérome Etienne for reviewing themanuscript.

References

1. Lowy, F. D. (1998). Staphylococcus aureus infections. NewEngland Journal of Medicine 339, 520–32.

2. NNIS System. (1999). National Nosocomial Infections Surveil-lance (NNIS) System report, data summary from January 1990–May 1999, issued June 1999. American Journal of Infection Control27, 520–32.

3. EARSS Management Team. (2000). Susceptibility test resultsof Staphylococcus aureus. EARSS Newsletter 2, 2–3.

4. Hiramatsu, K., Hanaki, H., Ino, T., Yabuta, K., Oguri, T. &Tenover, F. C. (1997). Methicillin-resistant Staphylococcus aureusclinical strain with reduced vancomycin susceptibility. Journal ofAntimicrobial Chemotherapy 40, 135–6.

5. Ploy, M. C., Grélaud, C., Martin, C., de Lumley, L. & Denis, F.(1998). First clinical isolate of vancomycin-intermediate Staphylo-coccus aureus in a French hospital. Lancet 351, 1212.

6. Smith, T. L., Pearson, M. L., Wilcox, K. R., Cruz, C., Lancaster,M. V., Robinson, B. et al. (1999). Emergence of vancomycin resist-ance in Staphylococcus aureus. New England Journal of Medicine340, 493–501.

7. Sieradzki, K., Roberts, R. B., Haber, S. W. & Tomasz, A.(1999). The development of vancomycin resistance in a patient withmethicillin-resistant Staphylococcus aureus infection. New EnglandJournal of Medicine 340, 517–23.

8. Centers for Disease Control and Prevention. (2000). Staphylo-coccus aureus with reduced susceptibility to vancomycin—Illinois,1999. Morbidity and Mortality Weekly Report 48, 1165–7.

9. Kim, M. N., Pai, C. H., Woo, J. H., Ryu, J. S. & Hiramatsu, K.(2000). Vancomycin-intermediate Staphylococcus aureus in Korea.Journal of Clinical Microbiology 38, 3879–81.

10. Bierbaum, G., Fuchs, K., Lenz, W., Szekat, C. & Sahl, H. G.(1999). Presence of Staphylococcus aureus with reduced suscept-ibility to vancomycin in Germany. European Journal of ClinicalMicrobiology and Infectious Diseases 18, 691–6.

11. Hiramatsu, K., Aritaka, N., Hanaki, H., Kawasaki, S., Hosoda,Y., Hori, S. et al. (1997). Dissemination in Japanese hospitals ofstrains of Staphylococcus aureus heterogeneously resistant tovancomycin. Lancet 350, 1670–3.

12. Howe, R. A., Bowker, K. E., Walsh, T. R., Feest, T. G. &MacGowan, A. P. (1998). Vancomycin-resistant Staphylococcusaureus. Lancet 351, 602.

13. Ariza, J., Pujol, M., Cabo, J., Pena, C., Fernandez, N., Linares,J. et al. (1999). Vancomycin in surgical infections due to methicillin-resistant Staphylococcus aureus with heterogeneous resistance tovancomycin. Lancet 353, 1587–8.



ownloaded from


O. Denis et al.

390

14. Geisel, R., Schmitz, F. J., Thomas, L., Berns, G., Zetsche, O.,Ulrich, B. et al. (1999). Emergence of heterogeneous intermediatevancomycin resistance in Staphylococcus aureus isolates in theDüsseldorf area. Journal of Antimicrobial Chemotherapy 43, 846–8.

15. Woodford, N., Warner, M. & Aucken, H. M. (2000). Vancomycinresistance among epidemic strains of methicillin-resistant Staphylo-coccus aureus in England and Wales. Journal of AntimicrobialChemotherapy 45, 258–9.

16. Wong, S. S., Ng, T., Yam, W., Tsang, D. N., Woo, P. C., Fung,S. K. et al. (2000). Bacteremia due to Staphylococcus aureus withreduced susceptibility to vancomycin. Diagnostic Microbiology andInfectious Disease 36, 261–8.

17. dos Santos Soares, M. J., da Silva-Carvalho, M. C., Ferreira-Carvalho, B. T. & Figueiredo, A. M. S. (2000). Spread of methicillin-resistant Staphylococcus aureus belonging to the Brazilian epidemicclone in a general hospital and emergence of heterogeneousresistance to glycopeptide antibiotics among these isolates. Journalof Hospital Infection 44, 301–8.

18. Kantzanou, M., Tassios, P. T., Tseleni-Kotsovili, A., Legakis,N. J. & Vatopoulos, A. C. (1999). Reduced susceptibility to vanco-mycin of nosocomial isolates of methicillin-resistant Staphylococcusaureus. Journal of Antimicrobial Chemotherapy 43, 729–31.

19. Trakulsomboon, S., Danchaivijitr, S., Rongrungruang, Y.,Dhiraputra, C., Susaemgrat, W., Ito, T. et al. (2001). First report ofmethicillin-resistant Staphylococcus aureus with reduced suscept-ibility to vancomycin in Thailand. Journal of Clinical Microbiology 39,591–5.

20. Marchese, A., Balistreri, G., Tonoli, E., Debbia, E. A. & Schito,G. C. (2000). Heterogeneous vancomycin resistance in methicillin-resistant Staphylococcus aureus strains isolated in a large Italianhospital. Journal of Clinical Microbiology 38, 866–9.

21. Deplano, A., Witte, W., Van Leeuwen, W. J., Brun, Y. & Struel-ens, M. J. (2000). Clonal dissemination of epidemic methicillin-resistant Staphylococcus aureus in Belgium and neighbouringcountries. Clinical Microbiology and Infection 6, 239–45.

22. National Committee for Clinical Laboratory Standards. (2001).Methods for Dilution Antimicrobial Susceptibility Tests for Bacteriathat Grow Aerobically—Fifth Edition: Approved Standard M7-A5.NCCLS, Wayne, PA, USA.

23. Tenover, F. C., Lancaster, M. V., Hill, B. C., Steward, C. D.,Stocker, S. A., Hancock, G. A. et al. (1998). Characterization ofstaphylococci with reduced susceptibilities to vancomycin and otherglycopeptides. Journal of Clinical Microbiology 36, 1020–7.

24. Chesneau, O., Morvan, A. & Solh, N. E. (2000). Retrospectivescreening for heterogeneous vancomycin resistance in diverseStaphylococcus aureus clones disseminated in French hospitals.Journal of Antimicrobial Chemotherapy 45, 887–90.

25. Patel, R., Uhl, J. R., Kohner, P., Hopkins, M. K. & Cockerill,F. R. (1997). Multiplex PCR detection of VanA, VanB, VanC-1, andVanC-2/3 genes in enterococci. Journal of Clinical Microbiology 35,703–7.

26. Mani, N., Tobin, P. & Jayaswal, R. K. (1993). Isolation andcharacterization of autolysis-defective mutants of Staphylococcusaureus created by Tn917-LacZ mutagenesis. Journal of Bacterio-logy 175, 1493–9.

27. Groupement pour le Dépistage l’Etude et la Prévention desInfections Hospitalières-Groep ter Opsporing, Studie en Preventievan de Infecties in de Ziekenhuizen GDEPIH-GOSPIZ. (1994).Guidelines for control and prevention of methicillin-resistantStaphylococcus aureus transmission in Belgian hospitals. ActaClinica Belgica 49, 108–13.

28. Howe, R. A., Wootton, M., Walsh, T. R., Bennett, P. M. &MacGowan, A. P. (2000). Heterogeneous resistance to vancomycinin Staphylococcus aureus. Journal of Antimicrobial Chemotherapy45, 130–2.

29. Aucken, H. M., Warner, M., Ganner, M., Johnson, A. P.,Richardson, J. F., Cookson, B. D. et al. (2000). Twenty months ofscreening for glycopeptide-intermediate Staphylococcus aureus.Journal of Antimicrobial Chemotherapy 46, 639–40.

30. Guerin, F., Buu-Hoi, A., Mainardi, J. L., Kac, G., Colardelle, N.,Vaupre, S. et al. (2000). Outbreak of methicillin-resistant Staphylo-coccus aureus with reduced susceptibility to glycopeptides in aParisian hospital. Journal of Clinical Microbiology 38, 2985–8.

31. Boyle-Vavra, S., Berke, S. K., Lee, J. C. & Daum, R. S. (2000).Reversion of the glycopeptide resistance phenotype in Staphylo-coccus aureus clinical isolates. Antimicrobial Agents and Chemo-therapy 44, 272–7.

32. Hubert, S. K., Mohammed, J. M., Fridkin, S. K., Gaynes, R. P.,McGowan, J. E. & Tenover, F. C. (1999). Glycopeptide-intermediateStaphylococcus aureus: evaluation of a novel screening methodand results of a survey of selected U. S. hospitals. Journal of Clin-ical Microbiology 37, 3590–3.

33. Walsh, T. R., Bolmstrom, A., Qwarnstrom, A., Ho, P., Wootton,M., Howe, R. A. et al. (2001). Evaluation of current methods fordetection of staphylococci with reduced susceptibility to glyco-peptides. Journal of Clinical Microbiology 39, 2439–44.

34. Burnie, J., Matthews, R., Jiman-Fatami, A., Gottardello, P.,Hodgetts, S. & D’Arcy, S. (2000). Analysis of 42 cases of septicemiacaused by an epidemic strain of methicillin-resistant Staphylo-coccus aureus: evidence of resistance to vancomycin. ClinicalInfectious Diseases 31, 684–9.

35. Pina, P., Marliere, C., Vandenesch, F., Bedos, J. P., Etienne, J.& Allouch, P. Y. (2000). An outbreak of Staphylococcus aureusstrains with reduced susceptibility to glycopeptides in a Frenchgeneral hospital. Clinical Infectious Diseases 31, 1306–8.

36. Centers for Disease Control and Prevention. (1997). Interimguidelines for prevention and control of staphylococcal infectionassociated with reduced susceptibility to vancomycin. Morbidity andMortality Weekly Report 46, 626–8, 635.

37. Sieradzki, K. & Tomasz, A. (1997). Inhibition of cell wall turn-over and autolysis by vancomycin in a highly vancomycin-resistantmutant of Staphylococcus aureus. Journal of Bacteriology 179,2557–66.

38. Cui, L., Murakami, H., Kuwahara-Arai, K., Hanaki, H. & Hira-matsu, K. (2000). Contribution of a thickened cell wall and itsglutamine non-amidated component to the vancomycin resistanceexpressed by Staphylococcus aureus Mu50. Antimicrobial Agentsand Chemotherapy 44, 2276–85.

39. Hanaki, H., Labischinski, H., Inaba, Y., Kondo, N., Murakami, H.& Hiramatsu, K. (1998). Increase in glutamine-non-amidated muro-



ownloaded from



391

peptides in the peptidoglycan of vancomycin-resistant Staphylo-coccus aureus strain Mu50. Journal of Antimicrobial Chemotherapy42, 315–20.

40. Boyle-Vavra, S., Labischinski, H., Ebert, C. C., Ehlert, K. &Daum, R. S. (2001). A spectrum of changes occurs in peptidoglycancomposition of glycopeptide-intermediate clinical Staphylococcusaureus isolates. Antimicrobial Agents and Chemotherapy 45, 280–7.

41. Finan, J. E., Archer, G. L., Pucci, M. J. & Climo, M. W. (2001).Role of penicillin-binding protein 4 in expression of vancomycinresistance among clinical isolates of oxacillin-resistant Staphylo-coccus aureus. Antimicrobial Agents and Chemotherapy 45, 3070–5.

42. Sieradzki, K., Wu, S. W. & Tomasz, A. (1999). Inactivation ofthe methicillin resistance gene mecA in vancomycin-resistantStaphylococcus aureus. Microbial Drug Resistance 5, 253–7.



ownloaded from




ownloaded from


Documents

Emergence of Vancomycin-Intermediate Staphylococcus aureus and S. sciuri , Greece