The first Staphylococcus aureus isolates with reduced susceptibility to vancomycin in Poland

Journal of Antimicrobial Chemotherapy (2002) 50, 1065 1069 DOI: 10.1093/jac/dkf252 The first Staphylococcus aureus isolates with reduced susceptibility to vancomycin in Poland Jolanta Krzysztoá-Russjan 1, Marek Gniadkowski 1, Hanna Poßowniak-Pracka 2, Elgbieta Hagmajer 2 and Waleria Hryniewicz 1 * 1 Sera and Vaccines Central Research Laboratory, ul. Cheßmska 30/34, 00-725 Warsaw; 2 The Maria Skßodowska-Curie Memorial Centre and Institute of Oncology, ul. Roentgena 5, 02-185 Warsaw, Poland Received 25 March 2002; returned 8 August 2002; revised 30 August 2002; accepted 16 September 2002 Introduction The first heterogeneously vancomycin-intermediate Staphylococcus aureus (hvisa) isolates to be identified in Poland were characterized. They were recovered in two Warsaw hospitals and, although in one of these the identification coincided with unsuccessful glycopeptide therapy, the selection conditions and clinical significance of the organisms remain unclear. Molecular typing revealed that the isolates were closely related and belonged to an international methicillinresistant S. aureus (MRSA) clone known as Iberian. The analysis of epidemiologically related MRSA isolates and some archival Polish MRSA isolates of the same clone indicated that the hvisa phenotype has been present in the MRSA populations of some hospitals in Poland since at least 1998. The first methicillin-resistant Staphylococcus aureus (MRSA) isolates with a specific mechanism of resistance to glycopeptides were described in 1997 in Japan 1 and have since been observed in other countries. 2,3 Two different phenotypes were identified among the glycopeptide-non-susceptible S. aureus strains. One of these is characterized by vancomycin MICs of 8 16 mg/l, and, according to the NCCLS, 4 such isolates are classified as vancomycin-intermediate S. aureus (VISA). 2,4 Isolates of the second phenotype demonstrate vancomycin MICs that are below the NCCLS breakpoint for intermediate strains (MICs 4 mg/l); however, they produce subpopulations of cells ( 10 6 ) that grow at a vancomycin concentration of 4 mg/l. These isolates are interpreted as heterogeneously vancomycin-intermediate S. aureus (hvisa). 2 In this study, we characterized the first hvisa isolates to be identified in Poland. Materials and methods Table 1. The analysis was commenced with eight isolates (isolates 1 8) collected from a single patient between June and August 2000 in an oncology centre in Warsaw (centre WA I). The next isolate (isolate 9) was recovered in June 2001 from a patient in a paediatric hospital in Warsaw (centre WA II). Twenty additional MRSA isolates (isolates 10 29), all from the collection of strains in the Sera and Vaccines Central Research Laboratory, were included in the study for comparative epidemiological analysis. Glycopeptide susceptibility testing MICs of vancomycin and teicoplanin were evaluated in accordance with NCCLS guidelines 4 and using Etest (AB Biodisk, Solna, Sweden), by both the standard procedure and the macromethod. 5 The screening test and population analysis of vancomycin susceptibility were performed as recommended by Hiramatsu et al. 1,3 S. aureus ATCC 29213 and S. aureus Mu3 (hvisa) and Mu50 (VISA) 1 were used as reference strains. Clinical isolates Twenty-nine MRSA isolates were analysed in this study; selected clinical data concerning the isolates are presented in PFGE PFGE analysis of the isolates was performed using a CHEF DRII apparatus (Bio-Rad, Hercules, CA, USA) as described... *Corresponding author. Tel: +48-22-841-33-67; Fax: +48-22-841-29-49; E-mail: waleria@urania.il.waw.pl... 1065 2002 The British Society for Antimicrobial Chemotherapy

Table 1. Study isolates: selected clinical data, glycopeptide susceptibility results, phenotypes and typing data MIC (mg/l) by dilution methods a MIC (mg/l) by Etest a,b vancomycin teicoplanin vancomycin teicoplanin Isolate c Centre Specimen Isolation date Screening test d CA broth CA broth BHI BHI hvisa phenotype e meca::tn554:: PFGE type 1066 1 WA I drain swab 05.06.00 + 2 2 4 8 + I::E::D6 2 WA I catheter tip 20.06.00 + 2 2 4 8 + I::E::D6 3 WA I respiratory tract 06.07.00 + 2 2 4 8 + I::E::D9 secretion 4 WA I blood 07.07.00 + 4 4 8 8 4 12 4 12 + I::E::D6 5 WA I bronchial tree 13.07.00 + 2 2 4 8 + I::E::D10 secretion 6 WA I abdominal 15.08.00 + 2 2 4 8 + I::E::D6 cavity swab 7 WA I sputum 15.08.00 + 4 4 8 8 4 12 4 16 + I::E::D6 8 WA I oral cavity 18.08.00 + 2 2 4 4 + I::E::D6 secretion 9 WA II throat swab 12.06.01 + 4 4 8 8 4 12 6 24 + I::E::D6 10 WA I eye socket swab 26.06.98 + 2 2 8 8 + I::E::D7 11 WA I tracheostomy 26.05.99 1 2 4 4 III::DD::B2 tube 12 WA I tracheostomy 15.06.00 + 2 2 8 8 + I::E::D6 tube 13 WA I wound 26.02.00 1 1 1 2 III::DD::B2 14 WA I wound 01.06.00 + 2 2 8 8 + I::E::D6 15 WA I blood 01.06.00 + 2 2 8 8 + I::E::D6 16 WA I urine catheter 26.06.00 + 2 2 8 8 + I::E::D6 17 WA I wound 11.01.01 + 2 2 8 8 + I::E::D6 18 WA I bronchial swab 08.01.01 1 1 1 2 III::DD::B7 19 WA I sputum 23.04.01 + 2 4 8 8 + I::E::D6 20 WA III urine 25.01.00 + 2 2 8 8 + I::E::D6 21 WA III eye socket swab 24.01.00 + 2 4 8 8 + I::E::D6 22 WA III wound 19.01.98 + 2 4 4 8 + I::E::D11 23 WA IV intubation tube 19.01.98 1 2 4 4 I::E::D6 24 WA IV wound 29.12.97 1 2 4 4 I::E::D7 25 WA IV wound 29.12.97 1 2 4 4 I::E::D2 26 WA V urine 17.10.96 1 2 4 4 I::E::D6 J. Krzysztoá-Russjan et al. Downloaded from http://jac.oxfordjournals.org/ at Pennsylvania State University on April 30, 2014

hvisa in Poland by Chung et al. 6 PFGE patterns were compared with the use of Molecular Analyst software, version 1.12 (Bio-Rad). 27 WA V pus (fistula) 27.09.94 1 2 4 4 I::E::D7 28 SI blood 12.02.94 1 2 4 4 I::E::D7 29 BB pus 08.09.97 1 2 4 4 I::J::D6 Mu3 f + 1 1 8 8 1 12 12 32 Mu50 f + 4 8 8 8 8 16 12 24 ATCC 29213 0.5 1 1 1 1 4 1 3 a CA, Mueller Hinton cation adjusted;, Mueller Hinton II; BHI, brain heart infusion. b Etest MICs were determined only for selected isolates. c Isolates 1 8 were derived from a single patient. d Plus and minus signs designate isolates that were positive and negative, respectively, in the vancomycin susceptibility screening test. 1,3 e Plus and minus signs designate isolates that were classified as hvisa and vancomycin-susceptible isolates, respectively. f Mu3 and Mu50 are the original hvisa and VISA strains, respectively. 1 Typing by ClaI restriction fragment length polymorphism (RFLP) analysis of meca and Tn554 loci RFLP analysis of meca and Tn554 loci was performed as reported by Leski et al. 7 Hybridization patterns were classified into ClaI meca and ClaI Tn554 types by comparison with the types described previously. 7 Results Isolates 1 8 MRSA isolates 1 8 from centre WA I were collected from a 45-year-old male patient with a stomach cancer. Three days after oesophagogastrectomy, the patient developed peritonitis and was transferred to the intensive care unit (ICU), where he spent 73 days undergoing numerous invasive procedures, including intubation, tracheostomy, intravenous central catheterization and total parenteral nutrition. The patient developed several infections caused by various microorganisms and was intensively treated with antibiotics. The glycopeptide treatment consisted of two courses of vancomycin (41 days altogether) and one course of teicoplanin (21 days). The patient improved following quinupristin/ dalfopristin therapy. MRSA isolates were recovered from various specimens at several time points over the whole period of treatment (Table 1). The first two isolates were cultured soon after his admission to the ICU. The next three isolates were identified 3 days after the first or during the second course of vancomycin therapy. The last three isolates were recovered several days after teicoplanin therapy, before and just at the beginning of treatment with quinupristin/dalfopristin. Table 1 and Figure 1 show the results of the analysis of isolates 1 8. Regarding MICs, the isolates behaved as susceptible to glycopeptides; 4 however, in the screening test, they Figure 1. Population analysis curves obtained for the selected clinical isolates and for the S. aureus ATCC 29213, Mu3 and Mu50 2 reference strains. 1067

J. Krzysztoá-Russjan et al. produced colonies on plates containing vancomycin 4 mg/l, with a frequency of 10 6. The population study revealed that cultures of the isolates contained fractions of cells that grew at a vancomycin concentration of 7 mg/l ( 10 7 ). According to the criteria of Hiramatsu et al., 1 they were classified as hvisa. PFGE typing demonstrated that isolates 1 8 were either indistinguishable or closely related to each other. The predominant pattern was identified in the database of Polish MRSA as subtype D6 (J. Krzysztoá-Russjan & W. Hryniewicz, unpublished results). All the isolates were characterized by ClaI meca RFLP type I and ClaI Tn554 RFLP type E. Therefore, they were specified by the combined RFLP type I::E::D, where I and E are international designations of ClaI meca and ClaI Tn554 types, 8 and D is a PFGE type designation in the context of the clonal structure of MRSA in Poland. Isolate 9 Isolate 9 from centre WA II was recovered from a throat swab of an asymptomatic newborn at the time of its admission from another hospital, WA III. The isolate was classified as putative hvisa by the WA II laboratory, which routinely tests MRSA on brain heart infusion (BHI) plates with vancomycin 4 mg/l. Table 1 and Figure 1 show the results of the analysis of the isolate. The glycopeptide susceptibility testing revealed that it behaved similarly to isolates 1 8 and could be interpreted as hvisa. 1 Moreover, the isolate was closely related to isolates 1 8 in typing and was classified into the I::E::D RFLP type, subtype I::E::D6. Isolates 10 22 Results of the analysis of isolates 1 8 and 9 indicated the necessity of a wider study, including additional MRSA isolates from centres WA I and WA III. Isolates 10 19 were collected from different patients in WA I between 1998 and 2001. Results of their analysis are shown in Table 1. The glycopeptide susceptibility data demonstrated that the majority of the isolates, including isolate 10 from 1998, were of the hvisa phenotype, 1 and all these could be classified into the I::E::D RFLP type, mostly subtype I::E::D6. The remaining non-hvisa isolates represented another RFLP type, III::B/DD::B. Similar data were obtained for isolates 20 22 from January 2000, which were the only MRSA isolates available from centre WA III (Table 1 and Figure 1). They demonstrated the hvisa phenotype 1 and belonged to the I::E::D RFLP type, including subtype I::E::D6. Isolates 23 29 Identification of the hvisa phenotype in MRSA isolates of the I::E::D clone in separate Warsaw hospitals was the reason for a retrospective study of other isolates of the same clone. Seven MRSA isolates, collected from 1994 to 1998 in four different hospitals (WA IV, WA V, SI and BB), were selected from the database of Polish MRSA (J. Krzysztoá-Russjan & W. Hryniewicz, unpublished results). They represented PFGE subtype D6 or closely related subtypes (Table 1) and belonged to ClaI meca type I and, except for one isolate, to ClaI Tn554 type E. None of the isolates was positive in the screening test for reduced susceptibility to vancomycin; therefore, they could not be classified as hvisa. Discussion This work is the first report on hvisa isolates in Poland. Initially, it was focused on the group of isolates collected in 2000 from a single patient in an oncology centre in Warsaw (WA I). The repetitive MRSA isolation, despite the prolonged glycopeptide treatment, indicated that it could have survived this therapy at certain locations for longer periods or that the patient was re-colonized and re-infected several times. The isolates fulfilled Hiramatsu s criteria for hvisa; 1 however, it is impossible to judge whether this phenotype was responsible for the failure of glycopeptide therapy. Molecular typing demonstrated that the patient was affected by a single hvisa strain, which slightly diversified with time. Because the first isolate was recovered at the beginning of the patient s stay in the ICU, the phenotype had most probably been acquired by the strain before colonization of the patient. This indicated that hvisa had not been noticed in the hospital for a certain time, which was later confirmed by analysis of the archival isolate from 1998. The hvisa phenotype correlated with the specific MRSA clone, as all the hvisa isolates analysed here were closely related to each other. They all represented variants of a strain of the combined I::E::D RFLP type, mostly subtype I::E::D6. PFGE type D MRSA isolates have been observed in Polish hospitals, mainly in Warsaw, since 1994 7 (subtype D6 since 1996), and they have usually represented the ClaI meca type I and ClaI Tn554 type E 7 (J. Krzysztoá-Russjan & W. Hryniewicz, unpublished results). Recently, the Polish PFGE type D MRSA clone was identified as belonging to the Iberian international clone. 8 The hvisa phenotype has not been an imminent characteristic of the I::E::D clone in Poland, as indicated by retrospective analysis of isolates identified in four hospitals from 1994 to 1998. The hvisa phenotype was originally identified in 1996 in the Mu3 MRSA strain in Japan, and its isolation coincided with the clinical failure of vancomycin therapy. 1 It has since been observed in MRSA isolates in other countries, 2,3 the majority of which were identified in retrospective studies, and it has been impossible to define clearly the clinical significance of the phenotype because of the lack of full clinical data. 9,10 What seems to be widely accepted is that hvisa 1068

hvisa in Poland may represent an intermediary stage in the evolution of VISA strains, which have been convincingly documented as non-responding to glycopeptide therapy. 1,2 Therefore, the precise monitoring of hvisa occurrence seems to be of high importance. Acknowledgements We would like to thank Keiichi Hiramatsu, who kindly provided the S. aureus Mu3 and Mu50 reference strains, Katarzyna Nowak for her excellent technical assistance, and microbiologists from all the centres for collecting isolates. We are also thankful to Tomasz déski for helpful discussions, and Barbara S. Ink for critical reading of the manuscript. This work was partially financed by the US Poland Maria Skßodowska-Curie Joint Fund II (MZ/NIH 98-324). References 1. Hiramatsu, K., Aritaka, N., Hanaki, H., Kawasaki, S., Hosoda, Y., Hori, S. et al. (1997). Dissemination in Japanese hospitals of strains of Staphylococcus aureus heterogeneously resistant to vancomycin. Lancet 350, 1670 3. 2. Tenover, F. C., Biddle, J. W. & Lancaster, M. V. (2001). Increasing resistance to vancomycin and other glycopeptides in Staphylococcus aureus. Emerging Infectious Diseases 7, 327 32. 3. Trakulsomboon, S., Danchaivijitr, S., Rongrungruang, Y., Dhiraputra, C., Susaemgrat, W., Ito, T. et al. (2001). First report of methicillin-resistant Staphylococcus aureus with reduced susceptibility to vancomycin in Thailand. Journal of Clinical Microbiology 39, 591 5. 4. National Committee for Clinical Laboratory Standards. (2001). Methods for Dilution Antimicrobial Susceptibility Tests for Bacteria that Grow Aerobically Fifth Edition: Approved Standard M7-A5. NCCLS, Villanova, PA, USA. 5. Walsh, T. R., Bolmström, A., Qwärnström, A., Ho, P., Wootton, M., Howe, R. et al. (2001). Evaluation of current methods for detection of staphylococci with reduced susceptibility to glycopeptides. Journal of Clinical Microbiology 39, 2439 44. 6. Chung, M., de Lencastre, H., Matthews, P., Tomasz, A., Adamsson, I., Aries de Sousa, M. et al. (2000). Multilaboratory Project Collaborators. Molecular typing of methicillin-resistant Staphylococcus aureus by pulsed-field gel electrophoresis: comparison of results obtained in a multilaboratory effort using identical protocols and MRSA strains. Microbial Drug Resistance 6, 189 98. 7. Leski, T., Oliveira, D., Trzciåski, K., Santos Sanches, I., Aires de Sousa, M., Hryniewicz, W. et al. (1998). Clonal distribution of methicillin-resistant Staphylococcus aureus in Poland. Journal of Clinical Microbiology 36, 3532 9. 8. Mato, R., Santos Sanches, I., Venditti, M., Platt D. J., Brown, A., Chung, M. et al. (1998). Spread of the multiresistant Iberian clone of methicillin-resistant Staphylococcus aureus (MRSA) to Italy and Scotland. Microbial Drug Resistance 4, 107 12. 9. Bierbaum, G., Fuchs, K., Lenz, W., Szekat, C. & Sahl, H. G. (1999). Presence of Staphylococcus aureus with reduced susceptibility to vancomycin in Germany. European Journal of Clinical Microbiology and Infectious Diseases 18, 691 6. 10. Ike, Y., Arakawa, Y., Ma, X., Tatewaki, K., Nagasawa, M., Tomita, H. et al. (2001). Nationwide survey shows that methicillinresistant Staphylococcus aureus strains heterogeneously and intermediately resistant to vancomycin are not disseminated throughout Japanese hospitals. Journal of Clinical Microbiology 39, 4445 51. 1069