Activity of Linezolid Against Gram-Positive Cocci Isolated in French Hospitals As Determined by Three In-Vitro Susceptibility Testing Methods C

View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector

ORIGINAL ARTICLE 10.1111/j.1198-743X.2004.00751.x

Activity of linezolid against Gram-positive cocci isolated in French hospitals as determined by three in-vitro susceptibility testing methods C. Muller-Serieys1, H. B. Drugeon2, J. Etienne3, C. Lascols4, R. Leclercq5, J. Nguyen6 and C. J. Soussy4

1Groupe Hospitalier Bichat, Paris, 2CHU Laennec, Nantes, 3Hoˆpital Edouard Herriot, Lyon, 4CHU Henri Mondor, Cre´teil, 5CHU Coˆte de Nacre, Caen and 6CHU Pitie´-Salpoˆtrie´re, Paris, France

ABSTRACT In total, 844 strains of Gram-positive cocci were collected from six university hospitals in France between September 1999 and January 2000. MICs of linezolid were determined: (i) for all strains by agar dilution (method A); (ii) by broth microdilution (method B) for staphylococci and enterococci; (iii) by Etest (method E) for b-haemolytic streptococci and Streptococcus pneumoniae. Susceptibility to other antibiotics was determined by the disk diffusion method. MIC50 and MIC90 values were identical (2 mg ⁄ L) for methicillin-susceptible Staphylococcus aureus (n ¼ 179) by methods A and B. Linezolid was active against methicillin-resistant S. aureus (n ¼ 117), with an MIC90 of 2 mg ⁄ L (methods A and B), but with a lower MIC50 of 1 mg ⁄ L by method A. Of the 200 coagulase-negative staphylococci, 56.5% were methicillin- resistant and 43.5% were methicillin-susceptible. Linezolid had similar in-vitro activity by methods A and B (MIC50 and MIC90 values of 1–2 mg ⁄ L), irrespective of methicillin susceptibility. The MIC90 of linezolid for all enterococci (150 Enterococcus faecalis and 50 Enterococcus faecium) was 2 mg ⁄ L by both methods. MICs of linezolid for b-haemolytic streptococci had a narrow range of 0.5–2 mg ⁄ L (method A) and 0.125–2 mg ⁄ L (method E). Pneumococci (n ¼ 118), including 67 penicillin G-intermediate and - resistant strains, were all inhibited by linezolid 2 mg ⁄ L (MIC90 of 2 mg ⁄ L by methods A and E). No strain had an MIC of > 2 mg ⁄ L by agar dilution or Etest, or of > 4 mg ⁄ L by broth microdilution. Overall, the study conﬁrmed the good in-vitro activity of linezolid and the very narrow range of MICs for Gram- positive cocci susceptible or resistant to other antibiotics, irrespective of the method used. Keywords Linezolid, antibacterial activity, Gram-positive cocci, testing methods Original Submission: 4 July 2002; Revised Submission: 25 October 2002; Accepted: 11 February 2003 Clin Microbiol Infect 2004; 10: 242–246

Linezolid is the ﬁrst member of a new class of [5–7]. Linezolid demonstrates good in-vitro synthetic antibacterial agents known as oxazolid- activity against Gram-positive anaerobes (MICs inones, which act by inhibiting initiation of protein of 1–2 mg ⁄ L) and some species of Gram-negative synthesis [1,2]. The targets of oxazolidinones bacilli, including Bacteroides spp., Fusobacterium appear not to be recognised by other antibacterial nucleatum and Prevotella spp. [8–10]. The aims of agents, since no cross-resistance has been found in this study were: (i) to assess the activity of strains resistant to other antibiotics [3,4]. The linezolid against a large number of recent clinical in-vitro and in-vivo spectrum of activity of linezo- isolates collected in six French university hospi- lid is primarily against Gram-positive cocci, inclu- tals; and (ii) to evaluate the in-vitro activity of ding methicillin-resistant Staphylococcus aureus linezolid by different methods. (MRSA), coagulase-negative staphylococci (CNS), and enterococci, especially vancomycin- and ampicillin-resistant strains and penicillin-suscept- MATERIALS AND METHODS ible and -resistant Streptococcus pneumoniae isolates Study design Six French teaching hospital laboratories were enrolled in the study. Each participating centre tested c.150 Gram-positive Corresponding author and reprint requests: C. Muller-Serieys, clinical isolates according to a standardised protocol. The Groupe Hospitalier Bichat-Claude Bernard, Laboratoire de isolates comprised (maximum number): methicillin-suscept- Bacte´riologie, 46 Rue Henri-Huchard, 75018 Paris, France ible S. aureus (MSSA) (30), MRSA (20), methicillin-susceptible E-mail: [email protected]_fr CNS (15), methicillin-resistant CNS (20), Enterococcus faecium

Ó 2004 Copyright by the European Society of Clinical Microbiology and Infectious Diseases Muller-Serieys et al. Linezolid activity against Gram-positive cocci isolated in France 243

(30), Enterococcus faecalis (10), other Enterococcus spp. (40), Quality assurance Strep. pneumoniae (20), and other Streptococcus spp. (viridans Internal quality control was carried out with S. aureus strains group or b-haemolytic streptococci) (5). Non-repetitive iso- ATCC 25923 and ATCC 29213, Strep. pneumoniae strain ATCC lates were collected between September 1999 and January 49619, and E. faecalis strain ATCC 29212. Each strain was tested 2000. ten times during the study by the MIC and disk-diffusion The strains were isolated from the following body sites: methods in each laboratory. blood (only isolates considered as clinically significant by local laboratory criteria), cerebrospinal fluid, skin and skin struc- tures (i.e., post-surgical wound, lower extremity infection site, RESULTS traumatic wound, cellulitis, abscess), bone and joint, intra- abdominal (i.e., peritonitis, intra-abdominal abscess), lower In total, 844 clinical strains (Table 1) were collec- respiratory tract (i.e., tracheal aspirate, bronchoalveolar lavage, ted and tested as consecutive isolates. bronchial brush, pleural fluid, sputum), and upper respiratory Among the 296 S. aureus isolates, 117 (39.5%) tract (i.e., sinus, middle ear). were methicillin-resistant (MRSA). Most MRSA isolates were also resistant to erythromycin Susceptibility testing methods (80.3%) and lincomycin (74.4%), indicating that Several different methods were used for antibiotic suscepti- resistance to macrolides was mainly of the consti- bility testing. tutive MLSB phenotype. Resistance rates to pefl- MIC determination oxacin were very high (93.2%), and the resistance MICs of linezolid were determined by agar dilution (method rate to gentamicin reached 34.2% among A) using Mueller-Hinton number 2 medium (bioMe´rieux; La MRSA strains. MSSA were resistant to erythro- Balme les Grottes, France) for staphylococci and enterococci, mycin less frequently (20.7%) and lincomycin % supplemented with defibrinated sheep blood 5 v ⁄ v for % Strep. pneumoniae and groups A, B, C and G streptococci. (8.5 ), and resistance to gentamicin and pefloxa- Isolates were incubated at 37 °C in ambient atmosphere for cin remained rare (0.6% and 4.5%, respectively). staphylococci and enterococci and in CO2 5% v ⁄ v for other Linezolid MIC ranges were similar for MSSA species. The inocula comprised 104 CFU ⁄ spot. In addition, and MRSA, with most MIC values between 1 MICs for staphylococci and enterococci were determined by and 2 mg ⁄ L (Fig. 1). MIC and MIC values of broth microdilution in a cation-adjusted Mueller-Hinton 50 90 broth incubated at 35 °C for 18–24 h in ambient atmosphere linezolid, measured by methods A and B, were (method B); MICs for Strep. pneumoniae and groups A, B, C similar for MSSA (2 mg ⁄ L). For MRSA, the and G streptococci were obtained by Etest (AB Biodisk, linezolid MIC50 was slightly lower (1 mg ⁄ L) by Solna, Sweden) (method E) on Mueller-Hinton number 2 method A. The MIC was 4 mg ⁄ L by method B agar with sheep blood 5% v ⁄ v and an inoculum of c. 108 CFU ⁄ mL. for six S. aureus isolates (four MSSA and two MRSA). Disk diffusion Of the 200 CNS isolates, 56.5% were methicillin- Antibiotic susceptibility testing by agar plate disk diffusion resistant and 43.5% were methicillin-susceptible. was performed in each participating laboratory. The antimi- Of the methicillin-resistant CNS, 61% were also crobial agents tested varied according to the species, but included b-lactams (penicillin, methicillin), macrolides (eryth- resistant to erythromycin, 35.4% to lincomycin, romycin, lincomycin, pristinamycin), a fluoroquinolone (pefl- 57.5% to gentamicin and 79.6% to pefloxacin. One oxacin), glycopeptides (vancomycin, teicoplanin) and an strain was resistant to teicoplanin, but all strains aminoglycoside (gentamicin). Zone diameters were measured were susceptible to vancomycin. Only 33.3% and after incubation for 18–24 h. Isolates were classified as susceptible, intermediate or resistant according to the guidelines 4.6% of methicillin-susceptible CNS were resist- of the Comite´ de l’Antibiogramme de la Socie´te´ Française de ant, respectively, to erythromycin and lincomycin. Microbiologie (CA-SFM) [11]. One strain was gentamicin-resistant and only four

Table 1. Distribution of strains by S. aureus CNS E. faecalis E. faecium Strep. pneumoniae Streptococci Total culture source Blood 76 87 47 13 30 10 263 Intra-abdominal 4 11 34 26 0 1 76 Skin and skin structure 50 9 23 2 0 8 92 Upper respiratory 17 2 1 0 14 2 36 Lower respiratory 79 9 3 2 58 1 152 Cerebrospinal ﬂuid 3 10 0 0 3 0 16 Bone and joint 18 15 3 0 0 2 38 Others 49 57 39 7 13 6 171

Total 296 200 150 50 118 30 844

CNS, coagulase-negative staphyloccoci.

Ó 2004 Copyright by the European Society of Clinical Microbiology and Infectious Diseases, CMI, 10, 242–246 244 Clinical Microbiology and Infection, Volume 10 Number 3, March 2004

100 200 80 150 60 100 Oxa S 40 Oxa S 20 50 Oxa R No. of strains Oxa R No. of strains 0 0 0.125 0.25 0.5 1 2 0.25 0.5 1 2 MIC (mg/L): method A MIC (mg/L): method A

50 200 150 40 100 Oxa S 30 50 Oxa R 20 Oxa S

No. of strains 0

No. of strains 10 0.25 0.5 1 2 4 Oxa R 0 MIC (mg/L): method B 0.25 0.5 1 2 4 MIC (mg/L): method B Fig. 1. Distribution of MICs of linezolid against S. aureus (n ¼ 296) by agar dilution (method A) and broth microdi- Fig. 2. Distribution of MICs of linezolid against coagulase lution (method B). negative staphylococci (n ¼ 200) by agar dilution (method A) and broth microdilution (method B). strains were resistant to peﬂoxacin. The linezolid MIC distributions are shown in Fig. 2. All isolates were susceptible to linezolid, with MICs of 0.125– to erythromycin and lincomycin reached 57.6%. 2mg⁄ L by method A, and 0.25–4 mg ⁄ Lby All the strains were susceptible to linezolid, with method B. The MIC50 and MIC90 values were MICs of 0.25–2 mg ⁄ L (Fig. 4). MIC50 and MIC90 similar for methicillin-resistant and methicillin- values were 1 and 2 mg ⁄ L, respectively, by susceptible CNS (1 and 2 mg ⁄ L, respectively) by method A, but the MIC50 appeared lower methods A and B. (0.38 mg ⁄ L) when measured by Etest. Of the 200 enterococci tested, E. faecalis accoun- The 30 b-haemolytic streptococci included four ted for 75% and E. faecium for 25%. Only 34.6% of erythromycin-resistant and three lincomycin- E. faecalis and 18% of E. faecium isolates remained resistant strains, thereby indicating a low susceptible to erythromycin. Approximately 10% prevalence of macrolide resistance. All 30 strains of the two species showed high-level resistance to were susceptible to cefotaxime and glycopeptides. gentamicin. Ampicillin resistance rates reached Linezolid was active against all isolates with 56% for E. faecium. All the enterococci were MICs of 0.5–2 mg ⁄ L by method A, and 0.125– susceptible to glycopeptides. The linezolid MIC 2mg⁄ L by method E. The distribution of linezo- ranges for enterococci, by methods A and B, are lid MICs for b-haemolytic streptococci groups A, shown in Fig. 3. Linezolid MIC50 and MIC90 B, C and G is shown in Fig. 5. The MIC90 value values for E. faecalis and E. faecium were both was similar by the two methods (2 mg ⁄ L), while 2mg⁄ L by method A, but the linezolid MIC90 for the MIC50 value was one dilution lower by E. faecalis was one dilution higher (4 mg ⁄ L) by method E (0.5 mg ⁄ L) compared to method A method B. (1 mg ⁄ L). Of 118 clinical isolates of Strep. pneumoniae, 45.8% and 11%, respectively, were either inter- DISCUSSION mediate or resistant to penicillin G (MIC of >1 mg⁄ L). Two strains were resistant to cefotax- Linezolid, an oxazolidinone, is being developed ime (MIC of > 2 mg ⁄ L) [11], while resistance rates to treat infections caused by Gram-positive

Ó 2004 Copyright by the European Society of Clinical Microbiology and Infectious Diseases, CMI, 10, 242–246 Muller-Serieys et al. Linezolid activity against Gram-positive cocci isolated in France 245

E. faecalis (n = 150)

150 150

100 100

50 50 No. of strains No. of strains 0 0 12 0.5 1 2 4 MIC (mg/L): method A MIC (mg/L): method B

E. faecium (n = 50) 40 50 20 Fig. 3. Distribution of MICs of line-

0 No. of strains 0 zolid against enterococci by agar No. of strains 1 2 1 2 dilution (method A) and broth microdilution (method B). MIC (mg/L): method A MIC (mg/L): method B

20 10 15 S 10 No. of strains I 5 Penicillin G 0 No. of strains R 0.5 1 2 0 0.25 0.5 1 2 MIC (mg/L): method A MIC (mg/L): method A

20 10 15 5 10 S No. of strains 5 I No. of strains Penicillin G R 0 0 0.125 0.38 0.75 2 0.19 0.38 0.75 2 MIC (mg/L): method E MIC (mg/L): method E

Fig. 4. Distribution of MICs of linezolid against S. pneu- Fig. 5. Distribution of MICs of linezolid against other moniae (n ¼ 118) by agar dilution (method A) and Etest streptococci (n ¼ 30) by agar dilution (method A) and (method E). Etest (method E). bacteria, including those resistant to the antibiot- Linezolid had potent activity against entero- ics currently available. In the present study, cocci, irrespective of vancomycin and ampicillin linezolid had the same activity against MSSA, susceptibilities. MIC90 values were similar for MRSA and CNS. These results are in agreement vancomycin-susceptible and vancomycin-resist- with previous studies [12–14]. Linezolid is active ant enterococci [16]. Linezolid was also very against glycopeptide non-susceptible staphylo- active against Strep. pneumoniae, regardless of cocci [15], although only one teicoplanin-resistant penicillin or erythromycin resistance, with MICs strain of CNS (susceptible to linezolid) was always £ 2mg⁄ L. Other studies have shown isolated in this study. similar ﬁndings [17,18]. However, in the present

Ó 2004 Copyright by the European Society of Clinical Microbiology and Infectious Diseases, CMI, 10, 242–246 246 Clinical Microbiology and Infection, Volume 10 Number 3, March 2004 study, the MICs appeared lower for Strep. pneu- 9. Goldstein EJ, Citron DM, Merriam CV. Linezolid activity compared to those of selected macrolides and other agents moniae when measured by Etest (MIC50 of ⁄ against aerobic and anaerobic pathogens isolated from soft 0.38 mg L). Linezolid was also very active against tissue bite infections in humans. Antimicrob Agents Chem- erythromycin-resistant b-haemolytic streptococci other 1999; 43: 1469–1474. [19,20]. 10. Edlund C, Oh H, Nord CE. In vitro activity of linezolid and For each group of bacteria included in this eperezolid against anaerobic bacteria. Clin Microbiol Infect study, the linezolid MICs never exceeded 4 mg ⁄ L, 1999; 5: 51–53. 11. Comite´ de l’Antibiogramme. Communique´ 2002. Paris, and linezolid was equally active against strains France: Socie´te´ Française de Microbiologie, 2002. that were multiply-resistant to other antibiotics. 12. von Eiff C, Peters G. Comparative in-vitro activities of These results suggest that linezolid is a valuable moxifloxacin, trovafloxacin, quinupristin ⁄ dalfopristin and antimicrobial agent for the treatment of serious linezolid against staphylococci. J Antimicrob Chemother 1999; 43: 569–573. infections caused by Gram-positive cocci. 13. Ballow C, Jones R, Biedenbach D, Bolmstrom A. Multi- centre evaluation of linezolid antimicrobial activity in Europe. Clin Microbiol Infect 2001; 7(suppl 1): 268. REFERENCES 14. Cuny C, Witte W. In vitro activity of linezolid against 1. Swaney SM, Aoki H, Ganoza MC, Shinabarger DL. The staphylococci. Clin Microbiol Infect 2000; 6: 331–333. oxazolidinone linezolid inhibits initiation of protein syn- 15. Rybak MJ, Hershberger E, Moldovan T, Grucz RG. In vitro thesis in bacteria. Antimicrob Agents Chemother 1998; 42: activities of daptomycin, vancomycin, linezolid, and qui- 3251–3255. nupristin-dalfopristin against staphylococci and entero- 2. Shinabarger D. Mechanism of action of oxazolidinone cocci, including vancomycin-intermediate and – resistant antibacterial agents. Expert Opin Invesigt Drugs 1999; 8: strains. Antimicrob Agents Chemother 2000; 44: 1062–1066. 1195–1202. 16. Henwood CJ, Livermore DM, Johnson AP, James D, 3. Fines M, Leclercq R. Activity of linezolid against Gram- Warner M, Gardiner A. Susceptibility of gram-positive positive cocci possessing genes conferring resistance to cocci from 25 UK hospitals to antimicrobial agents inclu- protein synthesis inhibitors. J Antimicrob Chemother 2000; ding linezolid. The Linezolid Study Group. J Antimicrob 45: 797–802. Chemother 2000; 46: 931–940. 4. Ford CW, Hamel JC, Stapert D et al. Oxazolidinones: new 17. Cercenado E, Garcia-Garrote F, Bouza E. In vitro activity of antibacterial agents. Trends Microbiol 1997; 5: 196–200. linezolid against multiply resistant Gram-positive clinical 5. Struwig MC, Botha PL, Chalkley LJ. In vitro activities of 15 isolates. J Antimicrob Chemother 2001; 47: 77–81. antimicrobial agents against clinical isolates of South 18. Wise R, Andrews JM, Boswell FJ, Ashby JP. The in-vitro African enterococci. Antimicrob Agents Chemother 1998; 42: activity of linezolid (U-100766) and tentative breakpoints. 2752–2755. J Antimicrob Chemother 1998; 42: 721–728. 6. Chien JW, Kucia ML, Salata RA. Use of linezolid, an 19. Betriu C, Redondo M, Palau ML et al. Comparative in vitro oxazolidinone, in the treatment of multidrug-resistant activities of linezolid, quinupristin–dalfopristin, moxifl- gram-positive bacterial infections. Clin Infect Dis 2000; 30: oxacin, and trovafloxacin against erythromycin-susceptible 146–151. and -resistant streptococci. Antimicrob Agents Chemother 7. Clemett D, Markham A. Linezolid. Drugs 2000; 59: 815– 2000; 44: 1838–1841. 827. 20. Johnson AP, Warner M, Livermore DM. Activity of li- 8. Zurenko GE, Yagi BH, Schaadt RD et al. In vitro activities of nezolid against multi-resistant gram-positive bacteria from U-100592 and U-100766, novel oxazolidinone antibacterial diverse hospitals in the United Kingdom. J Antimicrob agents. Antimicrob Agents Chemother 1996; 40: 839–845. Chemother 2000; 45: 225–230.

Ó 2004 Copyright by the European Society of Clinical Microbiology and Infectious Diseases, CMI, 10, 242–246