Influence of atmospheric conditions during incubation on the

JAC

Journal of Antimicrobial Chemotherapy (1997) 40, 599–608

Correspondence Influence of atmospheric conditions during incubation on the susceptibilities of Streptococcus pneumoniae isolates to five -lactam antibiotics J Antimicrob Chemother 1997; 40: 599–601 Monique Chomarata*, Lionel Cholleta, Michel Peyretb and Jean-Pierre Flandroisa a

Laboratoire de Microbiologie, Centre Hospitalier Lyon-Sud, 69310 Pierre-Bénite; bbioMérieux, La Balme les Grottes, France *Corresponding author. Tel: +33-04-7886-1234; Fax: +33-04-7886-3280. Sir, The National Committee for Clinical Laboratory Standards (NCCLS) has recommended that, when MICs are determined by the microbroth dilution method, the test medium should be supplemented with 5% lysed horse blood, that the atmosphere during incubation should be aerobic and, further, that a medium supplemented with sheep blood and a CO 2-enriched atmosphere be used for the disc diffusion test.1,2 The Société Française de Microbiologie (SFM), on the other hand, has not specified a methodology for susceptibility testing,3 although, in many studies carried out in France, Mueller–Hinton, supplemented with fresh horse blood, has been the medium used.4 The present study was undertaken to assess the influence of atmospheric conditions during incubation on the MICs of five -lactam antibiotics for strains of Streptococcus pneumoniae. The MICs for 37 clinical isolates of S. pneumoniae were determined by the agar dilution method. The antibiotics tested (and their sources) were as follows: amoxycillin and oxacillin from SmithKline Beecham (Nanterre, France), penicillin G and cefotaxime from Hoechst–Roussel (Paris, France) and imipenem from Merck Sharp & DohmeChibret (Paris, France). An inoculum of 5 104 cfu was applied with a Steers replicator to the surface of Mueller–Hinton agar supplemented with 5% lysed horse blood and containing serial two-fold dilutions of each antibiotic at concentrations ranging from 0.03 to 4 mg/L. The tests were carried out in triplicate, with one plate from each set being incubated under each of three atmospheric conditions (aerobic, anaerobic and aerobic enriched with 5% CO 2). The MIC was defined as the lowest concentration of each antibiotic that yielded no visible growth after

incubation at 35°C for 18 h. S. pneumoniae ATCC 49619 was used as a control. The MICs determined under the three atmospheric conditions were compared and discrepancies between the results obtained under the different conditions were expressed in terms of the number of two-fold dilutions. From these data, the percentage of isolates for which there was agreement (i.e., MICs differing by no more than one two-fold dilution) were calculated. Of the 37 strains of pneumococci, 13 were susceptible to penicillin (MICs 0.1 mg/L), eight were of intermediate susceptibility (MICs 0.1–1 mg/L) and 16 were resistant (MICs 1 mg/L). A correlation of the MICs determined under the various atmospheric conditions of incubation is shown in the Table. For all of the antibiotics tested, there was excellent agreement between the results, irrespective of the atmosphere of incubation. The poorest correlation was observed with oxacillin, but, even then, the MICs did not vary by more than one two-fold dilution. The percentage correlation for each comparison (i.e., MICs differing by no more than one two-fold dilution) was invariably 95% and, in most cases, 100%. According to interpretative breakpoint criteria recommended by the SFM, the isolates were placed into susceptibility categories (susceptible, intermediately susceptible and resistant), the results compared and the numbers of each type of discrepancy (minor, major and very major) noted. (For definitions of the three degrees of discrepancy, see the footnote to the Table). The number of minor discrepancies varied from none to four and there were no major or very major discrepancies (Table). Several methods of determining the susceptibilities of pneumococci to penicillin, each with its own medium and incubation conditions, have been described, but there have been few studies that have assessed the influence of the conditions of incubation on the outcome of testing. Swenson et al.4 compared susceptibilities to methicillin and oxacillin, as determined by the disc diffusion method in either ambient air or in an atmosphere enriched with CO2, and found no significant differences, although the diameters of the zones of inhibition around the oxacillin disc were slightly smaller when the plates were incubated in an atmosphere containing CO2 than in the ambient atmosphere. We have made a similar observation in respect of oxacillin, the MICs of this antibiotic being, on average, one two-fold dilution lower in air than in either an aerobic or a CO2-enriched atmosphere; however, the susceptibility category was the same regardless of the atmospheric conditions. Jorgensen et al. compared the MICs of penicillin, as determined by the Etest and microbroth dilution methods,

599 © 1997 The British Society for Antimicrobial Chemotherapy

Correspondence

600

Correspondence and reported a correlation of only 74% when the media were incubated under aerobic conditions and 92.5% when the atmosphere was enriched with CO2.5,6 The MICs, as determined by the Etest method, were a mean of one twofold dilution higher when the atmosphere contained CO2, compared with those determined in air. In contrast, we observed correlations of between 95% and 100%, regardless of the atmospheric condition. Moreover, the numbers of minor discrepancies detected by us were lower than those reported by other investigators.6 This study has demonstrated that, when determining the MICs of -lactams by the agar dilution method, the results are not influenced by the atmospheric conditions during incubation. However, this conclusion cannot be extrapolated to other classes of antibiotics.

Sir, Important variables of antimicrobial susceptibility testing, such as the MIC and MBC, are almost invariably determined with planktonic bacteria in broth culture. Bacteria within biofilms have been shown to be less susceptible to antibiotics; this has been attributed to the growth conditions and nutrient supply within the biofilms and to the presence of a glycocalyx which is impermeable to antimicrobial agents.1,2 It has been proposed that the term ‘biofilm eradicating concentration’ (BEC) should be used in place of the conventional MIC and MBC in studies in which the antimicrobial susceptibilities of bacteria in biofilms are determined.3 A drawback to the study of bacteria in biofilms has been the lack of a simple and reproducible technique. However, this has changed with the recent development of artificial biofilms or Sorbarods. A Sorbarod consists of a paper sleeve containing a comReferences pacted concertina of cellulose fibres. These simple biofilms 1. National Committee for Clinical Laboratory Standards. (1993). have been used to establish continuous culture biofilms of Methods for Dilution Antimicrobial Susceptibility Tests for Bacteria both Staphylococcus aureus and Pseudomonas aeruginosa.4 that Grow Aerobically—Third Edition; Approved Standard M7-A3. In the present study, the susceptibilities of a penicillinNCCLS, Villanova, PA. susceptible clinical isolate of Streptococcus pneumoniae to 2. National Committee for Clinical Laboratory Standards. (1993). four -lactams were determined in Sorbarod biofilms and Performance Standards for Antimicrobial Disk Susceptibility the results compared with the MICs and MBCs determined Tests—Fifth Edition; Approved Standard M2-A5. NCCLS, Villa- by the traditional broth dilution method and with MICs nova, PA. determined by the Etest method. 3. Soussy, C. J., Acar, J., Chardon, H., Choutet, P., Courvalin, P., The penicillin-susceptible strain of S. pneumoniae Dabernat, H. et al. (1996). Comité de l’Antibiogramme de la studied here was isolated from the sputum of a patient Société Française de Microbiologie. Communiqué 1996. Path with community-acquired pneumonia and the antibiotics ologie Biologie 44, 1–8. tested included benzylpenicillin (Britannia, Redhill, UK), 4. Swenson, J. M., Hill, B. C. & Thornsberry, C. (1986). Screening ampicillin and co-amoxiclav (SmithKline Beecham, Worpneumococci for penicillin resistance. Journal of Clinical Micro thing, UK) and cefuroxime (Glaxo, Greenford, UK). MICs biology 24, 749–52. and MBCs were determined by a broth-dilution method 5. Jorgensen, J. H., Howell, A. W. & Maher, L. A. (1991). Quantita- described by Holt & Brown.5 Etest strips were obtained tive antimicrobial susceptibility testing of Haemophilus influenzae from AB Biodisk, Solna, Sweden and the MICs were deterand Streptococcus pneumoniae by using Etest. Journal of Clinical mined according to the manufacturer’s instructions. S. Microbiology 29, 109–14. pneumoniae ATCC 671310 was used as a control for 6. Jorgensen, J. H., Ferraro, M. J., McElmeel, M. L., Spargo, J., both the broth dilution and Etest methods of susceptibility Swenson, J. M. & Tenover, F. C. (1994). Detection of penicillin and testing. extended-spectrum cephalosporin resistance among Strepto The Sorbarods were provided by Ilacon, Tonbridge, coccus pneumoniae clinical isolates by use of the Etest. Journal of Kent, UK. Sorbarod biofilms were prepared by a modificaClinical Microbiology 32, 159–63. tion of the published method.4 Cellulose Sorbarods (10 mm in diameter and 20 mm in length) were placed in 10 cm lengths of silicone tubing with an internal diameter of 10 The use of Sorbarod biofilms to study the mm. These were then connected by tubing (0.8 mm internal antimicrobial susceptibility of a strain of bore) to a supply of Brain Heart Infusion (BHI; Oxoid, Streptococcus pneumoniae Unipath, Basingstoke, UK) broth. The effluent end of the Sorbarod was connected via plastic adaptors to replaceable J Antimicrob Chemother 1997; 40: 601–602 150 mL sterile glass bottles, thereby enabling collection of R. K. Budhani and J. K. Struthers* effluent fluid at any time during the investigations. Each Sorbarod was inoculated with 3 mL of a 10–12 h broth Department of Medical Microbiology, Manchester culture of the bacterium in BHI broth and BHI ‘feed’ broth Royal Infirmary and the University of Manchester, was delivered by a 12-channel 205U peristaltic pump Manchester, UK (Watson Marlow, Falmouth, UK) at a flow rate of 0.1 mL/min. The whole apparatus with 12 biofilms was placed *Corresponding author. in an incubator at 37°C. At designated times, biofilm efflu601

Correspondence ent was collected over 15 min and the numbers of viable planktonic bacteria were determined. Individual Sorbarods were then harvested in 5 mL of BHI broth and vortexed to disintegrate the cellulose matrix and the numbers of viable organisms in the bacterial suspension were determined.5 Steady state growth of the bacteria in the Sorbarods was obtained within 24 h and maintained for at least 96 h. Viable counts in the biofilm vortexate were normally 1015 cfu/L, while those in the biofilm effluent were c.1013 cfu /L. After 24 h, individual biofilms were exposed to a single antibiotic concentration in BHI broth for 18 h. The numbers of viable bacteria in the biofilm and biofilm effluent were then determined as described above.5 The MICs and MBCs of the four -lactams for the pneumococcal isolate, as determined by the broth dilution and Etest methods, ranged from 0.015 mg/L to 0.06 mg/L. The BECs and effluent MBCs were 0.03 mg/L, i.e. the same as, or within one two-fold dilution of, the results obtained by the conventional methods. These observations suggest that biofilms confer no protection against the activities of the antibiotics tested on this particular strain of S. pneumoniae, even when the bacterial numbers in the biofilms are very high. We have demonstrated here that the Sorbarod biofilm model is both simple and well-suited to studying the invitro susceptibility of an important respiratory pathogen and allows reproducible growth patterns to be obtained. Moreover, with a 12-channel peristaltic pump, sufficient numbers of Sorbarod biofilms can be employed to cover the same range of concentrations that are used in conventional tests for determining MICs and MBCs, thereby facilitating the determination of the BEC of a particular antibiotic. Placing a T-connector upstream of the biofilm makes it relatively easy to replace BHI as the growth medium with one containing an antibiotic at a defined concentration. Although the Sorbarod method may be labourintensive, one of its principal advantages is the ease with which the conditions of growth can be altered and antibiotics introduced without disturbing the bacteria growing within the biofilm. The growth conditions within biofilms that are created by this model are therefore more likely to approximate those occurring in vivo than are those that prevail during conventional MIC and MBC studies. We propose that Sorbarods might be useful for determining the BECs of both existing and novel antimicrobial agents for a wide range of bacterial pathogens.

References 1. Duguid, I. G., Evans, E., Brown, M. R. W. & Gilbert, P. (1992). Effect of biofilm culture on the susceptibility of Staphylococcus epidermidis to tobramycin. Journal of Antimicrobial Chemotherapy 30, 803–10. 2. Nichols, W. W. (1991). Biofilms, antibiotics and penetration. Reviews in Medical Microbiology 2, 177–81.

3. Anwar, H. & Costerton, J. W. (1990). Enhanced activity of combination tobramycin and piperacillin for eradication of sessile biofilm cells of Pseudomonas aeruginosa. Antimicrobial Agents and Chemotherapy 34, 1666–71. 4. Hodgson, A. E., Nelson, S. M., Brown, M. R. W. & Gilbert, P. (1995). A simple in-vitro model for growth control of bacterial biofilms. Journal of Applied Bacteriology 79, 87–93. 5. Holt, A. & Brown, D. (1989). Antimicrobial susceptibilty testing. In Medical Bacteriology: A Practical Approach (Hawkey, P. M. & Lewis, D. A., Eds), pp. 167–94. IRL Press, Oxford.

A time–kill study to evaluate the in-vitro activity of clofazimine in combination with cefotaxime against a penicillin- and cefotaximeresistant strain of Streptococcus pneumoniae J Antimicrob Chemother 1997; 40: 602–604 S. van Niekerka*, F. Huygensb and C. E. J. van Rensburgb a

Department of Medical Microbiology and bInflam mation and Immunity Research Unit of the Medical Research Council, Department of Immunology, Insti tute of Pathology, University of Pretoria, PO Box 2034, Pretoria 0001, South Africa *Corresponding author. Tel: +27-12-319-2256; Fax: +27-12-319-4886. Sir, Streptococcus pneumoniae is an important cause of both upper and lower respiratory tract infections. In the USA, this pathogen causes 35–45% of episodes of acute otitis media in children,1 and in South Africa, the annual incidence of pneumococcal pneumonia is approximately 2.1 per 1000.2 The incidence of infections caused by strains of pneumococci that are resistant to penicillin, and, more recently, to third-generation cephalosporins such as cefotaxime and ceftriaxone has increased at an alarming rate and stimulated an urgent search for alternative therapeutic regimens. The riminophenazine, clofazimine, an agent that is commonly used in the treatment of patients with leprosy, and its analogues, are active against most Gram-positive bacteria. The present study was undertaken to investigate the activity of clofazimine in combination with a cell-wall active agent, cefotaxime, against a penicillin- and cefotaxime-resistant strain of pneumococcus. We chose to use time–kill studies for this purpose in order to assess the bactericidal activities of antibiotic regimes that might be candidates for the treatment of patients with serious pneumococcal infections.

602

Correspondence A clinical isolate of S. pneumoniae was identified according to standard laboratory methods. The oxacillin disc diffusion test was used initially to screen strains for resistance to penicillin. MICs and MBCs of penicillin (Novo Nordisk Pty Ltd, Johannesburg, Gauteng, South Africa) and cefotaxime (Roussel Laboratories, Sandton, Gauteng, South Africa) were determined by both the broth macrodilution method3 and the Etest method (AB Biodisk, Davies Diagnostics, Randburg, Gauteng, South Africa). Clofazimine (synthesized by Dr J. F. O’Sullivan, Depart-

ment of Chemistry, University College of Dublin, Republic of Ireland) was dissolved in 100% ethanol to give a final concentration of 2 g/L and the MICs of this drug were determined by the agar dilution method as recommended by the National Committee for Clinical Laboratory Standards.4 Killing curves were determined for cefotaxime and clofazimine, alone and in combination; the concentrations used were equivalent to 0.25 MIC of each drug. The methodology was that recommended by the NCCLS.5 Briefly, the initial inoculum was 108 cfu/L. Aliquots were obtained at 0, 2, 4, 6, 8 and 24 h, serial ten-fold dilutions were inoculated on to Mueller–Hinton agar supplemented with 5% horse blood and viable counts were determined after overnight incubation at 37°C. Each experiment was repeated four or five times and S. pneumoniae ATCC 49619 was used as a control. The MICs of the test strain for penicillin and cefotaxime were both 2 mg/L; this defines the organism as resistant to both agents according to NCCLS interpretative criteria for S. pneumoniae.6 The results of the time–kill studies are shown in the Figure. With synergy being defined as a reduction in the number of cfu/mL of 2 log 10, compared with the activity of the single most active component, it is evident that the combination of cefotaxime and clofazimine exhibited synergic activity against both the test and control strains after 8 h of antibiotic exposure; at this time no viable organisms were detected. The results of this in-vitro study confirm that the combination of cefotaxime and clofazimine exhibits rapid bactericidal activity against a clinical isolate of S. pneumo niae. This combination may therefore be an effective therapy for pneumococcal infections caused by resistant strains. However, even before clinical trials are undertaken, it will be necessary to evaluate the activity of the combination against other resistant isolates. In addition, the precise mechanism of the synergic activity of the two drugs warrants further investigations.

Acknowledgement This work was supported by a grant from Adcock Ingram Pharmaceuticals Ltd, South Africa.

References

Figure. Killing kinetics for clofazimine and/or cefotaxime against a penicillin- and cefotaximine-resistant clinical isolate of (a) S. pneumoniae (– –, clofazimine 0.25 mg/L; - - - - - - , cefotaxime 0.5 mg/L; —, the combination; and , control) and (b) S. pneumoniae ATCC 49619 (– –, clofazimine 0.125 mg/L; - - - - - - , cefotaxime 0.125 mg/L; —, the combination; and , control).

1. Block, S. L., Harrison, C. J., Hendrick, J. A., Tyler, R. D., Smith, R. A., Keegan, E. et al. (1995). Penicillin-resistant Streptococcus pneumoniae in acute otitis media: risk factors, susceptibility patterns and antimicrobial management. Pediatric Infectious Disease Journal 14, 751–9. 2. Klugman, K. P. (1990). Pneumococcal resistance to antibiotics. Clinical Microbiology Reviews 3, 171–96. 3. Van Rensburg, C. E. J., Joonè, G. K., O’Sullivan, J. F. & Anderson, R. (1992). Antimicrobial activities of clofazimine and B669 are

603

Correspondence by the National Committee for Clinical Laboratory Standards.3 4. National Committee for Clinical Laboratory Standards. (1993). Quinupristin/dalfopristin exhibited good activity against Performance Standards for Antimicrobial Susceptibility Testing— S. pneumoniae, with MICs falling in the range 0.25–4 mg/L; Fifth Informational Supplement: Update Tables for NCCLS the MIC50 and MIC90 were 0.5 mg/L and Antimicrobial Susceptibility Testing Standards. NCCLS, Villanova, 1 mg/L, respectively. The distribution of the MICs of PA. quinupristin/dalfopristin for pneumococci that were resist5. National Committee for Clinical Laboratory Standards. (1993). ant to either penicillin or erythromycin overlapped Methods for Determining Bactericidal Activity of Antimicrobial markedly with those for isolates that were susceptible to Agents; Approved Standard M26-P. NCCLS, Villanova, PA. these agents. For isolates of S. aureus, MICs were in the 6. National Committee for Clinical Laboratory Standards. (1995). range 0.25–1 mg/L (MIC and MIC , 1 mg/L). There was 50 90 Performance Standards for Antimicrobial Susceptibility Testing; no difference between staphylococci that were resistant to Approved Standard M2-A5. NCCLS, Villanova, PA. methicillin and those that were susceptible in terms of susceptibility to quinupristin/dalfopristin. Quinupristin/dalfopristin also exhibited good activity against several species of enterococci. The MICs for 50 of the 51 isolates of Enterococcus faecium were either 0.5 or 1 mg/L, the MIC for the remaining isolate being 4 mg/L. The MICs for the four strains of Enterococcus avium and six In-vitro activity of quinupristin/dalfopristin strains of Enterococcus gallinarum ranged from 1 to 2 mg/L (Synercid) against isolates of Streptococcus and from 2 to 4 mg/L, respectively, and the MICs for all five pneumoniae, Staphylococcus aureus and isolates of Enterococcus raffinosus were 2 mg/L. Enterococcus spp. Quinupristin/dalfopristin was much less active against the Enterococcus faecalis strains than other enterococci, MICs J Antimicrob Chemother 1997; 40: 604–605 falling within the range 8–32 mg/L. The MICs of quinupristin/dalfopristin for the five strains of Entero A. P. Johnson*, M. Warner and D. C. E. Speller coccus casseliflavus were between 2 and 8 mg/L. Antibiotic Reference Unit, Laboratory of Hospital Quinupristin/dalfopristin exhibited comparable activity Infection, Central Public Health Laboratory, Colin - against isolates of enterococci, regardless of their susceptibilities to other antimicrobial agents such as glycopeptides dale, London NW9 5HT, UK and gentamicin (high-level). Although a breakpoint that defines resistance to *Corresponding author. Tel: +44-181-200-4400; Fax: quinupristin/dalfopristin has not yet been formally +44-181-200-7449; E-mail: [email protected] agreed, some investigators have used a provisional value of 4 mg/L, based on pharmacokinetic parameters.4 According Sir, The emergence and spread of antibiotic-resistant to this breakpoint, all of the strains of S. pneumoniae, Gram-positive bacteria, including penicillin- and S. aureus, E. faecium, E. avium, E. gallinarum and E. erythromycin-resistant pneumococci, methicillin-resistant raffinosus tested in the present study would be considered Staphylococcus aureus (MRSA) and glycopeptide- susceptible to quinupristin/dalfopristin, whereas the 16 resistant enterococci, are posing increasingly difficult strains of E. faecalis would be classified as resistant. The therapeutic problems, particularly in hospitals.1,2 For this five E. casseliflavus strains clustered around the proreason, there is a continuing need for the development of visional breakpoint, with two being susceptible and three new antimicrobial agents with activities against bacteria resistant. Others have reported similar activities for that are resistant to currently available drugs. We report quinupristin/dalfopristin against these species.4–6 The activities of quinupristin/dalfopristin against a wide here the results of a study of the in-vitro activity of range of Gram-positive pathogens, including pneumococci, quinupristin/dalfopristin against isolates of Streptococcus staphylococci and some species of enterococci, mean that pneumoniae, S. aureus and Enterococcus spp. submitted this drug might be a useful addition to the current antibiotic to the Central Public Health Laboratory (CPHL) from hosarmamentarium. The observations made both in the pitals in the UK during 1995 and 1996. As a common reason present study and by others4–6 that quinupristin/ for referral of isolates to the CPHL is confirmation of antibiotic resistance, a significant proportion of the dalfopristin is active in vitro against bacteria that are isolates tested were resistant to one or more antibiotics resistant to other antibiotics, such as penicillin- and/or erythromycin-resistant pneumococci, MRSA and glyco(Table). peptide-resistant enterococci, suggest that this agent might Quinupristin/dalfopristin was supplied by Rhône– be particularly useful in clinical settings where the Poulenc Rorer (West Malling, UK) and the MICs therapeutic options are limited by problems of resistance. were determined by an agar dilution method described mediated by lysophospholipids. Chemotherapy 36, 2729–35.

Antimicrobial

Agents

and

604

Correspondence Table. Antimicrobial resistance patterns of the study strains

Bacterium (no. of strains)

Percentage of isolates of intermediate susceptibility or fully resistant to indicated agent Amp Pen Met Gen Van Tei Ery Cip Chl Rif Dox QD

S. pneumoniae (86) S. aureus (78) E. faecium (51) E. faecalis (16) E. avium (4) E. casseliflavus (5) E. gallinarum (6) E. raffinosus (5)

NT NT 100 0 0 0 0 5

26 100 NT NT NT NT NT NT

NT 85 NT NT NT NT NT NT

NT 30 45a 43a 25a 0a 0a 0a

0 0 80 69 50 100 100 40

NT 0 76 63 50 0 0 40

49 78 100 100 50 100 83 100

NT 63 100 88 100 80 83 60

NT NT 84 81 75 20 17 20

NT 9 90 50 0 60 67 100

NT NT 33 94 100 0 17 100

0 0 0 100 0 60 0 0

a

Refers to high-level gentamicin resistance. Abbreviations: Amp, ampicillin; Pen, penicillin; Met, methicillin; Gen, gentamicin; Van, vancomycin; Tei, teicoplanin; Ery, erythromycin; Cip, ciprofloxacin; Chl, chloramphenicol; Rif, rifampicin; Dox, doxycycline; QD, quinupristin/dalfopristin; NT, not tested.

Acknowledgement We thank Rhône–Poulenc Rorer for financial support for this work.

Antimicrobial activity of LB20304, a fluoronaphthyridone, tested against anaerobic bacteria

References

J Antimicrob Chemother 1997; 40: 605–607 Francesc Marco†, Mary S. Barrett and Ronald N. Jones*

1. Appelbaum, P. C. (1992). Antimicrobial resistance in Strepto coccus pneumoniae: an overview. Clinical Infectious Diseases 15, 77–83. 2. Johnson, A. P. & Woodford, N. (1993). Bacterial antibiotic resistance. Current Opinion in Infectious Diseases 6, 515–9. 3. National Committee for Clinical Laboratory Standards. (1993). Methods for Antimicrobial Susceptibility Tests for Bacteria that Grow Aerobically, Third Edition: Approved Standard M7-A3. NCCLS, Villanova, PA. 4. Johnson, C. C., Slavoski, L., Schwartz, M., May, P., Pitsakis, P. G., Shur, A. L. et al. (1995). In-vitro activity of RP59500 (quinupristin/dalfopristin) against antibiotic-resistant strains of Streptococcus pneumoniae and enterococci. Diagnostic Micro biology and Infectious Disease 21, 169–73. 5. Fass, R. J. (1991). In-vitro activity of RP59500, a semisynthetic injectable pristinamycin, against staphylococci, streptococci, and enterococci. Antimicrobial Agents and Chemotherapy 35, 553–9. 6. Collins, L. A., Malanoski, G. J., Eliopoulos, G. M., Wennersten, C. B., Ferraro, M. J. & Moellering, R. C. (1993). In-vitro activity of RP59500, an injectable streptogramin antibiotic, against vancomycin-resistant Gram-positive organisms. Antimicrobial Agents and Chemotherapy 37, 598–601.

Medical Microbiology Division, Department of Pathology, 5232 RCP, University of Iowa College of Medicine, Iowa City, IA 52242, USA †Present address: Microbiology Laboratory, Hospital Clinic, University of Barcelona, Villarroel 170, 08036-Barcelona, Spain. *Corresponding author. Tel: +1-319-356-2990; Fax: +1-319-356-4916. Sir, Most currently available fluoroquinolones exhibit limited or negligible activity against some Gram-positive aerobic and many anaerobic bacterial pathogens. In the last few years, however, newer fluoroquinolone compounds have been synthesized that have improved activity against these microorganisms while retaining broad-spectrum activity against Gram-negative bacilli.1–3 LB20304 is a novel fluoronaphthyridone that contains a pyrrolidine substitution.4 An animal model study has documented reduced central nervous system toxicity for LB20304 compared with ciprofloxacin or ofloxacin and good bioavailability after oral administration.5 LB20304 activity against most species of Enterobacteriaceae and Pseudomonas aeruginosa is two- to four-fold less than ciprofloxacin.6–8 Against Gram-positive organisms

605

Correspondence Table. Antimicrobial activity of LB20304 and five comparison compounds tested against 119 strains of anaerobic bacteria using reference methodsa Organism (no. tested)

Antimicrobial agent

Bacteroides fragilis (35)

LB20304 Sparfloxacin Trovafloxacin Cefoxitin Clindamycin Metronidazole LB20304 Sparvfloxacin Trovafloxacin Cefoxitin Clindamycin Metronidazole LB20304 Sparfloxacin Trovafloxacin Cefoxitin Clindamycin Metronidazole LB20304 Sparfloxacin Trovafloxacin Cefoxitin Clindamycin Metronidazole LB20304 Sparfloxacin Trovafloxacin Cefoxitin Clindamycin Metronidazole LB20304 Sparfloxacin Trovafloxacin Cefoxitin Clindamycin Metronidazole

B. fragilis group (15)b

Fusobacterium spp. (10)c

Prevotella spp. (22)d

Clostridium spp. (19)e

Peptostreptococcus spp. (18)

MIC (mg/L) 50% 90% 1 2 0.5 4 0.5 1 1 2 0.5 16 2 1 0.25 1 0.5 0.5 0.25 0.5 4 4 1 1 0.25 0.5 0.25 0.5 0.25 1 0.25 1 0.25 0.5 0.5 0.5 0.25 0.5

1 2 0.5 16 16 1 16 4 1 32 16 1 4 8 1 0.5 0.25 0.5 16 8 2 2 0.25 2 1 8 1 8 16 2 2 8 2 1 2 2

Range

% Susceptiblea

0.5–8 1– 8 0.25–4 2–16 0.25– 16 0.5–2 0.5– 16 1–8 0.25–4 2– 32 1– 16 0.5–2 0.25–8 0.5– 8 0.25–1 0.5–4 0.25 0.5 1–16 2–8 0.5–2 0.5–8 0.25 0.5–4 0.25–2 0.12–8 0.25–2 0.5– 32 0.25– 16 0.5–2 0.25–8 0.12– 8 0.25–8 0.5–2 0.25–.16 0.5– 32

NA NA NA 100 80 100 NA NA NA 80 80 100 NA NA NA 100 100 100 NA NA NA 100 100 100 NA NA NA 95 79 100 NA NA NA 100 94 94

% with MICa 1 mg/L 2 mg/L 91 23 97 NA NA NA 67 20 93 NA NA NA 80 50 100 NA NA NA 5 0 77 NA NA NA 95 89 95 NA NA NA 89 72 83 NA NA NA

91 94 97 NA NA NA 87 80 93 NA NA NA 80 80 100 NA NA NA 46 41 100 NA NA NA 100 89 100 NA NA NA 94 72 94 NA NA NA

a

Two concentrations were used to compare the fluoroquinolones only. NCCLS (1993) breakpoints were used for the non-quinolone drugs. NA, Not applicable. b Includes: Bacteroides disasonis (two strains), Bacteroides ovatus (four strains), Bacteroides thetaiotaomicron (two strains), Bacteroides uniformis (one strain), Bacteroides vulgatus (two strains), and other Bacteroides spp. (four strains). c Includes: Fusobacterium necrophorum (two strains), Fusobacterium nucleatum (five strains), Fusobacterium varium (one strain), and Fusobacterium spp. (two strains). d Includes: Prevotella bivia (five strains), Prevotella buccae (four strains), Prevotella loescheii (two strains), Prevotella melaninogenica (one strain), Prevotella denticola (one strain), and Prevotella spp. (nine strains). e Includes: Clostridium butyricum (one strain), Clostridium difficile (one strain), Clostridium perfringens (nine strains), Clostridium ramosum (one strain), Clostridium tertium (one strain), and other Clostridium spp. (six strains).

(including oxacillin-susceptible Staphylococcus aureus, sparfloxacin (four- to 32-fold) or trovafloxacin (equal to Streptococcus pneumoniae, and -haemolytic Strepto - eight-fold),6–8 but more limited activity was suggested coccus spp.) LB 20304 displays more potent activity than against a small sample of anaerobes.6 We have therefore 606

Correspondence tested the in-vitro activity of LB20304 against 119 anaerobic bacteria compared with that of sparfloxacin, trovafloxacin, and three reference drugs generally considered active against anaerobes (metronidazole, cefoxitin, and clindamycin). Antimicrobial compounds were obtained as follows: LB20304 from Biotech Research Institute (LG Chemical Ltd, Tae Jon, Korea); trovafloxacin from Pfizer Central Research (Groton, CT, USA) and sparfloxacin from Rhône–Poulenc Rorer (Collegeville, PA, USA). Metronidazole, cefoxitin, and clindamycin were supplied by their respective manufacturers. The 119 anaerobic strains were recent clinical isolates kept frozen at 70°C until tested. Susceptibility testing was performed according to the M11-A3 technique recommended by the National Committee for Clinical Laboratory Standards (NCCLS).9 In brief, this consisted of a reference agar dilution susceptibility testing method performed with a 105 cfu/spot inoculum, 48 h incubation, anaerobic conditions, and 5% sheep blood supplemented brucella agar. Quality control was assured when NCCLS approved limits were available with the following strains: Bacteroides fragilis ATCC 25285 and Bacteroides thetaiotaomicron ATCC 29741. The Table summarizes the MICs of six drugs tested against 100 clinical strains. At 2 mg/L, LB20304 inhibited 80%, 87% and 91% of Fusobacterium spp., B. fragilis group and B. fragilis isolates, respectively. At the same concentration, only 46% of Prevotella spp. were inhibited. The activity of LB20304 against all Gramnegative anaerobes was comparable to that of sparfloxacin, but was two- to 16-fold less than that of trovafloxacin. Against Clostridium spp. and Peptostrepto coccus spp., LB20304 was equipotent to trovafloxacin. Our findings for trovafloxacin activity generally agreed with the values reported by Sprangler et al.3 except for Peptostreptococcus spp. (MIC90: 2 mg/L vs 0.25 mg/L, respectively). The rank order of potency for the three quinolones against all strains tested was trovafloxacin LB20304 sparfloxacin, and that order was identical when the percentage of strains inhibited at 1 mg/L or 2 mg/L were considered. Among the reference compounds tested, metronidazole was the most active drug against all Gram-negative bacilli and Clostridium spp. (99% susceptible). Cefoxitin also showed good activity (97% of strains susceptible), and it was the only drug that inhibited all Peptostreptococcus spp. studied. Clindamycin was the least potent of the three non-quinolone comparison drugs (84% of strains susceptible). Therefore LB20304 not only exhibits broad-spectrum activity against Gram-negative and -positive aerobic bacteria,4,6–8 but also has moderate activity (by reference NCCLS methods) against anaerobic pathogens. Additional studies to evaluate the potential utility of this compound in clinical situations appear warranted.

References 1. Cohen, M. A., Yoder, S. L. & Talbot, G. H. (1996). Sparfloxacin worldwide in vitro literature: isolate data available through 1994. Diagnostic Microbiology and Infectious Disease 25, 53–64. 2. Gootz, T. D., Brighty, K. E., Anderson, M. R., Schmieder, B. J., Haskell, S. L., Sutcliffe, J. A. et al. (1994). In vitro activity of CP-99,219, a novel 7-(3-azabicylo[3.1.0]hexyl) naphthyridione antimicrobial. Diagnostic Microbiology and Infectious Disease 19, 235–43. 3. Sprangler, S. K., Jacobs, M. R. & Appelbaum, P. C. (1994). Activity of CP 99,219 compared with those of ciprofloxacin, grepafloxacin, metronidazole, cefoxitin, piperacillin, and piperacillin–tazobactam against 489 anaerobes. Antimicrobial Agents and Chemotherapy 38, 2471–6. 4. Kim, Y. K., Choi, H., Kim, S. H., Chang, J.-H., Nam, N.-H., Kim, Y.-Z. et al. (1995). Synthesis and antibacterial activities of LB20304: a new fluoronaphthyridone antibiotic containing novel oxime functionalized pyrrolidine. In Program and Abstracts of the 35th Interscience Conference on Antimicrobial Agents and Chemotherapy, San Francisco, CA, 1995. Abstract F204, p. 148. American Society for Microbiology, Washington, DC. 5. Kim, S. I., Kim, H. J., Kwak, J. H., Kim, I. C. & Lee, C. H. (1995). Safety evaluation of LB20304, a new quinolone antibiotic. Journal of Applied Pharmacology 3, 322–6. 6. Cormican, M. G. & Jones, R. N. (1997). Antimicrobial activity and spectrum of LB20304, a novel fluoronaphthyridone. Anti microbial Agents and Chemotherapy 41, 204–11. 7. Oh, J. I., Paek, K. S., Kim, M. Y., Seo, M. K., Lee, Y. H., Hong, C. H. et al. (1995). In vitro and in vivo antibacterial activities of LB20304, a new fluoronaphthyridone. In Program and Abstracts of the 35th Interscience Conference on Antimicrobial Agents and Chemotherapy, San Francisco, CA, 1995. Abstract F205, p. 148. American Society for Microbiology, Washington, DC. 8. Oh, J. I., Paek, K. S., Ahn, M. Y., Kim, M. J., Hong, C. Y., Kim, I. C. et al. (1996). In vitro and in vivo evaluations of LB20304, a new fluoronaphthyridone. Antimicrobial Agents and Chemotherapy 40, 1564–8. 9. National Committee for Clinical Laboratory Standards. (1993). Approved Standard M11-A3. Methods for Antimicrobial Suscept ibility Testing of Anaerobic Bacteria, Third Edition. NCCLS Villanova, PA.

Glycopeptide-resistant enterococci in Northern Ireland: first reported outbreak J Antimicrob Chemother 1997; 40: 607–608 E. A. Rizkalla, J. E. Moore, S. A. Marshall and P. G. Murphy Department of Bacteriology and Northern Ireland Public Health Laboratory, Belfast City Hospital, Lisburn Road, Belfast, BT9 7AD, UK

607

Correspondence Sir, We report here the first laboratory isolation of and outbreak caused by glycopeptide-resistant enterococci (GRE) in Northern Ireland. Although high-level gentamicin resistance in enterococci was first reported in Northern Ireland in 1991, 1 all strains until now have been susceptible to the glycopeptides; with the exception of one strain of Enterococcus faecium isolated in December 1991 from an abscess.2 In December 1996, a GRE was detected in the urine of a patient with a haematological malignancy. All patients and staff on the relevant ward were initially screened for faecal colonization followed by further screening of patients and the ward environment 1 month later. Surveillance specimens were obtained from 17 patients, 17 members of staff and 24 environmental sites. Altogether, ten specimens, seven from patients and three from the environment, including a mattress, a sluice bowl and a bed-frame, were positive for GRE. Coincident with these events, a patient from the affected ward was admitted to the intensive care unit where she subsequently died. Both GRE and a coagulase-negative staphylococcus were isolated from a postmortem blood culture obtained from this patient. Both isolates were identified by API STREP (bioMérieux) and their antimicrobial susceptibilities determined by the disc diffusion and broth microdilution methods. The MICs of teicoplanin for the strains were determined by the Etest method and molecular epidemiological typing was performed by the random amplification of polymorphic DNA (RAPD) method3, Basingstoke, UK with a 25-mer arbitrary oligonucleotide primer. The results of the various investigations identified all of the isolates as E. faecium exhibiting high-level resistance to vancomycin (MIC 64 mg/L) and teicoplanin (MIC 256 mg/L), consistent with the VanA phenotype.4 The strains were resistant to all but two of the antibiotics tested including gentamicin (resistant to a 100 g disc; MIC 32 mg/L); the exceptions were chloramphenicol (MIC 4 mg/L) and tetracycline (MIC 2 mg/L). RAPD analysis revealed two

distinguishable patterns amongst the 11 isolates examined, suggesting that more than one source was implicated. Implementation of heightened infection control precautions and restricted use of vancomycin and cephalosporins prevented further transmission of GRE on the outbreak ward. The province of Northern Ireland is geographically distinct from the rest of UK and there are many examples of epidemiological variation in antibiotic resistance here compared with Britain, as well as across the land border with the Republic of Ireland, where GRE have already been isolated (personal communication, E. G. Smyth). A previous all-Ireland survey of resistance amongst enterococci revealed that only 2% of isolates exhibited low-level resistance to vancomycin (4 mg/L MIC 8 mg/L) and that all strains were susceptible to teicoplanin (VanB phenotype); no strains isolated in Northern Ireland were categorized as GRE.5 The outbreak of GRE in the province is therefore of considerable significance from the point of view of both epidemiological surveillance and clinical practice.

References 1. Woodford, N., George, R. C., McNamara, E., Smyth, E., Namnyak, S. & Uttley, A. H. C. (1991). Enterococcus faecium with high-level resistance to gentamicin. Lancet 337, 1356. 2. Editorial. (1996). Vancomycin resistant enterococci in hospitals in the United Kingdom–1995. Communicable Diseases Report 6 (31), 1. 3. Williams, J. G. K., Hanafey, M. K., Rafalski, J. A. & Tingey, S. V. (1993). Genetic analysis using random amplified polymorphic DNA markers. Methods in Enzymology 218, 704–40. 4. Arthur, M. & Courvalin, P. (1993). Genetics and mechanisms of glycopeptide reistance in enterococci. Antimicrobial Agents and Chemotherapy 37, 2563–71. 5. McNamara, E. B., King, E. M. & Smyth, E. G. (1995). A survey of antimicrobial susceptibility of clinical isolates of Enterococcus spp. from Irish hospitals. Journal of Antimicrobial Chemotherapy 35, 185–9.

608