JAC Advance Access originally published online on February 28, 2006
Journal of Antimicrobial Chemotherapy 2006 57(5):992-998; doi:10.1093/jac/dkl052
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Antimicrobial susceptibility of anaerobic bacteria in New Zealand: 1999–2003
Department of Microbiology, LabPlus, Auckland District Health Board, Auckland, New Zealand
* Corresponding author. Tel: +64-9-379-7440; Fax: +64-9-307-8922; E-mail: sallyrob{at}adhb.govt.nz
Received 14 March 2005; returned 5 October 2005; revised 5 December 2005; accepted 7 February 2006
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResultsDiscussionTransparency declarationsReferences |
|---|
Objectives: Routine susceptibility testing of all anaerobic organisms is not advocated, but it is useful for laboratories to test periodically for anaerobic organisms and provide local susceptibility data to guide therapy. This study reports the national trend of antibiotic susceptibility of clinically significant anaerobes in New Zealand.
Methods: Clinical isolates were tested using standardized methods against a range of antibiotics commonly used to treat anaerobic infections. Susceptibility was determined using NCCLS criteria. The change in susceptibility trends between this study and earlier studies was measured by comparing the geometric mean of the MIC.
Results: A total of 364 anaerobes were tested. Penicillin had poor activity against Bacteroides spp., Prevotella spp., Eubacterium spp., Clostridium tertium and Veillonella spp. In general, Fusobacterium spp., Bacteroides ureolyticus, Propionibacterium spp., Clostridium perfringens and anaerobic streptococci isolates, with the exception of Peptostreptococcus anaerobius, were penicillin susceptible. Amoxicillin/clavulanate showed good activity against most anaerobes, but resistance was seen with Bacteroides fragilis group and P. anaerobius isolates. Cefoxitin was more active than cefotetan, particularly against non-B. fragilis species, Eubacterium spp. and P. anaerobius. Meropenem and imipenem showed good activity against all anaerobes, with only 2 and 4% of Bacteroides spp., respectively, showing resistance. With the exception of Propionibacterium acnes isolates, which are predictably resistant, metronidazole was active against all anaerobes tested. There has been little change in susceptibility since 1997.
Conclusions: Metronidazole, cefoxitin, piperacillin/tazobactam and amoxicillin/clavulanate remain good empirical choices when anaerobes are expected in our setting. No clinically relevant changes in susceptibility over time were found.
Keywords: anaerobes , susceptibility patterns , susceptibility trends
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionTransparency declarationsReferences |
|---|
Anaerobic organisms dominate our indigenous flora. They cause infections associated with surgical procedures and pulmonary, genitourinary, and skin and soft tissue infections.1 Antimicrobial therapy is often empirical and culture results become available some time after the initiation of antimicrobial therapy. For this reason, and given the cost issues and technical difficulties associated with the isolation and identification of anaerobes, it is not surprising that the work-up of clinical specimens for anaerobes is not performed routinely.
Antimicrobial resistance among anaerobic organisms is increasing and clinical failure has been reported in patients receiving inactive treatment.2
This study reports on the national trend of antibiotic susceptibility of clinically significant anaerobes in New Zealand and highlights changes in susceptibility profiles over time by comparing the results with those of previous studies.3,4
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionTransparency declarationsReferences |
|---|
Isolates
Non-duplicated anaerobic isolates from our laboratory, the National Reference Laboratory for anaerobes, or referred from other New Zealand laboratories between December 1999 and February 2003 were included in the study. Where there were sufficient stored isolates identified to the species level a representative sample was tested; otherwise the isolates were grouped by genus.
The anaerobes were isolated from the following clinical specimens: pus/wound/aspirates, 195 (53%); blood, 101 (28%); tissue, 39 (11%); synovial fluid, 5 (1%); and other sites, 24 (7%). Isolates were identified using standard methods,5 stored at –80°C in skimmed milk and subcultured twice before susceptibility testing was performed.
The isolates tested were Bacteroides fragilis (49), B. fragilis group (51), Bacteroides ureolyticus (10), Bacteroides spp. (10), Clostridium perfringens (20), Clostridium septicum (12), Clostridium tertium (6), Clostridium clostridioforme (14), Clostridium spp. (20), Eubacterium spp. (10) Fusobacterium spp. (47), Peptostreptococcus anaerobius (20), Peptostreptococcus asaccharolyticus (20), Peptostreptococcus magnus (20), Peptostreptococcus micros (21), Prevotella spp. (45), Fusobacterium necrophorum (25), Fusobacterium nucleatum (22), Propionibacterium spp. (10) and Veillonella spp. (9).
Susceptibility testing
Agar dilution. Susceptibilities for all isolates except C. septicum were determined using the NCCLS reference agar dilution procedure.6 Growth from a 40 h culture was suspended in thioglycolate broth to a density equivalent to a McFarland 0.5 standard then inoculated onto Brucella agar (Difco, Becton Dickinson, Sparks MD21152, USA) supplemented with 5% lysed sheep blood, 5 mg/L haemin and 1 mg/L vitamin K. Plates were inoculated with 
105 cfu per spot, using an antibiotic sensitivity replicator (H.I. Clements Pty Ltd., Sydney, Australia). All plates were incubated in an anaerobic cabinet (dw Scientific, Don Whitley Scientific Ltd, West Yorkshire, England) for 48 h.
Each batch included, as quality control isolates, B. fragilis ATCC 25285, Bacteroides thetaiotaomicron ATCC 29741 and Eubacterium lentum ATCC 43055.
Antibiotics tested were penicillin (Biochemie GmbH), amoxicillin/clavulanate (GlaxoSmithKline Pharmaceuticals), piperacillin-tazobactam (Wyeth), cefoxitin (Merck, Sharpe & Dohme), cefotetan (Wyeth), ceftriaxone (Roche), imipenem (Merck, Sharpe & Dohme), meropenem (AstraZeneca), clindamycin (Pharmacia) and metronidazole (Baxter). All antimicrobials were tested at doubling dilutions, from 128 to 0.06 mg/L (amoxicillin/clavulanate, clindamycin, imipenem, meropenem, metronidazole, penicillin) or 256 to 0.12 mg/L (cefotetan, cefoxitin, ceftriaxone, piperacillin/tazobactam).
Isolates were classified susceptible, intermediate or reduced susceptibility, or resistant to antibiotics using NCCLS interpretative criteria. Isolates were considered susceptible if the MIC for each antibiotic was as follows: penicillin 
0.5 mg/L, amoxicillin/clavulanate 4/2 mg/L, piperacillin/tazobactam 32/4 mg/L, cefoxitin 16 mg/L, cefotetan 16 mg/L, ceftriaxone 16 mg/L, clindamycin 2 mg/L, imipenem 4 mg/L, meropenem 4 mg/L and metronidazole 8 mg/L.
Etest method. C. septicum were tested using the Etest method (AB Biodisk, Solna, Sweden). Isolates were suspended in thioglyclolate broth to match the density of a McFarland 1 standard then inoculated onto Brucella agar supplemented with 5% lysed sheep blood, 5 mg/L haemin and 1 mg/L vitamin K. Penicillin and metronidazole were read after 24 h of incubation. Clindamycin was read at 48 h, following the manufacturer's instructions. B. fragilis ATCC 25285 was used as the control.
Comparison between testing periods
Comparisons of the results between study periods were undertaken for B. fragilis, C. perfringens, P. anaerobius and Propionibacterium acnes. The change in susceptibility trends between each study period was measured by comparing the geometric mean of the MIC.
Statistical methods
The geometric mean MIC results were compared using the Mann–Whitney U-test.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionTransparency declarationsReferences |
|---|
A total of 364 anaerobes were included in the study. The organisms tested; percentages of susceptible, intermediate and resistant organisms; MIC ranges; MIC50 and MIC90 for the anaerobic Gram-negative organisms and the anaerobic Gram-positive organisms are shown in Tables 1 and 2, respectively.
|
|
Penicillin had poor activity against Bacteroides spp. and Prevotella spp., with the exception of B. ureolyticus isolates, which were all penicillin susceptible. Penicillin resistance was found in the majority of Eubacterium spp. (90%) and Veillonella spp. (78%), with the MIC of penicillin between 1 and 4 mg/L. F. nucleatum isolates were all penicillin susceptible, but one F. necrophorum (4%) isolate was resistant to penicillin. All Propionibacterium spp. and C. perfringens were susceptible to penicillin. But only one-third (36%) of C. clostridioforme and no C. tertium isolates were susceptible to penicillin. Anaerobic streptococci were all susceptible to penicillin except for P. anaerobius.
Amoxicillin/clavulanate showed good activity against most anaerobes, but resistance was found in 10% of B. fragilis and 35% of P. anaerobius isolates. Only one isolate in the survey was resistant to piperacillin-tazobactam. This was a multiresistant B. fragilis, which was susceptible only to metronidazole.
Cefoxitin was more active than cefotetan, particularly against non-B. fragilis species, Eubacterium spp. and P. anaerobius. Ceftriaxone, as expected, showed poor activity against Bacteroides spp., except B. ureolyticus. However, Fusobacterium spp. and Veillonella spp. remained susceptible to ceftriaxone.
Meropenem and imipenem showed good activity against all anaerobes, with 2% and 4% of Bacteroides spp., respectively, showing resistance.
With the exception of P. acnes isolates, which were predictably resistant, metronidazole was active against all anaerobes tested.
There has been little change in susceptibility since 1997. The geometric mean MIC of clindamycin and metronidazole for B. fragilis increased from 0.25 to 1.29 mg/L (P < 0.001) and from 0.86 to 1.17 mg/L (P = 0.001), respectively, but remained within the susceptible range. No statistical change in geometric means of the antibiotics tested was observed for C. perfringens, P. anaerobius and P. acnes.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionTransparency declarationsReferences |
|---|
This is the third local survey looking at the antimicrobial susceptibility patterns of clinical isolates and it has allowed us to report on changes in susceptibility patterns over a 15 year period.
B. fragilis is the anaerobe most often isolated from intra-abdominal infections or from anaerobic bacteraemia.7 ß-Lactamase production, conferring resistance to penicillin, is common among B. fragilis. Metronidazole remains the most active agent; no resistant strains were reported locally. Meropenem and imipenem (carbapenems) had low levels of resistance at 2 and 4%, respectively. Resistance was more common amongst the other ß-lactam agents tested, cefoxitin, cefotetan and amoxicillin/clavulanate, with only 80–84% of isolates testing susceptible. Cefoxitin had greater activity than cefotetan.
Prevotella spp. were overall more susceptible than either B. fragilis or the B. fragilis group. A significant number of isolates produce ß-lactamase and only 16% were susceptible to penicillin.
With the exception of one isolate of F. necrophorum, all other Fusobacterium spp. tested were susceptible to all agents tested. Penicillin resistance owing to ß-lactamase production has been reported in F. nucleatum.8
C. perfringens isolates were susceptible to all agents tested. C. clostridioforme and C. tertium were more resistant to penicillin but were susceptible to carbapenems and metronidazole. The MIC90 of penicillin for C. tertium was 4 mg/L; using the previous NCCLS breakpoint of 8 mg/L these isolates would have been considered susceptible.6,9 The present breakpoint of 0.5 mg/L correlates well with ß-lactamase production in Prevotella spp. and Fusobacterium spp., and though C. clostridioforme produces ß-lactamase, to our knowledge this has not been shown in C. tertium.10,11
Eubacterium spp. are generally considered to be susceptible to penicillin. In this survey the MICs were in the range 1–4 mg/L, and therefore the isolates are shown to be resistant using current criteria but susceptible using older criteria.6,9 Good clinical data will be necessary to determine whether the current penicillin interpretation criterion is suitable for Eubacterium spp. and Veillonella spp.
The proposed reclassification of the Peptostreptococcus genus is based on molecular methods not available in most routine laboratories.12 In our laboratory all presumptive anaerobic Gram-positive cocci are tested against metronidazole; if it is resistant and grows under microaerophilic conditions, the isolate is called a microaerophilic Gram-positive coccus. In previous surveys identification criteria were less rigorous. Resistance to penicillin was seen with P. anaerobius (45%), MIC90 8 mg/L.
Comparison of the resistance rates between the different survey periods is limited by a number of factors. There have been changes to the methodology with the different media used. Supplemented BHI agar was used in the initial survey,3 supplemented WC agar in the second4 and supplemented Brucella agar in the present survey. While it has been reported that only minimal changes in MIC occur when any one of these media is used, small differences can make significant differences in reported parameters (MIC50, MIC90 and % S) when the MIC clusters around the breakpoint, as happens for some organism/antibiotic combinations.13 Some of the differences may also relate to differences in the identification of isolates.
There was no significant change documented in susceptibility between survey periods. Although for B. fragilis there was a significant increase in the geometric mean MIC for clindamycin, imipenem and metronidazole (P < 0.05), the increased geometric mean was within the susceptible range.
Based on our findings metronidazole, cefoxitin, piperacillin/tazobactam and amoxicillin/clavulanate remain good empirical choices when anaerobes are expected. The carbapenems should be reserved for situations where other organisms, especially resistant facultative Gram-negative bacilli, are expected. Ceftriaxone and less so cefotetan have poor activity and are not recommended for anaerobic cover.
|
Transparency declarations |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionTransparency declarationsReferences |
|---|
None to declare.
|
Acknowledgements |
|---|
We acknowledge Teena West for her assistance with the statistical analysis. No financial support was provided for this study.
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionTransparency declarationsReferences |
|---|
1. Finegold SM. Anaerobic bacteria: general concepts. In: Mandell GL, Bennett JE, Dolin R, eds. Mandell, Douglas and Bennett's Principles and Practice of Infectious Diseases, 5th edn. Philadelphia: Churchill Livingstone, 2000; 2519–37.
2. Nguyen MH, Yu VL, Morris AJ et al. Antimicrobial resistance and clinical outcome of Bacteroides bacteraemia: findings of a multicenter prospective observational trial. Clin Infect Dis 2000; 30: 870–6.[CrossRef][ISI][Medline]
3. Henderson G, Garner J, Morris AJ. Antimicrobial susceptibility of anaerobic bacteria in Auckland 1987–90. NZ Med J 1992; 105: 11–2.[Medline]
4. Shore KP, Pottumarthy S, Morris AJ. Susceptibility of anaerobic bacteria in Auckland: 1991–1996. NZ Med J 1999; 112: 424–6.[Medline]
5. Jousimies-Somer HR, Summanen P, Citron DM et al. Anaerobic Bacteriology Manual, 6th edn. Korea: Star Publishing Company, 2002.
6. National Committee for Clinical Laboratory Standards. Methods for Antimicrobial Susceptibility Testing of Anaerobic Bacteria—Sixth Edition: Approved Standard M11-A6. NCCLS, Wayne, PA, USA, 2004.
7. Goldstein EJC. Intra-abdominal anaerobic infections: bacteriology and therapeutic potential of newer antimicrobial carbapenem, fluoroquinolone and desfluoroquinolone therapeutic agents. Clin Infect Dis 2002;35 Suppl 1: 106–11.[CrossRef]
8.
Nyfors S, Kononen E, Syrjanen R et al. Emergence of penicillin resistance among Fusobacterium nucleatum populations of commensal oral flora during early childhood. J Antimicrob Chemother 2003; 51:107–12.
9. National Committee for Clinical Laboratory Standards. Performance Standards for Antimicrobial Susceptibility Testing; Sixth Information Supplement: Approved Document M100-S6. NCCLS, Wayne, PA, USA, 1995.
10. Kononen E, Saarela M, Kanervo A et al. ß-Lactamase production and penicillin susceptibility among different ribotypes of Prevotella melaninogenica simultaneously. Clin Infect Dis 1995; 20 Suppl 2: 364–6.
11.
Kononen E, Kanervo A, Salminen K et al. ß-lactamase production and antimicrobial susceptibility of oral heterogeneous Fusobacterium nucleatum populations in young children. Antimicrob Agents Chemother 1999; 43: 1270–73.
12.
Murdoch DA. Gram-positive anaerobic cocci. Clin Microbiol Rev 1998; 11: 81–120.
13. Roe DE, Finegold SM, Citron DM et al. Multilaboratory comparison of anaerobe susceptibility results using 3 different agar media. Clin Infect Dis 2002; 35 Suppl 1: 40–6.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  C.-Y. Liu, Y.-T. Huang, C.-H. Liao, L.-C. Yen, H.-Y. Lin, and P.-R. Hsueh Increasing Trends in Antimicrobial Resistance among Clinically Important Anaerobes and Bacteroides fragilis Isolates Causing Nosocomial Infections: Emerging Resistance to Carbapenems Antimicrob. Agents Chemother., September 1, 2008; 52(9): 3161 - 3168. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Nisbet, S. Briggs, R. Ellis-Pegler, M. Thomas, and D. Holland Propionibacterium acnes: an under-appreciated cause of post-neurosurgical infection J. Antimicrob. Chemother., November 1, 2007; 60(5): 1097 - 1103. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Kononen, A. Bryk, P. Niemi, and A. Kanervo-Nordstrom Antimicrobial Susceptibilities of Peptostreptococcus anaerobius and the Newly Described Peptostreptococcus stomatis Isolated from Various Human Sources Antimicrob. Agents Chemother., June 1, 2007; 51(6): 2205 - 2207. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||