M. D. Smith**, V. Wuthiekanun*, A. L. Walsh** and N. J. White***. 'Faculty of Tropical ... John Radcliffe Hospital, Oxford University, Oxford, UK ..... pseudomallei resistance to /J-lactam antibiotics due to alterations in the chromosomally encoded ...
Journal of Antimicrobial Chemotherapy (1996) 37, 611-615
In-vitro activity of carbapenem antibiotics against /Mactam susceptible and resistant strains of Burkholderia pseudomallei M. D. Smith**, V. Wuthiekanun*, A. L. Walsh** and N. J. White***
'Faculty of Tropical Medicine, Mahidol University, 420j6 Rajvithi Road, Bangkok 10400, Thailand;bCentre for Tropical Medicine, Nuffield Department of Clinical Medicine, John Radcliffe Hospital, Oxford University, Oxford, UK Four carbapenem antibiotics were tested for their in-vitro activities (MICs, MBCs and time-kill studies) against Burkholderia pseudomallei. The carbapenems were all more active than either ceftazidime or co-amoxiclav against strains of B. pseudomallei with a normal susceptibility pattern. Biapenem was the most active antibiotic tested. All four carbapenems retained bactericidal activity against 24 strains of B. pseudomallei with reduced susceptibility to ceftazidime and/or co-amoxiclav.
Introduction
Burkholderia (formerly Pseudomonas) pseudomallei is an environmental Gram-negative bacillus which is the cause of melioidosis, a potentially fatal infection of humans. Choice of treatment for this infection is limited because of the organism's resistance to aminoglycosides and the relatively poor activity of the fluoroquinolone compounds (Dance et al., 1989). Treatment with ceftazidime has reduced the mortality to approximately 40% (White et al., 1989) and a recent trial has shown that high dose co-amoxiclav has achieved a similar reduction in mortality (Suputtamongkol et al., 1994). However strains with resistance to one or both of these /Mactam antibiotics have emerged during treatment (Dance et al., 1991). Previous studies have shown that imipenem is one of the most active antibiotics against B. pseudomallei. In this study we have evaluated imipenem and other new carbapenem antibiotics for in-vitro activity against B. pseudomallei, including strains with reduced susceptibility or resistance to ceftazidime and co-amoxiclav.
Materials and methods
Bacterial strains Strains of B. pseudomallei, used in this study, were isolated from patients in Ubon Ratchatani, north-eastern Thailand. MICs were performed on 100 consecutive strains •Correspondence to: Dr N. J. White, Faculty of Tropical Medicine, Mahidol University, 420/6 Rajvithi Road, Bangkok 10400, Thailand. Tel. +66 (2) 246 0832; Fax: +66 (2) 246 7795.
0305-7453/96/030611 + 05 J12.00/0
611 %, 1996 The British Society for Antimicrobial Chemotherapy
612
M. D. Smith et al.
isolated during 1991-1992. Two of these strains were resistant to chloramphenicol and one strain was resistant to ceftazidime by disc diffusion. The remaining 97 strains were susceptible to the normal range of antibiotics with known activity against B. pseudomallei (Dance et al., 1989). Twenty of these strains were randomly selected for subsequent determination of MBCs. Subsequently, MICs and MBCs were determined for 24 strains of B. pseudomallei with reduced susceptibility or resistance to ceftazidime and/or co-amoxiclav (MIC ^ 16mg/L for both antibiotics). Reduced susceptibility to ceftazidime alone occurred in seven strains, to co-amoxiclav alone in seven strains, and to both antibiotics in ten strains. These strains have been collected since 1987, following the introduction of these antibiotics for the treatment of melioidosis. Three strains of B. pseudomallei were then selected for time-kill kinetic studies. These were 576a (normal susceptibility pattern), 704f (resistant to co-amoxiclav) and 979b (MIC = 256 mg/L for both ceftazidime and co-amoxiclav).
Antimicrobial agents The antibiotics used in this study were gifts from the manufacturers: amoxycillin, clavulanic acid (SmithKline Beecham Pharmaceuticals, UK), biapenem (Lederle Laboratories, UK), ceftazidime (Glaxo Laboratories Ltd., UK), imipenem (Merck Sharp & Dohme Ltd., UK), meropenem (ICI Pharmaceuticals, UK), panipenem (Sankyo Co. Ltd., Japan).
Susceptibility tests MICs were determined by an agar dilution method following NCCLS guidelines (1993), and MBCs were determined by a microtitre dilution method using Mueller-Hinton broth, as described previously (Dance et al., 1989). Time-kill kinetic studies were performed in Mueller-Hinton broth with a starting inoculum of approximately 5 x 105 cfu/mL, as described previously (Smith et al., 1994). Antibiotics were used at a concentration of 4 x MIC for each strain. Bactericidal activity was defined as ^ 3 log,0 reduction from the starting inoculum.
Table I. MICs (mg/L) for 100 consecutive strains and MBCs (mg/L) for 20 strains of B. pseudomallei Antibiotic Ceftazidime Co-amoxiclav Biapenem Imipenem Meropenem Panipenem
MIC»
MIGo
1 4 0.25 0.5 1 0.5
2 4 0.25 0.5 1 1
Range 0.25-32 0.5-8 0.06-1 0.12-1 0.25-1 0.06-2
MBC» 2 8 0.25 1 2 1
MBG» 8 8 0.5 1 2 2
Range 2-8 4-8 0.25-2 0.5-2 1-4
613
Susceptibility of B. pseudomallei to carbapenetns
Results and discussion The MIC and MBC results for 100 consecutive strains of B. pseudomallei are given in Table I. These strains were all fully susceptible to co-amoxiclav. One strain was resistant to ceftazidime (MIC = 32 mg/L). All strains were susceptible to each of the four carbapenems, with MIG» values lower than those for ceftazidime or co-amoxiclav. Twenty strains were randomly selected for MBC determination; these did not include the ceftazidime resistant strain. The MBC™ values for the carbapenems were ;> 4-fold lower than ceftazidime or co-amoxiclav. Of the four carbapenem antibiotics, biapenem was the most active against B. pseudomallei. Yamamoto et al. (1990) also found that imipenem, meropenem and panipenem had good activity against B. pseudomallei, although they did not test biapenem, nor did they determine MBCs. The MIC and MBC results for 24 strains of B. pseudomallei with reduced susceptibility or resistance to ceftazidime and/or co-amoxiclav are shown in Table II. The carbapenems retained their bactericidal activity. Biapenem was the most active of the four carbapenems. Imipenem and meropenem had equivalent activities against most
Table II. MICs and MBCs (mg/L) for 24 strains of B. pseudomallei with reduced susceptibilities to ceftazidime and/or co-amoxiclav Antibiotic
MIC»
MIQo
Range
All strains (n = 24) 64 16 ceftazidime 1-256 256 co-amoxiclav £512 1-£512 0.5 2 biapenem 0.12-2 1 4 imipenem 0.25-8 1 4 1-4 meropenem 1 16 panipenem 0.5-16 Reduced susceptibility to ceftazidime (n = 7) 32 256 ceftazidime 32-256 2 8 co-amoxiclav 1-8 0.5 biapenem 0.25 0.12-0.5 1 imipenem 0.25 0.25-1 1 2 1-2 meropenem 0.5 1 panipenem 0.5-1 Reduced susceptibility to co-amoxiclav (n = 7) 8 8 1-8 ceftazidime 256 co-amoxiclav 2:512 32-5:512 1 2 biapenem 0.5-2 2 4 1-4 imipenem 2 2 1-2 meropenem 8 16 panipenem 1-16 Reduced susceptibility to ceftazidime and co-amoxiclav (n 64 16 ceftazidime 16-256 256 512 co-amoxiclav 256-512 1 2 biapenem 0.5-2 1 8 imipenem 0.5-8 2 4 1-4 meropenem 4 16 panipenem 1-16
MBC*,
MBC*,
32
256
£512
£512
1 2 2 8
2 4 4 16
128 4 0.5 0.5 1 1
512 8 1 4 4 2
16
32
£512
£512
1 2 2 8
2 4 4 16
Range 4-512 2-£512 0.5-8 0.25-8 1-8
1-32 64-512 2-8
0.5-1 0.25-4 1-4 1-2
4-32 128-£512 1-2 2-4 1^»
2-16
= 10) 32
128
£512
£512
2 2 2 16
4 8 8 32
16-512 256-£512 1-8
0.5-8 1-8
4-32
614
M. D. Smith et at.
strains. Although the MICs for panipenem were higher than the other carbapenems they are still considered to be susceptible (T. Fukuoka, Sankyo Co., personal communication). The group of strains showing reduced susceptibility to ceftazidime had significant increases in MIC to biapenem, meropenem and panipenem (Kruskal-Wallis analysis of variance, P< 0.0001) although the absolute values remained within the susceptibility range. There was no significant increase in MICs for imipenem. However, two strains resistant to both ceftazidime and co-amoxiclav had MIC values for imipenem of 8 mg/L. This is only moderately susceptibility according to NCCLS guidelines, however the MBCs for imipenem were also 8 mg/L, and they remained fully susceptible to the other carbapenems. Time-kill studies for imipenem and meropenem against the susceptible strain 576a and the co-amoxiclav resistant strain 704f have been published previously (Smith et al., 1994). In the present study these were extended to include biapenem and panipenem, and a third strain of B. pseudomallei (979b) resistant to ceftazidime and co-amoxiclav. The MICs for the three test strains (576a, 704f and 979b) to biapenem were 0.25, 1, 1 mg/L respectively; to imipenem 1, 2, 0.5 mg/L; to meropenem 2, 2, 4 mg/L and to panipenem 0.5, 8 and 2 mg/L respectively. All time-kill studies used antibiotic concentrations of 4 x MIC. Biapenem and panipenem were significantly bactericidal for strains 576a and 704f; biapenem within 4 h (similar to imipenem) and panipenem within 6 h (similar to meropenem). In time-kill studies using strain 979b all four carbapenems were significantly bactericidal within 4 h. In contrast, our previous study showed that co-amoxiclav was not bactericidal when tested against the sensitive strain 576a and ceftazidime was not bactericidal for either 576a or 704f. B. pseudomallei is unusual among "pseudomonads" in that it produces a Richmond & Sykes Class I /f-lactamase which is inhibited by clavulanic acid (Livermore et al., 1987). Godfrey et al. (1991) have shown three different mechanisms of /Mactam resistance in B. pseudomallei which may result in resistance to co-amoxiclav, ceftazidime or both antibiotics. Insensitivity of the chromosomal /J-lactamase to clavulanic acid and alteration of the enzyme resulting in specific hydrolysis of ceftazidime alone would not be expected to reduce susceptibility to the carbapenems. Derepression of this chromosomal /J-lactamase produces resistance to both ceftazidime and co-amoxiclav, but this study shows that it does not significantly reduce susceptibility to carbapenems. Chen & Livermore (1994) have shown that the carbapenems overcome the mechanisms, especially /Mactamase production, that commonly produce resistance in Pseudomonas aeruginosa to other classes of anti-pseudomonal jS-lactams. The data in our study confirm that the carbapenems are the most active antibiotics against B. pseudomallei. In general, they retain bactericidal activity against strains resistant to ceftazidime and co-amoxiclav (the current treatments of choice for severe melioidosis), although MICs and MBCs are higher than for susceptible strains. The carbapenems are indicated for treatment of infection with these resistant strains; although expensive, imipenem is currently available for use within Thailand. Unfortunately, biapenem, the most active antibiotic against B. pseudomallei has been withdrawn from further development. A comparative trial is now needed to determine if the carbapenems will be a significant improvement in the treatment of all patients with severe melioidosis. However, it is possible that increased use of the carbapenems will lead to the development of resistance, as has happened with the use of ceftazidime and co-amoxiclav.
Susceptibility of B. pseudomallei to carbapenems
615
Acknowledgements
We are grateful to Dr Wipada Chaowagul, Dr Yupin Suputtamongkol and the staff of the Departments of Medicine and Microbiology, Sappasitprasong Hospital, Ubon Ratchatani, Thailand, for help in collecting strains. This study was part of the Wellcome-Mahidol University, Oxford Tropical Medicine Research Programme, funded by the Wellcome Trust of Great Britain. References Chen, H. Y. & Livermore, D. M. (1994). In-vitro activity of biapenem, compared with imipenem and meropenem, against Pseudomonas aeruginosa strains and mutants with known resistance mechanisms. Journal of Antimicrobial Chemotherapy 33, 949-58. Dance, D. A. B., Wuthiekanun, V., Chaowagul, W., Suputtamongkol, Y. & White, N. J. (1991). Development of resistance to ceftazidime and co-amoxiclav in Pseudomonas pseudomallei. Journal of Antimicrobial Chemotherapy 28, 321—4. Dance, D. A. B., Wuthiekanun, V., Chaowagul, W. & White, N. J. (1989). The antimicrobial susceptibility of Pseudomonas pseudomallei. Emergence of resistance in vitro and during treatment. Journal of Antimicrobial Chemotherapy 24, 295-309. Godfrey, A. J., Wong, S., Dance, D. A. B., Chaowagul, W. & Bryan, L. E. (1991). Pseudomonas pseudomallei resistance to /J-lactam antibiotics due to alterations in the chromosomally encoded /?-lactamase. Antimicrobial Agents and Chemotherapy 35, 1635—40. Livermore, D. M., Chau, P. Y., Wong, A. I. W. & Leung, L. 1C. (1987). 0-Lactamase of Pseudomonas pseudomallei and its contribution to antibiotic resistance. Journal of Antimicrobial Chemotherapy 20, 313-21. National Committee for Clinical Laboratory Standards. (1993). Methods for Dilution Antimicrobial Susceptibility Tests for Bacteria that Grow Aerobicallv: Approved Standard M7-A3. NCCLS, Villanova, PA. Smith, M. D., Wuthiekanun, V., Walsh, A. L. & White, N. J. (1994). Susceptibility of Pseudomonas pseudomallei to some newer /?-lactam antibiotics and antibiotic combinations using time-kill studies. Journal of Antimicrobial Chemotherapy 33, 145-9. Suputtamongkol, Y., Rajchanuwong, A., Chaowagul, W., Dance, D. A. B., Smith, M. D., Wuthiekanun, V. et at. (1994). Ceftazidime versus co-amoxiclav in the treatment of severe melioidosis. Clinical Infectious Diseases 19, 846-53. White, N. J., Dance, D. A. B., Chaowagul, W., Wattanagoon, Y., Wuthiekanun, V. & Pitakwatchara, N. (1989). Halving of mortality of severe melioidosis by ceftazidime. Lancet ii, 697-701. Yamamoto, T., Naigowit, P., Dejsirilert, S., Chiewslip D., Kondo, E., Yokota, T. et al. (1990). In vitro susceptibilities of Pseudomonas pseudomallei to 27 antimicrobial agents. Antimicrobial Agents and Chemotherapy 34, 2027-9. (Received 13 February 1995; returned 19 June 1995; revised 31 August 1995; accepted 15 October 1995)
