Southeast Asian J Trop Med Public Health

IN VITRO ACTIVITY OF OR SULBACTAM IN COMBINATION WITH OR AGAINST CLINICAL ISOLATES OF - RESISTANT BAUMANNII PRODUCING OXA-23 CARBAPENEMASES

Wichai Santimaleeworagun1, Payom Wongpoowarak1, Pantip Chayakul2, Sutthiporn Pattharachayakul1, Pimpimon Tansakul3 and Kevin W Garey4

1Department of Clinical Pharmacy, 3Department of Pharmacognosy and Pharmaceutical Botany, Faculty of Pharmaceutical Sciences, 2Department of Medicine, Faculty of Medicine, Prince of Songkla University, Songkhla, Thailand; 4Department of Clinical Sciences and Administration, College of Pharmacy, University of Houston, Houston, Texas, USA

Abstract. This study investigated the in vitro activity of colistin or sulbactam in combination with fosfomycin or imipenem against eight strains of carbapenem- resistant A. baumannii (CRAB). The eight CRAB clinical isolates were collected from hospitalized patients admitted to Songklanagarind Hospital in southern Thailand during January-December 2008. The isolates were divided into 4 different patterns of clonal relationships using the Repetitive Extragenic Palindromic-Polymerase Chain Reaction method (REP-PCR). The in vitro activity of combination antibacte- rial agents against theses isolates were determined by chequerboard and time-kill methods. All isolates producing OXA-23 carbapenemases were universally sus- ceptible to colistin but intermittently susceptible to other antimicrobial agents. A chequerboard assay showed the synergistic effects of sulbactam plus fosfomycin and colistin plus fosfomycin in 75% and 12.5% of isolates, respectively. Sulbactam 1 at a concentration of 1 x MIC plus fosfomycin at 1 x MIC or at /4 x MIC showed synergism in 75% and 37.5% of clinical isolates, respectively. Bactericidal activity was observed for up to 12 hours of incubation. There was no synergism between colistin and sulbactam, sulbactam and imipenem, and colistin and imipenem, against the tested isolates. Combined use of sulbactam and fosfomycin may pro- vide an alternative therapeutic option for CRAB infections. Key words: Acinetobacter baumannii, carbapenam resistant, colistin, sulbactam, fosfomycin, imipenem, carbapenemase

INTRODUCTION negative bacillus, causes nosocomial infections worldwide. In the United Acinetobacter baumannii, a gram- States, Acinetobacter species were ranked Correspondence: Wichai Santimaleeworagun, the second leading cause of mortality Department of Clinical Pharmacy, Faculty of (43.4%) in patients admitted to inten- Pharmaceutical Sciences, Prince of Songkla sive care units (Wisplinghoff et al, 2004). University, Songkhla 90112, Thailand. The National Antimicrobial Resistance Tel: 66 (0) 7442 8222; Fax: 66 (0) 7442 8222 Surveillance Center, Thailand (NARST) E-mail: [email protected] found A. baumannii was the fourth most

890 Vol 42 No. 4 July 2011 In vitro Activity of Antimicrobial Agent Combinations Against Crab common clinical pathogen isolated from some GNB, including extended-spectrum specimens studied in 2006 (NARST, b-lactamase-producers (de Cueto et al, 2009). A. baumannii related infections 2006; Falagas et al, 2008a). Falagas et al are associated with adverse outcomes, (2008b) found the MIC of fosfomycin including increased mortality, prolonged against MDR-AB strains ranged from 64 hospitalization, additional hospital costs, to >512 mg/l. In our setting, the MIC50 and reduced patient functional status (Lee and MIC90 of fosfomycin against MDR-AB et al, 2007; Tseng et al, 2007; Young et al, were 64 and 96 mg/l, respectively (Hor- 2007). At Songklanagarind Hospital, a tiwakul et al, 2009). This in vitro activity tertiary care hospital located in southern of fosfomycin against MDR-AB is within Thailand, Jamulitrat et al (2009) found possible therapeutic concentrations (Ban- mortality rates were higher in patients do and Toyoshima, 1984). Since the in vitro with carbapenem-resistant A. baumannii activity of fosfomycin in combination with (CRAB; 52.2%) than carbapenem-sensitive other agents against MDR-AB has not pre- A. baumannii bacteremia (19.9%). At pres- viously been reported, we investigated the ent, the optimum treatment for multi- combination of colistin or sulbactam with resistant A. baumannii (MDR-AB) has not fosfomycin or imipenem against in vitro been established. In previous in vitro stud- CRAB infections to evaluate the potential ies, colistin and rifampicin, sulbactam and synergistic or antagonistic effects of these imipenem and rifampicin and imipenem combinations. have been shown to have a synergistic ef- fect on MDR-AB (Song et al, 2007; Tripodi MATHERIALS AND METHODS et al, 2007). However, a synergistic effect with these drug combinations has not Bacterial strains been seen with resistant isolates. All clinical isolates were obtained In Thailand, colistimethate and from inpatients admitted at Songklana- /sulbactam are antimicro- garind Hospital, a university-affiliated bial agents used for treating multi-drug medical school in southern Thailand resistant gram-negative bacterial (GNB) during January-December 2008. CRAB infections, in particular, infections due to was defined as isolates resistant to imi- and A. bauman- . Resistance was defined as a MIC nii. Sulbactam, a b-lactamase inhibitor, is ≥ 16 mg/l according to the Clinical and combined with b-lactam (eg, Laboratory Standards Institute (CLSI, , cefoperazone) for inhibiting 2009). The clinical isolates of CRAB were b-lactamases, but sulbactam itself also obtained from sterile sites (eg, blood, cere- has activity against A. baumannii (Corbella brospinal fluid, bone marrow, peritoneal et al, 1998). The NARST survey in 2006 fluid, pleural fluid, synovial fluid). Eight (NARST, 2009) found 58% of A. baumannii CRAB specimens were collected; four isolates were susceptible to cefoperazone/ isolates had a MIC against imipenem of sulbactam. In contrast, at our institution 16 mg/l, and the others had a MIC ≥ 32 during 2004-2008, 90% of A. baumannii mg/l. MDR-AB was defined as isolates isolates were sensitive to colistin and cefo- resistant to carbapenem and at least two perazone/sulbactam (unpublished data). antibiotics from the following: piperacil- Fosfomycin has been shown to have ac- lin/, ciprofloxacin, , tivity against gram-positive and gentamicin, cefoperazone/sulbactam or

Vol 42 No. 4 July 2011 891 Southeast Asian J Trop Med Public Health

colistin. All CRAB strains were kept at tant genes, blaIMP-like, blaOXA-23, blaOXA-40, -20ºC until tested. blaOXA-58, blaSIM-1, blaVIM-1 and blaVIM-2, were designed from the nucleotide sequence of Polymerase chain reaction-based DNA each gene retrieved from GenBank. The fingerprint accession numbers of each gene are shown The REP-PCR method was used to in Table 1. Amplification was carried out interpret the clonal relationships of the under the following PCR conditions, 94ºC isolates. In this study, genomic DNA for 5 minutes, followed by 30 cycles of was extracted and the PCR technique first 94ºC for 1 minute then annealing for was used to amplify positions in the 2 minutes. The annealing temperatures REP region. Briefly, selected strains were were optimized for each primer set based cultured in Mueller-Hinton broth (Difco, on the melting temperature of the primer, Detroit, MI) and incubated at 37ºC for 18 followed by 72ºC for 2 minutes and a final hours. The genomic DNA was prepared extension at 72ºC for 10 minutes. The PCR using the DNeasy Qiagen kit (Qiagen, products were evaluated by agarose gel Valencia, CA). The purified DNA was electrophoresis, ligated with the Strata- kept at -20ºC until PCR testing was done. clone PCR cloning kit (Stratagene, La Jolla, The PCR mixture consisted of 1 µl of ge- CA) and transferred to E. coli. The plasmid nomic DNA, 10 mM of each forward and from the overnight E. coli culture was pu- reverse primer (REP-forward: 5’-IIIGC- rified and used for sequencing analysis. GCCGICATCAGGC-3’ and REP-reverse: 5’-ACGTCTTATCAGGCCTAC-3’) (Bou Determination of minimum inhibitory et al, 2000), 0.2 mM of deoxy-nucleotide concentration The MICs of the antimicrobial agents triphosphate and 1.5 mM MgCl2 in 5x GoTaq® buffer. The volume of GoTaq® were determined by the agar dilution polymerase (Promega, Madison, WI) method using Mueller-Hinton agar was 1.25 U. The total volume of the PCR (Difco) plates containing dilutions of the mixture was 50 µl. Thermocycling was antimicrobial agents, according to CLSI carried out with Takara thermocycler dice, protocol (CLSI, 2009). The lowest con- starting with denaturation at 94ºC for 10 centration of an antimicrobial agent that minutes, followed by 30 cycles of 94ºC for inhibited visible growth of the organism 1 minute, 45ºC for 1 minute (annealing was defined as the MIC. The antimicrobial temperature), 72ºC for 2 minutes and a agents used in this study were: fosfomy- final extension at 72ºC for 16 minutes. The cin disodium salt (Sigma Chemical, St PCR products were evaluated by electro- Louis, MO), supplemented with 25 mg/l of phoresis in 1.2% of agarose gel, stained glucose-6-phosphate (G-6-P sodium salt; with ethidium bromide (Invitrogen, Carls- Sigma Chemical), colistin sulfate (Sigma bad, CA). The bands of nucleotides were Chemical), sulbactam (DeF Pharmaceuti- detected by UV transillumination and cal Chemical Service, Shanghai, China), photographed with a camera. An isolate and imipenem (Merck, Rahway, NJ). The was considered a different clone if its REP- quality controlled strains was E. coli ATCC PCR pattern differed from the others by 25922, [Department of Medical Sciences more than three bands (Bou et al, 2000). Type (DMST) culture collection, Bang- kok, Thailand] was used to monitor the Amplification of resistant genes accuracy of the MIC determination. The The couple primers of drug resis- MIC results were interpreted according

892 Vol 42 No. 4 July 2011 In vitro Activity of Antimicrobial Agent Combinations Against Crab

Table 1 Oligonucleotide primer design for resistance genes in A. baumannii.

Gene Primer information Size of product Accession (bp) number

blaIMP-like F- 5‘CTRCCGCAGSAGMGBCTTTG3’ 587 AJ640197 R- 5‘AACCAGTTTTGCHTTACCAT3‘ AJ243491 AY590475 AF290912 EF127959 AB074436 AB184977

blaOXA-23 F- 5‘GGAATTCCATGAATAAATATTTTA3‘ 822 AJ132105 R- 5‘GGATCCCGTTAAATAATATTCAGC3‘

blaOXA-40 F- 5‘GGAATTCCATGAAAAAATTTATAC3‘ 828 AF509241 R- 5‘GGATCCCGTTAAATGATTCCAAGA3

blaOXA-58 F- 5‘GGAATTCCATGAAATTATTAAAAA3‘ 843 EU107365 R- 5‘GGATCCCGTTATAAATAATGAAAA3‘

blaSIM-1 F- 5‘GGAATTCCATGAGAACTTTATTGA3‘ 741 AY887066 R- 5‘GGATCCCGTTAATTAATGAGCGGC3‘

blaVIM-1 F- 5‘GGAATTCCATGTTAAAAGTTATTA3‘ 801 DQ112355 R- 5‘GGATCCCGCTACTCGGCGACTGAG3‘

blaVIM-2 F- 5‘GGAATTCCATGTTCAAACTTTTGA3‘ 801 AF291420 R- 5‘GGATCCCGCTACTCAACGACTGAG3‘

to CLSI breakpoint criteria (CLSI, 2009). of paired antimicrobial combinations in Colistin and sulbactam susceptibilities time-kill studies. Briefly, the checkerboard were classified using Comité de l’ An- technique consists of columns in which tibiogramme de la Société Française de each well is filled by twofold serial dilu- Microbiologie criteria (CA-SFM, 2009) tion with the first and rows in giving breakpoints for colistin and sulbac- which each well is filled by serial dilution tam of ≤ 2 and ≤ 8 mg/l, respectively. Due with the second antibiotic. This technique to the lack of a standard MIC breakpoint was performed with Mueller-Hinton for fosfomycin against A. baumannii, the broth (Difco, Detroit, MI), and samples MIC resistance breakpoint of ≥ 32 mg/l were incubated at 37ºC for 18 hours (Pillai was used for fosfomycin, the same as the et al, 2005). In this study, the synergistic breakpoint against Enterobacteriaceae effect for each of the antimicrobial com- and P. aeruginosa (CA-SFM, 2009). binations was determined: colistin with sulbactam, colistin with fosfomycin, Synergistic testing colistin with imipenem, sulbactam with The synergistic effect of antimicro- fosfomycin, and sulbactam with imipe- bial combinations were assessed using nem. The fractional inhibitory concentra- a checkerboard technique. This method tion indices (FICI) were calculated, where was used to select the synergistic effect FICI = MICA/MICA/B + MICB/MICB/A. The

Vol 42 No. 4 July 2011 893 Southeast Asian J Trop Med Public Health

Table 2 Characteristics of the eight carbapenem-resistant A. baumannii (CRAB) isolates studied.

Strains Warda Specimen Antimicrobial resistance profileb REP-PCR groupc

AB23 Med Blood AMK, AMP, CAZ, CIP, GEN, IPM, MEM, A SXT, TZP AB54 NICU Tissue AMK, AMP, CAZ, CIP, GEN, IPM, MEM, A SXT, TZP AB164 PICU Blood AMK, AMP, CAZ, CIP, GEN, IPM, MEM, B SCFP, SXT, TZP AB167 MICU Blood AMK, AMP, CAZ, CIP, GEN, IPM, MEM, B SCFP, SXT, TZP AB198 NeuSur CSF AMP, CAZ, CIP, IPM, MEM, SCFP, TZP D AB307 NICU Blood AMK, AMP, CAZ, SCFP, CIP, GEN, IPM, C MEM, SCFP, SXT, TZP AB313 Sur Ascitic fluid AMK, AMP, CAZ, CIP, GEN, IPM, MEM, A SCFP, SXT, TZP AB315 SICU Blood AMK, AMP, CAZ, CIP, GEN, IPM, MEM, C SCFP, SXT, TZP aMed, medicine; MICU, medical intensive care unit; NICU, neonatal intensive care unit; NeuSur, neurosurgery; SICU, surgical intensive care unit; Sur, surgery; bAMK, amikacin; AMP, ampicillin; CAZ, ; SCFP, cefoperazone-sulbactam; CIP, ciprofloxa- cin; GEN, gentamicin; IPM, imipenem; MEM, ; SXT, sulfamethoxazole-trimethoprim; TZP, -tazobactam cREP-PCR group, Repetative Extragenic Palindromic-Polymerase Chain Reaction used to determine the clonal relationships of the isolates. results of the FICI for each of the paired pretation of combinations of antibiotics antibiotics were interpreted as: ≤ 0.5 = and bactericidal activity. Synergy was ≥ synergistic, > 0.5 - 4.0 = no interaction, defined as a 2 log10 decrease in cfu/ml of and ≥ 4 antagonistic effect. the combination compared with the cfu/ Time-kill studies ml of the most active mono-antimicrobial agent. Bactericidal activity was defined as The chosen optimum combinations a ≥ 3 log decrease in cfu/ml compared of antibiotics depended on the synergistic 10 with the starting colony (Pillai et al, 2005). activity assessed using the checkerboard method, and the antibiotic concentration RESULTS would be based on 1 x MIC for colistin 1 and sulbactam and /4 x MIC for the The eight isolates of CRAB were combined drug in each synergistic drug- obtained from various sterile sites. Us- combination. The synergistic effects of the ing REP-PCR DNA profiling, there were drug combinations against the isolated 4 groups of clonal relationships divided strains were evaluated, together with into patterns A-D (Table 2 and Fig 1). The bactericidal activity at 0, 2, 4, 6, 12 and 24 MIC range was lowest for colistin (0.5-1 hours of incubation following the inter- mg/l) followed by imipenem (16-32 mg/l),

894 Vol 42 No. 4 July 2011 In vitro Activity of Antimicrobial Agent Combinations Against Crab

colistin and sulbactam, sulbactam and imipenem, and colistin and imi- penem. There were no antagonistic effects observed with any of the combinations of antimicrobial agents studied (Table 4). With the time-kill studies, the combination of sulbactam and fosfo- mycin at concentrations equal to the MIC of the organism had synergistic effects in 6 of 8 isolates (75%) (Fig 2A-H). Bactericidal activity within twelve hours was observed for almost tested isolates using these concentra- tions. Decreasing fosfomycin to a 1 concentration of /4 the MIC of the organism with sulbactam at the the MIC for the organism showed syn- ergy in 3 of 8 isolates (37.5%).

DISCUSSION A. baumannii is an important Fig 1–Pattern obtained with Repetitive Extragenic nosocomial infection in ventilator-as- Palindromic-Polymerase Chain Reaction (REP- sociated pneumonia, bacteremia, uri- PCR), evaluated by electrophoresis in 1.2% nary tract infection, meningitis, and agarose gel, stained with ethidium bromide. The letters above each lane indicate the strain. peritonitis (Bergogne-Bérézin and M, DNA molecular weight marker in kilobase Towner, 1996). Increased antibiotic re- unit (kb). Using the REP-PCR method, there sistance by A. baumannii dramatically were 4 groups divided into patterns A-D. limits the potential choices of anti- microbial agents (Bergogne-Bérézin and Towner, 1996). Several studies sulbactam (8-64 mg/l), and fosfomycin have reported the possible mechanisms of (128-512 mg/l) (Table 4). Upon evaluating resistance for A. baumannii. These include for resistant genes, all isolates had the enzymatic lysis [including oxacillinases blaOXA-23 gene but were negative for all the (OXA), MBL, Amp-C, aminoglycoside- other genes (blaOXA-40; blaOXA-58; blaIMP-like, modifying ], loss of porins, efflux blaSIM-1, blaVIM-1 and blaVIM-2). pump, or changing of binding The data from the checkerboard stud- proteins (Bergogne-Berezin and Towner, ies are shown in Table 3. A synergistic 1996; Bonomo and Szabo, 2006). This effect was seen for colistin and fosfomy- high level of A. baumannii resistance pres- cin in 1 out of 8 isolates (12.5%) and for ents clinical challenges for appropriate sulbactam and fosfomycin in 6 out of 8 treatment. Investigation of combined isolates (75%). There were no synergistic antimicrobial agents with activity against effects seen with the combinations of resistant strains of A. baumannii is needed.

Vol 42 No. 4 July 2011 895 Southeast Asian J Trop Med Public Health

Table 3 Determination of MICs of various antimicrobial agents against the eight clinical strains of carbapenem-resistant A. baumannii (CRAB) by agar dilution method.

MICs (mg/l) of various antimicrobial agents (interpretation)a Strains Imipenem Colistin Sulbactam Fosfomycin

AB23 32 (R) 1 (S) 8 (S) 128 (R) AB54 16 (R) 1 (S) 8 (S) 128 (R) AB164 16 (R) 1 (S) 32 (R) 128 (R) AB167 32 (R) 1 (S) 32 (R) 256 (R) AB198 32 (R) 1 (S) 32 (R) 128 (R) AB307 64 (R) 1 (S) 64 (R) 128 (R) AB313 16 (R) 1 (S) 16 (R) 128 (R) AB315 16 (R) 0.5 (S) 64 (R) 512 (R) aBreakpoint criteria were: imipenem: susceptible, ≤ 4 mg/l; intermediate, 8 mg/l; resistant, ≥ 16 mg/l according to CLSI. Colistin: susceptible, ≤ 2 mg/l; resistant, ≥ 4 mg/l according to CA-SFM. Sulbactam: susceptible, ≤ 8 mg/l according to CA-SFM. Fosfomycin: resistant, ≥ 32 mg/l according to CA-SFM.

Table 4 Testing of coupled antimicrobial agents against carbapenem-resistant A. baumannii (CRAB) clinical isolates by checkerboard technique.

FICI for combined antimicrobial agentsa Strains Colistin + Colistin + Colistin + Sulbactam + Sulbactam + sulbactam imipenem fosfomycin imipenem fosfomycin

AB23 0.96 (I) 0.94 (I) 1.17 (I) 0.63 (I) 0.32 (S) AB54 1.54 (I) 0.56 (I) 1.04 (I) 0.52 (I) 0.49 (S) AB164 0.87 (I) 0.89 (I) 1.38 (I) 0.62 (I) 0.48 (S) AB167 1.10 (I) 0.62 (I) 0.94 (I) 0.85 (I) 0.37 (S) AB198 0.94 (I) 0.52 (I) 0.85 (I) 0.75 (I) 0.28 (S) AB307 0.71 (I) 0.99 (I) 1.08 (I) 1.27 (I) 1.04 (I) AB313 0.75 (I) 0.71 (I) 0.49 (S) 0.51 (I) 0.47 (S) AB315 0.92 (I) 1.04 (I) 1.08 (I) 0.92 (I) 1.08 (I)

FICI, fractional inhibitory concentration index; ≤ 0.5, synergistic, S; > 0.5-4.0, indifferent, I; and > 4, antagonistic effect, AT

To date, the best antibiotic monotherapy Fosfomycin, a phosphonic acid de- or combination therapy for treating this rivative, inhibits bacterial pathogen is still unknown. In vitro stud- synthesis and has activity against gram- ies can be carried out to help clinical positive bacteria (eg, Staphylococcus spp, decisions regarding the most appropriate Streptococcus spp and Enterococcus spp) therapeutic regimen. and gram-negative bacteria, including

896 Vol 42 No. 4 July 2011 In vitro Activity of Antimicrobial Agent Combinations Against Crab

Log10 Log10

Log10 Log10

Log10 Log10

Log10 Log10

Fig 2–Time-kill curves of synergistic and bactericidal effects of combinations of antimicrobial agents against carbapenem-resistant A. baumannii clinical isolates: AB23 (A), AB54 (B), AB164 (C), AB167 (D), AB198 (E), AB307 (F), AB 313 (G), and AB315 (H). Open circles, sulbactam at a concentration of 1x MIC; filled squares, sulbactum at 1xMIC + fosfomycin at 1xMIC; filled 1 circles, sulbactam at 1xMIC + fosfomycin at /4 x MIC; open squares, fosfomycin at 1xMIC.

Vol 42 No. 4 July 2011 897 Southeast Asian J Trop Med Public Health

ESBL-producing microorganisms (Falagas of PBP, thus, we could not explain why the et al, 2008b). Falagas et al (2008a) reported susceptibility of organisms in our setting fosfomycin MIC against 30 A. bauman- differed from other setting; 2) the exact nii strains ranged from 64 to >512 mg/l synergistic mechanism of fosfomycin with

(MIC50: 256 mg/l; MIC90: >512 mg/l). In sulbactam was not clear in our study. We our study, the fosfomycin MIC ranged hypothesize fosfomycin likely inhibits from 128-512 mg/l (MIC50: 128; MIC90: 256 pyruvyl transferase responsible for trans- mg/l). The MIC of imipenem against the formation of N-acetylglucosamine into 8 strains of MBL-positive CRAB ranged N-acetylmuraminic acid which is the first from 16-64 mg/l. Of all the antibiotic step for cell wall synthesis (Greenwood, combinations studied, the combination 1994). b-lactam antibiotics inhibit the of sulbactam and fosfomycin showed the latter step of this process. Therefore the best synergistic effect and bactericidal inhibition of the same process of pepti- effect using the time kill-method. Con- doglycan synthesis, with different target centrations of these agents in our study mechanisms of action may explain the were within the therapeutic range. With synergistic effect of these two antibio- pharmacokinetic studies, the doses of fos- tics; 3) even though synergism between fomycin of 2 g and 8 g IV gave maximum fosfomycin and sulbactam was clearly blood concentrations of 145 mg/l and 692 demonstrated in our in vitro study, there mg/l, respectively. These concentrations were no clinical studies to confirm this were well above the concentrations used benefit,in vivo. Further studies are needed in our study (Bando and Toyoshima, 1984; to determine this. Sauermann et al, 2005). The use of fosfo- Sulbactam and fosfomycin had a syn- mycin with sulbactam may be possible ergistic bactericidal effect against CRAB in clinical practice. A previous study by isolates producing OXA-23 carbapen- Martinez-Martinez et al (1996) showed emase in our study. However, the use of the synergy of fosfomycin with amikacin combination antimicrobial regimens need or tobramycin were 44% and 32%, respec- to be tested in the clinical settings because tively, against 34 isolates of MDR-AB. Our differences in the epidemiology of resis- study adds to these previous findings by tant genes may influence the efficacy of demonstrating a synergistic effect be- antibiotic combination treatment. tween sulbactam and fosfomycin against CRAB. We found no synergism between ACKNOWLEDGEMENTS imipenem and sulbactam, unlike the find- ings of Song et al (2007) who found this We would like to thank the Office of combination had synergistic activity at the Higher Education Commission, Thai- 1 x MIC. No synergism was seen in our land for their support with a grant under study for the combinations of colistin and the Strategic Scholarships for Frontier sulbactam, and colistin and imipenem. Research Network. We are also grateful Our study had certain limitations: 1) for the funding received from the Gradu- we identified carbapenemase , ate School, Research and Development which are the most common mechanisms Office, and Faculty of Pharmaceutical in MDR-AB, but we did not study other Sciences, Prince of Songkla University, mechanisms of drug resistance, such as Songkhla, Thailand. We thank Dr Chan- loss of porins, efflux pump, or changing wit Tribuddharat, Dr Pannika Ritvirool

898 Vol 42 No. 4 July 2011 In vitro Activity of Antimicrobial Agent Combinations Against Crab and Dr Pintip Pongpet for providing the de Cueto M, Hernández JR, López-Cerero L, reference strains producing OXA-23, IMP Morillo C, Pascual A. Activity of fosfo- and VIM carbapenemases and also thank mycin against extended-spectrum beta- Dr Mariana Castanheira Kyungwon Lee lactamase producing Escherichia coli and for the positive strains producing OXA-40, Klebsiella pneumoniae. Enferm Infecc Micro- biol Clin 2006; 24: 613-6. OXA-58 ans SIM-1, respectively. We are thankful to Ms Thanaporn Hortiwakul for Falagas ME, Giannopoulou KP, Kokolakis GN, Rafailidis PI. Fosfomycin: use beyond uri- her help with the microbiological assays. nary tract and gastrointestinal infections. Clin Infect Dis 2008a; 46: 1069-77. REFERENCES Falagas ME, Kanellopoulou MD, Karageorgop- oulos DE, et al. Antimicrobial susceptibil- Bando T, Toyoshima H. and ity of multidrug-resistant Gram-negative clinical studies of fosfomycin in bile duct bacteria to fosfomycin. Eur J Clin Microbiol infections. Jpn J Antibiot 1984; 37: 1279-88. Infect Dis 2008b; 27: 439-43. Bergogne-Bérézin E, Towner KJ. Acinetobacter Greenwood D. Fosfomycin trometamol and spp. as nosocomial pathogens: micro- the single-dose treatment of cystitis. J Med biological, clinical, and epidemiological Microbiol 1994; 41: 293-4. features. Clin Microbiol Rev 1996; 9: 148-65. Hortiwakul T, Chayakul P, Ingviya N, Chayakul Bonomo RA, Szabo D. Mechanisms of multi- V. In vitro activity of colistin, fosfomy- drug resistance in Acinetobacter species cin, and piperacillin/tazobactam against and Pseudomonas aeruginosa. Clin Infect Dis Acinetobacter baumannii and Pseudomonas 2006; 43 (suppl 2): S49-56. aeruginosa in Songklanagarind Hospital, Bou G, Cerveró G, Domínguez MA, Quereda Thailand. J Infect Dis Antimicrob Agents C, Martínez-Beltrán J. PCR-based DNA 2009; 26: 91-6 fingerprinting (REP-PCR, AP-PCR) and Jamulitrat S, Arunpan P, Phainuphong P. At- pulsed-field gel electrophoresis character- tributable mortality of imipenem-resistant ization of a nosocomial outbreak caused nosocomial Acinetobacter baumannii blood- by imipenem- and meropenem-resistant stream infection. J Med Assoc Thai 2009; Acinetobacter baumannii. Clin Microbiol 92: 413-9. Infect 2000; 6: 635-43. Lee NY, Lee HC, Ko NY, et al. Clinical and Clinical and Laboratory Standards Institute economic impact of multidrug resistance (CLSI). Performance standards for antimi- in nosocomial Acinetobacter baumannii crobial susceptibility testing. Nineteenth bacteremia. Infect Control Hosp Epidemiol Informational Supplement M100-S19. 2007; 28: 713-9. Wayne, PA; CLSI, 2009. Martinez-Martinez L, Rodriguez G, Pascual Comité de l’Antibiogramme de la Societé A, Suárez AI, Perea EJ. In-vitro activity of Française de Microbiologie (CA-SFM). antimicrobial agent combinations against Recommandations 2009 (Édition de Jan- multiresistant Acinetobacter baumannii. J vier 2009). Paris: CA-SFM, 2009. [Cited Antimicrob Chemother 1996; 38: 1107-8. 2009 Oct 17]. Available from: URL: http:// National Antimicrobial Resistance Surveillance www.sfm.asso.fr/ Center (NARST). Result of antimicrobial Corbella X, Ariza J, Ardanuy C, et al. Efficacy of resistance surveillance. Nonthaburi: Na- sulbactam alone and in combination with tional Antimicrobial Resistance Surveil- ampicillin in nosocomial infections caused lance Center, National Institute of Health, by multiresistant Acinetobacter baumannii. Department of Medical Sciences, Ministry J Antimicrob Chemother 1998; 42: 793-802. of Public Health, Thailand. [Cited 2009

Vol 42 No. 4 July 2011 899 Southeast Asian J Trop Med Public Health

Oct 17]. Available from: URL: http://narst. combination against epidemic multi- dmsc.moph.go.th/ drug-resistant Acinetobacter baumannii Pillai SK, Moellering RC, Eliopoulos GM. An- isolates producing OXA-58 carbapen- timicrobial combinations. In: Lorian V, ed. emases. Int J Antimicrob Agents 2007; Antibiotics in laboratory medicine. Phila- 30: 537-40. delphia: Lippincott Williams & Wilkins, Tseng YC, Wang JT, Wu FL, Chen YC, Chie WC, 2005: 365-440. Chang SC. Prognosis of adult patients with Sauermann R, Karch R, Langenberger H, et al. bacteremia caused by extensively resistant Antibiotic abscess penetration: fosfomycin Acinetobacter baumannii. Diagn Microbiol levels measured in pus and simulated Infect Dis 2007; 59: 181-90. concentration-time profiles. Antimicrob Wisplinghoff H, Bischoff T, Tallent SM, Seifert Agents Chemother 2005; 49: 4448-54. H, Wenzel RP, Edmond MB. Nosocomial Song JY, Kee SY, Hwang IS, et al. In vitro activi- bloodstream infections in US hospitals: ties of carbapenem/sulbactam combina- analysis of 24,179 cases from a prospective tion, colistin, colistin/rifampicin combina- nationwide surveillance study. Clin Infect tion and tigecycline against carbapenem- Dis 2004; 39: 309-17. resistant Acinetobacter baumannii. J Antimi- Young LS, Sabel AL, Price CS. Epidemiologic, crob Chemother 2007; 60: 317-22. clinical, and economic evaluation of an Tripodi MF, Durante-Mangoni E, Fortunato outbreak of clonal multidrug-resistant R, Utili R, Zarrilli R. Comparative ac- Acinetobacter baumannii infection in a surgi- tivities of colistin, rifampicin, imipenem cal intensive care unit. Infect Control Hosp and sulbactam/ampicillin alone or in Epidemiol 2007; 28: 1247-54.

900 Vol 42 No. 4 July 2011