Department of Pathology, University of Alabama at Birmingham, Birmingham, Alabama, 1 and Merck and Co., Inc., West Point, Pennsylvania 2

ANTIMICROBIAL AGENTS AND CHEMOTHERAPY, May 1997, p. 881 885 Vol. 41, No. 5 0066-4804/97/$04.00 0 Copyright 1997, American Society for Microbiology Comparative In Vitro Antimicrobial Susceptibilities of Nosocomial Isolates of Acinetobacter baumannii and Synergistic Activities of Nine Antimicrobial Combinations MARISA B. MARQUES, 1 ENEIDA S. BROOKINGS, 1 STEPHEN A. MOSER, 1 PHILLIP B. SONKE, 2 AND KEN B. WAITES 1 * Department of Pathology, University of Alabama at Birmingham, Birmingham, Alabama, 1 and Merck and Co., Inc., West Point, Pennsylvania 2 Received 19 August 1996/Returned for modification 18 November 1996/Accepted 14 February 1997 The in vitro susceptibilities of 69 nosocomial Acinetobacter isolates were determined by the broth microdilution method. Fourteen (20%) isolates were resistant to at least two aminoglycosides and two extendedspectrum penicillins. Nine antimicrobial combinations were then tested for synergy against these 14 isolates by checkerboard titration: imipenem with ciprofloxacin, amikacin, and tobramycin and ampicillin-sulbactam, piperacillin-tazobactam, and ticarcillin-clavulanate with amikacin and tobramycin. Synergy was detected with one or more antimicrobial combinations against 9 of 14 (64%) isolates, partial synergy was detected with one or more combinations against all 14 isolates, and an additive effect alone was observed with two different combinations against two isolates. No antagonism was detected with any combination. Imipenem plus either amikacin or tobramycin resulted in a synergistic or partial synergistic response against all 14 isolates. Specific combinations showing synergy against A. baumannii isolates were imipenem with tobramycin (four isolates), imipenem with amikacin (three isolates), ampicillin-sulbactam with tobramycin (six isolates), ampicillinsulbactam with amikacin (three isolates), and ticarcillin-clavulanate with tobramycin (one isolate). Genotyping by randomly amplified polymorphic DNA analysis showed that 9 of the 14 isolates were of one strain, 4 isolates were of a second strain, and the remaining isolate was of a different strain. Eight of 14 (57%) patients infected with resistant A. baumannii isolates died. Only 3 of 14 patients had received a therapeutic regimen which was tested for synergy. Clinical studies are needed to determine the significance of these findings. Acinetobacter species are emerging as a major cause of nosocomial infections, particularly in intensive care and burn units, where antimicrobial use is greatest (29). Outbreaks of infections caused by multiple drug-resistant (MDR) Acinetobacter spp. have been widely reported (3, 10, 31, 33). Our institution experienced such an outbreak in the medical intensive care unit, with several fatalities, despite aggressive treatment (20). Concurrently, MDR Acinetobacter spp. were recovered from patients in other areas of the hospital. Due to the life-threatening potential of such infections, empiric treatment with broad-spectrum antimicrobial agents is mandatory while awaiting organism identification and in vitro susceptibility test results (1). However, increasing antimicrobial resistance in Acinetobacter spp. has effectively eliminated many treatment alternatives, raising concerns about optimum therapeutic regimens (9, 11, 15, 17, 18, 27, 30, 31, 34). Combination therapy including imipenem or another -lactam with an aminoglycoside such as amikacin has been recommended for empiric treatment when infection with an Acinetobacter sp. is suspected (1), but few studies have examined whether there is synergism between these or newer agents, including extended-spectrum penicillins, against these organisms. Historically, evaluations of interactions between antimicrobial agents against bacteria have been most valuable for endocarditis and infections in neutropenic patients (14, 19). However, for emerging MDR organisms, when there is a lack of prior information on the effectiveness of two agents combined, * Corresponding author. Mailing address: Department of Pathology WP 230, University of Alabama at Birmingham, 618 South 18th St., Birmingham, AL 35233-7331. Phone: (205) 934-6421. Fax: (205) 975-4468. E-mail: Waites @wp.path.uab.edu. or in the case of a poor therapeutic response, such investigations may also be useful in helping to guide therapy. Checkerboard titration, the technique used most often to assess antimicrobial combinations in vitro (16), was chosen to evaluate the inhibitory effects of nine combinations of antimicrobial agents against 14 MDR isolates of Acinetobacter baumannii. MATERIALS AND METHODS Bacteria. Sixty-nine Acinetobacter isolates obtained from blood, lung tissue, sputum, tracheal aspirates, urine, and wounds over a 6-month period in 1994 were stored at 70 C. Organisms were identified and antimicrobial susceptibilities were determined initially by using the MicroScan WalkAway 96 instrument (Dade MicroScan, West Sacramento, Calif.). Susceptibilities were redetermined by using frozen microtiter plates (PML Microbiologicals, Tualatin, Oreg.) to allow for the inclusion of additional antimicrobial agents. Prior to testing, the organisms were passaged twice onto Trypticase soy agar (TSA) containing 5% sheep blood (BBL, Cockeysville, Md.) and were incubated overnight. Suspensions in saline with a turbidity equivalent to that of a 0.5 McFarland standard were prepared, diluted, and inoculated into microtiter plates. The final inoculum density was 3 10 5 to 5 10 5 CFU/ml. The plates were incubated for 20 to 24 h and the MICs were read. The 12 antimicrobial agents tested were imipenem, ampicillin-sulbactam (A/S), piperacillin, piperacillin-tazobactam (P/T), ticarcillin-clavulanate (T/C), ciprofloxacin, ofloxacin, amikacin, gentamicin, tobramycin, rifampin, and trimethoprim-sulfamethoxazole. Fourteen isolates which were resistant to at least two aminoglycosides and two extended-spectrum penicillins were selected for use in synergy tests. Antimicrobial agents used in synergy tests. Standard powders of the following antimicrobial agents for laboratory use were obtained from their respective manufacturers: ciprofloxacin (Miles Inc., West Haven, Conn.), amikacin (Bristol- Myers Squibb Co., Princeton, N.J.), piperacillin and tazobactam (Wyeth-Ayerst Laboratories, Philadelphia, Pa.), ampicillin and sulbactam (Roerig Division, Pfizer Inc., New York, N.Y.), imipenem (Merck and Co., Inc., West Point, Pa.), and ticarcillin and clavulanate (SmithKline Beecham Pharmaceuticals, Philadelphia, Pa.). Stock solutions of ciprofloxacin, amikacin, piperacillin, tazobactam, ampicillin, and sulbactam were prepared according to published standards (23) and were frozen at 70 C until use. Stock solutions of imipenem, ticarcillin, and clavulanate were prepared fresh on the day of each assay. Tobramycin (Eli Lilly & Co., Indianapolis, Ind.) was obtained already in solution (10 mg/ml). The 881

882 MARQUES ET AL. ANTIMICROB. AGENTS CHEMOTHER. TABLE 1. Antimicrobial agents tested for synergy against 14 A. baumannii isolates Antimicrobial agent Concn tested ( g/ml) Peak concn in plasma ( g/ml) a Dose b Imipenem 0.25 32 57 750 mg A/S 0.5/0.25 32/16 109 150/48 88 2.0 g/1.0 g P/T 0.5/4 128/4 298/34 4.0 g/0.5 g T/C 4/2 256/2 324/8 3.0 g/0.1 g Ciprofloxacin 0.5 32 4 6 400 mg Amikacin 2 128 15 30 5.0 7.5 mg/kg Tobramycin 4 256 4 10 1.5 2.0 mg/kg a Data are from a previous publication (24). b The doses listed are those which are expected to produce concentrations in plasma in the indicated ranges (24). antimicrobial agents were stored according to the manufacturer s recommendations prior to dilution in cation-adjusted Mueller-Hinton broth (BBL). The following antimicrobial combinations were tested: imipenem with ciprofloxacin, amikacin, and tobramycin and A/S, P/T, and T/C with amikacin and tobramycin. For T/C and P/T, clavulanic acid and tazobactam were tested at fixed concentrations of 2 and 4 g/ml, respectively, while A/S was tested with ampicillin and sulbactam at a fixed 2:1 ratio (Table 1). Test methodology. Fourteen MDR isolates were further identified as A. baumannii (4 6). Synergy tests were performed in a 96-well microtiter plate (Dynatech Laboratories, Chantilly, Va.) containing two antimicrobial agents in twofold dilutions dispensed in a checkerboard fashion on the day of the assay. Each well contained 0.1 ml of individual antimicrobial combinations or broth controls. Suspensions with turbidities equivalent to that of a 0.5 McFarland standard were prepared to yield a final inoculum of 3 10 5 to 5 10 5 CFU/ml. The plates were incubated for 20 to 24 h, and the MICs were read. Growth and sterility controls were included in each plate. To check for purity, TSA was inoculated from the growth control well, and for inoculum verification, serial dilutions were plated onto TSA. Quality control strains were Pseudomonas aeruginosa ATCC 27853, Escherichia coli ATCC 25922, and E. coli ATCC 35218. Reproducibility of MICs. The MICs of each drug for the 14 MDR A. baumannii isolates were determined twice with MicroScan Overnight Neg Combo panels, once with frozen MIC panels read manually, and one or more times by the broth microdilution method in the synergy assays. Synergy tests were considered valid if the measured MICs of the individual antimicrobial agents were within 1 dilution of the values obtained in the original determinations of the MICs of the drugs tested alone. Interpretation of synergy tests. For the first clear well in each row of the microtiter plate containing both antimicrobial agents, the fractional inhibitory concentration (FIC) of each agent was calculated as follows (16): FIC of drug A (FIC A ) MIC of drug A in combination/mic of drug A alone, and the FIC of drug B (FIC B ) MIC of drug B in combination/mic of drug B alone. If the MIC of any agent alone occurred at the lowest or highest concentrations tested, the FIC was considered not determinable and synergy could not be assessed. The summation of both FICs ( FIC) in each well ( FIC FIC A FIC B ) was used Isolate TABLE 2. MICs of seven antimicrobial agents tested for 14 isolates of A. baumannii MIC ( g/ml) a IMI AMK TOB A/S P/T T/C CIP 1 4 32 32 16/8 128/4 256/8 32 2 4 64 32 16/8 128/4 256/8 32 3 4 64 32 8/4 128/4 256/4 32 4 4 16 32 8/4 128/4 256/8 32 5 4 64 32 8/4 128/4 256/8 32 6 4 64 32 8/4 128/4 256/8 32 7 4 64 32 16/8 128/4 256/8 32 8 1 4 16 2/1 0.5/4 256/8 32 9 4 32 32 16/8 128/4 256/8 32 10 4 64 32 16/8 128/4 256/8 32 11 2 16 128 4/2 128/8 256/8 32 12 0.5 128 32 4/2 16/4 256/8 32 13 4 32 128 32/16 128/4 256/8 32 14 4 32 64 8/4 128/4 256/8 32 a Abbreviations: IMI, imipenem; AMK, amikacin; TOB, tobramycin; CIP, ciprofloxacin. TABLE 3. Combinations of antimicrobial agents which resulted in synergistic, partially synergistic, or additive reactions Drug combination (concn [ g/ml]) No. of isolates a Syn Part syn Addit IMI AMK 0.25 2 1 1 0 0.25 4 0 1 0 0.25 8 0 1 0 0.25 16 0 2 0 0.5 2 1 0 0 0.5 4 1 0 0 1 4 1 0 0 1 8 0 1 0 2 2 0 4 0 2 4 0 5 0 2 8 0 12 0 2 16 0 11 0 Total b 3 11 0 0.25 4 1 0 0 0.25 8 1 0 0 1 8 3 1 0 2 4 0 7 0 2 8 0 6 0 4 4 0 1 0 Total b 4 9 0 1/0.5 2 1 0 0 1/0.5 8 0 1 0 2/1 8 1 1 0 2/1 16 0 2 0 4/2 2 1 0 0 4/2 4 1 1 0 4/2 8 0 2 0 4/2 16 1 1 1 8/4 2 0 2 0 8/4 4 0 1 0 8/4 8 0 1 0 8/4 16 0 0 0 16/8 8 0 1 0 Total b 3 8 1 0.5/0.25 8 0 0 0 1/0.5 8 0 1 0 2/1 8 1 1 0 4/2 4 2 1 0 4/2 8 5 4 0 8/4 4 0 4 0 8/4 8 0 0 0 Total b 6 7 0 ( g/ml) 16/2 2 0 1 0 64/2 16 0 1 0 128/2 4 0 1 0 128/2 8 0 1 0 128/2 16 0 2 0 Total b 0 5 0 64/2 4 1 0 0 128/2 4 0 1 0 128/2 8 0 3 1 Total b 1 4 1 P/T TOB 8/4 4 0 1 0 8/4 8 0 1 0 Total b 0 1 0 a Abbreviations: Syn, synergistic; Part syn, partially synergistic; Addit, additive. b Total number of isolates (some were inhibited by more than one concentration of the same combination).

VOL. 41, 1997 ANTIMICROBIAL SYNERGISM AGAINST A. BAUMANNII 883 TABLE 4. Biotypes, genotypes, and synergistic interactions for 14 A. baumannii isolates a Isolate no. Hospital unit Biotype DNA pattern Synergy Partial synergy 1 MICU 9 A IMI AMK A/S AUK 2 MICU 9 A IMI AMK 3 MICU 9 A IMI AMK 4 Non-ICU 9 A IMI AMK Additivity 5 Burn 9 A IMI AMK 6 Burn 9 A IMI AMK 7 SICU 9 A IMI AMK 8 MICU 1 B IMI AMK 9 Non-ICU 8 A IMI AMK 10 HTICU 9 A IMI AMK 11 MICU 1 B IMI AMK 12 SICU 9 Other IMI AMK P/T TOB 13 MICU 1 B IMI AMK 14 Non-ICU 1 B IMI AMK a Abbreviations: MICU, medical intensive care unit; SICU, surgical intensive care unit; HTICU, heart transplant intensive care unit; AMK, amikacin; IMI, imipenem; TOB, tobramycin. to classify the combination of antimicrobial agents at the given concentrations as synergistic ( FIC, 0.5), partially synergistic ( FIC, 0.5 and 1.0), additive ( FIC, 1.0), indifferent ( FIC, 1 and 4), or antagonistic ( FIC, 4). Synergistic, partially synergistic, or additive interactions were reported only if they occurred within the range of therapeutic concentrations achievable in human plasma, as indicated in Table 1. Furthermore, partial synergy was reported only if no synergy was detected at any clinically relevant concentration of that antimicrobial combination and additivity was reported only if no synergy or partial synergy was detected (see Tables 3 and 4). The peak concentration in plasma was used to define the upper limit of clinically relevant antimicrobial concentrations at which synergy occurred (13, 22, 25, 26, 28, 32). Although this is not the most important parameter for assessment of the clinical efficacy of -lactams, it is not appropriate to consider drug concentrations above their peak concentrations achievable in human plasma. Biotyping and DNA analysis. The 14 MDR isolates of A. baumannii were biotyped as described previously (4 6). Randomly amplified polymorphic DNA pattern analysis (RAPD) with four random primers was used to determine genotypes. Organism suspensions in saline subcultured onto TSA were centrifuged, and the pellet was resuspended in extraction buffer (35). The suspension was heated in a boiling water bath, extracted twice with phenol-chloroform and once with chloroform, and precipitated with ethanol. The absorbance of the recovered DNA was measured to assess its concentration and purity. PCR was performed by using 200 ng of DNA as a template and four separate primers (35) and Stoffel DNA polymerase in a model 480 thermocycler (Perkin-Elmer, Norwalk, Conn.) under the following conditions: 95 C for 3 min and 40 cycles of 95 C for 1 min, 37 C for 1 min, and 72 C for 8 min. The reaction mixtures were analyzed by electrophoresis on a 2% agarose gel, and the bands were visualized following staining with ethidium bromide. Patient outcomes. Clinical and treatment data were obtained by review of the patients medical records. RESULTS Antimicrobial susceptibilities. Only four drugs showed activity against the 14 MDR A. baumannii isolates. All isolates were susceptible to imipenem, eight (57%) were susceptible to A/S, three (21%) were susceptible to amikacin, and two (14%) were susceptible to P/T (23). The MICs obtained during checkerboard titrations of each of the seven antimicrobial agents tested alone are indicated in Table 2. Synergy tests. The synergistic, partially synergistic, or additive effects of nine antimicrobial combinations are indicated in Tables 3 and 4. Synergy was detected with one or more antimicrobial combinations against nine isolates. Partial synergy was detected with one or more antimicrobial combinations against all 14 isolates. Additivity with one or more antimicrobial combinations was detected against eight isolates. However, additivity in the absence of either synergy or partial synergy was detected against only two isolates, with one different combination showing additivity against each isolate. No antagonism was detected. Except for the specific combinations mentioned below, all others were indifferent. Imipenem with amikacin showed synergy against three isolates and partial synergy against 11 isolates, whereas imipenem with tobramycin showed synergy against 4 isolates and partial synergy against 9 isolates. A/S with amikacin showed synergy against three isolates and partial synergy against eight isolates. A/S with tobramycin showed synergy against six isolates and partial synergy against seven isolates. An additive effect alone was observed for A/S with amikacin against one isolate. T/C with tobramycin showed synergy against one isolate but showed no synergy with amikacin. There was partial synergy for T/C with amikacin and tobramycin, against five and four isolates, respectively. T/C with tobramycin was additive against one isolate. The MICs of T/C exceeded the highest concentration tested for 5 of 14 isolates, so that no interpretation was possible (Table 2). Interpretation of synergy tests for P/T and ciprofloxacin was more problematic since the MICs exceeded the highest concentration tested for all 14 isolates in the case of ciprofloxacin and for 12 isolates in the case of P/T. For another isolate the MIC was less than the lowest concentration of P/T tested. Against the single isolate for which data were interpretable, P/T with tobramycin showed partial synergy. Even though the FIC could not be calculated, the MICs of both ciprofloxacin and P/T in the combinations tested remained in the resistant

884 MARQUES ET AL. ANTIMICROB. AGENTS CHEMOTHER. TABLE 5. Patient characteristics and outcomes Patient no. Age (yr)/ gender Diagnoses Hospital day of first positive culture range, making it unlikely that these agents would have therapeutic value. In addition, for ciprofloxacin, testing of higher concentrations would have been nonproductive since the levels achievable in human plasma would have been greatly exceeded (24). Correlation between genotype and synergy. Table 4 presents the results of biotyping and DNA analysis and the synergistic configurations for all 14 isolates. Nine isolates were classified as biotype 9, four were classified as biotype 1, and one was classified as biotype 8. Nine isolates had similar DNA types according to RAPD data (pattern A), with eight of nine isolates designated as biotype 9. This RAPD pattern was also observed for the isolate classified as biotype 8. The four isolates classified as biotype 1 all belonged to a second RAPD pattern (pattern B). A single isolate was shown to be distinct from isolates of both patterns A and B, but it was classifiable as biotype 9. There was no correlation between the relatedness of the isolates and the response to any antimicrobial combination. Patient outcomes. As indicated in Table 5, 8 of 14 people (57%) with A. baumannii infection died; 5 patients died within a few days of detection of the organism. Among seven patients with bacteremia, only one (14%) (patient 12) survived. Only three patients (patients 8, 12, and 13) received combinations of the antimicrobial agents which were tested. Patient 8 received treatment with imipenem and amikacin, a combination which showed synergy against the organism isolated from the respiratory tract. The patient died several weeks later of unrelated complications. Patient 12, a survivor, received amikacin and imipenem, a combination shown to be partially synergistic against the organism recovered from blood. Patient 13 was treated with ciprofloxacin and piperacillin, in addition to imipenem, but died several weeks after the infection due to unrelated factors. DISCUSSION Empiric therapy for nosocomial infections must include broad-spectrum agents expected to cover the most likely organisms according to historical institutional susceptibilities. Body site(s) of MDR A. baumanii Treatment Outcome (hospital day) 1 58/male Sepsis, pneumonia 1st Blood, a CSF, b tracheal aspirate Piperacillin, tobramycin, Died (7th) imipenem, gentamicin 2 58/female Sepsis 18th Blood, a postmortem blood and lung Imipenem Died (19th) 3 55/female Respiratory failure 5th Postmortem blood a and lung Piperacillin, gentamicin Died (5th) 4 69/male Renal and heart failure 56th Urine None Survived 5 55/male Burns 90th Tracheal aspirate Imipenem Survived 6 46/female Burns 50th Blood, a sputum Piperacillin, tobramycin Died (61st) 7 79/female Cholecystitis 25th Hand Imipenem, gentamicin Survived 8 76/male Pneumonia, respiratory 22nd Tracheal aspirate, a sputum Imipenem, amikacin, Died (83rd) failure ceftazidime 9 43/female Hip fracture 5th Hip Amikacin Survived 10 54/male Sepsis, pneumonia 61st Bronchoalveolar lavage fluid, Ceftazidime, gentamicin Died (63rd) postmortem lung and blood 11 52/female Diabetes, sepsis 33rd Blood, a tracheal aspirate, leg wound Ceftazidime, gentamicin, Died (35th) imipenem 12 23/male Burns 17th Blood, a chest wound Amikacin, imipenem Survived 13 64/female Diabetes, respiratory 101st Leg wound Ciprofloxacin, piperacillin, Died (153th) failure imipenem 14 34/male Hand injury 6th Finger Debridement, cefazolin, gentamicin Survived a Source of organism tested for synergy in patients infected with multiple MDR A. baumanii isolates. b CSF, cerebrospinal fluid. Toxicity often limits the ability to increase doses of individual agents to achieve concentrations which might effectively eradicate organisms from patients with potentially life-threatening systemic infections. Therefore, severe infections, particularly in debilitated hosts, require aggressive treatment, usually involving at least two antimicrobial agents (1). Combinations of agents that exhibit synergy, partial synergy, or even additive activity could potentially reduce toxicity and improve outcomes for patients with difficult-to-treat infections. In studies of gramnegative bacilli, combinations of a -lactam and amikacin which were synergistic in vitro have been associated with significantly better outcomes than those achieved with nonsynergistic regimens (19). The presence or absence of synergy may also be a significant factor related to outcome for patients with neutropenia, shock, and other underlying conditions (19). A recent report showed synergy with imipenem and both amikacin and tobramycin against several isolates among a group of 25 MDR A. baumannii isolates (21). Combinations of imipenem plus amikacin and P/T plus amikacin have been shown to be more rapidly bactericidal against MDR A. baumannii than imipenem plus sulbactam and P/T plus sulbactam (7). Chang et al. (8) found that imipenem plus amikacin demonstrated synergy against 36% of 22 isolates of A. baumannii causing bacteremia and partial synergy against 50% of the isolates, whereas in this study the values for this combination were 21 and 79%, respectively. In their study the organisms were much more susceptible overall than those in our study. Those investigators also detected a synergistic bactericidal effect using time-kill curves and concluded that imipenem plus amikacin was the best choice among the four combinations tested (8). In our study, synergy or partial synergy was also detected for combinations of A/S, T/C, P/T, and aminoglycosides within therapeutic concentrations in plasma, even though one or both agents were inactive when they were tested alone. Tobramycin had no activity alone against MDR A. baumannii isolates, but synergy and/or partial synergy with other agents was detected

VOL. 41, 1997 ANTIMICROBIAL SYNERGISM AGAINST A. BAUMANNII 885 against several isolates, suggesting its potential as an alternative to amikacin. Among the three -lactam -lactamase inhibitor combinations tested, only A/S demonstrated substantial activity alone or synergistically with aminoglycosides. This observation is consistent with the experience of others (12, 31, 32). Urban et al. (31, 32) reported that sulbactam alone is highly active against Acinetobacter due to selective affinity to its penicillinbinding proteins, while clavulanate and tazobactam demonstrate considerably less activity. Although both sulbactam and tazobactam alone may exhibit intrinsic activity against some Acinetobacter spp., the difference between them is thought to be related to their relative pharmacokinetics (1, 30 32). Therefore, A/S might be considered for use in treatment when imipenem cannot be used or is contraindicated. Very limited conclusions concerning the potential synergy of P/T with aminoglycosides can be made because of the large number of indeterminate FICs. Unfortunately, there is no means to predict which antimicrobial agent-organism combinations will show synergy to guide empiric therapy, and there may not be a correlation between in vitro synergy and clinical efficacy (2, 19). Therefore, in vitro data must be validated by assessing the clinical performance of combinations of antimicrobial agents before specific recommendations to modify existing treatment guidelines for Acinetobacter infections are possible. ACKNOWLEDGMENTS This work was supported in part by grants from Wyeth-Ayerst Laboratories, Philadelphia, Pa., and Merck and Company, Inc., West Point, Pa. We gratefully acknowledge Yvonne Huang for providing technical support, Julie Mangino for helpful suggestions in selecting the antimicrobial combinations for synergy testing, and Michael Reinke, William Rogers, Edwin Swiatlo, and Thomas Rivers for helpful suggestions regarding the preparation of the manuscript. REFERENCES 1. Allen, D. M., and B. J. Hartman. 1995. Acinetobacter species, p. 2009 2013. In G. L. Mandell, J. E. Bennett, and R. Dolin (ed.), Principles and practice of infectious diseases. Churchill Livingstone, New York, N.Y. 2. Anstey, N. M. 1992. Use of cefotaxime for treatment of Acinetobacter infections. Clin. Infect. Dis. 15:374. 3. Bergogne-Berezin, E., M. L. Joly-Guillou, and J. F. Vieu. 1987. Epidemiology of nosocomial infections due to Acinetobacter calcoaceticus. J. Hosp. Infect. 10:105 113. 4. Bouvet, P. J. M., and P. A. D. Grimont. 1986. Taxonomy of the genus Acinetobacter with the recognition of Acinetobacter baumannii sp. nov. Acinetobacter haemolyticus sp. nov. Acinetobacter johnsonii sp. nov., and Acinetobacter junii sp. nov. and amended descriptions of Acinetobacter calcoaceticus and Acinetobacter lwoffii. Int. J. Syst. Bacteriol. 35:228 240. 5. Bouvet, P. J. M., and P. A. D. Grimont. 1987. Identification and biotyping of clinical isolates of Acinetobacter. Ann. Ińst. Pasteur Microbiol. 138:569 578. 6. Bouvet, P. J. M., S. Jeanjean, J.-F. Vieu, and L. Dijkshoorn. 1990. Species, biotype, and bacteriophage type determinations compared with cell envelope protein profiles for typing Acinetobacter strains. J. Clin. Microbiol. 28:170 176. 7. Caillon, J., F. Guerif, L. Raskine, J. Jouanelle, and Y. Xiong. 1995. Comparative bactericidal activity of imipenem, piperacillin tazobactam, ticarcillin clavulanic acid alone and in combination with amikacin or sulbactam against Acinetobacter baumannii with different resistance phenotypes, abstr. E-101, p. 103. In Program and abstracts of the 35th Interscience Conference on Antimicrobial Agents and Chemotherapy. American Society for Microbiology, Washington, D.C. 8. Chang, S. C., Y. C. Chen, K. T. Luh, and W. C. Hsieh. 1995. In vitro activities of antimicrobial agents, alone and in combination, against Acinetobacter baumannii isolated from blood. Diagn. Microbiol. Infect. Dis. 23:105 110. 9. Chopade, B. A., P. J. Wise, and K. J. Towner. 1985. Plasmid transfer and behaviour in Acinetobacter calcoaceticus EBF65/65. J. Gen. Microbiol. 131: 2805 2811. 10. Crombach, W. H. J., L. Dijkshoorn, M. van Noort-Klaassen, J. Niessen, and G. van Knippenberg-Gordebeke. 1989. Control of an epidemic spread of a multidrug-resistant strain of Acinetobacter calcoaceticus in a hospital. Intensive Care Med. 85:624 631. 11. Devaud, M., F. H. Kayser, and B. Bachi. 1982. Transposon-mediated multiple antibiotic resistance in Acinetobacter strains. Antimicrob. Agents Chemother. 22:323 329. 12. Douboyas, J., L. S. Tzouvelekis, and A. Tsakris. 1994. In-vitro activity of ampicillin/sulbactam against multi drug-resistant Acinetobacter calcoaceticus var. anitratus clinical isolates. J. Antimicrob. Chemother. 34:298 300. 13. Drusano, G. L. 1991. Human pharmacodynamics of -lactams, aminoglycosides and their combination. Scand. J. Infect. Dis. Suppl. 74:235 248. 14. Eliopoulos, G. M., and C. T. Eliopoulos. 1989. Antibiotic combinations: should they be tested? Clin. Microbiol. Rev. 1:139 156. 15. Goldstein, F. W., A. Labigne-Roussel, G. Gerbaud, C. Carlier, E. Collatz, and P. Courvalin. 1983. Transferable plasmid-mediated antibiotic resistance in Acinetobacter. Plasmid 10:138 147. 16. Isenberg, H. D. 1992. Synergism testing: broth microdilution checkerboard and broth macrodilution methods. In Clinical microbiology procedures manual, vol. 1, p. 5.18.1 5.18.28. American Society for Microbiology, Washington, D.C. 17. Joly-Guillou, M. L., E. Bergogne-Berezin, and N. Moreau. 1987. Enzymatic resistance to -lactams and aminoglycosides in Acinetobacter calcoaceticus.j. Antimicrob. Chemother. 20:773 776. 18. Joly-Guillou, M. L., E. Bergogne-Berezin, and J. F. Vieu. 1990. A study of the relationships between antibiotic resistance phenotypes, phage-typing and biotyping of 117 clinical isolates of Acinetobacter sp. J. Hosp. Infect. 16:49 58. 19. Klastersky, J., F. Meunier-Carpentier, and J. M. Prevost. 1977. Significance of antimicrobial synergism for the outcome of gram negative sepsis. Am. J. Med. Sci. 273:157 167. 20. Marques, M. B., J. E. Mangino, B. B. Hines, S. A. Moser, and K. B. Waites. 1995. Investigation of an outbreak of multidrug-resistant Acinetobacter baumannii in a medical intensive care unit, abstr. L-33, p. 115. In Abstracts of the 95th General Meeting of the American Society for Microbiology. American Society for Microbiology, Washington, D.C. 21. Martinez-Martinez, L., G. Rodriguez, A. Pascual, A. I. Suarez, and E. J. Perea. 1995. In vitro activity of several antimicrobial combinations against multiresistant Acinetobacter baumannii, abstr. E-104, p. 104. In Program and abstracts of the 35th Interscience Conference on Antimicrobial Agents and Chemotherapy. American Society for Microbiology, Washington, D.C. 22. Meyers, B. R., P. Wilkinson, M. H. Mendelson, S. Walsh, C. Bournazos, and S. Z. Hirschman. 1991. Pharmacokinetics of ampicillin-sulbactam in healthy elderly and young volunteers. Antimicrob. Agents Chemother. 35:2098 2101. 23. National Committee for Clinical Laboratory Standards. 1994. Performance standards for antimicrobial susceptibility testing; fifth informational supplement. NCCLS document M100-S5, vol. 14. National Committee for Clinical Laboratory Standards, Villanova, Pa. 24. Physicians Desk Reference. 1995. Physicians desk reference, 49th ed. Medical Economics Data Production Co., Montvale, N.J. 25. Rogers, J. D., M. A. P. Meisinger, F. Ferber, G. B. Calandra, J. L. Demetriades, and J. A. Bland. 1985. Pharmacokinetics of imipenem and cilastatin in volunteers. Rev. Infect. Dis. 7(Suppl. 3):S435 S446. 26. Scully, B. E., N. Chin, and H. C. Neu. 1985. Pharmacology of ticarcillin combined with clavulanic acid in humans. Am. J. Med. 79:39 43. 27. Seifert, H., R. Baginski, A. Schulze, and G. Pulverer. 1993. Antimicrobial susceptibility of Acinetobacter species. Antimicrob. Agents Chemother. 37: 750 753. 28. Sorgel, F., and M. Kinzig. 1993. The chemistry, pharmacokinetics and tissue distribution of piperacillin/tazobactam. J. Antimicrob. Chemother. 31(Suppl. A):39 60. 29. Tilley, P. A., and F. J. Roberts. 1994. Bacteremia with Acinetobacter species: risk factors and prognosis in different clinical settings. Clin. Infect. Dis. 18:896 900. 30. Traub, W. H., and M. Spohr. 1989. Antimicrobial drug susceptibility of clinical isolates of Acinetobacter species (A. baumannii, A. haemolyticus, genospecies 3, and genospecies 6). Antimicrob. Agents Chemother. 33:1617 1619. 31. Urban, C., E. Go, N. Mariano, B. J. Berger, I. Avraham, D. Rubin, and J. J. Rahal. 1993. Effect of sulbactam on infections caused by imipenem-resistant Acinetobacter calcoaceticus biotype anitratus. J. Infect. Dis. 167:448 451. 32. Urban, C., E. Go, N. Mariano, and J. J. Rahal. 1995. Interaction of sulbactam, clavulanic acid and tazobactam with penicillin-binding proteins of imipenem-resistant and -susceptible Acinetobacter baumannii. FEMS Microbiol. Lett. 125:193 198. 33. Vila, J., M. Almela, and M. T. Jimenez de Anta. 1989. Laboratory investigation of a hospital outbreak caused by two different multiresistant Acinetobacter calcoaceticus subsp. anitratus strains. J. Clin. Microbiol. 27:1086 1089. 34. Vila, J., A. Marcos, F. Marco, S. Abdalla, Y. Vergara, R. Reig, R. Gomez-Lus, and T. Jimenez de Anta. 1993. In vitro antimicrobial production of -lactamases, aminoglycoside-modifying enzymes, and chloramphenicol acetyltransferase by and susceptibility of clinical isolates of Acinetobacter baumannii. Antimicrob. Agents Chemother. 37:138 141. 35. Wang, G., T. S. Whittam, C. M. Berg, and D. E. Berg. 1993. RAPD (arbitrary primer) PCR is more sensitive than multilocus enzyme electrophoresis for distinguishing bacterial strains. Nucleic Acids Res. 21:5930 5933.