Branka Bedeni A*, Nata a Beadera and Ivojin Agarb

EFFECT OF INOCULUM SIZE ON THE ANTIBACTERIAL ACTIVITY

OF CEFPIROME AND CEFEPIME AGAINST KLEBSIELLA

PNEUMONIAE STRAINS PRODUCING SHV EXTENDED-

SPECTRUM -LACTAMASES

Branka Bedeni}a*, Nata{a Beadera and @ivojin @agarb

a Branka Bedeni}, MD, PhD, Nata{a Beader, MD, M. sc, Department of

Microbiology, School of Public Health “A. [tampar”, Medical School, The

University of Zagreb, Rockefeller Street 4, 10000 Zagreb, Croatia b @ivojin @agar, former (retired) Professor at Department of Microbiology,

School of Public Health “A. [tampar”, Medical School, The University of

Zagreb, Rockefeller Street 4, 10000 Zagreb, Croatia

*Corresponding adress: Branka Bedeni}, MD, PhD, Department of

Microbiology, School of Public Health “A. [tampar”, Medical School, The

University of Zagreb, Rockefeller Street 4, 10 000 Zagreb, CROATIA, Tel:

+385 1 46 84 443, FAX:+ 385 1 46 84 441

RUNNING TITLE: ANTIBACTERIAL ACTIVITY OF CEFPIROME AND CEFEPIME 2 3

SUMMARY

Objective: to determine the effects of varying inoculum size on in-vitro susceptibility of SHV extended-spectrum -lactamase (ESBL)- producing Klebsiella pneumoniae isolates to cefepime and cefpirome compared to previously established cephalosporins and aztreonam.

Methodology: Antibiotic susceptibilities were determined by disk-diffusion test, MIC broth microdilution method and in time-kill studies. The strains were classified in five groups acccording to the type of -lactamase they produce: SHV-2, SHV-5, SHV-12, putative SHV

ESBL producers and ESBL negative Klebsiellae.

Results: Antibacterial activity of cefpirome and cefepime was comparable to that of cefotaxime but it was significantly stronger than of ceftazidime and aztreonam. An inoculum effect was detected for all broad-spectrum cephalosporins but it was more pronounced with cefpirome and cefepime compared to older cephalosporins. A marked increase in MIC 50 and MIC90 was seen against K. pneumoniae strains producing all three types of SHV ESBLs at inoculum size of 107

CFU/ml compared to 103 CFU/ml. The disk-diffusion test turned out not to be enough sensitive method for detection of an inoculum effect, particularly for cefepime.

Conclusions:The present study found that most SHV producing Klebsiellae produce MICs of cefpirome that imply susceptibility at the low and moderate inoculum sizes and to cefepime at low inoculum size in spite of the fact that according to NCCLS all ESBL producers should be considered resistant to all cephalosporins, with independence of MIC values. With a high inoculum most of the of the strains seemed to be resistant to both antibiotics.

Furthermore bactericidal activity of cefpirome and cefepime against isogenenic E. coli strains producing SHV-2, SHV-4 and SHV-5 -lactamases respectively, was also inoculum- dependent. Bactericidal activity against SHV-4 and SHV-5 -lactamase producers was obtained only at low and moderate inocula while SHV-2 -lactamase producer was efficiantly killed with both antibiotics regardless of the inoculum size.

Keywords: extended-spectrum -lactamases, cefepime, cefpirome 4

INTRODUCTION

The plasmid-mediated extended-spectrum -lactamases confer resistance to oxymino- cephalosporins, such as cefotaxime, ceftazidime, and ceftriaxone and to monobactams such as aztreonam. These enzymes occur predominantly in Klebsiella species (1) and Escherichia coli, but they may also be present in other genera of the familiy Enterobacteriacaeae such as

Citrobacter, Serratia, Proteus, Salmonella, and Enterobacter (2-5).These enzymes are frequently derived from either a TEM or SHV related -lactamase.They hydrolyze oxymino - lactams at rates at least 10% of that observed for benzylpenicillin, in contrast to the parental

TEM-1 or SHV-1 enzymes, which recognize the expanded-spectrum -lactams poorly as substrates. As a result, MICs of the newer cephalosporins and aztreonam are elevated, although clinical resistance is not always evident (6-7).

Since the first discovery in Germany (8), they have been reported in many countries of the world (9).

Cefepime and cefpirome are new fourth generation cephalosporins which were believed to have enhanced activity against ESBL and Amp C–type -lactamase producing Enterobacteriaceae but according to the recent reports their activity is seriously compromised by ESBLs (10).

The aim of this investigation was to determine the effect of varying bacterial density suspensions on the in vitro antibacterial activity of cefpirome and cefepime against Klebsiella pneumoniae strains with previously characterized and sequenced SHV ESBLs (11-12), compared to previously established cephalosporins and aztreonam. 5

MATERIALS AND METHODS

Bacteria

The experiments were performed on a set of K. pneumoniae strains with previously characterized and sequenced -lactamases (11-12) which comprise :

 10 SHV-5 -lactamase- producing K. pneumoniae,

 20 SHV-2 + 1 SHV 2a -lactamase -producing K. pneumoniae,

 7 SHV-12 -lactamase- producing K. pneumoniae,

 39 putative SHV ESBL producing K. pneumoniae based on positive synergistic effect with

clavulanate and isoelectric focusing of 8 representative strains and their transconjugants (pI

of the enzymes was 8.2, enzymes of all strains had similar substrat profiles with high level

ceftazidime and aztreonam resistance)

 26 K. pneumoniae isolates highly susceptible to ceftazidime according to Kirby-Bauer disk-

diffusion method (diameter of the inhibition zones  27 mm) and thus considered to be

ESBL negative.

ESBL producing K. pneumoniae isolates were collected during 1994-1995. and 1997. from the

Sestre Milosrdnice Hospital and Dubrava Hospital iz Zagreb. ESBL-negative, ceftazidime- susceptible K. pneumoniae isolates were obtained from the Sestre Milosrdnice Hospital in

Zagreb during 1994-1995.

Reference strains (E. coli A15 R+) producing SHV-2, SHV-4 and SHV-5 -lactamases respectively, were used in time-kill studies. The strains were kindly provided by Prof. A.

Bauernfeind, Max von Pettenkofer Institute, Munich, Germany.

All isolates were identified by conventional biochemical tests.

Detection of ESBLs.

Production of ESBLs was determined by comparing ceftazidime MIC with and without clavulanic acid. Strains were further examined for SHV ESBLs by PCR/Nhe test (12-13). Due to the glycine (position 238) (SHV- non ESBL)  serine (position 238) (SHV-ESBL) mutation only PCR fragments from the genes coding for SHV-ESBLs are cleaved. 6

Characterization of -lactamases

Transfer of resistance determinants

K. pneumoniae isolates were investigated for the transferability of their resistance determinants.

Conjugation experiments were set up employing E. coli A15 R- strain resistant to rifampicin and

E. coli A15 R- resistant to sodium azide (14).

Preparation of -lactamases for isoelectric focusing.

Cells were harvested from 20 hour Brain-Heart Infusion broth (Oxoid CM 225) cultures by centrifugation, washed and resuspended in 0.5 ml phosphate buffer (0.1 M, pH 7) and - lactamase released by sonication. The samples were kept on ice during sonication which was performed with a W-385 sonicator (Heat Systems-Ultrasonics). After sonication samples were centrifuged at 13000 rpm for 15 minutes to remove debris and then stored at -20° C until use.

Analytical isoelectric focusing (IEF)

IEF was performed in agarose gel containing Pharmalyte, pH 3-9, (Pharmacia Biotech, St

Albans, Hertfordshire, UK). -lactamases were detected by staining the gel with nitrocefin (100 mg/ml, Oxoid SR 112). Preparations of strains known to produce TEM-1, TEM-2, SHV-1,

SHV-2, SHV-4 and SHV-5 were used as standards (15) .

-lactamase preparation for -lactamase assay.

Microorganisms were grown for 18 h at 37° C in a shaking incubator. After centrifugation for

15 minutes at 10 000 g, the cells were washed in 0.1 M phosphate buffer, pH 7, and resuspended in the same buffer to give 20 fold concentrated suspensions. The cells were disrupted by sonication for 7 minutes in an ice bath and the resulting preparations were centrifuged at 13 000 g for 20 minutes. The supernatans were used as crude enzyme preparations.

-lactamase assay

Vmax was measured from crude extracts by a macroiodomethric method (16), and expressed as a percentage of the Vmax for benzylpenicillin. Substrates were prepared as 5 mMl in 0.05 M phosphate buffer pH 7.0 Test and control flasks containing 5 ml substrate were equilibrated at 7

37 °C in a shaking water bath before adding 1 ml of enzyme to each of the test flasks.

Following a reaction period of 30 minutes 10 ml of iodine reagent (0.0166 M iodine, 0.06 M potassium iodide in 1.75 M sodium acetate buffer, pH 4.0) were added to stop the enzymatic reaction. A further incubation period (10 minutes for penicillins and 20 minutes for cephalosporins) was required for completition of the reaction between the hydrolysis products and iodine. 1 ml of enzyme was added in the control flask after incubation with iodine. The flasks were then removed from the water bath and titrated with sodium thiosulphate using a starch indicator.

Under these conditions 1 ml of 0.0166 M iodine reduced was equivalent to 2 ml of penicillin substrate or 4 ml of cephalosporin substrate.

One unit of b-lactamase (U) was defined as the amount of enzyme that hydrolysed 1 mmol of benzylpenicillin/min at pH 7 and 37° C.

DNA preparation

Total DNA for use as a template in PCR assays was prepared as described previously (12).

PCR and sequencing of blaSHV genes

PCR and sequencing of blaSHV genes was performed as described previously (12).

Susceptibility tests

Kirby-Bauer disk diffusion test.

The test was performed with three different inocula of 103, 105 and 107 CFU/ml. Inocula sizes were verified by serial dilution and plating techniques (17). Mueller-Hinton agar was used.

Ceftazidime (30 g), cefotaxime (30 g) and ceftriaxone (30 g) disks were purchased from

Imunoloski Zavod (Zagreb, Croatia). Cefepime (30 g) and aztreonam (30 g) disks were obtained from Becton-Dickinson. The test was interpreted according to the NCCLS (18). The strains were considered resistant if the zone diameter was equivalent or less than 14 mm for ceftazidime, cefotaxime, cefepime, cephalexin and cefuroxime, equivalent or less than 13 mm for ceftriaxone and equivalent or less than 15 mm for aztreonam. If there was no inhibition zone around the disk the zone diameter was read as 0. 8

Determination of minimum inhibitory concentrations (MIC)

MICs of cefuroxime, cephalexin, ceftazidime, cefotaxime, ceftriaxone, cefpirome, cefepime and aztreonam were determined by broth microdilution method in Mueller-Hinton broth, also with three different inocula of 103, 105 and 107 CFU/ml as described previously (19). The following

MIC resistance breakpoints were used: for cefuroxime, cephalexin, ceftazidime, cefpirome, cefepime and aztreonam 32 mg/L and for cefotaxime and ceftriaxone 64 mg/L. Antibiotic powders were obtained from the following sources: ceftazidime, cephalexin and cefuroxime,

Pliva, Zagreb, Croatia; cefepime and aztreonam, Bristol Myers Squibb; cefpirome and cefotaxime, Hoechst AG; and ceftriaxone Hoffmann - La Roche AG. An E. coli 25922 was used as the reference strain for quality control.

Determination of minimum bactericidal concentrations (MBC)

MBCs of ceftazidme, cefpirome and cefepime were determined by subculturing the wells which had no visible growth by removing 10 l from each clear well and spoting the sample on MH agar plate (20). The MBC was defined as the lowest concentration of the antibiotic that reduced the inoculum by 99.9% within 24 hours.

Time kill studies

Reference strains (E. coli A15 R+) producing SHV-2, SHV-4 and SHV-5 -lactamases were used in time-kill studies.

A bacterial inoculum in the logarithmic phase of growth was added to 25 ml flask containing antibiotic in the concentration equivalent to the break-point (32 mg/L). The bacterial concentration was adjusted to the final concentration of 103, 105 and 107 CFU/ml. The flasks were incubated in a shaking water bath at 37C, and samples were removed after 2, 4, 6 and 24 hours for enumeration by serial dilution and plating (21). MICs of ceftazidime, cefotaxime, ceftriaxone, cefpirome, cefepime and aztreonam were determined for the regrowth organisms and compared with those of the initial strains. 9

RESULTS

The results of the susceptibility tests are shown in the Tables 1, 2, 3 and 4. At the standard inoculum size of 105 CFU/ml, cefpirome and cefepime were visibly more active against SHV-2 producers than cefuroxime and cephalexin and equally active as cefotaxime and ceftriaxone

(Table 1). At the low inoculum all SHV-2 -lactamase producing isolates were susceptible to cefepime, cefpirome, cefotaxime and ceftriaxone (Table 2). On the other hand, at the high inoculum there was a higher percentage of strains resistant to cefepime and cefpirome than to cefotaxime and ceftriaxone (Table 2).

Cefpirome exhibited the strongest inhibitory effect agains SHV-5 -lactamase producers at the standard inoculum size while cefepime was equally active as cefotaxime and ceftriaxone.

However the cefepime MIC for 90% of such strains was at the resistance breakpoint (32 mg/L).

When the low inocum was applied all SHV-5- producing strains were susceptible to cefepime, cefpirome, and ceftriaxone while there was still a high percentage of strains resistant to ceftazidime, aztreonam and older-generation cephalosporins (Table 2). At the high inoculum cefpirome, cefotaxime and ceftriaxone more efficiently inhibited SHV-5 producers than cefepime. There was a higher percentage of strains resistant to cefpirome than to ceftriaxone and cefotaxime. Ceftazidime and aztreonam had high MICs whatever the inoculum tested

(Table 1). Disk-diffusion procedure did not detect the inoculum effect of cefepime (Table 3).

At the high inoculum size antibiograms still showed susceptibility while MICs were significantly elevated.

At the standard inoculum size cefpirome was slighty more active against SHV-12 producers than cefotaxime and ceftriaxone and markedly more active than ceftazidime, cefuroxime and cephalexin. The increase of the inoculum size to 107 CFU/ml decreased the activity of cefepime and cefpirome more significanty than that of cefotaxime and ceftriaxone.

Putative SHV ESBL producers were most efficiently inhibited by cefpirome, cefepime and cefotaxime at the moderate inoculum size. At the low inoculum size cefepime exibited stronger 10

antibacterial activity than cefotaxime, cefpirome and ceftriaxone. A marked reduction in MIC90 of cefepime and cefpirome was seen at inoculum density of 103 compared to 107.

Non ESBL producing Klebsiellae were uniformly susceptible to third and fourth generation cephalosporins when tested at standard and low inoculum size. Inoculum density over a range of

3 5 10 to 10 CFU/ml had little or no effect on MIC90 of most cephalosporins against ESBL negative Klebsiellae. However increasing the inoculum size to 107 CFU/ml clearly increased the concentration of ceftazidime, cefotaxime,cefepime, cefpirome and aztreonam required to inhibit

90% of the strains. For cefuroxime, cephalexin and ceftriaxone the MICs were similar regardless of the inoculum size.

The inoculum effect was strain dependent regardless of the type of -lactamase. It was more apparent when microdilution test was applied compared to disk-diffusion procedure (Table 3).

MBCs of ceftazidime, cefpirome and cefepime were usually the same as MICs of those antibiotics or at the most two dilution higher than MIC regardless of the inoculum density or the type of -lactamase (Table 4).

Analysis of time-kill curves (Figure 1) showed the SHV-5 producing strain exposed to ceftazidime and cefpirome at the breakpoint concentration had a regrowth after 24 hours regardless of the initial bacterial concentration. At the high initial inoculum there was no evidence of antibacterial activity. With cefepime, the low inoculum initial culture was sterile after 24 hours (Figure 1).

The SHV-2 producing strain was efficiantly killed with cefepime and cefpirome at the breakpoint concentration regardless of the starting inoculum size (Figure 2). This could be explained by the fact that cefpirome and cefepime MICs of that strain are significantly below the breakpoint concentration. Ceftazidime was immediately bactericidal at the low and moderate inoculum size. At the high inoculum size there was still 100 colonies after 24 hours.

With the SHV-4 producer there was a high rate of killing by cefpirome and cefepime at low and moderate inocula. At the high inoculum the both antibiotics were ineffective (Figure 3). 11

DISCUSSION

It is a well know fact that ESBL producing bacteria exhibit a pronounced inoculum effect against broad spectrum cephalosporins like ceftazidime, cefotaxime, ceftriaxone and cefoperazone (22,23) but there are not many reports on their inoculum effect against fourth generation cephalosporins: cefepime and cefpirome.

Numerous in-vitro studies have reported the enhanced activity of cefepime and cefpirome against Gram-negative organisms resistant to other extended-spectrum cephalosporins.However the in vitro activity of fourth generation cephalosporins is significantly decreased by the production of ESBLs by Enterobacteriaceae. This activity is greatly influenced by which specific ESBL is present. Although the in vitro activity of cefepime and cefpirome against most

ESBL producing Enterobacteriaceae exceeds that of ceftazidime (except CTX-M-type), it may be similar to that of cefotaxime against a number of strains. Since cefotaxime has been shown not to be generally efficiant against infections caused by ESBL producing Enterobacteriaceae despite MIC in the susceptible range, the clinical efficacy of the fourth generation cephalosporins must remain in question at this time. Data from in vitro studies suggest that clinical efficacy will probably depend upon which particular ESBL is being produced by the strain involved (10). However according to the last NCCLS edition ESBL producing strains should be considered resistant to all cephalosporins with independence of MIC values (19) but in the routine laboratory the important question is whether the organism is susceptible and certainly at the early stages it is not known and not relevant whether the organism is an ESBL producer. For that reason it is very important to use the suitable inoculum for the susceptibility testing of any organisms including ESBL producers. In the previous studies it was determined that cefepime is at least two fold more active against bacteria producing SHV-2, SHV-3 or

SHV-4 -lactamase than other cephalosporins including cefotaxime, cefpirome, ceftazidime and ceftriaxone (24) but the study was pursued on laboratory E. coli strains harbouring the mentioned ESBLs. Clinical K. pneumoniae isolates may have other resistance mechanisms like lack of porins which could influence the antibacterial activity of fourth generation cephalosporins. 12

The present study found that most SHV producing Klebsiellae were indeed susceptible to cefpirome at the low and moderate inoculum size and to cefepime at low inoculum size according to the MIC determination. Using a high inoculum most of the of the strains were resistant to both antibiotics. Discrepancies were found for cefepime between the results of the

MIC determination by broth microdilution method and disk-diffusion procedure when the high inoculum was applied. The size of the inhibition zones indicated susceptibility while MICs were above the susceptibility breakpoint. In principle the disk-diffusion test was found no to be an accurate method for detection of inoculum effect against SHV ESBL producing Klebsiellae.

The effect of inoculum density was observed to be strain dependent. It was more pronounced when the MIC for the particular strain was lower. The possible explanation is that the strains for which the cephalosporin MICs are lower produce less amount of -lactamase. As a result, higher inocula are needed to demonstrate abrogation of the inhibitory effect of a cephalosporin.

The inoculum effect was clearly observed in our time-kill studies on the bactericidal effect of cefepime. The drug was bactericidal using low inoculum but it was either bacteriostatic or permitted regrowth of organism with higher starting bacterial concentrations. There was no evidence that this regrowth was associated with emergence of a resistant mutant since the

MIC of cefepime for the regrowth organism remained the same as for the initial strain

(results not shown).

Factors causing an inoculum effect in vitro have not been clearly defined. However, it is most commonly associated with -lactam antibiotics and probably represents greater hydrolysis by the cumulative activity of -lactamase. The therapeutic significance of an observed inoculum effect is also uncertain. In theory, the inoculum effect may be an important factor in the management of infections involving high bacterial concentrations. The therapeutic failures of fourth generation cephalosporins in treatment of infections caused by ESBL producing bacteria could be in part explained by high densitiy of bacteria at the infection site bearing in mind that there may be other factors contributing to that like lack of porins. 13

ACKNOWLEDGEMENTS

I thank Dr. Herbert Heachler (Institut for Medical Microbiology, University of Zuerich,

Switzerland) for reading the manuscript. 14

REFERENCES

1. Katsanis G, Jacoby G. The frequency of extended-spectrum b-lactamases in isolates of

Klebsiella pneumoniae. J Antimicrob Chemother 1992;29:345-353.

2. Gianneli D, Tzelepi E, Tzouvelekis LS, Mentis AF, Nikolopoulou C. Dissemination of cephalosporin resistant Serratia marcescens strains producing plasmidic SHV b-lactamase in

Greek hospitals. Eur J Clin Microbiol Infect Dis 1994;13:764-767.

3. Morosini MI, Canton R, Martinez-Beltran J, Negri MC, Perez-Diaz JC, Baquero F, Blazquez

J. New extended-spectrum TEM type b-lactamase from Salmonella enterica subsp. enterica isolated in a nosocomial outbreak. Antimicrob Agents Chemoher 1995;39:458-461.

4. Neuwirth C, Siebor E, Lopez J, Pechinot A, Kazmiercak A. Outbreak of TEM-24 producing

Enterobacter aerogenes in an intensive care unit and dissemination of the extended-spectrum - lactamase to other members of the family Enterobacteriaceae. J Clin Microbiol 1996;34:76-79.

5. Palzkill T, Thomson KS, Sanders CC, Moland ES, Hang W, Milligan TW. New variant of

TEM-10 -lactamase gene produced by a clinical isolate of Proteus mirabilis. Antimicrob Agent

Chemother 1995;39:1199-1200.

6. Shannon K P, King A, Philips I, Nicolas M H, Philippon A. Importation of organisms producing broad-spectrum SHV-group b-lactamases into the United Kingdom. J Antimicrob

Chemother 1990; 25: 343-351.

7. Katsanis G P, Spargo J, Ferraro M J, Sutton L, Jacoby G A. Detection of Klebsiella pneumoniae and Escherichia coli strains producing extended-spectrum b-lactamases. J Clin

Microbiol 1994; 32(3):691-696.

8. Knothe H , Shah P, Krcmery V, Antal M, Mitsuhashi S. Transferable resistance to cefotaxime, cefoxitin, cefamandole and cefuroxime in clinical isolates of Klebsiella pneumoniae and Serratia marcescens. Infection 1983., 6: 315-317.

9. Jacoby G A, Medeiros A A. More extended-spectrum b-lactamases. Antimicrob Agent

Chemother 1991; 35(9): 1697-1704. 15

10. Sanders CC. In vitro activity of fourth generation cephalosporins against Enterobacteriaceae producing extended-spectrum -lactamases. J Chemother 1996;8(suppl.2): 57-63.

11. Bedenic B, Zagar Z. Extended-spectrum b-lactamases in clinical isolates of Klebsiella pneumoniae from Zagreb, Croatia. J Chemother 1998; 10(6):449-459.

12. Bedenic B, Randegger C, Stobberingh EE, Herbert Haechler. Molecular epidemiology of extended-spectrum -lactamases from Klebsiella pneumoniae strains isolated in Zagreb, Croatia

(unpublished results).

13. Nuesch-Inderbinen M, Haechler H, Kayser FH. Detection of extended-spectrum SHV - lactamases in clinical islates by a precise genetic method and comparison with E test. Eur J Clin

Infect Dis1996, 15:398-402.

14. Elwell LP, Falkow S. The characterization of R plasmids and the detection of plasmid- specified genes. In: Lorian Eds. Antibiotics in Laboratory Medicine, 2nd ed. Baltimore:

Williams & Wilkins, 1986: 683-721.

15. Shannon K, King A, Phillips I. -lactamases with high activity against imipenem and Sch

34343 from Aeromonas hydrophila. J Antimicrob Chemother 1986., 17: 45-50.

16. Perret CJ. Iodometric assay of Penicillinase. Nature 1954., 174: 1012-1013.

17. Johnson CC, Livornese L, Gold MJ, Pitsakis PG, Taylor S, Levinson ME. Activity of cefepime against ceftazidime-resistant Gram-negative bacilli using low and high inocula. J

Antimicrob Chemother 1995; 35:765-773.

18. NCCLS. Performance standards for antimicrobial disk susceptibility tests - fifth edition; approved standard. In: NCCLS document M2-A5. NCCLS. Villanova, Pa: 1993.

19. NCCLS. Methods for dilution antimicrobial susceptibility tests for bacteria that grow aerobically; approved standard-fifth edition. NCCLS document M7-A5.NCCLS:Wayne,

Pennsylvania;2000.

20. Stratton C W, Cooksey R C. Susceptibility tests: Special tests. In:Balows A, Hausler W,

Herrmann KL, Isenberg HD, Shadomy HJ, eds. Manual of Clinical Microbiology, 4th edn,

Washington DC: American Society for Microbiology: 1155-1156. 16

21. Krogstadt DJ, Moellering RC Jr. Combinations of antibiotics. In Lorian V, ed. Antibiotics in laboratory medicine, 2nd edn. Baltimore: Williams & Wilkins, 1986:537-78.

22. Fantin B, Pangon B, Potel G, Caron F, Vallee E, Vallois JM, Mohler J, Bure A, Philippon

A, Carbon C. Activity of sulbactam in combination with ceftriaxone in vitro and in experimental endocarditis caused by Escherichia coli producing SHV-2 like -lactamase.

Antimicrob Ag Chemother 1990;34(4):5811-586

23. Rice LB, Yso J, Klimm K, Eliopoulos GM, Moellering RC. Efficacy of different -lactams against an extended-spectrum -lactamase producing Klebsiella pneumoniae strain in the rat intra-abdominal abscess model. Antimicrob Ag Chemother 1991;35(6):1243-1244.

24. Jacoby GA, Carreras I. Activities of -lactam antibiotics against Escherichia coli strains producing extended-spectrum -lactamases. Antimicrobl Ag Chemother 1990; 34:858-862. 17

14. Arlet G, Brami G, Decre D, Flippo A, Gaillot O, Lagrange PH, Philippon A. Molecular characterisation by PCR-restriction fragment length polymorphism of TEM -lactamases.

FEMS Microbiology letters 1995; 134:203-208. 18

REFERENCES

1. Sirot, D. J, Sirot, J., Labia, R., Morand, A., Courvalin, P., Darfeuille-Michaud, A., Perroux

R. & Cluzel, R. (1987).Transferable resistance to third generation cephalosporins in clinical isolates of Klebsiella pneumoniae: identification of CTX-1, a novel b-lactamase. Journal of

Antimicrobial Chemotherapy 20, 323-334.

2. Gianneli, D., Tzelepi, E., Tzouvelekis, L.S., Mentis, A.F & Nikolopoulou, C.(1994)

.Dissemination of cephalosporin resistant Serratia marcescens strains producing plasmidic

SHV b-lactamase in Greek hospitals. European Journal of Clinical Microbiology and Infectious

Diseases 13, 764-767.

3. Morosini, M.I., Canton, R., Martinez-Beltran, J., Negri, M.C., Perez-Diaz, J.C., Baquero, F.

& Blazquez, J.(1995).New extended-spectrum TEM type b-lactamase from Salmonella enterica subsp. enterica isolated in a nosocomial outbreak. Antimicrobial Agents and Chemoherapy

39,458-461.

4. Neuwirth, C., Siebor, E., Lopez, J., Pechinot, A. & Kazmiercak, A. (1996). Outbreak of

TEM-24 producing Enterobacter aerogenes in an intensive care unit and dissemination of the extended-spectrum -lactamase to other members of the family Enterobacteriaceae. Journal of

Clinical Microbiology 34, 76-79.

5. Palzkill, T., Thomson, K.S., Sanders, C.C., Moland, E.S., Hang, W. & Milligan, T.W.

(1995).New variant of TEM-10 -lactamase gene produced by a clinical isolate of Proteus mirabilis. Antimicrobial Agents and Chemotherapy 39,1199-1200.

6. Shannon, K. P., King, A., Philips, I., Nicolas, M. H. & Philippon, A. (1990). Importation of organisms producing broad-spectrum SHV-group b-lactamases into the United Kingdom.

Journal of Antimicrobial Chemotherapy 25, 343-351.

7. Katsanis, G. P., Spargo, J., Ferraro, M. J., Sutton, L.& Jacoby, G. A. (1994) Detection of

Klebsiella pneumoniae and Escherichia coli strains producing extended-spectrum b-lactamases.

Journal of Clinical Microbiology 32:691-696. 19

8. Kliebe, C., Nies, B. A., Meyer, J. F., Tolxdorff-Neutzling, R. M. & Wiedemann, B. (1985)

Evolution of plasmid-coded resistance to broad-spectrum cephalosporins. Antimicrobial Agents and Chemotherapy 28, 302-307.

9. Knothe, H., Shah, P., Krcmery, V., Antal, M. & Mitsuhashi, S. (1983) Transferable resistance to cefotaxime, cefoxitin, cefamandole and cefuroxime in clinical isolates of Klebsiella pneumoniae and Serratia marcescens. Infection 6, 315-317.

10. Jacoby, G. A., Medeiros, A. A. More extended-spectrum b-lactamases. (1991).

Antimicrobial Agents and Chemotherapy 35, 1697-1704.

11. Sanders, C. C. (1996). In vitro activity of fourth generation cephalosporins against

Enterobacteriaceae producing extended-spectrum -lactamases. Journal of Chemotherapy 8

(suppl.2), 57-63.

12. Bedenic, B., Randegger, C., Haechler, H., Stobberingh, E. (2000). Unpublished results.

13. Nuesch-Inderbinen, M., Haechler, H. & Kayser, F.H. (1996). Detection of extended- spectrum SHV -lactamases in clinical isolates by a precise genetic method and comparison with E test. European Journal of Clinical Microbiology and Infectious Diseases 15, 398-402.

14. National Committee for Clinical Laboratory Standards. (1993). Performance standards for antimicrobial disk susceptibility tests - fifth edition; approved standard. In: NCCLS document

M2-A5. National Committee for Clinical Laboratory Standards, Villanova, Pa.

15. Jones, R. N., Barry, A. A., Gavan, T. L. & Washington, J. A. (1991)). Susceptibility tests: microdilution and macrodilution broth procedures. In Manual of Clinical Microbiology, 4th edn

(Balows, A., Ed.) pp. 972-977 . American Society for Microbiology, Washington DC.

16. Stratton, C. W. & Cooksey, R. C. (1991) Susceptibility tests: Special tests. In Manual of

Clinical Microbiology, 4th edn (Balows, A., Ed.) pp. 972-977. American Society for

Microbiology, Washington DC.

17. Krogstadt, D.J.& Moellering, R.C. Jr. (1986). Combinations of antibiotics. In Antibiotics in laboratory medicine, 2nd edn (Lorian, V., Ed.) 537-78. Williams & Wilkins, Baltimore, MD. 20

18. Fantin, B., Pangon, B., Potel, G., Caron, F., Vallee, E., Vallois, J.M., Mohler, J., Bure, A.,

Philippon, A. & Carbon, C. (1990) Activity of sulbactam in combination with ceftriaxone in vitro and in experimental endocarditis caused by Escherichia coli producing SHV-2 like - lactamase. Antimicrobial Agents and Chemotherapy 34,5811-586

19. Rice, L.B., Yso, J., Klimm, K., Eliopoulos, G.M. & Moellering, R.C. (1991).Efficacy of different -lactams against an extended-spectrum -lactamase producing Klebsiella pneumoniae strain in the rat intra-abdominal abscess model. Antimicrobial Agents and

Chemotherapy 35,1243-1244.

20. Jacoby, G. A. & Carreras, I. (1990). Activities of -lactam antibiotics against Escherichia coli strains producing extended-spectrum -lactamases. Antimicrobial Agents and

Chemotherapy 34,858-862. 21

FIGURES LEGEND

FIGURE 1. Effect of inoculum density on the antibacterial activity of cefpirome and cefepime against E. coli A 15 R+ producing SHV-5 -lactamase.