Skip Navigation


JAC Advance Access originally published online on March 29, 2007
Journal of Antimicrobial Chemotherapy 2007 59(5):926-933; doi:10.1093/jac/dkm073
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
59/5/926    most recent
dkm073v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Buijs, J.
Right arrow Articles by van der Meer, J. W. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Buijs, J.
Right arrow Articles by van der Meer, J. W. M.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author 2007. Published by Oxford University Press on behalf of the British Society for Antimicrobial Chemotherapy. All rights reserved. For Permissions, please e-mail: journals.permissions@oxfordjournals.org

Continuous administration of PBP-2- and PBP-3-specific ß-lactams causes higher cytokine responses in murine Pseudomonas aeruginosa and Escherichia coli sepsis

Jacqueline Buijs1,2,*, Anthonius S. M. Dofferhoff1,3, Johan W. Mouton1 and Jos W. M. van der Meer4,5

1 Department of Medical Microbiology and Infectious Diseases, Canisius-Wilhelmina Hospital, Nijmegen, The Netherlands 2 Department of Internal Medicine, Atrium Medical Centre, Heerlen, The Netherlands 3 Department of Internal Medicine, Canisius-Wilhelmina Hospital, Nijmegen, The Netherlands 4 Nijmegen University Centre for Infectious Diseases, University Medical Centre, Nijmegen, The Netherlands 5 Department of General Internal Medicine, University Medical Centre, Nijmegen, The Netherlands

* Correspondence address. Atrium Medical Centre, PO Box 4446, 6401CX Heerlen, The Netherlands. Tel: +31-45-5766666; Fax: +31-45-5713360; E-mail: j.buijs{at}atriummc.nl/ jacqbuijs{at}hotmail.com

Received 21 December 2006; returned 22 January 2007; revised 11 February 2007; accepted 12 February 2007


   Abstract
Top
 Abstract
Introduction
 Materials and methods
 Results
 Discussion
 Transparency declarations
 References
 
Objectives: Initial antibiotic treatment of severe infections can lead to clinical deterioration due to sudden endotoxin release and concomitant exaggerated inflammatory response. Antibiotic-induced morphological changes may contribute to this phenomenon. High-dose ceftazidime, which inhibits penicillin-binding protein (PBP)-1 in Gram-negative bacteria, causes quick bacteriolysis and low endotoxin release. Low-dose ceftazidime leads to PBP-3 inhibition, which causes bacterial filament formation, associated with high endotoxin releases. PBP-2-specific antibiotics induce spheroplasts, again associated with low endotoxin release. We hypothesized that antibiotic type, concentration and regimen influence bacterial morphology, endotoxin levels and inflammatory response.

Methods: Neutropenic mice with Escherichia coli or Pseudomonas aeruginosa sepsis were treated with ceftazidime or meropenem 10–320 mg/kg as an intravenous bolus or as continuous tail vein infusions of 0.1 mL/h. Four hours later, bacterial counts, morphology, plasma endotoxin, pro-inflammatory cytokines [tumour necrosis factor-{alpha} (TNF-{alpha}) and interleukin-6 (IL-6)] and antibiotic concentrations were measured.

Results: Continuous infusion of 80 mg/kg ceftazidime was the lowest dose preventing filaments in E. coli infections. Bolus treatment resulted in filament formation, irrespective of the dose. During continuous treatment, IL-6 and TNF-{alpha} concentrations were higher compared with bolus treatment and controls for both antibiotics and both strains. A clear relationship between cfu counts in muscle and circulating IL-6 was shown (r = – 0.579, P = 0.007), suggesting that plasma IL-6 is a valuable indicator of bacterial killing at the infection site.

Conclusions: Our findings show that not PBP affinity but the method of antibiotic administration is crucial during initial treatment of severe infections.

Keywords: ceftazidime , meropenem , TNF-{alpha} , IL-6 , endotoxin


   Introduction
Top
 Abstract
 Introduction
Materials and methods
 Results
 Discussion
 Transparency declarations
 References
 
Sepsis is a serious life-threatening condition needing prompt antibiotic treatment. Overall sepsis mortality is high (18% to 56%),1,2 with Gram-negative bacteria being causative in 20% of cases.3 Early and effective treatment, preventing severe inflammation and concomitant organ dysfunction, is essential. However, early antibiotic treatment may sometimes induce unexpected clinical deterioration, possibly because of an exaggerated inflammatory response.4–7

Endotoxin or lipopolysaccharide (LPS) is the central mediator in the pathogenesis of sepsis and its inflammatory response.8,9 There are strong relationships between endotoxin exposure and sepsis symptoms,10–12 as well as between bacterial endotoxin concentrations and production of pro-inflammatory cytokines, such as tumour necrosis factor-{alpha} (TNF-{alpha}) or interleukin-6 (IL-6).13–16 Bacterial endotoxin release after exposure to antibiotics depends on penicillin-binding protein (PBP)-binding characteristics of the antibiotic.17–19 PBPs are anchored in the bacterial cytoplasmic membrane. They determine cell shape, phage resistance, induction of capsule synthesis and regulation of autolysis.20 The ß-lactam antibiotic ceftazidime has a varying dose-dependent binding affinity for different PBPs. High ceftazidime concentrations cause PBP-1 inhibition, quick bacterial lysis without morphological changes and low endotoxin release. Low concentrations of ceftazidime cause PBP-3 binding, disabling bacterial septation capacity, leading to growth in filamentous forms (Figure 1).21,22 These filaments are associated with a high endotoxin release23,24 and high TNF-{alpha} and IL-6 production by exposed monocytes.25 In contrast, meropenem, which binds PBP-2, causes formation of large rounded cells, called spheroplasts.26 In contrast to filaments, spheroplasts are not associated with higher endotoxin releases.16,23


Figure 1
View larger version (111K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 1.. Filaments in thigh muscle (stained with haematoxylin–eosin).

 
In this study, we speculated that antibiotic type, initial dosing regimen and concentration levels may shift PBP affinities, alter bacterial morphological responses and thereby influence endotoxin release and clinical effect. We compared the effects of two different ß-lactam antibiotics (ceftazidime and meropenem) in two dosing schedules (continuous and intermittent) in sepsis induced by two strains of bacteria, using a murine model that has been previously shown to provide the most uniform sepsis response patterns.16


   Materials and methods
Top
 Abstract
 Introduction
 Materials and methods
Results
 Discussion
 Transparency declarations
 References
 
Mice

Female outbred Swiss mice (aged 6–8 weeks, weight 20–30 g) were used in all studies. Mice were bred by Harlan (Horst, The Netherlands) and arrived at the Central Animal Laboratory 2 weeks before the start of experiments. They were held at a specific pathogen-free unit in filter-top cages with sterilized sawdust. Food (Hope Farms, Woerden, The Netherlands) and acidified water were supplied ad libitum. The temperature was kept at 20–24°C, with a relative humidity of 50% to 60%, 15 air changes/h and light and dark periods of 14 and 10 h, respectively. All experiments were approved by the Ethics Committee on animal experiments of the University Medical Centre Nijmegen (18 March 2002, registration number KUNDEC 2002–25).

Induction of neutropenia and anaesthesia

On day –4, mice received cyclophosphamide (Endoxan-ASTA®; Dagra, Diemen, The Netherlands) 150 mg/kg subcutaneously to induce neutropenia. On day –1, a second dose of 100 mg/kg cyclophosphamide was given.27,28 Leucocyte counts confirmed a neutropenia of <0.1 x 106 leucocytes/L. During injections, infusions and blood collection, mice received inhalation anaesthesia with isoflurane. Body temperature was regulated by a heating mattress.

Microorganisms

Escherichia coli strain ATCC 25922 (Rosco Diagnostica, Taastrup, Denmark) was used. MICs of ceftazidime and meropenem were 0.25 and 0.025 mg/L, respectively, as determined by Etest and microdilution methods. Pseudomonas aeruginosa strain ATCC 27853 was used, with MICs of 1 mg/L of ceftazidime and meropenem. Bacteria were stored at –80°C and subcultured on blood agar 2 days before the experiment. Subsequently, overnight cultures were prepared by inoculation of two colonies in 10 mL of Mueller–Hinton broth. On test days, 2 h prior to inoculation, bacteria were diluted and resuspended in fresh Mueller–Hinton broth. At t = – 2 h, a standardized inoculum was injected into the left thigh muscle of each mouse.29–31

Antibiotics

Ceftazidime (GlaxoSmithKline B.V., Zeist, The Netherlands), with a dose-dependent affinity for PBP-3 and -1, and meropenem (AstraZeneca, Zoetermeer, The Netherlands), with a non-dose-related high affinity for PBP-2 and a low affinity for PBP-3, were used.32 A stock solution of each antibiotic of 5 mg/mL was divided into pyrogen-free Eppendorf tubes, stored at –80°C and used for all experiments. Thirty minutes before each experiment, one tube was defrosted and diluted in sterile 0.9% NaCl to the desired concentrations. In continuously treated mice, an intravenous (iv) cannula (Neoflon 0.7 x 19 mm, 24 gauche; Becton Dickinson Infusion Therapy AB, Helsingborg, Sweden) was inserted into the tail vein at t = 0 h. In the bolus group, an iv injection was given in the tail vein at t = 0 h

Blood and tissue collection and bacterial counts

Blood samples from the retro-orbital plexus were collected in sterile pyrogen-free Falcon tubes (Becton Dickinson, NJ, USA), containing 10 µL of a pyrogen-free heparin solution containing 5000 U of heparin/mL (Leo Pharmaceutical Products B.V., Weesp, The Netherlands), immediately stored on ice and centrifuged at 500 g for 5 min to obtain plasma samples. Plasma (100 µL) was diluted in 900 µL of 0.9% NaCl for endotoxin assays. Undiluted plasma samples were stored in sterile Eppendorf tubes at –80°C until assays were performed. Liver, spleen and thigh muscles were removed aseptically, immersed in saline, weighed and homogenized in a sterilized tissue grinder. Serial dilutions of homogenized tissues were plated on sheep blood agar. After overnight incubation at 37°C, cfu counts were performed and expressed as numbers of cfu/liver, cfu/spleen and cfu/g muscle.

Bacterial morphology

Bacterial morphology was judged by one of the investigators and a pathologist on haematoxylin–eosin-stained histological sections of thigh muscles, using light microscopy with a magnification of 400 x and 1000 x . Bacteria were defined as filaments when a length of three bacilli or more was observed.

Endotoxin and cytokine assays

Endotoxin concentrations of plasma were measured with a chromogenic limulus amoebocyte lysate (LAL) assay (Associates of Cape Cod Inc., Falmouth, MA, USA) according to the instructions of the manufacturer. Measurement of plasma TNF-{alpha} concentrations was performed by radioimmunoassay.33 IL-6 concentrations were determined in plasma by ELISA (Biosource International, Camarillo, CA, USA). Detection limits of the cytokine assays were 40 and 150 pg/mL, respectively.

Pharmacokinetics

Plasma concentrations of ceftazidime were checked by HPLC.34,35 For the HPLC assay, a Shimadzu with UV/Vis detector was used; flow 1.0 mL/min and pressure ± 120 bar. The method was linear over the dosing range studied, the lowest level of detection 0.1 mg/L, the intra- and inter-assay variance of measurements 1.1% to 3.8% (performed in triplicate on separate days) and spike recovery 98%. Pharmacokinetic parameters were calculated using the Winnonlin program (Pharsight Corp., USA). The Akaike value was used as a parameter to choose the best model describing these results.

Statistical analysis

Statistical analysis was performed using the SPSS package for Windows (SPSS 10.0, Chicago, IL, USA) and GraphPad Prism 3.0 (GraphPad software Inc., San Diego, CA, USA). Normal distribution was checked by Levene's test for equality of variances. Differences in viable counts and endotoxin and cytokine concentrations were analysed by multivariate analysis (ANOVA) for normally distributed parameters, followed, when appropriate, by post hoc t-tests. Parameters without a normal distribution were analysed with the non-parametric Kruskal–Wallis test, followed by the Mann–Whitney test to investigate differences among groups. Correlations among parameters were determined with the Pearson correlation test or Spearman's rank correlation test. Statistical tests were considered significant at P < 0.05.

Outline of the studies

Control studies. Baseline pharmacokinetics of a bolus injection of ceftazidime 80 mg/kg iv was documented in a group of 14 mice. After injections, at t = 1, 5, 10, 20, 30, 40 and 60 min, plasma samples were collected. Pharmacokinetic differences between infected and non-infected mice were determined in 36 neutropenic mice during continuous and intermittent schedules.

The course of parameters during untreated E. coli sepsis was separately studied in 10 mice. To evaluate effects of anaesthesia, injections and the presence of an infusion cannula on release of cytokines, a sham experiment was performed in 18 mice, receiving only saline as inoculum and saline treatments.

Main experiments: continuous versus intermittent dosing schedules of ceftazidime in E. coli sepsis. In seven experiments, a total of 83 neutropenic mice were challenged with 1 x 107 cfu of E. coli. Two hours after inoculation, mice received injections or continuous infusions of ceftazidime. In continuously treated mice, 20% was administered as a loading dose, and the remaining 80% was infused by a Terofusion syringe pump, model STC-521, for the next 4 h. Experiments were performed with 10, 20, 40 and 80 mg/kg ceftazidime (0.25, 0.50, 1.0 and 2.0 mg/mouse). Experiments with 40 and 80 mg/kg were performed twice. In mice treated with 20 and 80 mg/kg ceftazidime, endotoxin assays were performed, comparing the highest dose-level with lower doses. In an 80 mg/kg ceftazidime experiment, the time course of cytokine concentrations was also studied.

To check whether changes in cytokine patterns during continuous and intermittent schedules are unique for the combination of E. coli and ceftazidime, comparison studies of meropenem and ceftazidime were performed in infections with E. coli or P. aeruginosa at various concentrations.

Because the MIC of meropenem for E. coli is 10-fold lower, compared with that of ceftazidime, the dose of meropenem was adjusted in the E. coli experiments. For P. aeruginosa, MICs of both antibiotics are comparable, so equal doses could be applied.


   Results
Top
 Abstract
 Introduction
 Materials and methods
 Results
Discussion
 Transparency declarations
 References
 
Morphology and bacterial counts

Continuous infusion of 80 mg/kg ceftazidime was the lowest dose preventing filament formation in E. coli infections. Morphological studies showed bacterial debris and bacilli. In all other groups, including mice treated with bolus injections of 80 mg/kg ceftazidime, large filaments were seen (Figure 1). Meropenem induced spheroplast formation in all regimens in both infections.

In the E. coli experiments, mean values of muscle cfu/g were 1.7 x 109 (SD 1.1 x 109) and 5.2 x 107 (SD 4.3 x 107), respectively (P < 0.001), for untreated and treated E. coli infections at t = 4 h. cfu in muscle at t = 4 h did not differ between treatment regimens or antibiotic dose levels (Figure 2). cfu counts in liver and spleen of antibiotic-treated mice were significantly correlated (r = 0.995, P = 0.0001). cfu in muscles of treated mice were negatively correlated with IL-6 concentrations (r = – 0.579, P = 0.007).


Figure 2
View larger version (10K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 2.. cfu of E. coli in thigh muscles at t = 4 h after continuous or bolus injection of ceftazidime at various concentrations.

 
Varying antibiotic type in these E. coli experiments revealed the following results. Figure 3 demonstrates higher cfu after continuous meropenem treatment, compared with ceftazidime continuously (1.5 x 108 versus 2.8 x 106, P = 0.009). In the Pseudomonas experiments, lower cfu counts were shown during continuous treatment with either ceftazidime or meropenem, especially at doses of 320 mg/kg (P = 0.083 for ceftazidime, P = 0.05 for meropenem) (Figure 4).


Figure 3
View larger version (13K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 3.. cfu of E. coli and plasma TNF-{alpha} and IL-6 after 4 h of treatment with ceftazidime or meropenem.

 


Figure 4
View larger version (29K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 4.. Plasma IL-6 during P. aeruginosa infection after 4 h of treatment with 160 or 320 mg/kg ceftazidime or meropenem. cfu in thighs after 320 mg/kg.

 
During untreated E. coli sepsis, cfu in thigh muscle rose from 1.64 x 107 at t = – 2 h to 4.8 x 109 at t = 4 h. cfu in liver and spleen rose from 0 at t = – 2 h to 2.1 x 106 and 3.4 x 104, respectively, at t = 4 h (Figure 5). cfu counts in muscle and liver (r = 0.707, P = 0.022), muscle and spleen (r = 0.869, P = 0.001) and spleen and liver (r = 0.867, P = 0.001) were significantly correlated.


Figure 5
View larger version (8K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 5.. Plasma cytokines and cfu during untreated E. coli sepsis.

 
Endotoxin

Regardless of morphology, plasma endotoxin concentrations showed no differences between mice with a continuous or a bolus regimen of 80 mg/kg (Table 1). In the experiment with 20 mg/kg ceftazidime, plasma endotoxin levels were significantly higher in both bolus and continuous treatment groups, compared with non-treated controls (P = 0.048 and P = 0.008, respectively). With regard to endotoxin levels, bolus treatment did not differ from continuous treatment.


View this table:
[in this window]
[in a new window]

  		Table 1.. Endotoxin, TNF-{alpha} and IL-6 in E. coli sepsis after 4 h of continuous or bolus treatment at various dose levels

 
Cytokines

Plasma TNF-{alpha} and IL-6 concentrations were below the detection limits at t = 0.1, 0.5, 1, 2 and 4 h in all saline experiments. Cytokine concentrations declined during the natural course of infection (Figure 5). Spleen cfu were negatively correlated with IL-6 concentrations (r = –0.818, P = 0.004). At all concentrations of ceftazidime, plasma IL-6 levels were higher during continuous infusions (mean 14 261 pg/mL, SD 7578), compared with bolus-treated (mean 3563 pg/mL, SD 3263) and untreated mice (mean 2810 pg/mL, SD 2295) in E. coli sepsis (Table 1). P values were <0.0001 for the difference between continuous-treated mice and both bolus-treated and untreated mice. Mean IL-6 concentrations at 10, 20, 40 and 80 mg/kg dose levels were 15 268, 18 418, 9728 and 13 840 pg/mL for continuous treatments and 2250, 3843, 3565 and 2072 pg/mL for bolus treatments, respectively. P values for the difference between both treatments at each dose level were 0.140, 0.037, 0.326 and 0.001, respectively. At 80 mg/kg, continuous dosing also caused higher TNF-{alpha} concentrations (596 pg/mL) than bolus dosing did (97 pg/mL) (P = 0.017). Plasma TNF-{alpha} and IL-6 were significantly correlated (r = 0.698, P = 0.006).

Monitoring plasma cytokine concentrations for 4 h of treatment with 80 mg/kg ceftazidime revealed that at t = 0, 1 and 2 h, cytokines did not differ, reaching significant higher concentrations of IL-6 in continuously treated animals, compared with bolus treatment at t = 4 h. Plasma TNF-{alpha} showed an early peak, with declining values after t = 0.5 or 1 h for bolus or continuous treatment, respectively (Figure 6). Plasma IL-6 levels demonstrated remarkable overlap with plasma ceftazidime concentrations in continuous- and bolus-treated mice.


Figure 6
View larger version (9K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 6.. Course of cytokine concentrations for 4 h of sepsis treatment with 80 mg/kg ceftazidime iv.

 
Varying antibiotic type in E. coli experiments revealed that IL-6 and TNF-{alpha} responses were significantly higher in continuously treated mice (P < 0.05 for ceftazidime and meropenem) (Figure 3). Cytokine responses after continuous and bolus antibiotic treatment of Pseudomonas infections were comparable to those in E. coli infections (Figure 4). Higher doses of continuously given ceftazidime caused more pronounced IL-6 [mean 3538 pg/mL (SD 2029) for 320 g/kg and 930 pg/mL (SD 499) for 160 mg/kg, P = 0.02] and TNF-{alpha} responses [mean 285 (SD 122) for 320 mg/kg and 48 (SD 15) for 160 mg/kg, P = 0.018]. Using higher concentrations of meropenem continuously (320 mg/kg), higher cytokine concentrations were induced compared with the lower dose of 160 mg/kg (for IL-6, P = 0.083; for TNF-{alpha}, P = 0.020). Bolus treatment with ceftazidime showed no dose-dependent IL-6 response in contrast to bolus meropenem (P = 0.003).

Pharmacokinetics of ceftazidime

After 80 mg/kg iv, plasma peak levels of ceftazidime were 486 (SD 69.5) mg/L and plasma half-life was 10.8 min. Mean plasma concentration of ceftazidime at t = 4 h during continuous infusion of 80 mg/kg in infected mice was 67.6 mg/L (SD 41.7). A comparison of infected and non-infected mice revealed 30.1% lower mean plasma concentrations of ceftazidime in infected mice (P = 0.24). One hour after bolus injections, mean plasma concentrations of ceftazidime were 3.0 (SD 0.849) and 8.75 mg/L (SD 1.202) for infected and non-infected mice, respectively (P = 0.031). At t = 2 h, plasma concentrations were undetectable in both infected and non-infected mice.


   Discussion
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
Transparency declarations
 References
 
The present study on the influence of antibiotic type and dosing regimen on bacterial morphology, endotoxin release and cytokine response in murine sepsis shows a remarkable dependence of cytokine response on antibiotic treatment regimen: continuous antibiotic dosing causes higher cytokine levels, even at very low antibiotic concentrations. For E. coli, we titrated continuous dosing levels of ceftazidime to a cut-off point of 80 mg/kg at which induction of filaments could be prevented. Bolus treatment with the same dose never did prevent filament formation, even with very high peak levels. To our knowledge, this is the first study demonstrating these relations in vivo between bacterial morphology and type of antibiotic dosing schedule.

Using continuous dosing schedules, cytokine release was positively correlated to antibiotic dose. Since cfu counts showed an inverse relationship with plasma IL-6 concentrations, suggesting more effective bacteriolysis, continuous infusion at higher concentrations of antibiotics might be more effective in sepsis treatment. This finding corresponds to our previous in vitro findings, showing increased bacteriolysis of E. coli or P. aeruginosa exposed to continuous concentrations of ceftazidime at high concentration levels of 40 x MIC and up.23

Efficacy of treatment with ß-lactam antibiotics has been shown to depend on the time above MIC.36–38 In our model, plasma IL-6 levels corresponded to antibiotic concentration levels and are negatively correlated with muscle cfu, suggesting that IL-6 is useful as a parameter of bacterial killing, especially in the initial phase of treatment, when bacterial counts are still high and variable. In murine sepsis, Remick et al.39 also demonstrated a clear relationship of initial high IL-6 concentrations and mortality, whereas Pallua and von Heimburg40 showed that IL-6 in contrast to TNF-{alpha} shows significant elevations after LPS exposure. Both studies correspond to our findings.

Our study has several limitations. Bacterial counts are not reliable as parameters of bacterial load when morphological changes are present, as one filament is counted only as one cfu, but represents a large source of endotoxin.23 We were unable to demonstrate a relation between circulating endotoxin levels and bacterial morphology or treatment regimen. Clearance from the circulation, adherence of endotoxin molecules to tissue and plasma components, as well as retainment in the bacterial cell wall at the infection site may explain this finding. Vianna et al.41 also showed increased IL-6 responses without differences in plasma endotoxin concentrations during treatment of murine sepsis. In addition, the amount of endotoxin cannot be measured at the site of infection by LAL assays, as shown in our previous studies.16 Thus, plasma endotoxin is an indirect parameter, not adequately representing sepsis severity. Direct assays using PCR may enable direct quantification of bacterial and endotoxin load.

We conclude that antibiotic dosing regimen in sepsis has important consequences for bacterial morphology and cytokine response, particularly in the initial hours of therapy. High and continuous dosing schedules appear most advantageous in terms of bacteriolysis and cytokine responses. Future studies should extend these findings to human infections and explore whether alterations in bacterial morphology and subsequent enhancement of endotoxin release are related to cytokine responses and whether alternative dosing schedules alter the clinical course of sepsis.


   Transparency declarations
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Transparency declarations
References
 
None to declare.


   Acknowledgements
 
Parts of this study were presented at ICAAC 2001 (Chicago, Abstract A-38914-ASM), ECCMID 2002 (Milan, Abstract 1476) and ECCMID 2005 (Copenhagen, Abstract 1804). We would like to thank Carla Wauters, pathologist, who performed the histological judgements and photography. We also thank Bianca Lemmers, Maikel van Riel, Monique Bakker and Margot van den Brink of the Central Animal Laboratory of the Radboud University Nijmegen for their biotechnical assistance and microbiological work. We acknowledge Nico Jacobs for performance of the endotoxin assays and Maaike den Hollander for performance of the HPLC. We gratefully thank Ineke Verschueren of the Laboratory of Internal Medicine of the University Medical Centre Nijmegen for the performance of the cytokine assays.


   References
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Transparency declarations
 References
 
1 Brun-Buisson C, Doyon F, Carlet J, et al. (1995) Incidence, risk factors, and outcome of severe sepsis and septic shock in adults. A multicenter prospective study in intensive care units. French ICU Group for Severe Sepsis. JAMA 274:968–74.[Abstract]

2 Guidet B, Aegerter B, Gauzit R, et al. (2005) Incidence and impact of organ dysfunctions associated with sepsis. Chest 127:942–51.[CrossRef][ISI][Medline]

3 Sundararajan V, Korman T, Macisaac C, et al. (2006) The microbiology and outcome of sepsis in Victoria, Australia. Epidemiol Infect 134:307–14.[Medline]

4 Spink WW, Braude AI, Casteneda MR, et al. (1948) Aureomycin therapy in human brucellosis due to Brucella melitensis. JAMA 138:1145–8.[Medline]

5 Reilly J, Compagnon A, Tournier P, et al. (1950) Les accidents du traitement des fievres typhoides pas la chloromycetine. Ann Med 51:597–629.

6 Hopkin BD. (1977) Too-rapid destruction of gram-negative organisms. Lancet ii:603–4.

7 Netea MG, van der Meer JW, van Deuren M, et al. (2003) Proinflammatory cytokines and sepsis syndrome: not enough or too much of a good thing? Trends Immunol.. 24:254–8.

8 Leeson M, Fujihara Y, Morrison DC. (1994) Evidence for lipopolyaccharide as the predominant proinflammatory mediator in supernatants of antibiotic-treated bacteria. Infect Immun 62:4975–80.[Abstract/Free Full Text]

9 Kirikae T, Nakano M, Morrison DC. (1997) Antibiotic-induced endotoxin release from bacteria and its clinical significance. Microbiol Immunol 41:285–94.[ISI][Medline]

10 Casey LC, Balk RA, Bone RC. (1993) Plasma cytokine and endotoxin levels correlate with survival in patients with the sepsis syndrome. Ann Intern Med 119:771–8.[Abstract/Free Full Text]

11 Gardlund B, Siolin J, Nilsson A, et al. (1995) Plasma levels of cytokines in primary septic shock in humans: correlation with disease severity. J Infect Dis 172:296–301.[ISI][Medline]

12 Morrison DC. (1998) Antibiotic-mediated release of endotoxin and the pathogenesis of gram-negative sepsis. Prog Clin Biol Res 397:199–207.[ISI][Medline]

13 Shenep JL, Flynn PM, Barrett FF, et al. (1988) Serial quantitation of endotoxemia and bacteremia during therapy for gram-negative bacterial sepsis. J Infect Dis 57:565–8.

14 Silverstein R, Wood JG, Xue Q, et al. (2000) Differential host inflammatory responses to viable versus antibiotic-killed bacteria in experimental microbial sepsis. Infect Immun 68:2301–8.[Abstract/Free Full Text]

15 Rodriguez-Caballero A, Garcia-Montero AC, Bueno C, et al. (2004) A new simple whole blood flow cytometry-based method for simultaneous identification of activated cells and quantitative evaluation of cytokines released during activation. Lab Invest 84:1387–98.[CrossRef][ISI]

16 Buijs J, Dofferhoff AS, Mouton JW, et al. (2006) Pathophysiology of in-vitro induced filaments, spheroplasts and rod-shaped bacteria in neutropenic mice. Clin Microbiol Infect 12:1105–11.[CrossRef][ISI][Medline]

17 Hurley JC. (1992) Antibiotic-induced release of endotoxin: a reappraisal. Clin Infect Dis 15:840–54.[ISI][Medline]

18 Jackson JJ and Kropp H. (1992) ß-Lactam antibiotic-induced release of free endotoxin: in vitro comparison of penicillin-binding protein (PBP) 2-specific imipenem and PBP-3-specific ceftazidime. J Infect Dis 165:1033–41.[ISI][Medline]

19 Dofferhoff AS, Nijland J, de Vries-Hospers H, et al. (1991) Effects of different types and combinations of antimicrobial agents on endotoxin release from gram-negative bacteria: an in vitro and in vivo study. Scand J Infect Dis 23:745–54.[ISI][Medline]

20 Roll JT, Young KM, Czuprynski CJ. (1990) Human rTNF-{alpha} augments anti-bacterial resistance in mice: potentation of its effects by recombinant human rIL-1{alpha}. Immunol 69:316–22.[ISI][Medline]

21 Nguyen-Disteche M, Fraipont C, Buddelmeijer N, et al. (1998) The structure and function of Escherichia coli penicillin-binding protein 3. Cell Mol Life Sci 54:309–16.[CrossRef][ISI][Medline]

22 Buijs J, Dofferhoff AS, Mouton JW, et al. (2003) A quantitative assessment of dose-dependent ß-lactam-induced filament-formation in Gram-negative bacteria. Clin Microbiol Infect 9:Suppl 1, 88.

23 Dofferhoff AS and Buijs J. (1996) The influence of antibiotic-induced filament formation on the release of endotoxin from Gram-negative bacteria. J Endotox Res 3:187–94.[Free Full Text]

24 Jackson JJ and Kropp H. (1996) Differences in mode of action of ß-lactam antibiotics, influence on morphology, LPS release and in vivo antibiotic efficacy. J Endotox Res 3:201–18.[Abstract/Free Full Text]

25 Dofferhoff AS, Esselink MT, de Vries-Hospers HG, et al. (1993) The release of endotoxin from antibiotic-treated Escherichia coli and the production of tumour necrosis factor by human monocytes. J Antimicrob Chemother 31:373–84.[Abstract/Free Full Text]

26 Gould IM and MacKenzie FM. (1997) The response of Enterobacteriaceae to ß-lactam antibiotics–‘round forms, filaments and the root of all evil. J Antimicrob Chemother 40:495–9.[Free Full Text]

27 Van't Wout JW, Linde I, Leyh CJ, et al. (1988) Contribution of granulocytes and moncytes to resistance against experimental disseminated Candida albicans infections. Eur J Clin Microbiol Infect Dis 7:736–41.[CrossRef][ISI][Medline]

28 Netea MG, van Tits LJ, Curfs JH, et al. (1999) Increased susceptibility of TNF-{alpha} lymphotoxin-{alpha} double knockout mice to systemic candidiasis through impaired recruitment of neutrophils and phagocytosis of Candida albicans. J Immunol. 163:1498–505.

29 Kunst MW and Mattie H. (1978) Cefazolin and cephradine: relationship between serum concentrations and tissue contents in mice. Infection 6:166–70.[CrossRef][ISI][Medline]

30 Mattie H, van Dokkum A, Brus-Weijer L, et al. (1990) Antibacterial activity of four cephalosporins in an experimental infection in relation to in vitro effect and pharmacokinetics. J Infect Dis 162:717–22.[ISI][Medline]

31 Andes D and Craig WA. (2002) Animal model pharmacokinetics and pharmacodynamics: a critical review. Int J Antimicrob Agents 19:261–8.[CrossRef][ISI][Medline]

32 Sumita Y and Fukasawa M. (1995) Potent activity of meropenem against Escherichia coli arising from its simultaneous binding to penicillin-binding proteins 2 and 3. J Antimicrob Chemother 36:53–64.[Abstract/Free Full Text]

33 Netea MG, Demacker PN, Kullberg BJ, et al. (1996) Low-density lipoprotein receptor-deficient mice are protected against lethal endotoxinemia and severe Gram-negative infections. J Clin Invest 97:1366–72.[ISI][Medline]

34 Mouton JW, Horrevorts AM, Mulder PG, et al. (1990) Pharmacokinetics of ceftazidime in serum and suction blister fluid during continuous and intermittent infusions in healthy volunteers. Antimicrob Agents Chemother 34:2307–11.[Abstract/Free Full Text]

35 Ulett GC, Hirst R, Bowden B, et al. (2003) A comparison of antibiotic regimens in the treatment of acute melioidosis in a mouse model. J Antimicrob Chemother 51:77–81.[Abstract/Free Full Text]

36 Craig WA and Ebert SC. (1992) Continuous infusion of ß-lactam antibiotics. Antimicrob Agents Chemother 36:2577–83.[Free Full Text]

37 Craig WA. (1995) Interrelationship between pharmacokinetics and pharmacodynamics in determining dosage regimens for broad-spectrum cephalosporins. Diagn Microbiol Infect Dis 22:89–96.[CrossRef][ISI][Medline]

38 Craig WA. (2001) Does the dose matter? Clin Infect Dis 33:3, S233–7.[CrossRef][ISI][Medline]

39 Remick DG, Bolgos G, Copeland S, et al. (2005) Role of interleukin-6 in mortality from and physiologic response to sepsis. Infect Immun 73:2751–7.[Abstract/Free Full Text]

40 Pallua N and von Heimburg D. (2003) Pathogenic role of interleukin-6 in the development of sepsis. Part I: study in a standardized contact burn murine model. Crit Care Med 31:1490–4.[CrossRef][ISI][Medline]

41 Vianna RC, Gomes RN, Bozza PT, et al. (2004) Antibiotic treatment in a murine model of sepsis: impact on cytokines and endotoxin releases. Shock 21:115–20.[CrossRef][ISI][Medline]


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?



This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
59/5/926    most recent
dkm073v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Buijs, J.
Right arrow Articles by van der Meer, J. W. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Buijs, J.
Right arrow Articles by van der Meer, J. W. M.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?