JAC Advance Access originally published online on April 14, 2006
Journal of Antimicrobial Chemotherapy 2006 57(6):1128-1133; doi:10.1093/jac/dkl140
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Azithromycin iv pharmacodynamic parameters predicting Streptococcus pneumoniae killing in epithelial lining fluid versus serum: an in vitro pharmacodynamic simulation
Microbiological Department, School of Medicine, Universidad Complutense Avda Complutense s/n, 28040 Madrid, Spain
*Corresponding author: Tel: +34-91-3941508; Fax: +34-91-3941511; E-mail: jprieto{at}med.ucm.es
Received 23 February 2005; returned 19 January 2006; revised 24 January 2006; accepted 22 March 2006
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and methodsResultsDiscussionTransparency declarationsReferences |
|---|
Objectives: To investigate the azithromycin pharmacodynamic parameters predicting bacterial killing in epithelial lining fluid (ELF) versus serum against macrolide-susceptible and -resistant Streptococcus pneumoniae isolates (with different resistance genotypes), through the simulation of concentrations achieved after a 500 mg intravenous (iv) once a day regimen.
Methods: An in vitro computer-controlled pharmacodynamic simulation of human azithromycin concentrations in serum and ELF was carried out, and colony counts were determined over 24 h. Four strains with MIC values (mg/L) of 0.5 [mef(A) and erm(B) negative], 2 [mef(A) positive and erm(B) negative], 8 [mef(A) positive and erm(B) negative] and 256 [mef(A) negative and erm(B) positive] were used.
Results: Significant (P < 0.05) azithromycin antibacterial activity versus antibiotic-free controls was found in serum and ELF against the susceptible and mef(A) positive strains, but not against the erm(B) positive strain. AUC0–24/MIC values around or higher than 25 were needed to achieve (time to 99.9% reduction of initial inocula of around 6 h) and maintain (24 h inocula reduction 
3 log10cfu/mL) bactericidal activity without regrowth. This was achieved only with the susceptible strain in serum, but also with the mef(A) positive strain exhibiting an MIC of 2 mg/L in ELF.
Conclusions: The results of this study support that the suggested breakpoint for susceptibility (
2 mg/L) may be adequate to predict S. pneumoniae eradication with ELF but not with serum concentrations obtained after a 500 mg iv once a day regimen.
Keywords: in vitro models , antipneumococcal bactericidal activity , resistance genotypes
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsDiscussionTransparency declarationsReferences |
|---|
Because of its basic structure, azithromycin (but not other macrolides) is actively taken up by white blood cells1 that chemotactically carry the azalide to the site of infection. After contact with encapsulated microorganisms that are difficult to phagocytose, such as Streptococcus pneumoniae, the drug accumulated in the lysosome migrates to the edge of the cell and is released by degranulation.2 This extensive uptake and slow release of azithromycin from tissues,3 and a very large volume of distribution, are responsible for its long half-life [ranging from 14.9 to 76.8 h after a single 500 mg intravenous (iv) dose or at steady-state].4 These data together with protein binding (50% at concentrations from 0.02 to 0.05 mg/L, but only 7% at concentrations
1 mg/L)5 indicate that high levels of the free-drug are available at infectious sites after a 500 mg iv regimen, achieving serum Cmax values of 
4.0 mg/L,4 serum AUC0–24 ranging from 9.6 to 17.7 mg·h/L,4,6 and epithelial lining fluid (ELF) Cmax and AUC0–24 values of 2.86 and 45.8 mg·h/L, respectively.7
It has been suggested that, in respiratory tract infections, concentrations at the site of infection, such as ELF, are more predictive of antimicrobial effects of azithromycin against extracellular pathogens such as S. pneumoniae8 and that an in vivo AUC/MIC ratio of at least 10 is needed for an in vivo bacteriological response against S. pneumoniae in the immunocompetent host.7 A previous report on serum and ELF simulations of free-drug concentrations (using a single protein binding value of 50%) after azithromycin oral regimens showed absence of killing at 24 h against azithromycin-resistant S. pneumoniae (MIC 2 mg/L; serum or ELF AUC0–24/MIC < 5), regardless of the resistance genotype [mef(A) or erm(B)].9
In the present study we have explored the azithromycin pharmacodynamic parameters predicting bacterial killing in ELF versus serum against macrolide-susceptible and -resistant S. pneumoniae (with different resistance genotypes), through the simulation of concentrations achieved after a 500 mg iv regimen.
|
Material and methods |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsDiscussionTransparency declarationsReferences |
|---|
Strains
Four S. pneumoniae clinical isolates from the Spanish Pneumococcal Infection Network (G03/103) were used. MICs were determined five times following CLSI (formerly NCCLS) recommendations10 and modal values were considered. mef(A) and erm(B) genotypes were determined using PCR as described previously.11 Table 1 shows serotype, susceptibility and resistance genotype of the four strains used in the study.
|
Antibiotic
Azithromycin laboratory reference standard was supplied by Pfizer (Groton, CT, USA).
In vitro kinetic model
A previously described dynamic model was used in the study.12 Briefly, the model consisted of two compartments, with hollow fibre capsule filters (FiberFlo; Minntech Corp., Minneapolis, MN, USA) as the second compartment (which included the peripheral compartment as the infection site). The model was designed to expose bacteria to changing antibiotic concentrations, without dilution of the bacterial inoculum together with the antibiotic. A computer-controlled syringe pump (402 Dilutor Dispenser; Gilson S.A, Villiers-le-Bel, France) was used to simulate the 1 h infusion of azithromycin iv until the corresponding serum or ELF Cmax was achieved in the central compartment (Cc). Peristaltic pumps (Masterflex; Cole-Parmer Instrument Co., Chicago, IL, USA) were used to continually replace the antibiotic-containing medium with fresh medium, and to circulate the medium between and within the two compartments. Both compartments were maintained at 37°C for the duration of the experiment.
Flow rates in the peristaltic pumps were synchronized all over the simulated period using the Win Lin software (Cole-Parmer Instrument Co.).
After 4 h, to mimic the concentrations achieved in human ELF from 4 to 24 h7 an additional reservoir, subcompartment 0 (S0), was used. The volume of S0 was calculated according to the volume of Cc (600 mL) and the flow rate for azithromycin.8
Todd–Hewitt broth (Difco laboratories, Detroit, MI, USA) supplemented with 0.5% yeast extract (Difco laboratories) was used as the medium throughout the pharmacodynamic simulation.
Pharmacokinetic simulations
Steady-state pharmacokinetic profiles in serum and ELF after an azithromycin 500 mg iv regimen were simulated over 24 h.7,13 Target Cmax in serum and ELF were 3.63 and 2.86 mg/L, respectively.7,13 Target AUC0–24 in serum was 13.6 mg·h/L, representing the mean of published data (9.6–17.7 mg·h/L).4,6 Target AUC0–24 in ELF was 45.8 mg·h/L.7 The target serum concentrations were obtained from azithromycin iv serum concentration-versus-time data.13 The profile was divided into three quasi-linear portions with apparent half-lives of 0.39 h (Ke; 1.8 h–1) from 1 to 2 h, of 1.73 h (Ke; 0.4 h–1) from 2 to 3 h (corresponding to the initial rapid decline in concentration and redistribution into the tissues of azithromycin) and of 10.5 h (Ke; 0.06 h–1) from 3 to 24 h (representing further distribution and elimination). The clearance used in the ELF profile was calculated using a half-life of 68 h (Ke; 0.01 h–1).
Pharmacokinetic analysis
Pharmacokinetic analysis was performed in triplicate, in bacteria-free simulations under the same experimental conditions. Samples (0.5 mL) from the peripheral compartment were obtained at 0, 1, 1.5, 2, 2.5, 3, 4, 6, 8 and 24 h. Concentrations were determined in triplicate by bioassay using Micrococcus luteus ATCC 9341 as the indicator organism.14 Standards were prepared in Todd–Hewitt broth supplemented with 0.5% yeast extract. The lower limit of detection was 0.06 mg/L. The correlation coefficient was >0.99 and intra-day and inter-day coefficients of variation were 9.8% and 12.5%, respectively, at a level of 0.75 mg/L.
Azithromycin concentrations were analysed by a non-compartmental approach using WinNonlin Professional program (Pharsight, Mountain View, CA, USA). Cmax was obtained directly from the observed data. The area under the concentration–time curve from 0 to 24 h (AUC0–24) was calculated by the trapezoidal rule.
Experiments
Prior to each experiment, 1–2 colonies from a fresh passage on Mueller–Hinton agar supplemented with cations and 5% lysed sheep blood were inoculated in 50 mL of Todd–Hewitt broth supplemented with 0.5% yeast extract. The suspension was allowed to grow to a density of 107 cfu/mL, as measured using a UV-spectrophotometer (Hitachi U-1100). Of this inoculum, 30 mL was introduced into the peripheral compartment of the in vitro model 1 h prior to each simulation process to allow the microorganism to adapt to the medium. All initial inocula were in the range 107 to 3 x 107 cfu/mL.
Samples (0.5 mL) from the peripheral compartment were collected at 0, 2, 4, 6, 8, 10 and 24 h. Each sample was serially diluted 10-fold in 0.9% sodium chloride for bacterial counting in supplemented Mueller–Hinton agar with 5% sheep blood which was incubated at 37°C 5% CO2 for 24 h. The limit of detection was 20 cfu/mL.
Antibiotic-free simulations with each strain were used as controls.
All experiments were performed in triplicate.
Statistical analysis
Mean values of cfu/mL were calculated with the three values of colony counts at each time during the 24 h simulation. Reductions in initial inocula at 24 h were calculated by the difference between log10 cfu/mL initial inocula and log10 cfu/mL at 24 h. Percentages of reduction of initial inocula at the different sample times were calculated. The time to achieve 99.9% reduction (T99.9%) was determined using linear regression.
Antibacterial activity was also studied using the area under the killing curve (AUKC; cfu x h/mL) as a measure of global killing over 24 h. The bacterial growth in antibiotic-free simulations was used as control (AUKCK). Differences between AUKCK and AUKC obtained in simulations with azithromycin (AUKCAZ) were determined. Means ± standard deviations were calculated.
Comparisons of 24 h reductions in azithromycin simulations versus controls were analysed using the t-test. Inter-strain differences in capability of reduction of control growth curves over 24 h (AUKCK–AUKCAZ) and 24 h inocula reductions were analysed by the ANOVA. When the ANOVA P value was significant, contrast between groups was made by the use of the Tukey–Kramer test to adjust the type I experimental error.
|
Results |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsDiscussionTransparency declarationsReferences |
|---|
Experimental Tmax and Cmax (mg/L) in the pharmacodynamic simulations were 1 h and 3.59 ± 0.04 for serum and 24 h and 2.78 ± 0.05 for ELF. AUC0–24 (mg·h/mL) was 13.47 ± 0.85 and 49.09 ± 0.84 for serum and ELF, respectively. Serum experimental AUC0–24 was <1% lower than the target area, and ELF experimental AUC0–24 was 7% higher than the target area.
Figure 1 and Tables 2 and 3 show the antipneumococcal activity of the serum and ELF azithromycin profiles. In both cases, when exploring antibacterial activity as inocula reduction at 24 h, azithromycin exhibited significant (P < 0.05) activity versus antibiotic-free controls against the susceptible strain or those strains carrying the mef(A) gene. No differences (P = 0.77 and 0.46 for serum and ELF, respectively) versus antibiotic-free control were found with azithromycin against strain 4, harbouring the erm(B) gene. When exploring antibacterial activity through differences between AUKCK and AUKCAZ, significant (P < 0.05) differences were found in serum between all strains. These significant differences were maintained between all strains in the case of ELF, except in the case of strain 3 (MIC = 8 mg/L) versus strain 4 (MIC = 256 mg/L) where statistical differences disappeared.
|
|
|
As shown in Table 2, in serum, no bactericidal activity was obtained against strains 3 and 4 (MIC
8 mg/L), but bactericidal activity (T99.9% = 6.6 and 5.5 h) was obtained against strains 1 and 2. However, in the case of the susceptible strain (strain 1), the bactericidal activity (3 log10 reduction) obtained at 6.6 h was maintained and even increased at 24 h (5.6 log10 reduction), while in the case of strain 2 (exhibiting low resistance; MIC = 2 mg/L) the bactericidal activity achieved at 5.5 h was not maintained and regrowth occurred, the log10 reduction being 2.6 at 24 h.
In contrast, in ELF, azithromycin bactericidal activity was obtained against strains 1 and 2 at 6.9 and 6.3 h, respectively, and maintained and even increased at 24 h with 5.7 and 4.3 log10 reduction, respectively. Again, no bactericidal activity was obtained against strains with MIC 8 mg/L.
From the pharmacodynamic point of view, AUC/MIC values 24.5 related with attainment (at around 6 h) and maintenance over 24 h of bactericidal activity (3 log10 reduction), as occurred with strains 1 and 2 in ELF and with strain 1 in serum. Lower values were related with absence of bactericidal activity (strains 3 and 4) or with absence of maintenance over 24 h of the bactericidal activity obtained (regrowth with strain 2 in serum: T99.9% = 5.5 h, with a 24 h reduction of only 2.6 log10 cfu/mL).
|
Discussion |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsDiscussionTransparency declarationsReferences |
|---|
Macrolide resistance in S. pneumoniae in Spain is as high as 34.5%, most of the isolates (89.9%) being erm(B) positive,15 while in the USA, resistance prevalence is 31% and most of them (70.9%) belong to the mef(A) genotype.16 While the erm(B) genotype confers high-level resistance with MIC values >64 mg/L (that logically cannot be overcome by increasing antibiotic concentrations), the efflux resistance phenotype [mef(A) genotype] confers a much lower level of resistance. In the latter case, the question that arises is whether this low-level resistance can be overcome in body sites where the drug is concentrated.
Azithromycin is concentrated in ELF, with an AUC0–24 much higher than the one in serum. In the present study experimental ELF AUC0–24 was 3.64-fold higher than that in serum. Therefore if an AUC0–24/MIC of 30 is needed for bacterial eradication,17 this value would be more easily achieved by azithromycin in ELF than in serum against strains carrying the mef(A) gene, and with low-level resistance. Moreover, it has been suggested that concentrations at the site of infection are more predictive of antimicrobial efficacy of azithromycin against S. pneumoniae,8 when an AUC/MIC ratio of at least 10 is achieved.7 It should be considered that azithromycin antibacterial activity in ELF cannot be greatly modified by protein binding because the percentage is negligible (7%) when concentrations are more than 1 mg/L, and the ELF concentrations used in this study were higher than this value, as those determined in healthy adults.7 These facts may explain why reports on azithromycin failure in community-acquired respiratory tract infections are uncommon, although available,18,19 in areas where efflux is the prevalent phenotype of resistance.
In agreement with the previously reported value of AUC0–24/MIC needed for bacterial eradication, values around or more than 25 were associated with maintained bactericidal activity over 24 h, while regrowth or absence of bactericidal activity occurred with values <10. Although AUC0–24/MIC is the main pharmacodynamic parameter predicting efficacy for azithromycin (i.e. bactericidal activity and bacterial eradication), the pharmacokinetic profile should also be taken into account when examining the effects of similar AUC0–24/MIC against different strains.
In the present study, this can be seen when comparing the effects of the similar AUC0–24/MIC (6.7 and 6.12) obtained in serum with respect to strain 2, and in ELF with respect to strain 3 (both serotype 23). In the first case (serum simulation), bactericidal activity (3 log10 reduction; 99.9%) was obtained at 5.5 h against a strain with an MIC of 2 mg/L with azithromycin Cmax levels of 3.59 mg/L achieved 1 h after the start of the experiment. In the second case (ELF simulation) bactericidal activity (3 log10 reduction; 99.9%) was not obtained against a strain with an MIC of 8 mg/L with azithromycin Cmax levels of 2.78 mg/L achieved 24 h after the start of the experiment, because levels were never supra-inhibitory. In either of the two cases, eradication and bactericidal activity were not obtained at 24 h: mean initial inocula reduction (difference between log10 cfu/mL of initial inocula and log10 cfu/mL at 24 h) was only 2.6 log and 1.3 log, respectively, due to low AUC0–24/MIC values. Logically AUC0–24/MIC values <1 (with the highly resistant strain) were only associated with growth of initial inocula without differences versus controls.
An AUC0–24/MIC of 26.9 against the susceptible strain (MIC of 0.5 mg/L) seems adequate for the sustained bactericidal activity in serum. Against the other strains harbouring resistance genetic markers, azithromycin did not achieve experimental sustained bactericidal activity in serum because AUC0–24/MIC values were <10 and thus are considered resistant (2 mg/L) with the current CLSI breakpoints.10 This is not the case with azithromycin in ELF where, with the susceptible strain and the strain with an MIC of 2 mg/L, sustained bactericidal activity was obtained because AUC0–24/MIC values were around or higher than 25. From the bacterial eradication and pharmacodynamic perspective, the results of this study suggest that resistant strains with MICs of 2 mg/L, and thus intermediate-resistant strains with MICs of 1 mg/L, can be considered susceptible to the concentrations achieved in ELF after a 500 mg iv regimen. According to this, previous suggestions on breakpoint for susceptibility 2 mg/L based on the expected azithromycin tissue concentrations3 may be adequate in the case of ELF. These suggestions should not be extrapolated to bacteraemic pneumonia where failures have been described after macrolide treatment in patients infected by M phenotype strains exhibiting erythromycin MICs as low as 4 mg/L.18 This is consistent with the low AUC0–24/MIC and regrowth obtained in our serum experiments with the strain with an MIC of 2 mg/L.
|
Transparency declarations |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsDiscussionTransparency declarationsReferences |
|---|
None to declare.
|
Acknowledgements |
|---|
We thank A. Fenoll (Instituto de Salud Carlos III, Majadahonda, Madrid) for strain serotyping and for the critical review of the manuscript. This study was supported in part by Pfizer S.A. (Madrid, Spain) and by the Spanish Pneumococcal Infection Study Network (G03/103), Ministerio de Sanidad y Consumo. O. E. received scholarships from this Network.
Participants from the Spanish Pneumococcal Infection Study Network (G03/103) were as follows: Ernesto Garcia (Centro de Investigaciones Biológicas, Madrid, Spain); Julio Casal, Asuncion Fenoll and Adela G. de la Campa (Centro Nacional de Microbiología, Instituto de Salud Carlos III, Madrid, Spain); Emilio Bouza (Hospital Gregorio Marañon, Madrid, Spain); Fernando Baquero (Hospital Ramón y Cajal, Madrid, Spain); Francisco Soriano and José Prieto (Fundación Jiménez Díaz and Facultad de Medicina de la Universidad Complutense, Madrid, Spain); Roman Pallares (general coordinator) and Josefina Liñares (Hospital Universitari de Bellvitge, Barcelona, Spain); Javier Garau and Javier Martinez de la Casa (Hospital Mutua de Terrassa, Barcelona, Spain); Cristina Latorre (Hospital Sant Joan de Deu, Barcelona, Spain); Emilio Perez-Trallero (Hospital Donostia, San Sebastian, Spain); Juan Garcia de Lomas (Hospital Clinico, Valencia, Spain); and Ana Fleites (Hospital Central de Asturias, Oviedo, Spain).
|
References |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsDiscussionTransparency declarationsReferences |
|---|
1 Amsden GW. (1996) Erythromycin, clarithromycin, and azithromycin: are the differences real? Clin Ther 18:56–72.[ISI][Medline]
2
Glaude RP, Bright CM, Isaacson RE, et al. (1989) In vitro and in vivo uptake of azithromycin (CP-62,993) by phagocytic cells: possible mechanism of delivery and release at sites of infection. Antimicrob Agents Chemother 33:277–82.
3 Luke DR, Foulds G, Cohen SF, et al. (1996) Safety, toleration, and pharmacokinetics of intravenous azithromycin. Antimicrob Agents Chemother 40:2577–81.[Abstract]
4
Chiu LM, Menhinick AM, Johnson PW, et al. (2002) Pharmacokinetics of intravenous azithromycin and ceftriaxone when administered alone and concurrently to healthy volunteers. J Antimicrob Chemother 50:1075–9.
5 Lalak NJ and Morris DL. (1993) Azithromycin clinical pharmacokinetics. Clin Pharmcokinet 25:370–4.[Medline]
6 Garey KW and Amsden GW. (1999) Intravenous azithromycin. Ann Pharmacother 33:218–28.[Abstract]
7
Rodvold KA, Danziger LH, Gotfried MH. (2003) Steady-state plasma and bronchopulmonary concentrations of intravenous levofloxacin and azithromycin in healthy adults. Antimicrob Agents Chemother 47:2450–7.
8
Firsov AA, Zinner SH, Vostrov SN, et al. (2002) Comparative pharmacodynamics of azithromycin and roxithromycin with S. pyogenes and S. pneumoniae in a model that simulates in vitro pharmacokinetics in human tonsils. J Antimicrob Chemother 49:113–9.
9
Zhanel GG, DeCorby M, Noreddin A, et al. (2003) Pharmacodynamic activity of azithromycin against macrolide-susceptible and -resistant Streptococcus pneumoniae simulating clinically achievable free serum, epithelial lining fluid and middle ear fluid concentrations. J Antimicrob Chemother 52:83–8.
10 Performance Standards for Antimicrobial Susceptibility Testing: Fifteenth Informational Supplement. CLSI Document M100-S15 Clinical and Laboratory Standards Institute. CLSI, Wayne, PA, USA, 2005.
11
Hoban DJ, Wierzbowski AK, Nichol K, et al. (2001) Macrolide-resistant Streptococcus pneumoniae in Canada during 1998–1999: prevalence of mef(A) and erm(B) and susceptibilities to ketolides. Antimicrob Agents Chemother 45:2147–50.
12
Sevillano D, Calvo A, Giménez MJ, et al. (2004) Bactericidal activity of amoxicillin against non-susceptible Streptococcus pneumoniae in an in vitro pharmacodynamic model simulating the concentrations obtained with the 2000/125 mg sustained-release co-amoxiclav formulation. J Antimicrob Chemother 54:1148–51.
13 NDA 50-733/S-005. Zithromax. http://www.fda.gov/cder/foi/label/2001/50733s5lbl.pdf (23 January 2006, date last accessed).
14
Danesi R, Lupetti A, Barbara C, et al. (2003) Comparative distribution of azithromycin in lung tissue of patients given oral daily doses of 500 and 1000 mg. J Antimicrob Chemother 51:939–45.
15
Pérez-Trallero E, García-de-la-Fuente C, García-Rey C, et al. (2005) Geographical and ecological analysis of resistance, coresistance and coupled resistance to antimicrobials in and between respiratory pathogenic bacteria in Spain. Antimicrob Agents Chemother 49:1965–72.
16
Brown SD, Farrell DJ, Morrissey I. (2004) Prevalence and molecular analysis of macrolide and fluoroquinolone resistance among isolates of Streptococcus pneumoniae collected during the 2000–2001 PROTEKT US Study. J Clin Microbiol 42:4980–7.
17 Andes D. (2001) Pharmacokinetic and pharmacodynamic properties of antimicrobials in the therapy of respiratory tract infections. Curr Opin Infect Dis 14:165–72.[CrossRef][ISI][Medline]
18 Lonks JR, Garau J, Gómez L, et al. (2002) Failure of macrolide antibiotic treatment in patients with bacteremia due to erythromycin-resistant Streptococcus pneumoniae. Clin Infect Dis 35:556–64.[CrossRef][ISI][Medline]
19 Kelley MA, Weber DJ, Gilligan P, et al. (2000) Breakthrough pneumococcal bacteremia in patients being treated with azithromycin and clarithromycin. Clin Infect Dis 31:1008–11.[CrossRef][ISI][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||