Delivering Antibacterials to the Lungs

Delivering Antibacterials to the LungsConsiderations for Optimizing Outcomes

Mario Cazzola,1 Francesco Blasi,2 Claudio Terzano,3 Maria G. Matera4 and Serafino A. Marsico5

1 Department of Respiratory Medicine, Unit of Pneumology and Allergology, "A. Cardarelli" Hospital, Naples, Italy2 University of Milan, Institute of Care and Research, Polyclinic Hospital, Institute of Respiratory Medicine, Milan, Italy3 Department of Cardiovascular and Respiratory Sciences, University "La Sapienza", Rome, Italy4 Department of Experimental Medicine, Unit of Pharmacology, Second University of Naples, Naples, Italy5 Department of Cardiothoracic and Respiratory Sciences, Unit of Respiratory Medicine, Second University of Naples,

Naples, Italy

Contents Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2611. Concentrations Found in Lung Sites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2622. Therapeutic Significance of Pulmonary Disposition of Antibacterials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2633. Pharmacodynamic Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 264

3.1 Time-Dependent Killing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2653.2 Concentration-Dependent Killing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2663.3 Pharmacodynamic Modelling for Macrolides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267

4. Interrelationship Between Pharmacokinetics and Pharmacodynamics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2675. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 268

Abstract An important determinant of clinical outcome of a lower respiratory tract infection may be sterilization ofthe infected lung, which is also dependent on sustained antibacterial concentrations achieved in the lung. Forthis reason, recently there has been increased interest in measuring the concentration of antimicrobial agents atdifferent potential sites of infection in the lung. Levels of antibacterials are now measured in bronchial mucosa,epithelial lining fluid (ELF) and alveolar macrophages, as well as in sputum. Penicillins and cephalosporinsreach only marginal concentrations in bronchial secretions, whereas fluoroquinolones and macrolides have beenshown to achieve high concentrations. The extent of penetration of different antibacterials into the bronchialmucosa is relatively high. This is also true for β-lactams, although their tissue concentrations never reach bloodconcentrations. Antibacterials penetrate less into the ELF than into the bronchial mucosa, but fluoroquinolonesappear to concentrate more into alveolar lavage than into bronchial mucosa.

Pulmonary pharmacokinetics is a very useful tool for describing how drugs behave in the human lung, butit does not promote an understanding of the pharmacological effects of a drug. More important, instead, is thecorrelation between pulmonary disposition of the drug and its minimum inhibitory concentration (MIC) valuesfor the infectious agent. The addition of bacteriological characteristics to in vivo pharmacokinetic studies hastriggered a ‘pharmacodynamic approach’. Pharmacodynamic parameters integrate the microbiological activityand pharmacokinetics of an anti-infective drug by focusing on its biological effects, particularly growth inhi-bition and killing of pathogens.

Drugs that penetrate well and remain for long periods at the pulmonary site of infection often induce thera-peutic responses greater than expected on the basis of in vitro data. However, although the determination ofantibacterial concentrations at the site of infection in the lung has been suggested to be important in predictingthe therapeutic efficacy of antimicrobial treatment during bacterial infections of the lower respiratory tract,some studies have demonstrated that pulmonary bacterial clearance is correlated more closely to concentrationsin the serum than to those in the lung homogenates, probably because they better reflect antibacterial concenta-tion in the interstitial fluid.

THERAPY IN PRACTICE Am J Respir Med 2002; 1 (4): 261-2721175-6365/02/0004-0261/$25.00/0

© Adis International Limited. All rights reserved.

Several animal studies suggest that the pharmacokinetic pa-rameters of an antibacterial in lung tissue could be important intreatment outcome of experimental pneumonia.[1,2] From theseexperimental data, it appears that the pharmacokinetics of an anti-bacterial in pulmonary tissues can also be significant in treatmentoutcome of human pulmonary infections.

For this reason, recently there has been increased interest inmeasuring the concentration of antimicrobial agents at differentpotential sites of infection in the lung.[3] The site where the infec-tion develops inevitably influences the clinical presentation. His-tological studies suggest that both sputum and bronchial mucosamay be the sites of infection in acute exacerbations of chronicbronchitis and bronchiectasis. Epithelial lining fluid (ELF) andalveolar macrophages have been advocated as important infec-tion sites for common extracellular and intracellular pathogens,respectively, and may be better predictors of clinical efficacy inpneumonia.[4-6]

Initially, whole lung concentrations of the antibacterial werereported, but this methodology has been refined. In fact, duringthe last few years innovative methods for the pharmacologicalevaluation of antibacterial agents have been developed and tech-niques have been improved, although the in vivo determinationof concentrations at different sites of infection is plagued withcertain methodological difficulty[7] and needs further refinementand standardization. The relative ease with which respiratory tis-sues and fluids can be collected allows the obtaining of moreprecise information on the deposition of drugs within the respi-ratory tract, even considering ethical aspects. Levels are nowmeasured in bronchial mucosa, ELF and alveolar macrophages,as well as in sputum. Nevertheless, there are difficulties in eval-uating the real value of such data because the concept of antibac-terial ‘tissue pharmacokinetics’ is still controversial and the clin-ical significance of tissue concentrations has often beendebated.[8] Moreover, we still do not know if inflammation affectsantibacterial concentrations in the lung and what is the effect ofdisease on the penetration of the antibacterial.

1. Concentrations Found in Lung Sites

The concentration of antibacterials in the sputum has beenstudied extensively, but few antimicrobials penetrate well intobronchial secretions. Penicillins and cephalosporins reach onlymarginal concentrations in bronchial secretions; consequently,their sputum or bronchial secretion-to-simultaneous serum ratiosvary between 2 and 25%.[9-14] However, the percentages for theextent of penetration of piperacillin and tazobactam, as definedby the bronchial secretion-to-serum area under the curve ratio,have been calculated to be 35.7% and 78.42%, respectively.[15]

Apparently, there is wide variation in penetration into sputumeven for drugs that are similar. For example, serum concen-trations of ampicillin and amoxicillin following simultaneousintravenous administration are almost identical, but sputumamoxicillin concentrations are significantly higher than those ofampicillin.[12] Moreover, bacampicillin exhibits higher serumand bronchial concentrations than oral ampicillin.[16] Fluoro-quinolones have been shown to achieve high concentrations inbronchial secretions, 0.8 to 4.0 times greater than those in thebloodstream.[17-22] These high concentrations have been attrib-uted to an active process of transport across the airway epithe-lium. The concentrations of aminoglycosides and tetracyclines inthe bronchial secretions are approximately in the ratios of 6 to60% of the serum concentration.[23-27] Interestingly, antibacterialconcentrations in bronchial secretions in patients with pneumoniatreated with amikacin 7.5 mg/kg twice daily, ranged from 3 to 4mg/L whereas in patients with pneumonia treated with amikacin15 mg/kg once daily, the antibacterial concentrations in bronchialsecretions were more than 2-fold higher, being >8 mg/L at 12hours.[28] The serum-to-bronchial secretion concentration ratioremained relatively constant at about 30%. Macrolides exhibitvariable pulmonary penetration between 5 and 500%.[29-31] How-ever, penetration ratios >500% have been reported with azithro-mycin[32] and dirithromycin.[31,33] Unfortunately, sputum is nowconsidered an unsuitable fluid for pharmacokinetic studies, as itslack of homogeneity, dilution by saliva, pooling within the respi-ratory tract and, in addition, the instability of some antimicrobialagents in sputum lead to methodological and interpretationalproblems.[34]

The determination of antibacterial penetration into bronchialmucosa has been described as a more reliable indicator of pulmo-nary penetration than sputum concentration.[35] Unfortunately,the data in this field are limited, notwithstanding the fact that themethods for determining total concentrations in the samples ob-tained by means of bronchial biopsies during bronchoscopy arewell standardized and relatively simple to carry out.[36] The pen-etration of different antibacterials is relatively high (table I). Thisis also true for β-lactam agents, although their tissue concentra-tions never reach blood concentrations. In effect, β-lactams, al-though penetrating poorly into cells, accumulate in the bronchialmucosa to 35 to 60% of the concentration attained in the se-rum.[37-40] The penetration of fluoroquinolones and macrolides ishigher; for example, the levels of fluoroquinolones in the bron-chial mucosa are 1.5- to 3-fold greater than those in the blood-stream,[21,41-49] whereas the concentrations of macrolides in thebronchial mucosa are 1.7 to 8 times greater than those in theblood.[31-33,50-52] The behavior of azithromycin and dirithromycinis of particular interest because these drugs persist in the bron-

262 Cazzola et al.

© Adis International Limited. All rights reserved. Am J Respir Med 2002; 1 (4)

chial mucosa at high concentrations up to 72 to 96 hours after thelast administration, even in the presence of low or non-determin-able blood concentrations.[32,33] Drugs that show a high intracel-lular accumulation in vitro present higher concentrations in thebronchial mucosa. Many antibacterials are capable of accumulat-ing in cells, with the exception of β-lactam agents and aminogly-cosides. All data show that the concentration of hydrophilicantibacterials, such as β-lactams and aminoglycosides, in extra-cellular pulmonary fluids reflects serum concentrations fairlyfaithfully, whereas liposoluble antibacterials that are taken up bycells, such as fluoroquinolones, tetracyclines and macrolides,reach much higher concentrations in tissues than in blood.[4] Theprincipal criticism of the clinical significance of measurement ofantibacterial penetration into the bronchial mucosa is that it doesnot allow us to differentiate between intracellular and extracellu-lar components. This information is important because the abilityof antibacterials to enter and accumulate in cells influences, aspreviously stated, the levels of drug measured; moreover, thepenetration of different antibacterials into cells differs accordingto their class.[53]

As a result of improvements in the technique of broncho-alveolar lavage (BAL), it has been possible to obtain samplesfrom two distal sites in the lung, namely ELF and alveolar macro-phages. ELF and alveolar macrophages have been advocated asimportant sites of infection for common extracellular and intra-cellular pathogens, respectively.[1] The movement of antibacteri-als across the alveolar-capillary membrane is extremely difficultbecause the pulmonary capillary endothelium is non-fenestratedand the alveolar membranes are relatively impermeable becauseof the presence of many tight junctions (zonulae occludentes).[54]

As a consequence, sometimes antibacterials penetrate less intoELF than into bronchial mucosa[55] (table I). However, fluoro-quinolones appear to concentrate more into alveolar lavage thaninto bronchial mucosa,[41-49] probably because they have addi-tional mechanisms that allow the crossing of the membrane com-pared with simple passive diffusion, as is the case with β-lactams.[55] The hydrophilic nature of β-lactam agents leads to poorpenetration into the relatively impermeable alveolar space andthe ELF, reaching levels that are only 12 to 50% of serum concen-tration.[14,56-59] Macrolides have been found to concentrate intoELF,[31-33,60,61] with azithromycin showing a 7-fold and clar-

ithromycin a 5.7-fold increase compared with serum levels. Con-centration in ELF of the ketolide telithromycin, a semisyntheticderivative of the 14-membered ring macrolides, is 8 times higherthan that in serum.[62] It has been demonstrated that high peakserum concentrations of tobramycin are necessary to obtainmicrobiologically active concentrations at the alveolar level. Inpatients with pneumonia, the ratio of ELF to serum concentrationat peak serum time was 0.30,[63] whereas that for netilmicin was0.46.[27]

Although neutrophils are the prevalent cells in bacterial in-fections, in vitro studies show that alveolar macrophages concen-trate antibacterials in a manner similar to neutrophils. Conse-quently, the determination of concentration of antibacterials inmacrophages is extremely important because it indicates the pen-etration of these agents into phagocytic cells, which allows themto be useful in the treatment of infections caused by intracellularpathogens such as Legionella pneumophilaand Chlamydia pneu-moniae.[3] β-Lactam agents diffuse but do not accumulate intophagocytes, probably because of their acidic character, with theexception of clavulanic acid, which is detectable in macro-phages.[57] In any case, their activity at this site is negligible be-cause of the low pH. Aminoglycosides are too polar to pass acrossmembranes and are therefore taken up only slowly by endocyto-sis. Lincosamides, macrolides and fluoroquinolones all accumu-late in phagocytes.[64] Concentrations of azithromycin are up to23 times and clarithromycin about 70 times higher than those inserum,[61] whereas levofloxacin shows an 8-fold,[48] gatifloxacin35-fold[47] and moxifloxacin a 50-fold increase[46] compared withserum levels.

2. Therapeutic Significance of PulmonaryDisposition of Antibacterials

The therapeutic significance of antibacterial concentrationsis debatable. However, some studies have suggested that efficientantimicrobial penetration into potential sites of pulmonary infec-tion and its protracted permanence in active form are advanta-geous.[65] This is particularly true when the infection has notspread beyond the lung and there are no existing barriers to ade-quate lung penetration, such as lung abscess or necrotizing pneu-monia. The results of clinical trials of the intravenous azalideazithromycin in patients with community-acquired pneumonia

Table I. Comparison of penetration of fluoroquinolones, β-lactams and macrolides into different pulmonary sites

Class of antibacterial Bronchial mucosa Epithelial lining fluid Alveolar macrophages

Fluoroquinolones 1.5-3 : 1 3-12 : 1 10-34 : 1β-Lactam agents 0.35-0.6 : 1 0.25 : 1 0.1 : 1Macrolides 1.7-10 : 1 10+ : 1 20+ : 1

Delivering Antibacterials to the Lungs 263


(CAP) are paradigmatic of this assumption. These trials haverevealed that clinical outcomes were similar in bacteremic patientsand in a control group treated with cefuroxime with or withouterythromycin.[66] Since serum concentrations of azithromycinfall rapidly but antibacterial concentrations in the lung paren-chyma and phagocytes rise by several-fold, these results supportthe concept that pulmonary antibacterial levels may be the mostsignificant pharmacokinetic factor in the cure of patients withCAP.

Some other studies[67,68] have documented that drugs thatpenetrate well and remain for long periods at the pulmonary siteof infection often induce therapeutic responses greater than thatexpected on the basis of in vitro data. This finding seems to beindependent of the class of antibacterials used.

Thus, in patients with acute exacerbations of chronic bron-chitis, a 5-day treatment with dirithromycin 500 mg/day inducedeffective concentrations at different pulmonary sites of infectionup to 96 hours after the last dose. Symptom remission progressedtogether with the trend of pulmonary antibacterial concentra-tions, independently of the length of treatment[67] (figure 1). In-terestingly, notwithstanding its poor in vitro activity againstHaemophilus influenzae, dirithromycin was effective against thismicro-organism in vivo. This discrepancy could be due to the factthat the minimum inhibitory concentrations (MICs) of dirithro-mycin in the presence of serum in vivo are 5-fold lower than in

vitro in artificial media,[69] but the persistence of antimicrobialactivity in the tissues may also be inhibitory for this pathogen.[70]

In addition, the eradication of nosocomial pathogens such asEnterobacteriaceae or Pseudomonas aeruginosa correlates withthe existence of relatively high pulmonary concentrations ofceftriaxone (57.4 ± 13.3 mg/kg) or imipenem (6.6 mg/kg).[68]

Fluoroquinolones, which concentrate in pulmonary tissuesand fluids, reach levels that are sufficient to overcome a goodpercentage of cases of infection caused by Streptococcus pneu-moniae with high MICs toward these antibacterials, providingfurther evidence of the importance of effective pulmonary dispo-sition of antibacterials. For example, although the serum concen-tration of ciprofloxacin (1.19 ± 0.16 mg/L) is below the MIC90

(concentration required to inhibit 90% of pathogens) for S. pneu-moniae (2.0 mg/L), the concentrations reached in ELF are ap-proximately 3 mg/L.[41] Similarly, the mean concentration ofpefloxacin is 88.2 ± 10 mg/L in the ELF, whereas the mean serumconcentration was 6.67 ± 0.47 mg/L,[71] which is below the MIC90

for S. pneumoniae (8.0 mg/L). However, failures of ciprofloxacinand pefloxacin in the treatment of pneumococcal pneumoniahave been described notwithstanding the relevance of their accu-mulation in the lower respiratory tract. Baldwin et al.[41] havesuggested that subtherapeutic intrapulmonary concentrationsmight be responsible for these failures.

3. Pharmacodynamic Approach

The above-mentioned studies seem to indicate that drugswith higher intrapulmonary concentrations have greater clinicalefficacy. Nevertheless, the presence of a relationship betweenantibacterial levels at the site of infection and their clinical effi-cacy in the lung has not been clearly demonstrated because ofnumerous methodological difficulties.[4]

In particular, it has been suggested that experimentally de-termined ‘total tissue concentrations’ are not good indicators ofactivity, since they represent average values including un-specifically bound drug and not the concentrations actually pres-ent at the site of action.[8] For the same reason, the concept of‘tissue partition coefficients’ is inadequate, since it implies ho-mogeneous tissue concentrations. It is the aqueous unbound con-centration at the site of infection in the tissue that is most relevantfor the magnitude of antibiosis.[72] Thus, if overall concentrationsare measured, effect-site concentrations of drugs that equilibrateexclusively with the extracellular space, such as β-lactam agents,may be underestimated.[73] This, in turn, also leads to an overes-timation of effect-site concentrations of drugs that accumulateintracellularly. Unfortunately, the tissue-to-blood ratio is alsoconsidered to be of limited importance. In fact, the clinical utility

0

1

2

3

4

5

6

7

8

Sym

ptom

sco

re

1 2 3 4 5 6 7 8 9 10 20

Time (days)

Last doseLast day with effective

pulmonary concentrations

DyspneaCoughSputum

Fig. 1. Symptom scores, in patients with acute exacerbations of chronic bronchitis,after treatment with dirithromycin 500 mg/day for 5 days. Effective antibacterialconcentrations at different pulmonary sites of infection were achieved up to 96hours after the last dose. Symptom remission, evaluated by a decrease in arbitraryscores, progressed with trends in pulmonary antibacterial concentrations inde-pendently of the length of treatment (reproduced from Cazzola et al.,[67] withpermission).

264 Cazzola et al.


of penetration ratios is somewhat misleading, since bacterialeradication is a function of the drug concentration at the site ofinfection rather than penetration ratios.

Substantially, pulmonary pharmacokinetics is a very usefultool for describing how drugs behave in the human lung, but itdoes not promote an understanding of pharmacological effects ofa drug. More important, instead, is the correlation between pul-monary disposition of the drug and its MIC values for the infec-tious agent.[74]

Various antibacterials show a link between the correlationbetween tissue concentrations of the drug and the MIC values forthe infectious agent (known as inhibitory quotient or IQ or askilling ratio) and the eradication of the pathogens present in theairways.[68,75,76] The killing ratio is the best way to compare dis-similar antibacterials in terms of their likely efficacy in treatingpulmonary infections. Agents with high killing ratios (≥2) aremore likely to be associated with a favorable clinical outcomethan those with lower killing ratios.[77] For example, the peakconcentration in bronchial secretions-to-MIC ratio following asingle intravenous administration of ceftriaxone 1g in a patientwith acute exacerbation of chronic bronchitis shows thatceftriaxone is extremely active against H. influenzae and Klebsi-ella pneumoniae and partially active against S. pneumoniae, evenconsidering its concentrations at the site of infection[78] (figure2). It has also been documented that treatment is generally inef-fective when the peak tissue levels are equal to or below the MICfor the infecting bacteria. For example, in patients with exacer-bations of chronic bronchitis caused by H. influenzae, ampicillin(which has a MIC value of 0.12 mg/L against this bacterium) wasineffective at a dose of 200mg because its concentration in thesputum was below the MIC value. On the contrary, doses above400mg were effective.[79] Patients with pneumonia or exacerba-tion of chronic bronchitis responded adequately to 7 days of treat-ment with amoxicillin, particularly when the drug sputum con-

centrations exceeded 0.25 mg/L.[9] Conversely, sputum concen-trations of less than 0.5 mg/L following dosage of cefaclor of upto 500mg compared unfavorably with the MICs for H. influenzaeand Moraxella catarrhalis and were associated with clinical fail-ure.[80]

The results of these studies seem to indicate a good correla-tion between pulmonary concentrations of the drug and the MICfor the pathogens, but they have only correlated MIC values withthe peak concentration at the site of infection. In vivo, bacteriaare not exposed to constant but to constantly changing antibacte-rial concentrations with peaks and troughs.[81] Therefore, at thepulmonary site of infection, the pathogens are exposed to a gra-dient of antibacterial concentration according to the pharmaco-kinetics of the antibacterial.[81] The addition of bacteriologicalcharacteristics to in vivo pharmacokinetic studies has triggered a‘pharmacodynamic approach’. Pharmacodynamic parameters in-tegrate the microbiological activity and pharmacokinetics of ananti-infective drug by focusing on its biological effects, particu-larly growth inhibition and killing of pathogens. Therefore, theyallow better evaluation of the dosage regimen in conjunction withits clinical response. A series of so-called efficacy indices orsurrogate markers have been proposed by combining suitablepharmacokinetic parameters and susceptibility data (figure 3).[82]

3.1 Time-Dependent Killing

The pharmacodynamic parameters that determine antimicro-bial efficacy depend on the mechanism of antimicrobial killingobserved for the antibacterial, of which there are two primarypatterns (table II). Time-dependent killing is dependent on thetime that an antibacterial exceeds the MIC, but is independent ofthe antibacterial concentration, provided an inhibitory level hasbeen reached.[83] Antibacterials that show time-dependent kill-ing, such as β-lactam agents, but not carbapenems,[84] usually

IQ

100

10

1

0.1

0.01Streptococcuspneumoniae

Staphylococcusaureus

Haemophilusinfluenzae

Klebsiellapneumoniae

Pseudomonasaeruginosa

2 hours8 hours12 hours24 hours

Fig. 2. Inhibitory quotients [IQ; peak concentration in bronchial secretions to minimum inhibitory concentration (MIC) ratio] against various respiratory tract pathogensfollowing a single intramuscular administration of ceftriaxone 1g.[78]



have short postantibiotic effects, that is suppression of regrowthof bacteria after drug exposure. Consequently, regrowth of bac-teria, apart from Staphylococcus aureus,[85] occurs soon after theconcentrations in serum or tissue fall below the MIC. For thesedrugs, the time for which the concentration in serum exceeds theMIC (T>MIC) during the dosing interval has been shown to bethe key determinant of efficacy in a number of animal models ofinfection.[86] The proposed predictive values for T>MIC and bac-teriostatic effect in non-neutropenic hosts during intermittentdosing are 20 to 34% of the dosing interval for penicillins, 35 to55% for cephalosporins and 20 to 26% for carbapenems. Valuesfor selected pathogens may differ and include 24% for staphylo-cocci, 41% for streptococci, and 36% for Gram-negative bacilli.Extrapolation of animal data to neutropenic hosts suggests thatT>MIC should be 50 to 60% and 90 to 100%, respectively, forβ-lactam agents with and without a postantibiotic effect for thedesignated pathogen[87] (table III). It must be emphasized that time-dependent killing is characterized by maximum efficacy of anantibacterial at 2 to 4 times the MIC, an exposure profile achievedwhen 80 to 100% of the concentrations are above the MIC.[88]

This also coincides with an area under the concentration-timecurve (AUC)24/MIC ratio of 125 (where AUC24 is the AUC to 24hours). Further increases in concentration above these values donot kill bacteria more rapidly. In a study conducted by Schentaget al.,[89] there was a significant correlation between AUC for thetime interval that the concentrations are above the MIC dividedby the MIC value (AUIC), T>MIC and time to eradication. Fur-ther analysis of the data revealed that an AUIC >125 also corre-lated with microbiological response. However, the AUIC valueof 125 is derived from a study of nosocomial pneumonia, where

the line between success and failure is critically dependent ondrug activity alone. In less severely ill patients with better host

defense, it is conceivable that AUIC values as low as 30 to 60could be associated with successful outcome.[90]

3.2 Concentration-Dependent Killing

The second pharmacodynamic pattern is concentration-

dependent killing, where higher concentrations of antibacterialkill the pathogen more quickly and more completely. Antibacte-

rials that display concentration-dependent killing, such as amino-

glycosides and fluoroquinolones, show prolonged postantibiotic

effects. The goal for administration of these agents is to optimize

the dose and exposure to unbound drug concentrations. The phar-macodynamic parameters that correlate with successful clinical

and microbiological outcomes and prevent the emergence of bac-

terial resistance are unbound peak concentration (Cmax)/MIC andthe AUC24/MIC value.[91-93] Studies in dynamic in vitro models

of infection examined the relationship between dosage and bac-

terial killing. In these systems, bacteria are exposed to fluctuatingconcentrations of drug adjusted to simulate peak and trough se-

rum concentrations observed in humans. The models suggest thata Cmax/MIC ratio of 10 or above[94,95] and an AUIC ratio of 125

or above optimize rapid bacterial killing and prevent regrowth of

resistant Gram-negative bacterial subpopulations,[96] althoughthese agents may retain activity against Gram-positive micro-

organisms at lower AUC24/MIC ratios than required for Gram-

negative micro-organisms. Indeed, AUC24/MIC ratios as low as30 to 60 may maximize fluoroquinolone activity against S. pneu-

moniae[97] and ratios >57 indicate activity against S. aureus[98]

(table III). In any case, for concentration-dependent killers, higher

peaks translate into higher AUC/MIC ratios and longer T>MIC,

both of which are correlates of increased rates of killing.[86]

10

5

0

Peak

TroughAUC

AUIC24 =AUC24

MIC

MIC

0 2 4 6 8 10

Time (h)

Time above MIC

Ser

um c

once

ntra

tion

(μg/

ml)

Fig. 3. Surrogate markers proposed by combining suitable pharmacokinetic pa-rameters and susceptibility data. AUC = area under the concentration-time curve;AUC24 = area under the concentration-time curve to 24 hours; MIC = minimuminhibitory concentration; AUIC24 = area under the concentration-time curve forwhich antibacterial concentrations are above MIC divided by MIC to 24 hours.

Table II. Pharmacokinetic and pharmacodynamic parameters correlatingwith antibacterial efficacy in animal models of infection

Parameter Drugs

Time above the MIC Penicillins, cephalosporins,carbapenems, aztreonam, macrolides,clindamycin

24-hour AUC/MIC Aminoglycosides, fluoroquinolones,azithromycin, tetracyclines,vancomycin, quinapristin/dalfopristin

Cmax/MIC Aminoglycosides, fluoroquinolones

AUC = area under the concentration-time curve; Cmax = maximum concen-tration; MIC = minimum inhibitory concentration.

266 Cazzola et al.


3.3 Pharmacodynamic Modelling for Macrolides

Owing to the individual pharmacokinetic profiles of macro-lides, which exhibit a predominantly bacteriostatic effect, theirpharmacodynamic outcome parameters cannot be definitivelycharacterized as is possible for other antibacterial agents.[99]

Erythromycin and clarithromycin have concentration-inde-pendent bactericidal activity against streptococci and moderatepostantibiotic effects. The pharmacodynamic parameter that bestcorrelates with in vivo bacteriological response for these agentsis T>MIC. In contrast, the parameter that accounts for the phar-macodynamic effects of azithromycin and dirithromycin isAUC24/MIC. Azithromycin and dirithromycin have concentration-independent bactericidal activity and a prolonged postantibioticeffect. The duration of the postantibiotic effect correlates with itsAUC. In any case, owing to the pharmacokinetic profiles of mac-rolides, their activity relies not only on the measurable serumconcentrations, but also on the intracellular drug levels. In effect,for several macrolides, the classical pharmacodynamic model isincomplete, in that it does not fully describe or predict the suc-cessful clinical use of these drugs when the pathogen is an extra-cellular organism for which MICs are in the moderate range.Particularly for azithromycin, it has been speculated that the drugslowly diffuses out of the cell, resulting in high drug concentra-tions on the outer surface of the mammalian cell membrane (i.e.at the point of attachment of the pathogen to the solid surface ofthe cell); or it may involve the loading of granulocytes, whichmigrate to the local infection sites, go through an inflammatoryburst, and dump their load of drug extracellularly, the so-calledcellular drug delivery system.[100] The data suggest that indexingthe microbiological activities of many macrolides to concentra-tions in plasma may not be an appropriate method of pharmaco-dynamic modeling except when the target organism’s MIC islower than the average drug concentration in serum, plasma orinterstitial fluid. In this case drug concentrations in the centralcompartment are so overwhelmingly high, compared with the

MIC, that alternative models to explain the data are notneeded.[100]

It must be highlighted that Schentag et al.[101] have supportedthe validity of the AUIC as a universal parameter and attemptedto find a breakpoint value for it, to predict the efficacy of treat-ments of most antibacterial agents. They suggest a target AUICof ~125 regardless of S. pneumoniae versus Gram-negative or-ganisms.[102] However, some criticisms have arisen regarding theusefulness of this efficacy index, since the same AUIC value canbe obtained with different dosage regimens.[103,104]

4. Interrelationship Between Pharmacokineticsand Pharmacodynamics

A body of evidence confirms the effect of a satisfactory in-terrelationship between serum pharmacokinetics and pharmaco-dynamics on clinical and microbiological outcomes in patientswith lower respiratory tract infections. On the contrary, it is ex-tremely difficult to define the real effect of the interrelationshipbetween pulmonary pharmacokinetics and pharmacodynamicson clinical and microbiological outcomes. The majority of stud-ies have, in fact, only examined the interrelationship betweenserum pharmacokinetics and pharmacodynamics in patients withlower respiratory tract infections, probably because it is easierand ethical to sample blood than sputum, bronchial mucosa orELF.

Since β-lactam agents exert a dose-dependent bactericidaleffect on bacteria and do not have a significant postantibioticeffect, their levels at the site of infection should be above MICfor the entire length of treatment. A significant linear correlationexists between T>MIC and time to eradication of bacteria fromrespiratory secretions.[105] Nevertheless, the magnitude and du-ration by which concentrations must exceed the MIC remain con-troversial. In fact, papers that support the importance of theinterrelationship between pharmacokinetics and pharmaco-dynamics in inducing a good clinical and bacteriological out-

Table III. Magnitude of pharmacokinetic/pharmacodynamic (PK/PD) parameters determining efficacy against respiratory pathogens for different antibacte-rials

Drugs PK/PD parameter Magnitude for efficacy

β-Lactam agents with PAE T>MIC ≥50-60% of the dosing interval

β-Lactam agents without PAE T>MIC ≥90-100% of the dosing interval

Macrolides T>MIC ≥40-50% of the dosing interval

Azithromycin AUC24/MIC ≥25-30 (average 1 × MIC)

Fluoroquinolones AUC24/MIC ≥30-60 for Streptococcus pneumoniae

>57 for Staphylococcus aureus

≥125 for Gram-negative bacilli

AUC24 = area under the concentration-time curve to 24 hours; MIC = minimum inhibitory concentration; PAE = postantibiotic effect; T>MIC = time aboveMIC.



come are scarce.[106] In any case, experimental research hasshown that cephalosporins exert an in vivo bacteriostatic effecteven when their concentrations are above MIC for only 40% ofthe time between administrations, whereas maximal bactericidaleffect is obtained when concentrations are above MIC for 60 to70% of the time.[107] Therefore, the aim for a highly effectivedosing regimen would be to provide antibacterial levels abovethe MIC for at least 70% of the dosing interval.[85]

Ceftazidime reaches significantly higher levels than theMICs for the most common respiratory pathogens, at the potentialsites of lung infection, even 8 to 12 hours after the administrationof 1g intramuscularly[14] (figure 4). For this reason, a recent studyhas suggested the possibility of using ceftazidime in a single dailydose of 1g intramuscularly to treat patients with exacerbations ofchronic obstructive pulmonary disease who present only moder-ately impaired functional symptoms. On the contrary, this typeof therapeutic approach must be used with extreme caution inpatients with marked functional damage, although a satisfactoryclinical response may be obtained in some cases. In any case, thesmall number of patients included in this study does not allowsolid conclusions to be drawn.[108]

Although these data are intriguing, we must point out that,since studies on the tissue kinetics of β-lactam agents show evi-dence of a decline of antibacterial tissue levels parallel to serumconcentrations, it is always best to administer high concentrationsof the drug, particularly in treating patients hospitalized in inten-sive care units.[81] We must also emphasize that there is a closelink between antibacterial dosing and antibacterial resistance. Inparticular, resistance follows underdosing of the AUC in relationto the MIC.[109] A major component to the selective pressureargument is dosing. For years, we have under-appreciated the needto adjust antibacterial dosage as a means of lowering endemicresistance, and now we are beginning to suffer the consequences.It is now becoming urgent to manage resistance by managing thedosage and use patterns of our remaining agents.[110] With allβ-lactam agents, which have a slow, time-dependent antibacterialeffect, the aim must be to keep the antibacterial level above theMIC for the duration of therapy. Consequently, if the drug has along half-life, single doses at long intervals can be given. How-ever, if the half-life is short, the antibacterial should be givenfrequently, thus ensuring that it is maintained at concentrationsabove the MIC at all times in infected tissues.

Another study [111] has examined the relationship betweenthe bacterial susceptibility to cefaclor (MIC), achieved cefacloradvanced formulation (AF) serum and sputum concentrations,and in vivo eradication of the bacteria in patients with acute ex-acerbations of chronic bronchitis. Treatment was successful in allpatients, with percentage T>MIC in serum of >40%, whereas thetime that levels in sputum stayed above the MIC was not thepharmacodynamic parameter that correlated best with therapeu-tic efficacy for cefaclor.[111] This finding and the results of someother studies question the role of pulmonary pharmacodynamicsand seem to indicate that the correlation between clinical andmicrobiological outcomes and serum concentration is better thanthat between these outcomes and pulmonary levels of the anti-bacterial, probably because serum concentration better reflectsinterstitial fluid concentration.[112]

5. Conclusion

The determination of antibacterial concentrations at the siteof infection in the lung has been suggested to be important inpredicting the therapeutic efficacy of antimicrobial treatment ofbacterial infections of the lower respiratory tract.[74] Unfortu-nately, most studies are performed during steady state, in un-infected individuals. This could result in bias because the phar-macokinetics of antibacterials may be altered in individuals withan infection. It is also possible that tissue penetration at steadystate differs from that after a single dose and so several doses may

10

1

0.1

0.011 2 4 8 12

Time (h)

μg/m

l or

μg/g

Bronchial secretionBronchial mucosaEpithelial lining fluidHaemophilus influenzaeMoraxella catarrhalisStreptococcus pneumoniaeKlebsiella pneumoniaeStaphylococcus aureusPseudomonas aeruginosa

Fig. 4. Correlation between pulmonary concentration of intramuscular ceftazidime1g administered to patients with acute exacerbation of chronic bronchitis andminimum inhibitory concentration for relevant respiratory pathogens (reproducedfrom Cazzola et al.,[14] with permission).

268 Cazzola et al.


be needed to achieve steady state, which may also affect tissuepenetration. In any case, pulmonary pharmacokinetics does notpromote an understanding of pharmacological effects of a drug.More important, instead, is the correlation between pulmonarydisposition of the drug and its MIC values for the infectiousagent.[94] Unfortunately, the majority of studies have only exam-ined the interrelationship between serum pharmacokinetics andpharmacodynamics in patients with lower respiratory tract infec-tions. It is evident that studies looking at the effect of pulmonarypharmacodynamics, on clinical and microbiological outcomes,are absolutely necessary.

However, obtaining serial samples is difficult to impossiblein protected sites such as those in the lung (mainly bronchialmucosa), and also for ethical reasons, one must examine only asingle sample from each site of interest from each patient. Usinga site-to-plasma drug concentration ratio at a single timepoint asa measure of drug penetration is problematic, as there is oftensystem hysteresis that causes the ratio to change nearly continu-ously with time. While straightforward to perform as a study andalso straightforward to analyze, such investigations may give bi-ased estimates of drug penetration, depending on the samplingtime. On the other hand, a rise in MIC values reduces the timeabove MIC when an antibacterial is used at the same dosageregimen.[81] For example, the data from the Alexander Projectdemonstrated a decrease in time above MIC of relevant respira-tory pathogens when a comparison between 1992 and 1995 wasperformed[113] (figure 5).

It must be pointed out that a number of other factors mayinfluence how long a particular antibacterial remains above MIC,including properties of the drug itself and method of administra-tion.[114] High doses administered sufficiently often may com-pletely prevent any possibility of attaining a selective concentra-tion.[115] Alternatively, results of many animal and in vitro studiessuggest that continuous infusion may be extremely useful, at leastfor β-lactams.[116] Nevertheless, we must stress that for bacteriafor which the MIC is low, the daily dose may be substantiallylower than that used in conventional dosage regimens, while ininfections which are difficult to treat as a result of more resistantbacteria, continuous infusion may be more effective than anequivalent bolus dose.[117] Obviously, questions regarding thelevel by which the MIC needs to be exceeded and the effect ofintersubject variation in kinetics require prospective clinicalstudies.[87] Also, dosages and dosing intervals of antibacterialsfor lower respiratory tract infections should be designed not onlyaccording to their tissue penetration, but also with full consider-ation of the interrelationship between their pulmonary pharmaco-kinetics and pharmacodynamics.[118]

In any case, both pulmonary and serum interrelationshipsbetween serum pharmacokinetics and pharmacodynamics in pa-tients with lower respiratory tract infection supply useful infor-mation, but they are not devoid of gaps and do not allow us to beconfident that we are delivering antimicrobials to the lungs in thebest way for optimizing outcomes. In fact, we still lack an answerto the basic question: should antibacterial treatment choices be

S.p. 1992 S.p. 1995 H.i 1992 H.i. 1995 M.c. 1992 M.c. 1995

%T

>M

IC90

120

100

80

60

40

20

0

AmoxicillinCeftriaxoneAmoxicillin-clavulinic acidClarithromycinCefaclorErythromycinCefixime

Fig. 5. Time above minimum inhibitory concentration (% T>MIC) for Streptococcus pneumoniae (S.p.), Haemophilus influenzae (H.i.) and Moraxella catarrhalis (M.c.):comparison between 1992 and 1995.[113]



based on serum MICs, achievable lung levels at the site of infec-tion, or clinical outcome studies?

Acknowledgements

The authors have received no funding for the preparation of this manu-script and have no conflicts of interest directly relevant to it.

References1. Cazzola M. Problems and prospectives in the antibiotic treatment of lower respi-

ratory tract infections. Pulm Pharmacol 1994; 7: 139-522. Veber B, Vallée E, Desmont JM, et al. Correlation between macrolide lung phar-

macokinetics and therapeutic efficacy in a mouse model of pneumococcalpneumonia. J Antimicrob Chemother 1993; 32: 473-82

3. Honeybourne D. Antibiotic penetration in the respiratory tract and implicationsfor the selection of antimicrobial therapy. Curr Opin Pulm Med 1997; 3: 170-4

4. Valcke Y, Pauwels R, van der Straeten M. Pharmacokinetics of antibiotics in thelungs. Eur Respir J 1990; 3: 715-22

5. Baldwin DR, Honeybourne D, Wise R. Pulmonary disposition of antimicrobialagents in vivo: observations and clinical relevance. Antimicrob Agents Chem-other 1992; 36: 1176-80

6. Honeybourne D, Baldwin DR. The site concentrations of antimicrobial agents inthe lungs. J Antimicrob Chemother 1992; 30: 249-60

7. Cruciani M, Gatti G, Cazzadori A, et al. Pharmacokinetics of antimicrobial agentsin the respiratory tract. Zentralbl Bakteriol 1996; 284: 1-31

8. Bergogne-Bérézin E. New concepts in the pulmonary disposition of antibiotics.Pulm Pharmacol 1995; 8: 65-81

9. Stewart SM, Fisher M, Young JE, et al. Amoxycillin levels in sputum, serum andsaliva. Thorax 1974; 29: 110-4

10. Marlin GE, Burgess KR, Burgoyne J, et al. Penetration of piperacillin into bron-chial mucosa and sputum. Thorax 1981; 36: 774-80

11. Muller-Serieys C, Bergogne-Bérézin E, Rowan C, et al. Imipenem penetrationinto bronchial secretions. J Antimicrob Chemother 1987; 20: 618-9

12. Lovering AM, Pycock CJ, Harvey JE, et al. The pharmacokinetics and sputumpenetration of ampicillin and amoxycillin following simultaneous i.v. admin-istration. J Antimicrob Chemother 1990; 25: 385-92

13. Hill SL, Bilton D, Johnson MM, et al. Sputum and serum pharmacokinetics ofloracarbef (LY163892) in patients with chronic bronchial sepsis. J AntimicrobChemother 1994; 33: 129-36

14. Cazzola M, Matera MG, Polverino M, et al. Pulmonary penetration of ceftazidime.J Chemother 1995; 7: 50-4

15. Jehl F, Muller-Serieys C, de Larminat V, et al. Penetration of piperacillin-tazobactam into bronchial secretions after multiple doses to intensive care pa-tients. Antimicrob Agents Chemother 1994; 38: 2780-4

16. Bergogne-Berezin E, Morel C, Benard Y, et al. Pharmacokinetic study of beta-lactam antibiotics in bronchial secretions. Scand J Infect Dis 1978; Suppl. 14:267-72

17. Bergogne-Bérézin E, Berthelot G, Even P, et al. Penetration of ciprofloxacin intobronchial secretions. Eur J Clin Microbiol 1986; 5: 197-200

18. Davies B, Maesen FPV, Geraedts W, et al. Penetration of ofloxacin from serumto sputum. Drugs 1987; 34 Suppl. 1: 26-2

19. Bergogne-Bérézin E, Muller-Serieys C, Kafe H, et al. Penetration of lomefloxacininto bronchial secretions following single and multiple oral administration. AmJ Med 1992; 92 Suppl. 4A: S8-11

20. Fujita A, Miya T, Tanaka R, et al. Levofloxacin concentrations in serum, sputumand lung tissue: evaluation of its efficacy according to breakpoint. Jpn J Anti-biot 1999; 52: 661-6

21. Cazzola M, Matera MG, Tufano MA, et al. Pulmonary disposition of lomefloxacinin patients with acute exacerbation of chronic obstructive pulmonary disease.A multiple-dose study. J Chemother 2001; 13: 407-12

22. Begg EJ, Robson RA, Saunders DA, et al. The pharmacokinetics of oral fleroxacinand ciprofloxacin in plasma and sputum during acute and chronic dosing. Br JClin Pharmacol 2000; 49: 32-8

23. Pennington JE, Reynold HY. Concentrations of gentamicin and carbenicillin inbronchial secretions. J Infect Dis 1973; 128: 63-8

24. Gartmann J. Doxycycline concentrations in lung tissue, bronchial wall, bronchialsecretions. Chemotherapy 1975; 21: 19-26

25. Odio W, Vanlaer E, Klastersky J. Concentrations of gentamicin in bronchial se-cretions after intramuscular and endotracheal administration. J Clin Pharmacol1975; 15: 518-24

26. Dull WL, Alexander MR, Kasik JE. Bronchial secretion levels of amikacin. Anti-microb Agents Chemother 1979; 16: 767-71

27. Valcke YJ, Vogelaers DP, Colardyn FA, et al. Penetration of netilmicin in thelower respiratory tract after once-daily dosing. Chest 1992; 101: 1028-32

28. Santre C, Georges H, Jacquier JM, et al. Amikacin levels in bronchial secretionsof 10 pneumonia patients with respiratory support treated once daily versustwice daily. Antimicrob Agents Chemother 1995 39: 264-7

29. Marlin GE, Davies PR, Rutland J, et al. Plasma and sputum erythromycin concen-trations in chronic bronchitis. Thorax 1980; 35: 441-5

30. Bergogne-Bérézin E. Tissue distribution of macrolide antibiotics. In: Bryskier AJ,Butzler J-P, Neu HC, et al., editors. Macrolides: chemistry, pharmacology andclinical uses. Paris: Arnette Blackwell, 1993: 84

31. Matera MG, Tufano MA, Polverino M, et al. Pulmonary concentrations ofdirithromycin and erythromycin during acute exacerbation of mild chronicobstructive pulmonary disease. Eur Respir J 1997; 10: 97-102

32. Baldwin DR, Wise R, Andrews JM, et al. Azithromycin concentrations at the sitesof pulmonary infection. Eur Respir J 1990; 3: 886-90

33. Cazzola M, Matera MG, Tufano MA, et al. Pulmonary penetration of dirithro-mycin in patients suffering from acute exacerbation of chronic bronchitis. PulmPharmacol 1994; 7: 377-81

34. Honeybourne D. Antibiotic penetration into lung tissue. Thorax 1994; 49: 104-6

35. Wise R, Baldwin DR, Honeybourne D. Penetration of antibiotics into the bron-chial mucosa. Res Clin Forums 1990; 12 (4): 95-100

36. Baldwin DR, Andrews JM, Wise R, et al. Bronchoalveolar distribution of cefur-oxime axetil and in vitro efficacy of observed concentrations against respiratorypathogens. J Antimicrob Chemother 1992; 30: 377-85

37. Wise R, Honeybourne D. Antibiotic penetration into the respiratory tract. A basisfor rational therapy. J Chemother 1995; 4: 28-32

38. Wise R. The pharmacokinetics of the oral cephalosporins: a review. J AntimicrobChemother 1990; 26 Suppl. E: 13-20

39. Baldwin DR, Wise R, Andrews JM, et al. Concentrations of cefpodoxime in serumand bronchial mucosal biopsies. J Antimicrob Chemother 1992; 30: 67-71

40. Krumpe P, Lin C-C, Radwanski E, et al. The penetration of ceftibuten into therespiratory tract. Chest 1999 116: 369-74

41. Baldwin DR, Wise R, Andrews JM, et al. Comparative bronchoalveolar concen-trations of ciprofloxacin and lomefloxacin following oral administration. Res-pir Med 1993; 87: 595-601

42. Cook PJ, Andrews JM, Wise R, et al. Concentrations of OPC-17116, a newfluoroquinolone antibacterial, in serum and lung compartments. J AntimicrobChemother 1995; 35: 317-26

43. Wise R, Honeybourne D. A review of the penetration of sparfloxacin into thelower respiratory tract and sinuses. J Antimicrob Chemother 1996; 37: 57-63

44. Andrews JM, Honeybourne D, Brenwald NP, et al. Concentrations of trova-floxacin in bronchial mucosa, epithelial lining fluid, alveolar macrophages andserum after administration of single or multiple oral doses to patients undergo-ing fibre-optic bronchoscopy. J Antimicrob Chemother 1997; 39: 797-802

45. Andrews JM, Honeybourne D, Jevons G, et al. Concentrations of levofloxacin(HR 355) in the respiratory tract following a single oral dose in patients under-going fibre-optic bronchoscopy. J Antimicrob Chemother 1997; 40: 573-7

46. Soman A, Honeybourne D, Andrews J, et al. Concentrations of moxifloxacin inserum and pulmonary compartments following a single 400 mg oral dose in

270 Cazzola et al.


patients undergoing fibre-optic bronchoscopy. J Antimicrob Chemother 1999;44: 835-8

47. Honeybourne D, Andrews JM, Cunningham B, et al. The concentrations ofclinafloxacin in alveolar macrophages, epithelial lining fluid, bronchial mucosaand serum after administration of single 200 mg oral doses to patients undergo-ing fibre-optic bronchoscopy. J Antimicrob Chemother 1999 43: 153-5

48. Honeybourne D, Banerjee D, Andrews J, et al. Concentrations of gatifloxacin inplasma and pulmonary compartments following a single 400 mg oral dose inpatients undergoing fibre-optic bronchoscopy. J Antimicrob Chemother 2001;48: 63-6

49. Gotfried MH, Danziger LH, Rodvold KA. Steady-state plasma and intrapulmonaryconcentrations of levofloxacin and ciprofloxacin in healthy adult subjects. Chest2001; 119: 1114-22

50. Lamy P, Anthoine D, Zuck P, et al. Etude pharmacocinétique de la spiramycine enpathologie infectieuse respiratoire. Ann Med 1977; 16: 109-12

51. Chastre J, Brun P, Fourtillan JB, et al. Pulmonary disposition of roxithromycin (RU28965), a new macrolide antibiotic. Antimicrob Agents Chemother 1987; 31:1312-6

52. Mattie H, Hoogeterp JJ, Kaajan JPG, et al. The penetration of erythromycin intohuman bronchial mucosa. Br J Clin Pharmacol 1987; 24: 179-83

53. Cars O. Distribution of antibiotics into tissues and cells. Medical Masterclasses1993; 1 (3): 2-9

54. Staehelin LA. Structure and function of intercellular junctions. Intern Ren Cytol1974; 39: 191-283

55. Baldwin DR, Wise R, Andrews JM, et al. Concentrations of antimicrobials in thepulmonary alveolar epithelial lining. Res Clin Forums 1990; 12 (4): 103-13

56. Baldwin DR, Andrews JM, Wise R, et al. Bronchoalveolar distribution of cefurox-ime axetil and in-vitro efficacy of observed concentrations against respiratorypathogens. J Antimicrob Chemother 1992; 30: 377-85

57. Cook PJ, Andrews JM, Woodcock J, et al. Concentration of amoxycillin andclavulanate in lung compartments in adults without pulmonary infection. Thorax1994; 49: 1134-8

58. Cook PJ, Andrews JM, Wise R, et al. Distribution of cefdinir, a third generationcephalosporin antibiotic, in serum and pulmonary compartments. J AntimicrobChemother 1996; 37: 331-9

59. Allegranzi B, Cazzadori A, Di Perri G, et al. Concentrations of single-dosemeropenem (1 g iv) in bronchoalveolar lavage and epithelial lining fluid. JAntimicrob Chemother 2000; 46: 319-22

60. Panteix G, Harf R, de Montclos H, et al. Josamycin pulmonary penetration deter-mined by broncho-alveolar lavage in man. J Antimicrob Chemother 1988; 22:917-21

61. Conte Jr JE, Golden J, Duncan S, et al. Single-dose intrapulmonary pharmacoki-netics of azithromycin, clarithromycin, ciprofloxacin, and cefuroxime in volun-teer subjects. Antimicrob Agents Chemother 1996; 40: 1617-22

62. Khair OA, Andrews JM, Honeybourne D, et al. Lung concentrations oftelithromycin after oral dosing. J Antimicrob Chemother 2001; 47: 837-40

63. Carcas AJ, Garcia-Satue JL, Zapater P, et al. Tobramycin penetration into epitheliallining fluid of patients with pneumonia. Clin Pharmacol Ther 1999; 65: 245-50

64. Tulkens PM. Intracellular distribution and activity of antibiotics. Eur J ClinMicrobiol Infect Dis 1991; 10: 100-6

65. Honeybourne D. Antibiotic penetration in the respiratory tract and implications forthe selection of antimicrobial therapy. Curr Opin Pulm Med 1997; 3: 170-4

66. Plouffe J, Schwartz DB, Kolokathis A, et al. Clinical efficacy of intravenous fol-lowed by oral azithromycin monotherapy in hospitalized patients with commu-nity-acquired pneumonia. The Azithromycin Intravenous Clinical Trials Group.Antimicrob Agents Chemother 2000; 44: 1796-802

67. Cazzola M, Caputi M, Santangelo G, et al. A five-day course of dirithromycin inthe treatment of acute exacerbation of severe chronic obstructive pulmonarydisease. J Chemother 1997; 9: 279-84

68. Bergogne-Bérézin E. Predicting the efficacy of antimicrobial agents in respiratoryinfections – is tissue concentration a valid measure? J Antimicrob Chemother1995; 35: 363-71

69. Neu HC. Activity of macrolides against common pathogens in vitro. In: BryskierAJ, Butzler J-P, Neu HC, et al., editors. Macrolides: chemistry, pharmacologyand clinical uses. Paris: Arnette-Blackwell, 1993: 82

70. Sides GD, Cerimele BJ, Black HR, et al. Pharmacokinetics of dirithromycin. JAntimicrob Chemother 1993; 31 Suppl. C: 65-75

71. Panteix G, Harf R, Desnottes JF, et al. Accumulation of pefloxacin in the lowerrespiratory tract demonstrated by bronchoalveolar lavage. J Antimicrob Chem-other 1994; 33: 979-85

72. Nolting A, Costa TD, Vistelle R, et al. Determination of free extracellular concen-trations of piperacillin by microdialysis. J Pharm Sci 1996; 85: 369-72

73. Müller M, Haag O, Burgdorff T, et al. Characterization of peripheral-compartmentkinetics of antibiotics by in vivo microdialysis in humans. Antimicrob AgentsChemother 1996; 40: 2703-9

74. Bergogne-Bérézin E, Vallée E. Pharmacokinetics of antibiotics in respiratory tis-sues and fluids. In: Pennington JE, editor. Respiratory infections: diagnosis andmanagement. 3rd ed. New York: Raven Press, 1994: 715-40

75. Ellner PD, Neu HC. The inhibitory quotient. A method for interpreting minimuminhibitory concentration data. JAMA 1981; 246: 1575-8

76. Cazzola M. Pulmonary pharmacokinetics of dirithromycin allow a 5-day treatmentof acute bacterial exacerbation of chronic bronchitis. Drugs Today 1995; 31:105-9

77. Cunha BA. Antibiotic pharmacokinetic considerations in pulmonary infections.Semin Respir Infect 1991; 6: 168-82

78. Fraschini F, Scaglione F, Cogo R, et al. Bactericidal effect of ceftriaxone versusimipenem plus cilastin in bronchial secretion. Chemotherapy 1988; 34 Suppl.1: 3-15

79. Maesen FPV, Beeuwkes H, Davies BI, et al. Bacampicillin in acute exacerbationsof chronic bronchitis: a dose-range study. J Antimicrob Chemother 1976; 2:279-85

80. Maesen FPV, Geraedts WH, Davies BI. Cefaclor in the treatment of chronic bron-chitis. J Antimicrob Chemother 1990; 26: 456-7

81. Cazzola M, Matera MG, Noschese P. Parenteral antibiotic therapy in the treatmentof lower respiratory tract infections. Strategies to minimize the development ofantibiotic resistance. Pulm Pharmacol Ther 2000; 13: 249-56

82. Delacher S, Derendorf H, Hollenstein U, et al. A combined in vivo pharmacoki-netic-in vitro pharmacodynamic approach to simulate target site pharmacody-namics of antibiotics in humans. J Antimicrob Chemother 2000; 46: 733-9

83. Craig WA, Ebert SC. Killing and regrowth of bacteria in vitro: a review. Scand JInfect Dis 1990; 74 Suppl.: 63-70

84. Bustamante CI, Drusano GL, Tatem BA, et al. Post-antibiotic effect of imipenemon Pseudomonas aeruginosa. Antimicrob Agents Chemother 1984; 26: 678-82

85. Craig WA. Interrelationship between pharmacokinetics and pharmacodynamics indetermining dosage regimens for broad-spectrum cephalosporins. DiagnMicrobiol Infect Dis 1995; 22: 89-96

86. Craig WA. Pharmacokinetic/pharmacodynamic parameters: rationale for antibac-terial dosing of mice and men. Clin Infect Dis 1998; 26: 1-12

87. Turnidge JD. The pharmacodynamics of ß-lactams. Clin Infect Dis 1998; 27: 10-22

88. Drusano G L. Human pharmacodynamics of β-lactams, aminoglycosides and theircombination. Scand J Infect Dis 1990; 74: 235-48

89. Schentag JJ, Swanson DJ, Smith IL. Dual individualization: antibiotic dosage cal-culation from the integration of in-vitro pharmacodynamics and in-vivo phar-macokinetics. J Antimicrob Chemother 1985; 15 Suppl. A: 47-57

90. Nix DE, Schentag JJ. Role of pharmacokinetics and pharmacodynamics in thedesign of dosage schedules for 12-h cefotaxime alone and in combination withother antibiotics. Diagn Microbiol Infect Dis 1995; 22: 71-6

91. Lacy MK, Nicolau DP, Nightingale CH, et al. The pharmacodynamics of amino-glycosides. Clin Infect Dis 1998; 27: 23-7

92. Lode H, Borner K, Koeppe P. Pharmacodynamics of fluoroquinolones. Clin InfectDis 1998; 27: 33-9

93. Turnidge J. Pharmacokinetics and pharmacodynamics of fluoroquinolones. Drugs1999; 58 Suppl. 2: 29-36



94. Moore RD, Lietman PS, Smith CR. Clinical response to aminoglycoside therapy:

Importance of the ratio of peak concentration to minimum inhibitory concen-

tration. J Infect Dis 1987; 155: 93-9

95. Preston SL, Drusano GL, Berman AL, et al. Pharmacodynamics of levofloxacin.

JAMA 1998; 279: 125-9

96. Schentag JJ, Gilliland KK, Paladino JA. What have we learned from pharmaco-

kinetic and pharmacodynamic theories? Clin Infect Dis 2001; 32 Suppl. 1:

S39-46

97. Lister PD, Sanders CC. Pharmacodynamics of levofloxacin and ciprofloxacin

against Streptococcus pneumoniae. J Antimicrob Chemother 1999; 43: 79-86

98. Hoang AD, Peterson M, Hovde LB, et al. Investigation of AUC/MIC ratio as a

generic predictor of fluoroquinolone activity against Staphylococcus aureus

using trovafloxacin, ciprofloxacin, sparfloxacin, and levofloxacin in an in vitro

pharmacodynamic model [abstract]. Program and abstracts of the 98th General

Meeting of the American Society for Microbiology. Washington, DC: Ameri-

can Society for Microbiology, 1998: 50

99. Craig WA. Postantibiotic effects and the dosing of macrolides, azalides, and

streptogramins. In: Zinner SH, Young LS, Acar JF, et al., editors. Expanding

indications for the new macrolides, azalides, and streptogramins. New York:

Marcel Dekker, 1997: 27-38

100. Patel KB, Xuan D, Tessier PR, et al. Comparison of bronchopulmonary pharma-

cokinetics of clarithromycin and azithromycin. Antimicrob Agents Chemother

1996; 40: 2375-9

101. Schentag JJ, Nix DE, Adelman MH. Mathematical examination of dual individu-

alization principles (I): relationships between AUC above MIC and area under

the inhibitory curve for cefmenoxime, ciprofloxacin and tobramycin. DICP

1991; 25: 1050-7

102. Schentag JJ, Gilliland KK, Paladino JA. Fluoroquinolone pharmacodynamics [re-

ply]. Clin Infect Dis 2001; 33: 2092-6

103. Dalla Costa TD, Derendorf H. AUIC: a general target for the optimization of

dosing regimens of antibiotics? Ann Pharmacother 1996; 30: 1024-8

104. Drusano GL, Preston SL, Owens Jr RC, et al. Fluoroquinolone pharmacodynam-

ics. Clin Infect Dis 2001; 33: 2091-2

105. Odenholt-Tornqvist I, Lowdin E, Car O. Pharmacodynamic effects of subinhibi-

tory concentrations of β-lactam antibiotics in vitro. Antimicrob Agents Chem-

other 1991; 35: 1834-9

106. Cazzola M, Matera MG, Donner CF. Pharmacokinetics and pharmacodynamics

of newer oral cephalosporins: implication for treatment of community-acquired

lower respiratory tract infections. Clin Drug Invest 1998; 16: 335-46

107. Vogelman B, Gudmundsson S, Leggett J, et al. Correlation of antimicrobial phar-macokinetic parameters with therapeutic efficacy in an animal model. J InfectDis 1988; 158: 831-47

108. Cazzola M, Noschese P, Vinciguerra A, et al. Le correlazioni esistenti trafarmacocinetica polmonare e farmacodinamica permettono di ipotizzarel’utilizzo del ceftazidime alla dose di 1 g una sola volta al giorno nel trattamentodelle riacutizzazioni batteriche della broncopneumopatia cronica ostruttiva condanno funzionale moderato. Minerva Med 1998; 89: 15-22

109. Thomas JK, Forrest A, Bhavnani SM, et al. Pharmacodynamic evaluation of fac-tors associated with the development of bacterial resistance in acutely ill pa-tients during therapy. Antimicrob Agents Chemother 1998; 42: 521-7

110. Highet VS, Forrest A, Ballow CH, et al. Antibiotic dosing issues in lower respi-ratory tract infection: population-derived area under inhibitory curve is predic-tive of efficacy. J Antimicrob Chemother 1999; 43: 55-63

111. Cazzola M, Di Perna F, Boveri B, et al. Interrelationship between the pharmaco-kinetics and pharmacodynamics of cefaclor advanced formulation in patientswith acute exacerbation of chronic bronchitis. J Chemother 2000; 12: 216-22

112. Vallee E, Azoulay-Dupuis E, Bauchet J, et al. Kinetic disposition of temafloxacinand ciprofloxacin in a murine model of pneumococcal pneumonia. Relevancefor drug efficacy. J Pharmacol Exp Ther 1992; 262: 1203-8

113. Drusano GL, Craig WA. Relevance of pharmacokinetics and pharmacodynamicsin the selection of antibiotics for respiratory tract infections. J Chemother 1997;9 Suppl. 3: 38-44

114. Craig WA. Outpatient parenteral antibiotic therapy. Management of serious infec-tions. Part I: medical, socioeconomic, and legal issues. Selecting the antibiotic.Hosp Pract (Off Ed) 1993; 28 Suppl. 1: 16-20

115. Baquero F, Negri MC. Strategies to minimize the development of antibiotic resis-tance. J Chemother 1997; 9 Suppl. 3: 29-37

116. Vondracek TG. β-lactam antibiotics: is continuous infusion the preferred methodof administration? Ann Pharmacother 1995; 9: 415-24

117. Mouton JW, Vinks AA. Is continuous infusion of β-lactam antibiotics worth-while?. Efficacy and pharmacokinetic considerations. J Antimicrob Chemother1996; 38: 5-15

118. Cazzola M, Matera MG. Interrelationship between pharmacokinetics and pharma-codynamics in the design of dosage regimens for treating acute exacerbationsof chronic bronchitis. Respir Med 1998; 92: 895-901

Correspondence and offprints: Dr Mario Cazzola, Via del Parco Margherita24, 80121 Napoli, Italy.E-mail: [email protected]

272 Cazzola et al.


Delivering Antibacterials to the Lungs

Documents

Transcript of Delivering Antibacterials to the Lungs