Antimicrobial Agents in the Treatment of Bacterial Meningitis

Antimicrobial agents in the treatmentof bacterial meningitis

Scott W. Sinner, MDa, Allan R. Tunkel, MD, PhDb,*aDivision of Infectious Diseases, Drexel University College of Medicine,

245 North 15th Street, Mail Stop 487, Philadelphia, PA 19102-1101, USAbDepartment of Medicine, Drexel University College of Medicine,

245 North 15th Street, Mail Stop 487, Philadelphia, PA 19102-1101, USA

Although signicant advances have been made in the management ofbacterial meningitis, the disease has unacceptably high morbidity andmortality [1]. Clinical success in treating patients with bacterial meningitisrelies on familiarity with the pharmacologic principles of the commonlyused antimicrobial agents, which includes penetration of drug into thecerebrospinal uid (CSF) and the drugs activity in purulent CSF. Theseproperties, along with the mode of administration of the drug and theintrinsic pharmacodynamic relationships of CSF drug concentrations tobactericidal activity, help to dene the potential ecacy of a givenantimicrobial agent in the treatment of bacterial meningitis. Much of whatis known about these principles has been derived from the experimentalrabbit model of meningitis, rst described three decades ago [2].

Many changes in the epidemiology and etiology of bacterial meningitishave dictated modications in antimicrobial therapy over the last severalyears. One of the most important changes has been the dramatic decline inthe incidence of meningitis caused by Haemophilus inuenzae type b, asa result of the widespread use of the H inuenzae type b conjugate vaccine[3]. Since licensure of the pneumococcal conjugate vaccine, there has alsobeen a dramatic decline in the incidence of invasive pneumococcal infections[4], which may further change the epidemiology of pneumococcal meningitisin areas where vaccine use has been implemented. Another important

Infect Dis Clin N Am 18 (2004) 581602development has been the rise in the incidence of antimicrobial-resistantmeningeal pathogens, particularly Streptococcus pneumoniae, which has

* Corresponding author. Oce of Admissions, Drexel University College of Medicine,

2900 Queen Lane, Philadelphia, PA 19129.

E-mail address: [email protected] (A.R. Tunkel).

0891-5520/04/$ - see front matter 2004 Elsevier Inc. All rights reserved.doi:10.1016/j.idc.2004.04.005

been observed in the United States and throughout the world [5]. As a resultof this increasing problem of drug resistance, research in recent years hasfocused on nding new antimicrobial agents for the therapy of meningitiscaused by penicillin- and cephalosporin-resistant pneumococci.

Rather than attempt to review all aspects of the therapy of bacterialmeningitis, this article reviews the pharmacologic principles of use ofantimicrobial therapy, summarizes the specic animal data for selectedantimicrobial agents, and gives specic recommendations for therapy ofbacterial meningitis based on the isolated meningeal pathogen.

Basic pharmacologic principles

Cerebrospinal uid penetration

The rst pharmacologic factor that determines whether an antimicrobialagent is able to clear bacteria from the CSF is the drugs penetration acrossthe blood-brain barrier and into the subarachnoid space. In general, thefollowing drug characteristics aect the blood-brain barrier penetration andsubsequent CSF concentrations of the antimicrobial agent [1,69]:

1. Lipophilic agents (eg, uoroquinolones, chloramphenicol [Chloromy-cetin], rifampin [Rifadin, Rimactane], and the sulfonamides) diffuseacross the blood-brain barrier by transcellular pathways, reach peak CSFconcentrations (Cpeak) quickly, and penetrate relatively well, even in theabsence of meningeal inammation. Hydrophilic agents (eg, b-lactamsand vancomycin [Vancocin]), must enter the CSF through paracellulartight junctions, reach Cpeak more slowly, and have decreased penetration(sometimes markedly so) if the meninges are not inamed.

2. Drugs with a low molecular weight and relatively simple structure (eg,the uoroquinolones and rifampin) enter the CSF more readily thanlarger compounds or those with more complex chemical structures (eg,vancomycin).

3. Drugs that are highly protein-bound in serum (eg, ceftriaxone [Rocephin])display a lower degree of CSF penetration, because only the unboundfraction of the drug is available to cross the blood-brain barrier.

4. The choroid plexus contains an active transport system that can pumppenicillins and cephalosporins (and potentially aminoglycosides anduoroquinolones) from the CSF into the blood. Conversely, an activetransport system present in central nervous system capillaries cantransport penicillins and cephalosporins from the blood into the CSF,although this systems low afnity and capacity for drugs limits itsimportance in being able to achieve adequate CSF concentrations of

582 S.W. Sinner, A.R. Tunkel / Infect Dis Clin N Am 18 (2004) 581602a specic agent [10].

The percent CSF penetration of an individual antimicrobial agent may bemeasured in a number of dierent ways. If simultaneous CSF and serum

concentrations are measured to determine the percent penetration calculatedby the formula, (CSF concentration/serum concentration) 100, the resultsmay be misleading because the concentration-time curve of an antimicrobialin CSF usually lags behind that in serum [8]. This was demonstrated ina study of pediatric patients with meningitis in which the CSF penetration ofmeropenem (Merrem) was calculated at 7.8% when measured within 2hours of antimicrobial administration, but was recalculated at 23.9% whenmeasured after 2 hours [11]. Alternatively, one can calculate a percent CSFpenetration based on the Cpeak attained in both serum and in CSF. Thismethod is the one by which many of the published CSF penetration gureswere determined (Table 1). A more meaningful parameter for manyantimicrobial agents is the ratio of the area under the concentration curve(AUC) in CSF to that in serum. This method, however, requires multipleconcentration measurements in both CSF and serum, and is impractical forhuman pharmacokinetic studies.

Antimicrobial activity in purulent cerebrospinal uid

The second pharmacologic property of an antibacterial that helps todetermine its usefulness in treating meningitis is the activity of that agent inthe purulent environment of infected CSF. This may depend on a number ofparameters [1,69]:

1. The decreased CSF pH in bacterial meningitis contributes, at least in part,to the poor response seen with aminoglycosides, and perhaps also withclarithromycin (Biaxin), in experimental animal models of meningitis.

Table 1

Pharmacologic parameters of common antimicrobial agents used to treat bacterial meningitis

Percent CSF penetration Half-life (hours) in humansPharmacologic

Parametersa

Drug Animals Humans Serum CSF

Penicillin G 2.66 510 0.5 ND Time above MBC

Ampicillin 818.4 1335 0.71.4 2.13.6 Time above MBC

Ceftriaxone 2.712 1.516 5.410.9 16.8b Time above MBC

Gentamicin 18.928.7 030 23 ND CSFpeak/MBC

Ciprooxacin 1527.5 637 2.54 4.37.2b Time above MBC,

AUC-MBC

Vancomycin 8.413 153 48 2.84.1b Time above MBC

Rifampin 17.222 756 25.8 9.121b Paradoxicalc

TMP-SMX 1739 2435 11/9 ND ND

Abbreviations: AUC, area under the concentration curve; CSF, cerebrospinal uid; MBC,

minimum bactericidal concentration; ND, not determined; TMP-SMX, trimethoprim-sulfa-

methoxazole.a

583S.W. Sinner, A.R. Tunkel / Infect Dis Clin N Am 18 (2004) 581602Parameters that are most related to eective bactericidal activity in meningitis models.b Half-lives determined in patients with external ventriculostomies.c Paradoxical dose-response relationship has been seen with rifampin; see text for details.

Data from references [8,1226].

2. Highly protein-bound antimicrobial agents may have somewhat di-minished activity as a result of the high CSF protein concentrationsoften seen in bacterial meningitis, which result in decreased CSFconcentrations of free drug available to kill bacteria.

3. The relatively slow growth of bacteria in CSF may reduce the activity ofb-lactam agents, which are dependent on a brisk bacterial growth ratefor maximal activity.

4. Some antibacterials may be metabolized in the CSF space to less activecompounds (eg, cephalothin [Kein] to desacetylcephalothin). Incontrast, some antimicrobial agents still retain signicant antibacterialactivity after metabolism (eg, cefotaxime [Claforan] to desacetylcefotax-ime) that may be equal to that of the parent compound.

5. Certain antimicrobial combinations may be synergistic when co-admini-stered (eg, ampicillin [Omnipen, Principen] plus gentamicin [Garamycin]for Listeria monocytogenes and Streptococcus agalactiae meningitis),whereas others may be antagonistic (eg, chloramphenicol combined witheither penicillin or gentamicin).

6. The activity of some antimicrobial agents against certain bacterial strainsmay be decreased in the presence of high bacterial densities, the so-calledinoculum effect, in which high bacterial densities may result insignicant increases in the minimum inhibitory concentration (MIC) ofcertain antimicrobial agents against a specic microorganism.

Mode of administration

The choice of mode of antimicrobial administration is a third factorinuencing the potential success of a given agent in the treatment ofbacterial meningitis [1,6]. Intravenous antimicrobial agents may be admin-istered by intermittent bolus or by continuous infusion. The success of onemethod over another depends on the specic agent and the microorganismbeing studied because, just as in serum, various CSF pharmacodynamicparameters are important for dierent antibacterial agents (see Table 1)[8,1226]. Bolus administration of an antimicrobial agent leads to a higherCpeak, but may not maintain concentrations above the microorganismsminimum bactericidal concentration (MBC) for the entire dosing interval,whereas continuous infusion produces a lower Cpeak, but may maintainconcentrations above the MBC during nearly 100% of the dosing interval.The MBC, rather than the MIC, is often used in experimental meningitisstudies, because of the need for true bactericidal activity in this environmentwhere there is relatively little immune system function (ie, as a result of lowCSF concentrations of immunoglobulins and complement).

584 S.W. Sinner, A.R. Tunkel / Infect Dis Clin N Am 18 (2004) 581602Antimicrobial pharmacodynamics in cerebrospinal uid

A nal factor that may contribute to the response to antimicrobialtherapy in bacterial meningitis is pharmacodynamics, which is concerned

with the time course of antimicrobial activity at the site of infection (seeTable 1). For b-lactam agents, the following pharmacodynamic measuresseem to be interrelated: the ratio of Cpeak to the MBC for the isolate, theratio of the CSF AUC to the MBC, and the time that the b-lactam CSFconcentration remains above the MBC [8]. In a study of ceftriaxone therapyin the experimental rabbit model of cephalosporin-resistant pneumococcalmeningitis, these three parameters were interconnected, although the timeabove the MBC was the only factor that was independently correlated withthe bacterial killing rate (BKR) [27]; maximal BKR was only attained whenthe CSF ceftriaxone concentration exceeded the MBC for 95% to 100% ofthe dosing interval. Although not completely conrmed, vancomycin killingin CSF also seems to be mostly a time-dependent process [12].

For aminoglycosides, the concentration-dependent killing observed inserum is also seen in CSF, as was demonstrated in an experimental model ofEscherichia coli meningitis [20]. The CSF pharmacodynamics of uoroqui-nolones is more complicated, however, in that features of both time-dependency and concentration-dependency have been observed. Theexpected concentration-dependency of uoroquinolone killing was demon-strated in an experimental rabbit model of pneumococcal meningitis [21],but time above the MBC was also found to be important for bacterial killingby both trovaoxacin (Trovan) [28] and gatioxacin (Tequin) [29] in laterexperimental studies. The pharmacodynamics of rifampin are unusual, inthat a paradoxical decrease in BKR has been demonstrated at infusion ratesof 20 mg/kgh compared with 10 mg/kgh [13], possibly as a result of theneed for some amount of ongoing protein synthesis to maintain bacterialkilling.

Other pharmacodynamic properties of antimicrobial agents in CSF thathave been examined are the postantibiotic eect, which describes delayedregrowth of bacteria after removal of an antimicrobial agent [6], and thepost sub-MIC eect, which describes continued suppression of bacterialregrowth once antimicrobial concentrations fall below the MIC [8]. In invitro studies, most antibacterial agents display a signicant postantibioticeect against gram-positive organisms, whereas only aminoglycosides,uoroquinolones, and carbapenems exert a postantibiotic eect againstgram-negative organisms [30]. Studies of ampicillin in experimental animalmodels of pneumococcal meningitis suggest that slowly declining sub-MICantimicrobial concentrations in CSF are important in delaying bacterialregrowth, but that a true postantibiotic eect is unlikely to exist to anysignicant degree in CSF [31,32].

Experimental data for selected antibacterial agents

585S.W. Sinner, A.R. Tunkel / Infect Dis Clin N Am 18 (2004) 581602It is beyond the scope of this article to discuss all antimicrobial agents thathave been studied in experimental animal models of bacterial meningitis.Rather, the focus is on specic agents that are most commonly used in current

treatment regimens, and agents that may become important in the future.When relevant, the eects of adjunctive dexamethasone on ecacy of specicagents in experimental animal models are discussed, because this may haverelevance in the treatment of patients with bacterial meningitis.

b-Lactams

The b-lactam agents are among the most commonly used classes ofantibacterials in the treatment of bacterial meningitis. Despite some of thepotential drawbacks to their use (eg, low CSF penetration, hydrophilicchemistry, and optimal killing only at times when bacteria are in log phasegrowth), they have proved to be eective against a wide variety of meningealpathogens. This is because high systemic doses of b-lactams are generallywell-tolerated and can achieve CSF concentrations that are well above theMIC of sensitive pathogens. For example, high-dose intravenous penicillinG therapy (24 million units daily) can achieve serum penicillin concen-trations of 20 lg/mL, which translate to initial CSF concentrations ofapproximately 1 lg/mL in patients with meningitis. In the past, pneumo-cocci were uniformly susceptible to penicillin, with virtually all isolateshaving an MIC of less than or equal to 0.06 lg/mL, such that penicillin Ghad good microbiologic success in treating patients with pneumococcalmeningitis caused by susceptible strains [1]. Given the emergence ofantimicrobial resistance in meningeal pathogens, however, other agentshave been evaluated for their ecacy in the treatment of bacterialmeningitis.

Newer b-lactams that have been evaluated for the treatment of bacterialmeningitis include the carbapenems and advanced-generation cephalospor-ins. Of the carbapenems, imipenem-cilastatin (Primaxin) was the rststudied. Although imipenem has excellent in vitro activity against manyof the important meningeal pathogens, a high frequency of seizures inchildren receiving imipenem for meningitis [33] has largely precluded its usefor this infection. Meropenem is a carbapenem with less seizure proclivitythan imipenem [34], and is a therapeutic option in some patients withbacterial meningitis. In one in vitro study of the use of various antimicro-bials against penicillin- and cephalosporin-resistant S pneumoniae, merope-nem had comparable bactericidal activity with the combination ofvancomycin plus a third-generation cephalosporin, suggesting a possiblerole for meropenem in resistant pneumococcal meningitis [35]. Penicillin-resistant isolates, however, that have demonstrated reduced susceptibility tocarbapenems have been described. In one in vitro study [36], penicillin-nonsusceptible pneumococci displayed a higher likelihood of resistance tothe carbapenems (47%49%) than to either cefotaxime (15%) or chloram-

586 S.W. Sinner, A.R. Tunkel / Infect Dis Clin N Am 18 (2004) 581602phenicol (31%). In another in vitro study that examined 20 cefotaxime-resistant S pneumoniae isolates [37], 4 were intermediate and 13 wereresistant to meropenem, suggesting that meropenem is not a useful

alternative for treatment of pneumococcal isolates that are highly resistantto penicillin and cephalosporins. Meropenem has also been evaluated in ananimal model of bacterial meningitis, and was found to be more eectivethan ceftriaxone for therapy of penicillin-resistant S pneumoniae (PRSP);more eective than ceftazidime (Fortaz, Tazicef, Tazidime) for meningitiscaused by Pseudomonas aeruginosa; and equivalent to the combination ofampicillin and gentamicin for meningitis caused by L monocytogenes [38].

Ceftazidime is a third-generation cephalosporin with enhanced in vitroactivity against P aeruginosa, and has been shown to be ecacious in ananimal model of Klebsiella pneumoniae meningitis [39]. Cefepime (Max-ipime), a fourth-generation cephalosporin with excellent in vitro activityagainst a variety of meningeal pathogens, including aerobic gram-negativebacilli and PRSP, has also been studied in experimental animal models ofbacterial meningitis. Two experimental studies have demonstrated theecacy of cefepime in treatment of meningitis caused by PRSP. In the rststudy [40], the BKR of cefepime was superior to that of ceftriaxone orvancomycin against a PRSP strain (MIC values: penicillin, 4 lg/mL;ceftriaxone, 0.5 lg/mL; vancomycin, 0.120.25 lg/mL; cefepime, 0.5 lg/mL); statistical signicance was reached in the comparison of cefepime withceftriaxone, even though the in vitro activities of ceftriaxone and cefepimewere similar against the test strain. The in vivo superiority of cefepime maybe secondary to the enhanced CSF penetration of cefepime (20%) versusceftriaxone (9%) [41]. The addition of vancomycin to either cephalosporinincreased the BKR as compared with monotherapy, but combinationtherapy was not synergistic. In a second study [42], a PRSP isolate withinduced uoroquinolone resistance was evaluated (MIC values: penicillin, 4lg/mL; ceftriaxone, 0.5 lg/mL; cefepime, 0.5 lg/mL; vancomycin, 0.120.25lg/mL; trovaoxacin, 4 lg/mL; ciprooxacin [Cipro], 32 lg/mL); the BKRfor cefepime was similar to that for the combination of ceftriaxone andvancomycin, which is the current standard of care for treatment of patientswith meningitis caused by PRSP [1]. The uoroquinolone-resistant PRSPstrain was killed more slowly by ceftriaxone and cefepime in vitro than theparent uoroquinolone-susceptible strain; the in vitro killing rates forvancomycin were not dierent for the two strains. The underlyingmechanism linking uoroquinolone resistance in pneumococci to decreasedin vitro killing rates for the cephalosporins remains unclear.

Glycopeptides

As a result of drug resistance in pneumococci, vancomycin has become animportant antimicrobial agent in the treatment of bacterial meningitis. Oneimportant consideration in the use of vancomycin in experimental menin-

587S.W. Sinner, A.R. Tunkel / Infect Dis Clin N Am 18 (2004) 581602gitis is the decreased and unreliable CSF penetration in the presence ofdexamethasone. In some cases, this decreased penetration caused a decreasedBKR and delay in CSF sterilization compared with animals that did not

receive adjunctive dexamethasone [16,43]. In another study, the addition ofdexamethasone led to signicantly lower CSF vancomycin concentrations,but also signicantly decreased brain edema, without considerable alterationof the BKR [44]. Of additional concern is the description of a strain of Spneumoniae that displayed vancomycin tolerance [45,46], which is the abilityof the microorganism to survive in the presence of antimicrobial agentthrough loss of autolysin activity or triggering; experimental meningitiscaused by this tolerant strain was not responsive to vancomycin therapy. In1998, an evaluation of 138 pneumococcal isolates from patients withmeningitis revealed that 12 isolates were at least moderately vancomycin-tolerant [47]; although the mortality of patients with tolerant isolates wasnot signicantly increased, there was a prolonged length of hospital stay.Vancomycin has retained an important role in the treatment of pneumo-coccal meningitis, and combination therapy with vancomycin and ceftriax-one has been demonstrated to be synergistic against ceftriaxone-resistantpneumococcal meningitis in experimental animals without the adjunctiveuse of dexamethasone [48], and potentially additive in the killing ofceftriaxone-resistant pneumococci in an in vitro model that did includedexamethasone [49]. The combination of vancomycin and gentamicin wasfound to be synergistic against penicillin-resistant pneumococci in vitro andin a rabbit model of meningitis [50]; the BKR of vancomycin plusgentamicin was similar to that of vancomycin plus ceftriaxone. Interestingly,these results were achieved with an average Cpeak of gentamicin of 4.3 lg/mL, against a pneumococcal strain with a gentamicin MIC of 4 lg/mL.

A newer glycopeptide, oritavancin (LY333328), has been examined inexperimental meningitis caused by b-lactamsusceptible S pneumoniae [51].In this study, oritavancin was shown to release lower amounts of lip-oteichoic acid and teichoic acid in vitro than ceftriaxone, although this wasnot found in the in vivo model, possibly because of more rapid bactericidalactivity of oritavancin in vivo. The MIC and MBC of oritavancin ina representative b-lactamsusceptible pneumococcal isolate were very low(0.015 and 0.03 lg/mL, respectively); a 10 mg/kg dose of oritavancinachieved similar BKRs to a 12-hour infusion of ceftriaxone at 10 mg/kgh inrabbits, despite very low (1%5%) CSF penetration of this glycopeptide. Ofnote, the 40 mg/kg dose of oritavancin achieved signicantly higher BKRs,but the safety of this dose in humans has not yet been established.

Rifampin

Rifampin has been studied in experimental models of bacterial menin-gitis, mostly in the setting of combination therapy because rapid emergenceof bacterial resistance precludes the use of rifampin monotherapy in clinical

588 S.W. Sinner, A.R. Tunkel / Infect Dis Clin N Am 18 (2004) 581602settings. In a mouse model of pneumococcal meningitis, rifampin releasedlower amounts of lipoteichoic acid and teichoic acid, and had an improvedmortality rate, compared with ceftriaxone [52]. In an in vitro study of

combination antimicrobials, lipoteichoic acid and teichoic acid release wasfound to be diminished in the setting of pneumococcal exposure toceftriaxone, if the pneumococcal isolate was pre-exposed to rifampin [53];analogous results were demonstrated in the rabbit model of pneumococcalmeningitis, and this was associated with decreased numbers of apoptoticneurons at autopsy in animals in which ceftriaxone therapy was preceded byrifampin. In an earlier animal study of rifampin and ceftriaxone, with orwithout adjunctive dexamethasone, for pneumococcal meningitis, combi-nation therapy led to CSF sterilization, even with the pneumococcal MIC toceftriaxone of 4 lg/mL [16]. Another study revealed that the in vivo killingcurves for rifampin alone, rifampin plus ceftriaxone, and rifampin plusvancomycin were nearly identical for a penicillin- and ceftriaxone-resistantpneumococcal isolate [48]; each of these three rifampin-containing regimenshad killing rates at 24 hours similar to the vancomycin plus ceftriaxonecombination, and had higher BKRs than either ceftriaxone or vancomycinalone. In contrast, an in vitro model of antimicrobial therapy foramoxicillin-resistant S pneumoniae showed that the addition of rifampinto ceftriaxone resulted in at least a 10-fold decrease in killing rates comparedwith ceftriaxone monotherapy [49]. One additional study examined rifampinor vancomycin, or both, with or without dexamethasone, for the treatmentof PRSP and ceftriaxone-resistant S pneumoniae meningitis in rabbits [54].Without the addition of dexamethasone, the three antimicrobial regimensdisplayed similar BKRs, with no additive eect seen with use of thecombination regimen; the addition of rifampin to vancomycin, however,was able to reduce the negative eect of dexamethasone on vancomycinCpeak.

Fluoroquinolones

The uoroquinolones have recently undergone intensive study inexperimental animal models of PRSP meningitis, and for therapy ofmeningitis caused by other meningeal pathogens. Trovaoxacin, forexample, showed great promise for the treatment of bacterial meningitisin view of its broad in vitro activity and CSF pharmacokinetics [55], and wasstudied in experimental animal models of meningitis caused by S pneumo-niae [28], L monocytogenes [56], and other microorganisms. Reports ofserious liver toxicity, however, have limited its clinical usefulness. Sincethen, a number of newer uoroquinolones, particularly those with enhancedin vitro activity against S pneumoniae, have been evaluated for meningitistherapy in various experimental animal models (Table 2).

Clinaoxacin demonstrated excellent potency against S pneumoniae ina rabbit model of pneumococcal meningitis [48], displaying signicantly

589S.W. Sinner, A.R. Tunkel / Infect Dis Clin N Am 18 (2004) 581602more rapid bactericidal activity than either ceftriaxone, vancomycin,meropenem, or cefpirome, even against a more sensitive pneumococcalisolate. In a mouse model, clinaoxacin was found to have equal activity

against both ceftriaxone-susceptible and -resistant pneumococci [57],although ceftriaxone was 11- to 12-fold more potent than clinaoxacinagainst the ceftriaxone-susceptible strain, as measured by the mediancurative dose of the antimicrobial agent used in this study.

Gemioxacin (Factive) was found to be active against pneumococci inthe rabbit meningitis model, with bactericidal activity at a 5 mg/kghinfusion rate being almost equal to that of ceftriaxone at 10 mg/kgh [58].Coadministration of dexamethasone did not decrease the CSF penetrationof gemioxacin, nor did it alter its bactericidal activity. Lower degrees ofinitial lipoteichoic acid and teichoic acid release, as compared withceftriaxone therapy, were not accompanied by attenuation of the inam-matory response, as assessed by the CSF white blood cell count; CSFconcentrations of protein, lactate, or neuron-specic enolase; or by neuronalapoptosis.

Table 2

Summary of data for quinolones as monotherapy for experimental meningitis

Drug

Dose

(mg/kg)

CSF

penetration (%)

BKR

(log10 CFU/mL/h) Isolate studied

Clinaoxacin 20 18a ND PSSP

Gemioxacin 5 9a 0.25 from 012 h PSSP

28b

20c

Moxioxacin 10 44 without

dexamethasoneb0.32 from 012 h PSSP

34 with

dexamethasoneb

20 or 40 78 with meningitisc ND PSSP and PRSP

50 without

meningitisc

540 50 with meningitisc 1.72.4 from 03 h Escherichia coli

20 ND 0.92 from 03 h VTSP

0.50 from 010 h

Gatioxacin 7.530 2738a ND CRSP

15 ND 0.83 from 05 h E coli

15 49c 0.75 from 08 h PRSP

Garenoxacin 1030 1425c 0.81.3 from 05 h CRSP

0.50.7 from 010 h

20 then 10 ND 0.81.5 from 03 h VTSP

0.5 from 010 h

20 then 10 ND 0.95 from 03 h CRSP

0.65 from 010 h

Abbreviations: AUC, area under the concentration curve; BKR, bacterial killing rate; CFU,

colony-forming units; CRSP, ceftriaxone-resistant Streptococcus pneumoniae; CSF, cerebrospi-

nal uid; ND, not determined; PRSP, penicillin-resistant S pneumoniae; PSSP, penicillin-

susceptible S pneumoniae; VTSP, Vancomycin-tolerant S pneumoniae.a

590 S.W. Sinner, A.R. Tunkel / Infect Dis Clin N Am 18 (2004) 581602Determined as (peak CSF concentration/peak serum concentration) 100.b Determined as (CSF/serum concentrations at 12 h after a dose) 100.c Determined as (CSF AUC/serum AUC) 100.Data from references [29,48,57,5966].

Moxioxacin (Avelox) was evaluated in two separate rabbit models forthe treatment of meningitis caused by both penicillin-susceptible S pneumo-niae (PSSP) and PRSP. Against PSSP [59], the bactericidal activity of a 10mg/kgh infusion of moxioxacin, with or without dexamethasone, wasalmost as great as the bactericidal activity of ceftriaxone; doubling the doseof moxioxacin did not result in any measurable increase in the BKR. Aswith many other uoroquinolones, lipoteichoic acid and teichoic acidrelease were not signicantly decreased in comparison with ceftriaxone,but the Cpeak of lipoteichoic acid and teichoic acid were delayed. In ananimal model study that examined the ecacy of moxioxacin in PRSPmeningitis [60], moxioxacin was found to be as eective as ceftriaxone orvancomycin in sterilizing the CSF, and displayed dose-dependent bacteri-cidal activity. In addition, the bactericidal action of moxioxacin was morerapid than that of vancomycin, a nding that had been previouslydemonstrated [48]. Moxioxacin has also been evaluated in an animalmodel of E coli meningitis [61], in which the drug was shown to haveexcellent CSF penetration. As with most other uoroquinolones, theAUC:MBC ratio and the Cpeak:MBC ratio were interrelated in this study,and the time above the MBC was less strongly correlated with the BKR.Moxioxacin was found to be at least as eective as ceftriaxone, and moreeective than meropenem, in eradicating E coli from the CSF.

Gatioxacin has been evaluated in experimental rabbit models of both Spneumoniae and E coli meningitis. In a study of gatioxacin in experimentalcephalosporin-resistant pneumococcal meningitis, several interesting phar-macodynamic properties were noted [29]. The Cpeak:MBC, AUC:MBC, andtime above MBC were all signicantly interrelated (r = 0.94), and eachcorrelated signicantly with the BKR. Divided-dose regimens of gatiox-acin, resulting in greater time over the MBC but lower Cpeak:MBC ratios,were more eective in terms of bacterial clearance, as compared with single-dose regimens; maximal activity was achieved only when drug concen-trations exceeded the MBC for the entire dosing interval. Gatioxacin wasfound to be as eective as the combination of vancomycin plus ceftriaxoneagainst this cephalosporin-resistant isolate. Gatioxacin has also beenstudied alone and in combination with cefepime in experimental PRSPmeningitis [62]. The bactericidal activity of gatioxacin monotherapy in thismodel was excellent, and was signicantly greater than the standardcombination of ceftriaxone plus vancomycin (BKR of 0.75 versus 0.53log10 CFU/mL/h, respectively). The addition of cefepime in this animalmodel increased the BKR slightly (to 0.84 log10 CFU/mL/h), but truesynergy was demonstrated only in vitro. In an experimental model of b-lactamaseproducing E coli meningitis [63], gatioxacin was more rapidlybactericidal than cefotaxime, but similar to meropenem (bacterial killing

591S.W. Sinner, A.R. Tunkel / Infect Dis Clin N Am 18 (2004) 581602rates at 5 hours were 0.83 0.26, 0.46 0.3, and 0.73 0.17 log10 CFU/mL/h, respectively, for gatioxacin, cefotaxime, and meropenem; P= .03for gatioxacin versus cefotaxime).

Garenoxacin (BMS-284756 or T-3811ME), a des-uoroquinolone, is thenewest of the quinolone class to be evaluated for therapy of bacterialmeningitis. One study evaluated both the CSF pharmacokinetics ofgarenoxacin and its ecacy in experimental pneumococcal meningitis [64];CSF penetration and BKRs were similar for garenoxacin compared withother uoroquinolones, and bacterial killing was most closely correlatedwith the CSF AUC:MBC ratio. In contrast to other uoroquinolones[29,65], a potential sub-MBC eect was displayed with garenoxacin, becausebacterial clearance was observed at the 24-hour time period, despitegarenoxacin concentrations below the MBC. A regimen of 20 mg/kg ofgarenoxacin followed by a second dose of 10 mg/kg displayed similarbactericidal activity to the standard combination of vancomycin plusceftriaxone for ceftriaxone-resistant S pneumoniae. Garenoxacin and moxi-oxacin were also tested in an experimental model of vancomycin-tolerant Spneumoniae [66]; BKRs for both uoroquinolones exceeded that for thecombination of vancomycin plus ceftriaxone, although the dierences werenot statistically signicantly.

Specic antimicrobial therapy in patients with meningitis

Use of specic antimicrobial agents in patients with bacterial meningitistakes into account in vitro and animal model data in selecting the optimaltherapeutic regimen. Rather than giving treatment recommendations forevery bacterial pathogen that causes meningitis, the authors concentrate onthe specic microorganisms that have provided challenges for antimicrobialselection based on recent trends in in vitro susceptibility testing.

Streptococcus pneumoniae

Current guidelines from the National Committee for Clinical LaboratoryStandards dene a pneumococcus as susceptible, intermediate, or highlyresistant to penicillin when the isolate has a penicillin MIC of less than orequal to 0.06 lg/mL, 0.1 to 1 lg/mL, or greater than or equal to 2 lg/mL,respectively [1]. Because of the emergence of PRSP and cephalosporin-resistant S pneumoniae [5], penicillin can no longer be considered appropri-ate therapy for suspected or proved pneumococcal meningitis until in vitrosusceptibility testing is performed. The combination of vancomycin andceftriaxone is currently accepted as the standard for empiric therapy ofsuspected pneumococcal meningitis in locations where pneumococcal re-sistance to b-lactams is common. Once the organism and its in vitroantimicrobial susceptibilities are known, therapy may be appropriatelymodied (Table 3). High-dose parenteral penicillin therapy had been

592 S.W. Sinner, A.R. Tunkel / Infect Dis Clin N Am 18 (2004) 581602successfully used for pneumococcal meningitis in the era of predictablepenicillin susceptibility, and is still appropriate therapy for pneumococciwith a penicillin MIC of less than or equal to 0.06 lg/mL. If one chooses to

use a third-generation cephalosporin as alternative therapy for PSSP, thecephalosporin MIC should still be determined, because cases of ceftriaxone-nonsusceptible PSSP have been reported [67].

The third-generation cephalosporins, ceftriaxone and cefotaxime, areconsidered the agents of choice against pneumococci with intermediatesusceptibility to penicillin [1]. Cephalosporin monotherapy may also beadequate for pneumococci that are intermediately susceptible to the third-generation cephalosporins (MIC = 1 lg/mL) [68,69], although failures havebeen reported with high-dose cefotaxime therapy in children with meningitiscaused by cephalosporin-resistant S pneumoniae [70]. Vancomycin also hasa role in therapy of penicillin- or cephalosporin-resistant pneumococcalmeningitis, in combination with a third-generation cephalosporin. Vanco-mycin monotherapy cannot be recommended, because of unpredictable CSFpenetration (particularly in the presence of dexamethasone); poor perfor-mance in experimental animal models; and reports of clinical failures[16,43,44,71]. The combination of vancomycin plus a third-generationcephalosporin, in addition to demonstrating ecacy in animal models[48], has had an additive or synergistic eect in children with bacterialmeningitis [18,72].

For those patients allergic to, intolerant of, or nonresponsive to standardtherapies, there are a number of potential alternatives for the treatment ofpneumococcal meningitis. Meropenem has been studied for the treatment ofmeningitis both in children and adults, with microbiologic and clinicaloutcomes similar to the third-generation cephalosporins [73,74]. There arelittle data on use of meropenem, however, in patients with meningitis caused

Table 3

Specic antimicrobial therapy for pneumococcal meningitis

In vitro susceptibility Standard therapy Alternative therapies

Penicillin MIC 0.06 lg/mL Penicillin G or ampicillin 3rd P-Ceph,chloramphenicol

Penicillin MIC 0.11 lg/mL 3rd P-Ceph Meropenem,cefepime

Penicillin MIC 2 lg/mL Vancomycin plus 3rd P-Cepha 3rd P-Ceph plusuoroquinoloneb

Ceftriaxone MIC 1 lg/mL Vancomycin plus 3rd P-Cepha 3rd P-Ceph plusuoroquinoloneb

Abbreviations: 3rd P-Ceph, Third-generation parenteral cephalosporin, ceftriaxone, or

cefotaxime; MIC, mean inhibitory concentration.a The addition of rifampin might be considered in select patients. See text for details.b Currently available uoroquinolones with data for experimental pneumococcal meningitis

include gemioxacin, moxioxacin, and gatioxacin. Although recommended as alternative

agents in combination with standard therapy, there are currently no clinical data to support this

recommendation.

593S.W. Sinner, A.R. Tunkel / Infect Dis Clin N Am 18 (2004) 581602by penicillin- or cephalosporin-resistant pneumococci, such that routine useof meropenem cannot yet be recommended for patients with meningitiscaused by these resistant strains. Cefepime has been evaluated for empiric

therapy in children with bacterial meningitis [75]. Of the 345 childrenstudied, 26 had S pneumoniae isolated, 17 children received cefepime, and 9received a third-generation cephalosporin. Sixteen of the 17 children whoreceived cefepime cleared their CSF cultures promptly, and one had delayederadication. All pneumococcal isolates in this study were penicillin-susceptible. Clinical data for the addition of rifampin to vancomycin ora third-generation cephalosporin for patients with pneumococcal meningitisare lacking, and there are conicting results of the benets of rifampin in invitro and animal model studies [16,48,49,53,54]. It has been suggested thatrifampin be used for patients in whom the pneumococcal isolate is sensitiveto rifampin in vitro, and in patients who are not responding appropriately tostandard therapy [76].

Although animal model data for use of uoroquinolones in pneumococ-cal meningitis are impressive, there is a paucity of clinical data to warranttheir routine use in patients with bacterial meningitis. In a single trialexamining trovaoxacin as empiric therapy for childhood meningitis, a 94%microbiologic cure rate for trovaoxacin versus a 96% rate for thecomparator drug (ceftriaxone with or without vancomycin) was demon-strated in the 27% of the 311 children who had proved S pneumoniaemeningitis [77]. As more clinical trials are performed, the newer uoroqui-nolones may nd a role in the therapy of pneumococcal meningitis inpatients who are either intolerant of current therapies, or in those withpenicillin- or cephalosporin-resistant strains. Ongoing monitoring of uo-roquinolone susceptibility patterns is crucial, because uoroquinolone-resistant S pneumoniae have been reported [78], including one case of fatalpneumococcal meningitis caused by a levooxacin (Levaquin)-resistantstrain [79].

One important issue is whether adjunctive dexamethasone is appropriatein the treatment of patients with pneumococcal meningitis, given concernsthat dexamethasone may decrease antimicrobial penetration into CSF. Arecently published trial in adult patients with acute bacterial meningitisdemonstrated reduced mortality in patients with pneumococcal meningitiswho received adjunctive dexamethasone [80]. On the basis of these results,the use of adjunctive dexamethasone (given concurrently with or just beforethe rst dose of an antimicrobial agent) is recommended in adults withsuspected or proved pneumococcal meningitis [81]. It should be noted,however, that only 72% of pneumococcal isolates were tested for antimi-crobial susceptibility in this trial and all were susceptible in vitro topenicillin, a nding that is not expected in the United States or in manyother parts of the world. Because corticosteroids can decrease the CSFpenetration of many antimicrobials, particularly vancomycin, there areconcerns regarding the routine use of adjunctive dexamethasone in the

594 S.W. Sinner, A.R. Tunkel / Infect Dis Clin N Am 18 (2004) 581602treatment of patients with penicillin- or cephalosporin-resistant pneumo-coccal meningitis, in which vancomycin becomes a very important compo-nent of therapy. In some experimental rabbit models of meningitis,

decreased CSF penetration of vancomycin has been demonstrated in thepresence of dexamethasone and this has led to a delay in CSF sterilization[16,43]. Decreased CSF penetration of vancomycin was not seen in oneexperimental model [44], however, or in a study in children with bacterialmeningitis who received adjunctive dexamethasone [18]. Because of theseconcerns, a repeat lumbar puncture 36 to 48 hours after initiation ofantimicrobial therapy is recommended for any patient receiving adjunctivedexamethasone who is not improving as expected, or who has a pneumo-coccal isolate for which the cefotaxime or ceftriaxone MIC is 2 lg/mL orgreater [76,82]. In patients receiving dexamethasone for penicillin- orcephalosporin-resistant pneumococcal meningitis, careful observation andfollow-up are critical to determine whether use of dexamethasone may beassociated with adverse clinical outcome in these patients [81].

Neisseria meningitidis

Penicillin G and ampicillin have been the agents of choice for treatingmeningococcal meningitis, but emerging resistance to penicillin throughoutthe world [83] may necessitate a change in this choice of antimicrobialtherapy. One study from Spain [84] documented a rise in penicillin-resistantN meningitidis from 9.1% in 1986 to 71.4% in 1997. High-level penicillinresistance caused by b-lactamase activity, producing MICs as high as 256lg/mL, has also been reported [85], although not in the United States.Many, but not all, patients with penicillin-intermediate-resistant N menin-gitidis have responded well to penicillin therapy. One series from Spain [86],however, showed that reduced susceptibility to penicillin was signicantlyassociated with an increased risk of death or neurologic sequelae. Based onthese data, some authorities recommend a third-generation cephalosporin asthe most appropriate therapy for meningococcal meningitis [1], pendingresults of in vitro susceptibility testing. Resource-deprived areas of theworld have continued to use a single intramuscular dose of chloramphenicolin an oil-suspension for treatment of meningococcal meningitis. One largeobservational study from Sudan [87] demonstrated a 92.8% clinical successrate with chloramphenicol, compared with an 87.1% success rate for a 5-daycourse of parenteral penicillin, during a large outbreak of meningococcalmeningitis. High-level chloramphenicol resistance in meningococci has alsobeen described [88], however, mandating continued surveillance for theseisolates. Other options for the treatment of meningococcal meningitis mayinclude meropenem or the uoroquinolones, pending further study.

Aerobic gram-negative bacilli

595S.W. Sinner, A.R. Tunkel / Infect Dis Clin N Am 18 (2004) 581602The third-generation cephalosporins have revolutionized the treatment ofgram-negative bacillary meningitis, and are now the standard for therapybecause of signicantly improved mortality when compared with previousregimens (usually an aminoglycoside with or without chloramphenicol) [1].

Therapy with ceftazidime resulted in a 79% cure rate in a small series ofpatients with Pseudomonas meningitis, when used with or without anaminoglycoside [89]. A more recent study documented a 94.4% survivalrate for gram-negative meningitis in neonates when treated with thecombination of a third-generation cephalosporin and amikacin (Amikin)[90], although the study was retrospective and uncontrolled. The addition ofintrathecal or intraventricular aminoglycoside to a cephalosporin may stillbe considered in patients not responding to intravenous therapy [1]. Therewas one report of poorer outcomes, however, when intraventriculargentamicin was used in infants [91].

Several other antimicrobial agents have been evaluated for patients withmeningitis caused by aerobic gram-negative bacilli, mostly published as casereports or small case series. These have included cefepime for a cephalospor-inase-hyperproducing Enterobacter aerogenes [92], aztreonam (Azactam) forvarious gram-negative bacilli [93], meropenem for P aeruginosa [94,95],high-dose ampicillin-sulbactam (Unasyn) for a multidrug-resistant Acineto-bacter baumannii [96], colistin sulfomethate sodium (Coly-Mycin M) formultidrug-resistant P aeruginosa and A baumannii [97,98], and the uo-roquinolones for a variety of gram-negative bacilli [99,100]. The primaryarea of use for all of these agents is for therapy of multidrug-resistantorganisms, in which more standard therapies are not an option, or when thepatient is not responding as expected. Choice of empiric antimicrobialtherapy for meningitis caused by aerobic gram-negative bacilli should takeinto account the local susceptibility patterns of these microorganisms. Invitro susceptibility testing must then be performed on isolates to determinethe optimal therapeutic regimen.

Summary

Success in treating bacterial meningitis relies on familiarity with thepharmacology of the commonly used antimicrobial agents and the likelymicrobiologic etiology of the meningitis. Important pharmacologic param-eters relevant to acute bacterial meningitis include CSF penetration, thedrugs activity in purulent CSF, the mode of administration of the drug, andthe intrinsic pharmacodynamic relationships of CSF drug concentrations tobactericidal activity. The key changes in the microbiology of the syndromehave been the dramatic decline in the incidence of meningitis caused by Hinuenzae type b, the emergence of b-lactamresistant pneumococci, and theincreasing problem of resistant gram-negative bacilli in nosocomial menin-gitis. Many of the recent in vitro and animal model studies have focused onnewer antimicrobial agents (especially the uoroquinolones) thatmay be used

596 S.W. Sinner, A.R. Tunkel / Infect Dis Clin N Am 18 (2004) 581602to ght resistant pathogens, although their use in humans has largely beenconned to observational experience, as opposed to large controlled studies.

References

[1] Tunkel AR. Bacterial meningitis. Philadelphia: Lippincott Williams & Wilkins; 2001.

[2] Dacey RG Jr, Sande MA. Eect of probenecid on cerebrospinal uid concentrations

of penicillin and cephalosporin derivatives. Antimicrob Agents Chemother 1974;6:

43741.

[3] Schuchat A, Robinson K, Wenger JD, Harrison LH, Farley M, Reingold AL, et al.

Bacterial meningitis in the United States in 1995. N Engl J Med 1997;337:9706.

[4] Whitney CG, Farley MM, Hadler J, et al. Decline in invasive pneumococcal disease after

introduction of protein-polysaccharide conjugate vaccine. N Engl J Med 2003;348:

173746.

[5] Appelbaum PC. Resistance among Streptococcus pneumoniae: implications for drug

selection. Clin Infect Dis 2002;34:161320.

[6] Chowdhury MH, Tunkel AR. Antibacterial agents in infections of the central nervous

system. Infect Dis Clin North Am 2000;14:391408.

[7] Schmidt T, Tauber MG. Pharmacodynamics of antibiotics in the therapy of meningitis:

infection model observations. J Antimicrob Chemother 1993;31(Suppl D):6170.

[8] Lutsar I, McCracken GH Jr, Friedland IR. Antibiotic pharmacodynamics in

cerebrospinal uid. Clin Infect Dis 1998;27:111729.

[9] Andes DR, Craig WA. Pharmacokinetics and pharmacodynamics of antibiotics in

meningitis. Infect Dis Clin North Am 1999;13:595618.

[10] Spector R. Advance in understanding the pharmacology of agents used to treat bacterial

meningitis. Pharmacology 1990;41:1138.

[11] Dagan R, Velghe L, Rodda JL, Klugman KP. Penetration of meropenem into

cerebrospinal uid of patients with inamed meninges. J Antimicrob Chemother 1994;

34:1759.

[12] Ahmed A, Jafri H, Lutsar I, McCoig CC, Trujillo M, Wubbel L, et al. Pharmacody-

namics of vancomycin for the treatment of experimental penicillin- and cephalosporin-

resistant pneumococcal meningitis. Antimicrob Agents Chemother 1999;43:87681.

[13] Nau R, Kaye K, Sachdeva M, Sande ER, Tauber MG. Rifampin for therapy of

experimental pneumococcal meningitis in rabbits. Antimicrob Agents Chemother 1994;38:

11869.

[14] Gerding DN, Hughes CE, Bamberger DM, Foxworth J, Larson TA. Extravascular

antimicrobial distribution and the respective blood concentrations in humans. In: Lorain

V, editor. Antibiotics in laboratory medicine. 4th edition. Baltimore: Williams & Wilkins;

1996. p. 83599.

[15] Amsden GW. Tables of antimicrobial agent pharmacology. In: Mandell GL, Bennett JE,

Dolin R, editors. Mandell, Douglas and Bennetts principles and practice of infectious

diseases. 5th edition. New York: Churchill Livingstone; 2000. p. 551601.

[16] Paris MM, Hickey SM, Uscher MI, Shelton S, Olsen KD, McCracken GH Jr. Eect of

dexamethasone on therapy of experimental penicillin- and cephalosporin-resistant

pneumococcal meningitis. Antimicrob Agents Chemother 1994;38:13204.

[17] Nau R, Prange HW, Muth P, Mahr G, Menck S, Kolenda H, et al. Passage of cefotaxime

and ceftriaxone into cerebrospinal uid of patients with uninamed meninges.

Antimicrob Agents Chemother 1993;37:151824.

[18] Klugman KP, Friedland IR, Bradley JS. Bactericidal activity against cephalosporin-

resistant Streptococcus pneumoniae in cerebrospinal uid of children with acute bacterial

meningitis. Antimicrob Agents Chemother 1995;39:198892.

[19] Cherubin CE, Eng RH, Norrby R, Modai J, Humbert G, Overturf G. Penetration of

newer cephalosporins into cerebrospinal uid. Rev Infect Dis 1989;11:52648.

[20] Ahmed A, Paris MM, Trujillo M, Hickey SM, Wubbel L, Shelton SL, et al. Once-daily

597S.W. Sinner, A.R. Tunkel / Infect Dis Clin N Am 18 (2004) 581602gentamicin therapy for experimental Escherichia coli meningitis. Antimicrob Agents

Chemother 1997;41:4953.

[21] Nau R, Schmidt T, Kaye K, Froula JL, Tauber MG. Quinolone antibiotics in therapy of

experimental pneumococcal meningitis in rabbits. Antimicrob Agents Chemother 1995;

39:5937.

[22] Nau R, Prange HW, Martell J, Shari S, Kolenda H, Bircher J. Penetration of

ciprooxacin into the cerebrospinal uid of patients with uninamed meninges.

J Antimicrob Chemother 1990;25:96573.

[23] James JK, Palmer SM, Levine DP, RybakMJ. Comparison of conventional dosing versus

continuous-infusion vancomycin therapy for patients with suspected or documented

gram-positive infections. Antimicrob Agents Chemother 1996;40:696700.

[24] Pfausler B, Spiss H, Beer R, Kamp A, Engelhardt K, Schober M, et al. Treatment of

staphylococcal ventriculitis associated with external cerebrospinal uid drains: a pro-

spective randomized trial of intravenous compared with intraventricular vancomycin

therapy. J Neurosurg 2003;98:10404.

[25] Pfausler B, Haring HP, Kamp A, Wissel J, Schober M, Schmutzhard E. Cerebrospinal

uid (CSF) pharmacokinetics of intraventricular vancomycin in patients with staphylo-

coccal ventriculitis associated with external CSF drainage. Clin Infect Dis 1997;25:7335.

[26] Nau R, Prange HW, Menck S, Kolenda S, Visser K, Seydel JK. Penetration of rifampicin

into the cerebrospinal uid of adults with uninamed meninges. J Antimicrob Chemother

1992;29:71924.

[27] Lutsar I, Ahmed A, Friedland IR, Trujillo M, Wubbel L, Olsen K, et al. Pharmacody-

namics and bactericidal activity of ceftriaxone therapy in experimental cephalosporin-

resistant pneumococcal meningitis. Antimicrob Agents Chemother 1997;41:24147.

[28] Kim YS, Liu Q, Chow LL, Tauber MG. Trovaoxacin in treatment of rabbits with

experimental meningitis caused by high-level penicillin-resistant Streptococcus pneumo-

niae. Antimicrob Agents Chemother 1997;41:11869.

[29] Lutsar I, Friedland IR, Wubbel L, McCoig CC, Jafri HS, Ng W, et al. Pharmacody-

namics of gatioxacin in cerebrospinal uid in experimental cephalosporin-resistant

pneumococcal meningitis. Antimicrob Agents Chemother 1998;42:26505.

[30] Craig WA. Pharmacokinetic/pharmacodynamic parameters: rationale for antibacterial

dosing of mice and men. Clin Infect Dis 1998;26:112.

[31] Sande MA, Korzeniowski OM, Allegro GM, Brennan RO, Zak O, Scheld WM.

Intermittent or continuous therapy of experimental meningitis due to Streptococcus

pneumoniae in rabbits: preliminary observations on the postantibiotic eect in vivo. Rev

Infect Dis 1981;3:98109.

[32] Tauber MG, Zak O, Scheld WM, Hengstler B, Sande MA. The postantibiotic eect in the

treatment of experimental meningitis caused by Streptococcus pneumoniae in rabbits.

J Infect Dis 1984;149:57583.

[33] Wong VK, Wright HT Jr, Ross LA, Mason WH, Inderlied CB, Kim KS. Imipenem/

cilastatin treatment of bacterial meningitis in children. Pediatr Infect Dis J 1991;10:1225.

[34] Norrby SR, Gildon KM. Safety prole of meropenem: a review of nearly 5,000 patients

treated with meropenem. Scand J Infect Dis 1999;31:310.

[35] Kim BN, Woo JH, Kim YS, Ryu J, Kim MN, Pai CH. Time-kill studies of antimicrobial

combinations including cefotaxime, ceftriaxone, vancomycin and meropenem against

cephalosporin-resistant Streptococcus pneumoniae. Chemotherapy 2000;46:3038.

[36] Pikis A, Donkersloot JA, Akram S, Keith JM, Campos JM, Rodriguez WJ. Decreased

susceptibility to imipenem among penicillin-resistant Streptococcus pneumoniae.

J Antimicrob Chemother 1997;40:1058.

[37] Buckingham SC, Davis Y, English BK. Pneumococcal susceptibility to meropenem in

a mid-South childrens hospital. South Med J 2002;95:12936.

[38] Nairn K, Shepherd GL, Edwards JR. Ecacy of meropenem in experimental meningitis.

598 S.W. Sinner, A.R. Tunkel / Infect Dis Clin N Am 18 (2004) 581602J Antimicrob Chemother 1995;36(Suppl A):7384.

[39] Mizen L, Woodnutt G, Kernutt I, Catherall EJ. Simulation of human serum

pharmacokinetics of ticarcillin-clavulanic acid and ceftazidime in rabbits, and ecacy

against experimental Klebsiella pneumoniae meningitis. Antimicrob Agents Chemother

1989;33:6939.

[40] Gerber CM, Cottagnoud M, Neftel K, Tauber MG, Cottagnoud P. Evaluation of

cefepime alone and in combination with vancomycin against penicillin-resistant

pneumococci in the rabbit meningitis model and in vitro. J Antimicrob Chemother

2000;45:638.

[41] Modai J. Diusion of 3-quaternary ammonium cephem antibiotics into cerebrospinal

uid of patients with bacterial meningitis. J Chemother 1996;8(Suppl 2):8390.

[42] Cottagnoud P, Acosta F, Cottagnoud M, Tauber MG. Cefepime is ecacious against

penicillin- and quinolone-resistant pneumococci in experimental meningitis. J Antimicrob

Chemother 2002;49:32730.

[43] Cabellos C, Martinez-Lacasa J, Martos A, Tubau F, Fernandez A, Viladrich PF, et al.

Inuence of dexamethasone on ecacy of ceftriaxone and vancomycin therapy in

experimental pneumococcal meningitis. Antimicrob Agents Chemother 1995;39:215860.

[44] Martinez-Lacasa J, Cabellos C, Martos A, Fernandez A, Tubau F, Viladrich PF, et al.

Experimental study of the ecacy of vancomycin, rifampicin and dexamethasone in the

therapy of pneumococcal meningitis. J Antimicrob Chemother 2002;49:50713.

[45] Novak R, Henriques B, Charpentier E, Normark S, Tuomanen E. Emergence of

vancomycin tolerance in Streptococcus pneumoniae. Nature 1999;399:5903.

[46] McCullers JA, English BK, Novak R. Isolation and characterization of vancomycin-

tolerant Streptococcus pneumoniae from the cerebrospinal uid of a patient who

developed recrudescent meningitis. J Infect Dis 2000;181:36973.

[47] Rodriguez CA, Whitney CG, Toumanen EI. Prevalence of vancomycin tolerance in

pneumococci causing meningitis [abstract 1777]. In: Programs and abstracts of the 40th

Interscience Conference of Antimicrobial Agents and Chemotherapy. Toronto: ASM

Press; 2000. p. 105.

[48] Friedland IR, Paris M, Ehrett S, Hickey S, Olsen K, McCracken GH Jr. Evaluation of

antimicrobial regimens for treatment of experimental penicillin- and cephalosporin-

resistant pneumococcal meningitis. Antimicrob Agents Chemother 1993;37:16306.

[49] Fitoussi F, Doit C, Geslin P, Bingen E. Killing activities of trovaoxacin alone and in

combination with beta-lactam agents, rifampin, or vancomycin against Streptococcus

pneumoniae isolates with various susceptibilities to extended-spectrum cephalosporins at

concentrations clinically achievable in cerebrospinal uid. Antimicrob Agents Chemother

1999;43:23725.

[50] Cottagnoud P, Gerber CM, Cottagnoud M, Tauber MG. Gentamicin increases the

ecacy of vancomycin against penicillin-resistant pneumococci in the rabbit meningitis

model. Antimicrob Agents Chemother 2002;46:18890.

[51] Gerber J, Smirnov A, Wellmer A, Ragheb J, Prange J, Schutz E, et al. Activity of

LY333328 in experimental meningitis caused by a Streptococcus pneumoniae strain

susceptible to penicillin. Antimicrob Agents Chemother 2001;45:216972.

[52] Nau R, Wellmer A, Soto A, Koch K, Schneider O, Schmidt H, et al. Rifampin reduces

early mortality in experimental Streptococcus pneumoniae meningitis. J Infect Dis 1999;

179:155760.

[53] Gerber J, Pohl K, Sander V, Bunkowski S, Nau R. Rifampin followed by ceftriaxone for

experimental meningitis decreases lipoteichoic acid concentrations in cerebrospinal uid

and reduces neuronal damage in comparison to ceftriaxone alone. Antimicrob Agents

Chemother 2003;47:13137.

[54] Martinez-Lacasa J, Cabellos C, Martos A, Fernandez A, Tubau F, Viladrich PF, et al.

Experimental study of the ecacy of vancomycin, rifampicin and dexamethasone in the

therapy of pneumococcal meningitis. J Antimicrob Chemother 2002;49:50713.

599S.W. Sinner, A.R. Tunkel / Infect Dis Clin N Am 18 (2004) 581602[55] Cutler NR, Vincent J, Jhee SS, Teng R, Wardle T, Lucas G, et al. Penetration of

trovaoxacin into cerebrospinal uid in humans following intravenous infusion of

alatrooxacin. Antimicrob Agents Chemother 1997;41:1298300.

[56] Michelet C, Leib SL, Bentue-Ferrer D, Tauber MG. Comparative ecacies of antibiotics

in a rat model of meningoencephalitis due to Listeria monocytogenes. Antimicrob Agents

Chemother 1999;43:16516.

[57] Shapiro MA, Donovan KD, Gage JW. Comparative therapeutic ecacy of clinaoxacin

in a pneumococcal meningitis mouse model. J Antimicrob Chemother 2000;45:48992.

[58] Smirnov A, Wellmer A, Gerber J, Maier K, Henne S, Nau R. Gemioxacin is eective

in experimental pneumococcal meningitis. Antimicrob Agents Chemother 2000;44:

76770.

[59] Schmidt H, Dalho A, Stuertz K, Trostdorf F, Chen V, Schneider O, et al. Moxioxacin

in the therapy of experimental pneumococcal meningitis. Antimicrob Agents Chemother

1998;42:1397407.

[60] Ostergaard C, Sorensen TK, Knudsen JD, Frimodt-Moller N. Evaluation of

moxioxacin, a new 8-methoxyquinolone, for treatment of meningitis caused by

a penicillin-resistant pneumococcus in rabbits. Antimicrob Agents Chemother 1998;42:

170612.

[61] Rodriguez-Cerrato V, McCoig CC, Michelow IC, Ghaar F, Jafri HS, Hardy RD, et al.

Pharmacodynamics and bactericidal activity of moxioxacin in experimental Escherichia

coli meningitis. Antimicrob Agents Chemother 2001;45:30927.

[62] Perrig M, Acosta F, Cottagnoud M, Gerber CM, Tauber MG, Cottagnoud P. Ecacy of

gatioxacin alone and in combination with cefepime against penicillin-resistant

Streptococcus pneumoniae in a rabbit meningitis model and in vitro. J Antimicrob

Chemother 2001;47:7014.

[63] Lutsar I, Friedland IR, Jafri HS, Wubbel L, Ng W, Ghaar F, et al. Ecacy of

gatioxacin in experimental Escherichia coli meningitis. Antimicrob Agents Chemother

1999;43:18057.

[64] Rodriguez-Cerrato V, Ghaar F, Saavedra J, Michelow IC, Hardy RD, Iglehart J, et al.

BMS-284756 in experimental cephalosporin-resistant pneumococcal meningitis. Anti-

microb Agents Chemother 2001;45:3098103.

[65] McCoig CC, Wubbel L, Jafri HS, Lutsar I, Bastero R, Olsen K, et al. Pharmacodynamics

of trovaoxacin in experimental pneumococcal meningitis: basis for dosage selection in

children with meningitis. J Antimicrob Chemother 1999;43:6838.

[66] Rodriguez-Cerrato V, McCoig CC, Saavedra J, Barton T, Michelow IC, Hardy RD, et al.

Garenoxacin (BMS-284756) and moxioxacin in experimental meningitis caused by

vancomycin-tolerant pneumococci. Antimicrob Agents Chemother 2003;47:2115.

[67] De Champs C, Constantin JM, Bonnet R, Guelon D, Goldstein F, Sirot J. Decreased

susceptibility to extended-spectrum cephalosporins of a penicillin-susceptible Streptococ-

cus pneumoniae in meningitis. Infection 2000;28:589.

[68] Tan TQ, Schutze GE, Mason EO Jr, Kaplan SL. Antibiotic therapy and acute outcome of

meningitis due to Streptococcus pneumoniae considered intermediately susceptible to

broad-spectrum cephalosporins. Antimicrob Agents Chemother 1994;38:91823.

[69] Viladrich PF, Cabellos C, Pallares R, Tubau F, Martinez-Lacasa J, Linares J, et al. High

doses of cefotaxime in treatment of adult meningitis due to Streptococcus pneumoniae

with decreased susceptibilities to broad-spectrum cephalosporins. Antimicrob Agents

Chemother 1996;40:21820.

[70] Friedland IR, Klugman KP. Cerebrospinal uid bactericidal activity against cephalo-

sporin-resistant Streptococcus pneumoniae in children with meningitis treated with high-

dosage cefotaxime. Antimicrob Agents Chemother 1997;41:188891.

[71] Viladrich PF, Gudiol F, Linares J, Pallares R, Sabate I, Ru G, et al. Evaluation of

vancomycin for therapy of adult pneumococcal meningitis. Antimicrob Agents Chemo-

ther 1991;35:246772.

600 S.W. Sinner, A.R. Tunkel / Infect Dis Clin N Am 18 (2004) 581602[72] Doit C, Barre J, Cohen R, Bonacorsi S, Bourrillon A, Bingen EH. Bactericidal activity

against intermediately cephalosporin-resistant Streptococcus pneumoniae in cerebrospinal

uid of children with bacterial meningitis treated with high doses of cefotaxime and

vancomycin. Antimicrob Agents Chemother 1997;41:20502.

[73] Schmutzhard E, Williams KJ, Vukmirovits G, Chmelik V, Pfausler B, Featherstone A. A

randomised comparison of meropenem with cefotaxime or ceftriaxone for the treatment

of bacterial meningitis in adults. J Antimicrob Chemother 1995;36(Suppl A):8597.

[74] Odio CM, Puig JR, Feris JM, Khan WN, Rodriguez WJ, McCracken GH Jr, et al.

Prospective, randomized, investigator-blinded study of the ecacy and safety of

meropenem vs. cefotaxime therapy in bacterial meningitis in children. Pediatr Infect

Dis J 1999;18:58190.

[75] Saez-Llorens X, ORyan M. Cefepime in the empiric treatment of meningitis in children.

Pediatr Infect Dis J 2001;20:35661.

[76] Kaplan SL, Mason EO Jr. Management of infections due to antibiotic-resistant

Streptococcus pneumoniae. Clin Microbiol Rev 1998;11:62844.

[77] Saez-Llorens X, McCoig C, Feris JM, Vargas SL, Klugman KP, Hussey GD, et al.

Quinolone treatment for pediatric bacterial meningitis: a comparative study of

trovaoxacin and ceftriaxone with or without vancomycin. Pediatr Infect Dis J 2002;21:

1422.

[78] Chen DK, McGeer A, de Azavedo JC, Low DE. Decreased susceptibility of Streptococcus

pneumoniae to uoroquinolones in Canada. N Engl J Med 1999;341:2339.

[79] Wortmann GW, Bennett SP. Fatal meningitis due to levooxacin-resistant Streptococcus

pneumoniae. Clin Infect Dis 1999;29:1599600.

[80] de Gans J, van de Beek D. Dexamethasone in adults with bacterial meningitis. N Engl J

Med 2002;347:154956.

[81] Tunkel AR, Scheld WM. Corticosteroids for everyone with meningitis? N Engl J Med

2002;347:16135.

[82] Kaplan SL. Management of pneumococcal meningitis. Pediatr Infect Dis J 2002;21:

58991.

[83] van Deuren M, Brandtzaeg P, van der Meer JW. Update on meningococcal disease with

emphasis on pathogenesis and clinical management. Clin Microbiol Rev 2000;13:14466.

[84] Latorre C, Gene A, Juncosa T, Munoz C, Gonzalez-Cuevas A. Neisseria meningitidis:

evolution of penicillin resistance and phenotype in a childrens hospital in Barcelona,

Spain. Acta Paediatr 2000;89:6615.

[85] Oppenheim BA. Antibiotic resistance in Neisseria meningitidis. Clin Infect Dis 1997;

24(Suppl 1):S98101.

[86] Cubells CL, Garcia JJG, Martinez JR, Otin CL. Clinical data in children with

meningococcal meningitis in a Spanish hospital. Acta Paediatr 1997;86:269.

[87] Hussein AA, Abdel Rahman SI. Meningococcal meningitis epidemic: a new role for

single-dose oily chloramphenicol. Saudi Med J 2002;23:797801.

[88] Galimand M, Gerbaud G, Guibourdenche M, Riou JY, Courvalin P. High-level

chloramphenicol resistance in Neisseria meningitidis. N Engl J Med 1998;339:86874.

[89] Fong IW, Tomkins KB. Review of Pseudomonas aeruginosa meningitis with special

emphasis on treatment with ceftazidime. Rev Infect Dis 1985;7:60412.

[90] Dellagrammaticas HD, Christodoulou C, Megaloyanni E, Papadimitriou M, Kapetana-

kis J, Kourakis G. Treatment of gram-negative bacterial meningitis in term neonates with

third generation cephalosporins plus amikacin. Biol Neonate 2000;77:13946.

[91] McCracken GH Jr, Mize SG, Threlkeld N. Intraventricular gentamicin therapy in gram-

negative bacillary meningitis of infancy: report of the Second Neonatal Meningitis

Cooperative Study Group. Lancet 1980;1:78791.

[92] Rousseau JM, Soullie B, Villevielle T, Koeck JL. Eciency of cefepime in postoperative

meningitis attributable to Enterobacter aerogenes. J Trauma 2001;50:971.

601S.W. Sinner, A.R. Tunkel / Infect Dis Clin N Am 18 (2004) 581602[93] Kilpatrick M, Girgis N, Farid Z, Bishay E. Aztreonam for treating meningitis caused by

gram-negative rods. Scand J Infect Dis 1991;23:1256.

[94] Donnelly JP, Horrevorts AM, Sauerwein RW, De Pauw BE. High-dose meropenem in

meningitis due to Pseudomonas aeruginosa. Lancet 1992;339:1117.

[95] Chmelik V, Gutvirth J. Meropenem treatment of post-traumatic meningitis due to

Pseudomonas aeruginosa. J Antimicrob Chemother 1993;32:9223.

[96] Cawley MJ, Suh C, Lee S, Ackerman BH. Nontraditional dosing of ampicillin-sulbactam

for multidrug-resistant Acinetobacter baumannii meningitis. Pharmacotherapy 2002;22:

52732.

[97] Levin AS, Barone AA, Penco J, Santos MV, Marinho IS, Arruda EA, et al. Intravenous

colistin as therapy for nosocomial infections caused by multidrug-resistant Pseudomonas

aeruginosa and Acinetobacter baumannii. Clin Infect Dis 1999;28:100811.

[98] Jimenez-Mejias ME, Pichardo-Guerrero C, Marquez-Rivas FJ, Martin-Lozano D,

Prados T, Pachon J. Cerebrospinal uid penetration and pharmacokinetic/pharmacody-

namic parameters of intravenously administered colistin in a case of multidrug-resistant

Acinetobacter baumannii meningitis. Eur J Clin Microbiol Infect Dis 2002;21:2124.

[99] Schonwald S, Beus I, Lisic M, Car V, Gmajnicki B. Ciprooxacin in the treatment of

gram-negative bacillary meningitis. Am J Med 1989;87:248S9S.

[100] Krcmery V Jr, Filka J, Uher J, Kurak H, Sagat T, Tuharsky J, et al. Ciprooxacin in

treatment of nosocomial meningitis in neonates and in infants: report of 12 cases and

review. Diagn Microbiol Infect Dis 1999;35:7580.

602 S.W. Sinner, A.R. Tunkel / Infect Dis Clin N Am 18 (2004) 581602

Antimicrobial agents in the treatment of bacterial meningitisBasic pharmacologic principlesCerebrospinal fluid penetrationAntimicrobial activity in purulent cerebrospinal fluidMode of administrationAntimicrobial pharmacodynamics in cerebrospinal fluid

Experimental data for selected antibacterial agentsbeta-LactamsGlycopeptidesRifampinFluoroquinolones

Specific antimicrobial therapy in patients with meningitisStreptococcus pneumoniaeNeisseria meningitidisAerobic gram-negative bacilli

SummaryReferences