Anda di halaman 1dari 11

See

discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/233393830

Advances in treatment of bacterial meningitis

Article in The Lancet November 2012


DOI: 10.1016/S0140-6736(12)61186-6 Source: PubMed

CITATIONS READS

93 1,143

4 authors, including:

Diederik van de Beek Matthijs C Brouwer


Academisch Medisch Centrum Universiteit van Academisch Medisch Centrum Universiteit van
304 PUBLICATIONS 8,656 CITATIONS 138 PUBLICATIONS 2,092 CITATIONS

SEE PROFILE SEE PROFILE

All in-text references underlined in blue are linked to publications on ResearchGate, Available from: Matthijs C Brouwer
letting you access and read them immediately. Retrieved on: 03 November 2016
Series

Bacterial Meningitis 2
Advances in treatment of bacterial meningitis
Diederik van de Beek, Matthijs C Brouwer, Guy E Thwaites, Allan R Tunkel

Bacterial meningitis kills or maims about a fth of people with the disease. Early antibiotic treatment improves Lancet 2012; 380: 1693702
outcomes, but the eectiveness of widely available antibiotics is threatened by global emergence of multidrug-resistant See Comment page 1623
bacteria. New antibiotics, such as uoroquinolones, could have a role in these circumstances, but clinical data to This is the second in a Series of
support this notion are scarce. Additionally, whether or not adjunctive anti-inammatory therapies (eg, dexamethasone) three papers about bacterial
meningitis
improve outcomes in patients with bacterial meningitis remains controversial; in resource-poor regions, where the
disease burden is highest, dexamethasone is ineective. Other adjunctive therapeutic strategies, such as glycerol, Department of Neurology,
Center for Infection and
paracetamol, and induction of hypothermia, are being tested further. Therefore, bacterial meningitis is a substantial Immunity Amsterdam
and evolving therapeutic challenge. We review this challenge, with a focus on strategies to optimise antibiotic ecacy (CINIMA), Academic Medical
in view of increasingly drug-resistant bacteria, and discuss the role of current and future adjunctive therapies. Center, University of
Amsterdam, Amsterdam,
Netherlands
Introduction therapy. Furthermore, vancomycin is expensive and rarely (Prof D van de Beek MD,
Acute bacterial meningitis is a life-threatening infectious available in low-income countries.7 Alternative agents in M C Brouwer MD); Department
disease, the epidemiology of which has changed sub- these settings include an antipneumococcal uoro- of Infectious Diseases/Centre
stantially since the introduction of conjugate vaccines.13 quinolone (eg, moxioxacin) and rifampicin, although for Clinical Infection and
Diagnostics Research, Kings
Nevertheless, the disease continues to inict a heavy toll, clinical data to support the use of these drugs are scarce. College London, London, UK
including in high-income countries, causing substantial Rifampicin is inexpensive, widely available, penetrates (G E Thwaites MD); Guys and
morbidity and mortality.1,4 Early administration of anti- reasonably well into cerebrospinal uid (CSF), and usually St Thomas NHS Foundation
Trust, London, UK
biotics saves lives, but the global emergence of multidrug- has in-vitro activity against ceftriaxone-resistant pneumo-
(G E Thwaites); and Department
resistant bacteria threatens the eectiveness of many coccal strains.4,8 of Medicine, Monmouth
inexpensive and widely available antibiotics. The role of Listeria monocytogenes is noteworthy because of its Medical Center, Long Branch,
adjunctive anti-inammatory therapies is uncertain, resistance to cephalosporins. Amoxicillin or ampicillin NJ, USA (Prof A R Tunkel MD)
especially in resource-poor settings. For these reasons, are eective against Listeria spp and should be given to Correspondence to
bacterial meningitis is an evolving therapeutic challenge. immunosuppressed patients with meningitis who are at Prof Diederik van de Beek,
Department of Neurology,
In this review, we discuss the various treatment strategies risk of this infection, including pregnant patients and Center of Infection and
available, and draw attention to advances in antibiotic those older than 50 years. Immunity Amsterdam (CINIMA),
and adjunctive therapy. Academic Medical Center,
University of Amsterdam,
Optimisation of the delivery and eectiveness PO Box 22660, 1100DD
Initial empirical antibiotics of antibiotics Amsterdam, Netherlands
Early clinical suspicion of bacterial meningitis and rapid Optimisation of the delivery and eectiveness of anti- d.vandebeek@amc.uva.nl
administration of antibiotics is important to increase biotics are two key therapeutic challenges in bacterial
survival and reduce morbidity. In a prospective study of meningitis. Penetration across the bloodbrain barrier is
156 patients with pneumococcal meningitis admitted to important for successful treatment and depends on the
an intensive-care unit,5 a delay in antibiotic treatment of amount of disruption of the barriers integrity by inam-
longer than 3 h after arrival at the hospital was associated mation, and the size, charge, lipophilicity, protein-
with increased 3-month mortality. binding ability, and interaction with eux pumps of the
Administration of empirical antibiotics for patients antibiotic (table 2).8,9 However, clinical ecacy also
with bacterial meningitis should be based on local epi-
demiology, the patients age, and the presence of specic
underlying diseases or risk factors (table 1).4,6 In Search strategy and selection criteria
geographical regions with Streptococcus pneumoniae We searched the Cochrane Library (The Cochrane Library 2011, issue 1), Medline
(pneumococcal) strains that are resistant to penicillin and (1966 to March, 2012), and Embase (1974 to March, 2012). We used the search terms
cephalosporins (gure), patients older than 1 month with bacterial meningitis or meningitis with the terms therapy or antibiotics or
community-acquired bacterial meningitis should receive antimicrobial or treatment. We mainly selected articles published in the past
vancomycin plus a third-generation cephalosporin (either 5 years, but did not exclude commonly referenced and highly regarded older
cefotaxime or ceftriaxone). The decision of whether to use publications. We also searched the reference lists of articles identified by this search
vancomycin depends on the rate of resistance to third- strategy and selected those that we judged to be relevant. Review articles and book
generation cephalosporins. In areas where the prevalence chapters are cited to provide readers with more details and more references than can
of cephalosporin-resistant S pneumoniae is low (<1% be included in this paper. We modified our reference list on the basis of comments
resistance), a third-generation cephalosporin (either cefo- from peer reviewers.
taxime or ceftriaxone) usually suces as empirical

www.thelancet.com Vol 380 November 10, 2012 1693


Series

Bacterial pathogens Empirical therapy Intravenous dose (dose interval)


Community-acquired meningitis
Age <1 month Streptococcus agalactiae, Escherichia coli, Amoxicillin/ampicillin plus Age <1 week: ampicillin 150 mg/kg per day (8 h); cefotaxime 100150 mg/kg
Listeria monocytogenes cefotaxime, or amoxicillin/ per day (812 h); gentamicin 5 mg/kg per day (12 h)
ampicillin plus an aminoglycoside Age 14 weeks: ampicillin 200 mg/kg per day (68 h); gentamicin 75 mg/kg per
day (8 h); tobramycin 75 mg/kg per day (8 h); amikacin 30 mg/kg per day (8 h);
cefotaxime 150200 mg/kg per day (68 h)
Age 123 months S agalactiae, E coli, S pneumoniae, Neisseria Vancomycin plus a third-generation Vancomycin 60 mg/kg per day (6 h) to achieve serum trough concentrations of
meningitidis cephalosporin (either cefotaxime or 1520 g/mL; cefotaxime 225300 mg/kg per day (68 h); ceftriaxone
ceftriaxone)* 80100 mg/kg per day (1224 h)
Age 250 years S pneumoniae, N meningitidis Vancomycin plus a third-generation Children as above; adults: vancomycin 3060 mg/kg per day (812 h) to achieve
cephalosporin (either cefotaxime or serum trough concentrations of 1520 g/mL; ceftriaxone 4 g per day (12 h);
ceftriaxone)* cefotaxime 812 g per day (46 h); cefepime 6 g per day (8 h); ceftazidime 6 g
per day (8 h); amoxicillin or ampicillin 12 g per day (4 h); penicillin 24 million
units per day (4 h); meropenem 6 g per day (8 h)
Age >50 years S pneumoniae, N meningitidis, Vancomycin plus ampicillin plus a As for adults above
L monocytogenes, aerobic Gram-negative third-generation cephalosporin
bacilli (either cefotaxime or ceftriaxone)
Immunocompromised S pneumoniae, N meningitidis, Vancomycin plus ampicillin plus
state L monocytogenes, Staphylococcus aureus, either cefepime or meropenem
Salmonella spp, aerobic Gram-negative bacilli
(including Pseudomonas aeruginosa)
Recurrent S pneumoniae, N meningitidis, Vancomycin plus a third-generation
Haemophilus inuenzae cephalosporin (either cefotaxime or
ceftriaxone)
Health-care-associated meningitis
Basilar skull fracture S pneumoniae, H inuenzae, group A Vancomycin plus a third-generation
-haemolytic streptococci cephalosporin (either cefotaxime or
ceftriaxone)
Head trauma; Staphylococci (S aureus and Vancomycin plus ceftazidime,
post-neurosurgery coagulase-negative staphylococci), aerobic cefepime, or meropenem
Gram-negative bacilli (including P aeruginosa)

Preferred daily intravenous doses (and dosing intervals) apply to patients with normal renal and hepatic function. In patients with impaired renal function, the loading (initial) dose of the antibiotic is based on
the extracellular uid volume and is not changed in the case of decreased renal function; subsequent doses or dosing intervals need to be changed in patients with impaired renal function. *Add amoxicillin or
ampicillin if meningitis caused by L monocytogenes is also suspected.

Table 1: Empirical antibiotics for presumed bacterial meningitis by demography and risk factor

depends on the antibiotic CSF concentration and its A better understanding of the relation between CSF
bactericidal activity against causative bacteria.8 For concentration and antibiotic eectiveness could improve
example, although -lactam antibiotics penetrate poorly clinical outcomes. Almost 60 years ago, Eagle and
into the CSF, very eective bactericidal concentrations colleagues14 showed that penicillin killed bacteria more
can be achieved by administration of frequent and high eectively when given continuously rather than by bolus
systemic doses, which are generally well tolerated.8 injections; the best predictor of successful treatment was
Toxicity makes dose escalation dicult for the amino- the time that concentrations were maintained above the
glycosides, glycopeptides, and polymyxins; therefore, minimum inhibitory concentration (MIC). Some studies
intrathecal or intraventricular administration of these have investigated whether continuous infusions of these
agents might be needed to reach eective CSF con- antibiotics improve outcomes in patients with bacterial
centrations, although data to support the safety and meningitis.15 A possible benet of continuous cefotaxime
ecacy of this approach are scarce.10 The intrathecal infusion was suggested in a study of 723 African children
route resulted in high CSF aminoglycoside con- with bacterial meningitis randomly assigned to either
centrations in young children with gram-negative cefotaxime boluses or continuous cefotaxime infusion
meningitis,11 but a controlled, non-randomised study of for the rst 24 h of therapy;16 272 (38%) children died, but
intrathecal versus intravenous gentamicin in 117 infants the mode of cefotaxime administration did not signi-
with Gram-negative meningitis did not show clinical cantly change the proportion of children who died or
benet.12 Furthermore, in a randomised controlled trial were severely disabled by hospital discharge. However, a
of intraventricular versus systemic gentamicin, investi- planned subgroup analysis showed that children with
gators reported a substantially higher mortality rate in pneumococcal meningitis given continuous cefotaxime
patients receiving intraventricular gentamicin therapy infusion were signicantly less likely to die or have
(43% vs 13%).13 sequelae than were those given cefotaxime boluses.

1694 www.thelancet.com Vol 380 November 10, 2012


Series

<1%
15%
>5%
no data available

Figure: Global rates of pneumococcal penicillin resistance

Antibiotics for specic organisms (15 mg/kg loading dose, followed by a continuous infusion
Once a bacterial pathogen has been identied on a CSF of 60 mg/kg per day), led to adequate CSF vancomycin
Gram stain, or isolated and in-vitro susceptibility testing concentrations (mean 79 g/mL).20 Although clinical data
done, antibiotic therapy can be modied further for on the ecacy of rifampicin in patients with pneumococcal
optimum treatment (tables 3 and 4). meningitis are scarce, some authorities use this agent in
combination with a third-generation cephalosporin, with
Streptococcus pneumoniae or without vancomycin, in patients with pneumococcal
The treatment of pneumococcal meningitis has changed meningitis caused by strains that are likely to be highly
since the emergence of strains with reduced susceptibility resistant to penicillin or cephalosporins.4
to penicillin (gure); the prevalence of reduced suscept- Once the MIC of penicillin and third-generation cephalo-
ibility ranges from 25% to more than 50% in some US sporins is known, treatment can be modied accordingly
regions and is even higher in many other countries.17 (table 4). The Clinical and Laboratory Standards Institute
Penicillin resistance is a marker of decreased susceptibility has redened the in-vitro susceptibility breakpoints for
to other antibiotics, which could lead to treatment failures pneumococcal isolates from patients with meningitis as
in patients with pneumococcal meningitis.18 In areas either susceptible (MIC 006 g/mL) or resistant (MIC
with cephalosporin resistance, empirical therapy for 012 g/mL) to penicillin;21 for penicillin-resistant strains,
pneumococcal meningitis should consist of vancomycin the therapeutic approach depends on the degree of in-vitro
combined with either cefotaxime or ceftriaxone, pending susceptibility to the third-generation cephalosporins.
results of in-vitro susceptibility testing. Although rates of
pneumococcal meningitis have decreased since the Neisseria meningitidis
introduction of the heptavalent pneumococcal conjugate The current treatment recommendation for meningo-
vaccine, the number of patients with meningitis caused coccal meningitis is penicillin G, amoxicillin, or ampi-
by serotypes not covered by the vaccine, including cillin.3,4,6 However, meningococcal strains with reduced
resistant strains, has increased.19 Non-vaccine serotypes susceptibility to penicillin have been identied in many
are generally more susceptible to antibiotics than are countries. In a Spanish study,22 the investigators reported
vaccine serotypes, except for serotype 19A.19 an increase in the prevalence of meningococcal strains
Adequate doses of vancomycin are important to achieve with reduced susceptibility to penicillin from 91% in
appropriate CSF concentrations, because concomitant use 1986, to 714% in 1997. By contrast, intermediate suscept-
of adjunctive dexamethasone could reduce vancomycin ibility to penicillin (MIC >01 g/mL) has been reported
penetration into CSF. In a study of 14 patients with bac- in 34% of US meningococcal isolates and 2% of isolates
terial meningitis who were receiving adjunctive dexa- in sub-Saharan Africa.23,24 In one study,25 investigators
methasone, administration of intravenous vancomycin recorded an association between reduced susceptibility to

www.thelancet.com Vol 380 November 10, 2012 1695


Series

CSF penetration CSF penetration Comments on use of antibiotic class for meningitis treatment
(CSF:plasma)* in (drug in CSF:plasma)*
uninamed meninges in inamed meninges
-lactams Poor CSF penetration, but high systemic doses are well tolerated and attain CSF concentrations that greatly
exceed the MIC of susceptible bacteria. 40% of cefotaxime vs 90% of ceftriaxone is protein bound. Avoid
imipenem because it could lower the seizure threshold. Continuous infusions could enhance bacterial killing
Benzylpenicillin 002 01
Amoxicillin/ampicillin 001 005
Cefotaxime 01 02
Ceftriaxone 0007 01
Meropenem 01 03
Aminoglycosides Poor CSF penetration and toxicity limits increases in systemic doses. Consider intraventricular/intrathecal
delivery if needed
Gentamicin 001 01
Amikacin No data 01
Glycopeptides Poor CSF penetration and toxicity limits increases in systemic doses. Continuous infusions could enhance
bacterial killing. Limited data for intraventricular/intrathecal delivery
Vancomycin 001 02
Teicoplanin 001 01
Fluoroquinolones Good CSF penetration. Moxioxacin is an alternative agent for the treatment of penicillin-resistant
pneumococcal meningitis
Ciprooxacin 03 04
Moxioxacin 05 08
Levooxacin 07 08
Others
Chloramphenicol 06 07 Excellent CSF penetration, although toxicity concerns limit its use
Rifampicin 02 03 80% protein bound; CSF concentrations greatly exceed MIC of susceptible bacteria
Newer agents
Cefepime 01 02 Eective against penicillin-resistant pneumococcal meningitis
Linezolid 05 07 Case report/series suggest eectiveness for pneumococcal, staphylococcal, and enterococcal meningitis, although
high interindividual variability in CSF pharmacokinetics suggests therapeutic drug measurements could be needed
Daptomycin No data 005 Poor penetration, but CSF concentrations exceed MIC of susceptible bacteria; case reports/series suggest ecacy
in staphylococcal and enterococcal meningitis
Tigecycline No data 05 Good CSF penetration, but concentrations achieved at current standard doses could be insucient to ensure
bacterial killing

CSF=cerebrospinal uid. MIC=minimum inhibitory concentration. *Based on calculated area under the curve (AUC)CSF/AUCplasma when possible, but data are limited for most antibiotics and AUC cannot be
calculated on the basis of single CSF measurements. In these circumstances, CSF penetration is estimated from paired plasma and CSF measurements.

Table 2: Estimates of CSF penetration of antibiotics used for the treatment of bacterial meningitis8,9

in-vitro susceptibility testing are available. High-level


Antibiotic therapy
resistance to chloramphenicol (MIC 64 g/mL) has been
Gram-positive cocci in pairs Vancomycin plus a third-generation cephalosporin (either cefotaxime reported,26 but the incidence is low in most countries.27
or ceftriaxone)
Furthermore, ciprooxacin resistance has been described
Gram-negative cocci in pairs Third-generation cephalosporin (either cefotaxime or ceftriaxone)
in some regions of the USA,28 and has aected
Gram-positive bacilli Amoxicillin/ampicillin* or penicillin G*
recommendations for chemoprophylaxis. During menin-
Gram-positive cocci in chains Amoxicillin/ampicillin or penicillin G*
gococcal meningitis epidemics in resource-poor settings,
Gram-negative bacilli Third-generation cephalosporin one intramuscular injection of long-acting chloram-
*Consider the addition of an aminoglycoside. phenicol is sucient;27 an injection of ceftriaxone is
equally eective.29
Table 3: Recommended antibiotics in patients with community-acquired meningitis by result of
cerebrospinal uid Gram stain
Listeria monocytogenes
Amoxicillin, ampicillin, or penicillin G is the treatment
penicillin and an increased risk of death or neurological of choice for Listeria meningitis.30 Some authorities have
sequelae in children with meningococcal meningitis. recommended the addition of an aminoglycoside
Therefore, patients with meningococcal meningitis because of enhanced in-vitro killing and in-vivo synergy
should be treated empirically with a third-generation in animal models. No study has been done to compare
cephalosporin (cefotaxime or ceftriaxone) until results of amoxicillin or ampicillin alone versus amoxicillin or

1696 www.thelancet.com Vol 380 November 10, 2012


Series

ampicillin plus gentamicin, although retrospective


Recommended therapy Alternative therapies
clinical data suggest that the addition of gentamicin can
reduce mortality.31 By contrast, in a cohort of 118 patients Streptococcus pneumoniae

with listeriosis, the aminoglycoside-treated group Penicillin MIC 006 g/mL Penicillin G or amoxicillin/ Cefotaxime, ceftriaxone, chloramphenicol
ampicillin
had increased rates of kidney injury and mortality.32
Penicillin MIC 012 g/mL
Trimethoprim-sulfamethoxazole is an alternative treat-
Cefotaxime or ceftriaxone Cefotaxime or ceftriaxone Cefepime, meropenem
ment in patients who are allergic to or intolerant of MIC <10 g/mL
penicillin. In a retrospective study,33 treatment with Cefotaxime or ceftriaxone Vancomycin plus either Vancomycin plus moxioxacin
trimethoprim-sulfamethoxazole plus ampicillin was MIC 10 g/mL cefotaxime or ceftriaxone
associated with a lower antibiotic failure rate and fewer Neisseria meningitidis
neurological sequelae than was the combination of Penicillin MIC <01 g/mL Penicillin G or amoxicillin/ Cefotaxime, ceftriaxone, chloramphenicol
ampicillin plus an aminoglycoside. ampicillin
Penicillin MIC 01 g/mL Cefotaxime or ceftriaxone Cefepime, chloramphenicol,
Streptococcus agalactiae uoroquinolone, meropenem
The standard approach to the treatment of meningitis Listeria monocytogenes Amoxicillin/ampicillin or Trimethoprim-sulfamethoxazole
penicillin G
caused by group B streptococci is amoxicillin or ampicillin
Streptococcus agalactiae Amoxicillin/ampicillin or Cefotaxime, ceftriaxone, vancomycin
or penicillin G combined with an aminoglycoside.4 penicillin G
Vancomycin and third-generation cephalosporins are Haemophilus inuenzae
alternatives. Some group B streptococci are less sensitive
-lactamase negative Amoxicillin/ampicillin Cefotaxime, ceftriaxone, cefepime,
to penicillin (MIC 01210 g/mL) than others; the chloramphenicol, aztreonam, uoroquinolone
optimum regimen for these isolates is not clear and the -lactamase positive Cefotaxime or ceftriaxone Cefepime, chloramphenicol, aztreonam,
ecacy of the third-generation cephalosporins in this uoroquinolone
setting has not been established.34 -lactamase negative, Meropenem Fluoroquinolone
ampicillin resistant
Haemophilus inuenzae Staphylococcus aureus
Since the emergence of -lactamase-producing and Meticillin sensitive Nafcillin or oxacillin Vancomycin, linezolid, daptomycin
chloramphenicol-resistant strains of H inuenzae, third- Meticillin resistant|| Vancomycin Trimethoprim-sulfamethoxazole, linezolid,
daptomycin
generation cephalosporins have become standard treat-
Staphylococcus epidermidis|| Vancomycin Linezolid
ment. Third-generation cephalosporins are more eective
than second-generation cephalosporins (eg, cefuroxime)35 Enterobacteriaceae** Cefotaxime or ceftriaxone Aztreonam, uoroquinolone,
trimethoprim-sulfamethoxazole,
and chloramphenicol, even in patients with H inuenzae meropenem, ampicillin
type b meningitis caused by chloramphenicol-sensitive Pseudomonas aeruginosa Ceftazidime or cefepime Aztreonam, meropenem, ciprooxacin
strains.36 The rates of isolation of -lactamase-producing Acinetobacter baumannii** Meropenem Colistin (usually formulated as
strains vary worldwide (15% in the UK, 26% in the USA, colistimethate sodium), polymyxin B
31% in France, and 42% in Spain), with high rates (42%)
MIC=minimum inhibitory concentration. *In the absence of clinical data, recommendations for use of some agents are
for non-typeable strains in the USA.4 Chloramphenicol based on cerebrospinal uid penetration and ecacy in experimental animal models of bacterial meningitis. In-vitro
resistance is also a concern in resource-poor settings, activities of -lactam antibiotic agents against S pneumoniae are predictable within drug classes, but the relation
where the drug is often used as rst-line therapy for between penicillin and cefotaxime-ceftriaxone MICs is not linear. Addition of rifampicin can be considered if the
organism is susceptible, the expected clinical or bacteriological response is delayed, or the cefotaxime/ceftriaxone MIC
patients with bacterial meningitis. In Japan, the
of the pneumococcal isolate is >40 g/mL. No clinical data exist for use of this agent in patients with pneumococcal
prevalence of -lactamase-negative ampicillin-resistant meningitis; recommendation is based on cerebrospinal uid penetration and in-vitro activity against S pneumoniae.
H inuenzae meningitis has increased rapidly from 6% in Addition of an aminoglycoside should be considered; might need intraventricular or intrathecal administration in
2000 to 35% in 2004; many of these strains are also Gram-negative meningitis. ||Addition of rifampicin should be considered. **Choice of a specic agent should be based
on in-vitro susceptibility testing. Might also need to be administered by the intraventricular or intrathecal routes.
resistant to ceftriaxone.37
Table 4: Antibiotics for bacterial meningitis after microorganism identication and in-vitro
Aerobic Gram-negative bacilli susceptibility testing*
The emergence of multidrug-resistant Gram-negative
bacilli is worrying, especially in patients with health-care-
associated bacterial meningitis.9 Resistance to the third- colistin (usually formulated as colistimethate sodium) or
generation and fourth-generation cephalosporins, and polymyxin B should be given intravenously, and might be
carbapenems, has reduced the range of antibiotic options given by the intrathecal or intraventricular route. In one
available. Outbreaks of meningitis caused by Escherichia retrospective study of 51 patients with Acinetobacter
coli strains producing extended-spectrum -lactamases in meningitis,39 all eight patients given a combination of
neonatal wards can be dicult to control.38 In patients with intravenous and intrathecal colistin survived.
Acinetobacter baumannii meningitis, the most commonly
used empirical antibiotic is meropenem with or without Staphylococcus aureus
gentamicin or amikacin given either intraventricularly or S aureus meningitis occurs mainly after neurosurgical
intrathecally.9 If the organism is resistant to carbapenems, procedures or placement of CSF shunts.9 Treatment

www.thelancet.com Vol 380 November 10, 2012 1697


Series

should depend on the local prevalence of meticillin- of H inuenzae meningitis, and either cefepime or
resistant S aureus; antistaphylococcal penicillins are ceftazidime as empirical rst-line treatment in patients
more eective than is vancomycin for the treatment of with post-neurosurgical meningitis.6
severe S aureus disease, but empirical vancomycin can be
used until susceptibility testing results are ready.40 Carbapenems
Of the -lactams, the carbapenems possess the broadest
Duration of antibiotic therapy range of in-vitro activity against Gram-positive and
Antibiotics need enough time to kill all the bacteria and Gram-negative bacteria. Results from studies in human
prevent disease recurrence, but the timescale of this beings suggest that meropenem has better CSF pene-
process varies widely and depends on the causative tration than do imipenem and doripenem.8,46 In four
bacteria, disease severity, and antimicrobial agent used. controlled trials of 448 children and 58 adults, mero-
Uncomplicated meningococcal disease can be treated penem had similar ecacy and safety to cefotaxime or
eectively with one intramuscular dose of ceftriaxone or ceftriaxone, making meropenem the carbapenem of
oily chloramphenicol, both of which are recommended choice in the treatment of bacterial meningitis.8
by WHO in African meningococcal meningitis epi- The emergence of novel -lactamases with direct
demics.29,41 WHO recommends at least 5 days of carbapenem-hydrolysing activity has contributed to an
treatment in non-epidemic situations, in patients increased prevalence of carbapenem-resistant Entero-
younger than 24 months, or if fever, coma, or convulsions bacteriaceae.47
last for longer than 24 h.41 In a meta-analysis of ve
controlled trials investigating shorter (47 days) versus Fluoroquinolones
longer (714 days) antibiotic treatments for bacterial The uoroquinolones gatioxacin and moxioxacin
meningitis, investigators noted no dierence in penetrate the CSF eectively and have greater in-vitro
outcome.41,42 In a controlled trial in 1027 children with activity against Gram-positive bacteria than do their
bacterial meningitis caused by S pneumoniae, earlier counterparts (eg, ciprooxacin). Findings from
H inuenzae, or N meningitidis in Bangladesh, Egypt, experimental meningitis models suggested their ecacy
Malawi, Pakistan, and Vietnam, the investigators in S pneumoniae meningitis, including that caused
reported no dierences in treatment failure or relapse by penicillin-resistant and cephalosporin-resistant
between 5 days versus 10 days of ceftriaxone treatment.43 strains.48,49 Although one controlled trial suggested the
Nevertheless, many authorities in high-income countries uoroquinolone trovaoxacin mesilate to be as eective
recommend at least 7 days of treatment for haemophilus as ceftriaxone, with or without the addition of vanco-
and meningococcal meningitis, and 1014 days of mycin, for paediatric bacterial meningitis,50 no clinical
treatment for pneumococcal meningitis.3,6 trials describe the use of gatioxacin or moxioxacin to
treat bacterial meningitis in human beings. Trovaoxacin
New antibiotics for meningitis and gatioxacin have been associated with serious
The increasing prevalence of meningitis caused by hepatic toxicity and dysglycaemia, respectively, and were
resistant bacteria has led to the consideration of new withdrawn from many markets.51 The IDSA guidelines
antimicrobial agents for therapy, although data describing recommend moxioxacin as an alternative to third-
their role are generally limited to extrapolations from generation cephalosporins plus vancomycin for menin-
experimental animal models and case reports. We will gitis caused by S pneumoniae strains resistant to penicillin
limit our discussion to agents that have been assessed in and third-generation cephalosporins,6 although some
patients with bacterial meningitis. experts recommend that this agent should not be used
alone but rather should be combined with another drug
Cefepime (either vancomycin or a third-generation cephalosporin),
The fourth-generation cephalosporin cefepime has because of the absence of clinical data supporting its use.
broad-range activity and greater stability against
-lactamases, including those often produced by Daptomycin
Pseudomonas aeruginosa, than have agents from the Daptomycin is a cyclic lipopeptide with solely Gram-
preceding generation (eg, ceftriaxone and cefotaxime). positive activity. Although it penetrates the CSF poorly,
Findings from experimental meningitis models and experimental models indicate that CSF bactericidal
some human studies suggested that cefepime could have concentrations are achieved against most susceptible
better CSF activity than ceftriaxone, including against organisms, and daptomycin could have greater bac-
penicillin-resistant S pneumoniae;44,45 however, in two tericidal activity than vancomycin against -lactam-
controlled trials of 345 children with bacterial meningitis, resistant bacteria.52 Human data are limited to case
the investigators reported that cefepime has similar reports that describe the successful use of daptomycin
ecacy to cefotaxime and ceftriaxone.44,45 The Infectious (612 mg/kg once daily), usually combined with rifam-
Diseases Society of America (IDSA) guidelines recom- picin, for meningitis caused by meticillin-resistant
mend cefepime as a second-line agent in the treatment S aureus and vancomycin-resistant Enterococcus spp.53,54

1698 www.thelancet.com Vol 380 November 10, 2012


Series

Linezolid although the Vietnam trial66 did show that dexamethasone


Linezolid is an oxazolidinone that acts only on Gram- increased survival in patients with microbiologically
positive bacteria. It has never been assessed in a conrmed bacterial meningitis.
controlled trial in patients with bacterial meningitis, Investigators of an individual patient data meta-analysis
although some case reports have been published;55 of trials published since 2000 attempted to explain the
linezolid penetrates the CSF well and is associated with dierences between individual trial results.67 In this
cure rates of about 90%. Clinical studies have reported analysis of 2029 patients of all age groups from ve trials,
variable CSF penetration; about 50% of patients given treatment with adjunctive dexamethasone did not
standard doses (600 mg every 12 h) might not achieve signicantly reduce mortality, neurological disability, or
therapeutic CSF concentrations.56 Higher doses and CSF severe hearing loss in patients with bacterial meningitis.
concentration measurements might be needed to There were no signicant treatment eects in any of the
optimise linezolid therapy for bacterial meningitis. prespecied subgroups. A post-hoc analysis suggested that
adjunctive dexamethasone treatment reduced the rate of
Tigecycline hearing loss in survivors (odds ratio [OR] 077, 95% CI
Tigecycline is a glycycline antibiotic that is active against 060099; p=004). Adjunctive dexamethasone treatment
many Gram-positive and Gram-negative bacteria. Data was not associated with an increased risk of adverse events.
about its use in bacterial meningitis are limited mainly Guidelines recommend the use of adjunctive
to case reports describing tigecycline treatment for dexamethasone in patients with suspected or proven
multidrug-resistant Acinetobacter meningitis,57 some of community-acquired bacterial meningitis, but only in
which show that standard intravenous tigecycline doses high-income countries.6,68 Dexamethasone treatment
produce subtherapeutic CSF concentrations.57,58 should be started with or before the rst dose of antibiotics.
It should be given for 4 days at a dose of 06 mg per kg of
Adjunctive dexamethasone therapy bodyweight intravenously every day for children, and
Experimental animal models have shown that outcome 10 mg given intravenously every 6 h for adults. A controlled
from bacterial meningitis is related to the severity of study of 118 children with bacterial meningitis showed
inammation in the subarachnoid space and could 2-day and 4-day regimens of dexamethasone to be similarly
potentially be improved by modulation of the inam- eective.69 However, this study was underpowered, with
matory responseeg, with dexamethasone.59 Initial neurological sequelae or hearing loss occurring in 18%
trials suggested that dexamethasone reduced the risk of and 38% of patients in the 2-day and 4-day regimen
hearing loss in children with H inuenzae type b groups, respectively. Dexamethasone should be stopped if
meningitis.60 Additional data extended the likely benet the patient is discovered not to have bacterial meningitis
to children with S pneumoniae meningitis if dexa- or if the bacterium causing the meningitis is a species
methasone was given with or before the rst dose of an other than H inuenzae or S pneumoniae, although some
antibiotic agent.60 However, subsequent randomised experts advise that adjunctive treatment should be
controlled trials in Malawian and South American continued irrespective of the causative bacterium.3 A
children did not show a benet of dexamethasone.61,62 A recent study showed that adjunctive dexamethasone is
Cochrane meta-analysis published in 201060 showed that widely prescribed for Dutch patients with meningococcal
adjunctive dexamethasone treatment did not reduce meningitis and is not associated with harm.70
mortality in children with bacterial meningitis, but did Adjunctive dexamethasone therapy has been imple-
decrease hearing loss from 20% in the control group to mented on a large scale for patients with pneumococcal
15% in corticosteroid-treated children (risk ratio [RR] meningitis in some settings. In a nationwide obser-
074, 95% CI 062089). None of the included studies vational cohort study in the Netherlands,71 the drug was
investigated children younger than 1 month (neonatal given in 92% of meningitis episodes during 200609.
meningitis), and one randomised, but not placebo- This observational study reported a decrease in mortality
controlled, trial did not show a benet of dexamethasone from 30% to 20% after the introduction of adjunctive
in neonates.63 dexamethasone therapy (absolute risk dierence 10%,
For adults with community-acquired bacterial menin- 95% CI 417; p=0001).
gitis, the results of a European controlled trial showed that Cognitive decits occur often after bacterial meningitis,72
adjunctive dexamethasone, given before or with the rst and studies in animals have suggested that corticosteroids
dose of antibiotic therapy, was associated with a reduced can aggravate learning deciencies.59 A follow-up of the
risk of unfavourable outcome (15% vs 25%; RR 059, 95% European study in adults did not show dierences in
CI 037094) and a reduction in mortality (7% vs 15%, cognitive outcome between patients who received dexa-
048, 024096).64 This benecial eect was most obvious methasone and those who received placebo.73
in adults with pneumococcal meningitis, in whom the A potential rare complication of dexamethasone therapy
mortality rate decreased from 34% to 14%. However, in pneumococcal meningitis is delayed cerebral thrombosis,
randomised controlled trials in Malawi and Vietnam did although a causal relation between this complication and
not show that dexamethasone beneted adult patients,65,66 dexamethasone is dicult to establish.74 Delayed cerebral

www.thelancet.com Vol 380 November 10, 2012 1699


Series

thrombosis can occur 719 days after hospital admission in associated mortality rate means that the threshold at
patients with excellent initial recovery.74 which anticonvulsant therapy is started should be low.81
Studies published so far do not address two important Blood glucose concentrations need to be monitored and
questions: is dexamethasone eective after the rst normoglycaemia achieved.82 The goal of uid manage-
antibiotic dose; and is dexamethasone eective in ment should be to maintain a normovolaemic state; even
patients with septic shock? In experimental pneumo- in patients with severe hyponatraemia, uid maintenance
coccal meningitis, CSF bacterial concentrations at the therapy should be used, rather than uid restriction.3
start of treatment seemed to be a more important factor Monitoring of kidney function is also important,
aecting the antimicrobial-induced inammatory especially in patients who develop septic shock and in
response than the time when dexamethasone therapy those with pre-existing kidney disease. Repeat CSF
was started.75 An individual patient data meta-analysis analysis should only be done in patients whose condition
showed that dexamethasone reduced hearing loss, has not responded clinically after 48 h of appropriate
irrespective of whether the drug was given before or after antimicrobial therapy.
antibiotics.67 In patients with bacterial meningitis and
severe sepsis or septic shock, the survival benet in Novel therapeutic approaches
patients with pneumococcal meningitis who were given Investigators have used experimental meningitis models
adjunctive dexamethasone outweighed the risks asso- to study whether outcomes can be improved by modu-
ciated with high-dose steroids.71,76 lation of damage caused by reactive oxygen species, or by
inhibition of caspase or other mediators in the inam-
Other adjunctive therapies matory, coagulant, or complement cascades.59 Because
Glycerol is a hyperosmolar agent that has been used to bacteriolytic antibiotic regimens temporarily increase the
decrease intracranial pressure. Although glycerol had no release of bacterial components, investigators have used
benecial eect in experimental meningitis models,59 a animal studies to explore the role of non-bacteriolytic
randomised clinical trial in Finland suggested that this antibiotics in the treatment of bacterial meningitis.59 In a
drug might protect against sequelae in children with genetic association study in patients with bacterial
bacterial meningitis.77 A randomised controlled trial of meningitis,83 investigators reported that a common non-
654 children with bacterial meningitis in several South synonymous single nucleotide polymorphism in the gene
American countries showed a signicant decrease in for complement component 5 (C5) was associated with
sequelae.62 However, a randomised controlled trial of unfavourable clinical outcome. Consistent with these
265 Malawian adults with bacterial meningitis showed human data, C5a receptor-decient mice with pneumo-
that adjuvant glycerol was harmful and increased coccal meningitis had decreased brain damage, and
mortality.78 In children, the evidence is insucient to adjuvant treatment with C5-specic monoclonal anti-
justify routine glycerol treatment, but a randomised bodies prevented death in all wild-type mice with
controlled trial of this topic is ongoing in Malawi pneumococcal meningitis.83
(NCT00619203).
Despite some reported benecial eects of monitoring Conclusions and future challenges
and lowering of intracranial pressure in patients with Two main therapeutic strategies exist to improve the
bacterial meningitis,79 when and how it should be outcome of patients with bacterial meningitis: optimisation
undertaken is unclear.80 Randomised studies of various of antimicrobial killing with antibiotics, and reduction of
strategies to lower intracranial pressure have not been the inammatory response in the subarachnoid space with
done. Nevertheless, in patients with impending cerebral adjunctive agents such as dexamethasone. Optimisation of
herniation, monitoring of intracranial pressure and use the antibiotic eect depends on active antibiotic therapy
of osmotic diuretics to lower intracranial pressure could being started early in infection, usually before the causative
be considered, but outcomes are generally poor in this bacterium and its antibiotic susceptibility are known.
critically ill group of patients.80 Determination of which antibiotic agent will be most
Antipyretic treatments are often administered in eective is becoming ever more dicult in the face of
severely ill patients, but their eect on outcome is increasingly drug-resistant bacteria. Clinical data for new
uncertain. In a randomised controlled trial of 723 children antibiotics for bacterial meningitis have not kept pace with
with bacterial meningitis in Luanda, Angola, treatment the rise of resistance, and controlled trials exploring the
with paracetamol for the rst 48 h did not increase role of these agents are urgently needed. Dexamethasone
survival.16 Active cooling leading to hypothermia has is the only accepted adjunctive therapy for the treatment of
benecial eects in animals with pneumococcal menin- patients with bacterial meningitis, but it has shown
gitis.59 The results of a randomised clinical trial of obvious ecacy only in high-income countries. A greater
moderate hypothermia in patients with severe bacterial understanding of disease pathogenesis and patho-
meningitis are eagerly awaited (NCT00774631). physiology could explain why dexamethasone treatment
Patients with bacterial meningitis should be moni- benets some patients with bacterial meningitis, but
tored carefully. Seizures occur frequently and the high not others, and could help to identify new adjunctive

1700 www.thelancet.com Vol 380 November 10, 2012


Series

therapeutic strategies. In the near future, controlled trials 19 Hsu HE, Shutt KA, Moore MR, et al. Eect of pneumococcal
are needed to assess treatment modalities such as conjugate vaccine on pneumococcal meningitis. N Engl J Med 2009;
360: 24456.
induction of hypothermia, intracranial pressure manage- 20 Ricard JD, Wol M, Lacherade JC, et al. Levels of vancomycin in
ment, and specic monoclonal antibodies. However, the cerebrospinal uid of adult patients receiving adjunctive
greatest eect on the burden of illness due to bacterial corticosteroids to treat pneumococcal meningitis: a prospective
multicenter observational study. Clin Infect Dis 2007; 44: 25055.
meningitis is likely to be achieved through widespread use 21 Centers for Disease Control and Prevention. Eects of new
of vaccinations. penicillin susceptibility breakpoints for Streptococcus pneumoniae
United States, 20062007. MMWR Morb Mortal Wkly Rep 2008;
Contributors
57: 135355.
All authors contributed to writing and editing of the review, and all
22 Latorre C, Gene A, Juncosa T, Munoz C, Gonzalez-Cuevas A.
authors approved the nal version.
Neisseria meningitidis: evolution of penicillin resistance and
Conicts of interest phenotype in a childrens hospital in Barcelona, Spain. Acta Paediatr
We declare that we have no conicts of interest. 2000; 89: 66165.
23 Rosenstein NE, Stocker SA, Popovic T, Tenover FC, Perkins BA.
References Antimicrobial resistance of Neisseria meningitidis in the United
1 Thigpen MC, Whitney CG, Messonnier NE, et al. Bacterial States, 1997. The Active Bacterial Core Surveillance (ABCs) Team.
meningitis in the United States, 19982007. N Engl J Med 2011; Clin Infect Dis 2000; 30: 21213.
364: 201625.
24 Hedberg ST, Fredlund H, Nicolas P, Caugant DA, Olcen P,
2 van de Beek D, de Gans J, Spanjaard L, Weisfelt M, Reitsma JB, Unemo M. Antibiotic susceptibility and characteristics of
Vermeulen M. Clinical features and prognostic factors in adults Neisseria meningitidis isolates from the African meningitis belt,
with bacterial meningitis. N Engl J Med 2004; 351: 184959. 2000 to 2006: phenotypic and genotypic perspectives.
3 van de Beek D, de Gans J, Tunkel AR, Wijdicks EF. Antimicrob Agents Chemother 2009; 53: 156166.
Community-acquired bacterial meningitis in adults. N Engl J Med 25 Luaces CC, Garcia Garcia JJ, Roca MJ, Latorre Otin CL. Clinical data
2006; 354: 4453. in children with meningococcal meningitis in a Spanish hospital.
4 Brouwer MC, Tunkel AR, van de Beek D. Epidemiology, diagnosis, Acta Paediatr 1997; 86: 2629.
and antimicrobial treatment of acute bacterial meningitis. 26 Galimand M, Gerbaud G, Guibourdenche M, Riou JY, Courvalin P.
Clin Microbiol Rev 2010; 23: 46792. High-level chloramphenicol resistance in Neisseria meningitidis.
5 Auburtin M, Wol M, Charpentier J, et al. Detrimental role of N Engl J Med 1998; 339: 86874.
delayed antibiotic administration and penicillin-nonsusceptible 27 Stephens DS, Greenwood B, Brandtzaeg P. Epidemic meningitis,
strains in adult intensive care unit patients with pneumococcal meningococcaemia, and Neisseria meningitidis. Lancet 2007;
meningitis: the PNEUMOREA prospective multicenter study. 369: 2196210.
Crit Care Med 2006; 34: 275865.
28 Wu HM, Harcourt BH, Hatcher CP, et al. Emergence of
6 Tunkel AR, Hartman BJ, Kaplan SL, et al. Practice guidelines for ciprooxacin-resistant Neisseria meningitidis in North America.
the management of bacterial meningitis. Clin Infect Dis 2004; N Engl J Med 2009; 360: 88692.
39: 126784.
29 Nathan N, Borel T, Djibo A, et al. Ceftriaxone as eective as
7 Scarborough M, Thwaites GE. The diagnosis and management of long-acting chloramphenicol in short-course treatment of
acute bacterial meningitis in resource-poor settings. Lancet Neurol meningococcal meningitis during epidemics: a randomised
2008; 7: 63748. non-inferiority study. Lancet 2005; 366: 30813.
8 Nau R, Sorgel F, Eiert H. Penetration of drugs through the 30 Brouwer MC, van de Beek D, Heckenberg SG, Spanjaard L,
bloodcerebrospinal uid/bloodbrain barrier for treatment of central de Gans J. Community-acquired Listeria monocytogenes meningitis
nervous system infections. Clin Microbiol Rev 2010; 23: 85883. in adults. Clin Infect Dis 2006; 43: 123338.
9 Andes DR, Craig WA. Pharmacokinetics and pharmacodynamics of 31 Mylonakis E, Hohmann EL, Calderwood SB. Central nervous
antibiotics in meningitis. Infect Dis Clin North Am 1999; 13: 595618. system infection with Listeria monocytogenes. 33 years experience at
10 van de Beek D, Drake JM, Tunkel AR. Nosocomial bacterial a general hospital and review of 776 episodes from the literature.
meningitis. N Engl J Med 2010; 362: 14654. Medicine 1998; 77: 31336.
11 Rahal JJ Jr, Hyams PJ, Simberko MS, Rubinstein E. Combined 32 Mitja O, Pigrau C, Ruiz I, et al. Predictors of mortality and impact
intrathecal and intramuscular gentamicin for gram-negative of aminoglycosides on outcome in listeriosis in a retrospective
meningitis. Pharmacologic study of 21 patients. N Engl J Med 1974; cohort study. J Antimicrob Chemother 2009; 64: 41623.
290: 139498. 33 Merle-Melet M, Dossou-Gbete L, Maurer P, et al. Is
12 McCracken GH Jr, Mize SG. A controlled study of intrathecal amoxicillin-cotrimoxazole the most appropriate antibiotic regimen
antibiotic therapy in Gram-negative enteric meningitis of infancy. for listeria meningoencephalitis? Review of 22 cases and the
Report of the Neonatal Meningitis Cooperative Study Group. literature. J Infect 1996; 33: 7985.
J Pediatr 1976; 89: 6672. 34 Kim KS. Acute bacterial meningitis in infants and children.
13 McCracken GH Jr, Mize SG, Threlkeld N. Intraventricular Lancet Infect Dis 2010; 10: 3242.
gentamicin therapy in Gram-negative bacillary meningitis of 35 Schaad UB, Suter S, Gianella-Borradori A, et al. A comparison of
infancy. Report of the Second Neonatal Meningitis Cooperative ceftriaxone and cefuroxime for the treatment of bacterial meningitis
Study Group. Lancet 1980; 315: 78791. in children. N Engl J Med 1990; 322: 14147.
14 Eagle H, Fleischman R, Levy M. Continuous vs. discontinuous 36 Peltola H, Anttila M, Renkonen OV, on behalf of the Finnish Study
therapy with penicillin; the eect of the interval between injections Group. Randomised comparison of chloramphenicol, ampicillin,
on therapeutic ecacy. N Engl J Med 1953; 248: 48188. cefotaxime, and ceftriaxone for childhood bacterial meningitis.
15 Roberts JA, Webb S, Paterson D, Ho KM, Lipman J. A systematic Lancet 1989; 333: 128187.
review on clinical benets of continuous administration of -lactam 37 Hirakata Y, Ohmori K, Mikuriya M, et al. Antimicrobial activities of
antibiotics. Crit Care Med 2009; 37: 207178. piperacillin-tazobactam against Haemophilus inuenzae isolates,
16 Pelkonen T, Roine I, Cruzeiro ML, Pitkranta A, Kataja M, including -lactamase-negative ampicillin-resistant and
Peltola H. Slow initial -lactam infusion and oral paracetamol to -lactamase-positive amoxicillin-clavulanate-resistant isolates, and
treat childhood bacterial meningitis: a randomised, controlled trial. mutations in their quinolone resistance-determining regions.
Lancet Infect Dis 2011; 11: 61321. Antimicrob Agents Chemother 2009; 53: 422530.
17 Appelbaum PC. Resistance among Streptococcus pneumoniae: 38 Moissenet D, Salauze B, Clermont O, et al. Meningitis caused by
implications for drug selection. Clin Infect Dis 2002; 34: 161320. Escherichia coli producing TEM-52 extended-spectrum -lactamase
18 Klugman KP, Walsh AL, Phiri A, Molyneux EM. Mortality in within an extensive outbreak in a neonatal ward: epidemiological
penicillin-resistant pneumococcal meningitis. Pediatr Infect Dis J investigation and characterization of the strain. J Clin Microbiol
2008; 27: 67172. 2010; 48: 245963.

www.thelancet.com Vol 380 November 10, 2012 1701


Series

39 Rodriguez GA, Blanco A, Asensi V, et al. Multidrug-resistant 61 Molyneux EM, Walsh AL, Forsyth H, et al. Dexamethasone
Acinetobacter meningitis in neurosurgical patients with treatment in childhood bacterial meningitis in Malawi:
intraventricular catheters: assessment of dierent treatments. a randomised controlled trial. Lancet 2002; 360: 21118.
J Antimicrob Chemother 2008; 61: 90813. 62 Peltola H, Roine I, Fernandez J, et al. Adjuvant glycerol and/or
40 DeLeo FR, Otto M, Kreiswirth BN, Chambers HF. Community- dexamethasone to improve the outcomes of childhood bacterial
associated meticillin-resistant Staphylococcus aureus. Lancet 2010; meningitis: a prospective, randomized, double-blind,
375: 155768. placebo-controlled trial. Clin Infect Dis 2007; 45: 127786.
41 WHO. Standardized treatment of bacterial meningitis in Africa in 63 Daoud AS, Batieha A, Al-Sheyyab M, Abuekteish F, Obeidat A,
epidemic and non-epidemic situations. Geneva: World Health Mahafza T. Lack of eectiveness of dexamethasone in neonatal
Organization, 2007. bacterial meningitis. Eur J Pediatr 1999; 158: 23033.
42 Karageorgopoulos DE, Valkimadi PE, Kapaskelis A, Rafailidis PI, 64 de Gans J, van de Beek D. Dexamethasone in adults with bacterial
Falagas ME. Short versus long duration of antibiotic therapy for meningitis. N Engl J Med 2002; 347: 154956.
bacterial meningitis: a meta-analysis of randomised controlled trials 65 Scarborough M, Gordon SB, Whitty CJ, et al. Corticosteroids for
in children. Arch Dis Child 2009; 94: 60714. bacterial meningitis in adults in sub-Saharan Africa. N Engl J Med
43 Molyneux E, Nizami SQ, Saha S, et al. 5 versus 10 days of treatment 2007; 357: 244150.
with ceftriaxone for bacterial meningitis in children: a double-blind 66 Nguyen TH, Tran TH, Thwaites G, et al. Dexamethasone in
randomised equivalence study. Lancet 2011; 377: 183745. Vietnamese adolescents and adults with bacterial meningitis.
44 Miranda J, Tunkel AR. Strategies and new developments in the N Engl J Med 2007; 357: 243140.
management of bacterial meningitis. Infect Dis Clin North Am 2009; 67 van de Beek D, Farrar JJ, de Gans J, et al. Adjunctive
23: 92543. dexamethasone in bacterial meningitis: a meta-analysis of
45 Lodise TP Jr, Nau R, Kinzig M, Jones RN, Drusano GL, Sorgel F. individual patient data. Lancet Neurol 2010; 9: 25463.
Comparison of the probability of target attainment between 68 Visintin C, Mugglestone MA, Fields EJ, Jacklin P, Murphy MS,
ceftriaxone and cefepime in the cerebrospinal uid and serum Pollard AJ. Management of bacterial meningitis and meningococcal
against Streptococcus pneumoniae. Diagn Microbiol Infect Dis 2007; septicaemia in children and young people: summary of NICE
58: 44552. guidance. BMJ 2010; 340: c3209.
46 Nalda-Molina R, Dokoumetzidis A, Charkoftaki G, et al. 69 Syrogiannopoulos GA, Lourida AN, Theodoridou MC, et al.
Pharmacokinetics of doripenem in CSF of patients with Dexamethasone therapy for bacterial meningitis in children:
non-inamed meninges. J Antimicrob Chemother 2012; 67: 172229. 2- versus 4-day regimen. J Infect Dis 1994; 169: 85358.
47 Gupta N, Limbago BM, Patel JB, Kallen AJ. Carbapenem-resistant 70 Heckenberg SG, Brouwer MC, van der Ende A, van de Beek D.
Enterobacteriaceae: epidemiology and prevention. Clin Infect Dis Adjunctive dexamethasone in adults with meningococcal
2011; 53: 6067. meningitis. Neurology 2012; 79: 156369.
48 Ostergaard C, Sorensen TK, Knudsen JD, Frimodt-Moller N. 71 Brouwer MC, Heckenberg SG, de Gans J, Spanjaard L, Reitsma JB,
Evaluation of moxioxacin, a new 8-methoxyquinolone, for treatment van de Beek D. Nationwide implementation of adjunctive
of meningitis caused by a penicillin-resistant pneumococcus in dexamethasone therapy for pneumococcal meningitis. Neurology
rabbits. Antimicrob Agents Chemother 1998; 42: 170612. 2010; 75: 153359.
49 Lutsar I, Friedland IR, Wubbel L, et al. Pharmacodynamics of 72 Hoogman M, van de Beek D, Weisfelt M, de Gans J, Schmand B.
gatioxacin in cerebrospinal uid in experimental Cognitive outcome in adults after bacterial meningitis.
cephalosporin-resistant pneumococcal meningitis. J Neurol Neurosurg Psychiatry 2007; 78: 109296.
Antimicrob Agents Chemother 1998; 42: 265055. 73 Weisfelt M, Hoogman M, van de Beek D, de Gans J, Dreschler WA,
50 Saez-Llorens X, McCoig C, Feris JM, et al. Quinolone treatment for Schmand BA. Dexamethasone and long-term outcome in adults
pediatric bacterial meningitis: a comparative study of trovaoxacin with bacterial meningitis. Ann Neurol 2006; 60: 45668.
and ceftriaxone with or without vancomycin. Pediatr Infect Dis J 74 Schut ES, Brouwer MC, de Gans J, Florquin S, Troost D,
2002; 21: 1422. van de Beek D. Delayed cerebral thrombosis after initial good
51 Park-Wyllie LY, Juurlink DN, Kopp A, et al. Outpatient gatioxacin recovery from pneumococcal meningitis. Neurology 2009;
therapy and dysglycemia in older adults. N Engl J Med 2006; 73: 198895.
354: 135261. 75 Lutsar I, Friedland IR, Jafri HS, et al. Factors inuencing the
52 Egermann U, Stanga Z, Ramin A, et al. Combination of daptomycin anti-inammatory eect of dexamethasone therapy in experimental
plus ceftriaxone is more active than vancomycin plus ceftriaxone in pneumococcal meningitis. J Antimicrob Chemother 2003; 52: 65155.
experimental meningitis after addition of dexamethasone. 76 Moran JL, Graham PL, Rockli S, Bersten AD. Updating the
Antimicrob Agents Chemother 2009; 53: 303033. evidence for the role of corticosteroids in severe sepsis and septic
53 Lee DH, Palermo B, Chowdhury M. Successful treatment of shock: a Bayesian meta-analytic perspective. Crit Care 2010; 14: R134.
methicillin-resistant Staphylococcus aureus meningitis with 77 Kilpi T, Peltola H, Jauhiainen T, Kallio MJ. Oral glycerol and
daptomycin. Clin Infect Dis 2008; 47: 58890. intravenous dexamethasone in preventing neurologic and
54 Le J, Bookstaver PB, Rudisill CN, et al. Treatment of meningitis audiologic sequelae of childhood bacterial meningitis. The Finnish
caused by vancomycin-resistant Enterococcus faecium: high-dose Study Group. Pediatr Infect Dis J 1995; 14: 27078.
and combination daptomycin therapy. Ann Pharmacother 2010; 78 Ajdukiewicz KM, Cartwright KE, Scarborough M, et al. Glycerol
44: 200106. adjuvant therapy in adults with bacterial meningitis in a high HIV
55 Ntziora F, Falagas ME. Linezolid for the treatment of patients with seroprevalence setting in Malawi: a double-blind, randomised
central nervous system infection. Ann Pharmacother 2007; 41: 296308. controlled trial. Lancet Infect Dis 2011; 11: 293300.
56 Yogev R, Damle B, Levy G, Nachman S. Pharmacokinetics and 79 Lindvall P, Ahlm C, Ericsson M, Gothefors L, Naredi S, KoskinenLO.
distribution of linezolid in cerebrospinal uid in children and Reducing intracranial pressure may increase survival among patients
adolescents. Pediatr Infect Dis J 2010; 29: 82730. with bacterial meningitis. Clin Infect Dis 2004; 38: 38490.
57 Kim BN, Peleg AY, Lodise TP, et al. Management of meningitis due 80 van de Beek D, Weisfelt M, de Gans J, Tunkel AR, Wijdicks EF.
to antibiotic-resistant Acinetobacter species. Lancet Infect Dis 2009; Drug Insight: adjunctive therapies in adults with bacterial
9: 24555. meningitis. Nat Clin Pract Neurol 2006; 2: 50416.
58 Ray L, Levasseur K, Nicolau DP, Scheetz MH. Cerebral spinal uid 81 Zoons E, Weisfelt M, de Gans J, et al. Seizures in adults with
penetration of tigecycline in a patient with Acinetobacter baumannii bacterial meningitis. Neurology 2008; 70: 210915.
cerebritis. Ann Pharmacother 2010; 44: 58286. 82 Schut ES, Westendorp WF, de Gans J, et al. Hyperglycemia in
59 Mook-Kanamori BB, Geldho M, van der Poll T, van de Beek D. bacterial meningitis: a prospective cohort study. BMC Infect Dis
Pathogenesis and pathophysiology of pneumococcal meningitis. 2009; 9: 57.
Clin Microbiol Rev 2011; 24: 55791. 83 Woehrl B, Brouwer MC, Murr C, et al. Complement component
60 Brouwer MC, McIntyre P, de Gans J, Prasad K, van de Beek D. 5 contributes to poor disease outcome in humans and mice with
Corticosteroids for acute bacterial meningitis. pneumococcal meningitis. J Clin Invest 2011; 121: 394353.
Cochrane Database Syst Rev 2010; 9: CD004405.

1702 www.thelancet.com Vol 380 November 10, 2012