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

net/publication/316849934

An update on bacterial brain abscess in immunocompetent patients

Article  in  Clinical Microbiology and Infection · May 2017

DOI: 10.1016/j.cmi.2017.05.004

CITATIONS READS

3 332

9 authors, including:

Romain Sonneville Nur Benzonana

Hôpital Bichat - Claude-Bernard (Hôpitaux Universitaires Paris Nord Val de Seine) Lutfi Kirdar Kartal Education and Research Hospital
145 PUBLICATIONS   1,234 CITATIONS    19 PUBLICATIONS   254 CITATIONS   

SEE PROFILE SEE PROFILE

Jean-Marc Tadié Piau Caroline

Centre Hospitalier Universitaire de Rennes Centre Hospitalier Universitaire de Rennes
75 PUBLICATIONS   996 CITATIONS    34 PUBLICATIONS   96 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Occupational bloodborne infections View project

Nocardia View project

All content following this page was uploaded by Romain Sonneville on 19 August 2017.

The user has requested enhancement of the downloaded file.

 Clinical Microbiology and Infection 23 (2017) 614e620

Contents lists available at ScienceDirect

Clinical Microbiology and Infection

journal homepage: www.clinicalmicrobiologyandinfection.com

Review

An update on bacterial brain abscess in immunocompetent patients

6, C. Piau 7,
R. Sonneville 1, R. Ruimy 2, N. Benzonana 3, L. Riffaud 4, A. Carsin 5, J.-M. Tadie
6 6, *
M. Revest , P. Tattevin , the ESCMID Study Group for Infectious Diseases of the Brain
(ESGIB)
1)
Intensive Care Medicine and Infectious Diseases, AP-HP, Bichat Hospital, and UMR1148, LVTS, Sorbonne Paris Cit
e, INSERM/Paris Diderot University, Paris,
France
2) ^te d'Azur University, Nice, France
Microbiology, Archet Hospital, Nice Co
3)
Infectious Diseases and Clinical Microbiology, Dr Lütfi Kırdar Kartal Training and Research Hospital, Istanbul, Turkey
4)
Neurosurgery, Pontchaillou University Hospital, Rennes, France
5)
Radiology, Maison Blanche University Hospital, Reims, France
6)
Infectious Diseases and Intensive Care Unit, Pontchaillou University Hospital, Rennes, France
7)
Microbiology, Pontchaillou University Hospital, Rennes, France

a r t i c l e i n f o a b s t r a c t

Article history: Background: A brain abscess is a focal infection of the brain that begins as a localized area of cerebritis. In
Received 24 March 2017 immunocompetent patients, bacteria are responsible for >95% of brain abscesses, and enter the brain
Received in revised form either through contiguous spread following otitis, sinusitis, neurosurgery, or cranial trauma, or through
29 April 2017
haematogenous dissemination.
Accepted 1 May 2017
Available online 10 May 2017
Aims: To identify recent advances in the field.
Sources: We searched Medline and Embase for articles published during years 2012e2016, with the
Editor: Dr. C. Pulcini keywords ‘brain’ and ‘abscess’.
Content: The triad of headache, fever and focal neurological deficit is complete in ~20% of patients on
Keywords: admission. Brain imaging with contrastdpreferentially magnetic resonance imagingdis the reference
Abscess standard for diagnosis, and should be followed by stereotactic aspiration of at least one lesion, before the
Brain start of any antimicrobials. Efforts should be made for optimal management of brain abscess samples, for
Metronidazole reliable microbiological documentation. Empirical treatment should cover oral streptococci (including
Neurosurgery
milleri group), methicillin-susceptible staphylococci, anaerobes and Enterobacteriaceae. As brain ab-
Stereotactic aspiration
scesses are frequently polymicrobial, de-escalation based on microbiological results is safe only when
Streptococcus milleri
Third-generation cephalosporin aspiration samples have been processed optimally, or when primary diagnosis is endocarditis. Otherwise,
many experts advocate for anaerobes coverage even with no documentation, given the sub-optimal
sensitivity of current techniques. A 6-week combination of third-generation cephalosporin and metro-
nidazole will cure most cases of community-acquired brain abscess in immunocompetent patients.
Implications: Significant advances in brain imaging, minimally invasive neurosurgery, molecular biology
and antibacterial agents have dramatically improved the prognosis of brain abscess in immunocompe-
tent patients over the last decades. R. Sonneville, Clin Microbiol Infect 2017;23:614
© 2017 European Society of Clinical Microbiology and Infectious Diseases. Published by Elsevier Ltd. All
rights reserved.

Introduction Indeed, an unlimited ‘Pubmed search’ performed in March 2017

Among the ‘big three’ central nervous system infections, brain
abscess has always been the last in terms of research priority.
* Corresponding author. P. Tattevin, Infectious Diseases and Intensive Care Unit,
Pontchaillou University Hospital, 2, rue Henri Le Guilloux, 35033 Rennes Cedex,
France.
E-mail address: pierre.tattevin@chu-rennes.fr (P. Tattevin).
retrieved 68 710 papers with the keyword ‘meningitis’, 58 880 with
‘encephalitis’ and only 10 548 with ‘brain abscess’. However, brain
abscess is a fascinating disease, with complex diagnostic challenges
and potentially severe consequences [1e3]. Due to significant
progress in imaging studies, surgical techniques, microbiological
tests and antimicrobial agents use, based on pharmacokinetic/
pharmacodynamic principles, prognosis has improved over the last
decades [4]. Although brain abscess has been the subject of a few
recent reviews and meta-analysis elsewhere [1e3], none took into

http://dx.doi.org/10.1016/j.cmi.2017.05.004
1198-743X/© 2017 European Society of Clinical Microbiology and Infectious Diseases. Published by Elsevier Ltd. All rights reserved.
 R. Sonneville et al. / Clinical Microbiology and Infection 23 (2017) 614e620 615

account the multidisciplinary nature of this disease. As a group of
authors, with a combination of expertise in infectious diseases,
intensive care, microbiology, neurosurgery and imaging, we
reviewed the state-of-the art for brain abscess, in 2017, focusing on
immunocompetent patients, with special emphasis on recent pa-
pers. We searched Medline and Embase for articles published
during years 2012e2016, with the keywords ‘brain’ and ‘abscess’.
Papers that specifically addressed immunocompromised patients,
and case reports, were excluded. Although we applied no age re-
striction, our research retrieved no paper specifically dealing with
brain abscess in children.
Pathogenesis
Brain abscess is a focal infection of the brain, that begins as a
localized area of cerebritis, and develops into a collection of pus
surrounded by a well-vascularized capsule [5]. A landmark exper-
imental model with systematic histological and computed tomog-
raphy (CT) studies in dogs estimated the time course of untreated
streptococcal brain abscesses as follows: early cerebritis (days 1e3),
late cerebritis (days 4e9), early capsule formation (days 10e13) and
late capsule formation (after day 14) [6]. In lesions that were well
encapsulated (14 days and older), five distinct histological zones
were apparent: (a) a well-formed necrotic centre; (b) a peripheral
zone of inflammatory cells, macrophages and fibroblasts; (c) the
dense collagenous capsule; (d) a layer of neovascularization asso-
ciated with continuing cerebritis; and (e) astrogliosis, and cerebral
oedema external to the capsule. Other seminal studies in animals
have described valuable concepts for pathogenesis. First, the brain
is highly susceptible to bacterial infections once the bloodebrain
barrier has been crossed. Indeed, injections of 104 CFUs of Staphy-
lococcus aureus, or 106 CFUs of Escherichia coli, had no consequence
in skin tissues, whereas 102 CFUs of the same organisms were
sufficient to establish an abscess in brain tissues [7]. Second,
although strict anaerobes or microaerophilic bacteria were unable
to establish brain abscesses in a rat model, synergistic infectivity
between strict anaerobes and other bacteria is a key factor in
pathogenesis [8]. Third, capsule tends to be less robust on the
ventricular surface, compared with the cortical surface of a brain
abscess. This may be due to differences in vascularization between
grey and white matter, and may explain why the rupture of a brain
abscess mostly occurs into the ventricular systemdwith severe
consequencesdrather than into the subarachnoid space.
In immunocompetent patients, bacteria are responsible for
>95% of brain abscesses [2]. They enter the brain either through
contiguous spread (e.g. following otitis, mastoiditis, sinusitis,
neurosurgical procedures or cranial trauma) in 40%e50% of cases,
or through haematogenous dissemination in 30%e40% of cases,
especially in the case of infective endocarditis, in patients with
predisposing conditions associated with pulmonary circulation
shunt (e.g. congenital heart disease, pulmonary arteriovenous fis-
tulas, as in hereditary haemorrhagic telangiectasia), or as a conse-
quence of distant infectious foci (e.g. dental infection, pulmonary
abscess) [1,9]. For the management of an individual patient with a
brain abscess, the clinician should aim to identify the pathogenesis,
as this has consequences regarding the main bacteria to be sus-
pected, the investigations to be performed (e.g. echocardiography
when endocarditis is suspected), the therapeutic strategy (e.g.
source control, in selected cases, for otitis or mastoiditis), and
secondary prevention (e.g. reinforced dental hygiene). Table 1 de-
picts the major points to be considered during this important step.
Epidemiology
Limited data are currently available on the epidemiology of
brain abscess [9]. The incidence has been estimated at 0.3e0.9 per

Table 1
Pathogenesis of brain abscess and its consequences for patient management

Predisposing condition Bacteria Comments

Haematogenous (30%e40% of brain abscess in immunocompetent patients)

Infective endocarditis Staphylococcus aureus, oral streptococci, Even with no previously known
HACEK bacteria underlying condition,
echocardiography must be performed
in all patients with bilateral or
cryptogenic brain abscess
Pulmonary circulation shunts Polymicrobial, including streptococci, Careful skin examination to detect skin
(congenital heart disease, anaerobes (Actinomyces sp., Prevotella lesions related to hereditary
arteriovenous fistulas) sp., Bacteroides sp., Fusobacterium sp.) haemorrhagic telangiectasia (Rendu
eOslereWeber syndrome)
Dental infection Polymicrobial, mainly Streptococcus Preferentially affect frontal lobes
milleri group (S. anginosus, Orthopantogram and dentist
S. constellatus, S. intermedius), consultation to be performed
anaerobes (Actinomyces sp., Prevotella May require dental extraction
sp., Bacteroides sp., Fusobacterium sp.) Secondary prevention is key (reinforced
dental hygiene)
Contiguous spread from local infection (40%e50% of brain abscess in immunocompetent patients)
Otitis, mastoiditis, sinusitis Polymicrobial, mainly streptococci, Preferentially affect temporal lobes
Enterobacteriaceae, Streptococcus (otitis, mastoiditis, sphenoid sinusitis),
pneumoniae, anaerobes (Prevotella sp., or frontal lobes (other sinusitis)
Bacteroides sp.), Staphylococcus aureus Systematic ENT examination for
(sinusitis) patients with unilateral brain abscess
Selected cases may require surgery for
source control
Cranial trauma Polymicrobial, mainly Staphylococcus Risk factors: open wound contaminated
aureus, Streptococcus pyogenes, with environmental flora (telluric
anaerobes (Clostridium sp., Actinomyces bacteria), sub-optimal or delayed
sp.) wound care
Neurosurgery Polymicrobial, mainly Staphylococcus Risk factors: foreign devices, including
aureus, coagulase-negative ventricular derivation, bone graft, etc.
staphylococci, Enterobacteriaceae
Cryptogenic brain abscess (10%e20% of brain abscess in immunocompetent patients)

Abbreviations: ENT, ear, nose and throat; HACEK, Haemophilus spp., Aggregatibacter spp., Cardiobacterium spp., Eikenella corrodens and Kingella spp.
616 R. Sonneville et al. / Clinical Microbiology and Infection 23 (2017) 614e620
100 000 inhabitants per year in developed countries [10e12], with
a male-to-female sex ratio of 2: 1 to 3: 1, and a median age of
30e40 years. Of note, a meta-analysis of literature data found that
the distribution of bacterial pathogens was not significantly
different from one continent to another, and has been stable over
the last 60 years [2]. However, a study from Finland suggested that
the proportion of brain abscesses due to dental infections might be
on the rise [10]. On the other hand, the incidence of brain abscesses
may decrease when the general health of the population improves,
according to studies performed in the USA during years 1935e1981
[12], and in South Africa during years 1983e2002 [13].
A systematic review of 9699 cases of brain abscess reported
between 1935 and 2012 found that the most common predisposing
conditions were contiguous foci of infection: otitis or mastoiditis
(33%), sinusitis (10%) and meningitis (6%) [2]. Brain abscess was
related to haematogenous spread in 33% of cases, mostly with
endocarditis (13%), pulmonary infection (8%) or dental infection
(5%). Others were attributed to recent neurosurgery (9%), or cranial
trauma (14%), whereas the source could not be identified in 19% of
cases, so-called ‘cryptogenic brain abscess’.
The risk of brain abscess is estimated at 5%e9% in patients with
hereditary haemorrhagic telangiectasia [14], <5% in patients with
infective endocarditis overall (7% in critically ill patients) [15] and
0.2% after cranial surgery [16]. Brain abscess was encountered in
only 0.5% (14/950) of patients with bacterial meningitis in the
2006e2011 nationwide prospective cohort from the Netherlands
[17], and in 2% (6/252) of patients with neurolisteriosis in the
2009e2013 nationwide prospective cohort from France [18], much
less than the 10% proportion of brain abscess among patients with
neurolisteriosis previously estimated, based on literature review
[19].
Diagnosis
The classical triad of headache, fever and focal neurological
deficit, is present in only ~20% of patients with brain abscess on
admission. Headache is reported in 69% of patients, fever in 53%
and focal neurological deficit in 48% [1,2,4,5,13,20e23]. Other
common neurological symptoms include seizures (25%), and
altered consciousness (43%). The mean duration of symptoms
before diagnosis is 8.3 days [2]. The location of brain abscess im-
pacts the neurological presentation, and may be estimated from a
careful clinical examination, although major discrepancies between
clinical findings and imaging studies are not rare. In a patient with a
brain abscess, the abrupt onset of meningeal signs, associated with
worsening of headache and neurological status, should prompt
clinicians to order urgent brain imaging, to rule out rupture of an
abscess into the ventricular space with ventriculitis, carrying a high
risk of obstructive hydrocephalus, and lethality of 50%e85% [11,24].
Brain imaging is the cornerstone for the diagnosis of brain ab-
scess (Fig. 1). CT with contrast-enhancement typically shows, at
capsule stage, a peripheral rim-enhanced lesion including a hypo-
dense centre (central necrosis), surrounded by a variable hypo-
dense area (oedema). Brain abscesses appear as a single abscess in
most cases with CT (81% in the meta-analysis), and will be mostly
located in frontal (31%) or temporal (28%) lobes [2]. Magnetic
resonance imaging (MRI) should be preferred if readily available, as
it offers many advantages over contrast-enhanced CT, including (a)
better resolution, allowing identification of additional lesions; (b)
earlier detection of lesions at risk of complications; (c) lower
toxicity of the contrast-enhancement agent (gadolinium) [1,9].
Lastly, the combined use of morphological sequences, proton MR
spectroscopy, diffusion-weighted imaging, and the calculation of
the apparent diffusion coefficient has significantly improved the
performance of MRI, allowing sensitivity and specificity of 94% and
95%, respectively, to differentiate abscesses from tumours [25,26],
one of the major challenges with brain imaging until recently.
Pyogenic abscesses appear as a mass with a continuous capsule,
respectively hyperintense and hypointense in T1-weighted and T2-
weighted sequences, with regular rim enhancement after gado-
linium injection (Fig. 2). The necrotic centre is hypointense in T1-
weighted sequence, hyperintense in T2-weighted sequence and
hyperintense in diffusion-weighted imaging. Moreover, it typically
presents with low values of apparent diffusion coefficient, and
peaks corresponding to lipids, lactate and amino acids in MR
spectroscopy [25].
No biological test has any added value for the diagnostic
workout at this stage. Blood leucocyte count and serum C reactive
protein are increased in ~60% of patients, and may be abnormal in
most differential diagnoses. Blood cultures (40e60 mL) should be
collected before initiation of antibacterial treatment, as it was
found to be positive in 28% of patients with brain abscess (135/484)
in a recent meta-analysis [2]. Human immunodeficiency virus
testing should be proposed for all patients with brain abscess, as
cerebral toxoplasmosis may be the first opportunistic infection in
patients with previously undiagnosed human immunodeficiency
virus infection. Both tests are especially relevant in patients with
bilateral brain abscess (Fig. 3).
Neurosurgery
Progress in neurosurgical techniques is one of the key factor
behind the improved prognosis. Stereotactic surgeryda minimally
invasive form of surgical intervention that makes use of a three-
dimensional coordinate system to locate small targets inside the
bodydallows the aspiration of any brain abscess
1 cm with good
tolerability, regardless of its location [27]. Development of neuro-
navigation assistance in routine neurosurgical practice allows the
planning of the optimal trajectory from the point of brain entry to
the abscess, with the aim of avoiding any areas critical for neuro-
logical functions. As a consequence, the neurosurgical approach has
dramatically changed for brain abscesses: (a) total resection
through craniotomy is now rarely considered first, except for pa-
tients with a large multi-lobulated abscess and severe cranial hy-
pertension; (b) stereotactic aspiration by neuronavigation, either
CT- or MRI-guided, is indicated for all patients with undocumented
brain abscess
1 cm; (c) although not based on robust evidence, an
abscess size
2.5 cm is considered as a stand-alone indication for
drainage, even in patients with microbiological documentation (e.g.
through positive blood cultures) [28]; (d) drainage should be dis-
cussed in case of periventricular lesions at high risk of intraven-
tricular rupture, and in case of infections with difficult-to-treat
bacteria or fungi [1,9].
Other indications for neurosurgery include placement of an
external ventricular catheter for drainage and monitoring of
intracranial pressure, in all cases of abscess rupture into the ven-
tricular system, and in selected cases of large abscess with hydro-
cephalus (personal opinion). The importance of adequate sampling
of brain abscess tissues and fluids cannot be overstated: these are
among the most precious surgical samples, with high risk of op-
portunity loss in case of sub-optimal sampling and processing,
given the importance of accurate identification of the bacteria
involved, the polymicrobial context, the fragility of some pathogens
(e.g. strict anaerobes when exposed to room air) [29] and the broad
spectrum of differential diagnosis. Fast processing of surgical
specimen, good anaerobic culture conditions, and the use of blood-
culture bottles and molecular biology techniques when appro-
priate, improve the yield of microbiological diagnosis from brain
abscess samples [30]. In our institution, we build a specific protocol
for brain abscess sampling in the operating room, with a check list
 R. Sonneville et al. / Clinical Microbiology and Infection 23 (2017) 614e620 617

Clinical suspicion of brain abscess

Headache, fever, focal neurologic deficit, seizures, mental status changes, etc.

Search for predisposing condiƟons

ConƟguous spread of local infecƟon
- OƟƟs, sinusiƟs
- Recent neurosurgery
- Recent cranial trauma

CondiƟons at risk of hematogenous spread

- Hereditary hemorrhagic telangiectasia
- EndocardiƟs
- Pulmonary infecƟon
- Dental infecƟon Brain MRI readily available (<24 h)

YES NO

Urgent CT + contrast-enhancement

NO Brain imaging compaƟble with brain abscess

If first imaging was CT, order Brain MRI

If brain MRI not suggesƟve, look for other diagnosis YES

Collect 40-60 mL of blood cultures before iniƟaƟon of anƟbacterial treatment

HIV test
StereotacƟc aspiraƟon
- All abscess > 2.5 cm
- If no abscess > 2.5 cm, at least one of those > 1 cm (the most accessible)
- Microbiology tests: rouƟne (aero + anaerobes) +/- molecular biology (PCR 16S
rDNA, if negaƟve) + others if risk factors (tuberculosis, fungal, etc.)

Empirical anƟbacterial treatment

Third generaƟon cephalosporin (cefotaxime or ceŌriaxone) + metronidazole

Fig. 1. Algorithm for the management of patients suspected of brain abscess.

to ensure that the risk of non-contributing samples due to pre-
analytical adverse events will be kept to a minimum (Table 2).
Microbiology
Although significant progress has already been achieved, this
field will most likely improve dramatically during the next few
years (personal opinion). According to a recent systematic review
and meta-analysis, of the 6663 patients with brain abscess who had
samples submitted for cultures during years 1935e2012, 4543
(68%) yielded at least one potential pathogen, of which 902 (23%)
were polymicrobial. Streptococci predominated (n ¼ 2000; 34%),
mostly oral streptococci, including the milleri group, whereas
Streptococcus pneumoniae was isolated from only 2.4% of patients.
Staphylococci were second (n ¼ 1076; 18%), of which 84% were
Staphylococcus aureus. Gram-negative bacilli came third (n ¼ 861;
15%), mostly Enterobacteriaceae. However, these figures are prob-
ably not representative of the actual distribution of pathogens
involved in brain abscesses, for the following reasons: (a) a sub-
stantial proportion of these samples were obtained when patients
were already on antibacterial treatment, which dramatically de-
creases the sensitivity of cultures, especially for the most suscep-
tible bacteria; (b) limited information is available on sample
processing and culture techniques in these series, although it has a
618 R. Sonneville et al. / Clinical Microbiology and Infection 23 (2017) 614e620

Fig. 2. Typical brain abscess on magnetic resonance imaging. (a) Axial T1-weighted image: abscess with a thin hyperintense capsule (white arrow), hypointense central necrosis,
and a mass effect in the adjacent structures; (b) axial T2-weighted image: the same abscess with a thin hypointense capsule (white arrow), hyperintense central necrosis, and a
peripheral hyperintense oedema; (c) axial diffusion-weighted image: the centre of the abscess is hyperintense; (d) axial image, cartography of apparent diffusion coefficient (ADC):
the central component appears hypointense on ADC map; (e) axial T1-weighted image after gadolinium injection: the peripheral capsule (right arrow) is enhanced.

huge impact on culture yield, especially for strict anaerobes such as
Prevotella sp., Bacteroides sp. and Fusobacterium sp. [31e33].
The advent of molecular biology, and metagenome sequencing
have demonstrated the gap between bacterial species identified
through conventional cultures, and bacterial genetic material that
can be amplified from brain abscess samples. 16S rRNA-based
amplification, cloning and high-throughput sequencing have
dramatically increased the number of identified agents of brain
abscesses [29]. The systematic use of multiple 16S ribosomal DNA
sequencing on samples obtained from 71 patients with brain ab-
scess increased the proportion of patients with documentation
(from 66% to 83%), with the identification of 186 strains, compared
with only 58 with conventional cultures [34,35]. However, the
clinical significance of these large numbers of bacteria (as many as
16 distinct species in a single brain abscess), identified only through
metagenomics analysis, but not by conventional cultures, remains
to be proven. For example, Mycoplasma sp. were among the com-
monest bacterial species identified through molecular biology. Still,
>90% of brain abscesses resolve with a 6-week course of third-
generation cephalosporin combined with metronidazole [4],
although this regimen has no activity on Mycoplasma sp.

Antibacterial treatment

Although the level of evidence supporting the choice of anti-

Fig. 3. Contrast-enhanced computed tomography: bilateral brain abscesses with
hypodense centre (central necrosis), limited by a ring-shaped, contrast-enhanced
capsule, surrounded by a variable hypodense area (oedema).
bacterial regimens is limited, in the absence of randomized trials,
there is a global consensus for the following points: (a) antibacterial
treatment should be initiated as soon as possible, immediately after
stereotactic aspiration of one abscess has been performed, and
blood cultures have been sampled; (b) empirical treatment should
cover streptococci, staphylococci, strict anaerobes, and Enterobac-
teriaceae, pending results of aspiration samples and blood cultures;
(c) for community-acquired brain abscess, a combination of third-
generation cephalosporin (for adults, intravenous cefotaxime
8e12 g/day or ceftriaxone 4 g/day), and metronidazole (1.5 g/day),
would be recommended in most settings (personal opinion, and
[1,9]). One of the major controversies in this field is the opportunity
for de-escalation once microbiological results are available. Indeed,
 R. Sonneville et al. / Clinical Microbiology and Infection 23 (2017) 614e620 619

Table 2 patients with persistent fever, or any neurological event (personal

Suggested protocol for the processing of surgical samples (stereotactic aspiration or opinion).
craniotomy), based on authors' opinion
When antibacterial treatment can be safely targeted to the
Laboratory Samples Targets, comments pathogen(s) identified through cultures, basic pharmacokinetic/
Bacteriology* 10 mL to be directly Streptococci, staphylococci, pharmacodynamic principles apply, including (a) use drugs with
inoculated on blood- anaerobes (blood-culture significant diffusion through the bloodebrain barrier, although
culture bottles (aerobic/ bottles increases the yield, brain diffusion cannot be directly extrapolated from concentrations
anaerobic) especially if previous use of
measured in cerebrospinal fluid [37]; (b) a preferential use of
antibacterials)
Two separate dry tubes Gram staining agents that remain active in acid environments, which is the rule
At least 5 mL in each Cultures on routine media for bacterial abscess. For example, aminoglycosides will be avoided,
tube Molecular biology to be as probably not effective within abscess fluid and tissues, whereas
performed if cultures sterile
metronidazole (not affected by low pH) will be the preferential
after 48 h
Tests for mycobacteria if
drug for anaerobes (personal opinion). Another controversy relies
risk factors for tuberculosis in the preferential use of parenteral drugs. We routinely switch to
Histology One dry tube for fluid Differential diagnosis oral drugs, even early in the process, provided the bacteria targeted
sample (cancer) are susceptible to an effective oral drug, the drug is well absorbed,
If tissue, to be sent in
and adherence is not an issue. Table 3 depicts first-line antibacterial
sterile water
Mycology, Dry tube Only if treatment for the main pathogens isolated from brain abscess in
parasitology immunocompromised immunocompetent patients.
(human immunodeficiency
virus immunuppressive
agents, organ transplant, Symptomatic measures
malignant haemopathy)
Specific tests for Simple measures may be of importance in patients with intra-
Toxoplasma, invasive
cranial hypertension and/or altered mental status. Although not
mycosis (Aspergillus spp.)
Basic rules based on robust evidence, early intensive care unit admission
To be sent within 1 h to the microbiology laboratory, for immediate processing, should be considered in patients with co-morbidities, severe
24/24. neurological presentation (i.e. Glasgow Coma Scale 13 and/or
Two sets of blood cultures should be obtained, before and just after surgery, seizures), and/or cerebral lesion(s) with oedema and significant
optimally before antibiotic administration.
mass effect. A short course of adjunctive steroids may be proposed
Patient characteristics and clinical suspicions to be communicated to the
microbiologist. in case of warning signs of brain herniation. Primary seizure pro-
* phylaxis is not recommended (personal opinion).
The volume of samples to be inoculated on blood-culture bottles depends on the
volume that can be obtained, hence on abscess size.
Unmet needs and perspectives

given the large proportion of polymicrobial brain abscesses (>30%), Clinical research on brain abscess remains limited, in 2017, so
and the high risk of missing a co-pathogen in this context, many that unmet needs are quite large. In our opinion, the priorities
experts would advocate that, except for brain abscess associated
with endocarditis, it is reasonable to keep an adequate coverage for
anaerobes throughout the duration of treatment, even when ste- Table 3
reotactic aspiration found no anaerobes. Given this concern that First-line antibacterial treatment for main pathogens isolated from brain abscess in
immunocompetent patients
microbiological results might not identify all pathogens involved,
any additional suspicion that samples were sub-optimal (e.g. Bacteria Antibacterial treatment Comments
limited volume, samples obtained when antibacterial treatment Staphylococcus High doses i.v. oxacillin Brain diffusion of
was already initiated, long delay between sampling and analysis), aureus (150 mg/kg/day) cloxacillin and
should prompt clinicians to continue empirical treatment for the cefazolin probably
lower
whole 6-week duration (personal opinion). However, the potential
Streptococcus sp. High doses i.v. amoxicillin or To be adjusted
neurotoxicity of 6 weeks of metronidazole should not be under- ampicillin (200 mg/kg/day), or based on MIC
estimated. Hence, when anaerobic infections have been reliably penicillin G (300e400 000 UI/
ruled out, metronidazole may be discontinued. kg/day), or ceftriaxone
Most experts recommend 6-week duration for bacterial brain (2 g twice daily) or cefotaxime
(200 mg/kg/day)
abscess in immunocompetent patients, based on their clinical Strict anaerobes Metronidazole, 500 mg/kg i.v. or oral
experience, and practices for abscesses in other locations, but no (Prevotella sp., three times per day theoretically
clinical study specifically addressed treatment duration for brain Bacteroides sp., similar, in terms of
abscess. Likewise, indications for follow-up imaging studies still Fusobacterium sp.) brain and blood
concentrations
mostly rely on expert opinion, although a recent study found that
Actinomyces spp. High doses i.v. amoxicillin or
bacterial brain abscesses with favourable outcome decrease by 10% ampicillin (200 mg/kg/day), or
per week on MRI follow-up imaging studies [36]. In the absence of penicillin G (300e400 000 UI/
any systematic study in the field, it seems reasonable to perform kg/day)
systematic brain imaging at week 6 in patients with no clinical Enterobacteriaceae Ceftriaxone (2 g twice daily) or
cefotaxime (200 mg/kg/day)
complications. Some experts would continue antibacterial treat- HACEK bacteria Ceftriaxone (2 g twice daily) or Drug susceptibility
ment in patients with residual abscess cavity at 6 weeks, but we are cefotaxime (200 mg/kg/day) testing poorly
not aware of any study supporting these practices. Additional im- reliable
aging studies will be performed in the case of brain lesions at risk of Abbreviations: HACEK, Haemophilus parainfluenzae, Aggregatibacter spp., Car-
complications (e.g. periventricular location, size >2.5 cm), or in diobacterium spp., Eikenella corrodens and Kingella spp.; i.v., intravenous.
620 R. Sonneville et al. / Clinical Microbiology and Infection 23 (2017) 614e620
would be: (a) to gather a multidisciplinary, international group of
experts with the aim of designing practical guidelines for the
management of patients suspected of brain abscess; (b) to initiate
international prospective cohort studies with standardized collec-
tion of data, including 1-year follow up, to better characterize
clinical presentations, microbiology, medical and surgical treat-
ment approaches, prognostic factors and sequelae. These necessary
first steps would pave the way for interventional studies, including
randomized trials, to better define optimal surgical and medical
treatment, both poorly evidence-based, in 2017.
Conclusion
In conclusion, brain abscess is a fascinating disease, with a
prognosis that dramatically improved over the last decades due to
advances in brain imaging, minimally invasive neurosurgery and
better use of old and more recent antibacterial agents. In 2017,
when diagnosis is made ‘in time’, and basic rules of management
are applied, the rate of cure should be >90%.
Transparency declaration
All authors have stated that there are no conflicts of interest. No
external funding was received for this work.
References
[1] Brouwer MC, Tunkel AR, McKhann 2nd GM, van de Beek D. Brain abscess.
N Engl J Med 2014;371:447e56.
[2] Brouwer MC, Coutinho JM, van de Beek D. Clinical characteristics and outcome
of brain abscess: systematic review and meta-analysis. Neurology 2014;82:
806e13.
[3] Saez-Llorens X, Nieto-Guevara J. Brain abscess. Handb Clin Neurol 2013;112:
1127e34.
[4] Tattevin P, Bruneel F, Clair B, Lellouche F, de Broucker T, Chevret S, et al.
Bacterial brain abscesses: a retrospective study of 94 patients admitted to an
intensive care unit (1980 to 1999). Am J Med 2003;115:143e6.
[5] Mathisen GE, Johnson JP. Brain abscess. Clin Infect Dis 1997;25:763e79. quiz
80e1.
[6] Britt RH, Enzmann DR, Yeager AS. Neuropathological and computerized
tomographic findings in experimental brain abscess. J Neurosurg 1981;55:
590e603.
[7] Mendes M, Moore P, Wheeler CB, Winn HR, Rodeheaver G. Susceptibility of
brain and skin to bacterial challenge. J Neurosurg 1980;52:772e5.
[8] Onderdonk AB, Kasper DL, Mansheim BJ, Louie TJ, Gorbach SL, Bartlett JG.
Experimental animal models for anaerobic infections. Rev Infect Dis 1979;1:
291e301.
[9] Brouwer MC, van de Beek D. Epidemiology, diagnosis, and treatment of brain
abscesses. Curr Opin Infect Dis 2017;30:129e34.
[10] Laulajainen-Hongisto A, Lempinen L, Farkkila E, Saat R, Markkola A,
Leskinen K, et al. Intracranial abscesses over the last four decades; changes in
aetiology, diagnostics, treatment and outcome. Infect Dis (Lond) 2016;48:
310e6.
[11] Helweg-Larsen J, Astradsson A, Richhall H, Erdal J, Laursen A, Brennum J.
Pyogenic brain abscess, a 15 year survey. BMC Infect Dis 2012;12:332.
[12] Nicolosi A, Hauser WA, Musicco M, Kurland LT. Incidence and prognosis of
brain abscess in a defined population: Olmsted County, Minnesota,
1935e1981. Neuroepidemiology 1991;10:122e31.
[13] Nathoo N, Nadvi SS, Narotam PK, van Dellen JR. Brain abscess: management
and outcome analysis of a computed tomography era experience with 973
patients. World Neurosurg 2011;75:716e26. discussion 612e17.
View publication stats
[14] Dupuis-Girod S, Giraud S, Decullier E, Lesca G, Cottin V, Faure F, et al. Hem-
orrhagic hereditary telangiectasia (RendueOsler disease) and infectious dis-
eases: an underestimated association. Clin Infect Dis 2007;44:841e5.
[15] Sonneville R, Mirabel M, Hajage D, Tubach F, Vignon P, Perez P, et al.
Neurologic complications and outcomes of infective endocarditis in critically
ill patients: the ENDOcardite en REAnimation prospective multicenter study.
Crit Care Med 2011;39:1474e81.
[16] McClelland 3rd S, Hall WA. Postoperative central nervous system infection:
incidence and associated factors in 2111 neurosurgical procedures. Clin Infect
Dis 2007;45:55e9.
[17] Jim KK, Brouwer MC, van der Ende A, van de Beek D. Cerebral abscesses in
patients with bacterial meningitis. J Infect 2012;64:236e8.
[18] Charlier C, Perrodeau E, Leclercq A, Cazenave B, Pilmis B, Henry B, et al.
Clinical features and prognostic factors of listeriosis: the MONALISA national
prospective cohort study. Lancet Infect Dis 2017;17:510e9.
[19] Mylonakis E, Hohmann EL, Calderwood SB. Central nervous system infection
with Listeria monocytogenes. 33 years' experience at a general hospital and
review of 776 episodes from the literature. Medicine (Balt.) 1998;77:313e36.
[20] Lu CH, Chang WN, Lui CC. Strategies for the management of bacterial brain
abscess. J Clin Neurosci 2006;13:979e85.
[21] Seydoux C, Francioli P. Bacterial brain abscesses: factors influencing mortality
and sequelae. Clin Infect Dis 1992;15:394e401.
[22] Yang SY. Brain abscess: a review of 400 cases. J Neurosurg 1981;55:794e9.
[23] Chun CH, Johnson JD, Hofstetter M, Raff MJ. Brain abscess. A study of 45
consecutive cases. Medicine (Balt.) 1986;65:415e31.
[24] Takeshita M, Kawamata T, Izawa M, Hori T. Prodromal signs and clinical
factors influencing outcome in patients with intraventricular rupture of pu-
rulent brain abscess. Neurosurgery 2001;48:310e6. discussion 6e7.
[25] Muccio CF, Caranci F, D'Arco F, Cerase A, De Lipsis L, Esposito G, et al. Magnetic
resonance features of pyogenic brain abscesses and differential diagnosis
using morphological and functional imaging studies: a pictorial essay.
J Neuroradiol 2014;41:153e67.
[26] Reddy JS, Mishra AM, Behari S, Husain M, Gupta V, Rastogi M, et al. The role of
diffusion-weighted imaging in the differential diagnosis of intracranial cystic
mass lesions: a report of 147 lesions. Surg Neurol 2006;66:246e50. discussion
50e1.
[27] Barlas O, Sencer A, Erkan K, Eraksoy H, Sencer S, Bayindir C. Stereotactic
surgery in the management of brain abscess. Surg Neurol 1999;52:404e10.
discussion 11.
[28] Mamelak AN, Mampalam TJ, Obana WG, Rosenblum ML. Improved manage-
ment of multiple brain abscesses: a combined surgical and medical approach.
Neurosurgery 1995;36:76e85. discussion 6.
[29] Mishra AK, Dufour H, Roche PH, Lonjon M, Raoult D, Fournier PE. Molecular
revolution in the diagnosis of microbial brain abscesses. Eur J Clin Microbiol
Infect Dis 2014;33:2083e93.
[30] de Lastours V, Kalamarides M, Leflon V, Rodallec M, Vilgrain V, Nicolas-
Chanoine MH, et al. Optimization of bacterial diagnosis yield after needle
aspiration in immunocompetent adults with brain abscesses. Neurosurgery
2008;63(4 Suppl. 2):362e7. discussion 7e8.
[31] de Louvois J. Bacteriological examination of pus from abscesses of the central
nervous system. J Clin Pathol 1980;33:66e71.
[32] de Louvois J, Gortavai P, Hurley R. Bacteriology of abscesses of the central
nervous system: a multicentre prospective study. Br Med J 1977;2(6093):
981e4.
[33] Gortvai P, De Louvois J, Hurley R. The bacteriology and chemotherapy of acute
pyogenic brain abscess. Br J Neurosurg 1987;1:189e203.
[34] Al Masalma M, Armougom F, Scheld WM, Dufour H, Roche PH, Drancourt M,
et al. The expansion of the microbiological spectrum of brain abscesses with
use of multiple 16S ribosomal DNA sequencing. Clin Infect Dis 2009;48:
1169e78.
[35] Al Masalma M, Lonjon M, Richet H, Dufour H, Roche PH, Drancourt M, et al.
Metagenomic analysis of brain abscesses identifies specific bacterial associa-
tions. Clin Infect Dis 2012;54:202e10.
^tre P, Gauvrit JY, et al. Follow-up of
[36] Carsin A, Revest M, Le Reste PJ, Piau C, Filla
brain abscess volume by magnetic resonance imaging: a pilot study. Abstract
1190. ID Week, New Orleans, USA, 26e30 October 2016.
[37] Nau R, Sorgel F, Eiffert H. Penetration of drugs through the blood-
cerebrospinal fluid/bloodebrain barrier for treatment of central nervous
system infections. Clin Microbiol Rev 2010;23:858e83.