Child's Nervous System (2019) 35:1117–1128

https://doi.org/10.1007/s00381-019-04182-4

REVIEW ARTICLE

Brain abscess in pediatric age: a review

Chiara Mameli 1 & Teresa Genoni 1 & Cristina Madia 1 & Chiara Doneda 2 & Francesca Penagini 1 & Gianvincenzo Zuccotti 1

Received: 1 February 2019 / Accepted: 28 April 2019 / Published online: 6 May 2019
# Springer-Verlag GmbH Germany, part of Springer Nature 2019

Abstract
Objective The purpose of the paper is to examine the current state of the art about epidemiology, diagnosis, and treatment of this
infection.
Methods A review of the literature was performed through a PubMed search of original articles, case reports, and reviews using
the key words “brain abscess,” “cerebral abscess,” “brain infection,” “intracranial suppuration,” “otogenic brain abscess,” “otitis
complications,” and “sinusitis complications.”
Results Pediatric brain abscess is a rare but serious infection, often involving patients with specific risk factors and burdened by a
high risk of morbidity and mortality. Brain abscess incidence and mortality decreased over the years, thanks to improved
antibiotic therapy, new neurosurgical techniques, and the wide spread of vaccinations. There are no guidelines for the adequate
diagnostic-therapeutic pathway in the management of brain abscesses; therefore, conflicting data emerge from the literature. In
the future, multicentric prospective studies should be performed in order to obtain stronger evidences about brain abscesses
management. Over the next few years, changes in epidemiology could be observed because of risk factors changes.

Keywords Brain abscess . Intracranial infection . Children . Central nervous system infections

Introduction In this paper, we review and discuss the epidemiology,

pathogenesis, diagnosis, and treatment of BA in the pediatric
Brain abscess (BA) is a focal central nervous system infection population.
involving the cerebral parenchyma. This condition is infrequent
in the adult population and even rarer in pediatric patients [65], but
the risk is higher in some patient groups. Over the last decades, BA Epidemiology and risk factors
incidence decreased in some countries thanks to a reduction in
predisposing factors, such as untreated otorhinolaryngological in- Thanks to the health care quality improvement, in twentieth
fections and uncorrected congenital heart defects; at the same time, century, BA has become a rare disease with an overall inci-
other BA-associated conditions like immunosuppression (mainly dence of 0.3–1.8 per 100,000 inhabitants per year [19, 28, 37,
due to untreated human immunodeficiency virus (HIV) infection 48, 50, 51].
and transplants) have gained greater importance [67]. The incidence of BAs in children and adolescents is diffi-
Despite the advances in the diagnosis and treatment, BA cult to determine. Few studies have specifically explored the
remains a disease with significant morbidity and mortality, incidence of BAs in the pediatric population: Most of them are
both in adults and children [65]. small case series or monocentric retrospective studies, almost
exclusively focused on the situation in industrialized
countries.
* Chiara Mameli Overall, about 25% of BAs occur in children [64]. Recent
[email protected] studies show a considerable variability about which pediatric
age is most frequently affected; this may be explained by the
1
Department of Pediatrics, V. Buzzi Childrens’ Hospital, University of fact that some of them consider both adults and children, other
Milan, Milan, Italy report either the median age or an age interval; therefore, a
2
Pediatric Radiology and Neuroradiology Unit, Children Hospital V. comparison is difficult. Based on these reports, BA seems to
Buzzi, Milan, Italy occur most often between 4 and 10 years of age [19, 39, 56].
1118 Childs Nerv Syst (2019) 35:1117–1128

In Europe pediatric BA, prevalence ranges from 1.2% of BA can also be a rare complication of meningitis, particu-
cases in the under 7 years of age group of an Irish monocentric larly in neonates [59]. Other risk factors are described, in
study [60], to a British monocentric study reporting 32% of particular, the presence of foreign bodies in the airways [17],
cases in the 0–19 group [13]. The significant difference be- congenital dermal sinuses [47], and esophageal procedures
tween these studies is probably explained by the two different [41]. Infancy (age < 6 months) and male sex are also predis-
age groups considered and the different inclusion criteria, with posing factors [19, 26, 52, 56].
the first study considering only patients who underwent a The presence of single nucleotide polymorphisms of
surgical drainage, so pediatric BA—treated more ICAM-1 and MCP-1 genes has also been described as a risk
conservatively—may be underestimated. factor for developing cerebral abscesses [44].
During the neonatal age, BAs are even rarer, but they are Finally, BA as a complication of acute bacterial sinusitis
associated with a high risk of severe complications and mor- can be associated to a low socio-economic status even in de-
tality [14–38]. veloped countries [63].
An unexplained male predilection was reported in the pe- Pneumococcal vaccination is changing BA etiology, with
diatric population [11, 64]. initial reports of a reduced incidence of S. pneumoniae [56].
In developing countries, as expected, the incidence of BAs
is higher. In fact, in these countries, different types of patho-
gens are encountered, and the transmission of pathogens is Etiopathogenesis
increased due to deficient infrastructure; malnutrition leads
to amplified severity of common infections, and there is a lack A brain infection develops into a capsulated pus collection
of resources to treat this condition. BAs are reported to ac- inside the cerebral parenchyma within 14 days, evolving from
count for 8% of all intracranial mass lesions, and up to 50% early cerebritis to late capsular stage [7].
occurred in population aged 0–20 years [22, 43, 48]. The In 40–50% of cases, pathogens reach the brain via a con-
smallest incidence was reported in the South-East Asian coun- tiguous site, such as during middle ear, mastoid and paranasal
try Taiwan where only 5.4% of patients with BA were sinus infections, or through a skull discontinuity due to head
children. trauma or neurosurgery. In 30–40% of cases, they spread
Pediatric studies published in the last decade show that through blood flow from a distant focus of infection (e.g.,
BAs are often associated with predisposing conditions in dental abscess, endocarditis, lung, or cutaneous infections).
56–86% of cases [8, 19, 39, 52, 56, 64]. The remaining cases are of unknown origin [67].
Risk factors can be divided in acute events and chronic The location of the abscess depends on the origin of the
conditions. The first are mostly ear, sinus, and dental infec- primary infection. The most frequent site is the frontal lobe
tions (frequently related to poor dental hygiene or dental pro- (secondary to frontal or ethmoidal sinusitis or dental infec-
cedures), neurosurgical procedures, and head trauma. An in- tion), followed by parietal and temporal lobes (acute otitis
frequent but typical cause of trauma in children are injuries media, mastoiditis, sphenoidal sinusitis) and less frequent sites
with sharp toys or pencils [5]. Among chronic conditions, such as cerebellum and brainstem, from otogenic or hematog-
there is often the presence of a pulmonary circulation shunt, enous origin [8, 56, 64].
which could allow septic microemboli from distant foci of Multiple abscesses occur in 19–33% of total BAs [15, 52,
infection (endocarditis, osteomyelitis, pulmonary, and skin in- 56], often related to hematogenous spread, and they follow the
fections) to reach the brain without being stopped by pulmo- distribution of the middle cerebral artery.
nary phagocytic filter; this may happen in carriers of congen- Bacteria are by far the most common pathogens in immu-
ital heart defects (CHD) [68] and pulmonary arteriovenous nocompetent patients. A meta-analysis that included 6663
fistulas as in hereditary haemorrhagic telangiectasia [23]. adult and 1023 pediatric BAs—reported between 1935 and
Goodkin et al., describing twentieth century BA trends at 2012—in which pus or blood cultures were performed
Children’s Hospital Boston, found that congenital heart dis- showed that children shared similar etiology with adults, with
ease, otitis, and sinusitis decreased as a cause of intracranial pediatric cultures positive for Streptococcus spp. in 36% of
infection [26]. Other more recent studies show the same trend cases, followed by Staphylococcus spp. (18%) and gram-
in developed countries, thanks to early CHD correction and negative enteric bacteria (Proteus spp., Klebsiella
appropriate antimicrobial therapy [64]; they remain a signifi- pneumoniae, Escherichia coli, and Enterobacteriae) in 16%
cant problem in the developing world [68]. of cases [10]. Streptococcus spp. is commonly associated with
Another important predisposing condition are immunosup- sinusitis, otitis media, and endocarditis (the latter with the
pressive states, which can be rare like X-linked agammaglob- S. viridans group); a recent case series suggests a possible rise
ulinemia [31] or of increasing incidence like those associated in the incidence of BA due to group A streptococcus [12].
with a solid-organ or hematopoietic stem-cell transplantation Staphylococcus spp. (mainly aureus and epidermidis) in-
[6] and with HIV infection [25, 67]. fections are related to head trauma, surgery, or skin infections.
Childs Nerv Syst (2019) 35:1117–1128 1119

Neonatal BA can arise from Citrobacter or Proteus men- if the risk of finding negative samples is high [15, 26, 28, 39,
ingitis, less frequently from Escherichia coli and Serratia 56, 62]. However, it is important to underline that a lumbar
marcescens [26, 59]. puncture should not be routinely performed and is contraindi-
Pneumococcal vaccination may contribute in future years cated in the case of non-communicating obstructive hydro-
to a variation in BA incidence and etiology, as it is beginning cephalus and brain shift, which can be consequences of the
to be reported in some countries [52, 56]. mass effect of the BA [16]. Therefore, computed tomography
Finally, unusual pathogens like fungi, parasites, and (CT) is often the first imaging performed urgently, followed
mycobacteria accounted for less than 2% of total cases, but by a magnetic resonance (MR) to confirm the diagnosis and
this percentage was probably underestimated because many of deepen the characterization of the abscess [19, 62].
the analyzed studies considered only bacterial BA [10]. BAs The etiological diagnosis can be obtained with the culture
from opportunistic microorganisms are usually multiple. They of the pus extracted from the abscess; the sample can be ob-
can occur in HIV-positive children with a low CD4 count; the tained thorough stereotactic biopsy or aspiration rather than a
most common pathogens are Toxoplasma (although cerebral craniotomy with excision [3, 11, 15, 24, 26, 46, 52, 64], even
toxoplasmosis is less frequent than in adults), Nocardia and though in some cases neurosurgery is not indicated and ther-
Mycobacterium spp [25]. Fungal abscesses (mainly apy should be undertaken and continued empirically, with
Aspergillus or Candida) typically affect solid organ trans- monitoring of clinical and radiological improvement [11, 29,
plants recipients or children treated for leukemia [36]. 52, 62].

Blood and CSF samples

Clinical features
Despite the presence of the infection, the probability of find-
Clinical features depend on BA number, site, size, the involve- ing normal inflammatory markers could be high. Indeed,
ment of the surrounding area, and the microorganisms respon- Udayakumaran et al. [68] report only a 20% of pediatric car-
sible for the infection [34, 56]. The median duration from the diogenic BAs with altered blood tests (leukocyte count, CRP,
onset of symptoms to the diagnosis is 7–11 days [39, 56], but ESR). Regarding the evaluation of white blood cells, normal
the presentation can vary from indolent to very rapid. values are reported in about one fourth of cases by Atiq et al.
Patients are commonly symptomatic; signs and symptoms [3], in about two-thirds of cases by Shachor-Meyouhas et al.
can be divided in general and neurological. Fever is frequent, [64] and in half of adult patients by Helweg-Larsen et al. [28].
often associated with neurological symptoms in isolation or In the presence of altered white blood cell count, leukocytosis
combination, such as new onset headache (typically associat- is more frequently found than leukopenia [9, 10]. A normal
ed with vomiting), seizures, hemiplegia, cranial nerve palsies value of CRP is described in about one third of cases presented
and altered level of consciousness ranging from drowsiness to by Raffaldi et al. [56] and Helweg-Larsen et al. [28] and in
coma [15, 52, 64]. Neonates can have a bulging fontanel and/ 17% of patients with intracranial suppurations (BA and sub-
or an increase of the cranial circumference [39, 56]. dural empyemas) presented by Cole et al. [15], with most of
The clinical triad historically associated with BA (fever, these cases concerning single or even multiple BAs.
headache, neurological deficits) is only present in a small per- Even in cases when a lumbar puncture can be performed,
centage of cases in more recent retrospective studies [19, 52, results obtained with the chemical-physical examination
64]. Frontal abscesses can be symptomatic only when they (white blood cell count, glucose, and protein content) can
reach large dimensions [65]. show a great variability [24, 39, 56]. However, considering
the mean and median values reported in some studies, it is
possible to underline the more frequent presence of leukocy-
Diagnosis tosis, hyperproteinorrachia, and normal or reduced glucose
values [24, 28, 39, 56]. Normal CSF analyses are described
Blood tests (in particular leukocyte count, C-reactive protein, in 30% of cases by Shachor-Meyouhas et al. [64] and in 16%
erythrocyte sedimentation rate, and blood cultures) are the of patients in the systematic review by Brouwer et al. [10].
first and easiest examinations to perform urgently, to evaluate
the possible presence of altered inflammatory markers second- Cultural samples
ary to the infection and find any hematogenous dissemination
of the microorganism. Because of the wide range of micro-organisms that can be
Moreover, in the presence of signs and symptoms of neu- involved in the infection, cultures (for aerobic and anaerobic
rologic involvement, a lumbar puncture is useful to evaluate bacteria, Mycobacterium, fungi, protozoa), Gram and special
chemical-physical alterations of the cerebrospinal fluid (CSF) stains (for fungi, Mycobacterium, Nocardia) and polymerase
and to perform a CSF culture to find a specific etiology, even chain reaction should be performed on blood, CSF and pus of
1120 Childs Nerv Syst (2019) 35:1117–1128

the cerebral abscess (stereotactic biopsy or aspiration and Finally, a recent study by Liu et al. [40], performed
specimens from craniotomy) [1, 3, 4, 9, 11, 15, 24, 26, 28, on rat brains, has showed that a new MR technique,
39, 52, 62, 64]. based on bacterial chemical exchange saturation transfer
With regard to the isolation of micro-organisms from blood (bacCEST), is useful in detecting BAs (caused by
and CSF samples, the available data are contrasting. Raffaldi Staphylococcus aureus) and monitoring the infection
et al. [56] report a rate of positive blood cultures of 22.7%; a during the follow-up.
slightly higher rate is described by Lee et al. [39], with blood
culture positivity in 28.6% of cases. On the contrary, lower Further investigations
rates are reported by Canpolat et al. [11], with a 100% of blood
culture negativity in patients who did not undergo surgery, An otolaryngology evaluation with paranasal sinus and mid-
Auvichayapat et al. [4], with only 2.8% of blood growth, dle ear imaging should be taken into consideration [29, 35],
and Cole et al. [15], with 16% of positive blood cultures. because otogenic and sinogenic infections are one of the most
Positive CSF culture rates vary in the literature, with a frequent primary foci in patients with a diagnosis of BA [3, 19,
2.8% rate of micro-organisms growth described by 26, 52, 56, 62] and, in some cases, symptoms of sinusitis may
Auvichayapat et al. [4], a 24% and 33.3% shown in the studies be poor or misdiagnosed [61].
by Raffaldi et al. and Lee et al. [39, 56], respectively, reaching Similarly, a cardiologic evaluation with an echocardiogram
a value of 44% in the report by Cole et al. [15]. should be performed because of the high risk of association
The rate of micro-organism isolation from abscess samples with CHD or cardiac infections [3, 19, 26, 52, 56, 62].
is about 60–80%, with polymicrobial involvement in about Odontogenic origin of the infection is not particularly fre-
20–30% of cases [8, 11, 15, 19, 26, 28, 60]. However, some quent [24, 26, 56, 62]; however, a dental origin should be
studies in literature [39, 52] report abscess culture positivity in suspected and a dental evaluation should be performed in case
less than half of cases. no other focus is found, oral flora is isolated from the abscess,
an oral infection is found, or an oral intervention has been
performed [42].
Imaging Finally, the use of ultrasound, X-ray, or CT scan could be
useful to find other possible sources of infection, such as pul-
Frequently, the first type of imaging performed is a computed monary arteriovenous malformations [66], lung infections or
tomography (CT) [19, 39, 56, 64], as this is available as urgent malformations, bone, abdominal, or skin infections [20].
and it is able to clarify the characteristics of the suspected Figure 3 shows a hypothesis of diagnostic and therapeutic
lesion; according to what is reported by the ESCMID guide- flow chart in case a BA is suspected.
lines in 2016 [69], it is often performed before lumbar punc-
ture in the case of a patient presenting with Glasgow Coma
Scale < 10, focal neurological deficits, new-onset seizures, Treatment
and severe immunocompromised state.
In the study by Felsenstein et al. [19], magnetic resonance There is a lack of international guidelines about the proper
(MR) is suggested as the gold standard imaging for the diagno- management of BAs; therefore, in a consensus document in
sis, as in all the patients whose first imaging was MR, the BA 2010, Arlotti et al. [2] provided recommendations about in-
was detected; they also recommend its use in the follow-up. fection treatment that can be either only medical or both med-
The superiority of MR compared to CT is due to its better ical and surgical.
resolution, the lower toxicity of the contrast used, and the According to these recommendations, medical treatment
ability to identify lesions at risk of complications precociously alone may be considered in patients without severe neurolog-
[67]. Moreover, with the use of spectroscopy and diffusion- ical impairment at admission (GCS > 12), with a small abscess
weighted (DW) techniques, MR has high values of sensitivity (< 2.5 cm) or with multiple abscesses, with a diagnosed etiol-
and specificity in the differential diagnosis with cystic or neo- ogy and in case of contraindications to surgery; moreover,
plastic lesions [45, 53, 58]. antibiotics represent an adjuvant therapy after surgery for
The typical radiologic aspect of a pyogenic BA is a large BAs or BAs causing mass effect [2].
necrotic center with low signal at the DW-MR and a Case series of pediatric patients (Table 1) show how the
T2-hypointensity with enhancement for the peripheral percentage of parenchymal BAs not treated with neurosurgi-
capsule (Fig. 1); fungal abscesses show a hypointense cal intervention varies, ranging from 0% [3, 24, 64] to more
center in T2-weighed images (Fig. 2) and have a vari- than two-thirds of cases, as described for instance by Jain et al.
able expression in DW-MR; tubercular abscesses show in a cohort of patients with otogenic BAs (neurosurgery per-
hypointensity in T2-weighed images and a capsule en- formed in 36.1% of cases, but modified radical mastoidecto-
hancement, that is absent for the center [61]. my in all the patients) [48]. Not all the studies show the
Childs Nerv Syst (2019) 35:1117–1128 1121

Fig. 1 Brain abscesses from beta-hemolytic streptococcus infection. MR iso-hyperintense T2-weighted signal (arrows) and contrast enhancing reg-
imaging study in a 13-day-old baby affected by multiple small brain ular rim with uniform hypointense core on T1-weighted image (arrows); c
abscesses from beta-hemolytic streptococcus infection. a Axial T2 and the apparent diffusion coefficient (ADC) axial map demonstrates
b coronal contrast enhanced T1-weigthed sections showing multiple uniformely reduced value within the core of the lesions, compatible with
round shape lesions in basal ganglia region bilaterally, characterized by pus accumulation (arrows)

specific indications for the choice of a conservative strategy, describe the frontal lobe as the most frequently involved site
but, when reported, they usually meet the criteria recommend- in patients treated conservatively. Moreover, a lower tendency
ed by Arlotti et al. [67]. Additionally, Raffaldi et al. [10] to a conservative approach is reported before 1980 [24].

Fig. 2 Brain abscesses from aspergillus fumigatus infection. MR reduced value (solid nature) in the eccentric nodule with adjacent crescent
imaging study in a 13-year-old girl affected by multiple fully developed shape area of ADC increase (fluid nature), suggesting the nature
brain abscesses from aspergillus fumigatus infection. The heterogeneous (probable fungine) of the nodule; in d the contrast enhanced T1-
content of the abscesses is the main feature helping in differentiating from weighted axial scan shows remarkable signal increase within the
bacterial-pyogenic abscesses. a Axial T2 and b T1-weigthed sections abscess rim, but not in the eccentric amorphous nodule that appears to
showing multiple brain focal lesions, the larger one (arrows) be attached to the abscess wall. The presence of an eccentric solid nodule
characterized by regular iso-hypontense T2-weighted rim and of low ADC value and amorphous material may help in differentiating a
containing amorphous material in an eccentric nodule; c the fungine from a bacterial-pyogenic abscess
corresponding apparent diffusion coefficient (ADC) map demonstrating
1122 Childs Nerv Syst (2019) 35:1117–1128

Suspected brain abscess

(fever, headache, vomit, seizures, neurological

impairment, bulging fontanelle, increased
cranial circumference)

Blood examination Imaging:

(Complete blood count, CRP, ESR, culture) - MR (with diffusion-weighted and spectroscopy
techniques) if readily available.
CSF examination (if not contraindicated)

(Chemical-physical examination, culture) - CT scan in urgency (exclude intracranial hypertension,

MR not available).

Further investigations (depending on the suspected primary focus)

- Otolaryngology evaluation with paranasal sinus and middle ear imaging.

- Cardiological evaluation, echocardiogram.
- Odontoiatric evaluation.
- Skin inspection.
- Abdominal US, chest or bone X-Ray or CT scan.

Conservative treatment Neurosurgical treatment (with

culture of the abscess pus)
- No severe neurological impairment
at admission (GCS > 12).
- Small abscess (< 2.5 cm).
- Multiple abscesses. Aspiration Excision
- Detected pathogen.
- Contraindications for surgery. - Diameter ≥ 2.5 cm. - Posterior cranial fossa
- Deep or multiple lesions. location.
- Eloquent areas involved.
- Post-traumatic lesions.
- High risk of complications.
Close clinical, biochemical and - Multilobulated abscess.
radiological monitoring and, if no
improvement within 1-2 weeks,
consider neurosurgery

Antibiotic therapy

- Broad-spectrum, bactericidal and non-bacteriolytic agents, able to pass the

hematoencephalic barrier, safe to be administered at high dosages.
- Chosen empirically according to predisposing factors and changed
according to cultures results.
- Duration: 4-6 weeks for abscesses treated surgically; 6-8 weeks for
complicated or treated only medically abscesses.

Fig. 3 Flow-chart for the management of suspected brain abscess. CRP C-reactive protein, CSF cerebrospinal fluid, CT computed tomography, ESR
erythrocyte sedimentation rate, GCS Glasgow Coma Scale, MR magnetic resonance, US ultrasound,

Antibiotic therapy should be started empirically and towards bactericidal and nonbacteriolytic agents, using
then adjusted according to the results of the cultures molecules that can pass the blood-brain barrier (lipophil-
[42]. The choice of empiric therapy should rely on the ic, with low molecular mass and low plasma protein
predisposing conditions and consequently on the micro- binding) and sufficiently safe to be administered at high
organism thought to be involved, should be directed dosages even for a long time [49, 66, 67].
Table 1 Management of pediatric parenchymal brain abscesses

Surgical intervention Medical treatment

Neurosurgical intervention Type of abscess Antibiotics Type of abscess treated Empiric antibiotic therapy Duration of antibiotic
alone only medically therapy

Sahbudak Bal et al. N = 83.3% (93.3% burr hole – Not known N = 16.7% - Multiple abscesses - Third generation cephalosporins with vancomycin; - At least 6 weeks
2018 [62] 6.7% craniotomy) vancomycin and meropenem
Raffaldi et al. 2017 N = 76% (85% drainage – - Single abscess (78.3%) in N = 24% - Single abscess (57.9%) - IV: third generation cephalosporins alone or with - Mean: 59.5 ± 28.4
[56] 11.7% resection – 3.3% the frontal lobe (33.3%) in the frontal lobe metronidazole or with metronidazole and [15–159] days of
Childs Nerv Syst (2019) 35:1117–1128

other surgical intervention) - Diameter of 3 ± 1.6 cm (21%) vancomycin or teicoplanin; third generation overall therapy
[0.5–6 cm] - Diameter of cephalosporins with different drugs - Mean: 65.3 ± 21.2
1.7 ± 0.9 cm - Oral: Amoxicillin-clavulanate [30–119] days for
[0.3–3 cm] medical treatment
alone
- Mean: 57.8 ± 30.2
[15–159] days for
patients undergone
surgical
intervention
- Mean: 42.4 ± 18.3
[5–120] days iv
Jain et al. 2017 N = 36.1% (neurosurgical Poor response to N = 63.9% Not known - IV: ceftriaxone, vancomycin, metronidazole - 4–6 weeks iv
[29]a drainage) treatment, enlarging
abscesses with mass
effect, risk of
intraventricular rupture
Brizuela et al. 2017 N = 89.5% Not known N = 10.5% Not known - Vancomycin and meropenem; ceftriaxone and - Median: 56 [42–60]
[8] metronidazole with or without vancomycin days of overall
therapy
- Median: 42 [35–56]
days iv
Acar et al. 2016 [1] N = 83.9% (73% aspiration – Aspiration was preferred N = 16.1% Not known - IV: third generation cephalosporins with vancomycin - Median: 73
27% resection) [N = 15.4% for multiple abscesses and metronidazole; vancomycin and meropenem [28–540] days
(concomitant mastoidecto- and eloquent brain areas - Oral: Ciprofloxacin
my)] involved
Özsürekci et al. N = 76% (40% aspiration – Not known N = 24% - Small size, multiple - Third generation cephalosporins with vancomycin Not known
2012 [52] 36% resection) lesions, deep location, and metronidazole; penicillin G or
and good clinical ampicillin-sulbactam with chloramphenicol and/or
response to medical metronidazole, with or without amikacin
therapy
Cole et al. 2012 N = 88% (60% burr hole Not known N = 12% Single abscess but high - Third generation cephalosporins alone or with - Mean: 14.9 weeks
[15] aspiration – 13% stereotac- risk for surgery metronidazole, with or without amoxicillin or of overall therapy
tic surgery – 7% endocopic another drug; meropenem, rifampicin, clindamycinb - Mean: 9 weeks if
aspiration – 20% cranioto- medical therapy
my) alone
1123
Table 1 (continued)
1124

Surgical intervention Medical treatment

Neurosurgical intervention Type of abscess Antibiotics Type of abscess treated Empiric antibiotic therapy Duration of antibiotic
alone only medically therapy

- Mean: 16.9 weeks if

previous surgery
- Mean: 9.7 weeks iv
Sachor-Meyouhas N = 100% (70% burr hole Not known N = 0% - Chloramphenicol and cloxacillin or penicillin; third- 85.2% 6 weeks
et al. 2010 [64] drainage – 30% cranioto- generation cephalosporins with metronidazole - 14.8% 3–4 weeks
my) and/or vancomycin; meropenem alone or with van-
comycin or trimethoprim/sulphamethoxazole
Kao et al. 2008 N = 80% Not known N = 20% Small abscess, good Not known - Mean: 60 ± 30
[30] response to medical [35–105] days for
therapy, refusal of medical treatment
surgery alone
- Mean: 46 ± 11
[28–63] days for
patients undergone
surgical
intervention
Gelabert-González N = 100% (7.2% aspiration – Not known N = 0% - Third generation cephalosporins with vancomycin - Median: 6 weeks
et al. 2008 [24] 92.8% craniotomy) and metronidazole [4–12]
Auvichayapat et al. N = 85.3% (59.4% aspiration – Not known N = 14.7% Not known Not known Not known
2006 [4] 6.3% excision – both
34.3%)
Atiq et al. 2006 [3] N = 100% (100% burr hole) Diameter > 2 cm N = 0% - Metronidazole, third generation cephalosporins, - 14–40 days
cloxacillin, benzylpenicillin
Goodkin et al. N = 77% (93% aspiration – Not known N = 22% Multiple abscesses, Not known Not known
2004 [26]d 7% resection) sepsis or shock with
MOF, single and small
abscess
(diameter < 2 cm)

IV intravenous, MOF multi-organ failure, N number

a
All of the patients underwent radical mastoidectomy
b
Used for parenchymal abscesses and subdural empyemas
c
Not specified if empiric or specific antibiotic therapy
d
One patient was not treated with antibiotic
Childs Nerv Syst (2019) 35:1117–1128
Childs Nerv Syst (2019) 35:1117–1128 1125

In pediatric patients, the most common combination of 38, 70]. Lower percentages of neurosurgical treatments (aspi-
intravenous antibiotics is a third-generation cephalosporin to- ration or craniotomy) are described in the study by Jain et al.
gether with metronidazole and in some cases vancomycin [5, [29] in patients with intracranial complications of otitis media
9–11, 19, 20, 24, 37, 44, 46–48, 50]. Some studies [9, 37, 50] and in which, anyway, m ost patients underwent
also report the possible use of meropenem alone or in combi- mastoidectomy.
nation with other drugs and other studies [9, 12, 50] describe a
wider use of chloramphenicol in the past. Moreover,
Krzysztofiak et al. [68] show the effective use of linezolid in Prognosis
pediatric patients, with the evidence of clinical and radiolog-
ical improvement. Finally, a recent review of fungal brain Over the years, the mortality rate of BAs progressively de-
infections [70] reports that amphotericin B, voriconazole, creased and data reported in the literature describe a current
and fluconazole are the most frequently antimycotic agents rate lower than 10% [1, 8, 10, 15, 19, 29, 30, 51, 52, 56, 62,
recommended. 64, 69].
Antimicrobial agents should be started intravenously and Higher percentages (11–25%) are reported by Gelabert-
then can be switched to oral therapy, depending on clinical, González et al. (10.7%), Auvyachapat et al. (10.7%), Atiq
biochemical, and radiological improvement [10]. Arlotti et al. et al. (16%), and Goodkin et al. (24%) [3, 4, 24, 26].
[66] consider as adequate an overall treatment of 4–6 weeks The most frequent factors involved in higher rate of mor-
for patients that have undergone surgical treatment and a par- tality are delayed presentation and diagnosis, severe neurolog-
enteral therapy of 6–8 weeks for patients treated with just ical impairment, and development of complications [3, 4].
medical therapy or with complicated BAs. Felsenstein et al. In the cases described so far, involving studies with both low
[5] describe longer courses of antimicrobial therapy in chil- and high mortality rates, death occurred more frequently in
dren with immunodeficiency. Furthermore, it is worth noting patients with lower Glasgow Coma Scale score, intraventricular
that in the study by Raffaldi et al. [10], no significant differ- rupture of the abscess [3, 19, 28, 55, 62] and severe underlining
ence was shown in duration of therapy between only- conditions, such as CHD, congenital or acquired immunosup-
medically and surgically treated abscesses and, moreover, pression, organ transplantation, tumors in treatment with che-
Cole et al. [37] describe a longer course of antibiotic treatment motherapy, neonates born premature or with malformations [4,
in patients that have undergone surgery. 8, 11, 19, 26, 64]. The only patient who died for a parenchymal
There is indication for a short course of corticosteroids in abscess in the study by Cole et al. [15] had a primary immuno-
the presence of edema, and their use is contraindicated in the deficiency, and in the report by Acar et al. [1], of the two
absence of intracranial hypertension, because of the risk of patients who died (6.4%), one had a concomitant tubercular
delayed capsule formation, necrosis, and lower efficacy of infection in HIV and the other was immunosuppressed because
antibiotics [49, 70]. of a liver transplantation. Moreover, the study by Felsenstein
In the absence of clinical and radiological improvement et al. [19] reports a younger age at presentation and a more
within 1–2 weeks, a neurosurgical intervention should be con- immediate treatment (interpreted as a sign of severe clinical
sidered [66]. Depending on the characteristics, location, and condition) as associated with poor outcome. On the contrary,
number of the abscesses, there is the possibility to perform a Lee et al. [39] do not report association between outcome and
stereotactic or endoscopic aspiration of the abscess rather than factors such as neurological symptoms and level of conscious-
an open surgery (craniotomy with excision) [57]. Aspiration is ness at admission, presence of CHD, altered blood examina-
frequently considered the gold standard neurosurgical treat- tion, and type of treatment.
ment [21] and in a review by Ratnaike et al. [57], aspiration A full recovery rate from the infection of about 60–70% is
has been associated with a lower overall mortality rate when reported in the case of early diagnosis and proper therapy [10,
compared to excision. 15, 19, 39, 56, 62, 64, 68]. However, a percentage of less than
Stereotactic aspiration is indicated in the case of abscesses 50% of cases is described in other case series. These data may
≥ 2.5 cm, deep or multiple lesions, eloquent areas involve- be explained by delayed intervention, the presence of higher
ment, and high risk of complications; excision can be indicat- number of patients with neurological impairment at admission,
ed in the case of posterior cranial fossa location, post- and severe predisposing conditions or the patient’s choice to be
traumatic lesions, multilobulated or superficial abscesses, followed in a specialized center [1, 3, 8, 11, 26, 30].
and in the case of aspiration failure [2, 32, 54, 66]. Regarding otogenic BAs, a systematic review by Duarte
In the case series of pediatric patients reported in Table 1, et al. [18] describes meningitis, cerebral herniation, and death
more than three-quarters of patients were treated with a neu- as the most common acute complications. Clinical sequelae
rosurgical intervention, with less-invasive techniques pre- can be found in about 30% of patients and are mainly epilepsy,
ferred over craniotomy, and similar data can be found in stud- motor, visual and hearing deficits, hydrocephalus, and lan-
ies also involving adult patients over the years [4, 6, 7, 12, 13, guage impairment [1, 15, 26, 39, 56, 60, 62].
1126 Childs Nerv Syst (2019) 35:1117–1128

Conclusion 11. Canpolat M, Ceylan O, Per H, Koc G, Tumturk A, Kumandas S,

Patiroglu T, Doganay S, Gumus H, Unal E, Kose M, Gorkem SB,
Kurtsoy A, Ozturk MK (2015) Brain abscesses in children: results
Pediatric BA is an infrequent disease, still burdened by high of 24 children from a reference center in Central Anatolia, Turkey. J
morbidity and mortality, despite the advent of advanced diag- Child Neurol 30(4):458–467
nostic and therapeutic procedures. 12. Capua T, Klivitsky A, Bilavsky E, Ashkenazi-Hoffnung L, Roth J,
In the future, it would be interesting to evaluate Constantini S, Grisaru-Soen G (2018) Group a streptococcal brain
abscess in the pediatric population - case series and review of the
changes in the epidemiology of brain infections, taking literature. Pediatr Infect Dis J Feb 16:967–970. https://doi.org/10.
into account on one hand the progressive decrease of 1097/INF.0000000000001947
some predisposing factors, such as congenital heart dis- 13. Carpenter J, Stapleton S, Holliman R (2007) Retrospective analysis
ease (and immunosuppression induced by HIV infec- of 49 cases of brain abscess and review of the literature. Eur J Clin
Microbiol Infect Dis 26:1–11
tion), on the other hand the rise of antibiotic resistance,
14. Chowdhry SA, Cohen AR (2012) Citrobacter brain abscesses in
the decrease in vaccination coverage, and the growing neonates: early surgical intervention and review of the literature.
number of patients with iatrogenic immunodeficiencies. Childs Nerv Syst 28(10):1715–1722
Multicentric prospective studies are needed, in order to 15. Cole TS, Clark ME, Jenkins AJ, Clark JE (2012) Pediatric focal
achieve stronger evidence in particular on BA treatment. intracranial suppuration: a UK single-center experience. Childs
Nerv Syst 28:2109–2114
16. Costerus JM, Brouwer MC, van de Beek D (2018) Technological
Compliance with ethical standards advances and changing indications for lumbar puncture in neuro-
logical disorders. Lancet Neurol 17:268–278
Conflict of interest The authors declare that they have no conflict of 17. Dabbah H, Elias N, Bentur L (2009) Foreign-body aspiration as an
interest. aetiology for brain abscess. Arch Dis Child 94(8):601
18. Duarte MJ, Kozin ED, Barshak MB, Reinshagen K, Knoll RM,
Abdullah KG, Welling DB, Jung DH (2018) Otogenic brain ab-
scesses: a systematic review. https://doi.org/10.1002/lio2.150
References 19. Felsenstein S, Williams B, Shingadia D, Coxon L, Riordan A,
Demetriades AK, Chandler CL, Bassi S, Koutoumanou E,
1. Acar M, Sutcu M, Akturk H, Muradova A, Hancerli Torun S, Salman Stapleton S, Sharland M, Bryant PA (2013) Clinical and microbio-
N, Caliskan N, Izgi N, Somer (2016) Evaluation of Short-Term logic features guiding treatment recommendations for brain ab-
Neurological Outcomes in Children with Brain Abscesses. Turk scesses in children. Pediatr Infect Dis J 32(2):129–135
Neurosurg. https://doi.org/10.5137/1019-5149.JTN.18672-16.1 20. Frazier JL, Ahn ES, Jallo GI (2008) Management of brain abscesses
2. Arlotti M, Grossi P, Pea F, Tomei G, Vullo V, De Rosa in children. Neurosurg Focus 24(6):E8
FG, Di Perri G, Nicastri E, Lauria FNJ, Carosi G, Moroni 21. Gadgil N, Patel AJ, Gopinath SP (2013) Open craniotomy for brain
M, Ippolito G (2010) Consensus document on controversial abscess: a forgotten experience? Surg Neurol Int 4:3
issues for the treatment of infections of the central nervous 22. Gaskill SJ, Marlin AE (2008) Brain abscesses and encephalitis. In:
system: bacterial brain abscesses. Int J Infect Dis 14(4): Albright AL, Pollack IF, Adelson PD (eds) Principles and practice
S79–S92 of pediatric neurosurgery, ed 2. Thieme, New York, pp 1162–1181
3. Atiq M, Ahmed US, Allana SS, Chisti KN (2006) Brain abscess in 23. Gefen A, White A (2017) Asymptomatic pulmonary arteriovenous
children. Indian J Pediatr 73(5):401–404 malformations in children with hereditary hemorragic telangiecta-
4. Auvichayapat N, Auvichayapat P, Aungwarawong S (2007) sia. Pediatr Pulmonol 52:1194–1197
Brain abscess in infants and children: a retrospective study 24. Gelabert-González M, Serramito-García R, García-Allut A, Cutrín-
of 107 patients in Northeast Thailand. J Med Assoc Thail Prieto J (2008) Management of brain abscess in children. J Paediatr
90(8):1601–1607 Child Health 44(12):731–735
5. Bank DE, Carolan PL (1993) Cerebral abscess formation following 25. George R, Andronikou S, du Plessis J, du Plessis AM, Van Toorn R,
ocular trauma: a hazard associated with common wooden toys. Maydell A (2009) Central nervous system manifestations of HIV
Pediatr Emerg Care 9:285–288 infection in children. Pediatr Radiol 39(6):575–585
6. Baytan B, Evim MS, Güler S, Güneş AM, Okan M (2015) Acute 26. Goodkin HP, Harper MB, Pomeroy SL (2004) Intracerebral abscess
central nervous system complications in pediatric acute lympho- in children: historical trends at Children’s hospital Boston.
blastic leukemia. Pediatr Neurol 53(4):312–318 Pediatrics 113:1765–1770
7. Britt RH, Enzmann DR, Yeager AS (1981) Neuropathological and 27. Góralska K, Blaszkowska J, Dzikowiec M (2018) Neuroinfections
computerized tomographic findings in experimental brain abscess. caused by fungi. Infection. 46:443–459. https://doi.org/10.1007/
J Neurosurg 55:590–603 s15010-018-1152-2
8. Brizuela M, Pérez G, Martiren S, Varela Baino AN, Cedillo C, 28. Helweg-Larsen J, Astradsson A, Richhall H, Erdal J, Laursen A,
Ruvinsky S, Epelbaum C, Isasmendi A, Litterio M, Pinheiro JL, Brennum J (2012) Pyogenic brain abscess, a 15 year survey. BMC
Bologna R (2017) Absceso cerebral en niños: experiencia en diez Infect Dis 12:332
años en un hospital pediátrico de alta complejidad. Arch Argent 29. Jain A, Arora N, Meher R, Passey JC, Bansal R (2017)
Pediatr 115(4):e230–e232/e230 Intracranial complications of CSOM in pediatric patients: a
9. Brook I (2017) Microbiology and treatment of brain abscess. J Clin persisting problem in developing countries. Int J Pediatr
Neurosci 38:8–12 Otorhinolaryngol 100:128–131
10. Brouwer MC, Coutinho JM, van de Beek D (2014) Clinical char- 30. Kao KL, Wu KG, Chen CJ, Wu JJ, Tang RB, Chang KP, Wong TT
acteristics and outcome of brain abscess. Systematic review and (2008) Brain abscesses in children: analysis of 20 cases presenting
meta-analysis. Neurology 82:806–813 at a medical center. J Microbiol Immunol Infect 41:403–407
Childs Nerv Syst (2019) 35:1117–1128 1127

31. Kim JY, Park SY, Lee JM, Kim YK, Kim SY (2016) Intracranial 50. Nicolosi A, Hauser WA, Musicco M, Kurland LT (1991)
abscess as a complication of X-linked Agammaglobulinemia. Incidence and prognosis of brain abscess in a defined pop-
Childs Nerv Syst 32:2049–2051 u l a t i o n : O l m s t e d C o u n t y, M i n n e s o t a , 1 9 3 5 – 1 9 8 1 .
32. Kocherry XG, Hegde T, Sastry KVR, Mohanty A (2008) Efficacy Neuroepidemiology 10:122–131
of stereotactic aspiration in deep-seated and eloquent-region intra- 51. Ong CT, Tsai CF, Wong YS, Chen SCC (2017) Epidemiology of
cranial pyogenic abscesses. Neurosurg Focus 24(6):E13 brain abscess in Taiwan: a 14-year population-based cohort study.
33. Krzysztofiak A, Bozzola E, Lancella L, Quondamcarlo A, PLoS One 12(5):e0176705
Gesualdo F, Ugazio AG (2010) Linezolid therapy of brain abscess. 52. Özsürekci Y, Kara A, Cengiz AB, Çelik M, Özkaya-
Pediatr Infect Dis J 29(11):1063–1064 Parlakay A, Karadağ-Öncel E, Ceyhan M (2012) Brain ab-
34. Krzysztofiak A, Zangari P, De Luca M, Villani A (2017) Brain scess in childhood: a 28-year experience. Turk J Pediatr
abscesses: an overview in children. J Pediatr Infect Dis 14:002– 54(2):144–149
005. https://doi.org/10.1055/s-0037-1615786 53. Pal D, Bhattacharyya A, Husain M, Prasad KN, Pandey CM,
35. Kum RO, Özcan M, Ulusal T, Dündar G, Yılmaz YF, Ünal A Gupta RK (2010) In vivo proton MR spectroscopy evalua-
(2017) A rare complication of chronic otitis media: cerebellar ab- tion of pyogenic brain abscesses: a report of 194 cases. Am
scess. Turk Arch Otorhinolaryngol 55(3):140–143 J Neuroradiol 31:360–366
36. Lackner H, Sovinz P, Benesch M, Smolle-Jüttner F, Mokry M, 54. Pandey P, Umesh S, Bhat D, Srinivas D, Abraham R, Pillai S,
Schwinger W, Moser A, Urban C (2009) Management of brain Balasubramaniam A, Devi I, Sampath S, Chandramouli BA
abscesses in children treated for acute lymphoblastic leukemia. (2008) Cerebellar abscesses in children: excision or aspiration? J
Pediatr Blood Cancer 52(3):408–411 Neurosurg Pediatrics 1:31–34
37. Laulajainen-Hongisto A, Lempinen L, Färkkilä E, Saat R, 55. Patel AP, Masterson L, Deutsch CJ, Scoffings DJ, Fish BM (2015)
Markkola A, Leskinen K, Blomstedt G, Aarnisalo AA, Jero J Management and outcomes in children with sinogenic intracranial
(2016) Intracranial abscesses over the last four decades; changes abscesses. Int J Pediatr Otorhinolaryngol 79(6):868–873
in aetiology, diagnostics, treatment and outcome. Infect Dis (Lond) 56. Raffaldi I, Garazzino S, Castelli GG, Lipreri R, Lancella L,
48(4):310–316 Esposito S, Giannini AM, Montagnani C, Marseglia GL,
38. Lechowicz M, Dąbek K, Majewska U, Bekesińska-Figatowska M, Pignata C, Bernardi F, Tovo PA (2017) SITIP BRAIN
Borszewska-Kornacka MK, Bokiniec R (2017) Multiple brain ab- ABSCESSES REGISTRY. BRAIN abscesses in children:
scesses caused by Citrobacter Koseri in a preterm neonate - case an Italian multicentre study. Epidemiol Infect 145(13):
report. Pol J Radiol 82:837–841 2848–2855
39. Lee CG, Kang SH, Kim YJ, Shin HJ, Choi HS, Lee JH, Lee MH 57. Ratnaike TE, Das S, Gregson BA, Mendelow AD (2011) Review of
(2010) Brain abscess in Korean children: a 15-year single center brain abscess surgical treatment—78 years: aspiration versus exci-
study. Korean J Pediatr 53(5):648–652 sion. World Neurosurg 76(5):431–436
40. Liu J, Bai R, Li Y, Staedtke V, Zhang S, van Zijl PCM, Liu 58. Reiche W, Schuchardt V, Hagen T, Il’yasov KA, Billmann P, Weber
G (2018) MRI detection of bacterial brain abscesses and J (2010) Differential diagnosis of intracranial ring enhancing cystic
monitoring of antibiotic treatment using bacCEST. Magn mass lesions—role of diffusion-weighted imaging (DWI) and
Reson Med 80:662–671 diffusion-tensor imaging (DTI). Clin Neurol Neurosurg 112(3):
41. Louie JP, Osterhoudt KC, Christian CW (2000) Brain abscess fol- 218–225
lowing delayed endoscopic removal of an initially asymptomatic 59. Renier D, Flandin C, Hirsch E, Hirsch JF (1988) Brain
esophageal coin. Pediatr Emerg Care 16(2):102–105 abscesses in neonates. A study of 30 cases. J Neurosurg
42. Maraki S, Papadakis IS, Chronakis E, Panagopoulos D, Vakis A 69(6):877–882
(2016) Aggregatibacter aphrophilus brain abscess secondary to pri- 60. Roche M, Humphreys H, Smyth E, Phillips J, Cunney R,
mary tooth extraction: case report and literature review. J Microbiol McNamara E, O'Brien D, McArdle O (2003) A twelve-year review
Immunol Infect 49(1):119–122 of central nervous system bacterial abscesses; presentation and
43. Menon S, Bharadwaj R, Chowdhary A, Kaundinya DV, aetiology. Clin Microbiol Infect 9(8):803–809
Palande DA (2008) Current epidemiology of intracranial ab- 61. Rossi A, Martinetti C, Morana G, Severino M, Tortora D (2016)
scesses: a prospective 5 year study. J Med Microbiol 57(Pt Neuroimaging of infectious and inflammatory diseases of the pedi-
10:1259–1268 atric cerebellum and brainstem. Neuroimaging Clin N Am 26(3):
44. Mishra P, Prasad KN, Singh K, Sahu RN, Ojha BK (2016) 471–487
Association of ICAM-1 (K469E) and MCP-1 −2518 a>G gene 62. Sahbudak Bal Z, Eraslan C, Bolat E, Avcu G, Kultursay N,
polymorphism with brain abscess. J Neuroimmunol 292:102–107 Ozkinay F, Kurugol Z, Vardar F (2018) Brain abscess in
45. Muccio CF, Caranci F, D’Arco F, Cerase A, De Lipsis L, Esposito children: a rare but serious infection. Clin Pediatr (Phila)
G, Tedeschi E, Andreula C (2014) Magnetic resonance features of 57(5):574–579
pyogenic brain abscesses and differential diagnosis using morpho- 63. Sedaghat AR, Wilke CO, Cunningham MJ, Ishman SL
logical and functional imaging studies: a pictorial essay. J (2014) Socioeconomic disparities in the presentation of
Neuroradiol 41(3):153–167 acute bacterial sinusitis complications in children.
46. Muzumdar D, Jhawar S, Goel A (2011) Brain abscess: an overview. Laryngoscope 124(7):1700–1706
Int J Surg 9(2):136–144 64. Shachor-Meyouhas Y, Bar-Joseph G, Guilburd JN, Lorber A,
47. Naderi S, Nejat F, Shahjouei S, El Khashab M (2012) Cranial der- Hadash A, Kassis I (2010) Brain abscess in children – epidemiol-
mal sinus: presentation, complications and management. Pediatr ogy, predisposing factors and management in the modern medicine
Neurosurg 48(2):86–92 era. Acta Paediatr 99(8):1163–1167
48. Nathoo N, Nadvi SS, Narotam PK, van Dellen JR (2011) Brain 65. Sheehan JP, Jane JA, Ray DK, Goodkin HP (2008) Brain abscess in
abscess: management and outcome analysis of a computed tomog- children. Neurosurg Focus 24:E6. https://doi.org/10.3171/FOC/
raphy era experience with 973 patients. World Neurosurg 75(5–6): 2008/24/6/E6
716–726 66. Shovlin CL, Condliffe R, Donaldson JW, Kiely DG, Wort SJ, on
49. Nau R, Seele J, Djukic M, Eiffert H (2018) Pharmacokinetics and behalf of the British Thoracic Society (2017) British Thoracic
pharmacodynamics of antibiotics in central nervous system infec- Society clinical statement on pulmonary arteriovenous
tions. Curr Opin Infect Dis 31(1):57–68 malformations. Thorax 72:1154–1163
1128 Childs Nerv Syst (2019) 35:1117–1128

67. Sonneville R, Magalhaes E, Meyfroidt G (2017) Central nervous 70. Zhang Z, Cai X, Li J, Kang X, Wang H, Zhang L, Yan R, Gao N,
system infections in immunocompromised patients. Curr Opin Crit Liu S, Yue S, Zhang J, Yang S, Yang X (2016) Retrospective anal-
Care 23(2):128–133 ysis of 620 cases of brain abscess in Chinese patients in a single
68. Udayakumaran S, Onyia CU, Kumar RK (2017) Forgotten? Not center over a 62-year period. Acta Neurochir 158:733–739
yet. Cardiogenic brain abscess in children: a case series based re-
view. World Neurosurg 107:124–129
69. Van de Beek D, Cabellos C, Dzupova O, Esposito S, Klein M, Publisher’s note Springer Nature remains neutral with regard to
Kloek AT, Leib SL, Mourvillier B, Ostergaard C, Pagliano P, jurisdictional claims in published maps and institutional affiliations.
Pfister HW, Read RC, Resat Sipahi O, Brouwer MC, for the
ESCMID Study Group for Infections of the Brain (ESGIB)
(2016) ESCMID guideline: diagnosis and treatment of acute bacte-
rial meningitis. Clin Microbiol Infect 22:S37–S62