You are on page 1of 9

Q J Med 2002; 95:501–509

Original papers
Bacterial brain abscess: microbiological features,
epidemiological trends and therapeutic outcomes

Downloaded from by guest on 16 June 2019

C.-H. LU 1 , W.-N. CHANG 1 , Y.-C. LIN 2 , N.-W. TSAI 1 , P.-C. LILIANG 2 , T.-M. SU 2 ,
C.-S. RAU 2 , Y.-D. TSAI 2 , C.-L. LIANG 2 , C.-J. CHANG 3 , P.-Y. LEE 4 , H.-W. CHANG 5
and J.-J. WU 6
From the Departments of 1Neurology, 2Neurosurgery, 3Pediatric Neurology, and 4Pharmacy,
Chang Gung Memorial Hospital-Kaohsiung, Kaohsiung, 5Department of Biological Science,
National Sun Yat-Sen University, and 6Department of Medical Technology, National Cheng
Kung University Medical College, Tainan, Taiwan

Received 14 February 2002 and in revised form 22 April 2002

Background: Despite the advent of modern neuro- increased in the second study period. Viridans
surgical techniques, new antibiotics, and powerful streptococci and Klebsiella pneumoniae were the
imaging technologies, brain abscess remains a two prevalent pathogens associated with haemato-
potentially fatal central nervous system infection. genous spread. Metastatic septic abscess, a devas-
Aim: To determine the epidemiological trends, tating complication of K. pneumoniae septicaemia,
prognostic factors, and outcomes of bacterial brain frequently occurs in diabetic patients, with a high
abscess, to improve the therapeutic strategy for this mortality rate. Viridans streptococci were the
disease. most prevalent pathogens from infection in para-
Design: Retrospective hospital-based epidemiology nasal sinusitis, but no fatality occurred. In recent
study. years, head trauma and/or post-neurosurgical states
Methods: Over a period of 15 years (1986–2000), have become important predisposing factors,
123 patients were retrospectively identified as and nosocomial infections also play an important
having brain abscesses at Kaohsiung Chang Gung role.
Memorial Hospital. To compare changes over time, Discussion: Despite the availability of new anti-
the appearance of disease among our patients was biotics and the development of better neurosurgical
divided into two time periods: 1986–1993 and techniques, therapeutic outcomes of brain abscess
1994–2000. showed no significant change when comparing the
Results: The prevalence rate of brain abscesses two study periods, and only the presence of septic
caused by Gram-negative organisms significantly shock influenced outcome.

Despite the advent of modern neurosurgical tech- fatal central nervous system (CNS) infection.1–4 In
niques, new antibiotics, and new powerful imaging 1973, Kao et al.5 reported the clinical features of
technologies, brain abscess remains a potentially 26 cases of brain abscess with surgical intervention;

Address correspondence to Dr W.-N. Chang, Department of Neurology, Chang Gung Memorial Hospital, 123 Ta Pei
Road, Niao Sung Hsiang, Kaohsiung Hsien, Taiwan. e-mail:
ß Association of Physicians 2002
502 C.-H. Lu et al.

however, a large series study of brain abscess in Vitex). Patients were considered to have mixed
Taiwan is still lacking. A hospital-based study infections if at least two bacterial organisms
provides accurate information about localization were isolated from the initial cultures. Patients
of brain abscess, predisposing factors, clinical who were initially treated at other hospitals but
features, the prevalence rate of implicated bacterial subsequently transferred to our hospital for further
pathogens, and causes of fatality. In this study, therapy were also included in this study, with initial
we investigated differences between nosocomial clinical data collected at those hospitals used for
and community-acquired infections, predisposing analysis.
factors, clinical and neuroimaging findings, com- Brain abscesses were defined as nosocomial
plications, and therapeutic outcomes, in order according to the 1988 guidelines of the Centers
to improve the therapeutic strategies for this for Disease Control.7 Brain abscesses related to

Downloaded from by guest on 16 June 2019

potentially fatal disease. head trauma with skull fracture or neurosurgical
procedures, were classified as post-neurosurgical.
Other patients who presented with no distinctive
characteristics, and/or who had not undergone
Methods invasive procedures, were classified as having the
spontaneous form. Patients with evidence of brain
Over a period of 15 years (January 1986 to
abscesses not due to bacterial pathogens were
December 2000), 123 patients were retrospectively
excluded from this study.
identified as having brain abscesses at Kaohsiung
The mental status for patients was classified into
Chang Gung Memorial Hospital. The largest med-
two groups: (i) clear consciousness; and (ii) inatten-
ical centre in southern Taiwan, this facility is a
tion, confusion, clouded consciousness, stupor, or
2482-bed acute-care teaching hospital, which
coma. Surgical intervention and/or antibiotic ther-
provides both primary and tertiary referral care
apy were the mainstays of our treatment. Antibiotic
services. Southern Taiwan consists of two cities and
susceptibility was determined using the Kirby-Bauer
three counties (Kaohsiung Municipality and Hsien,
Tainan City and Hsien, and Pingtung Hsien) with disc diffusion method (Mueller-Hinton II agars;
a population of approximately 5 441 000 as of Becton Dickinson). Appropriate antimicrobial
October 1998.6 The annual population increase therapy was defined as the administration of one
for Taiwan has been reported to be 7.52%.6 To or more antimicrobial agents, which demonstrated
compare changes over time, the appearance of effectiveness against bacterial pathogens, as deter-
disease among our patients was divided into two mined from susceptibility tests, and which were
time periods: the first time period was 8 years capable of passing through the blood-brain barrier
(January 1986–December 1993) and the was in adequate amounts. Surgical treatment consisted
7 years (January 1994–December 2000). of either aspiration or excision of the abscess.
Criteria of bacterial brain abscess for inclusion Aspiration was used to aspirate the content of the
were: (i) characteristic computerized tomography abscess with a ventricular catheter via a burr hole
(CT) and/or magnetic resonance imaging (MRI) or small craniotomy, which left the capsule alone.
findings; (ii) evidence of brain abscess seen during Craniotomy and resection of the abscess was
surgery or histopathological examination; and (iii) defined as evacuation. In our institution, those
classical clinical manifestations including head- patients presenting in good neurological condition
ache, fever, localized neurological signs and/or with a well-formed abscess in the right hemisphere
consciousness disturbance.7–9 All materials from underwent evacuation, whereas those presenting
bacterial brain abscesses were cultured for aerobic significant surgical risks or those with an abscess
and anaerobic bacteria, mycobacteria, and fungi. that was deep-seated or in the left hemisphere
Aerobic culture media included blood, chocolate, underwent aspiration. Recovery was evaluated
EMB, CNA media, and thioglycolate broths. 3 months after patient discharge.
Anaerobic cultures were processed in Brucellae Data, including initial clinical manifestations,
agar plate, BBE/PEA media, and thioglycolate acquisition of infection, type of infection, numbers
broths. BHI agar, mycobiotic media, SDA media, of abscesses, various treatment regimens, and fatal-
and SCG media were used for fungal isolation. ities in the two study periods, were analysed using
Lowenstein-Jensen media and middle 7H11 media the x2 test or Fisher’s exact test. Data for age, and
were used for mycobacterial isolation. Organisms interval between onset and diagnosis between
were identified by API (Analytab Products), fatal and non-fatal groups were compared using
RapID-ANA II identification system (Innovative Diag- Student’s t-test. Stepwise logistic regression was
nostic Systems), and ID 32 C (VITEK, bioMérieux used to evaluate the relationship between clinical
Bacterial brain abscess 503

factors and the mortality rate adjusting for other In the first study period, 58% (40/69) of cases
potential confounding factors. All analyses were involved a single pathogen, with viridans strepto-
conducted using SAS (1990).10 cocci being the most prevalent (23%), followed
by Bacteroides species (18%), S. aureus (10%), and
Corynebacterium species (10%). Fourteen percent
Results (10/69) of patients had mixed infections, and
26% (18/69) of patients had a negative culture. In
The 123 patients included 92 males (mean age the second study interval, 70% (38/54) of cases
43 years; range 1 month–80 years) and 31 females were infected by a single pathogen, with viridans
(mean age 41 years; range 2 months–73 years) streptococci again the most prevalent (29%),
(Figure 1). Of these 123 patients, 103 had community- followed by K. pneumoniae (26%). Eleven percent

Downloaded from by guest on 16 June 2019

acquired infections, while the other 20 were (6/54) of patients had mixed infections, and 20%
diagnosed with nosocomial infections. Sixty-nine (11/54) of patients had a negative culture.
cases appeared in the first time period (January Clinical manifestations of these 123 patients are
1986–December 1993), with the other 54 in the listed in Table 3. Fever was found in 71 patients
second time period (January 1994–December 2000) and headache in 68. Forty-nine patients were
(Table 1). The ratios of community-acquired to admitted with disturbed consciousness. Twenty-
nosocomial meningitis for these two time periods one patients suffered septic shock, 13 of whom
were 58:11 and 44:10, respectively. subsequently died. Seizures and/or status epilepti-
Associations for portal of entry and causative cus occurred in 23 patients, of whom five died.
pathogens are listed in Table 2. Portal of entry for Concomitant bacterial meningitis was found in 29
infection in 94 culture-positive brain abscesses patients. Of these 29, 24 had community-acquired
included haematogenous spread (n = 32), postneuro- infections and five had acquired the infections
surgical states (n = 17), contiguous infection from nosocomially. Four of the five nosocomially-
parameningeal foci such as an otogenic origin infected patients had undergone neurosurgical
(n = 13), paranasal sinusitis (n = 9), and unknown procedures as the underlying condition; the other
(n = 24). In this study, Klebsiella pneumoniae and had had a stroke. Of these 29 cases with con-
viridans streptococci were the two most prevalent comitant bacterial meningitis, five were identified
pathogens associated with haematogenous spread, as K. pneumoniae infection and revealed multiple
accounting for 52% (15/29) of the episodes. septic abscesses on further examination, including
Staphylococcus aureus, K. pneumoniae, and viridans liver abscess, renal abscess, lung abscess, para-
streptococci were the three common pathogens spinal abscess, endophthalmitis, pyomyositis, purulent
in postneurosurgical states, accounting for 47% pericarditis, and septic arthritis. All five of these
(7/15) of the episodes. Viridans streptococci were patients were diabetic. Other clinical manifes-
the most prevalent pathogens from paranasal sinusi- tations in these 123 cases included hemiparesis,
tis infections, while Proteus species were the stiff neck, facial palsy, hemiparaesthesia, and
common pathogens of otogenic origin, in brain disturbance of speech and vision.
abscesses with polymicrobial infection. The mean The locations of brain abscesses of these 123
interval between onset of symptoms to detection cases were supratentorial in 118 cases and infra-
of brain abscess was 21 days (range 1–700 days). tentorial in five (Table 4). In total, 99 (80.5%) had a
single brain abscess and 24 (19.5%) had multiple
brain abscesses. The most common sites for brain
abscess were the frontal lobe (33%), followed by
the temporal lobe (20%) and temporo-parietal lobe
(10%). The left hemisphere was involved more than
the right with 63 on the left, 47 on the right, two on
the midline, six bilaterally on both hemispheres,
and five on the infratentorial area.
Antimicrobial therapy, with or without surgical
intervention (aspiration or total excision), was the
cornerstone of treatment in these 123 patients.
Seventeen patients received antimicrobial therapy
alone, four of whom had multiple pyogenic brain
abscesses; the other 13 were associated with
poor systemic conditions. Penicillin G with chlor-
Figure 1. Age distribution of patients with brain abscess. amphenicol was the mainstay of initial empiric

Table 1 Causative organisms, January 1986 through December 2000

Organisms Jan 1986–Dec 1993 (n = 69) Jan 1994–Dec 2000 (n = 54) Total Total deaths
n = 123 n = 21
Community Nosocomial Deaths Community Nosocomial Deaths
n = 58 n = 11* n = 13 n = 44 n = 10** n=8

Gram-negative bacilli (n = 27)

Klebsiella pneumoniae 2 1 2 7 3 2 13 4
Pseudomonas aeruginosa 1 0 0 1 1 1 3 1
Escherichia coli 1 0 1 2 0 0 3 1
Salmonella spp. 0 1 0 1 0 0 2 0
Proteus spp. 0 1 0 1 0 0 2 0
Enterobacter spp. 0 1 0 0 0 0 1 0
Klebsiella oxytoca 0 0 0 1 0 0 1 0
Vibrio cholerae non-O1 1 0 1 0 0 0 1 1
Pasteurella spp. 0 0 0 1 0 0 1 0
Streptococcus species (n = 21)
Viridans streptococci 8 1 0 10 1 0 20 0
non-A, non-B, and non-D streptococci 0 0 0 1 0 0 1 0
C.-H. Lu et al.

Staphylococcus species (n = 9)
Staphylococcus aureus 3 1 2 1 1 0 6 2
Coagulase-negative Staphylococcus 1 0 1 2 0 1 3 2
Corynebacterium spp. (n = 4) 4 0 1 0 0 0 4 1
Anaerobes (n = 17)
Bacteroides spp. 7 0 2 0 0 0 7 2
Fusobacterium spp. 1 1 0 1 0 0 3 0
Peptostreptococcus spp. 0 0 0 1 0 0 1 0
Peptococcus spp. 3 0 1 0 0 0 3 1
Propionibacterium spp. 2 0 0 1 0 0 3 0
Mixed bacterial pathogens (n = 16) 8 2 0 4 2 2 16 2
Negative culture (n = 29) 16 2 2 9 2 2 29 4

*All eleven patients contracted the infection after trauma or neurosurgical procedures. **Except for two patients with K. pneumoniae infection, all the others contracted the
infection after trauma or neurosurgical procedures.

Downloaded from by guest on 16 June 2019

Bacterial brain abscess 505

Table 2 Relation between predisposing factors and causative pathogens

Organisms Haematogenous Postneurosurgical Contiguous infection Unknown

spread states***
Otogenic Paranasal
infection sinusitis

Aerobic Gram-negative bacilli

Klebsiella pneumoniae 9* (3) 2 1 0 1 (1)
Pseudomonas aeruginosa 0 1 (1) 0 0 2
Escherichia coli 2 (1) 0 0 0 1
Salmonella spp. 1 1 0 0 0

Downloaded from by guest on 16 June 2019

Proteus spp. 0 1 1 0 0
Enterobacter spp. 0 1 0 0 0
Klebsiella oxytoca 0 0 1 0 0
Vibrio cholerae non-O1 1(1) 0 0 0 0
Pasteurella spp. 0 0 0 0 1
Streptococcus spp.
Viridans streptococci 6** 2 0 6 6
non-A, non-B, and non-D streptococci 1 0 0 0 0
Staphylococcus spp.
Staphylococcus aureus 1 (1) 3 0 0 2
Coagulase-negative staphylococci 2 (1) 0 0 0 1 (1)
Corynebacterium spp. 2 (1) 0 0 0 2
Bacteroides spp. 1 1 2 (1) 1 2 (1)
Fusobacterium 1 1 1 0 0
Peptostreptococcus spp. 1 0 0 0 0
Peptococcus spp. 1 0 1 0 1 (1)
Propionibacterium spp. 0 0 0 2 1
Mixed bacterial pathogens 31 (1) 42 53 (1) 0 44
Negative culture 0 2 5 (2) 2 20 (2)

Deaths are shown as figures in parentheses. *Seven of the nine had diabetes mellitus as the underlying disease. **Of these
six, five had heart diseases, including infectious endocarditis (1), tetralogy of Fallot (2), Ebstein’s anomaly (1) and atrial
septal defect (1). ***Included craniotomy due to head trauma (6), intracerebral or subdural haematoma (4), brain tumour (4),
ventriculoperitoneal shunt for hydrocephalus (1). 1Bacteroides melaninogenicus, non-A, non-B, and non-D streptococci;
viridans streptococci, Pseudomonas species; Peptostreptococcus, Bacteroides melaninogenicus. 2Proteus mirabilis,
Enterobacter cloacae; Staphylococcus aureus, viridans streptococci; Proteus mirabilis, Escherichia coli, Serratia
marcescens; Peptostreptococcus, Bacteroides fragilis. 3Proteus vulgaris, Klebsiella oxytoca; Bacteroides fragilis, Proteus
mirabilis, Pseudomonas spp., Enterococcus spp., Group D streptococci; Bacteroides spp; Escherichia coli, Proteus mirabilis,
Corynebacterium, Bacteroides spp.; Peptococcus, Veillonella spp.; Enterococcus, Staphylococcus aureus, non-A, non-B,
and non-D streptococci; viridans streptococci, Klebsiella pneumoniae. 4Bacillus, Fusobacterium nucleatum; Prevotella
intermedia, Fusobacterium nucleatum, viridans streptococci; Bacteroides fragilis, Proteus mirabilis; viridans streptococci,

antimicrobial treatment of bacterial brain abscess both antimicrobial therapy and aspiration died. All
unless Staphylococcus or Gram-negative bacilli six patients who received antimicrobial therapy and
were more likely to be the responsible pathogens. aspiration with subsequent total excision survived,
The choice of final antibiotics was guided by the while there were six fatalities among the 17 patients
final culture results. With this regimen of therapy, who received antimicrobial therapy alone.
12 patients experienced recurrence of brain Mortality rates of patients by group of causative
abscesses, and they received repeated aspiration organism were as follows: Gram-negative bacilli,
and/or total excision of brain abscesses. In total, 27% (7/26); Streptococcus spp., 0% (0/21);
102/123 patients survived. Mortality rates for the Staphylococcus spp., 33% (3/9); Corynebacterium
first and second study periods were 19% and 15%, spp., 25% (1/4); anaerobic pathogens, 17% (3/18);
respectively (Figure 2). Nine of 67 patients who mixed bacterial species, 13% (2/16); and negative
received both antimicrobial therapy and total cultures, 14% (4/29) (Table 1). Of the 21 patients
excision died. Six of 33 patients who received who died, 13 had fulminant clinical courses and
506 C.-H. Lu et al.

Table 3 Clinical manifestations

Community-acquired Nosocomial Total (n = 123)

(n = 102) (d = 20) (n = 21) (d = 1) (d = 21)

Fever/chills 60 11 71
Headache 63 5 68
Disturbed consciousness 44 15 59
Hemiparesis 44 10 54
Nausea/vomiting 35 4 39
Stiff neck 33 3 36
Speech disturbance 22 3 25

Downloaded from by guest on 16 June 2019

Seizure 18 5 23
Septic shock 20 1 21
Visual disturbance 17 1 18
Facial palsy 8 1 9
Hemiparaesthesia 7 1 8

d, deaths.

Table 4 Localization of brain abscesses deterioration of systemic conditions, three had

intracerebral haemorrhage after surgical interven-
Localization Number of cases tions, and four had brain swelling with subsequent
(n = 123) (%) herniation. Among the 102 survivors, 64 resumed
normal lives or had mild mental abnormalities, 27
Single site
had focal neurological deficits with moderate
Frontal lobe 33
Temporal lobe 20 disability, and 11 had severe disability or were in
Temporo-parietal area 10 a persistent vegetative state.
Basal ganglion 8 The potential prognostic factors of these 123
Occipital lobe 6 patients are listed in Table 5. According to the
Parieto-occipital area 6 statistical analysis, only the presence of septic shock
Fronto-parietal area 5 (p-0.001) had a significant influence on therapeu-
Parietal lobe 4 tic outcomes. Variables, including age at infection,
Cerebellum 4 sex, acquisition of infection, type of infection, type
Cerebellopontine angle 1 of treatment, headache, fever, septic shock, seizure,
Third ventricle 1
concomitant meningitis, mental status, and the
Intrasellar area 1
interval between onset and diagnosis, were used
Total 99 (80.5%)
Multiple sites 24 (19.5%) in the stepwise logistic regression, and the presence
of septic shock (p = 0.0001) was the only factor
predictive of mortality.

The causative pathogens of bacterial brain abscess
vary with time period, geographic distribution, age,
underlying medical and/or surgical conditions, and
mode of infection.1–5 The relative frequency of
mixed bacterial infections and limitations of anae-
robic culture techniques also influence the patho-
gen prevalence in different studies.1–14 A study
from China reported that Gram-negative bacilli,
with Proteus, Enterobacter and Alcaligenes species
being common, accounted for 21% (82/400) of the
implicated pathogens of brain abscess from 1952
Figure 2. Numbers of fatal and non-fatal cases of to 1972.14 In one early study of brain abscess from
bacterial brain abscess, distributed over 15 years. Taiwan, Gram-negative bacilli, Staphylococcus
Bacterial brain abscess 507

Table 5 Prognostic factors

Prognostic factor Patients Dead Alive p OR 95%CI Statistical test

(1) Age at infection (mean, years)

All 49.0 40.8 0.079 Student’s t
(2) Sex
Male 92 16 76 0.875 1.10 0.37–3.28 x2
Female 31 5 26
(3) Acquisition of infection
Community 102 20 82 0.121 4.88 0.62–38.54 Fisher’s exact
Nosocomial 21 1 20

Downloaded from by guest on 16 June 2019

(4) Type of infection
Spontaneous meningitis 104 20 84 0.192 4.29 0.54–34.03 Fisher’s exact
Postneurosurgical meningitis 19 1 18
(5) Types of treatment
Excision and antimicrobial 66 9 57*** 0.165 2.15 0.11–41.33 Fisher’s exact
Aspiration and antimicrobial 34 6 28 2.97 0.15–59.55
Antimicrobial alone 17 6 11 7.35 0.35–152.53
Aspiration then excision plus 6 0 6
(6) Clinical manifestations
Yes 68 10 58 0.477 0.69 0.27–1.77 x2
No 55 11 44
Yes 71 15 56 0.163 2.05 0.74–5.72 x2
No 52 6 46
Septic shock
Yes 21 13 8 -0.001** 19.10 6.11–59.63 Fisher’s exact
No 102 8 94
Yes 23 5 18 0.542 1.46 0.47–4.50 Fisher’s exact
No 100 16 84
Focal neurological deficits*
Yes 78 14 64 0.734 1.19 0.44–3.20 x2
No 45 7 38
Concomitant meningitis
Yes 29 3 26 0.40 0.49 0.13–1.79 x2
No 94 18 76
Mental status
Group I 64 11 53 0.972 1.02 0.40–2.60 x2
Group II 59 10 49
(7) Interval between onset and diagnosis (mean, days)
All 10.6 23.2 0.11 Student’s t

*Focal neurological deficits included hemiparesis, facial palsy, hemiparesthesia, and disturbance of speech and vision.
**Statistically significant. ***Relative to aspiration then excision plus antibiotics treatment.

species, and Streptococcus species accounted for remained the most prevalent pathogens in both
40% (10/25) of the episodes,5 and these three time periods. Among the Gram-negative bacilli,
groups of pathogens also accounted for 46% K. pneumoniae, E. coli, P. aeruginosa, and other
(56/123) of the pathogens implicated in this study. rare Gram-negative organisms accounted for more
Of the Streptococcus species, viridans streptococci than 21% of implicated pathogens. There was a
were the most frequently found, at 13% (9/69) significantly increased prevalence of Gram-negative
and 22% (12/54) of the implicated pathogens in organisms of from 13% (9/69) in the first to 31%
the first and second study time periods, respec- (17/54) in the second time period.
tively. Although there was a change in the relative This study shows that clinical presentations and
frequency of pathogens, viridans streptococci underlying disorders vary greatly in brain abscesses
508 C.-H. Lu et al.

due to different species. K. pneumoniae was the However, the presence of fever, headache, seizure,
most prevalent pathogen associated with haemato- and focal neurological signs showed no statistical
genous spread or postneurosurgical states. Multiple significance on the prognosis in our study. The
metastatic septic abscesses were common in interval between onset of symptom and signs and
patients with brain abscess due to K. pneumoniae. diagnosis was of no statistical significance in our
Although the reason is unknown, this association study, as in a previous study.12
of Klebsiella infection with metastatic abscesses Metastatic abscesses are commonly located in
has been noted in other reports from Taiwan.15–17 the parietal, frontal, or temporal lobes,12,14 and
Of the brain abscesses caused by Streptococcus multiple brain abscesses are usually the result
species, viridans streptococci, the most prevalent of metastatic spread from remote primary foci,
members of the flora in the upper respiratory tract, accounting for 6% to 22% of cases.12,14 In this

Downloaded from by guest on 16 June 2019

gastrointestinal tract, and female genital tract, study, the most common cause of brain abscesses
accounted for 95% of strains. This study also was haematogenous spread, accounting for 26%
shows that viridans streptococci are the prevalent (32/123) of the episodes, a figure higher than the
pathogens with haematogenous spread secondary study from China.14 Almost all abscesses were
to cardiogenic origin or from infection of paranasal found in the frontal, temporal, and parietal lobes,
sinusitis; this finding is consistent with those of and multiple abscesses were present in 19.5%
other reports.18,19 In this study, no deaths due (24/123). Although brain abscesses associated with
to viridans streptococci occurred, which may well neurosurgical procedures were once considered
be due to the low virulence of viridans strepto- uncommon,2,14 they accounted for 12% of the
cocci.18,20 In the brain abscesses of otogenic origin, episodes in this study. The increasing frequency
Proteus species were the most commonly impli- of neurosurgical procedures may be due to the
cated pathogens among the brain abscesses with increasing number of neurosurgeons, the improve-
polymicrobial infection, and this phenomenon was ment of neurosurgical facilities, and the large
also noted in other studies.13,21 This study also number of patients with head injury from motor-
showed that S. aureus, viridans streptococci, and K. cycle accidents. Unlike the data shown in one study
pneumoniae were the most prevalent pathogens in from China,14 where the majority of the abscesses
patients with post-neurosurgical states. were temporal and cerebellar, suggesting otogenic
Regarding the portal of entry, brain abscess is spread, our patients had more frontal disease.
almost always secondary to a focus of suppuration This may reflect the trend towards better and
elsewhere in the body, and may develop either more aggressive management of otogenic disorders
by a contiguous focus of infection, head trauma, between the 1970s and the 1990s.
or haematogenous spread from a distant focus. Treatment of brain abscess requires a combina-
Although head trauma or neurosurgical procedures tion of antimicrobials, surgical intervention, and
were once unusual as causes of brain abscesses, eradication of primary infected foci. In Taiwan,
they were important in this study. As a previous initial empirical antibiotics with third-generation
report12 demonstrated and our study confirms, when cephalosporins in combination with metronidazole,
haematogenous spread is the source of the brain should be considered for the majority of abscesses
abscess, this infection is frequently community- cases resulting from infection with Gram-negative
acquired, and is commonly associated with under- bacilli and streptococcal species. Different the-
lying diseases and with more fulminant courses, rapeutic regimens show no significant statistic
as evidenced by higher rates for shock and death. differences in prognosis in other reports,12 and
By contrast, postneurosurgical causes and a were also of no statistical significance in this
contiguous focus of infection are associated with a study.
favourable outcome. In summary, the clinical presentations, under-
In our 123 cases with brain abscesses; fever, lying conditions and therapeutic outcomes vary
headache, altered consciousness, and hemiparesis greatly in brain abscesses, according to the different
were the most common manifestations, occurring pathogens. In recent years, head trauma and/or
in 58%, 55%, 50%, and 44% of our patients, post-neurosurgical states have become important
respectively. The presence of focal neurological predisposing factors of brain abscess, and noso-
findings (e.g. hemiparesis, hemisensory deficits, and comial infections are also important. Despite the
aphasia) depends on the location of the abscess, availability of new antibiotics and the development
and these signs were seen in approximately 63% of better neurosurgical techniques, therapeutic
(78/123) of our cases, while 37% (45/123) had outcomes of brain abscess have not shown a
no localized signs. The presence of septic shock statistically significant change, and only the
significantly influenced the outcome of our patients. presence of septic shock is a prognostic factor.
Bacterial brain abscess 509

12. Chun CH, Johnson JD, Hofstetter M, Raff MJ. Brain abscess: a
References study of 45 consecutive cases. Medicine (Baltimore) 1986;
1. Bidzinski J, Koszewski W. The value of different methods
of treatment of brain abscess in the CT era. Acta Neurochir 13. de Louvis J, Gortvai P, Hurley R. Bacteriology of abscesses of
the central nervous system: a multicentre prospective study.
(Wien) 1990; 105:117–20.
Br Med J 1977; 2:981–4.
2. Mathisen GE, Johnson JP. Brain abscess. Clin Infect Dis
14. Yang SY. Brain abscess: a review of 400 cases. J Neurosurg
1997; 25:763–81.
1981; 55:794–9.
3. Rosenblum ML, Hoff JT, Norman D, Weinstein PR, Pitts L.
Decreased mortality from brain abscess since advent of 15. Tang LM, Chen ST, Hsu WC, Chen CM. Klebsiella
meningitis in Taiwan: an overview. Epidemiol Infect 1997;
computerized tomography. J Neurosurg 1978; 59:658–63.
4. Stephanov S. Surgical treatment of brain abscess. Neurosurg
1988; 22:724–30. 16. Wang JH, Liu YC, Lee SS, Yen MY, Chen YS, Wang JH,

Downloaded from by guest on 16 June 2019

Wann SR, Lin HH. Primary liver abscess due to
5. Kao MC. Brain abscess: a clinical analysis of 26 cases with Klebsiella pneumoniae in Taiwan. Clin Infect Dis 1998;
surgical treatment. Bull Tokyo Med Dent Univ 1973; 6:1434–8.
17. Wang LS, Lee FY, Cheng DL, Liu CY, Hinthorn DR, Lost PM.
6. Ministry of the Interior, ROC. Taiwan-Fukien Demography Klebsiella pneumoniae bacteremia: analysis of 100 episodes.
Quarterly 1999; 25(4):14–16. J Formosan Med Assoc 1990; 89:756–63.
7. Garner JS, Jarvis WR, Emori TG, Horan TC, Hughes JM. CDC 18. Enting RH, de Gans J, Blankevoort JP, Spanjaard L.
definitions for nosocomial infections. Am J Infect Control Meningitis due to viridans streptococci in adult. J Neurol
1988; 16:128–40. 1997; 244:435–8.
8. New PFJ, Davis KR, Ballantine HT Jr. Computed tomography 19. Pruitt A, Rubin MH, Karchmer AW, Duncan GW.
in cerebral abscess. Radiology 1976; 121:641–6. Neurologic complications of bacterial endocarditis.
9. Rosenblum ML, Hoff JT, Norman D, Edwards MS, Berg BO. Medicine (Baltimore) 1978; 57:329–43.
Nonoperative treatment of brain abscesses in selected 20. Cabello C, Viladrich PF, Corredoira J, Verdaguer R, Ariza J,
high-risk patients. J Neurosurg 1980; 52:217–25. Gudiol F. Streptococcal meningitis in adult patients: current
10. SAS. SAS User’s Guide. SAS Statistical Institute, Cary NC, epidemiology and clinical spectrum. Clin Infect Dis 1999;
1990. 28:1104–8.
11. Beller AJ, Sahar A, Praiss I. Brain abscess: review of 89 cases 21. Kangsanarak J, Navacharoen N, Fooanant S, Ruckphaopunt
over a period of 30 years. J Neurol Neurosurg Psych 1973; K. Intracranial complications of suppurative otitis
36:757–68. Media: 13 years’ experience. Am J Otol 1995; 16:104–9.