J Neurosurg

J Neurosurg
Pediatrics
Pediatrics
6:000–000,
6:38–42, 2010

The treatment and outcome of postmeningitic subdural

empyema in infants
Clinical article

Zhuo-Hao Liu, M.D.,1 Nan-Yu Chen, M.D., 2 Po-Hsun Tu, M.D.,1 Shih-Tseng Lee, M.D.,1
and Chieh-Tsai Wu, M.D.1
1
Department of Neurosurgery, and 2Division of Infectious Diseases, Department of Internal Medicine, Chang
Gung Memorial College of Medicine, Hospital and Chang Gung University, Linkou, Taiwan

Object. The management of subdural empyema (SDE) has been debated in the literature for decades. Cranioto-
my and bur hole drainage have been shown to achieve a favorable outcome. However, there is a lack of comparative
data for these modes of management of SDE subsequent to meningitis in infants.
Methods. The authors conducted a retrospective review of 33 infants identified with SDE due to meningitis at
the Department of Neurosurgery, Chang Gung Memorial Hospital between 2000 and 2006. Preoperative clinical
presentation, duration of symptoms, radiological investigations, CSF data, and postoperative outcome were analyzed
and compared between these 2 surgical groups.
Results. At diagnosis, there were no differences between the groups in age, weight, degree of consciousness,
CSF analysis, or duration of fever. The outcome data showed no difference in the number of days until afebrile, num-
ber of days of postsurgical antibiotic treatment, neurological outcome, recurrence rate, or complication rate. There
was only 1 death in the series.
Conclusions. Subdural empyema due to meningitis in infants is unique with respect to the pathophysiology, pre-
sentation, and treatment of SDE. Early detection and removal of SDE provide a favorable outcome in both surgical
intervention groups. Bur hole drainage is less invasive, and it is possible to expect a clinical outcome as good as with
craniotomy in postmeningitic SDE. (DOI: 10.3171/2010.4.PEDS09433)

Key Words      •      subdural empyema      •      surgical treatment      •      bur hole      •     

craniotomy      •      infant meningitis      •      outcome

A
n SDE is one of the most serious neurosurgical
emergencies. It is a fulminant form of intracranial
infections and accounts for 15–20% of all local-
ized intracranial infections.6,13 A delay in diagnosis and
treatment can lead to significant morbidity and mortal-
ity. The incidence of SDE in children is lower than that
in adults.13 Although meningitis is an unusual cause of
SDE in adults, it is the most frequent cause in infants.
Postmeningitic SDE in infants is unique with respect to
the pathophysiology, presentation, and treatment. The
mechanism of the development of SDE is through infec-
tion of sterile, reactive subdural effusions, which appear
in 40–60% of all cases of meningitis in infants.14 The op-
timal surgical approach in patients with postmeningitic
SDE (drainage via craniotomy or bur hole) is controver-
sial.2,5,12
Abbreviations used in this paper: GCS = Glasgow Coma Scale;
GOS = Glasgow Outcome Scale; SDE = subdural empyema.
Although there have been various studies on SDE,
there is a paucity of literature regarding the surgical
treatment results of SDE related to infant meningitis.
However, no current data support a specific duration of
antibiotic therapy and clinical outcome between the dif-
ferent surgical approaches. Therefore, we conducted a
retrospective study comparing craniotomy to bur holes
with subdural drainage surgery in infants with postmen-
ingitic SDE.
Methods
This retrospective case series was conducted at the
Chang Gung Memorial Hospital. The study population
consisted of patients younger than 2 years with a diag-
nosis of SDE who underwent surgical intervention. All
medical records and radiological materials were re-
viewed. An empyema was defined as a pus accumulation
in the subdural space as noted on enhanced CT scanning.

38 J Neurosurg: Pediatrics / Volume 6 / July 2010

Surgical outcome of postmeningitic subdural empyema in infants

Frank pus was identified during surgery (bur hole with TABLE 1: Patient characteristics at diagnosis*
subdural drainage and/or craniotomy). Between 2000 and
2006, 41 patients had undergone surgical treatment due Group
to an SDE. We excluded patients with isolated cerebral Bur Hole Craniotomy p
abscess, sinusitis, infection that developed after neurosur- Variable (18 patients) (15 patients) Value
gery, head trauma, otorhinolaryngeal infection, and other
mean age (mos) 5.2 ± 0.89 4.9 ± 0.83 0.79
systemic infections. All cases of SDE in infants were due
to meningitis, and the CSF culture was positive at the sex (M/F) 11:7 11:4
time when meningitis was diagnosed. mean weight (g) 6651 ± 520 6651 ± 564 0.43
The medical records, surgical notes, and discharge GCS score
summaries of all the patients were analyzed retrospec-    15 8 6
tively. On diagnosis, age, weight, GCS scores, microbio-    13 or 14 5 7
logical studies, duration of symptoms before intervention,
   9–12 4 1
and CSF data before intervention were collected. After
intervention, the number of days of postoperative antibi-    3–8 1 1
otic treatment, the number of days until the patient was mean preop duration of fever 11.6 ± 0.79 12.5 ± 1.96 0.69
afebrile, the need for a second procedure, and the GOS   (days)
score were reviewed retrospectively. mean CSF leukocyte count 4391 ± 2407 5442 ± 3886 0.82
The data were analyzed using the Student t-test and   (/µl)
the chi-square test. Statistical significance was set at p < mean CSF protein level 457 ± 152 202 ± 32 0.12
0.05.   (mg/dl)
mean CSF glucose (mg/dl) 25 ± 8.3 11 ± 3.1 0.12
Results SDE involved (unilat/bilat) 2/16 8/7
mean SDE size (mm) 10.35 ± 1.15 11.49 ± 1.64 0.57
Patient Characteristics
Between January 2000 and December 2006, 41 pa- *  Values represent the number of patients unless stated otherwise.
tients were diagnosed with SDE and underwent surgical Mean values are presented as the mean ± SEM.
intervention. Eight patients were excluded from the study.
Of these 8 patients, the SDE was diagnosed after neu-
rosurgical intervention in 4, after sinusitis-related SDE mon site of pus collection. In 1 case, it was associated
without CSF culture in 2, and after meningitis subsequent with pus collection in the parafalcine region. There was
to chronic subdural hemorrhage in 2. Thirty-three pa- no significant difference in the size of the SDE between
tients were enrolled in the study (22 boys and 11 girls, the 2 groups. However, SDEs that involved the bilateral
age range 10 days–16 months [mean 5.1 ± 3.49 months]). convexity were more prominent in the bur hole group
Medical and surgical treatment data of the patients with than in the craniotomy group.
early-onset infection are shown in Table 1. At diagnosis,
there were no significant differences between the groups Microbiological Findings
in terms of age, body weight, CSF white blood cell count, The CSF culture was positive in all patients. Haemo-
CSF protein level, CSF sugar level, level of conscious- philus influenzae type b was the most common pathogen
ness, and days since fever onset. (10 patients [32.3%]), followed by Streptococcus pneumo-
nia, Escherichia coli, and group B streptococci. Organ-
Clinical Presentation isms identified in the CSF are shown in Table 3. The pus
In 33 patients, the main clinical features observed cultures were positive in 7 patients in the bur hole group
were fever in 31 (94%), seizure in 19 (58%), decreased and in 4 patients in the craniotomy group.
level of consciousness in 19 (58%), neurological deficit in
7 (21%), and bulging fontanel in 7 (21%). The majority of TABLE 2: Clinical features at presentation
patients presented with fever, but 2 patients did not de-
velop fever. Seven patients had a neurological deficit; 4 of Group
them had hemiparesis and 3 of them had a cranial nerve Bur Hole Craniotomy
palsy. Complete data of both groups regarding the clinical Feature (18 patients) (15 patients)
presentation at diagnosis are shown in Table 2.
fever 17 14
Imaging Investigations depressed level of conscious (GCS 10 9
An intracranial SDE was diagnosed when contrast-   score <15)
enhanced CT scans showed the presence of a hypo- to seizure 11 8
isodense subdural collection with enhancement.21 All pa- bulging fontanel 4 3
tients underwent either CT scanning or MR imaging in neurological deficit 5 2
the postoperative and follow-up periods. The distribution
vomiting 3 5
of SDEs according to their sites is given in Table 1. The
cerebral convexity (frontotemporal) was the most com- neck rigidity 5 1

J Neurosurg: Pediatrics / Volume 6 / July 2010 39


TABLE 3: Patient microbiology
Group
Bacterium Bur Hole Craniotomy
Haemophilus influenza type b 3 7
Streptococcus pneumoniae 5 1
Escherichia coli 3 2
group B streptococci 3 1
Neisseria menigitidis 1 2
Salmonella spp. 2 2
Staphylococcus aureus 1 0
Treatment Course
All patients received an antibiotic regimen for menin-
gitis, which was modified if required, following the culture
and sensitivity report; the patients underwent follow-up
with CSF culture and imaging studies. We then performed
ultrasonography for the initial evaluation of the subdural
fluid. The size of the infectious collection was determined
readily by using imaging studies. In cases in which pro-
gressive subdural fluid accumulation and SDE was highly
suspected, CT scanning of the brain with enhancement was
performed for a definite diagnosis. One of the surgical in-
dications in our series was progressive SDE accumulation
despite antibiotic treatment, which indicated poor medical
treatment control (Fig. 1). The other surgical indication was
that the patient suffered high intracranial pressure due to
the mass effect of SDE. Focal neurological abnormalities
such as hemiparesis, dysphasia, cranial neuropathies, and
seizure, have been shown to be due to local pressure on the
underlying cortex because of the lesion (Fig. 2). Following
diagnosis, all patients underwent an emergency operation.
The initial surgical intervention (bur hole with subdural
drainage and craniotomy) varied between the individual
cases and depended on the surgeon’s preference. The mean
duration between fever onset and surgical intervention was
11.6 days and 12.5 days in the bur hole and craniotomy
groups, respectively. Antiepileptic medication was started
in patients who presented with seizures. One case was asso-
ciated with ventriculitis complicated by hydrocephalus and
required the placement of an external ventricular drain.
Fig. 1.  This 4-month-old boy suffered tachypnea and vomiting. Fever workup and CSF study revealed Streptococcus pneu-
moniae meningitis. The fever persisted despite antibiotic treatment with vancomycin and Claforan for 2 weeks.  A and B: Brain
CT scanning without (A) and with (B) contrast showed subdural fluid collection over the bilateral frontotemporal region, with the
enhancement of the inner membrane.  C: Postoperative MR image demonstrating no subdural effusion collection.
40
Z. H. Liu et al.
Outcome
Postoperative CT scans were obtained, and 5 patients
were subjected to further surgery. Two patients in the
Total craniotomy group and 3 patients in the bur hole group
10 had recollection of the SDE and required surgical inter-
6 vention. These patients underwent craniotomy, and no
5
further recurrences were observed. The outcome data are
shown in Table 4. There was no difference in the duration
4 of postoperative antibiotic treatment after intervention,
3 GOS scores, or number of days until afebrile. One pa-
4 tient died in the craniotomy group during the admission
1 course of severe sepsis and multiple organ failure.
The analysis of complications in both groups revealed
that the majority of complications were wound infection
(3 patients in the bur hole group and 1 patient in the cran-
iotomy group). Seven patients had neurological sequelae
in the follow-up period, 2 patients presented with hemi-
paresis, and 5 patients suffered from hearing impairment
during the follow-up period. The percentage of patients
with good outcomes (that is, those with GOS scores of 4
or 5) was 78% at the time of the last outpatient visit. The
duration of follow-up was 2–8 years.
Discussion
Epidemiological Analysis
Subdural empyema due to meningitis is a rare, life-
threatening complication of meningitis in infants. Infants
are highly predisposed to the development of SDE in
association with meningitis.7,16 Subdural collections are
a common complication of meningitis in young infants.
They occur in 39–60% of infants with proven pyogenic
meningitis.17,18 However, SDE due to purulent meningi-
tis is extremely rare and occurs as a complication in ap-
proximately 1–2% of all cases of infantile bacterial men-
ingitis.9,19 Many cases of SDE in infants occur when a
subdural effusion related to meningitis becomes infected.
The male predominance in the present series is in concor-
dance with that in previous pediatric series.1,4,20
Symptoms/Signs
Subdural empyema during infancy occurs in infants
with bacterial meningitis. It subsequently manifests with
persistent fever, declining neurological status, and/or sei-
J Neurosurg: Pediatrics / Volume 6 / July 2010
Surgical outcome of postmeningitic subdural empyema in infants
Fig. 2.  This 4-week-old boy suffered low-grade fever and irritable crying with right limb myoclonic jerklike movements. The
CSF study and culture data revealed Escherichia coli meningitis. Persistent fever and uncontrolled seizure attacks developed in
the patient despite antibiotic treatment with meropenem and vancomycin for 13 days.  A and B: Brain CT scanning without (A)
and with (B) contrast demonstrated extraaxial subdural fluid collection with marginal well enhancement about 10 mm in thick-
ness, along left frontotemporal convexity and left sylvian fissure. A left temporal craniotomy and removal of SDE was performed
immediately.  C: Postoperative contrast CT showed no residual SDE.
zures. Both infections have similar clinical features such
a fever, seizures, and focal neurological deficits, which
make the distinction difficult.14 In the present study, the
clinical observation revealed a high fever that was present
in most patients because of the infection. Not all patients
developed a seizure or a focal neurological deficit. Hence,
a combination of imaging modalities and clinical evalua-
tion is necessary for the early diagnosis of SDE.
Diagnosis of SDE
Cases in which infants have an unsatisfactory response
to appropriate antibiotic therapy during acute bacterial
meningitis should raise the suspicion of SDE. Reactive
subdural effusions due to meningitis are much more com-
mon than SDE in infants.9,19 Cranial ultrasonography is a
reliable imaging modality in infants that may distinguish
between reactive subdural effusion and SDE.4 Cranial
ultrasonography seems to be cost-effective and has been
recommended as a screening tool for every infant with
bacterial meningitis.3 In our study, all patients were first
suspected to have a subdural effusion based on the brain
sonography findings. Although CT reconfirmation is usu-
ally necessary, brain ultrasonography cannot only be per-
formed easily at bedside at the onset, but it is also used to
assess the amount of empyema that remains after treatment.
Early detection with frequent ultrasonographic monitoring
of the subdural effusions might lessen the neurological se-
quelae in infantile and neonatal purulent meningitis. The
time interval observed in our series between the onset of
the first symptoms of meningitis (fever, drowsiness, and
seizure) and the diagnosis of SDE ranged from 2 days to 4
weeks. Radiographic imaging should be performed in all
patients in whom SDE is suspected. Although MR imag-
ing is more sensitive in showing parenchymal abnormali-
ties, cranial CT scanning is often performed as the first
neuroimaging modality because of the more widespread
availability and the need for a rapid diagnosis.
Treatment of SDE
Despite modern diagnostic and therapeutic capabili-
ties, SDE still carries a significant risk of mortality. The
treatment goal is removal of the pus and eradication of
J Neurosurg: Pediatrics / Volume 6 / July 2010
the source of infection. Few reports have suggested that
some episodes of SDE can be managed conservatively
with antibiotics alone.6,10,11 The choice of antibiotics for
the treatment of SDE should be tailored according to the
suspected route of infection. The subsequent selection of
antibiotics should be based on the available culture and
sensitivity results.8 The duration of antibiotic treatment
often depends on the rate of resolution of the SDE (based
on radiological images).
The surgical approach to SDE is a debatable issue.
Bur hole drainage and craniotomy have been used for
the drainage of SDE.6,13 In literature reviews, intermittent
transfontanel needle aspiration of the SDE may provide a
means of removal of pus,5,13 but it must be performed with
great caution. Some pediatric series emphasized the im-
portance of craniotomy in achieving a better outcome.1,20
The majority of the reports were across all age groups and
various pathogeneses. In our present study, we focused on
the postmeningitic SDE treatment. The mechanism of the
development of postmeningitic SDE is through infection of
the sterile, reactive subdural effusions.14 The infection then
presumably extends through the arachnoid membrane and
into the subdural space.14 However, the pathophysiology of
SDE due to sinus infections, head trauma, or after neuro-
surgery is different. In these cases, the pus spreads into the
subdural space from the sinus, an otorhinogenic source, or
the epidural space by either direct spread associated with
bone erosion or by indirect spread via retrograde throm-
TABLE 4: Clinical outcomes
Group
Bur Hole Craniotomy p
Parameter (18 patients) (15 patients) Value
mean no. of days postop until 7.4 ± 0.79 7.6 ± 1.2 0.91
  afebrile
mean duration of postop antibiot- 20.6 ± 2.23 23.2 ± 2.08 0.40
  ics (days)
GOS Score 4 or 5 14 12 0.75
secondary intervention for residu- 3 2 0.79
  al empyema
41

bophlebitis, which involves the interconnecting venous
systems of the extra- and intracranial spaces. Management
of the primary source of the infection plays an important
part in the treatment of SDE. The reactive subdural effu-
sion is the primary source of infection in postmeningitic
SDE. Bur holes with continuous drainage through a sub-
dural tube may alleviate the source of infection until the
meningitis is well controlled. On the other hand, in cases
of SDE due to other infectious sources, bur hole drainage is
insufficient to obliterate the infectious sources.
Subdural empyema due to meningitis was found in the
cerebral convexity in most of our patients, except for 1. The
use of bur holes with irrigation may be more efficacious for
drainage at the early stage of postmeningitic SDE when the
pus is liquid and can be easily accessed.
The average duration of symptoms in SDE is 2 weeks
and can range from 1 to 8 weeks.8 In our series, SDE was
treated about 12 days after the symptoms of meningitis de-
veloped. Recently, Yilmaz et al.20 reported a retrospective
study of 28 patients identified with SDE associated with
meningitis. In that study, the average time interval be-
tween symptoms and diagnosis of SDE was approximately
2 months. Bur hole drainage seemed insufficient in that
study because the duration of the symptoms was longer
than that reported in other studies on SDE.8 In our pres-
ent study, the outcome data showed no difference in recur-
rence rate, number of days until afebrile, number of days of
postoperative antibiotic treatment, and GOS score between
the bur hole and craniotomy groups. With early detection,
bur hole drainage may be appropriate in the management
of postmeningitic SDE.
Patient Outcomes
In our series, 1 patient (3%) died of disseminated sep-
sis. The patient was resistant to antibiotic treatment and
died of multiple organ failure. The outcome of postmen-
ingitic SDE is difficult to evaluate because of the similar
symptoms of sequelae of meningitis and those of brain
compression by mass effect. The underlying mechanism
of brain injury is SDE, which is related to thrombophle-
bitis, venous infarction, and toxin-reduced local effects,
where a mass effect may be of lesser importance.15 A fa-
vorable outcome was achieved in 73% of patients in our
present study, which is in accordance with the majority of
the previous studies on pediatric SDE.1
Conclusions
Postmeningitic SDE in infants is unique with respect
to the pathophysiology, presentation, and treatment of
SDE. The clinical presentation of meningitis often masks
the presence of SDE in infants. Early detection as well
as evacuation of SDE provides a favorable outcome. Bur
hole drainage under ultrasonographic, CT, or MR imag-
ing guidance is less invasive, and it is possible to remove
subdural pus to the same extent as by craniotomy. Bur
hole drainage should be considered as an effective and
safe method for the surgical management of postmenin-
gitic SDE in infants when detected early.
Disclosure
The authors report no conflict of interest concerning the mate-
42
Z. H. Liu et al.
rials or methods used in this study or the findings specified in this
paper.
Author contributions to the study and manuscript preparation
include the following. Conception and design: Wu, Liu. Acquisition
of data: Wu, Chen. Analysis and interpretation of data: Liu. Drafting
the article: Liu. Critically revising the article: Tu. Reviewed final
version of the manuscript and approved it for submission: all au­­
thors. Statistical analysis: Liu, Chen. Study supervision: Lee.
References
  1.  Banerjee AD, Pandey P, Devi BI, Sampath S, Chandramouli
BA: Pediatric supratentorial subdural empyemas: a retrospec-
tive analysis of 65 cases. Pediatr Neurosurg 45:11–18, 2009
  2.  Bok AP, Peter JC: Subdural empyema: burr holes or cranioto-
my? A retrospective computerized tomography-era analysis
of treatment in 90 cases. J Neurosurg 78:574–578, 1993
  3.  Chang YC, Huang CC, Wang ST, Chio CC: Risk factor of
complications requiring neurosurgical intervention in infants
with bacterial meningitis. Pediatr Neurol 17:144–149, 1997
  4.  Chen CY, Huang CC, Chang YC, Chow NH, Chio CC, Zim-
merman RA: Subdural empyema in 10 infants: US character-
istics and clinical correlates. Radiology 207:609–617, 1998
  5.  Curless RG: Subdural empyema in infant meningitis: diagnosis,
therapy, and prognosis. Childs Nerv Syst 1:211–214, 1985
  6.  Dill SR, Cobbs CG, McDonald CK: Subdural empyema: analy-
sis of 32 cases and review. Clin Infect Dis 20:372–386, 1995
  7.  Feuerman T, Wackym PA, Gade GF, Dubrow T: Craniotomy
improves outcome in subdural empyema. Surg Neurol 32:
105–110, 1989
  8.  Hall WA, Truwit CL: The surgical management of infections
involving the cerebrum. Neurosurgery 62 (Suppl 2):519–531,
2008
  9.  Han BK, Babcock DS, McAdams L: Bacterial meningitis in
infants: sonographic findings. Radiology 154:645–650, 1985
10.  Heran NS, Steinbok P, Cochrane DD: Conservative neurosur-
gical management of intracranial epidural abscesses in chil-
dren. Neurosurgery 53:893–898, 2003
11.  Leys D, Destee A, Petit H, Warot P: Management of subdural
intracranial empyemas should not always require surgery. J
Neurol Neurosurg Psychiatry 49:635–639, 1986
12.  Mauser HW, Tulleken CA: Subdural empyema. A review of 48
patients. Clin Neurol Neurosurg 86:255–263, 1984
13.  Nathoo NNS, Nadvi SS, van Dellen JR, Gouws E: Intracranial
subdural empyemas in the era of computed tomography: a re-
view of 699 cases. Neurosurgery 44:529–536, 1999
14.  Pattisapu JV, Parent AD: Subdural empyemas in children. Pe-
diatr Neurosci 13:251–254, 1987
15.  Rich PM, Deasy NP, Jarosz JM: Intracranial dural empyema.
Br J Radiol 73:1329–1336, 2000
16.  Smith HP, Hendrick EB: Subdural empyema and epidural ab-
scess in children. J Neurosurg 58:392–397, 1983
17.  Snedeker JD, Kaplan SL, Dodge PR, Holmes SJ, Feigin RD:
Subdural effusion and its relationship with neurologic seque-
lae of bacterial meningitis in infancy: a prospective study. Pe-
diatrics 86:163–170, 1990
18.  Syrogiannopoulos GA, Nelson JD, McCracken GH Jr: Sub-
dural collections of fluid in acute bacterial meningitis: a re-
view of 136 cases. Pediatr Infect Dis 5:343–352, 1986
19.  Tunkel AR, Hartman BJ, Kaplan SL, Kaufman BA, Roos KL,
Scheld WM, et al: Practice guidelines for the management of
bacterial meningitis. Clin Infect Dis 39:1267–1284, 2004
20.  Yilmaz N, Kiymaz N, Yilmaz C, Bay A, Yuca SA, Mumcu
C, et al: Surgical treatment outcome of subdural empyema: a
clinical study. Pediatr Neurosurg 42:293–298, 2006
21.  Zimmerman RD, Leeds NE, Danziger A: Subdural empyema:
CT findings. Radiology 150:417–422, 1984
Manuscript submitted September 26, 2009.
Accepted April 7, 2010.
Address correspondence to: Chieh-Tsai Wu, M.D., Department of
Neurosurgery, Chang Gung Memorial Hospital, Chang Gung Med­
ical College and University, 5 Fu-Shin Street, Kwei-Shan Hsiang,
Taoyuan Hsien, 333, Taiwan. email: liuch@adm.cgmh.org.tw.
J Neurosurg: Pediatrics / Volume 6 / July 2010