Original Paper
Pediatr Neurosurg 2009;45:345–349
DOI: 10.1159/000257522

Received: February 3, 2009
Accepted after revision: June 29, 2009
Published online: November 11, 2009

Is Postoperative CT Scanning Predictive of
Subdural Electrode Placement Complications
in Pediatric Epileptic Patients?
Carlo Giussani a, c Tanya Filardi a Krishnapundha Bunyaratavej a, d
Jeffrey C. Mai a Masahiro Ogino a, e Stephanie Greene a, b Samuel R. Browd a
Anthony M. Avellino a Richard G. Ellenbogen a Jeffrey G. Ojemann a
a

Department of Neurological Surgery, Seattle Children’s Hospital, University of Washington School of Medicine,
Seattle, Wash., and b Neurosurgery, University of Pittsburgh, Pittsburgh, Pa., USA; c Neurochirurgia, Dipartimento
di Neuroscienze e Tecnologie Biomediche, Università degli Studi Milano-Bicocca, Ospedale San Gerardo,
Monza, Italia; d Division of Neurosurgery, Department of Surgery, Chulalongkorn University, Bangkok , Thailand;
e
Department of Neurosurgery, Dokkyo University School of Medicine, Tochigi, Japan

Key Words
Intractable seizures ⴢ Subdural electrodes ⴢ Hemorrhagic
complications ⴢ Midline shift

Abstract
Aims: To understand the reliability of postoperative CT scans
to predict the development of intracranial hemorrhagic
complications associated with subdural electrode implants
for monitoring intractable seizure, we reviewed the data of
a consecutive series of children treated at our institution.
Methods: Forty children (mean age: 11.4 years) with subdural electrode implants were reviewed. The immediate postoperative CT scans were evaluated for the presence of hemorrhagic complications and/or brain swelling resulting in a
midline shift. Results: Twenty-six patients (65%) presented a
postoperative midline shift (range = 2–10 mm; mean shift =
4.0 mm). Two children had a midline shift of 15 mm. Two patients with a shift of !5 mm at the first CT scan required a
repeat craniotomy. These patients experienced worsening
neurologic symptoms in a delayed fashion on postoperative

© 2009 S. Karger AG, Basel
1016–2291/09/0455–0345$26.00/0
Fax +41 61 306 12 34
E-Mail karger@karger.ch
www.karger.com

Accessible online at:
www.karger.com/pne

days 1 and 4, respectively. This was correlated to an increase
in midline shift of 15 mm. Conclusions: Subdural electrode
implants in children are safe. The presence of a midline shift
of !5 mm is common postoperatively. The presence and extension of the midline shift at the first CT scan does not seem
to be predictive of the development of symptomatic complications with a mass effect. Complications happened in a delayed fashion.
Copyright © 2009 S. Karger AG, Basel

Introduction

The efficacy of epilepsy surgery depends on the capability to demarcate the epileptic zone that is defined as
‘the area necessary and sufficient for initiating seizures

The authors dedicate this report to the memory of Dr. Ted Roberts
who directly cared for many of the patients in this series. His vision
and dedication remain an inspiration to all of us.

Jeffrey G. Ojemann
Department of Neurological Surgery
4800 Sand Point Way NE, Mailstop W-7729
Seattle, WA 98105 (USA)
Tel. +1 206 987 4525, Fax +1 206 987 3925, E-Mail jeff.ojemann@seattlechildrens.org

and whose removal or disconnection is necessary for abolition of seizures’ [1]. Although scalp EEG, along with
other neuroimaging techniques, helps to effectively identify the general area of a seizure focus [2, 3], invasive techniques have proven to be a far superior method for accurate localization [4]. Unfortunately, several reports
described complications related to the placement of subdural electrodes both in the adult and in the pediatric
population [5–14]. Frequently, reported complications
include infection of the wound and bone flap (2–15%),
CSF leaks through the skin insertion sites of the electrode
cables (30%) and formation of acute symptomatic epidural (1.8–2.5%) and, particularly, subdural hematomas (2–
14%) along with symptomatic brain swelling (8%) [5–14].
While the implantation of subdural electrodes is generally a safe procedure in children with a low rate of clinically relevant hemorrhagic complications [10–13], the development of acute hemorrhagic complications and brain
swelling can be life-threatening conditions that present
mostly in the immediate postoperative period. The performance of postoperative CT scanning after craniotomies for subdural grid implants has become routine. The
majority of patients with implanted subdural electrodes
develop postoperative subacute extra-axial collections
(EAC) with a common midline shift. EAC are composed
of serosanguinous fluid and CSF. It has been hypothesized that the electrodes may slide over the cerebral cortex, especially during seizures, and damage bridging
veins [8]. However, EAC are regularly seen on CT scans
performed in the early postoperative time period before
any seizures develop. Another possible explanation is
that the implanted electrodes do not completely conform
to the shape of the brain surface, creating spaces between
the electrodes and the dura, and between the dura and
the cranium, in which CSF or blood products can collect
[15]. Even if the usual nature of EAC is benign, the formation of delayed intracranial hemorrhagic complications
and/or brain swelling has been reported [8, 9]. Mocco et
al. [15] demonstrated in a patient population composed
mainly of adults that there is no significant radiographic
difference between symptomatic and asymptomatic EAC.
Therefore, the authors concluded that ‘the presence of an
EAC is not a specific indication … for decompression’.
To understand the correlation between EAC, midline
shift and clinically evident complications after the implantation of subdural electrodes, along with the safety of
this invasive diagnostic technique in pediatric patients,
we reviewed the radiological and clinical postoperative
data of a consecutive group of children with intractable
epilepsy treated at our institution.
346

Pediatr Neurosurg 2009;45:345–349

Material and Methods
Patient Population
From October 1997 to May 2004, 40 consecutive patients with
intractable epilepsy underwent a craniotomy for the placement of
subdural electrodes (subdural grids and subdural strips). All the
patients underwent the placement of a combination of subdural
grids and strips. All the patients who had been submitted to the
placement of subdural strips by the use of simple burr holes without craniotomy were excluded from the analysis.
The patient population age ranged from 2 to 17 years (mean
age at the time of surgery was 11.4 years). In our series there were
21 males and 19 females.
Surgical Technique
The aim of our subdural electrode implant is to cover the cerebral cortex with grids and strips as completely as possible, including the lateral, basal and medial surface of the brain. To
achieve this goal, we usually perform a large frontotemporoparietal craniotomy, recently with an osteoplastic bone flap, leaving
the bone flap attached to the temporal muscle. Our aim is to decrease the risk of infection and resorption of the bone flap by
maintaining intact the vascularization coming from the deep
temporalis fascia. In fact, as previously reported by our team, the
risk of cranial bone resorption after a decompressive craniotomy
in children is very high, reaching 50% of cases [16].
The dura is opened, and several grids and strips are positioned
to cover the cortical surface. In particular, subtemporal, orbitofrontal and interhemispheric strips are always placed along with
a 64-electrode convexity grid. We secure the convexity grid to the
dura with a suture to reduce the risk of the grid sliding over the
surface of the brain in case of seizures. After adequate hemostasis
has been obtained, we perform a watertight closure of the dura.
The bone flap is secured to the edges of the craniotomy with a
single-point fixation by the use of 1 plate with screws. In this way
the pressure of the bone flap on the brain covered by the grids is
reduced, at the same time avoiding that the bone flap floats over
the dura. In the experience of our group, it is important to resecure the bone flap to avoid the patients injuring themselves in case
of seizures. Nonetheless, we try to reduce the mass effect produced by the grids using a single-point fixation with a single
straight plate in order to allow the bone flap to be slightly displaced in case of EAC with mass effect. A musculocutaneous closure is performed in multiple layers to reduce the risk of CSF leaking through the skin and around the exit sites of the electrode
cables. In fact, CSF leaks around electrode exit sites have been
proposed as one of the possible mechanisms of subdural hematoma development [8].
The leads are usually continuously connected to an EEG monitoring device in the intensive care unit immediately after a full
recovery from anesthesia to the baseline neurologic condition.
The patients spend the first 12–24 h in the intensive care unit.
Postoperative CT Scan Review
The postoperative CT scans of the patients were reviewed to
evaluate the development of postoperative complications. At our
institution, all the patients undergoing craniotomies are submitted
to a postoperative CT scan within 1 h of the completion of surgery
to check for the presence of acute epidural, subdural or intraparenchymal hemorrhagic complications. In the case of craniotomies for

Giussani et al.

Color version available online

subdural electrode placement, we look for complications of iatrogenic origin, for instance the damage to a midline or base-of-skull
bridging vein during the positioning of the electrodes.
Attention was paid during the review to the development of
postoperative hemorrhagic complications and/or brain swelling.
In particular, all the immediate postoperative CT scans were reviewed, evaluating the presence of EAC, epidural and subdural
hematomas, or brain swelling causing a midline shift. A midline
shift of 15 mm was considered significant [17–19], but managed
conservatively if asymptomatic.

Results

Forty consecutive patients were reviewed. A midline
shift was recognizable in 26 patients. In these patients,
the midline shift ranged from 2 to 10 mm with a mean
midline shift of 4.0 mm. Only 2 patients showed a midline shift of 15 mm (6 and 10 mm). They did not require
a revision surgery due to the absence of related symptoms
(fig. 1). If the 14 patients with no shift are considered as
well, the median shift was 2 mm.
Of the 24 patients with a recognizable midline shift
of !5 mm, 2 patients required repeat craniotomy due to
the development of increased subdural fluid collections/
hemorrhage, with a progressive shift to 15 mm and symptoms of increased intracranial pressure. In both cases, the
patients developed worsening neurologic symptoms in a
delayed fashion (after 1 and 4 days, respectively). Both
patients underwent evacuation of the collection: 1 patient with a reimplantation of the electrodes and 1 with a
definitive epilepsy surgery. They both recovered completely.
The 2 symptomatic cases that required a postimplant
revision surgery will be discussed.
Case 1
The first case was a 13-year-old boy with a history of
intractable partial seizures since he was 18 months old.
The first CT scan after the right subdural electrode implant showed the presence of right EAC with a maximal
thickness of 8 mm and a midline shift of 4 mm (fig. 2a).
However, on postoperative day 4, the patient developed a
severe headache with worsening lethargy. A new CT scan
showed an acute subdural hematoma with a maximal
thickness of 15 mm and a midline shift of 7 mm (fig. 2b).
Due to the worsening of the headache, the patient underwent a new operation through the same skin incision and
bone flap. Under the dura that was reopened using the
same incision, an organized subdural hematoma was
evacuated. The brain was soft and pulsating after the heSubdural Electrode Placement
Complications and CT

Fig. 1. Initial postoperative CT scan of a patient with left subdural
electrodes, demonstrating a midline shift of 15 mm. This patient
was neurologically intact and did not require a reoperation.

a

b

Fig. 2. a Initial postoperative CT scan of a 13-year-old male with
intractable partial seizures who underwent right subdural electrode implantation. It demonstrates the presence of right EAC
with a maximal thickness of 8 mm and a midline shift of 4 mm.
b Repeat postoperative CT scan performed on day 4 after surgery
due to the development of a worsening headache and lethargy. It
shows an acute subdural hematoma with a maximal thickness of
15 mm and a midline shift of 7 mm that required a reoperation.

matoma evacuation. The patient did not receive a new
electrode implant even though the monitoring had not
been a contributory factor. He directly underwent definitive epilepsy surgery based on MRI anatomical data and
on previous EEG data.
Pediatr Neurosurg 2009;45:345–349

347

Case 2
The second case was a 17-year-old boy with a history
of intractable complex partial seizures with secondary
generalization who had previously undergone a partial
right frontal lobectomy for intractable seizures. His seizures recurred, and grids, strips and depth electrodes
were implanted. He developed a worsening headache on
postoperative day 1. The first CT scan after surgery
showed the presence of right frontal EAC with a maximal
thickness of 6 mm near the midline border of the craniotomy and a midline shift of 4 mm with no related symptoms. A new CT scan performed the following day showed
the presence of a minimal evolution of the EAC that measured a maximum of 8 mm with a midline shift of 6 mm.
The second CT scan did not show a new acute subdural
hematoma, but there was an increase in serosanguinous
fluid over the grids. Due to the worsening of the headache
and the development of lethargy, the patient underwent
repeat craniotomy through the same skin incision and
bone flap. Under the dura that was reopened using the
same incision, a fluid serosanguinous collection was rapidly evacuated. The brain was soft and pulsating after the
evacuation of the serosanguinous collection. The electrodes were replaced due to the lack of conclusive electrocorticographic data. He underwent definitive epilepsy
surgery after 1 week of monitoring.
In both cases requiring reoperations, it was not possible to find a clear source of the bleeding during the revision surgery.

Discussion

The need to submit a pediatric patient with intractable
epilepsy to an invasive monitoring procedure by the use
of subdural electrodes requires a careful balance of the
risks and benefits [5–14].
The review of the pediatric patients with intractable
epilepsy who underwent placement of subdural electrodes at our hospital for invasive monitoring demonstrated that it is common to see a midline shift of !5 mm
on an immediate postoperative CT scan. The majority of
our pediatric patients (26) presented at the first postoperative CT scan a midline shift of !5 mm with a mean of
4 mm. Nevertheless, only 2 patients out of 40 required a
revision craniotomy due to the development of neurologic symptoms in the postoperative period.
Our findings are in agreement with other studies conducted on pediatric patients demonstrating that the
placement of subdural grids and strips performed via cra348

Pediatr Neurosurg 2009;45:345–349

niotomy is generally a safe procedure in children with
intractable epilepsy who require invasive seizure monitoring [10–13]. One explanation advocated for the relative
safety of subdural electrode implants in pediatric patients
is the greater plasticity of pediatric brain parenchyma
and vascular structures compared to adults [13].
The 2 patients who required a revision surgery both
presented with a midline shift of !5 mm at the first postoperative CT scan. Neither patient in the present series
with an initial shift of 15 mm (6 and 10 mm, respectively) developed related postoperative symptoms. These 2
opposite findings suggest that the presence of a postoperative midline shift as well as its extent are not predictive
of the development of symptomatic hemorrhagic complications with a mass effect. In agreement with other authors [15], we think that signs of EAC with a mass effect
are commonly seen on the first postoperative CT scan
and, by themselves, do not represent a concern.
The 2 cases that were complicated by the development
of voluminous EAC and subdural hematoma developed
symptoms related to increased intracranial pressure in a
delayed fashion 1 day and 4 days after the subdural electrode implants, respectively. As previously reported, this
timing pattern could be related to the slow and continuous CSF leakage around the electrode exit sites. The leaking could lead to a unilateral alteration of the CSF dynamic and the production of an ex vacuo shift toward the
side of the subdural electrode with subsequent brain shift
and tearing of bridging veins [8, 14]. The leakage of CSF
through the cable exit sites is a well-known event related
to subdural electrode implants [8, 14], with an occurrence
rate of 30% in pediatric patients [13]. Nevertheless, other
factors likely play a role in symptomatic extra-axial hemorrhagic complications, given the low rate of reoperations [10–13].

Conclusion

The present series reviews pediatric patients who underwent placement of subdural electrodes for seizure
monitoring. Subdural electrode implants performed in
children by craniotomy are a safe procedure. The operation is characterized by a low percentage of symptomatic
hemorrhagic complications that usually present in a delayed fashion. It is a common finding at the initial postoperative CT scan to see a midline shift of !5 mm with
EAC; importantly, the presence and the extent of the
midline shift at the initial CT scan are not predictive of
the development of delayed complications.
Giussani et al.

References
1 Luders HO, Engel J Jr, Munari C: General
principles; in Engel J Jr (ed): Surgical Treatment of the Epilepsies, ed 2. New York, Raven
Press, 1993, pp 137–153.
2 Won HJ, Chang KH, Cheon JE, Kim HD, Lee
DS, Han MH, Kim IO, Lee SK, Chung CK:
Comparison of MR imaging with PET and
ictal SPECT in 118 patients with intractable
epilepsy. AJNR Am J Neuroradiol 1999; 20:
593–599.
3 Lee DS, Lee SK, Lee MC: Functional neuroimaging in epilepsy: FDG PET and ictal
SPECT. J Korean Med Sci 2001;16:689–696.
4 Jayakar P, Duchowny M, Resnick TJ: Subdural monitoring in the evaluation of children
for epilepsy surgery. J Child Neurol 1994;
9(suppl 2):61–66.
5 Wiggins GC, Elisevich K, Smith BJ: Morbidity and infection in combined subdural grid
and strip electrode investigation for intractable epilepsy. Epilepsy Res 1999; 37:73–80.
6 Zaccariotti VA, Pannek HW, Holthausen H,
Oppel F: Evaluation with subdural plates in
children and adolescents. Neurol Res 1999;
21:463–474.
7 Bruce DA, Bizzi JW: Surgical technique for
the insertion of grids and strips for invasive
monitoring in children with intractable epilepsy. Childs Nerv Syst 2000;16:724–730.

Subdural Electrode Placement
Complications and CT

8 Lee WS, Lee JK, Lee SA, Kang JK, Ko TS:
Complications and results of subdural grid
electrode implantation in epilepsy surgery.
Surg Neurol 2000;54:346–351.
9 Hamer HM, Morris HH, Mascha EJ, Karafa
MT, Bingaman WE, Bej MD, Burgess RC,
Dinner DS, Foldvary NR, Hahn JF, Kotagal
P, Najm I, Wyllie E, Lüders HO: Complications of invasive video-EEG monitoring with
subdural grid electrodes. Neurology 2002;
58:97–103.
10 Onal C, Otsubo H, Araki T, Chitoku S, Ochi
A, Weiss S, Elliott I, Snead OC 3rd, Rutka JT,
Logan W: Complications of invasive subdural grid monitoring in children with epilepsy. J Neurosurg 2003;98:1017–1026.
11 Simon SL, Telfeian A, Duhaime AC: Complications of invasive monitoring used in intractable pediatric epilepsy. Pediatr Neurosurg 2003;38:47–52.
12 Johnston JM Jr, Mangano FT, Ojemann JG,
Park TS, Trevathan E, Smyth MDJ: Complications of invasive subdural electrode monitoring at St. Louis Children’s Hospital, 1994–
2005. Neurosurgery 2006;105:343–347.
13 Musleh W, Yassari R, Hecox K, Kohrman M,
Chico M, Frim D: Low incidence of subdural
grid-related complications in prolonged pediatric EEG monitoring. Pediatr Neurosurg
2006;42:284–287.
14 Fountas KN, Smith JR: Subdural electrodeassociated complications: a 20-year experience. Stereotact Funct Neurosurg 2007; 85:
264–272.

15 Mocco J, Komotar RJ, Ladouceur AK, Zacharia BE, Goodman RR, McKhann GM 2nd:
Radiographic characteristics fail to predict
clinical course after subdural electrode
placement. Neurosurgery 2006;58:120–125.
16 Grant GA, Jolley M, Ellenbogen RG, Roberts
TS, Gruss JR, Loeser JD: Failure of autologous bone-assisted cranioplasty following
decompressive craniectomy in children and
adolescents. J Neurosurg 2004; 100:163–168.
17 Marshall LF, Marshall SB, Klauber MR,
Clark MvB, Eisenberg HM, Jane JA, Luerssen TG, Marmarou A, Foulkes MA: A new
classification of head injury based on computerized tomography. J Neurosurg 1991;
75(suppl):S14–S20.
18 Marshall LF, Marshall SB, Klauber MR, van
Berkum Clark M, Eisenberg H, Jane JA, Luerssen TG, Marmarou A, Foulkes MA: The
diagnosis of head injury requires a classification based on computed axial tomography. J
Neurotrauma 1992;9:287–292.
19 Yuh EL, Gean AD, Manley GT, Callen AL,
Wintermark M: Computer-aided assessment
of head computed tomography (CT) studies
in patients with suspected traumatic brain
injury. J Neurotrauma 2008;25:1163–1172.

Pediatr Neurosurg 2009;45:345–349

349

Copyright: S. Karger AG, Basel 2009. Reproduced with the permission of S. Karger AG, Basel. Further
reproduction or distribution (electronic or otherwise) is prohibited without permission from the copyright
holder.