Seizure 45 (2017) 28–35

Contents lists available at ScienceDirect

Seizure
journal homepage: www.elsevier.com/locate/yseiz

Review

A systematic review of epileptic seizures in adults with subdural

haematomas
Sae-Yeon Wona,* , Juergen Konczallaa , Daniel Dubinskia , Adriano Cattania , Colleen Cucac,
Volker Seiferta , Felix Rosenowb , Adam Strzelczykb , Thomas M. Freimana
a
Department of Neurosurgery, Goethe University, Frankfurt am Main, Germany
b
Department of Neurology and Epilepsy Center, Frankfurt Rhine-Main Goethe University, Frankfurt am Main, Germany
c
Department of Anaesthesiology, Intensive Care Medicine and Pain Therapy, University Hospital Frankfurt am Main, Frankfurt am Main, Germany

A R T I C L E I N F O A B S T R A C T

Article history: Background: Posttraumatic epileptic seizures (PTS) are a serious complication in patients with subdural
Received 23 September 2016 haematoma (SDH). However, to date, several studies have shown discordances about SDH-associated
Received in revised form 10 November 2016 seizures in terms of incidence, risk factors and prophylactic antiepileptic treatment.
Accepted 20 November 2016
Objective: The aim of this study was to analyse the incidence, risk factors of PTS and the role of
Available online xxx
prophylactic antiepileptic treatment in patients with SDH.
Data sources: A systematic literature review examining PTS in patients with SDH was performed using
Keywords:
PubMed gateway, Cochrane Central Register of Controlled Trials, and Excerpta Medica dataBASE between
Epileptic seizure
Subdural haematoma
September 1961 and February 2016. Search terms included subdural haematoma, seizure, epilepsy,
Risk factors prophylactic antiepileptic drugs, anticonvulsive medication, and risk factors.
Prophylactic antiepileptic treatment Data selection: Human-based clinical studies focusing on epileptic seizures in patients with SDH.
Incidence of seizure Data extraction and synthesis: PRISMA statements were used for assessing data quality. Two independent
reviewers extracted data from included studies and disagreement was solved by consensus. Twenty-four
studies were identified for inclusion into the study.
Results: Overall incidence of early PTS (ePTS) and late PTS (lPTS)/2 years was 28% and 43% in acute SDH
(aSDH) whereas the incidence of e- and lPTS was lower in chronic SDH (cSDH; 5.3% vs. 10%). Overall risk
factors for PTS in patients with aSDH were: 24 h postoperative Glasgow Coma Score (GCS) score below 9
(OR 10.5), craniotomy (OR 3.9), preoperative GCS below 8 (OR 3.1). In patients with cSDH the risk factors
were alcohol abuse (OR 14.3), change of mental status (OR 7.2), previous stroke (OR 5.3) and density of
haematoma in computer tomography (OR 3.8). Age, sex, haematoma size/side and midline shifts were not
significant risk factors for PTS in both types of SDH. In prevention of PTS phenytoin and levetiracetam
showed similar efficacy (OR 1.3), whereas levetiracetam was associated with significantly lower adverse
effects (OR 0.1).
Limitations: Most of the studies were of retrospective nature with a small sample size. Due to the
inclusion criteria, some studies had to be excluded and that might lead to selection bias.
Conclusions: PTS are a serious complication in patients with SDH, particularly in aSDH. The “prophylactic
use” of antiepileptic drugs might be beneficial in patients with cumulative risk factors.
ã 2016 Published by Elsevier Ltd on behalf of British Epilepsy Association.

1. Introduction is reported to be 24% in acute SDHs (aSDHs) and 11% in chronic

Post-traumatic epileptic seizures (PTSs) are a serious compli-
cation after traumatic brain injury (TBI), particularly in patients
with subdural haematomas (SDHs). The prevalence of PTSs in SDHs
* Corresponding author at: Department of Neurosurgery, Goethe University,
Schleusenweg 2 16, D 60528 Frankfurt am Main, Germany.
Fax: +49 0 696301 6279.
E-mail address: Sae-Yeon.Won@kgu.de (S.-Y. Won).
SDHs (cSDHs) [1–3]. Despite the high prevalence of PTSs in SDHs,
which patients are at risk of developing PTSs and could eventually
profit from a prophylactic antiepileptic medication remains a
matter of speculation. In fact, some retrospective studies showed
that older age, chronic alcoholism and the severity of trauma were
associated with PTSs [4,8–11]; however, classifying PTSs is more
complex than it seems at first sight. In the literature, PTSs are
classified by arbitrary time limits into three groups: immediate
seizures occur within 24 h after insult; early acute symptomatic
seizures within 7 days after insult; and late unprovoked seizures

http://dx.doi.org/10.1016/j.seizure.2016.11.017
1059-1311/ã 2016 Published by Elsevier Ltd on behalf of British Epilepsy Association.
 S.-Y. Won et al. / Seizure 45 (2017) 28–35
7 days after insult. It seems important to differentiate between
those groups due to the different underlying pathophysiology,
which might be associated with different seizure recurrence rates.
Unfortunately, the time limits are not used consistently: some
authors delineate between early and late seizures at 14 days.
Compared to early PTSs (ePTSs), late PTSs (lPTSs) have a much
higher risk of recurrence of seizures (<20% compared to >60%).
Although the risk of PTS is highest within 24 h after TBI as an acute
symptom, lPTSs can occur up to 20 years after TBI, thereby
suggesting clinical alertness. In particular, injuries involving SDHs
are likely to increase both ePTSs and lPTSs in children and adults
compared to other TBIs [3]. The prognosis after an SDH depends on
multiple factors such as age, Glasgow coma scale (GCS), size of
haemorrhage and time from the SDH until treatment, but the
occurrence of a PTS is an independent marker of poor functional
and social outcome [5–8,10,19].
Antiepileptic drugs (AEDs) are standard treatment for lPTSs, but
their use in ePTSs remains controversial. There have been some
prospective studies verifying the positive effect of prophylactic
AEDs like phenytoin, which was largely used in the past as the gold
standard in reducing ePTSs [13]. Recently, more tolerable AEDs like
levetiracetam, which have a similar efficacy, were introduced into
the market to replace the standard AEDs [55]. However, to date,
there have been no prospective controlled studies describing the
value of prophylactic AEDs exclusively focusing on SDHs. One
retrospective study showed a significant reduction of epileptic
seizures by administering prophylactic AEDs [2,15], but other
retrospective studies were not able to show any advantage of
prophylactic AEDs [16,17,25]. Therefore, further studies are needed
to clarify the ongoing debate about the benefit of using
prophylactic AEDs in patients with SDHs.
During the last few years, several studies have reported a
correlation between SDHs and epileptic seizures, but many studies
disagree about SDH-associated seizures. In the present study, we
performed the first systematic review on epileptic seizures in
SDHs. The objective of this study was to identify and summarize
the incidence and risk factors of SDH-associated seizures and the
current state of prophylactic use of AEDs.
2. Methods
A systematic review was performed following the guidelines of
Preferred Reporting Items for Systematic Reviews and Meta-
Analyses (PRISMA) and was approved by the Clinical Ethics
Committee of the University of Frankfurt (Nr.509/15) [32,51].
2.1. Data sources and data searches
A broad search for all studies describing epileptic seizures in
patients with SDHs using the PubMed gateway of the MEDLINE
database, the Cochrane Central Register of Controlled Trials
(CENTRAL) and the Excerpta Medica dataBASE (EMBASE) was
performed from 1961 to February 2016. Additionally, we
researched the reference lists of those studies. The following
keywords were used individually or in combination with “AND”:
subdural haematoma; SDH (acute/chronic); seizure; epilepsy;
prophylactic antiepileptic drugs; anticonvulsive medication; risk
factors of seizure.
2.2. Study selection
Inclusion required human-based clinical studies that focused
on epileptic seizures in patients with SDHs. Exclusion criteria
included studies reporting only paediatric patients (<18 years old),
language other than English, redundant data and deficient
insufficient disaggregation to identify incidence or risk factors of
29
PTSs. Case reports, presentations and conference abstracts were
excluded as well. The study selection process is illustrated in Fig. 1.
2.3. Data extraction and synthesis
Two reviewers performed eligibility assessment independently
in an unblinded standardized manner. Two independent review
authors acquired the literature, selected the studies and extracted
data from the included studies. Disagreement between reviewers
was resolved by consensus [51]. In the case of no agreement, third
third author was authorized to decide.
In total, 982 studies were identified. After screening the titles
and abstracts, 42 studies were identified as addressing subdural
haematomas and epileptic seizures. Among those studies, 18 were
excluded (not in English, redundant, not disaggregated, no
manuscript available) and 24 were included. Among these studies,
17 investigated seizure incidence, 11 explored risk factors for
seizures and 10 studies examined prophylactic antiepileptic
treatment. For all parameters, we distinguished between aSDHs
and cSDHs. The incidence that we describe in this review is a crude
incidence estimate and not from population-based studies.
Furthermore, we tried to analyse immediate, early and late PTSs
separately; however, the majority of the studies considered
immediate PTSs and ePTSs as one category. Due to the lack of
granularity provided in the reviewed literature we were only able
to distinguish between ePTSs and lPTSs.
2.4. Statistical analysis
GraphPad Prism (6.0, GraphPad Software Inc., USA) was used for
statistical analysis. Parametric data were analysed between group
differences using an unpaired t-test. For categorical variables, we
used Fisher’s exact or the Chi-square test. For dichotomy risk
factors, we counted odds ratios (ORs) with 95% confidence
intervals (CI95). A p-value  0.05 was regarded as statistically
significant.
3. Results
In total, 17 studies reporting incidence of seizures in patients
with acute and chronic SDHs were analysed [3,5,10–
12,15,17,21,25,27–30,33–37]. Two of the 17 studies (11.7%) reported
PTSs in SDHs, without differentiating the type of SDH. Nine of the
17 studies (52.9%) distinguished between ePTSs and lPTSs, but
none of them differentiated between immediate PTSs and ePTSs.
Therefore immediate PTSs and ePTS were considered as one
category. Other studies did not differentiate between these entities
and referred to PTS as one entity, which we classified as
unspecified PTS. Eleven of the 17 studies (64.7%) analysed
additionally risk factors for PTSs in SDHs. Apart from two
prospective studies [11,39], all other studies were retrospectively
analysed with an evidence level of 3.
3.1. Incidence of PTS in patients with acute subdural haematoma
The results are illustrated in Table 1. In total, three studies were
identified as describing the incidence of PTS in patients with
aSDHs. About 40 years ago, Jennett [36] reported that 58 out of
159 patients with aSDHs (36%) developed ePTSs, which was the
highest incidence among all intracranial haematomas. Recently,
similar results (24%, 25%) were reported by Temkin and Rabinstein
et al. [3,10]. The mean time to occurrence of ePTS was one day,
indicating the highest incidence within 24 h after onset [5].
Furthermore, two of those studies analysed cumulative lPTS
incidence in aSDHs within a two-year period. In both studies, the
incidence of lPTS was more than 40% [3,36]. One prospective study
30 S.-Y. Won et al. / Seizure 45 (2017) 28–35

Fig. 1. Flow chart summarizing the study selection process. Inclusion criteria required human-based clinical studies focusing on epileptic seizures in patients with subdural
haematomas. Exclusion criteria required studies reporting on paediatric patients only (<18 years old), language other than English, redundant data or deficientinsufficient
disaggregation to identify the incidence or risk factors of epileptic seizure in subdural haematomas. Overall, 24 studies met the inclusion and exclusion criteria.

observing lPTSs in mixed SDHs found that the incidence varied
between 15% and 28%, depending on the treatment [11].
3.2. Incidence of PTS in patients with chronic subdural haematoma
We identified 12 studies focusing on PTSs in patients with
cSDHs. Five of the 12 studies (41.7%) distinguished between ePTSs
and lPTSs whereas other studies distinguished between pre- and
post-operative seizures. The overall incidence of PTS in patients
with cSDHs is in general lower than with aSDHs; it varies between
2% and 42% [15–17,25–30,34,35,37,38], and the incidence of post-
operative seizures is reported as between 1% and 23% [25–30]. The
incidence of ePTS ranged between 3.6% and 42% whereas the
incidence of lPTS ranged between 5% and 20.6% within a three-year
period [25,30,35,37,38]. Presumably, the wide range of PTS
incidence between studies may be due to the different surgical
treatments, the status of patients and the severity of the head
injury. Hirakawa et al. [28] compared 309 patients with cSDHs
after different surgical treatments and found that patients
undergoing burr hole treatment had a decreased incidence of
epileptic seizures compared to patients with craniotomy and
haematoma-membrane resection (capsulotomy) [25,28]. More-
over, patients with a history of alcohol abuse or severe head trauma
had a higher PTS incidence (42%) in combination with cSDH [4,37].
3.3. Risk factors for PTS in patients with acute subdural haematoma
To date, there have been few studies identifying risk factors for
PTS in patients with SDH. Differentiating aSDH from cSDH, the
results and statistical analysis are summarized in Tables 2 and 3.
Significant risk factors for ePTS in patients with aSDH were a
preoperative GCS score < 8, a post-operative GCS score  8 within
24 h and the need for a craniotomy [10]. Similar correlating results
were reported between craniotomy and lPTS within a two-year
period [3,11]. Age, haematoma size, haematoma side and midline
shifts were not significant risk factors for ePTS in patients with
aSDH [10]. Additional risk factors such as age, black ethnicity,
secondary SDH and major organ dysfunction were reported in a
large population-based study, but the authors did not differentiate
between types of SDH and PTS [5].
3.4. Risk factors for PTS in patients with chronic subdural haematoma
For patients with cSDH, the risk factors for PTS were different to
those for aSDH. One prospective study reported alcoholism as a
risk factor for lPTS in cSDH [38]; however, since most of the studies
did not distinguish between the type of PTS, we could only identify
risk factors for unspecified PTS: previous stroke, specific symptoms
such as change of mental status (e.g. disorientation, frontal lobe
 S.-Y. Won et al. / Seizure 45 (2017) 28–35 31

Table 1
Incidence of PTS in patients with subdural haematoma.

Authors Year Age group No. Type of SDH Early PTS (%) Late PTS (%) Unspecified PTS (%)
Jennett et al. 1975 Adult 159 Acute –
Temkin et al. 2003 Adult ? Acute –
Rabinstein et al. 2010 Adult 134 Acute/acute-on-chronic –

Cole et al. 1961 Adult 50 Chronic

Hirawaka et al. 1972 Adult 270 Chronic
Luxon et al. 1979 Adult 194 Chronic
Robinson et al. 1984 Adult 133 Chronic
Kotwica et al. 1991 Adult 131 Chronic
Rubin et al. 1993 Adult 143 Chronic
Ohno et al. 1993 Adult 56 Chronic
Sabo et al. 1995 Adult 98 Chronic
Dudek et al. 1999 Adult 34 Chronic
Chen et al. 2004 Adult 128 Chronic
Grobelny et al. 2009 Adult 88 Chronic
Huang et al. 2011 Adult 100 Chronic

Seifi et al. 2014 Adult 1,583,255 Mixed – – 0.5

Englander et al. 2003 Adult 276 Mixed (mostly acute) 3 13.8/2 y –

SDH: subdural haematoma; PTS: post-traumatic seizure.

syndrome), CT findings (Hounsfield unit (HU) density, brain
atrophy), craniotomy and mean Glasgow outcome scale by
discharge [15,17,28,30,35,37,38]. Furthermore, the incidence of
unspecified PTS in left-sided haematoma was slightly higher but
did not reach significance [15,17,35]. Similarly to aSDH, age,
haematoma size and midline shifts were not significant risk factors
for unspecified PTS in patients with cSDH [17,30].
3.5. Prophylactic antiepileptic treatment in acute subdural
haematomas
To date, there has only been one recently published retrospec-
tive study by Radic et al. [18] evaluating prophylactic antiepileptic
treatment in patients with aSDHs. In this study, levetiracetam and
phenytoin were compared and were shown to have equal efficacy
in reducing PTSs (OR 1.3, 95% CI 0.7–2.4, p = ns), whereas
levetiracetam had a significantly lower adverse event profile (OR
0.1, 95% CI 0.1–0.4, p < 0.0001). Interestingly, in just one subgroup
with midline shift in the CT scan, the levetiracetam group had a
higher risk of electrographic epileptic seizures than the phenytoin
group (OR n.a., p = 0.03). Similar results regarding the benefit of
prophylactic AEDs (phenytoin or levetiracetam) in reducing PTSs
were reported in the past; however, they did not distinguish
between the different TBI entities [12,13,24,40,42].
3.6. Prophylactic antiepileptic treatment in chronic subdural
haematomas
For patients with cSDHs, there are controversial statements in
terms of prophylactic anticonvulsive medication. A retrospective
study focusing on surgically treated patients with cSDHs showed
effective reduction of seizures by using prophylactic phenytoin
medication (OR 0.1, 95% CI 0.0–0.4, p = 0.0002); however, the
benefit in terms of outcome is not precisely described [15]. Another
study identified prophylactic AED treatment as a predictor of lower
seizure incidence in cSDHs (OR 0.2, 95% CI 0.1–0.7, p = 0.013), but
the outcome at discharge did not differ (OR n.a., p = ns). In contrast,
a retrospective study showed no benefit in patients treated
prophylactically with AEDs (OR 1.3, 95% CI 0.4–3,9, p = ns) [16]. Due
to the variable incidence of epileptic seizures in patients with
cSDHs, there is currently no consensual recommendation for the
use of prophylactic AEDs in these patients [16,25].
4. Discussion
In the present study, we report a systematic review of the
incidence and risk factors of PTS as well as prophylactic AED
treatment in patients with SDHs. Overall, SDHs, especially aSDHs,
were associated with high incidence of PTS [1,4]. The incidence of
ePTS in patients with aSDHs was approximately 28% and in
patients with cSDHs 10%. Furthermore, the incidence of lPTS in
patients with aSDHs was 43% within 2 years of follow-up whereas
in patients with cSDHs it was only 5.3% within 3 years of follow-up,
indicating a lower risk of developing epilepsy.
Depending on the type of SDH (acute vs. chronic), there are
different risk factors likely to cause PTSs: in patients with aSDHs, in
particular for ePTSs, risk factors were mainly dependent on the
severity of the TBI as graded by the preoperative GCS score and the
24-h post-operative GCS score, both below 8, as well as the need for
a craniotomy [4,9,10,12,20].
In the case of lPTSs, craniotomy was found to be the most
important risk factor. Compared to ePTSs, the recurrence rate of
seizures was shown to be much higher in patients with lPTSs
[52,53]. An increased risk of lPTS compared to the normal
population persisted for at least 20 years. [4,9]. Additionally, the
recurrence of lPTS was related to the severity of the aSDH. Sixty-
nine per cent of the patients with aSDHs and skull fractures
developed secondary PTS within 6 months and 86% within 2 years
[12].
As a pathophysiological mechanism for the high epileptogenity
of aSDH it is postulated that the haematoma itself irritates the
cortical surface with blood compounds and later with its
 32
Table 2
Risk factors for seizures in acute and mixed SDHs, odds ratio (OR), confidence intervals and p-values.

Authors Year Cohort information Risk factors for ePTS Risk factors for lPTS Risk factors for unspecified PTS OR (95% CI) p-Value*
Rabinstein et al. [10] 2010 134 SDH patients Age N.a. n.s.
-41 aSDH patients Sex 0.865 (0.54–2.114) n.s.
-93 acute-on-chronic SDH patients aSDH 2.593 (1.152–5,833) 0.03
CT scans:
-Haematoma side 1.493 (0.678–3.286) n.s.
-Mean SDH volume N.a. n.s.
-Mean midline shift N.a. n.s.
Craniotomy 3.85 (1.688–8.779) 0.001.
GCS score <8 preop. 3.130 (1.038–9.437) 0.05

S.-Y. Won et al. / Seizure 45 (2017) 28–35

GCS score 8, 24 h post-op. 10.544 (3.035–36.634) 0.0001

Temkin [3] 2003 783 TBI patients Treatment N.a. <0.001

-? aSDH patients -Craniotomy: 44%
-Conservative: 19%
-No SDH: 15%

Englander et al. [11] 2003 647 TBI patients Treatment

-276 mixed SDH -Surgery: 27.8% 2.26 (1.13–4.52) 0.03
Conservative: 15.3%
-No SDH: 9.5% 0.514 (0.307–0.861) 0.01

Seifi et al. [5] 2014 1,583,255 mixed SDH Age (per year) 0.99 (0.99–0.99) <0.0001
Patients Black 2.08 (1.952–2.216) <0.0001
-0.5% Status epilepticus 1 SDH/2 SDH 0.555 (0.53–0.581) <0.0001
Major organ dysfunction
-Haematologic
-Metabolic 1.945 (1.807–2.093) <0.0001
-Renal 4.185 (3.818–4.587) <0.0001
-CNS 2.263 (2.132–2.403) <0.0001
-Respiratory 3.348 (3.247–3.64) <0.0001
6.176 (5.889–6.477) <0.0001

SDH: subdural hematoma; PTS: post-traumatic seizure; ePTS: early PTS; lPTS: late PTS; 95% CI: 95% confidence interval.
*
Unpaired t-test for parametric statistics, Fisher’s exact and Chi-square test for categorical variables; not significant (n.s., p > 0.05).
 S.-Y. Won et al. / Seizure 45 (2017) 28–35 33

Table 3
Risk factors for seizures in chronic SDHs, odds ratio (OR), confidence intervals and p-values.

Authors Year Cohort Risk factors Risk factors Risk factors for OR (95% CI) p-Value*
information for ePTS for lPTS unspecified PTS
Cole et al. [37] 1961 50 cSDH Age N.a. n.s.
patients
History of N.a. n.s.
epilepsy

Hirakawa et al. [28] 1972 309 cSDH patients Burr hole 0.453 (0.195–1.0517) 0.06
Craniotomy 6.248 (2.772–14.081) 0.0001

Sabo et al. [15] 1994 98 cSDH Age, history of trauma, N.a. n.s.
patients Markwalder scale, location of
clot, type of surgery

Dudekt et al. [38] 1999 34 cSDH Alcohol 14.25 (1.469–138.278) 0.01

patients abusus

Chen et al. [35] 2004 128 cSDH CT scans:

patients
-Density
Hypodense 1.178 (0.133–10,471) n.s.
(n = 1/16 seizure)
Isodense 0.198 (0.037–1.063) 0.052
(n = 2/83 seizure)
Mixed density 5.12 (1.076–24.373) <0.05
(n = 4/29 seizure)
-SDH side
Right 0.292 (0.034–2.502) n.s.
Left 3.802 (0.709–20.397) n.s.
Bilateral 0.529 (0.061–4.575) n.s.

Grobelny et al. [17] 2009 88 cSDH Age N.a. n.s.

patients
Sex (male) 0.254 (0.069–0.932) 0.06
Mean GCS at admission N.a. n.s.
CT scans:
-SDH side
Right 0 0.03
Left 2.741 (0.749–10.172) n.s.
Bilateral 1.429 (0.38–5,371) n.s
-Mean SDH volume N.a. n.s.
-Mean midline shift N.a. n.s.

Huang et al. [30] 2011 100 cSDH Age N.a. n.s.

patients
Sex 0.912 (0.179–4,654) n.s.
Symptoms (mental change) 7.224 (1.481–35.241) 0.008
Mean GCS at admission N.a. 0.008
CT scans:
-Density of haematoma (HU) N.a. 0.051
-Severe brain atrophy
-Haematoma mass effect 3.802 (1.032–14.012) 0.05
-Haematoma volume N.a. n.s.
Mean GOS at discharge N.a. n.s.
Stroke N.a. 0.016
5,347 (1,409–20.291) 0.019
*
Unpaired t-test for parametric statistics, Fisher’s exact and Chi-square test for categorical variables; not significant (n.s., p > 0.05). SDH: subdural hematoma; PTS: post-
traumatic seizure; ePTS: early PTS; lPTS: late PTS; 95% CI: 95% confidence interval.

degradation products [20]. In particular, the decomposition of
haemoglobin on the cortical surface is highly epileptogenic [12,22].
Thus, blood evacuation through craniotomy or craniectomy might
reduce the risk of PTS; however, intracranial surgery was found to
be an independent predictor of lPTS [10,11,28]. Englander et al. [11]
speculate that the correlation between surgery and PTS might be
explained by sudden decompression through craniotomy that
might cause an additional intraparenchymal injury through a
sudden negative deceleration. Consequently, this might promote
neuronal damage and lead to the creation of an epileptogenic focus
[5]. On the other hand, the high incidence of PTS after craniotomy
might also be explained by the simple fact that the craniotomy
itself is a surrogate marker of the severity of a brain injury, which
accompanies structural parenchyma damage, which is known to be
epileptogenic.
Unfortunately, a substantial number of studies dealing with
cSDHs did not differentiate between the types of PTS. Therefore we
had to summarize the risk factors for unspecified PTSs in cSDHs. In
contrast to aSDHs, the main risk factors were alcohol abuse,
specific symptoms such as mental disorders, previous strokes and
CT abnormalities (Hounsfield Unit density irregularities, brain
atrophy). Similarly to aSDHs, other risk factors for PTSs in cSDHs
were craniotomy and low GCS at patient discharge. Alcoholism as a
patient-related risk factor was known to be an important predictor
of lPTS. A recent meta-analysis by Samokhvalov et al. [44] reported
a strong correlation between alcohol use and seizures with a
34 S.-Y. Won et al. / Seizure 45 (2017) 28–35
pooled OR of 2.19 across all studies. A pathophysiological
mechanism for lPTS in patients with chronic alcoholism was
suspected to involve multiple head traumas, structural changes
such as brain atrophy, changes in neurotransmitter pathways and
metabolic changes, thereby lowering the convulsive threshold
[5,44,45]. Interestingly, older age was not a significant risk factor
despite the fact that brain atrophy was a significant parameter for
epileptic seizure [15,17,30,37]. In this context, it seems to be
important not to classify general brain atrophy as a predictor of
seizures, but to evaluate the cause of brain atrophy, such as alcohol
consumption or previous stroke [30]. Furthermore, HU density of
SDH was related to seizure occurrence in cSDH. Patients with
mixed-density haematomas were more prone to seizures than
patients with low-isodense haematomas. A pathophysiological
explanation could be that blood degradation products may lead to
fibrinogen degradation products, which contribute to membrane
formation as well as irritation of the brain parenchyma. On the
other hand, high activation of tissue plasminogen activators may
lead to more rebleeding and consequently to a larger haematoma
mass effect [31,35,46,47].
The diagnosis of seizures depends on either clinical seizure
manifestation or seizure patterns on an EEG. In particular,
subclinical seizures are one of the problems leading to delayed
treatment and worse outcome. Several case reports described
epileptiform abnormalities such as periodic lateralized epilepti-
form discharges (PLEDs), which are periodic spikes or sharp waves
occurring in 1–3 s intervals, in patients with SDH [33,48–50].
PLEDs might be a sign of a continuum between ictal and interictal
status and could be a biomarker of an increased probability of the
occurrence of subclinical seizures [48]. A retrospective study
reported very high incidence of EEG abnormalities (e.g. slowing,
asymmetries, epileptiform discharges, PLEDs) on EEGs in 21 out of
24 patients (87%) with aSDHs [33], implying that the occurrence
ofPTSs in patients with aSDHs might be higher than noticed in
routine clinical treatmentin clinical routine. EEG abnormalities and
the occurrence of PTSs were associated with unfavourable functional
outcome and impairment of quality of life [10,33]. Currently, the use
of AEDs in patients with recurrent PTSs might be advisable, but there
is no consensus on the use of prophylactic AEDs.
Initially, the benefit of prophylactic AEDs was described in
patients with TBI. Temkin et al. [13] analysed, in a double-blind
fashion, the occurrence of PTSs after TBI in 404 patients either
treated with phenytoin or placebo and found that patients
undergoing prophylactic phenytoin treatment had significantly
fewer ePTSs than placebo-treated patients with 0.25 RR (4% vs.
14%). A previous study had shown no significant difference
between those groups; however, the incidence of PTS was very
low, making an interpretation of the effect of phenytoin difficult
(4% PTS after TBI) [37]. A meta-analysis comparing AEDs reported a
promising effect of phenytoin- and carbamazepine-induced
reduction of ePTS after TBI. However, the incidence of lPTS was
not significantly reduced by using prophylactic AEDs [12,24,39].
One study observing the use of phenytoin even reported delay in
functional recovery after brain damage; therefore, the American
Academy of Neurology recommended prophylactic use of phenyt-
oin within the first 7 days after TBI to decrease the risk of PTS, but
not beyond 7 days [23,54]. Meanwhile, phenytoin has mostly been
replaced by new-generation AEDs like levetiracetam, since
phenytoin has significantly more side effects, such as hepatic
cytochrome P450 induction, cutaneous reaction, cardiac compli-
cation, thrombocytopenia and drugs interaction, which led in one
study to a discontinuation rate of 40% [40,41].
Comparing levetiracetam and phenytoin, levetiracetam
appeared to have a similar efficacy in reducing PTSs with
significantly lower complication rates (except for gastrointestinal
problems) and better long-term outcome [14,18,40,42,43]. In
addition, levetiracetam is not subject to therapeutic drug-level
monitoring, as phenytoin is [40,42], leading to better compliance.
However, prophylactic AEDs are only used in the short term; hence
the results should be interpreted with caution in case of
extrapolating the data into long-term treatment.
Unfortunately, there is currently no prospective, randomized
study describing the use of prophylactic AEDs in patients with
exclusively SDHs. In the case of aSDHs, prophylactic AED
treatmentsuch as levetiracetam seems to be beneficial in reducing
PTSs and improving functional outcome. In the case of cSDHs, several
retrospective studies reported contrary results, but one should note
that no differentiation between high-risk and low-risk patients was
made when administering prophylactic AEDs [15,16,25].
The present study has limitations that should be addressed.
First, the incomplete recruitment due to our inclusion and
exclusion criteria caused a selection bias. Second, most of the
studies were of a retrospective nature and had small sample sizes.
Third, there were limited data for certain subgroups, leading to
imputation of missing values. Despite the limitations, our study is
the first systematic review to summarize specific seizure occur-
rence in SDHs, and the results were able to consistently illustrate
the correlation between SDHs and PTSs.
5. Conclusion
The overall incidence of ePTS and lPTS was higher in patients
with aSDHs than in those with cSDHs. Risk factors for PTSs in
aSDHs were preoperative and 24 h post-operative low GCS and
craniotomy, whereas in cSDHs the risk factors were alcohol abuse,
change of mental status, previous stroke and haematoma density
on the CT. The results of our study suggest, first, differentiating the
type of SDH, and secondly, summarizing the risk factors for ePTSs
and lPTSs in order to decide either for or against the use of
prophylactic AEDs. In patients with severe aSDHs, a prophylactic
anticonvulsive treatment might be considered for a limited, short
period of time, e.g. 1 week. In patients with cSDHs, the decision to
apply prophylactic AEDs should be made individually depending
on the cumulative risk factors. To verify the beneficial effect of
prophylactic AEDs, further randomized controlled studies are
warranted in the future.
Ethical publication statement
We confirm that we have read the journal’s position on issues
involved in ethical publication and affirm that this report is
consistent with those guidelines.
Funding source
This research did not receive any specific grant from funding
agencies in the public, commercial or not-for-profit sectors.
Acknowledgements
We thank Marina Heibel and Anne Sicking for their excellent
technical support.
References
[1] Hamasaki T, Yamada K, Kuratsu J. Seizures as a presenting symptom in
neurosurgical patients: a retrospective single-institution analysis. Clin Neurol
Neurosurg 2013;115:2336–40.
[2] Huang KT, Bi WL, Abd-El-Barr M, et al. The neurocritical and neurosurgical care
of subdural hematomas. Neurocrit Care 2015.
[3] Temkin NR. Risk factors for posttraumatic seizures in adults. Epilepsia 2003;44
(Suppl. 10):18–20.
 S.-Y. Won et al. / Seizure 45 (2017) 28–35
[4] Annegers JF, Hauser WA, Coan SP, et al. A population-based study of seizures
after traumatic brain injuries. N Engl J Med 1998;338:20–4.
[5] Seifi A, Asadi-Pooya AA, Carr K, et al. The epidemiology, risk factors, and impact
on hospital mortality of status epilepticus after subdural hematoma in the
United States. Springerplus 2014;3:332.
[6] Phuenpathom N, Choomuang M, Ratanalert S. Outcome and outcome
prediction in acute subdural hematoma. Surg Neurol 1993;40:22–5.
[7] Howard 3rd MA, Gross AS, Dacey Jr. RG, et al. Acute subdural hematomas: an
age dependent clinical entity. J Neurosurg 1989;71:858–63.
[8] Servadei F. Prognostic factors in severely head injured adult patients with
acute subdural haematomas. Acta Neurochir (Wien) 1997;139:279–85.
[9] Wiedemayer H, Triesch K, Schafer H, et al. Early seizures following non-
penetrating traumatic brain injury in adults: risk factors and clinical
significance. Brain Inj 2002;16:323–30.
[10] Rabinstein AA, Chung SY, Rudzinski LA, et al. Seizures after evacuation of
subdural hematomas: incidence, risk factors, and functional impact. J
Neurosurg 2010;112:455–60.
[11] Englander J, Bushnik T, Duong TT, et al. Analyzing risk factors for late
posttraumatic seizures: a prospective, multicenter investigation. Arch Phys
Med Rehabil 2003;84:365–73.
[12] Haltiner AM, Temkin NR, Dikmen SS. Risk of seizure recurrence after the first
late posttraumatic seizure. Arch Phys Med Rehabil 1997;78:835–40.
[13] Temkin NR, Dikmen SS, Wilensky AJ, et al. A randomized, double-blind study of
phenytoin for the prevention of post-traumatic seizures. N Engl J Med
1990;323:497–502.
[14] Szaflarski JP, Sangha KS, Lindsell CJ, et al. Prospective, randomized, single-
blinded comparative trial of intravenous levetiracetam versus phenytoin for
seizure prophylaxis. Neurocrit Care 2010;12:165–72.
[15] Sabo RA, Hanigan WC, Aldag JC. Chronic subdural hematomas and seizures:
the role of prophylactic anticonvulsive medication. Surg Neurol 1995;43:579–
82.
[16] Rubin G, Rappaport ZH. Epilepsy in chronic subdural haematoma. Acta
Neurochir (Wien) 1993;123:39–42.
[17] Grobelny BT, Ducruet AF, Zacharia BE, et al. Preoperative antiepileptic drug
administration and the incidence of postoperative seizures following burr
hole-treated chronic subdural hematoma. J Neurosurg 2009;111:1257–62.
[18] Radic JA, Chou SH, Du R, et al. Levetiracetam versus phenytoin: a comparison of
efficacy of seizure prophylaxis and adverse event risk following acute or
subacute subdural hematoma diagnosis. Neurocrit Care 2014;21:228–37.
[19] Asikainen I, Kaste M, Sarna S. Early and late posttraumatic seizures in
traumatic brain injury rehabilitation patients: brain injury factors causing late
seizures and influence of seizures on long-term outcome. Epilepsia 1999;40
(May (5)):584–9.
[20] Frey LC. Epidemiology of posttraumatic epilepsy: a critical review. Epilepsia
2003;44(Suppl. 10):11–7.
[21] Glotzner FL, Haubitz I, Miltner F, et al. Seizure prevention using carbamazepine
following severe brain injuries. Neurochir (Stuttg) 1983;26:66–79.
[22] Hammond EJ, Ramsay RE, Villarreal HJ, et al. Effects of intracortical injection of
blood and blood components on the electrocorticogram. Epilepsia 1980;21
(February (1)):3–14.
[23] Chang BS, Lowenstein DH, Quality Standards Subcommittee of the American
Academy of Neurology. Practice parameter: antiepileptic drug prophylaxis in
severe traumatic brain injury: report of the Quality Standards Subcommittee
of the American Academy of Neurology. Neurology 2003;60:10–6.
[24] Temkin NR. Antiepileptogenesis and seizure prevention trials with antiepi-
leptic drugs: meta analysis of controlled trials. Epilepsia 2001;42:515–24.
[25] Ohno K, Maehara T, Ichimura K, et al. Low incidence of seizures in patients with
chronic subdural haematoma. J Neurol Neurosurg Psychiatry 1993;56:1231–3.
[26] Ratilal BO, Pappamikail L, Costa J, et al. Anticonvulsants for preventing seizures
in patients with chronic subdural haematoma. Cochrane Database Syst Rev
2013;6:CD004893.
[27] Luxon LM, Harrison MJ. Chronic subdural haematoma. Q J Med 1979;48:43–53.
[28] Hirakawa K, Hashizume K, Fuchinoue T, et al. Statistical analysis of
chronicsubdural hematoma in 309 adult cases. Neurol Med Chir (Tokyo)
1972;12:71–83.
[29] Kotwica Z, Brzeinski J. Epilepsy in chronic subdural haematoma. Acta
Neurochir (Wien) 1991;113:118–20.
[30] Huang YH, Yang TM, Lin YJ, et al. Risk factors and outcome of seizures after
chronic subdural hematoma. Neurocrit Care 2011;14:253–9.
35
[31] Brodersen P, Gjerris F. Regional cerebral blood flow in patients with chronic
subdural hematomas. Acta Neurol Scand 1975;51(March (3)):2339.
[32] Moher D, Liberti A, Tetzlaff J, et al. Preferred reporting items for systematic
reviews and meta-analyses: the PRISMA statement. J Clin Epidemiol 2009;62
(October (10)):1006–12, doi:http://dx.doi.org/10.1016/j.jclinepi.2009.06.005.
[33] Rudzinski LA, Rabinstein AA, Chung SY, et al. Electroencephalographic findings
in acute subdural hematoma. J Clin Neurophysiol 2011;28:633–41.
[34] Robinson RG. Chronic subdural hematoma: surgical management in
133 patients. J Neurosurg 1984;61:263–8.
[35] Chen CW, Kuo JR, Lin HJ, et al. Early post-operative seizures after burr-hole
drainage for chronic subdural hematoma: correlation with brain CT findings. J
Clin Neurosci 2004;11:706–9.
[36] Jennett B. Epilepsy and acute traumatic intracranial haematoma. J Neurol
Neurosurg Psychiatry 1975;38:378–81.
[37] Cole M, Spatz E. Seizures in chronic subdural hematoma. N Engl J Med
1961;265:628–31.
[38] Dudek H, Michno T, Michalski J, et al. Late posttraumatic epilepsy in patients
with an alcoholic problem treated surgically for posttraumatic chronic
subdural hematomas. Rocz Akad Med Bialymst 1999;44:119–27.
[39] Young B, Rapp RP, Norton JA, et al. Failure ofprophylactically administered
phenytoin to prevent late posttraumatic seizures. J Neurosurg 1983;58
(February (2)):236–41, doi:http://dx.doi.org/10.3171/jns.1983.58.2.0236.
[40] Zafar SN, Khan AA, Ghauri AA, et al. Phenytoin versus leviteracetam for seizure
prophylaxis after brain injury—a meta analysis. BMC Neurol 2012;12:30.
[41] Allen JP, Ludden TM, Burrow SR, et al. Phenytoin cumulation kinetics. Clin
Pharmacol Ther 1979;26:445–8.
[42] Szaflarski JP, Sangha KS, Lindsell CJ, et al. Prospective, randomized, single-
blinded comparative trial of intravenous levetiracetam versus phenytoin for
seizure prophylaxis. Neurocritical Care 2010;12(April (2)):165–72, doi:http://
dx.doi.org/10.1007/s12028-009-9304-y.
[43] Lim DA, Tarapore P, Chang E, et al. Safety and feasibility of switching from
phenytoin to levetiracetam monotherapy for glioma-related seizure control
following craniotomy: a randomized phase II pilot study. J Neurooncol
2009;93:349–54.
[44] Samokhvalov AV, Irving H, Mohapatra S, et al. Alcohol consumption,
unprovoked seizures, and epilepsy: a systematic review and meta-analysis.
Epilepsia 2010;51:1177–84.
[45] Dam AM, Fuglsang-Frederiksen A, Svarre-Olsen U, et al. Late-onset epilepsy:
etiologies, types of seizure, and value of clinical investigation, EEG, and
computerized tomography scan. Epilepsia 1985;26:227–31.
[46] Lim DJ, Chung YG, Park YK, et al. Relationship between tissue plasminogen
activator, plasminogen activator inhibitor and CT image in chronic subdural
hematoma. J Korean Med Sci 1995;10:373–8.
[47] Nomura S, Kashiwagi S, Fujisawa H, et al. Characterization of local
hyperfibrinolysis in chronic subdural hematomas by SDS-PAGE and immuno-
blot. J Neurosurg 1994;81:910–3.
[48] Westmoreland BF. Periodic lateralized epileptiform discharges after evacua-
tion of subdural hematomas. J Clin Neurophysiol 2001;18:20–4.
[49] Chu NS. Acute subdural hematoma and the periodic lateralized epileptiform
discharges. Clin Electroencephalogr 1979;10:145–50.
[50] Mehryar GR, McIntyre HB. Periodic lateralized epileptiform discharges
associated with subdural hematoma. Bull Los Angeles Neurol Soc
1975;40:8–1251.
[51] Liberati A, Altman DG, Tetzlaff J, et al. The PRISMA statement for reporting
systematic reviews and meta-analyses of studies that evaluate health care
interventions: explanation and elaboration. PLoS Med 2009;6(July (7)):
e1000100.
[52] Hart YM, Sander JW, Johnson AL, et al. National General Practice Study of
Epilepsy: recurrence after a first seizure. Lancet (London, England) 1990;336
(November (8726)):1271–4.
[53] Hesdorffer DC, Benn EK, Cascino GD, et al. Is a first acute symptomatic seizure
epilepsy? Mortality and risk for recurrent seizure. Epilepsia 2009;50(May
(5)):110208.
[54] Naidech AM, Kreiter KT, Janjua N, et al. Phenytoin exposure is associated with
functional and cognitive disability after subarachnoid hemorrhage. Stroke: J
Cerebral Circ 2005;36(March (3)):583–7.
[55] Ertl J, Hapfelmeier J, Peckmann T, et al. Guideline conform initial monotherapy
increases in patients with focal epilepsy: a population-based study on German
health insurance data. Seizure 2016;41:9–15.