Epilepsia, 39(9):978-982, 1998
Lippincott Williams & Wilkins, Philadelphia
0 International League Against Epilepsy
Nonepileptic Seizures After Head Injury
Lauren E. Westbrook, Orrin Devinsky, and Romergryko Geocadin
Department o f Neurology, New York University School o f Medicine, New York, New York, U.S.A.
Summary: Purpose: To examine the role of head injury as a
risk factor in the development of nonepileptic seizures (NES).
Specifically, we will determine the relative frequency of head
injury among NES patients referred to our center and will de-
scribe several pertinent clinical features and personal charac-
Methods: Retrospective record review of patients referred to
our center for evaluation of seizures over a 4-year period. All
patients with NES were evaluated as in a previously described
protocol, which included intensive video EEG monitoring,
provocation by suggestion, and psychiatric interview. All NES
patients with a history of head injury were extracted for this
Results: Of 102 patients with NES, nearly one-third (32%)
had an antecedent head injury; 52% were male, mean age was
34 years, and 12% had coexisting epilepsy. Multiple psychiat-
ric disorders were not uncommon (79%), and a history of abuse
was found in 35%. All but four patients had documented fi-
nancial gain from their injury. Follow-up at 1 year found poor
long-term outcome with lasting disability; despite that, the ma-
jority (91%) of head injuries were minor.
Conclusions: Our preliminary findings suggest that prior
head injury is associated with the development of NES and may
contribute to the pathogenesis of NES in vulnerable patients.
Head injury and sexual or physical abuse appear to occur in
comparable proportions in patients with NES. This suggests
that head injury and abuse may be equally important risk fac-
tors in the development of NES. Key Words: Nonepileptic
seizures-Psychogenic seizures-Head trauma-Risk factors.
NONEPILEPTIC SEIZURES AFTER
Nonepileptic seizures (NES) are paroxysmal events
that resemble epileptic seizures but do not result from a
primary abnormality of excessive neuronal activity. The
origins of NES include a wide variety of psychogenic
and physiologic disorders (1). NES are common, ac-
counting for between 10-20% of all inpatient hdeo-EEG
monitoring unit admissions (2,3), and up to 37% of pa-
tients diagnosed with possible epilepsy in a city hospital
general neurology clinic (4).
NES of psychogenic etiology have captured the atten-
tion of psychiatrists and neurologists, but many myster-
ies about this disorder remain. Many characteristics of
individuals who experience psychogenic NES have been
well documented. For example, NES tends to present in
young women in their late 20s to early 30s (1). Depres-
sion is a frequent concomitant diagnosis in NES (3,
personality disorders occur in 8-54% of NES patients
Accepted April 24, 1998.
Address correspondence and reprint requests to Dr. L. E. Westbrook
at Department of Neurology, Hospital for Joint Diseases, 301 East 17th
St., Room 204, New York, NY, 10003, U.S.A.
Dr. Geocadin is currently at the Division of Neurosciences, Critical
Care, Johns Hopkins Hospital, Baltimore, Maryland, 21287 U.S.A.
(1). Recovery or immediate improvement can occur
when the NES diagnosis is presented (1,6,7). For most
with recent onset of NES and some with chronic NES,
long-term functioning can be quite good with appropriate
psychiatric treatment (1,7).
Research is just beginning to identify risk factors for
NES. Sexual and physical abuse are associated with the
development of NES. A. history of physical or sexual
abuse was found in 32% of NES patients from our center
(8); slightly higher rates have been reported elsewhere
(9). This rate sharply contrasts with the much smaller
proportion (9%) of patients with epilepsy who have com-
parable sociodemographic backgrounds and who also re-
ported past abuse (8). Epilepsy may be another risk fac-
tor for NES. Rates of co-occurrence vary between stud-
ies, but 10-35% is the usual range (6). No other
predisposing factors for NES are well documented.
Risk factors in the development of epilepsy have been
more thoroughly studied. Head injury is a significant risk
factor in the pathogenesis of epilepsy (10). Relative risk
depends on severity and length of time since trauma.
Annegers and colleagues (1 1) report from their popula-
tion-based study of seizures after head trauma that the
risk was 7.1% within 1 year after severe head injury. The
relative risk after 1 year is considerably lower after mod-
erate (0.7-4.0%) and mild (0.1-1.5%) injury (11,12).
NONEPILEPTIC SEIZURES AFTER HEAD INJURY
Thirty-seven percent of those who develop posttraumatic
seizures do so within 1 year (1 1). The risk of developing
posttraumatic epilepsy a5 years after the event is -25%
(12) for all severity groups combined. The risk of sei-
zures within 5 years after severe head injury is 11.5%;
for moderate head injury, 1.6%; and for mild head injury,
0.6% (1 1).
Although head trauma is common, little is known
about patients who develop NES after documented in-
jury. This preliminary report explores the role of head
injury as a possible risk factor in the development of
NES. In this article we will (a) determine the relative
frequency of head injury among patients with NES re-
ferred for evaluation to a major tertiary care epilepsy
center, and (b) describe clinical features of the head
trauma, NES-related characteristics, and other individual
factors of patients with NES after head injury.
We retrospectively reviewed records of patients ad-
mitted to our Comprehensive Epilepsy Center for evalu-
ation of seizures between January 1990 and July 1994.
All patients with a diagnosis of NES were identified and
those with a history of head injury were extracted for this
report. Every patient admitted to our center undergoes
the same systematic questioning about prior head inju-
ries, whether the topic of head injury arises spontane-
ously or not during the course of the clinical interview. In
all cases, patients attributed the cause of their events to
the antecedent head trauma. All patients underwent in-
tensive video-EEG monitoring and provocation testing
with suggestion by the NES protocol previously de-
scribed (13). Those few patients who did not have a
positive provocation test were diagnosed with NES by
clinical features, video-EEG findings, and psychiatric
evaluation. NES patients with coexisting epilepsy were
included. Follow-up after 1 year either by subsequent
clinic appointment or by telephone was attempted in all
cases. Long-term functioning after diagnosis was as-
sessed by patient report of current employment status
and the existence of ongoing financial benefits.
All patients were interviewed by a psychiatrist who
specializes in assessing people with epilepsy. Psychiatric
diagnoses were made by DSM-111-R criteria using the
Structured Clinical Interview for DSM (SCID); diag-
noses tended to be inclusive. Whenever possible, patients
were divided into two NES groups based on their psy-
chiatric diagnosis: conversion NES and nonconversion
NES. More detailed information describing how this dis-
tinction is made has been published (13). We classified
NES as nonconversion if the axis I DSM-111-R disorder
was not conversion disorder and could completely ex-
plain the behavioral manifestation of the nonepileptic
Head injury was defined as a nonpenetrating injury
with contact or acceleratioddeceleration trauma to the
head. We specified a priori that the injury be nonpen-
etrating due to the high correlation of penetrating head
injury with subsequent development of epilepsy. How-
ever, no patients were actually excluded due to this cri-
terion. Head injury was graded according to the Head
Injury Severity Scale (HISS) (14). The HISS is a five-
interval severity scale based on initial postresuscitation
Glasgow Coma Scale score and the associated compli-
cations. Patients with minor head injury are divided into
minimal, mild, and moderate severity categories. Mini-
mal injury corresponds to a Glasgow Coma Scale inter-
val of 15, with no loss of consciousness (LOC) or am-
nesia and no other indications for admission. Mild injury
corresponds to a 14 (or a 15 plus amnesia) with brief (<5
min) LOC or impaired alertness of memory and an in-
tracranial lesion on neuroimaging study. The Glasgow
Coma Scale interval of 9-13 (or LOC >5 min or focal
neurologic deficit) corresponds to the moderate severity
category, with the added complication of an intracranial
lesion evident on neuroimaging. The serious head injury
category on the HISS includes severe (scale interval 5-
8), with complications such as brainstem hemorrhage or
effacement of brainstem cistern, and critical (scale inter-
val of 3-4), with loss of pupillary reflex or severe non-
neurologic injury. Severity level of head injury did not
restrict patient selection.
Among 102 inpatients diagnosed with NES, 33 (32%)
had a documented head injury. Individual sociodemo-
graphic characteristics of these patients are consistent
with the population served by this center. There were 17
males (52%) and 16 females (48%). Mean age was 34
years and ranged from 17 to 57 years. Over half (58%)
were married and 85% were white. Most patients (88%)
had at least a high school education and many (59%)
were college educated. Most patients (85%) were em-
ployed or in school at the time of the head injury: only
five were known to be unemployed or not in school.
Those who were employed before head injury appeared
to be largely skilled workers; a few patients were pro-
fessionals. Four patients (12%) had a diagnosis of coex-
isting epilepsy confirmed.
Spontaneous NES was observed in all patients and
video-EEG monitoring was consistent with NES in each
instance. Provocation testing with suggestion induced
NES typical of spontaneous NES in 24 (77%) patients.
Psychiatric history (inpatient or outpatient treatment)
was positive in 21 (64%). Eight of these patients (38%)
had at least one previous psychiatric hospitalization, six
of whom had attempted suicide at least once in the past.
Epiiepsia, Vol. 39, No. 9. 1998
L. E. WESTBROOK ET AL.
Positive history of physical or sexual abuse was elicited
in 11 of the 31 patients (35%) for whom this information
Psychiatric diagnosis was documented for all but one
patient. A principal (primary) diagnosis of conversion
disorder was made in 85% of patients (28), while 15%
(five patients) were diagnosed with nonconversion NES.
Factitious disorder or malingering accounted for four of
the five patients with nonconversion NES; panic disorder
was diagnosed in the other. More than three-quarters of
all patients (79%) had more than one psychiatric diag-
nosis at the time of NES diagnosis: anxiety disorder,
52%; personality disorder, 42%; mood disorder, 39%;
other somatoform disorders, 9%; drug and alcohol de-
pendence, 6%; attention deficit disorder, learning dis-
ability, and dissociative disorder, 3% each.
Notably, all patients or their representatives (e.g., par-
ent, spouse) considered the head injury to cause the sei-
zures. Onset of NES after head injury ranged from im-
mediate to >9 years, but largely occurred within 1 year
(81%). Nine of 32 patients (28%) developed events
within the first week of their injury; 11 of 32 patients
(34%) developed events between 1 week and 6 months
of their injury; 6 of 32 (19%) within 6 months to 1 year;
4 of 32 (13%) within 1-5 years of injury; and 2 of 32
(6%) developed NES >5 years after head trauma. Over
one-third of NES patients (34%) had been having the
episodes for >10 years.
The majority (91%) of head injuries were minor as
categorized by the HISS: 11 were considered minimal;
17 were considered mild; and 2 were moderate. Only
three patients had severe head injuries, although none
were critical. Twelve patients (36%) had head injury
from motor vehicle accidents, including one patient who
was involved in a car accident on the job. Nine patients
(27%) had work-related head injuries, 10 (30%) had as-
sault-related head injuries, and two (6%) had a sports-
related injury. There was no LOC in 12 cases. Among
those patients who did have LOC at the time of head
injury, most (16 of 20) lasted 4
in a coma after head injury: one for 3 days with antero-
grade memory impairment for 1 month, the other for 7
days. One patient experienced 10-12 h of LOC, while
another had amnesia for 3 days after the trauma with no
LOC. Those with co-morbid epilepsy all had either mild
or minimal head injuries.
Other injuries were varied when they did occur, and
only two could be considered serious. Neuroimaging
studies were normal in 28 patients (85%). Findings were
positive in five patients: of those, three had moderate to
severe head trauma. Initial medical treatment ranged
from none to ICU admission, and included ER evaluation
and hospital admission. All ICU admissions (total of
three) were for severe or moderate head injury. The three
min. Two patients were
hospital admissions were regular admissions and were
for mild and minimal injuries.
All but four of the 31 patients for whom this informa-
tion was available had documented financial gain from
their injury that could be considered an external incen-
tive responsible for the persisting disability as mani-
fested by NES. At the time of their evaluation, 87% were
receiving some type of financial remuneration in the
form of disability benefits and workman’s compensation,
were in the process of litigation for damages. or had
submitted disability claims. Disability and employment
status 1 year after the NES diagnosis could only be as-
sessed in 22 patients due to patient refusal, incomplete or
unconfirmed information, or inability to contact the pa-
tient. Follow-up showed that 20 (91%) had poor long-
term outcomes: 17 were still disabled in that they were
not functioning at premorbid levels (defined as resump-
tion of previous job or comparable work), and all con-
tinued to receive financial and personal benefit. One pa-
tient (with factitious disorder) died of an intentional drug
overdose. Only two of the patients (7%) who had been
employed at the time of the head injury were still gain-
fully working 1 year later.
Results of this preliminary study imply that head
trauma may be a frequent precipitant to psychogenic
NES. Head trauma was identified as the cause of the
“seizures” by the patient or family in nearly one-third of
all NES cases referred to our center for evaluation. This
proportion is comparable to other reports of postconcus-
sive head trauma patients experiencing epilepsy-like
symptom constellations (1 5). It is also approximately
equal to the reported rates of sexual abuse in NES (8),
suggesting that head trauma and abuse may be equally
important risk factors in the development of NES.
The descriptions generated by this report identify sev-
eral characteristics of patients with NES after head injury
that do not appear typical of those with posttraumatic
epilepsy or of the general NES population. The primary
purpose of describing these characteristics is to stimulate
hypotheses for future work examining the relative role of
vqious risk factors in the development of NES after head
injury versus NES without associated head trauma and
posttraumatic epilepsy. First, injury severity levels were
predominantly mild or minimal (85%) in the patients
studied here. This is in contrast to the relative risk of
epilepsy after mild or minimal head injury, which is
widely accepted to be no greater than the risk for the
general population (12,16). The probability of develop-
ing epileptic seizures correlates with the severity of the
head injury (1 1,17,18): from 2% with moderate trauma
to 12% for survivors of severe trauma (18). Taken to-
Epilepsia, Vol. 39, No. 9, 1998
NONEPILEPTIC SEIZURES AFTER HEAD INJURY
gether, these findings suggest that mild head injury plays
a substantially larger role as a risk factor in the epide-
miology of NES than it does in epilepsy. We did find,
however, that a small proportion of NES patients actually
had severe head injuries (9%). It has also been reported
that minor head trauma may cause subsequent epilepsy
in selected cases (19). Therefore, the nature and severity
of the head injury may not point to the correct seizure
type in every case.
Second, NES usually presents in people who are in
their late 20s or early 30s (6,13), but the mean age of
patients in our sample was somewhat older at 34 years.
Third, contrary to the female gender predominance (3: 1)
of most NES series (2,13), patients in this study were
equally likely to be male. In head trauma and other types
of accidents, males tend to predominate at an approxi-
mate ratio of 2.5: 1 (lo). Sex-specific prevalence rates of
epilepsy show fairly equal proportions of males to fe-
males (1 2), resembling the distribution we found here.
Finally, while the most frequently diagnosed nonconver-
sion NES disorders are reported to be anxiety, psychosis,
and impulse control (13), our preliminary data suggest
that factitious disorders and malingering may be more
common nonconversion diagnoses among patients with
NES after head injury than reported in the general NES
population. This would be consistent with the prepon-
derance of external monetary motives present in the
population studied here.
The mechanism by which head injury leads to the
elaboration of psychogenic NES remains unknown. The
association between organic brain disease and conver-
sion symptomatology has been studied and described
since the time of Charcot (Charcot's Syndrome or Gross
Hysteria Syndrome) (6,1520). The vulnerability of the
brain to behavioral dysfunction after neurologic insult is
often cited as a causal pathogenesis model (6,20,21).
Other explanatory models include symptom modeling
(6,22), particularly well-suited to account for the high
rate of co-occurrence between epilepsy and NES, and
even the toxic effects of the antiepileptic medications
(23). Psychiatric vulnerability (24) may also be an etio-
logic factor in the psychiatric disability associated with
NES after head injury. Preexisting mental illness or ge-
netic tendencies may predispose a person to develop psy-
chiatric symptomatology after neurological insult.
While many legitimate disability claims and litigations
are initiated by people with head injury, the observation
that nearly every patient in this study could benefit fi-
nancially from being disabled suggests another explana-
tory factor. Whether external monetary incentives actu-
ally drive psychiatric disorders causing NES, at least
after head injury, is not known. Head injury may present
an opportunity to achieve both primary and secondary
gain for the individual. The underlying financial com-
pensation, the relief from the pressures of work, the per-
sonal investment in medical explanations for symptoms
encouraged in litigious atmospheres, and the concomi-
tant delay of psychiatric treatment may all contribute to
the enduring illness and disability. Consistent with this
explanation is the poor functional prognosis noted in our
patients 1 year after NES diagnosis.
In summary, our exploratory study suggests that prior
head injury is associated with the development of NES
and may contribute to the pathogenesis of NES in vul-
nerable patients. Future comparison studies with larger
samples will be needed to further elucidate the risk of
head injury relative to other factors and to statistically
compare the characteristics of patients with NES after
head injury with other NES and epilepsy groups. Until
then, clinical evaluation of seizure patients referred after
minor head injury will continue to require careful ap-
praisal and individualized patient assessment.
1. Devinsky 0, Thacker K. Nonepileptic seizures. Neurol Clin 1995;
2. Krumholz A, Niedermeyer E. Psychogenic seizures: a clinical
study with follow-up data. Neurology 1983;33:498-502.
3. Lesser RP. Psychogenic seizures. In: Pedley TA, Meldrum BS,
eds. Recenf advances in epilepsy, vol. 12. Edinburgh: Churchill-
4. Bazil CW, Kotari M, Luciano D, et al. Provocation of nonepileptic
seizures by suggestion in a general seizure population. Epilepsia
5. Roy A, Barris M. Psychiatric concepts in psychogenic non-
epileptic seizures. In: Rowan AJ, Gates JR, eds. Non-epileptic
seizures. Boston: Butterworth-Heinemann, 1993: 143-52.
6. Alper K. Nonepileptic seizures. Neurol Clin 1994; 12: 153-73.
7. Buchanan N, Snars J. Pseudoseizures (non epileptic attack disor-
der)-clinical management and outcome in 50 patients. Seizure
1993;2: I4 1-6.
8. Alper K, Devinsky 0, Perrine K, Vazquez B, Luciano D. Nonepi-
leptic seizures and childhood sexual and physical abuse. Neurology
9. Betts T, Boden S. Diagnosis, management and prognosis of a
group of 128 patients with nonepileptic attack disorder. Part 11.
Previous childhood sexual abuse in the etiology of these disorders.
0. Jennett B, Teasdale G. Management of head injuries. Philadelphia:
F. A. Davis, 1981:281-7.
1. Annegers JF, Grabow JD, Groover RV, Laws ER, Elveback LR,
Kurland LT. Seizures after head trauma: a population study. Neu-
2. Hauser WA. Hesdorffer DC. Epilepsy. Frequency, causes and con-
sequenies. New York: Demos, 1990:72-3.
3. Alper K, Devinsky 0, Perrine K, Vazquez B, Luciano D. Psychi-
atric classification of nonconversion nonepileptic seizures. Arch
Neurol l995;52: 199-201.
14. Stein SC, Spettell C. The head injury severity scale (HISS): a
practical classification of closed head injury. Bruin Injury 1995;9:
15. Novelly RA. Cerebral dysfunction and cognitive impairment in
nonepileptic seizure disorders. In: Rowan AJ, Gates JR, eds. Non-
epileptic seizures. Boston: Butterworth-Heinemann, 1993:233-42.
16. Annegers JF. The epidemiology of epilepsy. In: Wyllie E, ed. The
treatment of epilepsy. Principles and practices. Philadelphia: Lea
& Febiger, 1993:157-64.
Epilepsia, Vol. 39, NO. 9, IYYX
L. E. WESTBROOK ET AL.
17. Jennett B, Frankowski RF. The epidemiology of head injury. In:
Braakman R, ed. Handbook of clinical neurology, vol. 13. New
York: Elsevier, 1990 1-16.
18. Weiss GH, Feeney DM, Caveness WF, et al. Prognostic factors for
the occurrence of posttraumatic epilepsy. Arch Neurol 1983;40:7-
19. Devinsky 0. Epilepsy after minor head trauma. Epilepsia 1996;9:
20. Eames P. Hysteria following brain injury. J Neurol Neurosurg
Psychiatry 1992;55: 1046-53.
21. Fenton GW. Epilepsy and hysteria. Br J Psychiatry 1986;146:28-
22. Toone BK. Disorders of hysterical conversion. In: Bass C, ed.
Somatization: physical symptoms and psychological illness. Lon-
don: Blackwell, 1990:207-34.
23. Niedermeyer E, Blumer D, Holscher E, Walker BA. Classical hys-
terical seizures facilitated by anticonvulsant toxicity. Psychiatr
Clin 1970;3:7 1-84.
24. Lishman WA. Physiogenesis and psychogenesis in the ‘post con-
cussional syndrome.’ Br J Psychiarry 1988; 153:460-9.
Epilepsia, Vd. 39, No. 9, 1998