www.elsevier.com/locate/braindev

Brain & Development 30 (2008) 527–532

Original article

Seizure precipitants in children with intractable epilepsy

Peng-Cheng Fang a, Yung-Jung Chen b,*, Inn-Chi Lee c

a Department of Pediatrics, Sin-Lau Christian Hospital, Tainan, Taiwanb Department of Pediatrics, Medical Center, National Cheng Kung University, 138 Sheng-Li Road, Tainan 70428, Taiwan

c Department of Pediatrics, Chung-Shan Medical University Hospital, Taichung, Taiwan

Received 6 April 2007; received in revised form 3 January 2008; accepted 16 January 2008

Abstract

Purpose: To investigate the seizure precipitants in children with intractable epilepsy, and to determine any distinctive clinical fea-tures contributing to seizures in these patients. Methods: A questionnaire and seizure diary prepared by the parents of the patients.Demographic and seizure data were reviewed. Results: Of 120 patients with intractable epilepsy, 74 (62%) had one (n = 43), two(n = 23), or three seizure precipitants (n = 8). The three most common precipitants were illness or fever (32%), sleep deprivation(13%), and menstruation (10%). Of these precipitants, inducing factors (endogenous origin) were more common than triggering fac-tors (exogenous origin): 73% versus 27%, respectively. Three distinctive clinical features – neurological abnormalities (P = 0.01),status epilepticus (P = 0.017), and abnormal neuroimaging (P = 0.007) – were significantly more common in patients with thanin patients without precipitants. Conclusions: Prompt recognition and management of seizure precipitants has practical implicationsfor treating patients with refractory epilepsy. Such patients can be counseled to avoid specific precipitants.� 2008 Elsevier B.V. All rights reserved.

Keywords: Children; Epilepsy; Intractable; Seizure precipitant

1. Introduction

Epilepsy is a chronic neurologic condition character-ized by recurrent, unprovoked seizures requiring contin-uous medication for long-term management. However,10–20% of children with epilepsy may not have remis-sion despite appropriate therapy [1]. Such cases areoften termed intractable. Intractability, however, is notevenly distributed among patients with epilepsy. It ismore common in those with mental retardation, neuro-logical deficits, or both, and generally in patients withdetectable structural brain damage. Some children withrefractory seizure are able to induce or control seizuresby means of mental or physical activity [2,3].

0387-7604/$ - see front matter � 2008 Elsevier B.V. All rights reserved.

doi:10.1016/j.braindev.2008.01.004

* Corresponding author. Tel.: +886 6 235 3535; fax: +886 6 2753083.

E-mail address: pcyj@mail.ncku.edu.tw (Y.-J. Chen).

Seizure precipitants are defined as a variety of envi-ronmental phenomena or specific functional alterationsof the body that precede the onset of an epileptic attackand are considered factors that lead to epileptic seizures.These precipitants include both seizure-inducing (envi-ronmental or endogenous origin) and seizure-triggeringfactors (chemical or physiologic origin). The mechanismsby which seizure precipitants may reduce seizure thresh-old or modulate its occurrence are not fully understood.Epilepsy is a group of heterogeneous disorders, each witha different pathophysiology. Thus, seizure precipitantsare individually variable. The incidence of precipitantsmay be of particular value in suggesting possibilitiesfor seizure control as a supplement to conventional man-agement, especially in patients with medically refractoryepilepsy. The present study aimed to investigate the sei-zure precipitants in children with intractable epilepsyand to determine the characteristics of the clinical vari-ables in patients with seizure-precipitating factors.

Table 1The number of patients with each precipitant

Precipitantsa Total patients (n = 74)

Seizure-inducing factors 83 (73%)Fever or illness 37 (32%)Sleep deprivation 15 (13%)Menstruation 11 (10%)Emotional stress 9 (8%)Sleep 6 (5%)Fatigue 3 (3%)Fright 2 (2%)

Seizure-trigger factors 31 (27%)Loud noise 9 (8%)Change weather 6 (5%)Television 5 (4%)Heat or humidity 4 (4%)Touch 4 (4%)Sunshine, flickering light 3 (3%)

a Thirty-one patients had more than one precipitant: 2 precipitants(23), 3 precipitants (8).

528 P.-C. Fang et al. / Brain & Development 30 (2008) 527–532

2. Patients and methods

We recruited 798 consecutive children with epilepsywho had been treated and followed-up between 1 Sep-tember 2003 and 31 August 2005 at the National ChengKung University Hospital, a tertiary public referral hos-pital serving the southern region of Taiwan. All partici-pants fulfilled the following criteria of intractableepilepsy [1]: (1) were between 5 and 16 years old, (2)had at least one seizure in the past month despite treat-ment with two or more antiepileptic drugs (AEDs), (3)had an average of one seizure per month for >18months, and (4) had no more than a 3-month seizure-free hiatus during those 18 months. Exclusion criteriawere (a) febrile seizures or seizures classified as acutesymptomatic or situation-related, (b) noncompliancewith physician-prescribed AED treatment.

The demographic and clinical characteristics of thestudy population were collected. Information on theprecipitants that were identified as associated withseizures the previous 6 months was obtained using aquestionnaire that provided a list of possible seizure-pre-cipitants. The parents of all the patients were given adiary in which they recorded details of each seizure overthe following month. This was used as a check on thereliability of the information from the seizurequestionnaires.

In this study, precipitants were defined as those fac-tors (from the list of possible precipitants) present dur-ing at least two seizures and that apparently inducedan increase >50% above the patient’s baseline frequencyof seizures. Patients with intractable epilepsy were clas-sified into two groups: patients with precipitants andpatients without precipitants. The clinical and neuroim-aging features of these patients were compared and ana-lyzed to find the determinants which might havecontributed to seizure recurrence. Institutional reviewboard approval was given for this study, and informedconsent was obtained from parents.

2.1. Statistical analysis

The clinical data of the patients in each group wereanalyzed using one-way analysis of variance (ANOVA).Categorical variables between groups were analyzedusing contingency tables. Statistical significance wasset at P < 0.05.

3. Results

We enrolled 120 patients with intractable epilepsy(71 boys and 49 girls; mean age, 12.8 ± 2.7 years;range, 5.2–15.4 years) for this study. Overall, 74patients (62%) indicated seizure precipitants: one preci-pitant, 43 (58%); two precipitants, 23 (31%); and threeprecipitants, 8 (11%) (Table 1). At least 9% of the

patients or their parents listed, in descending order, ill-ness or fever, sleep deprivation, or menstruation. Tele-vision, sunshine or flickering light, heat or humidity,touch, fatigue, and fright were less frequently listed.Of the 37 patients who reported seizures induced by ill-ness or fever, 26 had a respiratory system infection and11 had gastroenteritis.

Precipitating factors were divided into seizure-induc-ing and seizure-trigger factors. Seizure-inducing factorswere significantly more often identified than were sei-zure-trigger factors (n = 83, 73% versus n = 31, 27%).The three most common (n = 63, 55%) seizure-inducingfactors were illness or fever, sleep deprivation, and men-struation, the most common seizure-trigger factor(n = 9, 8%) was loud noise.

3.1. Seizure types and epileptic syndromes of patients with

precipitants

The most common seizure types were partial seizureswith secondary generalization (frequency = 51%) andcomplex partial seizures (frequency = 33%) (Table 2).More than half the patients (n = 61, 56%) were symp-tomatic localization-related epilepsy, and more than aquarter (n = 32, 27%) were cryptogenic localization-related epilepsy. Temporal lobe epilepsy (n = 42, 69%)was the most common type of localization-related epi-lepsy, frontal lobe epilepsy the second most common(n = 12, 19%), occipital lobe epilepsy the third, (n = 6,10%), and parietal lobe epilepsy the fourth (n = 1, 2%).

Between the two groups with seizure precipitants thetypes of seizures and epileptic syndromes were compara-ble, except that symptomatic temporal lobe epilepsy wasmore common in patients with a seizure-inducing factorthan in patients with seizure-triggering factors(P = 0.018).

Table 2Distributions of seizure types and epileptic syndromes in 74 patientswith seizure-inducing factors or seizure-trigger factors

Seizure type Seizure-inducingfactors(n = 83)

Seizure-triggerfactors(n = 31)

Totalfactors(n = 114)

SeizuresPrimary

generalization3 (3%) 2 (6%) 5 (4%)

Simple partialseizure

11 (13%) 3 (10%) 14 (12%)

Complexpartial seizure

29 (35%) 8 (26%) 37 (33%)

Secondarygeneralization

40 (48%) 18 (58%) 58 (51%)

Epileptic syndromesLocalized-related

Cryptogenic 20 (24%) 5 (16%) 25 (22%)Symptomatic 46 (55%) 21 (68%) 67 (59%)

Generalized epilepsySymptomatic 15 (18%) 5 (16%) 20 (18%)Unclassified 2 (2%) 0 2 (2%)

P.-C. Fang et al. / Brain & Development 30 (2008) 527–532 529

3.2. Seizure frequency

Of the 74 patients with seizure precipitants, 54 (73%)patients had a moderate monthly rate of seizures (2–4/month), and 20 (27%) had a high rate (>1/week) (Table3). A remarkable increase in the frequency of seizureevents when precipitating factors occurred: the meanrose to 7.4 ± 2.1/month and 18.4 ± 4.4/month(P < 0.0001).

3.3. Antiepileptic drug treatment

Thirty-two patients were taking two AEDs, 22 weretaking three AEDs, and 20 were taking more than three.Carbamazepine (n = 53) and sodium valproate (n = 49)were the most frequently prescribed medications. Theothers were lamotrigine (n = 34), clobazam (n = 28),clonazepam (n = 19), topiramate (n = 17), phenobarbi-tone (n = 5), and phenytoin (n = 4).

3.4. EEG

Serial EEGs were done on all patients (mean, 8.3;range, 5–18). There were epileptiform discharges onthe EEGs of all 120 patients: 102 had focal spikes and

Table 3Number of seizures in 74 patients at baseline and during seizure precipitatio

Group Baseline (mean ± SD/month)

Moderatea (n = 54) 3.1 ± 0.6 (range, 2–4)Highb (n = 20) 11.2 ± 4.5 (range, 5–20)

a Moderate frequency: 2–4 seizures per month.b High frequency: >5 seizures per month.* Two-tailed t-test.

18 had generalized spikes and waves. Of the patientswith focal epileptiform activity, 62 had seizure precipit-ants and 40 had none. There were no significant differ-ences in the location of focal spikes between patientswith and patients without precipitants.

3.5. Comparison of clinical features between epileptic

children with and without precipitants

There was no significant difference between patientswith and without seizure precipitants in gender, age ofonset of epilepsy, duration of epilepsy, family historyof epilepsy, history of neonatal seizures, average seizurefrequency, or types of seizure and epileptic syndrome(Table 4); however, symptomatic temporal lobe epilepsywas more common in patients with than without precip-itants (P = 0.019).

There were three statistically significant and distinc-tive differences between the two groups: neurologicalabnormalities (P = 0.01), history of status epilepticus(P = 0.017), and abnormal neuroimaging (P = 0.007)were found more often in patients with than withoutprecipitants.

We found a higher incidence of neurological deficitsin patients with than without precipitants. Sixty-sevenof the 120 (56%) patients with seizure precipitants hadsubnormal intelligence quotients (IQ < 70) (56%): 47 ofthe 74 (64%) patients with precipitants and 20 of 46(43%) patients without precipitants (P = 0.03). Thirty-one (31/120, 26%) patients were severely mentalretarded (IQ < 30): 26 (84%) of them had seizure-precip-itating factors and 5 (16%) had none (P = 0.03). Therewere no differences, however, in the motor deficitsbetween the two groups. Of the 26 patients with severemental retardation, 19 (73%) had two precipitantsand 7 (17%) had three precipitants. Hence, a largerpercentage of patients who reported two (83%) or three(88%) factors provoking seizure were severely mentalretarded.

4. Discussion

In the present study we demonstrated that 74 (62%)of 120 children with intractable epilepsy cited at leastone precipitating factor for seizure. The prevalence rateis consistent with previous studies reported [4,5]. How-ever, the difference in prevalence rates found between

n

During precipitation (mean ± SD/month) P value*

7.4 ± 2.1 (range, 3–12) <0.000118.4 ± 4.4 (range, 12–26) 0.0001

Table 4Comparison of clinical features between epileptic children with and without precipitants

Clinical variables With precipitants (n = 74) Without precipitants (n = 46) P value

Gender (male/female) 32/42 17/29 NSAge epilepsy diagnosed (yrs)

Mean ± SD (year) (range) 4.6 ± 3.3 (0.1–11) 4.8 ± 3.7 (0.1–11.8) NSAverage of seizure frequency

Mean ± SD (month) (range) 5.3 ± 4.3 (2–20) 5.7 ± 2.9 (3–19) NSDuration of follow-up

Mean ± SD (year) (range) 7.7 ± 3.2 (3–14) 8.1 ± 3.4 (4–14) NS

AED (2/c 3) 32/42 21/25 NSNeurological abnormality 55 (74%) 23 (50%) 0.01

Mental retardation 29 15Motor deficits 8 3Global deficits 18 5

Family history of epilepsy 3 (4%) 1 (2%) NSHistory of neonatal seizures 3 (4%) 5 (11%) NSHistory of SE 25 (34%) 6 (13%) 0.017

Seizure types NSSPS 9 (12%) 4 (9%)CPS 26 (35%) 16 (35%)PGS 2 (3%) 2 (4%)SGS 25 (34%) 18 (39%)CPS + SGS 12 (16%) 6 (13%)

Epileptic syndrome NSCryptogenic localized 20 (27%) 19 (41%)

TLE/XTLE 11/9 11/8

Symptomatic localized 41 (56%) 19 (41%)TLE/XTLE 31/10 8/11 0.019

Symptomatic generalized 12 (16%) 6 (13%)Unclassified 1 (1.4%) 2 (4%)

Neuroimaging (total) (n = 59) (n = 39) 0.007Normal 6 (10%) 13 (33%)Abnormal 53 (90%) 26 (67%)

AED, antiepileptic drugs; CPS, complex partial seizure; PGS, primary generalized seizure; SE, status epilepticus; SGS, secondary generalized seizure;SPS, simple partial seizure; TLE, temporal lobe epilepsy; XTLE, ex-temporal lobe epilepsy.

530 P.-C. Fang et al. / Brain & Development 30 (2008) 527–532

our study and that of Verduyn et al. [3], 401 (90%) of446 children with epilepsy, may be attributable to agedifferences or differences in the severity of seizure disor-ders between the two groups.

The seizure precipitants and their order of frequencyin this study were similar to those reported in previousstudies [3–7], except the commonest precipitant of sei-zures. Illness or fever was the most frequently reportedprecipitant in the present study, whereas it was stressin previous studies [3–5]. Stress accounted for 20–30%of precipitants, and was most often reported in adultpatients with epilepsy [4,5,14–16] than in children [3].The discrepancy of the results may be due to differingstudy populations and methods of data collection. Pre-vious studies [4,5,8] reported that under stress condi-tions, seizure frequency increased primarily in adultpatients.

One study [8] using a questionnaire-based survey ofthe perception of stress and seizure on 89 adult patientswith epilepsy found that 58 (64%) patients believed that

stress increased the frequency of their seizures. There-fore, stress reduction may be regarded as an additionalpotential therapeutic modality for epilepsy.

The mechanisms by which seizure precipitants mayreduce the seizure threshold or modulate its occurrenceare not fully understood. Illness or fever was the mostfrequently cited precipitant in the present study eventhough we excluded patients under 5 years old, in whomfebrile convulsion is usually seen. Thus, none of ourpatients had a situation-related epileptic syndrome,namely febrile seizure. Three possible mechanisms havebeen proposed to explain the occurrence of seizures insuch patients with illness or fever. First, a high incidenceof illness may be related to a susceptible condition ofimmune-compromised children with refractory epilepsy[9,10]. Second, an acute febrile infection such as gastro-enteritis may provoke a seizure in susceptible patients,especially in patients with a developmental disability[11–13]. Twenty-six (70%) of 37 patients who reportedillness or fever as precipitants of their seizures had a

P.-C. Fang et al. / Brain & Development 30 (2008) 527–532 531

respiratory infection, and 11 (30%) had gastroenteritis.Third, 89 (74%) of our study population had neurolog-ical deficits or were mental retarded, or both, generallywith detectable brain damage. Patients with neurologi-cal deficits were inactive or in poor health; therefore,they had a higher susceptibility to illness than healthypatients did. Fever or illness may have reduced the sei-zure threshold and precipitated seizures in thesepatients.

Some of the reported precipitants often coexisted inthe same individuals [3,14]. Tan et al. [14] reported thatemotional stress, compliance, and sleep deprivationinterplay in adult patients with combinations of precip-itants. Therefore, a precipitant may increase seizure fre-quency on its own as well as interact with otherprecipitants to induce seizures. The significant correla-tions between these precipitants suggest that their mech-anisms of action may be related and act similarly insusceptible syndromes, which could indicate that theseprecipitants represent different facets of the same patho-physiology, as in the case of their interactions in sleepdisorders [17]. The high levels of inter- and intra-individ-ual variability in seizure precipitants provide a furtherindication of the complexity of the relationship betweenseizure precipitants and seizure recurrence.

Patients with seizure clustering may have poorer epi-lepsy control than patients without clustering do. There-fore, seizure clustering appears to be a marker forintractable epilepsy. Most studies have not identifiedspecific precipitants for seizure clustering [12,13].Because the patients in the present study all had intrac-table epilepsy, we found a higher-than-expected inci-dence of seizure clustering. In some cases of refractoryepilepsy, it is possible to control seizures by modifyingthe stimuli, the precipitants [18]. It may be of practicalvalue to suggest possibilities for seizure control as a sup-plement to the conventional management of epilepsy.

The differential distribution of endogenous precipit-ants provides insights into the pathophysiology of spe-cific epilepsy. Frucht et al. [4] found that stress,fatigue, and sleep deprivation positively correlated witheach other and were often reported by patients withsymptomatic localization-related epilepsies. The signifi-cant correlations among these three precipitants suggestthat their mechanisms of action may be related and actsimilarly in susceptible syndromes. This finding mightindicate that these precipitants represent different facetsof the same pathophysiology, as in the case of theirinteractions in sleep disorders [17,19,20]. In the presentstudy, although we found no significant difference in epi-leptic syndromes, we did find that more than half thepatients with precipitants had symptomatic localiza-tion-related epilepsy, 55% with inducing factors and68% with triggering factors. The most common epilepticsyndrome in these patients was symptomatic temporallobe epilepsy (31/41 = 76%).

Seizure-inducing factors originate from endogenousor environmental events, and seizure-trigger factorsoriginate from exogenous or physical condition; the for-mer was more common than the latter in the presentstudy. A seizure elicited by some specific stimulus orevent is called ‘‘reflex epilepsy”. The types of stimuli thatevoke seizures are flickering light, complex activities,movement, auditory stimulation, and tapping andtouching [21–25]. In startle epilepsies, a sudden unex-pected noise may cause someone with epilepsy to bestartled or to jump, thereby inducing a tonic seizure.Such seizures are relatively common in patients withcerebral palsy and Lennox–Gastaut syndrome [26].Accordingly, there are maneuvers that seem to reducethe risk of photosensitive seizures, including monocu-lar-complete occlusion, increased distance from avideo-display terminal, avoidance of bright and flicker-ing lights, and ambient back lighting [21–25]. Sleepdeprivation is an easily preventable seizure precipitantin most cases [18,19].

Intractable epilepsy is commonly believed to be themain cause of uncontrolled seizures and hospitalizationin epileptic patients. Tan et al. [14] examined 40 adultepileptic patients admitted with seizures and found thatthree-quarters of the seizures leading to admission wereassociated with potentially preventable precipitants.They concluded that patient education may be impor-tant for decreasing the occurrence of seizures and, there-fore, hospital admissions.

Precipitants are thought to be avoidable throughactions taken by patients or physicians. Aird [2] foundgood results in a study of 500 drug-resistant with refrac-tory epilepsy: 17% became seizure-free and 25% experi-enced significantly reduced seizure frequency, achievedsimply by promoting moderate lifestyle changes.

Previous studies on the seizure precipitants were eval-uated using patient [4,5,7,8,14,27–29] or parental reports[3]. Information obtained from patient reports in dairiesmay at times be inaccurate because patients either maynot be aware of all the seizures they had or they mayhave forgotten some of them [5]. There are several pit-falls when interpreting self-reported information on sei-zure precipitants. One is that it is difficult to determinewhich the actual precipitant was when several arereported to have simultaneously occurred [5]. When epi-leptic patients are unable to determine or unaware of therelevant precipitants, for example, young children orwith mentally retarded patients, it is necessary to getthat information from caregivers [3,30]. In the presentstudy we investigated the precipitating factors usingreports from the parents of our patients, more thantwo thirds of whom were mentally retarded and mostof whom were unable to clearly describe the event thatprecipitated the seizure. Furthermore, all these patientshad intractable epilepsy and a great number of seizuresdespite most of them being given multiple-drug therapy.

532 P.-C. Fang et al. / Brain & Development 30 (2008) 527–532

The parents were both experienced observers and morereliable reporters than their children were; therefore,we accepted their ability to accurately identify whichprecipitants exacerbated their children’s seizures.

Status epilepticus is a symptom of brain dysfunction.It is more common in those with mental retardation,multiple handicaps, and refractory epilepsy. Steffenburget al. [6] reported in their population-based study of 98mentally retarded children that 45% had intractable sei-zures [6]. Spector et al. [7] reported that 91% of 100adults with poorly controlled epilepsy recognized atleast one provocative factor cited. Status epilepticushas been reported in more than two-thirds of patientswith Lennox–Gastaut syndrome [31]. Similarly, in thisstudy we found that a greater rate of patients with thanwithout precipitants had a history of status epilepticus.

There are few studies [2,14] that compare clinical fea-tures between precipitant-positive and precipitant-nega-tive patients. In the present study, we found that threedistinctive clinical features – neurological deficits,abnormal neuroimaging, and a history of status epilep-ticus – were more common in patients with precipitants.These findings revealed that epileptic patients withremarkably structural brain lesions have a greater inci-dence of precipitant-exacerbated seizures, and that moreprecipitants are reported in patients with more severebrain dysfunction. However, further studies on largepopulations and a wide spectrum of epileptic syndromesare necessary to confirm these findings.

In conclusion, the majority of patients with intracta-ble epilepsy have specific precipitants. Informationabout these seizure precipitants may help physiciansimprove seizure control in their patients, especiallypatients with refractory epilepsy, by teaching themhow to avoid specific precipitants.

References

[1] Berg AT, Kelly MM. Defining intractability: comparisons amongpublished definitions. Epilepsia 2006;47:431–6.

[2] Aird RB. The importance of seizure-inducing factors in thecontrol of refractory forms of epilepsy. Epilepsia 1983;24:567–83.

[3] Verduyn CM, Stores G, Missen A. A survey of mothers’impressions of seizure precipitants in children with epilepsy.Epilepsia 1988;29:251–5.

[4] Frucht MM, Quigg M, Schwaner C, Fountain NB. Distributionof seizure precipitants among epilepsy syndromes. Epilepsia2000;41:1534–9.

[5] Nakken KO, Solaas MH, Kjeldsen MJ, Friis ML, Pellock JM,Corey LA. Which seizure-precipitating factors do patients withepilepsy most frequently report? Epilepsy Behav 2005;6:85–9.

[6] Steffenburg U, Hedstrom A, Lindroth A, Wiklund LM, HagbergG, Kyllerman M. Intractable epilepsy in a population-based seriesof mentally retarded children. Epilepsia 1998;39:767–75.

[7] Spector S, Cull C, Goldstein LH. Seizure precipitants andperceived self-control of seizures in adults with poorly-controlledepilepsy. Epilepsy Res 2000;38:207–16.

[8] Haut SR, Vouyiouklis M, Shinnar S. Stress and epilepsy: a patientperception survey. Epilepsy Behav 2003;4:511–4.

[9] Gross-Tsur V, Shalev RS, Kazir E, Engelhard D, Amir N.Intravenous high-dose gammaglobulins for intractable childhoodepilepsy. Acta Neurol Scand 1993;88:204–9.

[10] Billiau AD, Wouters CH, Lagae LG. Epilepsy and the immunesystem: is there a link? Eur J Pediatr Neurol 2005;9:29–42.

[11] Millichap JG, Millichap JJ. Role of viral infections in the etiologyof febrile seizures. Pediatr Neurol 2006;35:165–72.

[12] Uemura N, Okamura A, Negoro T, Watanabe K. Clinicalfeatures of benign convulsions with mild gastroenteritis. BrainDev 2002;24:745–9.

[13] Narchi H. Benign afebrile cluster convulsions with gastroenteritis:an observation study. BMC Pediatr 2004;4:2.

[14] Tan JH, Wilder-Smith E, Lim EC, Ong BK. Frequency ofprovocative factors in epileptic patients admitted for seizures: aprospective study in Singapore. Seizure 2005;14:464–9.

[15] Neugebauer R, Paik M, Hauser WA, Nadel E, Leppik I, SusserM. Stressful life events and seizure frequency in patients withepilepsy. Epilepsia 1994;35:336–43.

[16] Temkin NR, Davis GR. Stress as a risk factor for seizures amongadult with epilepsy. Epilepsia 1984;25:450–6.

[17] Waters WF, Adams Jr SG, Binks P, Varnado P. Attention, stressand negative emotion in persistent sleep-onset and sleep-mainte-nance insomnia. Sleep 1993;16:128–36.

[18] Ellingson RJ, Wilken K, Bennett DR. Efficacy of sleep depriva-tion as an activation procedure in epilepsy patients. J ClinNeurophysiol 1984;1:83–101.

[19] Malow BA, Passaro E, Milling C, Minecan DN, Levy K. Sleepdeprivation does not affect seizure frequency during inpatientvideo-EEG monitoring. Neurology 2002;59:1371–4.

[20] Wilkins AJ, Darby CE, Binnie CD, Stefansson SB, Jeavons PM,Harding GF. Television epilepsy – the role of pattern. ElectroenClin Neuro 1979;47:163–71.

[21] Harding W, Jeavons J. Photosensitive epilepsy. London: MacKeithPress; 1992, p. 125–50.

[22] Millett CJ, Fish DR, Thompson PJ, Johnson A. Seizures duringvideo-game play and other common leisure pursuits in knownepilepsy patients without visual sensitivity. Epilepsia 1999;40(Sup-pl. 4):59–64.

[23] Fylan F, Harding GF, Edson AS, Webb RM. Mechanisms ofvideo-game epilepsy. Epilepsia 1999;40(Suppl. 4):28–30.

[24] Inoue Y, Fukao K, Araki T, Yamamoto S, Kubota H,Watanabe Y. Photosensitive and nonphotosensitive electronicscreen game-induced seizures. Epilepsia 1999;40(Suppl. 4):8–16.

[25] Shiraishi H, Fujiwara T, Inoue Y, Yagi K. Photosensitivity inrelation to epileptic syndromes: a survey from an epilepsy centerin Japan. Epilepsia 2001;42:393–7.

[26] Saenz-Lope E, Herranz FJ, Masdeu JC. Startle epilepsy: a clinicalstudy. Ann Neurol 1984;16:78–81.

[27] Spatt J, Langbauer G, Mamoli B. Subjective perception of seizureprecipitants: results of a questionnaire study. Seizure1998;7:391–5.

[28] Antebi D, Bird J. The facilitation and evocation of seizures. Aquestionnaire study of awareness and control. Br J Psychiat1993;162:759–64.

[29] Hayden M, Penna C, Buchanan N. Epilepsy: patient perceptionsof their condition. Seizure 1992;1:191–7.

[30] Cull CA, Fowler M, Brown SW. Perceived self-control of seizuresin young people with epilepsy. Seizure 1996;5:131–8.

[31] Beaumanoir A, Foletti G, Magistris M, Volanschi D. Statusepilepticus in Lennox–Gastaut syndrome. In: Niedermeyer E,Degen R, editors. The Lennox–Gastaut syndrome. NewYork: Alan R Liss; 1988. p. 283–99.