Suicide attempts in adult patients with idiopathic generalized epilepsy


  • Eriko Hara md, phd,

    1. Department of Clinical Neurophysiology & Epilepsies, Guy's & St Thomas' NHS Foundation Trust, London,
    2. Section of Psychiatry and Behavioral Science, Tokyo Medical and Dental University Graduate School, Tokyo,
    Search for more papers by this author
  • Nozomi Akanuma md, phd,

    Corresponding author
    1. Department of Clinical Neurophysiology & Epilepsies, Guy's & St Thomas' NHS Foundation Trust, London,
    2. Lambeth Rehabilitation Services, South London & Maudsley NHS Foundation Trust,
    3. Department of Clinical Neuroscience, Institute of Psychiatry, King's College London, London,
    Search for more papers by this author
  • Naoto Adachi md, phd,

    1. Adachi Mental Clinic, Sapporo,
    Search for more papers by this author
  • Koichiro Hara md, phd,

    1. Department of Clinical and Experimental Epilepsy, Institute of Neurology, University College London, London, UK and
    2. Department of Psychiatry, Asai Hospital, Chiba, Japan
    Search for more papers by this author
  • Michael Koutroumanidis md

    1. Department of Clinical Neurophysiology & Epilepsies, Guy's & St Thomas' NHS Foundation Trust, London,
    Search for more papers by this author

Nozomi Akanuma, MD, PhD, Lambeth Rehabilitation Services, South London & Maudsley NHS Foundation Trust, 308/312 Brixton Road, London SW9 6AA, UK. Email:


Aims:  To study clinical features of adult patients with idiopathic generalized epilepsy with special attention to suicidal behavior.

Methods:  We reviewed the medical records of 145 consecutive adult patients with electro-clinically confirmed idiopathic generalized epilepsy and identified those with a history of at least one attempted suicide. Clinical variables in relation to their epilepsy and psychiatric conditions were analyzed.

Results:  Seven patients (4.8%) had a history of suicide attempts with drug overdose, and one of these patients committed suicide after multiple attempts. All attempts were made interictally without direct relation to their epileptic seizures. All had at least one co-morbid mental disorder (two with dual diagnosis). Although their psychiatric diagnoses varied, they all appeared to have increased emotional instability and poor impulse control. Only one patient's attempt was directly associated with her co-morbid depression.

Conclusions:  Physicians managing people with epilepsies should be aware of psychiatric disturbances and suicidal behavior in idiopathic generalized epilepsy.

PATIENTS WITH CHRONIC epilepsy may have an increased risk (five times or more) of attempting or committing suicide compared to the general population.1–3 Such an increased risk is sometimes related to co-morbid psychiatric illnesses, most commonly depression. Some studies have suggested that the risk is particularly high in patients with temporal lobe epilepsy (TLE),1,4 but this has not been universally accepted.5,6 In contrast to TLE, little is known about suicidal behavior of patients with idiopathic generalized epilepsy (IGE).

Patients with IGE have long been thought to have a generally favorable seizure control without cognitive impairment or psychiatric co-morbidity. However, recent observations have suggested that adults with IGE may have considerable rates of mental and psychosocial problems,7–10 which could lead to an increased risk of attempting suicide. We studied a cohort of consecutive adult IGE patients with special attention to suicide attempts.


We reviewed the medical records of IGE patients aged 16 or over who were consecutively referred to and followed up in the tertiary Epilepsy Clinic of St Thomas' Hospital between January 1990 and December 2004,10 to identify those with a history of at least one attempted suicide. Most of the referrals to our epilepsy clinic come from the area's general practitioners, the neurological and other specialist services of Guy's and St Thomas' Hospitals including the pediatric services, and the accident and emergency department of St Thomas' Hospital through a dedicated ‘first seizure’ clinic. Approximately 10% are tertiary referrals from outside the catchment area of the two hospitals including overseas referrals. The diagnosis of IGE is based on the official clinical and electroencephalogram (EEG) diagnostic criteria set by the International League Against Epilepsy.11,12 Our clinical and EEG methodology on syndromic diagnosis and classification of IGE has been detailed elsewhere.13 Patients are not only prospectively studied but also retrospectively investigated through old medical notes, past EEG reports or actual tracings to better characterize their epilepsy and its natural history. The follow-up duration included periods during which relevant information was confirmed in the records. We reviewed and evaluated the age at onset and the duration of epilepsy, the seizure types and IGE sub-syndromes, and the outcome of seizure-control. Satisfactory seizure-control was defined as having no seizures of any type during at least one year from last follow up, and such patients were classified as seizure-free. Follow up shorter than two years was deemed insufficient for a reliable definition of outcome; therefore, patients falling in this category were excluded from the analysis.

Psychiatric diagnoses were made in accordance with the Tenth Revision of the International Classification of Diseases and Related Health Problems.14 For the patients reviewed or seen for the first time during the last five years psychiatric diagnoses were made after clinical interviews with the qualified neuropsychiatrists of our team (E. H. and N. Ak.), whereas for the few remaining patients whose last clinical follow up was before 2000, psychiatric diagnoses were based on past medical correspondence and previous hospital or community psychiatric reviews.

In all patients with at least one attempted suicide who had complete psychiatric history with formal psychiatric assessments and follow up, we evaluated additional parameters: seizure types and frequency, particularly around the time of the attempted suicides; antiepileptic and psychotropic medication; psychiatric state and relevant past history; drugs taken for overdose; previous overdose attempts; and triggers of the attempts.


We enrolled 145 consecutive IGE patients (46 men and 99 women, mean age 34.3, median 30.0, range 16 to 72 years) with follow up of more than two years (mean 17.3, median 15.0, range 2 to 51 years) and complete medical records. There were 41 patients with juvenile myoclonic epilepsy, 23 with juvenile absence epilepsy, three with clinical phenotypes overlapping juvenile myoclonic and absence epilepsy, 12 with IGE with generalized tonic-clonic seizures (GTCS) only, eight with childhood absence epilepsy, 16 with the syndrome of phantom absences,15 13 with eyelid myoclonia with absences, seven with perioral myoclonia with absences, five with reflex (mainly photosensitive) epilepsy, and 17 with clinical–EEG phenotypes that were not possible to classify in the known IGE sub-syndromes. Regarding seizure types, 111 patients had absences, 62 had myoclonic seizures and 138 had GTCS; 22 patients had only one seizure type, 82 had two and 41 had three types of seizure. Thirty-six patients (24.8%) had co-morbid mental disorders (four patients with dual diagnosis): two substance misuse disorders (F1x), two schizophrenia spectrum disorders (F2x), 18 mood disorders (F3x), 10 neurotic disorders (F4x), seven personality disorders (F6x) and one case of behavioral and emotional disorders (F9x). Forty-eight patients (33.1%) were seizure-free and 97 were not seizure-free.

Seven patients (six women and one man, 4.8%) attempted suicide at least once, and one of them completed suicide (Table 1). All used drug overdoses as the means. Six patients took an overdose of antiepileptic drugs, and one took antidepressants. The doses taken were relatively high in four patients, low in one and unknown in two.

Table 1.  Clinical and psychiatric characteristics of seven patients who took a drug overdose as serious suicide attempts
Case/age (y)/sex1/41/F2/51/M3/39/F4/43/F5/26/F6/34/F7/24/F
  • Reported in Reference14.

  • This patient later committed suicide under uncertain circumstances (information not available).

  • CNZ, clonazepam; ESM, ethosuximide; F, female; GTCS, generalized tonic-clonic seizures; ICD-10, Tenth Revision of the International Classification of Diseases and Related Health Problems; IGE, idiopathic generalized epilepsy; LTG, lamotrigine; M, male; PB, phenobarbitone; PHT, phenytoin; VPA, sodium valproate.

Demographic characteristics
Epilepsy onset age / duration of epilepsy / duration of follow up (y)5 / 36 / 2118 / 33 / 2611 / 28 / 1814 / 29 / 85 / 21 / 2110 / 24 / 243 / 21 / 21
IGE sub-syndrome/ conditionsUnclassified IGE with photosensitivityGTCS with phantom absencesJuvenile myoclonic epilepsyJuvenile myoclonic epilepsyUnclassified IGE with absencesJuvenile absence epilepsyEyelid myoclonia with absences
Absences / myoclonus / GTCS− / + / ++ / − / ++ / + / ++ / + / ++ / − / ++ / + / ++ / + / +
Characteristics related to suicide attempts
Number of suicide attempts22MultipleMultiple112
Age at the latest attempt26403642212918
Seizure control at the time of the attempt(s)Monthly GTCSSeizure freeSeizure freeSeizure freeWeekly absencesWeekly absences and monthly GTCSYearly GTCS
Antiepileptic / psychotropic medicationPHT / amitriptylineVPA / nonePB, PHT, CNZ / citalopramVPA, CNZ / fluoxetineESM, LTG, CNZ / noneVPA, LTG, ESM / noneVPA, CNZ, LTG / none
Trigger of the latest attemptRelationship break-upJob change and relapse of his mother's bipolar disorderFamily issuesArgument with her husbandDivorce and consequent troubles with ex-husbandSelf-discontinuation of fluoxetineMother's suicide
Drugs taken in the latest attemptAmitriptyline 1.5 gVPA 18 gPB 3.5 g, PHT 12 g, indomethacine 8.3 gVPA 20 g, CNZ 60 mgESM, LTG, CNZ, Dose not knownVPA 3 g, LTG 25 mg, ESM 500 mgVPA, CNZ, LTG (with excess of alcohol), Dose not known
Psychiatric diagnosis (ICD-10)DysthymiaEmotionally unstable personality disorderSomatization disorder / Emotionally unstable personality disorderDysthymia / Emotionally unstable personality disorderEmotionally unstable personality disorderDepressionAdjustment disorder

The age at the most recent suicidal attempt ranged from 18 to 42 years, the duration of epilepsy from 21 to 36 years, and the follow-up period from 8 to 26 years. None of the epilepsy-related factors (age of seizure onset, IGE sub-syndromes, types, frequency and control of seizures, and types and dosage of antiepileptic drugs) was common across them. Around the time of their suicide attempts, three patients were seizure-free and four were not seizure-free. The proportion of seizure freedom (3 out of 7, 42.9%) was comparable to that in those without suicidal attempts (32.6%). Careful assessment of the medical records in relation to the timing of the suicide attempts showed that no attempt was associated with increased frequency of absences, absence or myoclonic status, or convulsive seizures.

All seven patients met at least one category (two patients with dual diagnosis) of mental disorders interictally: three had mood disorders, two neurotic disorders and four personality disorders. Although their psychiatric diagnoses varied, they appeared to have in common emotional instability and poor impulse control according to their psychiatric or psychological assessments. Only one patient's attempt (Patient 6, Table 1) was directly associated with a co-morbid psychiatric disorder (a relapse of depression). In the other six patients, attempts were triggered by some psychosocial events; their suicidal ideations occurred suddenly and diminished quickly. Five patients attempted suicide recurrently. One of them (Patient 4, Table 1) eventually committed suicide. The event came to our attention two years later, and the circumstances are unknown despite our efforts.


We present seven patients who attempted suicide with drug overdose out of 145 consecutive adults with electro-clinically confirmed diagnosis of IGE and long follow up. The rate of suicide attempts was 4.8% during our follow-up period. In the general population, the lifetime prevalence of suicide attempts is 2–4%,16,17 usually observed in people aged 15 to 49 years with greatest risk of its completion in the middle-aged or older. Given the median age of 30 years in our series, the rate of 4.8% appears to be higher and it could increase with age. Previous studies on patients with epilepsy have reported five- to 25-fold increased suicide rates, with the highest risk in patients with TLE.1 Suicide attempts are also more frequent in patients with epilepsy than in comparatively handicapped controls with other chronic disabilities,18 while attempts by overdose are again more frequent in patients with complex partial seizures.5 To our knowledge, this is the first observational study of suicidal behavior in patients with clinically-EEG confirmed diagnosis of IGE, and suggests an increased rate although not as high as in focal epilepsies.

The suicide attempts in our patients were not related to the type, frequency or timing of seizures, the particular IGE sub-syndromes or the antiepileptic medication. Similarly, lack of association between suicide attempts and factors directly related to epilepsy itself has been noted in patients with focal epilepsies and complex partial seizures.5 Suicidal behavior does not appear to be a direct consequence of epilepsies per se.

All seven patients with suicidal attempts had at least one co-morbid mental disorder, accounting for 19.4% (7 out of 36) among patients with psychiatric co-morbidity. This may suggest that psychiatric co-morbidity in IGE patients is associated with suicidal attempts. With the exception of Patient 6, our patients' attempts were primarily associated with interictal psychopathological characteristics, which are in keeping with the personality traits characterized in some IGE patients, including lack of self-control, impressionability and emotional lability.7,9,19 Conversely, possible causative psychiatric conditions of suicide attempts in focal epilepsies include interictal severe mood disturbance or psychosis.4,5 Moreover, a significant proportion of suicide acts in patients with TLE are reportedly associated with their peri-ictal affective or psychotic symptoms.2,4 Although suicidal behavior in epilepsies in general is more likely related to their psychopathological disturbances than epilepsy-related factors, underlying psychopathologies may be different between focal epilepsies and IGE.

Notably, six out of the seven patients were women and the majority of attempted suicides failed. While a higher rate of suicide attempts in women than men has been reported in focal epilepsies,2 it is not known whether a similar tendency exists in IGE. Our result may reflect the female preponderance in our IGE samples but could equally reflect the higher prevalence of impulsive suicide attempts by self-poisoning in the female population.17

There are certain methodological problems in this report, some of which seem to be common in such studies.6 Analysis of different parameters at different time points prevents comparisons between studies and extrapolation of findings. There is also a selection bias towards patients who tend to present with intractable seizures and/or medical and psychiatric co-morbidity in tertiary epilepsy clinics like ours. The aim of this study, however, is to alert psychiatrists and epileptologists, particularly those who work in liaison with physicians, about the considerable risk of these patients for attempting suicide. The literature is surprisingly poor in this field, and the general perception that IGE patients are usually free from psychiatric or behavioral problems may deflect these patients from receiving necessary assessment and care. Attempted suicide is the most important risk factor for completed suicide,20 and further studies are needed to elucidate suicidal behavior in IGE patients.