Adult-Onset Idiopathic Generalized Epilepsy: Clinical and Behavioral Features


Address correspondence and reprint requests to Dr. O. Devinsky at Neurology Department, NYU-Mt Sinai Comprehensive Epilepsy Center, 560 1st Avenue, Rivergate Building, 4th Floor, New York, NY 10016, U.S.A. E-mail:


Summary:  Purpose: To identify and define clinical and behavioral features of patients with adult-onset idiopathic generalized epilepsy (IGE).

Methods: We reviewed the charts of 313 IGE patients at the NYU Comprehensive Epilepsy Center over the past 5 years to identify patients with adult onset (18 years old or older). We excluded patients with childhood or adolescent symptoms that suggested absence, myoclonic, or tonic–clonic seizures, as well as those with a history of significant head injury or other known causes of localization-related epilepsy.

Results: Forty-two (13.4%) patients had a clear onset of IGE in adulthood; average age of onset was early 20s (mean, 23.8 years; range, 18–55 years). Twenty-one patients had adult myoclonic epilepsy (AME, 50%), and three had generalized tonic–clonic seizures on awakening (GTCS-A, 7%). More than two thirds (n = 30) are well controlled with current antiepileptic drugs (AEDs), and almost 90% are currently employed (n = 37). One third were diagnosed and treated for mental disorders, including depression (n = 12), anxiety (n = 7), obsessive–compulsive personality disorder (n = 2), and postictal psychosis (n = 1).

Conclusions: Adult-onset IGE is associated with a good prognosis. An association may exist between psychological disorders, psychotropic medication, and level of seizure control in adults with IGE.

Idiopathic generalized epilepsy (IGE) typically appears within the first two decades of life. Types of seizures, distinct EEG patterns, age at onset, and circadian rhythmicity of seizures are used to classify specific syndromes. These patients usually are considered to have an excellent prognosis. Most are well controlled on one antiepileptic drug (AED), cognitive function is normal, and the frequency of depression and anxiety is lower than that in patients with localization-related epilepsy (1,2). However, some long-term follow-up studies suggest higher rates of persistent seizures despite AED therapy, failure to outgrow epilepsy, and an increased frequency of problems in academic, social, and behavioral functioning (3,4).

The diagnosis of IGE becomes more problematic when symptoms appear in adulthood. Some investigators question whether IGE syndromes with adult onset exist (5). The definitive appearance of generalized seizures during adolescence or adulthood may lead momentary lapses of attention in childhood, considered normal daydreaming spells, to be reclassified as absence seizures. Apparent adult onset of IGE may be a localization-related epilepsy related to traumatic brain injuries sustained in the recent past or may reflect a failure to recognize IGE symptoms present since childhood or adolescence. However, evidence is growing in support of adult-onset IGE. Gilliam et al. (6) recently described adult myoclonic epilepsy as a distinct syndrome that occurs with no previous childhood or adolescent seizures.

We studied a consecutive group of patients with onset of IGE in adulthood to define further the prognosis, clinical features, and behavioral features of this population. We sought information on provocative factors for first seizures within the adult IGE population and compared these factors with those of other patients with new-onset seizures (7–9). We examined rate of psychological complaints, level of cognitive performance, use of psychotropic medications, and seizure control to determine whether a relation exists.


We reviewed the outpatient charts of 646 epilepsy patients evaluated at the NYU Comprehensive Epilepsy Center between 1995 and 2000. Patients came from a variety of sources, including patients being seen for new onset of epilepsy and referrals with medically refractory epilepsy. Patients whose clinical history supported a localization-related epilepsy, who had never had an abnormal EEG, whose EEG showed focal slowing or focal epileptiform activity, who were developmentally delayed, who had a potentially epileptogenic brain lesion, or who had resective brain surgery were excluded from this study, leaving 313 cases. Patients younger than 18 years at the time of their first diagnosed seizure were excluded, leaving 42 cases. Each EEG report (video, routine, and ambulatory) was reviewed, and the presence of interictal generalized spike and slow-wave activity and its frequency was recorded. Patients whose generalized spike and slow-wave discharges were never ≥3 Hz were excluded.

We reviewed the history of seizures and clinical diagnosis of each patient. The diagnosis of adult myoclonic epilepsy (AME) was made if the patient experienced myoclonic jerks (not limited to sleep or transition into sleep) and at least one other seizure type. Generalized tonic–clonic seizures on awakening (GTCS-A) was diagnosed if the majority of seizures occurred within 90 min of waking. The gender, age at onset, and family history of each patient was recorded. Time of day and type of initial seizure were recorded. Loss of consciousness associated with head trauma was recorded, and those patients who had a loss of consciousness 30 s within 4 years of seizure onset were excluded.

We categorized current seizure activity as grade I (full control), grade II (fair control), or grade III (poor control). These were arbitrary categories assigned to estimate the potential impact of a patient's seizure activity on daily living. “Current” was defined as the past 12 months. Grade I was defined as experiencing no seizures or only infrequent (fewer than one per month) absence or myoclonic seizures. Grade II was defined as a moderate decrease (>75%) in the number of seizures from prior baseline, with no more than one TCS per year and no more than one absence or myoclonic seizure per day. Grade III was used when patients had ≥75% reduction in seizure frequency compared with baseline, with more than one TCS per year or an average of more than one myoclonic or absence seizure per day. Current and past AED treatment history was recorded.

Precipitating factors before the initial seizure, such as sleep deprivation, increased stress, and consumption of two or more alcoholic beverages or an illegal substance were recorded. A final category of “other” was listed, which consisted of atypical factors that might have contributed to the initial seizure (such as an overdose of over-the-counter medicine).

Patient charts also were reviewed for current and past psychiatric history. Psychiatric diagnosis, individual therapy sessions attended, and psychotropic treatment of patient complaints (past and present) were recorded. Psychiatric diagnoses based on uncertain criteria or originating from an unclear or nonpsychological source were excluded. Psychotropic medication was categorized as effective, somewhat effective, or ineffective in treating psychic complaints as per patient report and physician interview of patient and family members; in many cases these findings were supplemented with standardized inventories.

Marital status, level of education, number of children, and current employment status were noted. Patients currently attending college or presently caring for children were considered employed. Neuropsychological test results also were examined, where available.


Among 313 patients with IGE, 42 (13.4%) patients had onset in adulthood. Table 1 summarizes the demographics of these patients. The average age of the patient at seizure onset was 23.8 years (range, 18–55 years); median age at onset was 20 years. Twenty-three patients were female (55%) and nineteen were male (45%). Generalized spike and slow-wave activity ≥3 Hz was seen in all patients. Twenty-one (50%) patients have AME, three (7%) have GTCS-A, and the remaining eighteen (43%) have an IGE that does not fit either category. Eleven (26%) patients have a family history of epilepsy; in six of these 11 patients, the family history was consistent with a generalized epilepsy. Ambiguous periods of blanking out or “daydreaming” were reported in three of our 42 patients; however, none was typical of absence seizures, as patients were fully responsive during these spells.

Table 1.  Characteristics of adult IGE population and selected subpopulations
Current use of
  1. Patients may be included in multiple subpopulations.

  2. IGE, idiopathic generalized epilepsy; AED, antiepileptic drug.

Number of patients42102110
Percentage of IGE adult-onset population100%23.8%50.0%23.8%
Average age at onset23.824.823.724.5
Seizure control    
 Grade I306135
 Grade II6252
 Grade III6233
Clinical presentation    
 Tonic–clonic seizures4110209
 Myoclonic seizures216216
 Absence seizures16575
Drug therapy regimen    
 No AEDs1
 One AED (monotherapy)266143
 Two AEDs10245
 Three AEDs5232
Precipitants of first seizure    
 No factor16454
 Sleep deprivation6153
 Sleep deprivation and alcohol intake512
 Sleep deprivation and increased stress322
 Increased stress4232
 Alcohol intake/illegal drug use32

Twenty-two (52%) patients are single, 17 (40%) patients are currently married, and one (2%) patient is widowed. Divorce occurred in three (7%) patients, one of whom remarried. Thirty-seven (88%) patients are currently employed, including six (14%) college students and two (5%) homemakers. Four (10%) patients are not employed, and one (2%) patient is retired. Twenty-eight (67%) patients have attended at least one semester of college; five (12%) of these have a bachelor's degree, and three (7%) have a master's or doctoral degree. For 11 patients, educational status is not known.

Nine (21%) patients have had a neuropsychological evaluation, including a standard measure of cognitive function. Of those patients, seven have an average full-scale IQ score (between 90 and 110), one has a low-average full-scale IQ score (between 80 and 90), and one has a borderline full-scale IQ score (between 70 and 80).

Twenty (48%) patients are well controlled (grade I) with no AEDs (n = 1) or monotherapy (n = 19); 10 additional patients (n = 30 total, 71%) are well controlled. Valproic acid (VPA) is the most frequently prescribed medication; more than two thirds of monotherapy patients and 26 patients in total are using this AED. Further information regarding AEDs used for monotherapy is summarized in Table 2.

Table 2.  AEDs prescribed to patients with adult-onset IGE
  1. AED, antiepileptic drug; IGE, idiopathic generalized epilepsy.

Medications for monotherapy 
 Valproic acid (Depakote)18
 Topirimate (Topamax)3
 Phenytoin (Dilantin)2
 Lamotrigine (Lamictal)1
 Acetazolamide (Diamox)1
 Clonazepam (Klonopin)1

Factors that may have precipitated the first seizure include sleep deprivation (n = 14; 33%), patient report of increased stress (n = 8; 19%), and consumption of two or more alcoholic beverages within the 24 h preceding the seizure (n = 8, 19%). Frequently, combinations of these factors led to seizure onset. Sixteen (38%) patients reported no precipitating factors. The first seizure was a GTC in 79% of patients (n = 33), myoclonic jerks or myoclonic jerks followed by a GTC in 12% of patients (n = 5), and was unclear or unknown in 10% of the patients (n = 4). More than one fourth of patients (n = 11) experienced their first seizure within 90 min of awakening. Patients diagnosed with AME were no more sensitive to sleep deprivation than were those who had other diagnoses (p = 0.54).

Diagnosis and treatment of mental disorders occurred in more than one third of the population. Five (12%) patients have been to a therapist at least twice for individual counseling sessions. Fifteen (36%) patients have taken psychotropic medications at some point. Ten (24%) patients have taken medication to treat depression or depressive symptoms, seven (17%) to treat anxiety, one (2%) to control obsessive–compulsive symptoms, and one (2%) for psychotic symptoms.

Ten (24%) patients are currently taking psychotropic medication. Six take one psychotropic, and the remaining four take two. Half use selective serotonin reuptake inhibitors (SSRIs) to treat psychological complaints. More than two thirds (n = 7) reported current psychotropic treatment to be effective, and one third (n = 3) reported current psychotropic treatment to be ineffective.

Among the 10 patients taking psychotropics, five have grade I seizure control, two have grade II, and three have grade III. Patients taking psychotropic medication were no more likely to have grade II or III seizure control than was the population as a whole (p = 0.26). However, patients taking two psychotropics seem to have less seizure control (three of four or 75% have grade II or III control) than do patients taking a single psychotropic (two of six or 33.3% have grade II or III control). Additionally, those patients who are not treated effectively with psychotropics seem to have poor seizure control as well (two of the three reporting ineffective psychotropic treatment had grade II or III seizure control).


Adult onset occurred in 13% of our patients with IGE; this supports the existence of a clinically distinct syndrome, which is consistent with reports of other investigators (6,10). These patients have a good prognosis, with most experiencing good to excellent seizure control with a single AED and being able to support themselves. Psychiatric disorders such as depression and anxiety are epidemiologically similar to the general epilepsy population; symptoms typically are well treated by a single psychotropic medication.

Clinical, behavioral, and EEG findings were similar among patients with a family history of seizures, those with AME, and the adult-onset IGE population as a whole. Demographic and social features, use and response to AEDs (∼65% in all three groups are taking no or one AED; 65% have grade I control) and age at seizure onset (low 20s) were comparable. Precipitating factors for the first seizure were reported by 61.9% of our group, with sleep deprivation being the most common factor. This is consistent with current theories regarding the sensitivity of generalized epilepsies to changes in sleep patterns (8). Contrary to recent studies, patients with AME were no more sensitive to sleep deprivation as a contributing factor for their first seizure than were others (p = 0.54)

Depression and anxiety were the most common psychological disorders diagnosed, consistent with other epilepsy syndromes. Because most adult IGE patients were seizure free and self-supported, depression and anxiety may be primarily related to primary brain dysfunction or medication effects rather than ongoing seizures or disability. According to several community based studies, between 10 and 22% of epilepsy patients experience some level of depression, and between 10 and 25% experience significant anxiety (11–13). These ranges are lower than figures suggested for hospital-based samples, which typically include more severe, intractable epilepsy cases. Most of our patients were treated successfully for their psychic and seizure needs and thus are more similar to a community sample. The prevalence of adult-onset IGE patients treated for depression (23.8%) and anxiety (16.6%) in our study is similar to the ranges of the community-based studies.

Patients taking psychotropics had seizure control and combination AED therapy rates similar to the adult IGE population, suggesting that psychotropic medication can be effectively used to treat psychological symptoms without fear of affecting seizure control. Our study was consistent with several other studies stating that psychotropic medications do not directly adversely affect seizure control (14,15). The trend toward poor seizure control with the use of multiple psychotropics raises an alternate explanation. All of our patients currently taking multiple psychotropics also were taking multiple AEDs. Patients taking multiple AEDs may have more difficult-to-control epilepsy than those effectively treated with monotherapy; poor seizure control could then lead to increased cognitive and behavioral dysfunction. The use of multiple AEDs is linked with self-reported decreased life satisfaction and poorer health (16). Disruption of sleep, circadian rhythm, and wakefulness, which can result from AEDs or frequent seizures, also could influence quality of life and psychological complaints (17).

Further research is needed to confirm our observation on the relationship of seizure frequency and psychic complaints to use of multiple psychotropics and AEDs in patients with IGE. Our sample size did not allow the use of more sensitive statistical analysis. If supported, the association between multiple AEDs, poor seizure control, and increased behavioral and cognitive disturbances could be attributed to a more refractory condition, psychological stress, drug interactions, or changes in seizure threshold induced by psychotropic agents.