Clinical manifestations

Presentation and course

Juvenile myoclonic epilepsy typically appears in the second decade. The age of onset ranges from 8 to 24 years, with peak onset between 12 and 18 years (21). It is characterized by myoclonic seizures, associated at times with generalized tonic-clonic seizures or absence seizures.

A detailed medical history with a specific focus on seizure types, age at onset, timing, and triggers is fundamental for the diagnosis (28). Comorbidities and family history are also important. The EEG can provide valuable information for the diagnosis, including specific idiopathic generalized epilepsy syndromes, and, therefore, contributes to the best course of treatment and management.

The cardinal seizure type is that of myoclonic jerks characterized by sudden, brief, bilaterally symmetrical and synchronous muscle contractions. The intermittent and involuntary shocklike contractions ("secousse") affect mainly the shoulder and upper extremities; much less commonly, the lower extremities or the entire body may also be involved. Myoclonic seizures may occur either singly or in clusters (02; 50). Objects may be thrown as a result of jerks, or, rarely, patients may fall. Consciousness remains unimpaired during myoclonic seizures, even if they occur in series or in myoclonic status epilepticus (51).

Because the jerks in juvenile myoclonic epilepsy are brief, without impairment of awareness, and take place in the morning soon after awakening, they are often interpreted as "nervousness" or "clumsiness" until generalized convulsions occur (25). Many patients ignore mild events and are unaware that they are abnormal (90). Therefore, physicians need to ask specifically about jerks early in the morning, which in association with the characteristic EEG pattern, makes it difficult to miss this diagnosis.

Generalized tonic-clonic seizures appear after the onset of myoclonic seizures in most cases (02). The average interval between the first myoclonic seizures and the beginning of tonic-clonic seizures is 3.3 years (50). A convulsive seizure may begin with a series of myoclonic jerks of increasing intensity, followed by generalized myoclonus and then by generalized tonic-clonic seizures. This characteristic picture has been described as "myoclonic grand mal," "impulsive grand mal," and "clonic-tonic-clonic seizures" (50).

Both the myoclonic seizures and generalized tonic-clonic seizures have a special circadian pattern, ie, they frequently occur on or soon after awakening, either from all-night sleep or from a nap (02). The myoclonic and generalized tonic-clonic seizures are frequently precipitated by early awakening, sleep deprivation, emotional stress, alcohol consumption, recreational drug use, or photic provocation (25; 51). Occasionally, they may also occur sporadically during the daytime, or when the patients are tired and relaxed. Photosensitivity occurs in approximately one third of the patients (02; 25; 36; 28). Seizure precipitation by complex neuropsychological tasks, especially when they require some kind of motor action ("praxis-induction"), has been shown to be associated with juvenile myoclonic epilepsy in up to half of the patients and may be associated with less favorable response to pharmacotherapy (64; 94). As peri-oral reflex myoclonias elicited by talking and reading are also frequent in this syndrome (65), the presence of some reflex epileptic traits seems to be one of its characteristics (07; 90).

Absence seizures, usually of typical variants, occur in 15% to 30% of patients and begin at a mean age of 11.5 years (02; 21), although these are infrequent and shorter than in childhood absence epilepsy or in juvenile absence epilepsy, and frequently go unnoticed by patients and family members (90).

Prognosis and complications

Juvenile myoclonic epilepsy does not remit spontaneously in most patients (48). Studies have shown that relapse rate after discontinuation of antiseizure medications is high, varying from 78% to 90% (50; 81; 84). However, some studies indicate that lifelong therapy may not be necessary for all patients (75). A population-based study showed that one third of patients did not require antiseizure medication treatment 25 years after onset of seizures, with 17% being free of all seizure types and 13% with only myoclonus (12). Another study showed that after a mean follow up of 45 years, 59% of patients were seizure-free for five years or more, and 28% of those were off medication (83). Hofler and colleagues showed that after a period of up to 36 years, 62% of patients with juvenile myoclonic epilepsy were free of all types of seizures for more than one year, 53% for more than two years with antiseizure medication, and 9% of patients were seizure-free for more than two years without medication (49). Pietrafusa and colleagues evaluated the long-term prognosis in 61 patients with juvenile myoclonic epilepsy and found that 40 (65%) had a 5-year terminal remission with a mean age at last seizure of 27.4 years (76). After withdrawal of antiseizure medications in 13 of these seizure-free patients, six had seizure recurrence, whereas seven remained seizure-free. Interestingly, the mean age at antiseizure medications withdrawal was longer in the subgroup of seven patients who remained seizure-free (21 vs. 35.7 years, p < 0.05) (76). Taken together, these long-term follow up studies indicate that the longer the patients are seizure-free on antiseizure medication, the more likely it is for them to continue seizure-free after medication withdrawal (12; 83; 49; 37; 76).

The presence of absence seizures and high frequency of generalized tonic-clonic seizures at onset were associated with poorer prognosis (83; 49; 37; 76). Patients with prolonged epileptiform discharges in ambulatory EEG and psychiatric comorbidities also had a poorer prognosis (09; 93). A metaanalysis found that psychiatric comorbidities, catamenial epilepsy, focalities on EEG, history of childhood absence epilepsy, family history of epilepsy, status epilepticus, and febrile seizures were independent predictors of pharmacoresistance in juvenile myoclonic epilepsy (84).

Clinical vignette

A 30-year-old woman was referred for consultation in the epilepsy clinic. She had a history of generalized tonic-clonic seizures since the age of 14. Seizures occurred mostly soon after awakening or after stressful moments, particularly when she was very tired or had lack of sleep. She brought a couple of normal EEGs and a normal MRI and said that her previous doctors, including a neurologist, did not know what she had because all the exams were normal. She was using carbamazepine and had taken phenytoin and phenobarbital in the past without good seizure control.

Her neurologic examination was normal, and during the interview she confirmed that she had had myoclonic jerks in the morning, but she never thought that this was relevant. Doctors never asked her about these jerks, which did not bother her much.

She was told that her diagnosis was juvenile myoclonic epilepsy. An EEG after partial sleep deprivation showed typical 4-Hz generalized spike and polyspike-and-slow waves with predominance over the frontal regions.

Because she was overweight and had plans to get pregnant, and the myoclonic jerks were not very frequent, she was prescribed lamotrigine, with a very slow titration up to 300 mg a day. She has been free of generalized tonic-clonic seizures and has had rare early morning myoclonus.

This is a typical history, and it is usually easy to make a diagnosis of juvenile myoclonic epilepsy. It is still often misdiagnosed and mistreated by many physicians, even though it is one of the most common forms of epilepsy in young adults. Early morning myoclonic jerks are an important feature and should always be asked of patients with a history of generalized tonic-clonic seizures.

Biological basis

Etiology and pathogenesis

Juvenile myoclonic epilepsy is an electroclinical epilepsy syndrome with a strong genetic component following a complex pattern of inheritance, with a polygenic basis and an environmental contribution (101; 90; 55; 06). Not a single case associated with organic brain pathology has been reported, although a history of febrile or isolated seizures in childhood can be obtained in a minority of cases (50; 51; 48).

Up to 50% of patients have first- and second-degree relatives with epileptic seizures, and 80% of symptomatic and 6% to 16% of asymptomatic siblings have diffuse 4- to 6-Hz polyspike-and-wave complexes on their EEG (63).

The concordance rate for monozygotic twins is 0.7 to 1.0, whereas that for dizygotic twins is the same as in siblings (40). Despite long and intense genetic research, the genetics of the syndrome are not fully clarified. A polygenic mode of inheritance is the more likely cause as the different reflex epileptic traits that are frequent parts of the phenotype are also genetically determined. Evidence indicates that one gene involved in juvenile myoclonic epilepsy may be located on the short arm of chromosome 6 (26), but it was not found in all investigated families (30). Another candidate locus was found on chromosome 15q (29) but, again, could not be found ubiquitously (79). A relationship appears to exist between juvenile myoclonic epilepsy and other age-related genetic generalized epilepsies such as childhood absence epilepsy, epilepsy with generalized tonic-clonic seizures alone, and early childhood myoclonic epilepsy because more than one phenotype may exist in the same family (101).

Several genes predisposing to juvenile myoclonic epilepsy have been identified over the last decades, although results are not always replicated in different studies indicating a high genetic heterogeneity (31; 45; 86; 43). Among these genes, mutations in genes encoding for subunits of the gamma-aminobutyric acid (GABA)A receptor and in the Myoclonin1/EFHC1 gene have been reported in patients with juvenile myoclonic epilepsy (20; 23; 55; 86; 60). EFHC1 is involved in the regulation of cell division and with the process of neuronal radial migration during embryogenesis (19; 23). However, the gene-disease relationship between EFHC1 and juvenile myoclonic epilepsy is still unclear (39). Thus, the inclusion of EFHC1 in gene panels for genetic testing should be limited to research purposes (19; 39).

Conventional neuroimaging evaluation is normal (48); however, studies using quantitative structural and functional neuroimaging have shown subtle, but consistent, abnormalities in patients with juvenile myoclonic epilepsies that help to better understand the pathophysiology of this disease (95; 99; 24; 32; 27; 10; 98; 72; 74).

Wandschneider and colleagues used quantitative postprocessing techniques of high-resolution structural MRIs to measure the cortical surface curvature, morphology, and the geodesic distance, a surrogate marker of cortico-cortical connectivity in patients with juvenile myoclonic epilepsy and their unaffected siblings (98). They found that patients and siblings had increased cortical curvature and surface, as well as changes in geodesic distance that were more prominent in prefrontal and cingulate cortices, which are areas undergoing late maturation. Their results indicate that these MRI biomarkers of brain development and connectivity are likely heritable and, thus, relate to the underlying disease neurobiology. However, cortical thinning in frontocentral and occipital cortices occurred only in patients with juvenile myoclonic epilepsy but not in their siblings, probably representing a marker of seizures as opposed to the epileptogenic pathophysiology (98).

There has been growing neuroimaging evidence for a thalamic-cortico-frontal network dysfunction in juvenile myoclonic epilepsies (95; 24; 32; 27). These abnormalities do not appear to be secondary to seizures and may be genetically determined (58; 99). Indeed, unaffected siblings of juvenile myoclonic epilepsy patients showed abnormal fMRI co-activation in the primary motor cortex and supplementary motor area, as well as abnormal functional connectivity between motor and prefrontal cognitive networks, similarly as in juvenile myoclonic epilepsy patients (97).

Microstructural white matter and gray matter abnormalities were present in patients with new-onset juvenile myoclonic epilepsy using diffusion tensor imaging (DTI), indicating these abnormalities exist since the beginning of the disease (27). In addition, there is evidence for a network dysfunction, with hyperconnectivity involving primary motor, parietal, and subcortical regions that correlate with cognitive task performance (11).

Studies have shown that patients with all subtypes of genetic generalized epilepsies present clear reductions in functional connectivity within the default mode network, which includes the precuneus, posterior cingulate cortex, the medial prefrontal cortex, parietal cortex, and inferior temporal cortex (74). Functional connectivity changes in the default mode network may be related to the cognitive deficits, anxiety and depression in patients with juvenile myoclonic epilepsies and other idiopathic generalized epilepsies (34).

Epidemiology

The reported incidence of juvenile myoclonic patients among all epilepsies has increased from 2.7% in 1957, to 5.4% in 1977, and to 11.9% in 1984, as awareness of this disease has increased (50; 22). Today, it is estimated that the prevalence of juvenile myoclonic patients is around 5% to 10% of all patients with epilepsy, with a clear predominance of women (13). Diagnosis may still be commonly missed, largely because patients fail to report their myoclonic jerks, and physicians fail to specifically ask about them. In one study, definitive diagnosis of a series of patients referred to an epilepsy center was delayed by a mean of 14.5 years (41). The point prevalence was estimated at 5.6 out of 10,000 among people with less than 30 years of age in Norway. Juvenile myoclonic epilepsy constituted 9.3% of all epilepsies in that age group (88). Juvenile myoclonic epilepsy is the most common subsyndrome of the genetic generalized epilepsies spectrum, representing 42.4% of all genetic generalized epilepsies in the study by Gesche and colleagues (37).

Differential diagnosis

Confusing conditions

Myoclonic seizures occur in patients with other idiopathic generalized epilepsies, such as the juvenile absence epilepsy and epilepsy with generalized tonic-clonic seizures alone, but are mild and have no particular circadian distribution and are not the most prominent features of these syndromes. In epilepsy with eyelid myoclonia syndrome, there are frequent episodes of fast (greater than 4-Hz) rhythmic jerks of the eyelids and upward deviation of the eyeballs (28).

Lennox-Gastaut syndrome, epilepsy with myoclonic-astatic seizures, and epilepsy with myoclonic absences begin in childhood rather than adolescence and are associated with more frequent seizures and cognitive impairment. In the latter two, the myoclonic seizures usually involve the face and are associated with absences, whereas consciousness is not impaired during myoclonic seizures of juvenile myoclonic epilepsy (25; 90).

Postanoxic or posthypoxic myoclonus can be acute or chronic. Acute posthypoxic myoclonus occurs within hours of the hypoxia while the patient is still unconscious, has a generalized multifocal distribution, is elicited by external stimulus or movements, and commonly demonstrates generalized EEG abnormalities, including epileptiform discharges, but acute posthypoxic myoclonus may also be of subcortical nature (nonepileptic myoclonus). Chronic posthypoxic myoclonus, or Lance-Adams syndrome, develops after the patient recovers consciousness and presents as multifocal action myoclonus (44).

Familial adult myoclonic epilepsy or adult myoclonic epilepsy with cortical tremor is associated with prominent cortical tremor and may mimic juvenile myoclonic epilepsy (42).

In progressive myoclonic epilepsies, nonepileptic multifocal myoclonus and epileptic myoclonus often coexist and are clinically different from those in juvenile myoclonic epilepsy. Progressive neurologic deterioration, dementia, action myoclonus, and ataxia are important features in progressive myoclonic epilepsies and never occur in juvenile myoclonic epilepsy (71).

Diagnostic workup

Because the jerks in juvenile myoclonic epilepsy are brief, without loss of consciousness, and take place in the morning soon after awakening, they are often interpreted as "nervousness" or "clumsiness" until generalized convulsions occur (25). Many patients ignore mild events and are unaware that they are abnormal (90).

The EEG background activity is normal during wakefulness and sleep, except in case of antiseizure medication intoxication or inadequate treatment.

Seventy-four percent of patients have epileptiform interictal EEG patterns (69). Typical interictal abnormalities consist of brief (1 to 3 seconds) 3- to 6-Hz generalized discharges of spike and polyspike-and-slow waves that are usually asymptomatic, and sometimes may appear asymmetrical or with pseudo focalities (59). Focal EEG features are relatively common and may contribute to misdiagnosis of juvenile myoclonic epilepsy as focal epilepsy; however, these focalities do not influence treatment response (82; 54). Classical 3-Hz spike-and-wave complexes or 3-Hz polyspike-and-wave complexes characteristic of typical absence seizures may occur in some teenagers.

The ictal EEG is characterized by 10- to 16-Hz medium-high amplitude spikes followed by irregular slow waves. The number of spikes ranges from 5 to 20 per episode and correlates with the intensity rather than the duration of each seizure (52).

Provocative measures, such as photic stimulation and sleep deprivation, can help elicit the characteristic EEG abnormalities. Recording during the process of awakening may be necessary for the diagnosis (50). Sleep deprivation is very effective in precipitating seizures. Praxis-induction is found by neuropsychological testing during EEG (64), and perioral reflex myoclonias, unless observed during the interview with the patient, require a video-EEG investigation that includes reading aloud and free talking (65). Polyspike-and-wave discharges are seen more frequently after nocturnal awakening than after morning awakening (50). The incidence of photosensitivity has been reported to be as high as 30.5% (50).

Neurologic examination and neuroimaging studies are normal in this syndrome. When the clinical presentation and EEGs are typical for juvenile myoclonic epilepsy, imaging is not required for clinical management. However, MRI should be considered if atypical features or pharmacoresistant seizures are present, or if there is persistent focal slowing on EEG (48).

One study showed unrelated or nonspecific MRI abnormalities in up to 24% of juvenile myoclonic epilepsy patients (08). There was a higher proportion of EEG focalities in patients with abnormal MRI, which were, in most part, concordant with the location of the MRI findings (08). As mentioned previously, these findings may predict poorer response to antiseizure medications (84).

Management

Both medical treatment and counseling are important in the management of juvenile myoclonic epilepsy. Monotherapy with sodium valproate is the most effective in controlling myoclonic, generalized tonic-clonic, and absence seizures (38). Due to its reduced serum fluctuations and single daily dosage, divalproex sodium extended-release formulation may significantly improve tremor, weight gain, and gastrointestinal complaints while maintaining similar seizure control as valproate traditional formulation (57). Low doses of valproate (500 to 1000 mg) may be sufficient for initial treatment in juvenile myoclonic epilepsy (46). However, up to 30% of patients with juvenile myoclonic epilepsy do not respond to valproate and other monotherapies (01).

However, the significantly higher risk of fetal malformations (92) and of worse cognitive outcome as well as higher frequency of autism spectrum disorder and attention-deficit or hyperactivity disorder in children born to women who used valproate limit its use in young women (17; 100). On the other hand, studies suggest that avoiding valproate might be associated with unsatisfactory seizure control in women with juvenile myoclonic epilepsy who are of childbearing potential (15; 38). In women of childbearing potential, levetiracetam and lamotrigine should represent the first-choice treatment (01; 04; 68).

Levetiracetam is a good alternative to valproate, particularly when lamotrigine exacerbates myoclonus (89). One study showed a lower retention rate for levetiracetam than valproate due to poorer seizure control during long-term follow-up, but patients with levetiracetam achieved more myoclonic seizure freedom than those with valproate (78). There is some evidence that levetiracetam may have better efficacy than lamotrigine in monotherapy for patients with juvenile myoclonic epilepsy (66). However, psychiatric adverse effects are common with levetiracetam and may lead to treatment discontinuation; it should be avoided in patients with antecedent of depression (56). An alternative to levetiracetam is brivaracetam. Brivaracetam is well tolerated and has shown a good efficacy for treating patients with juvenile myoclonic epilepsy and other forms of genetic generalized epilepsies (87; 33). The occurrence of behavioral adverse events appears to be less frequent with brivaracetam than with levetiracetam, and the switch between these two medications can be made easily (87).

Lamotrigine is a good choice for treating patients with juvenile myoclonic epilepsy, particularly in those with psychiatric comorbidities, but it may exacerbate myoclonus (62). Clonazepam is a useful add-on therapy for myoclonus and can be used in combination with lamotrigine to avoid its promyoclonic effects. The combination of lamotrigine and valproate may be an alternative for patients who fail seizure control with monotherapy. One study showed that the combination of lamotrigine and valproate was successful in 69% as compared to 9% with all other combinations in patients who failed to respond to valproate in monotherapy (03). In addition, only one of 24 (4%) patients became seizure-free after failing the valproate and lamotrigine combination (03).

Topiramate and zonisamide may be an alternative for some patients who do not tolerate valproate, particularly patients with severe weight gain due to valproate (73; 09).

Benzodiazepines may have an adjunctive effect for short periods and are sometimes good for intermittent use in specific stressful situations. Clonazepam may be beneficial for controlling myoclonic seizures but not the generalized tonic-clonic seizures. Complete elimination of myoclonic seizures could cause additional disability by depriving some patients of the warning jerks that herald the onset of generalized tonic-clonic seizures (70).

Seizures are usually precipitated by fatigue, noncompliance, stress, sleep deprivation, and alcohol consumption. Successful treatment is contingent on acceptance of appropriate limitations on lifestyle. Up to one third of patients with juvenile myoclonic epilepsy may have difficulty in treating seizures, and the presence of absence seizures, psychiatric comorbidities, earlier age at seizure onset, and praxis-induced seizures appears to indicate poorer outcome (85).

Outcomes

Juvenile myoclonic epilepsy is a life-long condition, and although a significant proportion of patients remain seizure-free on medication, 70% to 90% will have relapse seizures with the withdrawal of antiseizure medication (50; 81; 12; 83; 96; 37; 38).

Complete remission of generalized tonic-clonic seizures under antiseizure medication significantly increased the chance for complete seizure freedom, and the occurrence of EEG photoparoxysmal responses significantly increases the risk of seizure recurrence after antiseizure medication discontinuation (35). The presence of absence seizures is an independent predictor of poorer outcome (05; 83; 37).

In a meta-analysis, Stevelink and colleagues estimated that 35% of patients with juvenile myoclonic epilepsy were pharmacoresistant; however, the proportion of patients with pharmacoresistance varied greatly among the studies analyzed (85). In an epidemiologic study, Gesche and colleagues estimated that 12.1% of patients with genetic generalized epilepsies were pharmacoresistant, with an average burden of seizures of 2.2 generalized tonic-clonic seizures per year (37). Although the overall frequency of seizure freedom was high, only 18.6% of the patients achieved seizure freedom with acceptable side effects at the first treatment trial in a cohort of 328 adult patients with idiopathic generalized epilepsy (38).

In the long-term follow-up, most patients with juvenile myoclonic epilepsy have a fluctuating pattern, with variable intervals of seizure control and seizure relapse that may subside over time (34; 16).

Intelligence usually remains normal in patients with juvenile myoclonic epilepsy (50; 77). However, associated immature personality, emotional instability, inadequate social adjustment, impulsivity, depression, and personality disorders occur frequently irrespective of seizure control (99; 61). Together with higher seizure frequency, impulsive behavior correlates with worse social adjustment (67). In one study, praxis-induced seizures were more common in males, and patients with reflex traits presented higher rates of persistent myoclonia and more frequent psychiatric comorbidities (14). Patients who do not respond well to antiseizure medications are more likely to have depressive and anxiety symptoms than those free of seizures (34).

Special considerations

Pregnancy

A major concern is the high risk of valproate causing major congenital malformations (92) as well as the cognitive and behavioral outcome of children who were exposed to valproate in utero (17; 100), and it should be avoided in girls and women of childbearing potential. Levetiracetam and lamotrigine are considered the safest choices for women of childbearing potential with juvenile myoclonic epilepsy (68). A study comparing carbamazepine, phenobarbital, valproic acid, and lamotrigine confirmed particularly higher malformation rates in women taking valproic acid at doses higher than 1500 mg per day, whereas lamotrigine at a dose of less than 300 mg per day had the lowest rate of malformations (91). However, lamotrigine at doses higher than 300 mg per day was no safer than valproic acid at doses lower than 700 mg per day, demonstrating that both the type and dose of antiseizure medication are important to consider in women of childbearing potential.