JLE

Epileptic Disorders

MENU

Epilepsy and mental retardation: genetic syndromes Volume 8, supplément 1, Supplement 1, April 2006

Auteur(s) : Marie-Odile Livet1,2, Nathalie Villeneuve2,3, Géraldine Daquin3, Marie-Anne Cournelle1, Joselle Mancini2, Anne Moncla2

1CH Aix-en-Provence
2Hôpital La Timone, Marseille
3Hôpital Henri-Gastaut, Centre Saint-Paul, Marseille, France

Epilepsy and mental retardation are often associated. The specific study of clinical manifestations, and subsequently electroencephalographic characteristics of the epilepsy can make an important contribution to the etiologic diagnosis of the underlying pathology.We know that for patients with mental retardation and for their families, the search for etiology is of primary importance. It should be undertaken as early as possible, but is sometimes only accomplished by default in adulthood. This search implies a rigorous investigation that is essentially clinical.This manuscript deals with a rare but privileged situation where patients present certain genetic syndromes with mental retardation, for which epilepsy has a rather characteristic electro-clinical profile, very helpful for diagnostic purposes (Gobbi et al. 2002).Moreover, these “genetic epileptic syndromes” with a known chromosomal anomaly are of obvious interest for research dealing with the genetic causes of the epilepsies (Singh et al. 2002).After looking at the diagnostic procedure involving a patient presenting with mental retardation and epilepsy, we will describe major genetic syndromes with specific epilepsy.

Epilepsy with mental retardation: how to make the diagnosis?

As is the case for all mental retardation cases, the diagnostic procedure is primarily and essentially clinical, based on anamnesis and clinical exam. It confirms mental deficiency and orients etiologic diagnosis according to its chronology: perinatal, postnatal or antenatal (Des Portes et al. 2002). In this context, recognition by the physician of certain characteristic phenotypes, whether it be dysmorphisms, behaviors or a particular type of epilepsy, can sometimes suggest a diagnosis early in the course of the disorder. Identifying etiology is crucial.

In cases of developmental delay in a child, it makes it possible to give parents better answers to their questions: “Why? Where does his problem come from? What will be his evolution? What can be done to help his development? Is there a risk if we have another child?” This request for genetic counseling, coming from parents or from other family members, is more and more often what motivates etiologic research.

The stages of this search for etiology are well known (Des Portes et al. 2002). First, questioning the parents is essential and requires careful listening to their answers, as well as consideration for what they need to know. Family history is compiled using a family tree extending over three generations; the child’s personal and developmental history is taken.

Complete clinical examination includes growth curves, cranial perimeter, search for dysmorphic features, neurological examination proper as well as, language, expression and comprehension tests, assessment of ability to communicate and evaluation of behavior. Certain specific behavior traits are characteristic and can give an immediate indication of the diagnosis for some genetic syndromes with “behavioural phenotype” (Cassidy and Morris 2002).

When epilepsy is associated with mental retardation, history taking is, once again, essential, but too often neglected. The history should include age at onset of seizures, their exact description, time of day of their occurrence, their duration and frequency, and possible triggering factors. Waking and sleep EEGs, often completed by video EEG recordings, identify ictal and interictal anomalies.

These clinical data make it possible to determine the need for more specific complementary investigations: brain imaging, genetic or metabolic tests.

Syndrome-etiology approach

For several genetic syndromes caused either by chromosomal structure anomalies, most often deletions, or by genetic mutations, diagnosis can be made using specific genetic tests that are fast and reliable: search for a deletion using FISH analysis, or molecular biology search for mutations. But the clinical examination remains an essential prerequisite to these specific investigations. The clinical picture is comprised of the child’s developmental profile, possible distinctive behavioral traits, observation of characteristic dysmorphisms not obvious in a young child but evolving with age, and sometimes an electroclinical profile specific to epilepsy.

Close cooperation between neuropediatricians and geneticists is imperative in these cases.

Angelman syndrome

Angelman syndrome is the most frequent and best known genetic syndrome with specific epilepsy. Pampiglione and Martinez (1983) and Boyd et al. (1988) described the characteristics of the EEG and stressed its importance for early diagnosis.

Consensus diagnostic criteria for the syndrome have been developed (Williams et al. 1995). Constant criteria are: severe mental deficiency, absent or minimal language, ataxia with saccadic walking, very characteristic behaviour with hyperexcitability and easy laughter. Signs present in 80% of cases are: moderate postnatal microcephaly, epilepsy and, abnormal, highly suggestive EEG findings, facial dysmorphism evolving with age (( figure 1 )). Sleep problems are very common (Clayton-Smith and Laan 2003). The particularity of the syndrome resides in the fact that its genetic origin involves various mechanisms, but has as a constant the absence of contribution from genes in the 15q11-q12 region of maternal origin; microdeletion of 15q11-q12 on maternal chromosome 15 in 70% of cases; as well as paternal isodisomia of chromosome 15 in 5% of cases; mutations in the central region of the imprint, responsible for an isolated methylation anomaly in 5% of cases; or mutation of the UBE-3A gene in 15% of cases.

Cases with no identified abnormality are more and more rare (Clayton-Smith and Laan 2003).

This syndrome, and Prader-Willi syndrome (15q11-q12 microdeletion on paternal chromosome 15) were the first models of differential parental imprint: certain genes in the 15q11-q12 region are exclusively expressed by maternal chromosome 15, others by paternal chromosome 15. Methylation bears witness to the imprint, and methylated genes are inactive. The UBE-3A gene receives a specific cerebral imprint, and is expressed in the cerebellum and the hippocampus exclusively from the maternal allele (Albrecht et al. 1997).

Epileptic seizures (Viani et al. 1995) are present in 90% of cases and generally start before the age of three years.

Infantile spasms have been reported, but they remain the exception. In most cases, the child presents with atypical absence, or atonic or myoclonic seizures. Generalized tonic-clonic seizures and hemiclonic seizures, febrile and non-febrile, can occur. Myoclonic status has been observed.

Non specific but very characteristic electroencephalographic abnormalities are present in over 90% of cases, even before seizures. They can suggest an Angelman syndrome diagnosis when an EEG has been requested for an infant with psychomotor delay. The EEG shows diffuse and continuous slow and sharp waves at a frequency of 4 to 6 Hz. Other patterns are also suggestive of Angelman: Complexes of slow 3 to 4 Hz waves and spike waves, with occipital predominance, occurring upon closing of the eyes, or complexes of very high amplitude 2 to 3 Hz slow waves and spike waves, with bifrontal predominance (( figure 2 )).

A particular feature of Angelman syndrome is intermittent “tremor”; these are rapid rhythmic myoclonic jerks unrelated to EEG abnormalities. Retrograde averaging techniques (Guerrini et al. 1996) have documented the cortical origin of this type of myoclonus. It should also be kept in mind that a continuum may exist between the myoclonic jerks and a myoclonic status.

When this occurs, the child no longer shows his usual hyperactivity, he becomes too quiet, unresponsive and “absent”, with continuous erratic myoclonus.

Angelman syndrome epilepsy has a particular pharmacologic profile. Valproate, benzodiazepines, ethosuccimide and lamotrigine are effective treatments, while vigabatrin and carbamazepine usually aggravate the condition. The effectiveness of levetiracetam seems variable, but controlled data is lacking.

Evolution is most often characterized by a decrease in seizures at adolescence. Slow wave activity at 4 to 6 Hz wave activity disappears after the age of 12 years, but intermittent bifrontal slow waves can persist. In some cases, seizures continue in adulthood (Laan et al. 1996).

It has now been clearly demonstrated that the Angelman phenotype, and particularly epilepsy, have variable degrees of severity, based on genetic type: cases with deletion are the most serious, followed by UBE3A mutations, and lastly by imprint isodisomias and mutations (Minassian et al. 1998, Moncla et al. 1999). Risk of recurrence also depends on genetic type: it is very low for deletions or de novo isodisomias, while imprint and UBE3A mutations depend on dominant imprinted maternal transmission (Moncla et al. 1999).

The gene or genes responsible for epilepsy and for myoclonias are not yet known. The UBE3A gene is a possibility; a murine model has been studied (Miura et al. 2002). Interactions between UBE3A and the cluster of encoding genes for β3, α5 and γ3 subunits of the GABA A receptor, located in the deleted region, cannot be excluded.

Chromosome 15 inversion-duplication syndrome

The presence of a supernumerary chromosome 15 produces a tetrasomy or quadruple dose effect in the 15q11-q13 region, which is also involved in both Angelman and Prader-Willi syndromes; the additional fragment is of maternal origin. This relatively common syndrome is usually associated with severe mental retardation with microcephaly and autistic features. Dysmorphism is absent or uncharacteristic, limited to rough features. But epilepsy is constant and severe. It starts early, often in the first months of life. Spasms, atypical absences, tonic and atonic seizures have been described. The EEG shows atypical hypsarrhythmia with large amplitude diffuse slow spike waves and/or multifocal abnormalities (figure 3 and 4) (Buoni et al. 2000, Takeda et al. 2000, Borgatti et al. 2001).

However, less severe cases have been reported, with epilepsy starting in adulthood and characterized by atypical absences and moderate mental retardation (Chifari et al. 2002).

The 15q11-q13 region is also involved in 15q11-q13 duplications. These interstitial duplications are related to chromosome 15 of maternal origin. They have been reported in children with autistic symptoms, with variable mental retardation and without dysmorphisms. Epileptic seizures are rare but the EEG can show centrotemporal spike waves activated during sleep (Bundey et al. 1994).

FISH analysis of the 15q11-q12 region is recommended in genetic research related to autistic patients (Keller et al. 2003).

Wolf-Hirschorn syndrome

This syndrome is caused by chromosome 4p deletion, and is much more rare compared to the two syndromes previously described (Sgro et al. 1995, Battaglia et al. 1999). It is characterized by severe mental deficiency, in utero and postnatal developmental delay, microcephaly, dysmorphism with “Greek warrior helmet” nose, hypertelorism, high frontal hairline, large ears and, sometimes, cleft lip or palate.

Organ malformations, particularly cardiac, are common. One third of patients die in the first year. MRI can show gyration abnormalities, corpus callosum hypoplasia, and dentate nucleus dysplasia. Epilepsy is constant from the first year, with partial motor seizures, myoclonus or atypical absences easily triggered by eyes closure.

The EEG shows generalized large amplitude sharp wave discharges at closing of the eyes, and multifocal irritative signs. Valproate is effective against seizures, while carbamazepine aggravates them. Thus, the epilepsy has electroclinical characteristics close to those seen in Angelman syndrome. Genes encoding GABA A receptor subunits are located in 4p12-13, near the critical 4p16-3 region of the syndrome. The role of a GABA A receptor functional deficit in the pathogenesis of the epilepsy should be considered.

Ring chromosome 20 syndrome: an example of mosaicism

This anomaly is characterized mainly by a very particular epilepsy syndrome associated to behavioral problems; mental deficiency is light to moderate, and dysmorphic features are not present (Roubertie et al. 2000, Latour et al. 2002, Petit et al. 1999, Biraben et al. 2001, Augustjin et al. 2001).

The epilepsy starts in childhood and usually is very drug resistant. Critical manifestations (Latour et al. 2002, Petit et al. 1999, Biraben et al. 2001) are: absences with motor automatisms, gait ataxia, intense fear, sometimes hallucinations lasting a few seconds or long periods. Prolonged confused states lasting 30 minutes to 1 hour or longer, with slow, laborious speech, perioral or palpebral clonic spasms, and complex automatisms have been observed (( figures 5A ), ( B ), ( C )). The particularity of these manifestations is that they are often triggered by emotions, frustration, conflict… so that they can be mistaken for non epileptic seizures. Tonic or tonic-clonic seizures occur more rarely. Seizures occurring at night have sometimes been observed (Augustjin et al. 2001).

Ictal EEG shows long sequences of slow, rhythmic bifrontal waves or slow spike sharp waves with bifrontal predominance. Intercritical tracings can be normal or can show slow wave complexes associated with spikes predominant in the frontal regions (( figure 6 )). These abnormalities are activated upon falling asleep and during slow, light sleep, and disappear in the other stages of sleep (Latour et al. 2002).

After a period of normal psychomotor development, mental retardation, usually moderate, appears after the age five or six years, especially with personality problems: impulsiveness, contrariness, aggressivity, making social integration difficult.

Cranial MRI is normal. It is the unusual nature of this epilepsy, with its prolonged, repeated seizures lasting over 30 minutes, and its EEG features, that should motivate a karyotype search for ring mosaic chromosome 20 syndrome, present in a variable, usually small number of mitoses (( figure 5D )). When this percentage is predominant, mental retardation is more likely to be associated (Biraben et al. 2001).

We know that two epileptic syndrome genes are located on the distal portion of the long arm of chromosome 20: the gene for autosomal dominant nocturnal frontal epilepsy, and the gene for benign familial neonatal convulsions. But the origin of ring chromosome 20 syndrome is still unknown.

Fragile X syndrome

Fragile X syndrome, considered the second cause of genetically determined mental retardation, is included in our presentation due to its frequency and the particular characteristics of its epilepsy, which can facilitate the diagnosis (Musumeci et al. 1988, 1991, and 1999). This diagnosis is a possibility when the patient is a boy with mental and language retardation, with behavioral problems, agitation, instability and, in many cases, macrocephaly. Dysmorphism is also present: long face with high forehead, large ears and hyperlaxity. In this context, family history should be examined for previous mental retardation compatible with X-related transmission that could also have affected girls to a lesser extent. Diagnosis is confirmed by a molecular biology search for gene FMR-1 mutations.

Epilepsy (Musumeci et al. 1988, 1991, and 1999) is present in 20 to 25% of cases. It starts between the ages of two and 12 years, with partial, simple and complex seizures, or with generalized seizures. The EEG shows high-voltage biphasic or triphasic spikes, central or centrotemporal, but sometimes multifocal. These abnormalities are very clearly activated during sleep. This EEG pattern is very similar to that seen in benign partial epilepsy with centrotemporal spikes.

Evolution of the epilepsy shows the same similarity: rare and mild seizures disappearing in adulthood, along with the EEG abnormalities (Musumeci et al. 1991). Relational studies have excluded linkage between the FMR-1 gene and benign partial epilepsy with centrotemporal spikes (( figure 7 )) (Musumeci et al. 1999).

Many other chromosomal disorders include epilepsy as an important component.

We can mention, for example, ring chromosome 14 syndrome, where epilepsy is severe and constant, with very severe mental retardation, microcephaly and dysmorphisms: 12p trisomy, chromosome 1, 1p and 1q terminal deletions (Singh et al. 2002)…

It is essential to identify the chromosomal regions whose abnormalities are significantly associated with epilepsy, and which can be targets for the epilepsy gene search.

But it is also essential to rigorously characterize, both clinically and by EEG, the epilepsy associated with each rare chromosomal syndrome. This is indispensable for identifying specific epileptic syndromes.