- Email: [email protected]
When should clinicians search for GLUT1 deficiency syndrome in childhood generalized epilepsies?
e u r o p e a n j o u r n a l o f p a e d i a t r i c n e u r o l o g y x x x ( 2 0 1 4 ) 1 e6
Official Journal of the European Paediatric Neurology Society
Original article
When should clinicians search for GLUT1 deficiency syndrome in childhood generalized epilepsies? Sebastien Lebon a,*, Philippe Suarez b, Semsa Alija b, Christian M. Korff c, Jo€el Fluss c, Danielle Mercati d, Alexandre N. Datta e, Claudia Poloni a, Jean-Pierre Marcoz f, Ana Belinda Campos-Xavier b, Luisa Bonafe b, Eliane Roulet-Perez a a
Pediatric Neurology and Neurorehabilitation Unit, Department of Pediatrics, Centre Hospitalier Universitaire Vaudois, Lausanne, Switzerland b Centre for Molecular Diseases, Lausanne University Hospital, Lausanne, Switzerland c Child Neurology, University Hospitals, Geneva, Switzerland d Children's Hospital Neuchatel, Neuchatel, Switzerland e Pediatric Neurology and Development Unit, University Children's Hospital, Basel, Switzerland f Children's Unit, H^ opital du Valais, Sion, Switzerland
article info
abstract
Article history:
GLUT1 deficiency (GLUT1D) has recently been identified as an important cause of gener-
Received 23 June 2014
alized epilepsies in childhood. As it is a treatable condition, it is crucial to determine which
Received in revised form
patients should be investigated.
12 November 2014
Methods: We analyzed SLC2A1 for mutations in a group of 93 unrelated children with
Accepted 24 November 2014
generalized epilepsies. Fasting lumbar puncture was performed following the identification of a mutation. We compared our results with a systematic review of 7 publications of series
Keywords:
of patients with generalized epilepsies screened for SLC2A1 mutations.
GLUT1 deficiency
Results: We found 2/93 (2.1%) patients with a SLC2A1 mutation. One, carrying a novel de novo
SLC2A1
deletion had epilepsy with myoclonic-atonic seizures (MAE), mild slowing of head growth,
Generalized epilepsy
choreiform movements and developmental delay. The other, with a paternally inherited missense mutation, had childhood absence epilepsy with atypical EEG features and paroxysmal exercise-induced dyskinesia (PED) initially misdiagnosed as myoclonic seizures. Out of a total of 1110 screened patients with generalized epilepsies from 7 studies, 2.4% (29/1110) had GLUT1D. This rate was higher (5.6%) among 303 patients with early onset absence epilepsy (EOAE) from 4 studies. About 50% of GLUT1D patients had abnormal movements and 41% a family history of seizures, abnormal movements or both. Conclusion: GLUT1D is most likely to be found in MAE and in EOAE. The probability of finding GLUT1D in the classical idiopathic generalized epilepsies is very low. Pointers to
* Corresponding author. Pediatric Neurology and Neurorehabilitation Unit, Department of Pediatrics, Centre Hospitalier Universitaire Vaudois (CHUV), 1011 Lausanne, Switzerland. Tel.: þ41 213143563; fax: þ41 213143572. E-mail address: [email protected] (S. Lebon). http://dx.doi.org/10.1016/j.ejpn.2014.11.009 1090-3798/© 2014 European Paediatric Neurology Society. Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Lebon S, et al., When should clinicians search for GLUT1 deficiency syndrome in childhood generalized epilepsies?, European Journal of Paediatric Neurology (2014), http://dx.doi.org/10.1016/j.ejpn.2014.11.009
2
e u r o p e a n j o u r n a l o f p a e d i a t r i c n e u r o l o g y x x x ( 2 0 1 4 ) 1 e6
GLUT1D include an increase in seizures before meals, cognitive impairment, or PED which can easily be overlooked. © 2014 European Paediatric Neurology Society. Published by Elsevier Ltd. All rights reserved.
1.
Introduction
Glucose transporter type 1 deficiency (GLUT1D) leads to insufficient glucose for brain metabolism and has long been associated with severe seizures, microcephaly, motor disorders and developmental delay.1 Recently GLUT1D has been identified as a cause of generalized epilepsies such as socalled “idiopathic” generalized epilepsies (IGE), early-onset absence epilepsy (EOAE) and epilepsy with myoclonic-atonic seizures (MAE).2e5 Diagnosis is based on the presence of low glucose in the cerebrospinal fluid (CSF),6,7 and a mutation in SLC2A1 (solute carrier family 2 member 1) is found in the majority of cases.1 The ketogenic diet (KD) or its modified forms can usually control seizures and abnormal movements and improve cognitive function.8e10 When seizures start in infancy with an encephalopathic phenotype,11 they will readily prompt a complete diagnostic work-up, but when they present later as a generalized epilepsy, the challenge for clinicians caring for children and adolescents is to decide which patients should be investigated. We screened SLC2A1 in a cohort of children and adolescents with either new onset or established generalized epilepsies according to clinical and electro-encephalographic (EEG) features. Our findings were compared to a systematic review of 7 cohorts of patients with generalized epilepsies screened for SLC2A1 mutations from the literature.
2.
Materials and methods
2.1.
Patients
Patient data and blood samples were collected at the Lausanne University Hospital and other Swiss pediatric neurology ^ tel, Sion, Luzern and outpatient clinics (Geneva, Basel, Neucha Zurich) between September 2011 and May 2013. Inclusion criteria were: children and adolescents aged 18 years, a diagnosis of generalized epilepsy on clinical and EEG criteria (interictal or ictal generalized (poly-) spike and (poly-) spikewaves). Exclusion criteria were focal epilepsies, symptomatic epilepsies or the family refusal to participate in the study. The protocol was approved by the ethics committees of all participating centers. Informed consent from the parents or legal guardian was obtained for all patients. Probands were phenotyped including seizure, neurological and developmental histories and examination by a pediatric neurologist. Data were reported on a standardized form including the following: family history of epilepsy or abnormal movements, developmental history, learning disabilities, abnormal movements, seizure manifestations, EEG findings
and whether a diagnosis of GLUT1D was suspected before enrolment.
2.2.
Electroclinical definitions
Probands were categorized, whenever possible, into the classical idiopathic generalized epilepsy syndromes of childhood absence epilepsy (CAE), juvenile absence epilepsy (JAE), juvenile myoclonic epilepsy (JME) and epilepsy with generalized tonic-clonic seizure alone (EGTCA) according to the International League Against Epilepsy (ILAE) classification.12,13 For instance, a patient was classified in the CAE group only when he/she fulfilled the following: at least daily typical absences accompanied by bilateral, regular, symmetric, generalized 3e4 Hz spike-waves with normal background on EEG, normal development and neurologic examination.14 A child was classified into the early onset absence epilepsy (EOAE) group only when absence seizures started before four years of age but otherwise fulfilled the criteria of CAE. The generalized epilepsy syndrome of epilepsy with myoclonic-atonic seizures (MAE) is characterized by afebrile seizures starting between one and five years of age with multiple seizures types including myoclonic-atonic, atonic, or myoclonic seizures with or without generalized tonic-clonic seizures, and absence seizures. One-third has earlier febrile seizures. The EEG shows generalized spike- or polyspike and wave discharges and often background slowing.15 Children may have abnormal development prior to seizure onset or present subsequent cognitive impairment. Patients not fitting into MAE or classical IGE, such as those with absence seizures with eyelid myoclonia (Jeavons syndrome) or a combination of absence, tonic-clonic, tonic or atonic seizures, were denoted as unclassified IGE when development and neurological examination were normal. Patients with abnormal neurological findings and/or abnormal development before seizure onset were classified into generalized epilepsy of unknown origin. A fasting lumbar puncture (LP) was performed in patients in whom a SLC2A1 mutation was found.
2.3.
Systematic review
PubMed was searched by using the terms SLC2A1-gene and epilepsy. Relevant references mentioned in the articles were also included. Patients who were described in the literature more than once were included only once. We focused on publications reporting series of epileptic patients with generalized seizures screened for SLC2A1 (Table 1) and looked for features that may point to GLUT1D: additional seizures type, movement disorders, family history of epilepsy,
Please cite this article in press as: Lebon S, et al., When should clinicians search for GLUT1 deficiency syndrome in childhood generalized epilepsies?, European Journal of Paediatric Neurology (2014), http://dx.doi.org/10.1016/j.ejpn.2014.11.009
PED (1), Ataxia (1) NA 7 (12.7) 4 (2.1) 55 188
GTCS (2) NA
1 (1) 95 probands
EOAE
MAE
IGE
EOAE/CAE þ JAE familial IGE
EOAE EOAE
Suls et al.3
Mullen et al.5
Arsov et al.2
Muhle et al.19
Striano et al.17
Arsov et al.4 Agostinelli et al.18
GTCS: generalized tonic clonic seizure, IGE: idiopathic generalized epilepsy, EOAE: early onset absence epilepsy, CAE: childhood absence epilepsy, MAE: myoclono-atonic epilepsy, JAE: juvenile absence epilepsy, PED: paroxysmal exercise-induced dyskinesia, ID: intellectual disability, DD: developmental delay, NA: not available, HC: head circumference.
ID (2), DD (1) NA
no
Epilepsy þ borderline intellect (1) Epilepsy (1), PED (1) NA no
2 (7.7)/0 26/124
no
ID (1) PED (1), Ataxia (1)
7 (1.4) 504
GTCS (1)
4 (5) 84
Myoclonic seizure (1)
no
Tremor (1), Dysarthria (3), Ataxia (2), PED (2) PED (1)
Epilepsy (2), PED (2), Epilepsy þ PED (1) Epilepsy (1)
Cognitive decline (2), HC growth slowing (1) no
Epilepsy (3) Mild ataxia (2), PED (1)
GTCS (3), Myoclonic seizure (1) no 4 (11.7) 34
Family history (N) Movement disorders (N) Additional seizure types (N)
Additional features of patients with GLUT1D (SLC2A1þ)
Patients GLUT1D (SLC2A1þ) N (%) N Selection criteria Author
Table 1 e Systematic review of literature cohorts with generalized epilepsies screened for SLC2A1.
ID (2)
Others (N)
e u r o p e a n j o u r n a l o f p a e d i a t r i c n e u r o l o g y x x x ( 2 0 1 4 ) 1 e6
3
abnormal movements or both and others neurological and/or relevant developmental findings.
2.4.
Genetic analysis
All samples were analyzed at the laboratory of the Centre for Molecular Diseases at the Lausanne University Hospital. Genomic DNA was extracted from peripheral blood leukocytes using standard procedures. Primers for SLC2A1 were designed to cover all coding regions and intron-exon boundaries (Ensembl access number, ENSG00000117394; ENST00000426263). Fragments were amplified by PCR and were sequenced on a 3500 Genetic Analyzer (Life Technologies). Multiplex Ligation-dependent Probe Amplification (MLPA) was performed on an ABI Prism 310 Genetic Analyzer (Life Technologies) according to the manufacturer's instructions.
3.
Results
Ninety-three patients with generalized epilepsies were tested for SLC2A1 mutations. Fifty-seven cases were recruited in Lausanne and 36 in other Swiss centers. They all had generalized seizures including absence, tonic-clonic, clonic, atonic and myoclonic seizures, either alone or in combination. Twenty-five patients had a family history of epilepsy and 1 had a family history of epilepsy and a movement disorder. Eighty patients (86%) had IGEs: CAE in 14 of 93 (15%), EOAE in 4 of 93 (4.3%); JAE in 3 of 93 (3.2%), JME in 10 of 93 (10.7%), EGTCS in 6 of 93 (6.4%), MAE in 26 of 93 (28%) and epilepsy with myoclonic absences in 2 (2.1%). Unclassified IGE was diagnosed in 18 of 93 (19.3%) including 13 patients with a combination of absence and tonic-clonic seizures, 2 patients with atypical absences with eyelid myoclonia (Jeavons syndrome), two with absence epilepsy with photosensitivity, one with reflex absence epilepsy. Ten of 93 (10.7%) had a generalized epilepsy of unknown origin. In 21/93 (22.5%) patients, GLUT1D was suspected by the referring physician. Among them, 17 patients had IGEs including 4 CAE, 1 JAE, 9 MAE, 3 unclassified IGE and 4 generalized epilepsies of unknown origin. Refractory seizures were significantly more prevalent in these patients compared to the group of patient not thought likely to have GLUT1D (Fisher exact test, p ¼ 0.002). SLC2A1 variants, including one not previously reported in databases of human genetic variation, were identified in 2/93 patients (2.1%).
3.1.
Patient 1
This boy had a heterozygous de novo deletion in exon 4 with a frameshift leading to a premature stop codon (c.348delC; p.Lys117Serfs*6). His family history was unremarkable. He had a few febrile seizures from 6 to 18 months. Then he had multiple afebrile seizures including absence, myoclonic and myoclonic-atonic seizures. EEGs showed 3e4 Hz generalized spike and wave complexes activated by sleep on a slow background. At 4 years, he had global developmental delay with poor gross and fine motor skills and he was non-verbal. Clinical examination revealed ataxia, brisk deep tendons
Please cite this article in press as: Lebon S, et al., When should clinicians search for GLUT1 deficiency syndrome in childhood generalized epilepsies?, European Journal of Paediatric Neurology (2014), http://dx.doi.org/10.1016/j.ejpn.2014.11.009
4
e u r o p e a n j o u r n a l o f p a e d i a t r i c n e u r o l o g y x x x ( 2 0 1 4 ) 1 e6
reflexes and choreiform movements. Mild deceleration of head growth from the tenth to the third percentile was observed. Magnetic resonance imaging (MRI) of the brain showed delayed myelination of U-fibers in both frontotemporal areas. LP performed after only 2 h fasting showed low cerebrospinal fluid (CSF) glucose at 2 mmol/L (2.4e3.8 mmol/L)16 and low CSF: blood glucose ratio of 0.4. The KD was started and the child became seizure-free 1 month later, with marked improvement of his abnormal movements. The seizures are now controlled for 2 years. The child remains non-verbal with mild choreiform movements and ataxia.
3.2.
Patient 2
The boy was seen at the age of 14 years for a second opinion on familial generalized epilepsy and had a missense mutation in exon 5. The heterozygous c.634C > T mutation changed an arginine to a cysteine residue (p.Arg212Cys) and was inherited from his father. He had a history of jerks in the lower limbs since the age of 4, which were first interpreted as epileptic myoclonus. The patient reported involuntary jerks in both lower limbs, often occurring when he felt hungry or after a short walk before meal sometimes leading to falls. They were actually paroxysmal dyskinesia triggered by hunger or exercise on family videos. He developed brief episodes of absence seizures with upward eye deviations at 7 years. The EEG showed irregular (2e4 Hz) generalized poly-spike-waves on a normal background. At 14 years, he had 2 generalized tonicclonic seizures triggered by exercise and delayed meals. Symptoms and epileptiform discharges were overall carbohydrate responsive. Wechsler Intelligence Scale for Children Fourth Edition (WISC-IV) showed heterogeneous scores with a verbal comprehension index (VCI) of 84, perceptual reasoning index (PRI) of 65, working memory index (WMI) of 76, processing speed index at 55 and learning difficulties. His father had paroxysmal exercise induced dyskinesia (PED) of indeterminate age of onset and had no history of seizures. The monozygotic father's twin-brother had epileptic seizures and PED and his son (proband's cousin) had absences since the age of 5 years with mild intellectual disability and PED. A fasting LP showed a CSF glucose of 2.4 mmol/L and a CSF:blood glucose ratio of 0.43, which were both below the tenth percentile for the age.7 Six months after introduction of the KD, the patient's seizures and PED were fully controlled and this effect was sustained at 1 year follow-up.
3.3.
Systematic review (Table 1)
The literature searched yielded a total of 7 articles describing series of patients with generalized epilepsies and screened for SLC2A1 mutations. Two studies ascertained patients for IGEs,2,17 3 for EOAE,3,4,18 1 for EOAE and IGEs19 and 1 for MAE.5 Out of a total of 1110 patients screened for SLC2A1 mutations, 29 patients (2.6%) had GLUT1D. This rate was 2-fold higher (5.6%, p ¼ 0.01) among 303 patients from the 4 EOAE cohorts. In this subgroup, additional seizures were reported in 35% (6/17). There was a 5-fold increase risk to find GLUT1D in the EOAE compared to the IGEs (17/303 versus 8/723 respectively, Fisher exact test p < 0.001). Forty-five percent of GLUT1D patients had abnormal movements, 41% had a family history of seizures,
abnormal movements or both. Twenty seven percent had developmental delay/intellectual disability, but since these studies were focusing on epilepsy and not cognition, this value could have been underestimated.
4.
Discussion
In our cohort of 93 children and adolescents with generalized epilepsies, we found 2 patients (2.1%) harboring a SLC2A1 mutations. They were both suspected clinically to have GLUT1D on the basis of history and clinical examination. In terms of epilepsy syndrome classification, patient 1 has MAE and patient 2 has unclassified familial IGE. No patient with a classical IGE syndrome had GLUT1D. Our cases illustrate two of the GLUT1D phenotypes, one being close to the “classical developmental encephalopathy”, and the second with a milder familial form.9,11 Patient 1 has a previously unreported de novo deletion in exon 4 of SLC2A1, leading to a premature stop codon in addition to the pathognomonic finding of hypoglycorrhachia. The missense mutation p.Arg212Cys in patient 2 has been previously described and related to an early-onset severe phenotype, whereas our case corresponds to a later-onset milder phenotype reported by Leen et al.9 We are aware of the limitations and biases of our study: our cohort of patients is small and we have a higher than usual proportion of patients with MAE15: this is due to a referral bias, since these patients are now increasingly being investigated for GLUT1D and our study offered an opportunity for free diagnosis. However, even if our cohort is not representative of a wider population of generalized epilepsies, our patients have been recruited from routine pediatric neurology outpatient clinics, and the rate of 2.1% of GLUT1D was similar to that found in our systematic review (2.4%; 2/93 versus 29/1110 respectively; Fisher exact test, p ¼ 1).
4.1.
Generalized epileptic syndromes and GLUT1D
4.1.1.
Epilepsy with myoclonic atonic seizures (MAE)
GLUT1D was previously identified the cause of MAE in 4/84 patients.5 Our case had mild slowing of head growth and choreiform movements; both are atypical features for MAE. However, MAE patients can sometimes display multiple segmental myoclonus responsible for an ataxic gait (“pseudoataxia”) or choreiform movements. Therefore, identifying signs specifically suggesting GLUT1D in MAE may be tricky. Mullen et al. underline that 3 of their patients were undistinguishable from “classical” MAE from the clinical and EEG point of view.5 We suggest to follow their recommendations to search for GLUT1D in all patients with MAE.
4.1.2.
Idiopathic generalized epilepsies
GLUT1D can cause IGE syndromes of JME, JAE and CAE.2,17,19 In a recent study, Arsov et al. focused on a large cohort of 504 patients with IGE and found 7 (1.4%) patients with GLUT1D. Among them, 1 had PED and 5/7 had a family history of epilepsy, PED or both.2 PED may be mistaken for myoclonic seizures as occurred in our patient 2. This diagnosis can easily be missed in busy clinics. A careful description is essential, as
Please cite this article in press as: Lebon S, et al., When should clinicians search for GLUT1 deficiency syndrome in childhood generalized epilepsies?, European Journal of Paediatric Neurology (2014), http://dx.doi.org/10.1016/j.ejpn.2014.11.009
e u r o p e a n j o u r n a l o f p a e d i a t r i c n e u r o l o g y x x x ( 2 0 1 4 ) 1 e6
well as a video whenever possible. In summary, the probability of finding a GLUT1D in patients with classical IGE syndromes remains low, after a critical case review. This was confirmed by Muhle et al. who did not find any positive patients in a cohort of 124 typical CAE and JAE.19
4.1.3.
Early-onset absence epilepsy
EOAE is the epilepsy category in which GLUT1D has been reported in more than 10% of cases, and this finding was replicated in 3 different studies (Table 1).3,4,19 Agostinelli et al. recently studied a large cohort of 188 patients with EOAE in whom SLC2A1 was screened, divided into “pure EOAE” (n ¼ 111), adopting strict clinical and EEG criteria for CAE except for the age of onset14 and “non pure EOAE” including all other cases (n ¼ 77): 4/188 patients (2.1%) had GLUT1D and were all part of the “non pure EOAE” group (4/77; 5.2%). The authors concluded that patients with “pure EOAE” represent a homogeneous group with favorable outcome whereas the “non pure” group is heterogeneous, with various possible underlying etiologies, among them structural abnormalities and GLUT1D.18 Our 4 patients with EOAE were also “pure” and had no mutation. When pooling 303 patients with EOAE from 4 studies,3,4,18,19 only 5% had GLUT1D which is lower than the initial finding of 10%.3,4,19 Hence GLUT1D has to be searched for in children with EOAE especially when abnormal developmental or neurological features or additional seizure types like tonic-clonic and myoclonic seizures are found. An atypical EEG may be a further clue, since polymorphic and irregular discharges on an abnormal background have been reported in GLUT1D with early onset absence seizures.20
4.2.
Which patients to test and what test to perform?
Despite its apparent low rate as a cause of generalized epilepsy, the issue of making a diagnosis of GLUT1D is critical because of its implications for treatment and prognosis. This can make a tremendous difference for patients as illustrated by our two cases. KD and its variants are efficacious10 but may however be difficult to maintain in the long term, especially in older children and adolescents. Since seizures are not always refractory to AEDs and patients may only have minor cognitive difficulties or movement disorders, the burden of the KD may not seem justified.2,21 One may then ask if it is really important to undertake costly genetic analyses for GLUT1D. Of note, in milder forms of GLUT1D, seizures can decrease spontaneously with age (sometimes with the later onset of PED), suggesting that brain maturation or other compensatory mechanisms play a role that is not yet well understood.8 Mild forms of GLUT1D seem not lead to cognitive decline during adulthood.22 From our study and literature review, we suggest the following recommendations: in generalized epilepsies, testing for GLUT1D should be performed if seizures are not controlled and if there is a developmental delay and/or a movement disorder. Seizures before meals, and non-epileptic paroxysmal event triggered by fasting or exercise are good cues.11,23 A family history of movement disorder and/or seizures, especially when suggesting autosomal dominant inheritance, is also important.2,22 If the patient has a classical IGE
5
syndrome with controlled seizures, investigations for GLUT1D are not indicated. Hypoglycorrhachia being the distinctive biomarker for GLUT1D, LP remains an easy, cheap and quick way to make the diagnosis. However, CSF glucose level can be mildly lowered (2.2e2.8 mmol/l)6 especially in milder forms of GLUT1D. On another hand about 5e10% of GLUT1D patients do not carry a SLC2A1 mutation.24 Therefore, the combination of LP and genetic testing is currently the most reliable diagnostic approach. The discovery of hypoglycorrhachia in a patient with refractory seizures and cognitive decline will allow rapid initiation of the KD with the hope of improvement before the genetic result is available. SLC2A1 sequencing can be proposed first when seizures are less severe and treatment is less urgent. It is likely that, with the rapid development of next generation sequencing technologies becoming more and more readily available and cheaper, the current recommendations to search for GLUT1D will change over time.
Conflict of interest None of the authors have any conflict of interest to disclose.
Acknowledgment We would like to thank all participating family members. This work was funded by the Research Funds of the Pediatric Department at the Lausanne University Hospital. Christian M Korff is supported by Swiss National Found 140332. We also thank Dr J Kroell (Swiss Epilepsy Center, Zurich, Switzerland), Dr T Schmitt-Mechelke (Children's hospital Lucerne, Switzerland) and Dr AO Rossetti (Lausanne University hospital, Switzerland) for their participation to recruit patients. We are very grateful to Dr Ingrid Scheffer for her helpful comments.
references
1. Klepper J, Leiendecker B. GLUT1 deficiency syndromee2007 update. Dev Med Child Neurol 2007;49:707e16. 2. Arsov T, Mullen SA, Rogers S, et al. Glucose transporter 1 deficiency in the idiopathic generalized epilepsies. Ann Neurol 2012;72:807e15. 3. Suls A, Mullen SA, Weber YG, et al. Early-onset absence epilepsy caused by mutations in the glucose transporter GLUT1. Ann Neurol 2009;66:415e9. 4. Arsov T, Mullen SA, Damiano JA, et al. Early onset absence epilepsy: 1 in 10 cases is caused by GLUT1 deficiency. Epilepsia 2012;53:e204e7. 5. Mullen SA, Marini C, Suls A, et al. Glucose transporter 1 deficiency as a treatable cause of myoclonic astatic epilepsy. Arch Neurol 2011;68:1152e5. 6. De Vivo DC, Wang D. Glut1 deficiency: CSF glucose. How low is too low? Rev Neurol Paris 2008;164:877e80. 7. Leen WG, Wevers RA, Kamsteeg EJ, et al. Cerebrospinal fluid analysis in the workup of GLUT1 deficiency syndrome: a systematic review. JAMA Neurol 2013;70:1440e4.
Please cite this article in press as: Lebon S, et al., When should clinicians search for GLUT1 deficiency syndrome in childhood generalized epilepsies?, European Journal of Paediatric Neurology (2014), http://dx.doi.org/10.1016/j.ejpn.2014.11.009
6
e u r o p e a n j o u r n a l o f p a e d i a t r i c n e u r o l o g y x x x ( 2 0 1 4 ) 1 e6
8. Pong AW, Geary BR, Engelstad KM, et al. Glucose transporter type I deficiency syndrome: epilepsy phenotypes and outcomes. Epilepsia 2012;53:1503e10. 9. Leen WG, Klepper J, Verbeek MM, et al. Glucose transporter-1 deficiency syndrome: the expanding clinical and genetic spectrum of a treatable disorder. Brain 2010;133:655e70. 10. Ito Y, Oguni H, Ito S, et al. A modified Atkins diet is promising as a treatment for glucose transporter type 1 deficiency syndrome. Dev Med Child Neurol 2011;53:658e63. 11. Pearson TS, Akman C, Hinton VJ, et al. Phenotypic spectrum of glucose transporter type 1 deficiency syndrome (Glut1 DS). Curr Neurol Neurosci Rep 2013;13:342. 12. Engel Jr J. International league against E. A proposed diagnostic scheme for people with epileptic seizures and with epilepsy: report of the ILAE task force on classification and terminology. Epilepsia 2001;42:796e803. 13. Berg AT, Berkovic SF, Brodie MJ, et al. Revised terminology and concepts for organization of seizures and epilepsies: report of the ILAE commission on classification and terminology, 2005e2009. Epilepsia 2010;51:676e85. 14. Panayiotopoulos CP. Typical absence seizures and related epileptic syndromes: assessment of current state and directions for future research. Epilepsia 2008;49:2131e9. 15. Kelley SA, Kossoff EH. Doose syndrome (myoclonic-astatic epilepsy): 40 years of progress. Dev Med Child Neurol 2010;52:988e93.
16. Leen WG, Willemsen MA, Wevers RA, et al. Cerebrospinal fluid glucose and lactate: age-specific reference values and implications for clinical practice. PLoS One 2012;7:e42745. 17. Striano P, Weber YG, Toliat MR, et al. GLUT1 mutations are a rare cause of familial idiopathic generalized epilepsy. Neurology 2012;78:557e62. 18. Agostinelli S, Accorsi P, Beccaria F, et al. Clinical dissection of early onset absence epilepsy in children and prognostic implications. Epilepsia 2013;54:1761e70. 19. Muhle H, Helbig I, Froslev TG, et al. The role of SLC2A1 in early onset and childhood absence epilepsies. Epilepsy Res 2013;105:229e33. 20. Roulet-Perez E, Ballhausen D, Bonafe L, et al. Glut-1 deficiency syndrome masquerading as idiopathic generalized epilepsy. Epilepsia 2008;49:1955e8. 21. Mullen SA, Suls A, De Jonghe P, et al. Absence epilepsies with widely variable onset are a key feature of familial GLUT1 deficiency. Neurology 2010;75:432e40. 22. De Giorgis V, Veggiotti P. GLUT1 deficiency syndrome 2013: current state of the art. Seizure 2013;22:803e11. 23. Urbizu A, Cuenca-Leon E, Raspall-Chaure M, et al. Paroxysmal exercise-induced dyskinesia, writer's cramp, migraine with aura and absence epilepsy in twin brothers with a novel SLC2A1 missense mutation. J Neurol Sci 2010;295:110e3. 24. Klepper J. Absence of SLC2A1 mutations does not exclude Glut1 deficiency syndrome. Neuropediatrics 2013;44:235e6.
Please cite this article in press as: Lebon S, et al., When should clinicians search for GLUT1 deficiency syndrome in childhood generalized epilepsies?, European Journal of Paediatric Neurology (2014), http://dx.doi.org/10.1016/j.ejpn.2014.11.009
Official Journal of the European Paediatric Neurology Society
Original article
When should clinicians search for GLUT1 deficiency syndrome in childhood generalized epilepsies? Sebastien Lebon a,*, Philippe Suarez b, Semsa Alija b, Christian M. Korff c, Jo€el Fluss c, Danielle Mercati d, Alexandre N. Datta e, Claudia Poloni a, Jean-Pierre Marcoz f, Ana Belinda Campos-Xavier b, Luisa Bonafe b, Eliane Roulet-Perez a a
Pediatric Neurology and Neurorehabilitation Unit, Department of Pediatrics, Centre Hospitalier Universitaire Vaudois, Lausanne, Switzerland b Centre for Molecular Diseases, Lausanne University Hospital, Lausanne, Switzerland c Child Neurology, University Hospitals, Geneva, Switzerland d Children's Hospital Neuchatel, Neuchatel, Switzerland e Pediatric Neurology and Development Unit, University Children's Hospital, Basel, Switzerland f Children's Unit, H^ opital du Valais, Sion, Switzerland
article info
abstract
Article history:
GLUT1 deficiency (GLUT1D) has recently been identified as an important cause of gener-
Received 23 June 2014
alized epilepsies in childhood. As it is a treatable condition, it is crucial to determine which
Received in revised form
patients should be investigated.
12 November 2014
Methods: We analyzed SLC2A1 for mutations in a group of 93 unrelated children with
Accepted 24 November 2014
generalized epilepsies. Fasting lumbar puncture was performed following the identification of a mutation. We compared our results with a systematic review of 7 publications of series
Keywords:
of patients with generalized epilepsies screened for SLC2A1 mutations.
GLUT1 deficiency
Results: We found 2/93 (2.1%) patients with a SLC2A1 mutation. One, carrying a novel de novo
SLC2A1
deletion had epilepsy with myoclonic-atonic seizures (MAE), mild slowing of head growth,
Generalized epilepsy
choreiform movements and developmental delay. The other, with a paternally inherited missense mutation, had childhood absence epilepsy with atypical EEG features and paroxysmal exercise-induced dyskinesia (PED) initially misdiagnosed as myoclonic seizures. Out of a total of 1110 screened patients with generalized epilepsies from 7 studies, 2.4% (29/1110) had GLUT1D. This rate was higher (5.6%) among 303 patients with early onset absence epilepsy (EOAE) from 4 studies. About 50% of GLUT1D patients had abnormal movements and 41% a family history of seizures, abnormal movements or both. Conclusion: GLUT1D is most likely to be found in MAE and in EOAE. The probability of finding GLUT1D in the classical idiopathic generalized epilepsies is very low. Pointers to
* Corresponding author. Pediatric Neurology and Neurorehabilitation Unit, Department of Pediatrics, Centre Hospitalier Universitaire Vaudois (CHUV), 1011 Lausanne, Switzerland. Tel.: þ41 213143563; fax: þ41 213143572. E-mail address: [email protected] (S. Lebon). http://dx.doi.org/10.1016/j.ejpn.2014.11.009 1090-3798/© 2014 European Paediatric Neurology Society. Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Lebon S, et al., When should clinicians search for GLUT1 deficiency syndrome in childhood generalized epilepsies?, European Journal of Paediatric Neurology (2014), http://dx.doi.org/10.1016/j.ejpn.2014.11.009
2
e u r o p e a n j o u r n a l o f p a e d i a t r i c n e u r o l o g y x x x ( 2 0 1 4 ) 1 e6
GLUT1D include an increase in seizures before meals, cognitive impairment, or PED which can easily be overlooked. © 2014 European Paediatric Neurology Society. Published by Elsevier Ltd. All rights reserved.
1.
Introduction
Glucose transporter type 1 deficiency (GLUT1D) leads to insufficient glucose for brain metabolism and has long been associated with severe seizures, microcephaly, motor disorders and developmental delay.1 Recently GLUT1D has been identified as a cause of generalized epilepsies such as socalled “idiopathic” generalized epilepsies (IGE), early-onset absence epilepsy (EOAE) and epilepsy with myoclonic-atonic seizures (MAE).2e5 Diagnosis is based on the presence of low glucose in the cerebrospinal fluid (CSF),6,7 and a mutation in SLC2A1 (solute carrier family 2 member 1) is found in the majority of cases.1 The ketogenic diet (KD) or its modified forms can usually control seizures and abnormal movements and improve cognitive function.8e10 When seizures start in infancy with an encephalopathic phenotype,11 they will readily prompt a complete diagnostic work-up, but when they present later as a generalized epilepsy, the challenge for clinicians caring for children and adolescents is to decide which patients should be investigated. We screened SLC2A1 in a cohort of children and adolescents with either new onset or established generalized epilepsies according to clinical and electro-encephalographic (EEG) features. Our findings were compared to a systematic review of 7 cohorts of patients with generalized epilepsies screened for SLC2A1 mutations from the literature.
2.
Materials and methods
2.1.
Patients
Patient data and blood samples were collected at the Lausanne University Hospital and other Swiss pediatric neurology ^ tel, Sion, Luzern and outpatient clinics (Geneva, Basel, Neucha Zurich) between September 2011 and May 2013. Inclusion criteria were: children and adolescents aged 18 years, a diagnosis of generalized epilepsy on clinical and EEG criteria (interictal or ictal generalized (poly-) spike and (poly-) spikewaves). Exclusion criteria were focal epilepsies, symptomatic epilepsies or the family refusal to participate in the study. The protocol was approved by the ethics committees of all participating centers. Informed consent from the parents or legal guardian was obtained for all patients. Probands were phenotyped including seizure, neurological and developmental histories and examination by a pediatric neurologist. Data were reported on a standardized form including the following: family history of epilepsy or abnormal movements, developmental history, learning disabilities, abnormal movements, seizure manifestations, EEG findings
and whether a diagnosis of GLUT1D was suspected before enrolment.
2.2.
Electroclinical definitions
Probands were categorized, whenever possible, into the classical idiopathic generalized epilepsy syndromes of childhood absence epilepsy (CAE), juvenile absence epilepsy (JAE), juvenile myoclonic epilepsy (JME) and epilepsy with generalized tonic-clonic seizure alone (EGTCA) according to the International League Against Epilepsy (ILAE) classification.12,13 For instance, a patient was classified in the CAE group only when he/she fulfilled the following: at least daily typical absences accompanied by bilateral, regular, symmetric, generalized 3e4 Hz spike-waves with normal background on EEG, normal development and neurologic examination.14 A child was classified into the early onset absence epilepsy (EOAE) group only when absence seizures started before four years of age but otherwise fulfilled the criteria of CAE. The generalized epilepsy syndrome of epilepsy with myoclonic-atonic seizures (MAE) is characterized by afebrile seizures starting between one and five years of age with multiple seizures types including myoclonic-atonic, atonic, or myoclonic seizures with or without generalized tonic-clonic seizures, and absence seizures. One-third has earlier febrile seizures. The EEG shows generalized spike- or polyspike and wave discharges and often background slowing.15 Children may have abnormal development prior to seizure onset or present subsequent cognitive impairment. Patients not fitting into MAE or classical IGE, such as those with absence seizures with eyelid myoclonia (Jeavons syndrome) or a combination of absence, tonic-clonic, tonic or atonic seizures, were denoted as unclassified IGE when development and neurological examination were normal. Patients with abnormal neurological findings and/or abnormal development before seizure onset were classified into generalized epilepsy of unknown origin. A fasting lumbar puncture (LP) was performed in patients in whom a SLC2A1 mutation was found.
2.3.
Systematic review
PubMed was searched by using the terms SLC2A1-gene and epilepsy. Relevant references mentioned in the articles were also included. Patients who were described in the literature more than once were included only once. We focused on publications reporting series of epileptic patients with generalized seizures screened for SLC2A1 (Table 1) and looked for features that may point to GLUT1D: additional seizures type, movement disorders, family history of epilepsy,
Please cite this article in press as: Lebon S, et al., When should clinicians search for GLUT1 deficiency syndrome in childhood generalized epilepsies?, European Journal of Paediatric Neurology (2014), http://dx.doi.org/10.1016/j.ejpn.2014.11.009
PED (1), Ataxia (1) NA 7 (12.7) 4 (2.1) 55 188
GTCS (2) NA
1 (1) 95 probands
EOAE
MAE
IGE
EOAE/CAE þ JAE familial IGE
EOAE EOAE
Suls et al.3
Mullen et al.5
Arsov et al.2
Muhle et al.19
Striano et al.17
Arsov et al.4 Agostinelli et al.18
GTCS: generalized tonic clonic seizure, IGE: idiopathic generalized epilepsy, EOAE: early onset absence epilepsy, CAE: childhood absence epilepsy, MAE: myoclono-atonic epilepsy, JAE: juvenile absence epilepsy, PED: paroxysmal exercise-induced dyskinesia, ID: intellectual disability, DD: developmental delay, NA: not available, HC: head circumference.
ID (2), DD (1) NA
no
Epilepsy þ borderline intellect (1) Epilepsy (1), PED (1) NA no
2 (7.7)/0 26/124
no
ID (1) PED (1), Ataxia (1)
7 (1.4) 504
GTCS (1)
4 (5) 84
Myoclonic seizure (1)
no
Tremor (1), Dysarthria (3), Ataxia (2), PED (2) PED (1)
Epilepsy (2), PED (2), Epilepsy þ PED (1) Epilepsy (1)
Cognitive decline (2), HC growth slowing (1) no
Epilepsy (3) Mild ataxia (2), PED (1)
GTCS (3), Myoclonic seizure (1) no 4 (11.7) 34
Family history (N) Movement disorders (N) Additional seizure types (N)
Additional features of patients with GLUT1D (SLC2A1þ)
Patients GLUT1D (SLC2A1þ) N (%) N Selection criteria Author
Table 1 e Systematic review of literature cohorts with generalized epilepsies screened for SLC2A1.
ID (2)
Others (N)
e u r o p e a n j o u r n a l o f p a e d i a t r i c n e u r o l o g y x x x ( 2 0 1 4 ) 1 e6
3
abnormal movements or both and others neurological and/or relevant developmental findings.
2.4.
Genetic analysis
All samples were analyzed at the laboratory of the Centre for Molecular Diseases at the Lausanne University Hospital. Genomic DNA was extracted from peripheral blood leukocytes using standard procedures. Primers for SLC2A1 were designed to cover all coding regions and intron-exon boundaries (Ensembl access number, ENSG00000117394; ENST00000426263). Fragments were amplified by PCR and were sequenced on a 3500 Genetic Analyzer (Life Technologies). Multiplex Ligation-dependent Probe Amplification (MLPA) was performed on an ABI Prism 310 Genetic Analyzer (Life Technologies) according to the manufacturer's instructions.
3.
Results
Ninety-three patients with generalized epilepsies were tested for SLC2A1 mutations. Fifty-seven cases were recruited in Lausanne and 36 in other Swiss centers. They all had generalized seizures including absence, tonic-clonic, clonic, atonic and myoclonic seizures, either alone or in combination. Twenty-five patients had a family history of epilepsy and 1 had a family history of epilepsy and a movement disorder. Eighty patients (86%) had IGEs: CAE in 14 of 93 (15%), EOAE in 4 of 93 (4.3%); JAE in 3 of 93 (3.2%), JME in 10 of 93 (10.7%), EGTCS in 6 of 93 (6.4%), MAE in 26 of 93 (28%) and epilepsy with myoclonic absences in 2 (2.1%). Unclassified IGE was diagnosed in 18 of 93 (19.3%) including 13 patients with a combination of absence and tonic-clonic seizures, 2 patients with atypical absences with eyelid myoclonia (Jeavons syndrome), two with absence epilepsy with photosensitivity, one with reflex absence epilepsy. Ten of 93 (10.7%) had a generalized epilepsy of unknown origin. In 21/93 (22.5%) patients, GLUT1D was suspected by the referring physician. Among them, 17 patients had IGEs including 4 CAE, 1 JAE, 9 MAE, 3 unclassified IGE and 4 generalized epilepsies of unknown origin. Refractory seizures were significantly more prevalent in these patients compared to the group of patient not thought likely to have GLUT1D (Fisher exact test, p ¼ 0.002). SLC2A1 variants, including one not previously reported in databases of human genetic variation, were identified in 2/93 patients (2.1%).
3.1.
Patient 1
This boy had a heterozygous de novo deletion in exon 4 with a frameshift leading to a premature stop codon (c.348delC; p.Lys117Serfs*6). His family history was unremarkable. He had a few febrile seizures from 6 to 18 months. Then he had multiple afebrile seizures including absence, myoclonic and myoclonic-atonic seizures. EEGs showed 3e4 Hz generalized spike and wave complexes activated by sleep on a slow background. At 4 years, he had global developmental delay with poor gross and fine motor skills and he was non-verbal. Clinical examination revealed ataxia, brisk deep tendons
Please cite this article in press as: Lebon S, et al., When should clinicians search for GLUT1 deficiency syndrome in childhood generalized epilepsies?, European Journal of Paediatric Neurology (2014), http://dx.doi.org/10.1016/j.ejpn.2014.11.009
4
e u r o p e a n j o u r n a l o f p a e d i a t r i c n e u r o l o g y x x x ( 2 0 1 4 ) 1 e6
reflexes and choreiform movements. Mild deceleration of head growth from the tenth to the third percentile was observed. Magnetic resonance imaging (MRI) of the brain showed delayed myelination of U-fibers in both frontotemporal areas. LP performed after only 2 h fasting showed low cerebrospinal fluid (CSF) glucose at 2 mmol/L (2.4e3.8 mmol/L)16 and low CSF: blood glucose ratio of 0.4. The KD was started and the child became seizure-free 1 month later, with marked improvement of his abnormal movements. The seizures are now controlled for 2 years. The child remains non-verbal with mild choreiform movements and ataxia.
3.2.
Patient 2
The boy was seen at the age of 14 years for a second opinion on familial generalized epilepsy and had a missense mutation in exon 5. The heterozygous c.634C > T mutation changed an arginine to a cysteine residue (p.Arg212Cys) and was inherited from his father. He had a history of jerks in the lower limbs since the age of 4, which were first interpreted as epileptic myoclonus. The patient reported involuntary jerks in both lower limbs, often occurring when he felt hungry or after a short walk before meal sometimes leading to falls. They were actually paroxysmal dyskinesia triggered by hunger or exercise on family videos. He developed brief episodes of absence seizures with upward eye deviations at 7 years. The EEG showed irregular (2e4 Hz) generalized poly-spike-waves on a normal background. At 14 years, he had 2 generalized tonicclonic seizures triggered by exercise and delayed meals. Symptoms and epileptiform discharges were overall carbohydrate responsive. Wechsler Intelligence Scale for Children Fourth Edition (WISC-IV) showed heterogeneous scores with a verbal comprehension index (VCI) of 84, perceptual reasoning index (PRI) of 65, working memory index (WMI) of 76, processing speed index at 55 and learning difficulties. His father had paroxysmal exercise induced dyskinesia (PED) of indeterminate age of onset and had no history of seizures. The monozygotic father's twin-brother had epileptic seizures and PED and his son (proband's cousin) had absences since the age of 5 years with mild intellectual disability and PED. A fasting LP showed a CSF glucose of 2.4 mmol/L and a CSF:blood glucose ratio of 0.43, which were both below the tenth percentile for the age.7 Six months after introduction of the KD, the patient's seizures and PED were fully controlled and this effect was sustained at 1 year follow-up.
3.3.
Systematic review (Table 1)
The literature searched yielded a total of 7 articles describing series of patients with generalized epilepsies and screened for SLC2A1 mutations. Two studies ascertained patients for IGEs,2,17 3 for EOAE,3,4,18 1 for EOAE and IGEs19 and 1 for MAE.5 Out of a total of 1110 patients screened for SLC2A1 mutations, 29 patients (2.6%) had GLUT1D. This rate was 2-fold higher (5.6%, p ¼ 0.01) among 303 patients from the 4 EOAE cohorts. In this subgroup, additional seizures were reported in 35% (6/17). There was a 5-fold increase risk to find GLUT1D in the EOAE compared to the IGEs (17/303 versus 8/723 respectively, Fisher exact test p < 0.001). Forty-five percent of GLUT1D patients had abnormal movements, 41% had a family history of seizures,
abnormal movements or both. Twenty seven percent had developmental delay/intellectual disability, but since these studies were focusing on epilepsy and not cognition, this value could have been underestimated.
4.
Discussion
In our cohort of 93 children and adolescents with generalized epilepsies, we found 2 patients (2.1%) harboring a SLC2A1 mutations. They were both suspected clinically to have GLUT1D on the basis of history and clinical examination. In terms of epilepsy syndrome classification, patient 1 has MAE and patient 2 has unclassified familial IGE. No patient with a classical IGE syndrome had GLUT1D. Our cases illustrate two of the GLUT1D phenotypes, one being close to the “classical developmental encephalopathy”, and the second with a milder familial form.9,11 Patient 1 has a previously unreported de novo deletion in exon 4 of SLC2A1, leading to a premature stop codon in addition to the pathognomonic finding of hypoglycorrhachia. The missense mutation p.Arg212Cys in patient 2 has been previously described and related to an early-onset severe phenotype, whereas our case corresponds to a later-onset milder phenotype reported by Leen et al.9 We are aware of the limitations and biases of our study: our cohort of patients is small and we have a higher than usual proportion of patients with MAE15: this is due to a referral bias, since these patients are now increasingly being investigated for GLUT1D and our study offered an opportunity for free diagnosis. However, even if our cohort is not representative of a wider population of generalized epilepsies, our patients have been recruited from routine pediatric neurology outpatient clinics, and the rate of 2.1% of GLUT1D was similar to that found in our systematic review (2.4%; 2/93 versus 29/1110 respectively; Fisher exact test, p ¼ 1).
4.1.
Generalized epileptic syndromes and GLUT1D
4.1.1.
Epilepsy with myoclonic atonic seizures (MAE)
GLUT1D was previously identified the cause of MAE in 4/84 patients.5 Our case had mild slowing of head growth and choreiform movements; both are atypical features for MAE. However, MAE patients can sometimes display multiple segmental myoclonus responsible for an ataxic gait (“pseudoataxia”) or choreiform movements. Therefore, identifying signs specifically suggesting GLUT1D in MAE may be tricky. Mullen et al. underline that 3 of their patients were undistinguishable from “classical” MAE from the clinical and EEG point of view.5 We suggest to follow their recommendations to search for GLUT1D in all patients with MAE.
4.1.2.
Idiopathic generalized epilepsies
GLUT1D can cause IGE syndromes of JME, JAE and CAE.2,17,19 In a recent study, Arsov et al. focused on a large cohort of 504 patients with IGE and found 7 (1.4%) patients with GLUT1D. Among them, 1 had PED and 5/7 had a family history of epilepsy, PED or both.2 PED may be mistaken for myoclonic seizures as occurred in our patient 2. This diagnosis can easily be missed in busy clinics. A careful description is essential, as
Please cite this article in press as: Lebon S, et al., When should clinicians search for GLUT1 deficiency syndrome in childhood generalized epilepsies?, European Journal of Paediatric Neurology (2014), http://dx.doi.org/10.1016/j.ejpn.2014.11.009
e u r o p e a n j o u r n a l o f p a e d i a t r i c n e u r o l o g y x x x ( 2 0 1 4 ) 1 e6
well as a video whenever possible. In summary, the probability of finding a GLUT1D in patients with classical IGE syndromes remains low, after a critical case review. This was confirmed by Muhle et al. who did not find any positive patients in a cohort of 124 typical CAE and JAE.19
4.1.3.
Early-onset absence epilepsy
EOAE is the epilepsy category in which GLUT1D has been reported in more than 10% of cases, and this finding was replicated in 3 different studies (Table 1).3,4,19 Agostinelli et al. recently studied a large cohort of 188 patients with EOAE in whom SLC2A1 was screened, divided into “pure EOAE” (n ¼ 111), adopting strict clinical and EEG criteria for CAE except for the age of onset14 and “non pure EOAE” including all other cases (n ¼ 77): 4/188 patients (2.1%) had GLUT1D and were all part of the “non pure EOAE” group (4/77; 5.2%). The authors concluded that patients with “pure EOAE” represent a homogeneous group with favorable outcome whereas the “non pure” group is heterogeneous, with various possible underlying etiologies, among them structural abnormalities and GLUT1D.18 Our 4 patients with EOAE were also “pure” and had no mutation. When pooling 303 patients with EOAE from 4 studies,3,4,18,19 only 5% had GLUT1D which is lower than the initial finding of 10%.3,4,19 Hence GLUT1D has to be searched for in children with EOAE especially when abnormal developmental or neurological features or additional seizure types like tonic-clonic and myoclonic seizures are found. An atypical EEG may be a further clue, since polymorphic and irregular discharges on an abnormal background have been reported in GLUT1D with early onset absence seizures.20
4.2.
Which patients to test and what test to perform?
Despite its apparent low rate as a cause of generalized epilepsy, the issue of making a diagnosis of GLUT1D is critical because of its implications for treatment and prognosis. This can make a tremendous difference for patients as illustrated by our two cases. KD and its variants are efficacious10 but may however be difficult to maintain in the long term, especially in older children and adolescents. Since seizures are not always refractory to AEDs and patients may only have minor cognitive difficulties or movement disorders, the burden of the KD may not seem justified.2,21 One may then ask if it is really important to undertake costly genetic analyses for GLUT1D. Of note, in milder forms of GLUT1D, seizures can decrease spontaneously with age (sometimes with the later onset of PED), suggesting that brain maturation or other compensatory mechanisms play a role that is not yet well understood.8 Mild forms of GLUT1D seem not lead to cognitive decline during adulthood.22 From our study and literature review, we suggest the following recommendations: in generalized epilepsies, testing for GLUT1D should be performed if seizures are not controlled and if there is a developmental delay and/or a movement disorder. Seizures before meals, and non-epileptic paroxysmal event triggered by fasting or exercise are good cues.11,23 A family history of movement disorder and/or seizures, especially when suggesting autosomal dominant inheritance, is also important.2,22 If the patient has a classical IGE
5
syndrome with controlled seizures, investigations for GLUT1D are not indicated. Hypoglycorrhachia being the distinctive biomarker for GLUT1D, LP remains an easy, cheap and quick way to make the diagnosis. However, CSF glucose level can be mildly lowered (2.2e2.8 mmol/l)6 especially in milder forms of GLUT1D. On another hand about 5e10% of GLUT1D patients do not carry a SLC2A1 mutation.24 Therefore, the combination of LP and genetic testing is currently the most reliable diagnostic approach. The discovery of hypoglycorrhachia in a patient with refractory seizures and cognitive decline will allow rapid initiation of the KD with the hope of improvement before the genetic result is available. SLC2A1 sequencing can be proposed first when seizures are less severe and treatment is less urgent. It is likely that, with the rapid development of next generation sequencing technologies becoming more and more readily available and cheaper, the current recommendations to search for GLUT1D will change over time.
Conflict of interest None of the authors have any conflict of interest to disclose.
Acknowledgment We would like to thank all participating family members. This work was funded by the Research Funds of the Pediatric Department at the Lausanne University Hospital. Christian M Korff is supported by Swiss National Found 140332. We also thank Dr J Kroell (Swiss Epilepsy Center, Zurich, Switzerland), Dr T Schmitt-Mechelke (Children's hospital Lucerne, Switzerland) and Dr AO Rossetti (Lausanne University hospital, Switzerland) for their participation to recruit patients. We are very grateful to Dr Ingrid Scheffer for her helpful comments.
references
1. Klepper J, Leiendecker B. GLUT1 deficiency syndromee2007 update. Dev Med Child Neurol 2007;49:707e16. 2. Arsov T, Mullen SA, Rogers S, et al. Glucose transporter 1 deficiency in the idiopathic generalized epilepsies. Ann Neurol 2012;72:807e15. 3. Suls A, Mullen SA, Weber YG, et al. Early-onset absence epilepsy caused by mutations in the glucose transporter GLUT1. Ann Neurol 2009;66:415e9. 4. Arsov T, Mullen SA, Damiano JA, et al. Early onset absence epilepsy: 1 in 10 cases is caused by GLUT1 deficiency. Epilepsia 2012;53:e204e7. 5. Mullen SA, Marini C, Suls A, et al. Glucose transporter 1 deficiency as a treatable cause of myoclonic astatic epilepsy. Arch Neurol 2011;68:1152e5. 6. De Vivo DC, Wang D. Glut1 deficiency: CSF glucose. How low is too low? Rev Neurol Paris 2008;164:877e80. 7. Leen WG, Wevers RA, Kamsteeg EJ, et al. Cerebrospinal fluid analysis in the workup of GLUT1 deficiency syndrome: a systematic review. JAMA Neurol 2013;70:1440e4.
Please cite this article in press as: Lebon S, et al., When should clinicians search for GLUT1 deficiency syndrome in childhood generalized epilepsies?, European Journal of Paediatric Neurology (2014), http://dx.doi.org/10.1016/j.ejpn.2014.11.009
6
e u r o p e a n j o u r n a l o f p a e d i a t r i c n e u r o l o g y x x x ( 2 0 1 4 ) 1 e6
8. Pong AW, Geary BR, Engelstad KM, et al. Glucose transporter type I deficiency syndrome: epilepsy phenotypes and outcomes. Epilepsia 2012;53:1503e10. 9. Leen WG, Klepper J, Verbeek MM, et al. Glucose transporter-1 deficiency syndrome: the expanding clinical and genetic spectrum of a treatable disorder. Brain 2010;133:655e70. 10. Ito Y, Oguni H, Ito S, et al. A modified Atkins diet is promising as a treatment for glucose transporter type 1 deficiency syndrome. Dev Med Child Neurol 2011;53:658e63. 11. Pearson TS, Akman C, Hinton VJ, et al. Phenotypic spectrum of glucose transporter type 1 deficiency syndrome (Glut1 DS). Curr Neurol Neurosci Rep 2013;13:342. 12. Engel Jr J. International league against E. A proposed diagnostic scheme for people with epileptic seizures and with epilepsy: report of the ILAE task force on classification and terminology. Epilepsia 2001;42:796e803. 13. Berg AT, Berkovic SF, Brodie MJ, et al. Revised terminology and concepts for organization of seizures and epilepsies: report of the ILAE commission on classification and terminology, 2005e2009. Epilepsia 2010;51:676e85. 14. Panayiotopoulos CP. Typical absence seizures and related epileptic syndromes: assessment of current state and directions for future research. Epilepsia 2008;49:2131e9. 15. Kelley SA, Kossoff EH. Doose syndrome (myoclonic-astatic epilepsy): 40 years of progress. Dev Med Child Neurol 2010;52:988e93.
16. Leen WG, Willemsen MA, Wevers RA, et al. Cerebrospinal fluid glucose and lactate: age-specific reference values and implications for clinical practice. PLoS One 2012;7:e42745. 17. Striano P, Weber YG, Toliat MR, et al. GLUT1 mutations are a rare cause of familial idiopathic generalized epilepsy. Neurology 2012;78:557e62. 18. Agostinelli S, Accorsi P, Beccaria F, et al. Clinical dissection of early onset absence epilepsy in children and prognostic implications. Epilepsia 2013;54:1761e70. 19. Muhle H, Helbig I, Froslev TG, et al. The role of SLC2A1 in early onset and childhood absence epilepsies. Epilepsy Res 2013;105:229e33. 20. Roulet-Perez E, Ballhausen D, Bonafe L, et al. Glut-1 deficiency syndrome masquerading as idiopathic generalized epilepsy. Epilepsia 2008;49:1955e8. 21. Mullen SA, Suls A, De Jonghe P, et al. Absence epilepsies with widely variable onset are a key feature of familial GLUT1 deficiency. Neurology 2010;75:432e40. 22. De Giorgis V, Veggiotti P. GLUT1 deficiency syndrome 2013: current state of the art. Seizure 2013;22:803e11. 23. Urbizu A, Cuenca-Leon E, Raspall-Chaure M, et al. Paroxysmal exercise-induced dyskinesia, writer's cramp, migraine with aura and absence epilepsy in twin brothers with a novel SLC2A1 missense mutation. J Neurol Sci 2010;295:110e3. 24. Klepper J. Absence of SLC2A1 mutations does not exclude Glut1 deficiency syndrome. Neuropediatrics 2013;44:235e6.
Please cite this article in press as: Lebon S, et al., When should clinicians search for GLUT1 deficiency syndrome in childhood generalized epilepsies?, European Journal of Paediatric Neurology (2014), http://dx.doi.org/10.1016/j.ejpn.2014.11.009