J Neurol (2014) 261:2009–2015 DOI 10.1007/s00415-014-7441-5
Severe familial paroxysmal exercise-induced dyskinesia Pawel Tacik • Sebastian Loens • Christoph Schrader Sabine Gayde-Stephan • Saskia Biskup • Dirk Dressler
Received: 17 March 2014 / Revised: 10 July 2014 / Accepted: 11 July 2014 / Published online: 7 August 2014 Ó Springer-Verlag Berlin Heidelberg 2014
Abstract Familial paroxysmal exercise-induced dyskinesia (PED) is a rare movement disorder that is mostly caused by mutations in the solute carrier family 2, member 1 (SLC2A1) gene and inherited in an autosomal dominant manner. Clinical, laboratory, and genetic studies were performed in three family members. The proband’s symptoms were recorded in a private video. He was placed on clonazepam. The 42-year-old proband presented with a 34-year-history of ‘‘dancing fits’’ suggesting a psychogenic aetiology. They occurred spontaneously or were triggered by physical exercise with a frequency up to six episodes per month, duration up to 30 min and no impairment of consciousness. Cerebrospinal fluid-(CSF)-to-blood glucose ratio was slightly reduced (0.59) and electroencephalograms were unremarkable. His 63-year-old father had less severe symptoms with spontaneous recovery before age of 45. The proband and his 38-year-old only brother also reported daily absence episodes early in the morning with an onset at age three and spontaneous recovery before age 15. Genetic testing revealed a novel c.972G[A, p.S324S heterozygous variant in the SLC2A1 gene in three patients. No splicing defects at the RNA level could be
Electronic supplementary material The online version of this article (doi:10.1007/s00415-014-7441-5) contains supplementary material, which is available to authorized users. P. Tacik (&) S. Loens C. Schrader S. Gayde-Stephan D. Dressler Movement Disorders Section, Department of Neurology, Hannover Medical School, Carl-Neuberg-Str. 1, 30625 Hannover, Germany e-mail: [email protected]
S. Biskup Center for Genomics and Transcriptomics (CeGaT GmbH), Paul-Ehrlich-Str. 23, 72076 Tu¨bingen, Germany
demonstrated. Five milligrams per day of clonazepam allowed for excellent control of PED. PED may produce a broad range of bizarre movements mimicking psychogenic movement disorders. A positive family history suggests an organic aetiology. PED can effectively be treated with clonazepam. Clinical manifestations, autosomal dominant inheritance and CSF findings suggest a causative role of the SLC2A1 gene, although no splicing defect at the RNA level could be demonstrated for the novel variant. Additional studies such as exome sequencing are indicated. Keywords PED EOAE SLC2A1 DYT18 GLUT1DS Clonazepam
Introduction Paroxysmal dyskinesias are transient, involuntary movements that are not associated with epileptiform discharges or impairment of consciousness [1, 3]. They may clinically present as dystonia, choreoathetosis, ballism or a mixture of these and may be subdivided based upon precipitating factors into: paroxysmal kinesigenic dyskinesia (Kertesz type), paroxysmal non-kinesigenic dyskinesia (Mount and Reback type) and rarer paroxysmal exercise-induced dyskinesia (PED, or Lance type, or dystonia 18) . In PED, episodes last from 5 min to an hour, are triggered by prolonged physical exercise, are typically restricted to the exercised limbs and most frequently affect the lower extremities [2–4, 11]. Initially, PED was generally considered a separate entity, as only cases with movement disorders were reported. Later co-occurrence of epilepsy, migraine, haemolytic anaemia, and aggressive behaviour were found [1, 4, 5]. Since 2008, PED has been genetically-linked to mutations in the solute carrier family 2, member 1 (SLC2A1) gene on chromosome
1p34.2, and lead to reduced glucose transport into the brain [1, 4, 5]. Although SLC2A1 mutations are likely to be the most common cause of PED , familial and sporadic cases linked to other genes [10, 12] or still unknown genetic causes [1, 4, 11, 17] have been described. Symptomatic cases have also been reported . SLC2A1 mutations have been previously found to be causative of the glucose transporter-1 deficiency syndrome (GLUT1-DS or De Vivo disease) that is characterised by infantile refractory seizures, acquired microcephaly, severe developmental delay, dysarthria and complex movement disorders [7, 17]. Although PED and GLUT1-DS were initially seen as separate syndromes , PED is now more often regarded as part of the rarer ‘‘non-classical’’ phenotypic spectrum of GLUT1-DS . The previous definition of GLUT1-DS falls into the category of the severe, ‘‘classical’’ GLUT1-DS phenotype [7, 8] that accounts for 80–90 % of all reported GLUT1-DS cases [5, 7]. Current proposals to classify GLUT-DS phenotypes are symptom-oriented and include: (1) epileptic syndromes, (2) continuous and/or paroxysmal movement disorders (including other paroxysmal events such as weakness, lethargy, somnolence, sleep disturbances, migraine, and dyspraxia), (3) haemolytic anaemia, (4) intellectual impairment or other psychiatric alterations and (5) acquired microcephaly . So far, over 100 different mutations in the SLC2A1 gene have been identified  in approximately 400  patients with GLUT1-DS. The mutations include large-scale deletions, missense, nonsense, frame shift, splice-site and translation initiation mutations . Heterozygous missense mutations result in more benign clinical presentations [19, 20], such as intermittent dyskinesias. Ninety percent of GLUT1-DS cases are attributed to de novo heterozygous mutations, whereas only 10 % of GLUT1-DS cases, especially those with a ‘‘non-classical’’ phenotype, have a positive family history and are most commonly inherited in an autosomal dominant manner [8, 9, 20]. Incidence of GLUT1-related PED is difficult to estimate. PED often occurs together with other movement disorders or additional GLUT1-related features. In a group of patients, mostly with more than one motor abnormality within the ‘‘classical’’ GLUT1-DS phenotype, 8 % showed paroxysmal dyskinesia . Besides, the symptoms can change in the course of GLUT1-DS [8, 21]. Finally, subtle PED can be overlooked [16, 22] or misdiagnosed as simple action dystonia [6, 17]. GLUT1-related ‘‘pure’’ PED [2, 6, 17] defined as a paroxysmal, exercise-induced movement disorder that consists predominantly of dystonia and/or choreoathetosis without additional GLUT1-related features has been reported in approximately 20 patients [4, 6, 13, 16, 17, 22]. Herein, we report three family members with GLUT1DS and a novel variant in the SLC2A1 gene.
J Neurol (2014) 261:2009–2015
Methods Clinical, genealogical and laboratory investigations Genealogical and clinical evaluations of three family members were performed, between February 2012 and May 2014 in the Movement Disorders Section at the Hannover Medical School in Germany, by means of medical chart reviews, interviews of the patients, neurological examinations and analysis of clinical symptoms of the proband recorded in a private video. He was placed on clonazepam. His diagnostic procedures prior to his first visit to the clinic included brain magnetic resonance imaging (MRI), electroencephalographic studies (EEG), cerebrospinal fluid (CSF), blood count and clinical biochemistry analyses. Molecular genetic studies Prior to the first visit to the clinic, the proband was tested for episodic ataxias (KCNA1, 5, 6 and CACNA1A genes), paroxysmal kinesigenic and non-kinesigenic dyskinesia (PRRT2 and MR1 genes). Genomic DNA was extracted from blood samples by the standard procedures. Automated Sanger sequencing was performed in the Center for Genomics and Transcriptomics (CeGaT) in Tu¨bingen, Germany, to study the entire coding sequence including exon–intron boundaries of each of the 10 exons of the SLC2A1 gene. After PCR-amplification using flanking intronic primers (can be provided upon request) purified PCR products were sequenced in both directions using Big Dye Terminator version 3.1 Cycle Sequencing kit (Applied Biosystems). An ABI3730xl Genetic Analyzer (Applied Biosystems) was used on the purified sequencing reaction products and DNA sequence variants were identified with SeqPilot software (JSI) in all the patients. For RNA sequencing, RNA was isolated using the PAXgene Blood RNA Kit and (PreAnalytix, 74104) PAXgene Blood RNA Tubes (BD, 762174) according to the manufacturer’s protocol. cDNA synthesis was performed with the Transcriptor High Fidelity cDNA Synthesis Kit from Roche (#05 081 955 001) using 300 ng of RNA. After PCR-amplification using exonic primers in the SLC2A1 mRNA sequence (can be provided upon request) purified PCR products were visualised and then sequenced in both directions using Big Dye Terminator version 3.1 Cycle Sequencing kit (Applied Biosystems). An ABI3730xl Genetic Analyzer (Applied Biosystems) was used on the purified sequencing reaction products and RNA sequence variants were identified with SeqPilot software (JSI).
J Neurol (2014) 261:2009–2015
Results Genealogical, clinical and laboratory investigations Patient 1 A 42-year-old male proband presented with a history of ‘‘dancing fits’’ beginning at the age of eight. They were initially observed in the morning without any apparent cause. Since adolescence, they had also been trigged by physical exertion at any time during the day. Other trigger factors, such as hunger, exposure to cold or emotional stress, coffee and alcohol, were denied. The fits were painless and their frequency normally ranged from one to six episodes per month with a duration up to 30 min, although the severity and duration was proportionally dependent on physical activity intensity. In the 3 weeks prior to his first visit, the severity had increased and frequency reached three episodes a week without any apparent cause. The exercise-induced spells occur during or directly after exertion and usually affected muscles that had previously been activated. The lower extremities were most frequently involved giving an impression of dancing. Falling to the ground would not stop the fits. The upper limbs, trunk, neck and even face were also involved in most fits. Frequent prodromal symptoms included inner disquiet and tingling sensation in the limbs. Voluntary suppression of the fits was not possible. Consciousness was never impaired. A private video revealed a broad range of symmetrical, inconsistent and bizarre movements of his extremities, trunk, neck and facial muscles (Video 1). They affected his lower and upper extremities equally. The movements were mainly choreoathetotic, sometimes more ballistic and occasionally resembled myoclonic jerks. Dystonic postures could also be observed whilst lying and standing. Grasping his hands behind his back during trunk flexion or opisthotonus seemed to bring relief. The patient maintained verbal contact with his mother. The fits took about 20 min and spontaneously ceased. His neurological status between fits was normal. EEG studies, including one recorded during an episode of the ‘‘dancing fits’’, were normal. Previous treatment attempts with unknown doses of acetazolamide, phenytoin, lorazepam, carbamazepine and valproate were ineffective. After 2 months on clonazepam, he became free of fits at a 3 mg/day dose for 1 year. Later, minor fits led to further dose increase up to 5 mg/day. This allowed excellent control of the fits such that the patient refused additional treatment options. Besides, the patient reported daily absence episodes with abrupt onset and duration up to 10 s between the ages
Fig. 1 Pedigree structure of the family. Standard symbols were used. Round symbols indicate females, squares males, diagonal lines indicate the individual is deceased. Diamonds were used to disguise gender. The solid arrowhead indicates the proband. Black full-filled symbols indicate individuals with clinical features of GLUT1-DS. Asterisks indicate those family members that underwent genetic testing for SLC2A1 gene mutations
three and 13. Episodes consisted of staring, confusion and unresponsiveness without motor abnormalities. Initially, up to five subsequent episodes were seen daily, always early in the morning after awakening. EEG was normal. However, it was never performed during absence episodes. Therapy with valproate was ineffective. Later, the episodes became more infrequent and spontaneously ceased. Brain MRI and CSF parameters were normal, including glucose and lactate, but the CSF-to-blood glucose ratio was slightly reduced (0.59, normal [0.6). Blood tests showed hypochromic anaemia with low levels of haemoglobin (7.4 g/dl; normal 13.4–17.1), iron (3 lmol/l; normal 14–27), and ferritin (11 lg/l; normal 27–365), but normal values for lactate dehydrogenase, bilirubin, haptoglobin, folate and vitamin B12. There was no history of learning disability, psychiatric disorders, other paroxysmal events, delay in motor development or microcephaly. His occipitofrontal circumference was at the 90th percentile . The patient was treated with methotrexate for his Crohn’s disease. He had no children. Patient 2 The 38-year-old only brother of patient 1 presented with a history of absence episodes between the ages three and 14. The episodes had similar characteristics to those of patient 1, but lasted longer and initially occurred up ten times a day with more profound impairment of consciousness. Diagnostic procedures included normal EEGs performed
J Neurol (2014) 261:2009–2015
Table 1 Summary of clinical, molecular, and diagnostic data of three patients Patient 1 (42 years)
Patient 2 (38 years)
Patient 3 (63 years)
EOAE, 3–13 years, early in the morning after awakening, duration up to 10 s, initial frequency up to 5 episodes/day, spontaneous recovery
EOAE, 3–14 years, early in the morning after awakening, duration up to 30 s, frequency up to 10 episodes/day, more profound impairment of consciousness, spontaneous recovery
Movement disorders/ other paroxysmal events
PED, since the age of 8 years, initially in the morning, later during the day following exercise or spontaneously triggered; duration up 30 min, frequency up to 6 episodes/month; no further paroxysmal events
PED, 7–45 years of age, initially in the morning, in adolescence during the day triggered by physical exertion or spontaneously; duration up to 15 min, frequency up to 1–2 episodes/month; spontaneous recovery; no other paroxysmal events
No (iron deficiency anaemia was diagnosed)
No (90th percentile: HC 60 cm, H 185 cm, W 103 kg) 
No (90th–97th percentile: HC 61 cm, H 189 cm, W 85 kg)
No (90th percentile: HC 60 cm, H 180 cm, W 120 kg)
Normal (including one study during PED episode)
Slightly reduced CSF-to-blood glucose ratio of 0.59 (normal[0.6) and normal values for glucose (2.3 mmol/l, range 1.6–3.6) and lactate (1.2 mmol/1, range 1.1–1.9)
Neurological examination between paroxysms
Failed treatment attempts
Acetazolamide, phenytoin, carbamazepine, valproate
Not applicable (no treatment attempts)
Response to clonazepam
Excellent control of PED
EOAE early onset absence epilepsy, PED paroxysmal exercise-induced dyskinesia, CSF cerebrospinal fluid, HC occipitofrontal head circumference, H height, W weight
between absence episodes. A trial of valproate was ineffective. The patient reported good scholastic performance. There was no history of ‘‘dancing fits’’ or any other GLUT1-related symptoms. His occipitofrontal circumference was between the 90th and 97th percentile. His neurological status was normal. His two daughters, 13 and 8 years of age, were described as healthy. Patient 3 The 63-year-old father of patients 1 and 2 presented with a history of ‘‘dancing fits’’ between the ages of seven and 45. They initially occurred early in the morning. In adolescence, they were triggered by physical exertion. They had a
lower frequency, shorter duration and milder severity than the fits of patient 1. With time, they became less frequent and disappeared. No diagnostic procedures were performed and the patient was never treated. There was no history of additional GLUT1-related symptoms. His occipitofrontal circumference was at the 90th percentile. His neurological status was normal. No additional affected relatives were reported. Pedigree structure is presented in Fig. 1. Molecular genetic studies Prior genetic tests showed no pathogenic mutations. Current genetic testing for mutations in the SLC2A1 gene
J Neurol (2014) 261:2009–2015
revealed a novel c.972G[A, p.S324S heterozygous variant in exon 7 at an exon–intron boundary in all three patients. This variant was not found in control individuals. RNA sequencing from blood showed no alternative splicing (data not shown). Other tissues were not available. Summary of clinical, molecular and diagnostic data of all patients is presented in Table 1.
Discussion The clinical picture and course of all three family members are consistent with GLUT1-DS. They illustrate a phenotypic diversity that in GLUT1-DS may refer to both intraand inter-familial heterogeneity caused by identical SLC2A1 mutations . ‘‘Dancing fits’’ of patient 3 fit well with ‘‘pure’’ PED. Like in most cases with GLUT1-related PED, his PED become less severe with age and even spontaneously ceased without any prior treatment attempts [1, 4]. In contrast, PED seen in patient 1 had a rare course and worsened with age, temporarily becoming very frequent . In view of absent additional symptoms in his current clinical presentation, the proband’s fits also correspond well with ‘‘pure’’ PED. Its onset was typical around the age of 8 years . Compared to previously reported GLUT1-related ‘‘pure’’ PED cases, the proband’s spells were more severe and unusual in several aspects. The episodes were not always restricted to the exercised limbs , and facial and trunk muscles were also affected . They often occurred without an apparent cause, but with prodromal symptoms. There was no asymmetry of his complaints, and dystonia was not the predominant feature [4, 17, 22]. However, his dystonic manifestation was more severe [6, 22] and did not mimic simple action dystonia [6, 17]. At first sight, the fits suggested a psychogenic aetiology. There were also some similarities with the ‘‘classical’’ GLUT1-DS phenotype, such as unpredictable fluctuation of the symptoms, without influence of fasting or food intake . Approximately 30 % of GLUT1-related ‘‘pure’’ PED cases have been sporadic and the remaining 70 % of cases come from four families [4, 6, 13, 16, 17, 22]. Treatment options for PED still remain limited  and observations on effective drugs are contradictory. In the proband, clonazepam resulted in excellent control of PED. Trials of valproate and carbamazepine failed both in the proband and other patients [13, 17], but carbamazepine was helpful in another case . Neither a ketogenic diet , nor levodopa [13, 17] have been tried in patient 1. Contrary to some reports [14, 17], acetazolamide was ineffective. Like in most GLUT1-related ‘‘pure’’ PED cases, the proband did not show any abnormalities in neurological
examination between fits. Typically, his EEG studies, including one recorded during a spell, were normal . In the overwhelming majority of patients with both ‘‘classical’’  and ‘‘non-classical’’  GLUT1-DS phenotypes, brain MRI was normal. Only in one case with ‘‘pure’’ PED, mild cerebellar involvement was seen clinically and on MRI . The corticostriate pathway, frontal lobe and cerebellum were suggested to play a role in the pathogenesis of PED . The most common clinical GLUT1-DS feature that affects approximately 90 % of individuals is seizures [7, 19]. Absence seizures are very common in the ‘‘non-classical’’ phenotype with a mean age at onset of 2 years , whereas generalised tonic–clonic seizures dominate in the ‘‘classical’’ phenotype and occur at the age of 8 months . Both brothers’ absence episodes were without an atonic or tonic component. Onset before the age of 4 years is consistent with early onset absence epilepsy (EOAE). Although generalised spike-waves on EEG were not reported , no ictal EEG was recorded. Besides, a normal interictal EEG was the most common EEG finding in GLUT1-DS . In both brothers, the seizures occurred early in the morning before breakfast, typical of GLUT1DS, due to hypoglycemia, which additionally impairs glucose transport into the brain. Approximately 12 % of EOAE are estimated to be caused by GLUT1 deficiency . Both brothers completely recovered from seizures before the age of 15 years. This clinical course confirms a previous observation that epilepsy in GLUT1-DS not infrequently becomes less severe or disappears during adolescence despite of treatment with anti-epileptic drugs or a ketogenic diet . Other patients with ‘‘pure’’ PED were also diagnosed in infancy as having myoclonic  or partial  seizures that ceased in early adolescence giving way to PED. Both the brothers had a poor response to treatment attempts with unknown doses of valproate. In one report, two patients with GLUT1-related EOAE became seizure-free following a combination therapy with valproate and lamotrigine or levetiracetam, whereas seizure freedom could not be achieved with valproate alone in another case . In most GLUT1-DS cases, there is evidence of hypoglycorrhachia, and the CSF-to-blood glucose ratio is a superior diagnostic tool to the absolute CSF glucose level. In the ‘‘classical’’ phenotype, the ratio is lower than 0.37. In milder ‘‘non-classical’’ phenotypes, especially those characterised by movement disorders without epilepsy like in our proband, the ratio can be up to 0.59 [4, 7]. All three patients were normocephalic. Although no neuropsychological tests were performed, there was no anamnestic evidence of any form of intellectual or behavioural impairment in them.
The proband was diagnosed with anaemia. However, the available blood test findings were consistent with iron deficiency anaemia that most probably resulted from his Crohn’s disease. Haemolytic anaemia is a rare GLUT1-DS feature that depends on the type of a SLC2A1 mutation (e.g., in deletion mutations ). Both our family and the previously reported familial cases with GLUT1-related ‘‘pure’’ PED showed a pattern suggesting autosomal dominant inheritance. All described cases with GLUT1-related ‘‘pure’’ PED showed heterozygous missense SLC2A1 mutations (c.283T[A (p.S95I), 284C[T (p.S95I), c.418G[A (p.V140M) ; c.998G[A (p.R333Q) ; c.493G[A (p.V165I) ; c.950A[C (p.N317T) ; c.998G[A (p.R333Q) ; c.971C[T (p.S324L) ). In our family, the novel c.972G[A, p.S324S SLC2A1 heterozygous missense variant was found in all three patients. No additional family members were either genetically tested or reported to show any GLUT1related complaints. However, asymptomatic adult SLC2A1 mutation carriers from affected families have already been identified [1, 4]. This novel variant was not present in controls and was evaluated by the MutationTaster program to be diseasecausing. It would be presumed that this variant interferes with the splice donor site and leads to aberrant splicing. However, subsequent RNA sequencing from blood failed to detect a splicing defect as the variant was expressed in a heterozygous state and no alternate amplicon sizes were observed. Possible alternative explanations include nonsense mediated mRNA decay and tissue specific expression. Besides SLC2A1 mutations, PED has been reported in a few familial and sporadic cases with unknown genetic defects [1, 4, 11, 17] as well as in two families linked to two other genes [10, 12]. In these cases, PED is mostly accompanied by additional symptoms, such as restless legs syndrome (GTP cyclohydrolase 1 gene  ) and rolandic epilepsy (autosomal recessive inheritance linked to a 6-cM region on chromosome 16 ). We conclude that PED may produce a broad range of bizarre movements, which can mimic psychogenic movement disorders. A positive family history suggests an organic aetiology. Establishing the correct diagnosis is important, since PED can effectively be treated with clonazepam. Although no splicing defect at the RNA level could be demonstrated for the novel SLC2A1 c.972G[A, p.S324S variant, together, the clinical picture (including heterogeneous phenotypes within one family), autosomal dominant inheritance, and CSF findings are consistent with a diagnosis of GLUT1-DS. Additional studies, such as exome sequencing, are indicated to exclude other genetic variations with major clinical impact.
J Neurol (2014) 261:2009–2015 Acknowledgments We thank the patients for their assistance in preparing this manuscript. Conflicts of interest disclose.
PT, SL, CS, SGS, SB, DD have nothing to
References 1. Weber YG, Storch A, Wuttke TV, Brockmann K, Kempfle J, Maljevic S, Margari L, Kamm C, Schneider SA, Huber SM, Pekrun A, Roebling R, Seebohm G, Koka S, Lang C, Kraft E, Blazevic D, Salvo-Vargas A, Fauler M, Mottaghy FM, Mu¨nchau A, Edwards MJ, Presicci A, Margari F, Gasser T, Lang F, Bhatia KP, Lehmann-Horn F, Lerche H (2008) GLUT1 mutations are a cause of paroxysmal exertion-induced dyskinesias and induce hemolytic anemia by a cation leak. J Clin Invest 118:2157–2168 2. Demirkiran M, Jankovic J (1995) Paroxysmal dyskinesias: clinical features and classification. Ann Neurol 38:571–579 3. Groffen AJ, Klapwijk T, van Rootselaar AF, Groen JL, Tijssen MA (2013) Genetic and phenotypic heterogeneity in sporadic and familial forms of paroxysmal dyskinesia. J Neurol 260:93–99 4. Suls A, Dedeken P, Goffin K, Van Esch H, Dupont P, Cassiman D, Kempfle J, Wuttke TV, Weber Y, Lerche H, Afawi Z, Vandenberghe W, Korczyn AD, Berkovic SF, Ekstein D, Kivity S, Ryvlin P, Claes LR, Deprez L, Maljevic S, Vargas A, Van Dyck T, Goossens D, Del-Favero J, Van Laere K, De Jonghe P, Van Paesschen W (2008) Paroxysmal exercise-induced dyskinesia and epilepsy is due to mutations in SLC2A1, encoding the glucose transporter GLUT1. Brain 131:1831–1844 5. Leen WG, Klepper J, Verbeek MM, Leferink M, Hofste T, van Engelen BG, Wevers RA, Arthur T, Bahi-Buisson N, Ballhausen D, Bekhof J, van Bogaert P, Carrilho I, Chabrol B, Champion MP, Coldwell J, Clayton P, Donner E, Evangeliou A, Ebinger F, Farrell K, Forsyth RJ, de Goede CG, Gross S, Grunewald S, Holthausen H, Jayawant S, Lachlan K, Laugel V, Leppig K, Lim MJ, Mancini G, Marina AD, Martorell L, McMenamin J, Meuwissen ME, Mundy H, Nilsson NO, Panzer A, Poll-The BT, Rauscher C, Rouselle CM, Sandvig I, Scheffner T, Sheridan E, Simpson N, Sykora P, Tomlinson R, Trounce J, Webb D, Weschke B, Scheffer H, Willemsen MA (2010) Glucose transporter-1 deficiency syndrome: the expanding clinical and genetic spectrum of a treatable disorder. Brain 133:655–870 6. Bovi T, Fasano A, Juergenson I, Gellera C, Castellotti B, Fontana E, Tinazzi M (2011) Paroxysmal exercise-induced dyskinesia with self-limiting partial epilepsy: a novel GLUT-1 mutation with benign phenotype. Parkinsonism Relat Disord 17:479–481 7. De Giorgis V, Veggiotti P (2013) GLUT1 deficiency syndrome 2013: current state of the art. Seizure 22:803–811 8. Leen WG, Taher M, Verbeek MM, Kamsteeg EJ, van de Warrenburg BP, Willemsen MA (2014) GLUT1 deficiency syndrome into adulthood: a follow-up study. J Neurol 261:589–599 9. Yang H, Wang D, Engelstad K, Bagay L, Wei Y, Rotstein M, Aggarwal V, Levy B, Ma L, Chung WK, De Vivo DC (2011) Glut1 deficiency syndrome and erythrocyte glucose uptake assay. Ann Neurol 70:996–1005 10. Guerrini R, Bonanni P, Nardocci N, Parmeggiani L, Piccirilli M, De Fusco M, Aridon P, Ballabio A, Carrozzo R, Casari G (1999) Autosomal recessive rolandic epilepsy with paroxysmal exerciseinduced dystonia and writer’s cramp: delineation of the syndrome and gene mapping to chromosome 16p12–11.2. Ann Neurol 45:344–352
J Neurol (2014) 261:2009–2015 11. Bhatia KP, Soland VL, Bhatt MH, Quinn NP, Marsden CD (1997) Paroxysmal exercise induced dystonia: eight new sporadic cases and a review of the literature. Mov Disord 12:1007–1012 12. Dale RC, Melchers A, Fung VS, Grattan-Smith P, Houlden H, Earl J (2010) Familial paroxysmal exercise-induced dystonia: atypical presentation of autosomal dominant GTP-cyclohydrolase 1 deficiency. Dev Med Child Neurol 52:583–586 13. Baschieri F, Batla A, Erro R, Ganos C, Cordivari C, Bhatia KP (2014) Paroxysmal exercise-induced dystonia due to GLUT1 mutation can be responsive to levodopa: a case report. J Neurol 261:615–616 14. Anheim M, Maillart E, Vuillaumier-Barrot S, Flamand-Rouvie`re C, Pineau F, Ewenczyk C, Riant F, Apartis E, Roze E (2011) Excellent response to acetazolamide in a case of paroxysmal dyskinesias due to GLUT1-deficiency. J Neurol 258:316–317 15. Arsov T, Mullen SA, Damiano JA, Lawrence KM, Huh LL, Nolan M, Young H, Thouin A, Dahl HH, Berkovic SF, Crompton DE, Sadleir LG, Scheffer IE (2012) Early onset absence epilepsy: 1 in 10 cases is caused by GLUT1 deficiency. Epilepsia 53:e204– e207 16. Mullen SA, Suls A, De Jonghe P, Berkovic SF, Scheffer IE (2010) Absence epilepsies with widely variable onset are a key feature of familial GLUT1 deficiency. Neurology 75:432–440
2015 17. Schneider SA, Paisan-Ruiz C, Garcia-Gorostiaga I, Quinn NP, Weber YG, Lerche H, Hardy J, Bhatia KP (2009) GLUT1 gene mutations cause sporadic paroxysmal exercise-induced dyskinesias. Mov Disord 24:1684–1688 18. Pong AW, Geary BR, Engelstad KM, Natarajan A, Yang H, De Vivo DC (2012) Glucose transporter type I deficiency syndrome: epilepsy phenotypes and outcome. Epilepsia 53:503–510 19. Brockmann K (2009) The expanding phenotype of GLUT1deficiency syndrome. Brain Dev 31:545–552 20. Wang D, Pascual JM, Yang H, Engelstad K, Jhung S, Sun RP, De Vivo DC (2005) Glut-1 deficiency syndrome: clinical, genetic, and therapeutic aspects. Ann Neurol 57:111–118 21. Pons R, Collins A, Rotstein M, Engelstad K, De Vivo DC (2010) The spectrum of movement disorders in Glut-1 deficiency. Mov Disord 25:275–281 22. Afawi Z, Suls A, Ekstein D, Kivity S, Neufeld MY, Oliver K, De Jonghe P, Korczyn AD, Berkovic SF (2010) Mild adolescent/ adult onset epilepsy and paroxysmal exercise-induced dyskinesia due to GLUT1 deficiency. Epilepsia 51:2466–2469 23. Nguyen AKD, Simard-Meilleur AA, Berthiaume C, Godbout R, Mottron L (2012) Head circumference in Canadian male adults: development of a normalized chart. Int J Morphol 30:1474–1480