sign in sign up
<<
Home , Biotin, Pyridoxal phosphate, Pyridoxine

PERSPECTIVES 659

Neonatal first few days of life after birth and are ...... intractable to drugs. EEG is abnormal with no normal background activity and bursts of spike-wave dis- Treatable neonatal epilepsy charges. Neuroimaging is normal at the onset. There is no response to intravenous Robert Surtees, Nicole Wolf . However, there is a prompt and lasting response to oral pyridoxal phos- ...... phate (hence the alternative name of pyridoxal -responsive ) Perspective on the paper by Bok et al (see page 687) given at a dose of 10 mg/kg. Like PDE, treatment of PNPO deficiency with pyr- reatable metabolic causes of early epilepsy the first dose of pyridoxine can idoxal phosphate can cause transient onset epilepsy (in the first few also cause cerebral depression, which is cerebral depression. Again treatment is Tmonths of life) are uncommon but more likely if the infant is receiving antic- life-long, requiring at it is important to diagnose them because onvulsant drugs. Children with PDE a dose of 30–50 mg/kg/day. With prompt delay in specific treatment commonly require life-long treatment with pyridoxine treatment the prognosis is generally good. results in poor neurological and cognitive at a dose of ,15 mg/kg/day up to 500 mg/ However, if left untreated, most cases of outcome. Indeed, some of these day. With prompt treatment prognosis is PNPO deficiency are fatal in the first year, are fatal if left untreated. The disorders generally good, although there may be and rare survivors have severe neurological are listed in table 1 and are divided into later learning difficulties, particularly with handicap and profound brain atrophy.56 -responsive and other metabolic language. If treatment is delayed by Some biochemical markers of PNPO epilepsies. months or years, children develop severe deficiency are evident in cerebrospinal 5 We are not going to discuss all of these learning difficulties, four limb motor dis- fluid and plasma. Because disorders, but will confine ourselves to order and sensory disturbances. crosses from cerebrospinal fluid into those presenting at or very close to birth. The major cause of PDE has been neural cells as phosphorylated esters of The investigation and treatment of the recently identified2 as being due to muta- pyridoxine and as well as other epilepsies are summarised in table 2. tions in the ALDH 7A1 gene which pyridoxal, PNPO is necessary to convert encodes a central nervous system a- the pyridoxine phosphate and pyridoxa- PYRIDOXINE-DEPENDENT aminoadipic semialdehyde dehydrogen- mine phosphate to the active EPILEPSY ase. This results in accumulation of a- pyridoxal phosphate. Pyridoxal phos- Typical neonatal onset pyridoxine-depen- aminoadipic semialdehyde (a-AASA). a- phate is the cofactor for over 100 dent epilepsy (PDE) presents in the first AASA is in reversible equilibrium with involved in and amine meta- few days of life with multiple piperideine-6-carboxylate which can con- bolism, three of which are L-aromatic types which are intractable to anticonvul- dense with pyridoxal phosphate and amino acid decarboxylase (AAAD), the sant drug treatment. Experienced inactivate its cofactor activity. The build- cleavage system and threonine mothers may also recognise seizures in up of a-AASA and its spill-over into dehydratase. Dysfunction of the latter utero from about 20 weeks of gestation. plasma and urine make it a specific and two enzymes results in a mild increase Approximately one third of infants with sensitive marker of PDE caused by a- in the concentrations of glycine and aminoadipic semialdehyde dehydrogen- threonine, respectively, in both plasma PDE will also have features of a neonatal 2 (hyperalertness, irritabil- ase deficiency. Another metabolite more and cerebrospinal fluid. AAAD dysfunc- ity and a stimular sensitive startle) which proximal in this degradation path- tion causes a reduction in cerebrospinal can be accompanied by systemic features. way is pipecolic acid, and this is also fluid concentrations of homovanillic acid Neuroimaging may show structural brain raised in PDE but is less specific and (the stable acidic metabolite of dopa- 3 abnormalities, hydrocephalus or haemor- sensitive than a-AASA. Both markers mine) and 5-hydroxyindoleacetic acid rhage. EEG is abnormal but has no remain raised after treatment with pyr- (the stable acidic metabolite of ) diagnostic features.1 The seizures and idoxine. This is demonstrated in this issue and an increased concentration of 3- EEG abnormalities respond promptly of ADC by Bok et al who have shown that methoxytyrosine. These biochemical (within minutes) to 100 mg of intrave- all patients in the Netherlands with abnormalities are not present in all 6 nous pyridoxine. However, in about 20% clinically definite PDE (according to patients. However, if they are present, 4 of infants with pyridoxine-dependent Baxter’s criteria ) and six of eight patients they reverse with treatment. with clinically probable or possible PDE The birth incidence of PNPO deficiency had raised urinary and plasma a-AASA is not known. Table 1 Treatable metabolic causes concentrations (one patient was not of early onset epilepsy tested and one was negative). In these -RESPONSIVE patients plasma, but not urinary, concen- Vitamin-responsive epilepsies SEIZURES trations of pipecolic acid were also raised. Pyridoxine-dependent epilepsy Folinic acid responsive seizures is the least Bok et al also calculate that PDE caused by Pyridox(am)ine phosphate oxidase well recognised cause of vitamin-responsive deficiency (pyridoxal phosphate- a-aminoadipic semialdehyde dehydro- neonatal epilepsy.78Infants present in the responsive epilepsy) genase deficiency has a birth incidence Folinic acid-responsive seizures first few days of life with multiple seizure in the Netherlands of at least 1:276 000. deficiency types which are intractable to anticonvul- Other metabolic epilepsies sant drugs. They may also have features of a Glucose transporter 1 deficiency deficiency syndromes PYRIDOX(AM)INE PHOSPHATE neonatal encephalopathy. EEG shows Creatine deficiency syndromes OXIDASE DEFICIENCY abnormal background activity with multi- Untreated phenylketonuria Pyridox(am)ine phosphate oxidase focal spike-wave complexes but no diag- (PNPO) deficiency presents as foetal dis- nostic features. Neuroimaging is normal. tress in late . Seizures start in the There can sometimes be a transient

www.archdischild.com 660 PERSPECTIVES

Table 2 Investigation and treatment of metabolic causes of early onset epilepsy

Disorder Investigation Treatment

Biotinidase deficiency Trial 5 mg twice daily Biotin 5–10 mg twice daily Plasma biotinidase activity GLUT1 deficiency CSF glucose ,2.2 mM, CSF/plasma glucose ,0.5 Ketogenic diet Red cell glucose transport Mutation analysis of SLC2A1 gene Serine deficiencies Reduced fasting plasma serine (and variably glycine) Serine 200–600 mg/kg/day Reduced CSF serine and glycine (glycine 200–300 mg/kg/day) Fibroblast 3-phosphoglycerate dehydrogenase or phosphoserine aminotransferase activity Mutation analysis of PHGDH and PSAT1 genes Creatine deficiencies Urine creatine/creatinine ratios, urinary guanidinoacetate/creatinine ratio GAMT deficiency: creatine 500–2000 mg/kg/ Reduced CSF creatine day, arginine-restricted diet with Reduced cerebral creatine plus creatinine (1H-MRS) supplementation (100 mg/kg/day). Fibroblast guanidinoacetate methyltransferase activity Mutation analysis of GAMT, AGAT and SLC6A8 genes Untreated PKU Plasma phenylalanine Phenylalanine-restricted diet Mutation analysis of PAH gene

1 AGAT, L-arginine:glycine amidinotransferase; CSF, cerebrospinal fluid; GAMT, guanidinoacetate methyltransferase; GLUT1, glucose transporter 1; H-MRS, proton magnetic resonance spectroscopy. response to pyridoxine. The seizures calcium and magnesium disturbance available, pyridoxine should be given as promptly respond to folinic acid at a dose have been excluded or corrected, we first line therapy. It must, however, not of 2.5–5 mg twice daily. However, seizures believe the next line of investigation and be forgotten that unsuccessful treatment can recur later, sometimes responding to treatment in neonatal epilepsy is to with pyridoxine does not exclude PNPO increases in folinic acid alone (up to 8 mg/ initiate a vitamin trial. We suggest giving deficiency. kg/day) and sometimes also requiring antic- two doses of pyridoxal phosphate (10 mg/ Thirdly, delayed diagnosis of these onvulsant drug therapy. If left untreated, kg/dose) 2 h apart, and, if the epilepsy conditions can cause severe neurological folinic acid-responsive seizures is a fatal persists, two doses of folinic acid (5 mg) handicap and early death. disorder. Even with treatment and control 6 h apart. EEG monitoring is helpful but Arch Dis Child 2007;92:659–661. of the epilepsy, there is appreciable mortal- not mandatory. If there is no response to doi: 10.1136/adc.2007.116913 ity and survivors have global learning the , anticonvulsant therapy difficulties. should be introduced and further inves- ...... Folinic acid-responsive seizures does tigations into the cause of the epilepsy Authors’ affiliations have a biochemical marker in cerebrospinal carried out. Robert Surtees, Neurosciences Unit, UCL fluid, but its nature is unknown. It is There are three lines of reasoning Institute of Child Health, London, UK detected by high performance liquid chro- behind this view as follows. Nicole Wolf, Department of Paediatric matography with electrochemical detection Firstly, it has recently become clear that Neurology, University Children’s Hospital, under the conditions required to measure there are physiological reasons why antic- Heidelberg, Germany homovanillic acid and 5-hydroxyindolea- onvulsant drugs which act as GABA- cetic acid. This unknown compound is not agonists might not be as effective in Correspondence to: Professor Surtees, The normally present in cerebrospinal fluid and newborns as in older children and adults. Wolfson Centre, Mecklenburgh Square, London WC1N 2AP, UK; [email protected] decreases once treatment with folinic acid These anticonvulsant drugs include phe- has been initiated. nobarbitone and benzodiazepines. There Competing interests: None. The birth incidence of folinic acid- is also increasing evidence from animal responsive seizures is not known. studies that phenobarbitone, phenytoin and the benzodiazepines are toxic to the REFERENCES APPROACH TO DIAGNOSIS newborn brain, causing increased neural 1 Baxter P, Griffiths P, Kelly T, et al. Pyridoxine- There are two possible approaches to dependent seizures: demographic, clinical, MRI and apoptosis. These findings suggest that psychometric features, and effect of dose on diagnosing these treatable neonatal epi- standard anticonvulsant treatment of intelligence quotient. Dev Med Child Neurol lepsies. The first is to give the relevant neonatal seizures may not be as effective 1996;38:998–1006. vitamins and observe the response. In our or as safe as previously believed.910 2 Mills PB, Struys E, Jakobs C, et al. Mutations in antiquitin in individuals with pyridoxine-dependent view, this is the correct and also the Secondly, there is no biochemical or seizures. Nat Med 2006;12:307–9. easiest approach. The second method is to chemical reason to believe that pyridoxal 3 Plecko B, Hikel C, Korenke GC, et al. Pipecolic acid investigate the biochemical markers for phosphate will not be as effective as as a diagnostic marker of pyridoxine-dependent each condition. This will require the help epilepsy. Neuropediatrics 2005;36:200–5. pyridoxine in the treatment of PDE. PDE 4 Baxter P. Epidemiology of pyridoxine dependent of a specialised (often highly specialised) can be distinguished from PNPO defi- and pyridoxine responsive seizures in the UK. Arch metabolic laboratory. Because this ciency later by measuring urinary a- Dis Child 1999;81:431–3. approach will take time and typical 5 Mills PB, Surtees RA, Champion MP, et al. AASA excretion, and a firm diagnosis Neonatal epileptic encephalopathy caused by biochemical abnormalities will not be made by mutation analysis of DNA. It is mutations in the PNPO gene encoding present in all cases,6 we advise treatment also important to remember that a- pyridox(am)ine 59-phosphate oxidase. Hum Mol before starting such investigations. aminoadipic semialdehyde dehydrogen- Genet 2005;14:1077–86. 6 offmann GF, Schmitt B, Windfuhr M, et al. ase deficiency is not the only cause of Pyridoxal 59-phosphate may be curative in early- APPROACH TO NEONATAL PDE, and a small proportion of cases onset epileptic encephalopathy. J Inherit Metab Dis EPILEPSY will be missed by biochemical analysis 2007;30:96–9. 7 Torres OA, Miller VS, Buist NM, et al. Folinic acid- Once precipitating conditions such as unless pyridoxal phosphate is tried. If responsive neonatal seizures. J Child Neurol infection, hypoglycaemia or electrolyte, pyridoxal phosphate is not easily 1999;14:529–32. www.archdischild.com PERSPECTIVES 661

8 Nicolai J, van Kranen-Mastenbroek VH,Wevers RA, et 9 Rennie J, Boylan G. Treatment of neonatal seizures. 10 Bittigau P, Sifringer M, Ikonomidou C. Antiepileptic al. Folinic acid-responsive seizures initially responsive Arch Dis Child Fetal Neonatal Ed drugs and apoptosis in the developing brain. to pyridoxine. Pediatr Neurol 2006;34:164–7. 2007;92:F148–50. Ann N Y Acad Sci 2003;993:103–14.

Children’s medicines available to prescribers, although not ...... always for the most important medi- cines.14 15 The largest category considered for paediatric studies was cardiovascular ‘‘Bonne Anne´e’’, ‘‘Gutes Neues Jahr’’? drugs (n = 36), including 22 anti-hyper- tensives, suggesting choice was not due to Will 2007 be a ‘‘Happy New Year’’ for paediatric need.15 The financial gains were higher for ‘‘blockbuster’’ drugs since children’s medicines in Europe? the 6 months’ extension of patent protec- tion applied to all formulations of the T Stephenson drug type, whether appropriate for pae- diatric use or not. The current US regula- ...... tions have a ‘‘sunset’’ clause in 2007 unless renewed by Congress. New European legislation has the potential to have an enormous The US measures brought little benefit impact on how paediatric medicines are studied and used to European children. International com- panies appear unwilling to voluntarily submit US data to support authorisation ew European Union (EU) paedia- medicine is given as a liquid whereas for children in the EU, presumably tric regulations to encourage more the licence is for a tablet) or outside the because of the lack of financial incentives research on children’s medicines terms of the licence (off-label; eg, a N 1 in Europe. Progress in Europe appeared became law on 26 January 2007. The different dose for a different age group). unlikely without legislation. main proposals are as follows. First, there Half of prescriptions for children and 90% will be incentives to encourage the of prescriptions for neonates are for drugs pharmaceutical industry to test products which have not been licensed for that THE NEW EUROPEAN LEGISLATION for use in children. Secondly, the use.6–8 Every day, paediatricians face the The key measures included in the new European Agency for the Evaluation of dilemma of whether it is more unethical European paediatric regulations cover Medicinal Products (EMEA) will host an to prescribe a drug that has not been both older and newer medicines. expert paediatric committee to oversee tested or to deny treatment to a sick child. and assess paediatric investigation plans Whether a medicine is beneficial and For newer medicines for developing medicines for children. safe in children is best determined by All applications for a marketing author- Thirdly, a European clinical trials network studying it in children. Where the market isation for a new medicine, including will be set up to foster collaboration on is large (eg, antibiotics) or the price is orphan medicines, must contain the paediatric studies throughout the EU. high (eg, surfactant), companies do con- results of all studies and information Fourthly, medicines licensed for paedia- duct well planned studies in children and required in a previously agreed paediatric tric use should be sold with a special label secure licences, suggesting that the investigation plan (PIP). The PIP will indicating that fact. industry can overcome ethical pro- contain a full proposal of all the studies This legislation has the potential to 910 blems and litigation risks if the price (and their timings) necessary to support have an enormous impact on how pae- 11 is right. However, because of the cost of the paediatric use of an individual pro- diatric medicines, both old and new, are 12 developing drugs and the small market duct and will cover all paediatric age studied and used. Is this good news for in children, recouping research costs groups and all necessary age-appropriate Europe’s 140 million children? before the patent expires is uncertain. formulations. This is the default position of the regulation. THE DRIVERS FOR THE NEW THE US EXPERIENCE There is a system of waivers for EUROPEAN LEGISLATION The ‘‘paediatric exclusivity’’ provision in medicines unlikely to benefit children so Evidence-based prescribing for children is the Food and Drug Administration that children are not studied unnecessa- compromised by lack of sufficient data on Modernisation Act 1997 provided an rily. There is also a system of deferrals of many drugs.2–4 Although there is an incentive (6 months added to market the requirement to study the childhood elaborate system for testing drugs before exclusivity or patent protection on the population to ensure that medicines are they are used in adults, many drugs given active drug) for companies who per- tested in children only when it is safe to to children have not been tested in this formed clinical studies in children.13 The do so and to prevent the requirements population at all, even though children incentive was granted irrespective of delaying the authorisation of medicines can react differently to drugs.5 Moreover, whether the results demonstrated safety for adults. In this case, the paediatric data it is often financially unrewarding to and efficacy in children or, equally will be provided after the initial market- conduct research on medicines for chil- importantly, suggested the drug should ing authorisation has been granted. But dren because the children’s market is not be used in children. unless a deferral or a waiver has been smaller than that of adults. As of 2004, the data generated had led granted, paediatric data must be provided If evidence for the safety, efficacy and to revised labelling (ie, new dose, route or in all applications for the authorisation of acceptable risk/benefit of a drug exists, indication) for 63 medicines and 661 new medicinal products. This applies then that drug can be licensed for use as a studies had been requested. The incen- from 26 July 2008. medicine with indications, doses and tives have therefore been successful in the The requirement also applies to side-effect warnings. However, prescrib- US in stimulating paediatric studies and applications to add a new indication, ing can also be unlicensed (eg, the ensuring that more paediatric data are new formulation or new route of

www.archdischild.com