Epilepsy Flashcards
infantile and childhood epileptic encephalopathies
a group of epilepsies,
estimated to affect ~1.2/1000 live births
typified by multiple types of seizures
within the first years of life, developmental delay and resistance to anti-epileptic drugs
Diverse underlying aetiologies and overlapping phenotypes make their investigation
challenging
In ICEE, the epileptic activity contributes to cognitive / behavioural impairments above and
beyond what might be expected from the underlying pathology alone
also such thing as epileptic and developmental encephalopathies (EDEs) where developmental impairment occurs as a direct result of the genetic variant, in addition to the effect of excessive epileptic activity
Many will have an acquired (e.g. hypoxic ischaemic injury) or easily identifiable genetic
cause (e.g. Down syndrome)
Where there is no identifiable acquired cause, the evolution of trio whole exome and whole
genome sequencing (WES/WGS), has increased the diagnostic yield significantly, although it remains less than 60%
initial clinical assessment of ICEE
meticulous history is imperative, including history of assisted conception (including donor
egg, sperm or embryos) and ante/perinatal periods. A thorough family tree should be
constructed
Parental international travel and country
of birth should be noted in view of geographical variations in infectious diseases and newborn screening programs
A full developmental assessment of all domains should be conducted including the trajectory
of developmental delay or regression. Height, weight and head circumference should be
plotted on standardised growth charts
A thorough clinical examination of all body systems should be conducted, looking for signs which may help tailor investigations. The assessment for dysmorphism should include the face, hair, nails, skin, eyes, ears, hands, feet, fat distribution and genitalia, with particular emphasis on the skin for neuro-cutaneous
syndromes
Clinicians should assess the level of intellectual disability and for the presence
of autistic spectrum disorders. All children should be referred for ophthalmological and
audiological assessment.
Abnormal clinical signs should initiate specific testing such as:
- Cutaneous stigmata or dysmorphism - referral to a geneticist;
- Spleno/hepatomegaly - referral to a metabolic specialist
- Movement disorders should prompt clinicians to consider specific metabolic
and/or genomic analyses
first line testing for ICEEs - eeg and neuroimaging
Electrophysiology- EEG- is essential in the investigation of aetiology. It may allow broader
classification into a focal or generalised epilepsy and can be consistent with electro-clinical
syndromes such as Lennox Gastaut or of a genetic disorder with specific investigation e.g.
Dravet syndrome.
The clinician should note both a standard recording, as well as whether sleep was achieved
and for older children activation techniques such as photic stimulation / hyperventilation.
Where an ictal recording has not been obtained there should be consideration for
repeat/longer recording
We recommend consideration of MRI imaging in all children, even where the aetiology is clear
on clinical grounds– e.g. tuberous sclerosis/Down syndrome; sedation or GA is justified for this where needed
MRI sequence protocol should follow that stipulated by the regional epilepsy surgery
centre. recommended that a final conclusion is not drawn until the images have been
reviewed and the sequences judged appropriate by an expert in the neuroimaging appearances
of ICEE
first line testing for ICEE - genomics
Due to the high genetic heterogeneity of ICEE, genomic testing should be implemented early
in the investigative process; start with ICEE panel, array-CGH, and storing parent samples
Array comparative genomic hybridisation (array-CGH) can be ordered once the family have
been appropriately consented. Array-CGH detects imbalances in chromosomal material.
Array-CGH can detect genomic imbalances as small as 100,000 base pairs, however it is unable
to detect pathogenic single or multi-nucleotide variants, triplet repeat expansions and
methylation defects. Array-CGH may also miss large copy number variants in the presence of
low level mosaicism and will never detect a copy number variant from a blood sample if the
abnormal cell line is no longer present in lymphocytes. Array-CGH has been estimated to
provide a definitive diagnosis in <5% of cases of early infantile epileptic encephalopathy
(EIEE), therefore we recommend ordering next generation sequencing (NGS) and array-CGH
concomitantly
NGS using exome sequencing, limiting the interpretation of the exome to a targeted panel of genes known to cause ICEE, has helped to further increase the diagnostic yield although this remains <60%
recommended to parental DNA alongside first line investigations, as we have had
difficulty obtaining both parental samples later and note the importance of trio analysis to
improve diagnostic yield
first line testing for ICEE - biochem
Genomic and biochemical testing for metabolic conditions may yield false negative/positive
results and so should be performed concurrently
First line tests aim to exclude basic and reversible metabolic derangement(s), assist in tailoring further investigation, screen for rarer metabolic conditions and prioritise those which are treatable
Full blood count
* U+E, LFT, blood
gas/bicarbonate,
glucose, calcium,
magnesium,
ammonia, plasma
lactateΔ
* Plasma amino
acidsΔ
, biotinidase
* Acylcarnitines
* Urine organic
acidsΔ
* Urine and plasma
creatine and
guanidinoacetat
Cerebral creatine deficiency syndromes may present in children with seizures, intellectual
disability and sometimes movement disorders. Treatment of guanidinoacetate
methyltransferase deficiency has been shown to reduce or eliminate seizures in 67% of patients,
while current treatment of creatine transporter deficiency has shown improvement in only a
few individuals
mito disease may show up in lactate and urine organic acid profile
Patients with biotinidase deficiency show an excellent response to oral biotin and delay in
treatment leads to irreversible neurological disease; Dermatitis and alopecia may be
present. Since oral biotin supplementation does not affect biotinidase activity, treatment can be
commenced pending the biotinidase result
CSF can assess for f mitochondrial disease
(simultaneous CSF/plasma lactate), GLUT-1 deficiency (simultaneous CSF/plasma glucose),
phosphoglycerate dehydrogenase deficiency, non-ketotic hyperglycinaemia (paired CSF and
plasma amino acids/serine/glycine) and vitamin dependant epilepsies
Treatment with pyridoxine/pyridoxal phosphate, folinic acid and biotin should be initiated in
children <1year of age, except those with infantile spasms (IS) alone; Where there is no improvement, they should be
stopped after 14 days, awaiting confirmatory negative biochemical/genomic analyses; CSF should ideally be obtained prior to commencing pyridoxal phosphate/pyridoxine and
folinic acid (but treatment should not be unnecessarily delayed); For pyridoxine
dependant epilepsy urinary alpha-aminoadipic semialdehyde (α-AASA), should remain
positive post treatment, as an alternative diagnostic test
These treatable conditions screened for using CSF analysis are unlikely to present with IS alone or >1year of age and are likely to be detected using WES/WGS, thus weighed against the risks of general anaesthesia/processing difficulties, CSF analysis should be reserved as a
second line test in children >1 year of age
finally urine sulfocysteine will assess for sulfite oxidase and molybdenum co-factor deficiencies.
There have been reports of clinical improvement in type A molybdenum cofactor deficiency
following intravenous cyclic pyranopterin monophosphate
second line testing for ICEE - eegs, neuroimaging, and genetics
eegs: check that both an ictal recording as well as sleep has been obtained. Where
this has not been achieved or where there remains diagnostic uncertainty, recommend to do a repeat/prolonged recording
Before repeating MRI, clinicians should discuss with an expert in the radiology of ICEE if all
appropriate sequences were performed, their quality is sufficient and if another imaging
modality is needed e.g. CT to show calcification in tuberous sclerosis.
If these are deemed unremarkable, repeat scan would be indicated where the initial MRI was
performed within the first two years of life, when lack of myelin can mask radiological
stigmata of ICEE. If the original scan is of high quality and was performed after the age of
three years, clinicians should consider repeat MRI at 3T, which may identify subtle
abnormalities e.g. cortical dysplasias, which were not evident on the original series
genetics: Trio WES or WGS improves diagnostic speed and yield over that attained with an epilepsy panel; ‘Trio’ describes the method of genomic analysis whereby the child’s genomic
variants are compared to the parental variants using a specific bioinformatic approach. The
analysis may be applied to the entire exome or genome, rather than being limited to a gene
panel, hence it is sometimes referred to as ‘gene agnostic’. It allows rapid identification of de
novo changes (which cannot be clarified without parental analysis) now recognised to cause a
large proportion of serious paediatric genetic disorders; recommended to refer to clinical genetics in all cases where an acquired cause is unlikely.
They would review the likely implications of results, and whether variants are likely to be
disease causing. At present, paediatricians are unlikely to have access to trio WES/WGS, but
can order ICEE panels. Currently, clinical geneticists have limited access to trio (agnostic)
WES and can initiate this testing if needed
second line testing for ICEE - biochem
CSF studies where not already done
Thyrotoxic and Hashimoto’s encephalopathy are rare and should be associated with other
clinic signs, thus thyroid function testing is included as a second line investigation unless
there are clinical reasons to test sooner e.g. maternal history of Graves’ disease
Transferrin glycoform testing for congenital disorders of glycosylation, which often present
with a multisystem process and are largely without treatment, is reserved as second line.
Transferrin glycoforms are unreliable in the first three weeks of life due to the influence of
maternal transferrin.
Very long chain fatty acid analysis to test for peroxisomal disorders such as Zellweger’s
syndrome, have been selected as second line in view of their rarity, likelihood of other
clinical features predicating analysis e.g. dysmorphic features and lack of efficacious
treatment. This will also detect X-linked adrenoleukodystrophy, which in a small proportion of boys can initially present with seizures alone, but are likely to have a suggestive MRI on first line testing
third line testing for ICEE
Where first and second line testing fail to identify an aetiology, we recommend careful
review of results and further discussion with the family. Without new findings on history,
examination and review of previous investigations, the yield will be low. There can be risks to a procedure e.g. muscle biopsy under general anaesthetic and difficulty interpreting the
results. Muscle biopsy can be considered; it has a low, but not negligible risk and is unlikely
to provide a specific diagnosis/therapy, with an estimated additional diagnostic yield of
1%
We recommend careful on-going review to ensure the child does not have features of a
degenerative, potentially treatable, condition e.g. late infantile neuronal lipofuscinosis. We
have recommended many, but not all possible metabolic investigations, and at this stage
others should be considered; e.g. lysosomal enzymes to identify lysosomal storage disorders
(many, but not all will be detected by trio WGS)
ICEE investigation where even third line has failed
first careful re-analysis of the case to ensure this is correct. With so many investigations, results can be mis-documented or misunderstood. Peer review, including experts in the neurophysiology, neuroimaging and genomics of ICEE is essential
If a diagnosis is not forthcoming, we expect significant improvements in yield of both
genomic and other investigations at least every 2 years, so consideration to
review the process in that time frame is appropriate and should include assessment of
advances in technology and bioinformatics, knowledge of the genetic aetiology of
conditions/specific phenotypes, the patients emerging phenotype and changes in the family
history (e.g. newly affected members)
Where the aetiology remains unclear, paediatricians should reassure families that they can still treat the epilepsy with evidence-based therapies. Families should receive
multidisciplinary input from allied health professionals, including psychology, education and specialist therapists
managing treatment resistance ICEE where aetiology identified
first step should be to challenge the diagnosis; particularly HIE. We recommend instituting appropriate first-line therapies for the relevant diagnosis whilst discussing with families the risk versus benefits of the above tiered testing. This will vary between cases for example there is over diagnosis of HIE in children with underlying disorders e.g. pyridoxine dependent epilepsy. Conversely, in children with Down syndrome presenting with infantile spasms, treatment resistance is common, other causation extremely unlikely and investigation potentially traumatic
ICEE UMP therapy for CAD
(UMP) has been demonstrated as an effective and safe treatment for children with CAD (carbamoyl-phosphate synthetase 2, aspartate
transcarbamylase and dihydroorotase) deficiency
Some have suggested a 6 month trial of uridine in all cases of neonatal seizures and children with developmental delay, seizures and anisopoikilocytosis/anaemia, pending negative genetic results
not incorporated this into our standard recommendations due to the lack of long-term data. However, clinicians should engage in careful discussion with geneticists regarding CAD deficiency and related VUS, and consider uridine supplementation pending genetics,
particularly in refractory ICEE or where the phenotype is indicative
glutamate and glut transporters
Glutamate and aspartate are non-essential amino acids that do not cross the blood-brain barrier. They are synthesised from glucose and a variety of other precursors within the brain. Synthetic and metabolic enzymes for glutamate and aspartate have been localised to neurons and glial cells. (Glutamic acid is in a metabolic pool with a-ketoglutaric acid and glutamine.) A large fraction of the glutamate released from nerve terminals probably is taken up into glial cells, where it is converted into glutamine. Glutamine then cycles back to nerve terminals, where it participates in the transmitter pools of glutamate and GABA; transmitter pool of glutamate is stored in synaptic vesicles that actively accumulate glutamate through a Mg2+/ATP-dependent process. Substances that destroy the electrochemical gradient inhibit this uptake mechanism. The concentration of glutamate within synaptic
vesicles is thought to be very high >20mM; glut/asp are zwitterions so cant cross PM; Uptake
mechanisms have an important role in regulating the extracellular concentrations of glutamate and aspartate in the brain. At least two families of
glutamate transporter have been localised to the plasma membrane of neurons and astrocytes. Only the Na+-dependent glutamate transporter is coupled to the electrochemical gradient that
permits transport of glutamate and aspartate against their concentration gradients; n the CNS, glutamate transporter-1 (GLT-1) and glutamate-aspartate transporter (GLAST) are expressed preferentially in glial cells, GLT-1 has its highest expression in the thalamus and cerebellum, with lower levels in the hippocampus, cortex and striatum. GLAST immunoreactivity is found
predominantly in the cerebellum and less so in the forebrain. Excitatory amino acid carrier-1 (EAAC1) is
expressed predominantly in neurons, most prominently in the hippocampus and not in glial cells; glut in brain tissue at 1-2mM
glut transporters major role limit glut/asp conc in ecf to prevent excessive GLUTr stim; Net transport of glutamate is increased by high intracellular K+; upon dissociation of glutamate and Na+ from the transport machinery, cytoplasmic K+ binds to
the protein to be recycled into the extracellular compartment - 3Na, 1H, 1 glut in and 1 K out; OH or HCO3 accompanies K out
AED mechanisms - NaV
e native sodium channel comprises a single alpha-subunit protein, which contains the poreforming region and voltage sensor, associated with one or more accessory beta-subunit
proteins which can modify the function of the alpha-subunit but are not essential for basic
channel activity. There are four predominant sodium channel alpha-subunit genes expressed
in mammalian brain, denoted SCN1A, SCN2A, SCN3A and SCN8A, which encode the
channels Nav1.1, Nav1.2, Nav1.3 and Nav1.6, respectively. These channels are expressed
differentially in the nervous system. Nav1.3 expression is mainly restricted to the early stages
of development, while Nav1.1 is the major sodium channel in inhibitory interneurons and
Nav1.2 and Nav1.6 are expressed in the AIS of principal excitatory neurons - former more in immature brain, latter predominates later
phenytoin and
carbamazepine are archetypal sodium channel blockers, a mechanism they share with the
newer drugs, lamotrigine, felbamate, topiramate, oxcarbazepine, zonisamide, rufinamide,
lacosamide, and eslicarbazepine acetate. There is also anecdotal evidence to suggest that
sodium valproate and gabapentin have inhibitory effects on neuronal sodium channels
Antiepileptic agents with sodium channel blocking properties have
highest affinity for the channel protein in the inactivated state and binding slows the
conformational recycling process. As a result, these drugs produce a characteristic voltageand frequency-dependent reduction in channel conductance, resulting in a limitation of
repetitive neuronal firing, with little effect on the generation of single action potentials.
Further complexity is added by the existence of multiple inactivation pathways. Although
most sodium channel blocking AEDs target the fast inactivation pathway, lacosamide appears
to enhance slow inactivation and there is preliminary evidence to suggest that eslicarbazepine
acetate may do likewise. The clinical implications of this distinction remain unclear but it has
been proposed that the slow inactivation pathway is more prominent during prolonged
depolarisation, as might be expected during epileptiform discharges
AED mechanisms - CaV
voltage-gated
calcium channels comprise a single alpha-subunit, of which at least seven are known to be
expressed in mammalian brain. There are also accessory proteins, including beta- and alpha2-
delta-subunits, that modulate the function and cell-surface expression of the alpha-subunit
but which are not necessarily essential for basic channel functionality. Voltage-gated calcium
channels are commonly distinguished on the basis of their biophysical properties and patterns
of cellular expression. High-voltage-activated (HVA) channels respond to strong
depolarisations and are involved in both pre-synaptic neurotransmitter release (N-, P/Q-, and
R-type) and the processing of synaptic inputs at the somatodendritic level (L-type). In
contrast, the low-voltage-activated (LVA) channel opens in response to modest
depolarisations at or below resting membrane potential and gives rise to transient (T-type)
currents which participate in intrinsic oscillatory activity. The T-type channel is highly
expressed on the soma and dendrites of thalamic relay and reticular neurones where it has
been postulated to underpin the rhythmic 3 Hz spike-wave discharges that are characteristic
of absence seizures
efficacy of ethosuximide and zonisamide in generalised
absence epilepsy is believed to be mediated by blockade of the LVA T-type calcium channel
in the soma and dendrites of thalamic relay and reticular neurones. There is anecdotal
evidence that sodium valproate may have a similar action. Lamotrigine limits
neurotransmitter release by blocking both N- and P/Q-types of the HVA calcium channel and
levetiracetam exerts a partial blockade of N-type calcium currents, suggesting a selective
effect on an as yet unidentified sub-class of this particular channel type, although its main mechanism is through binding SVA2 to modulate synaptic vesicular release of NTs. Phenobarbital,
felbamate, and topiramate are also believed to influence HVA calcium channel conductance
gabapentin and pregabalin also exert their effects via
HVA calcium channels, but rather than interacting with a traditional channel sub-type such
as N- or L-type, they appear to bind to an accessory subunit termed alpha2-delta-1, which can
modulate the function of various native channels. This subunit is upregulated in dorsal root
ganglion cells of the spinal cord in response to nerve injury, with selective calcium channel
blockade via the alpha2-delta-1 subunit explaining the efficacy of gabapentin and pregabalin
in the treatment of neuropathic pain
AED mechanisms - GABA
GABAA receptor is a ligand-gated ion channel,
comprising five independent protein subunits arranged around a central anion pore permeable
to chloride and bicarbonate. Nineteen GABAA receptor subunits have been identified to date which come together
as heteromeric pentamers to form functional channels. GABAA receptors mediating transient,
rapidly desensitising currents at the synapse (phasic receptors) typically comprise two alpha-,
two beta-, and one gamma-subunit, whereas those at extra-synaptic sites and mediating longlasting, slowly desensitising currents (tonic receptors) preferentially contain alpha4- and alpha6-subunits and a delta-subunit in place of the gamma-subunit. In contrast, the GABAB
receptor is coupled, via a G-protein, to potassium channels which mediate slow
hyperpolarisation of the post-synaptic membrane. This receptor is also found pre-synaptically
where it acts as an auto-receptor, with activation limiting further GABA release. GABA is
removed from the synaptic cleft into localised nerve terminals and glial cells by a family of
transport proteins, denoted GAT-1, GAT-2, GAT-3, and BGT-1. Thereafter, GABA is either
recycled to the readily releasable neurotransmitter pool or inactivated by the mitochondrial
enzyme GABA-transaminase
Activation of the ionotropic GABAA receptor resulting in an enhanced
response to synaptically released GABA is a major AED mechanism. Barbiturates (e.g.
phenobarbital, primidone) and benzodiazepines (e.g. diazepam, clobazam, clonazepam) share
this effect, but they bind to distinct sites on the receptor complex and differentially influence
the opening of the chloride ion pore. All GABAA receptors containing at least one alpha- and
one beta-subunit appear susceptible to activation by barbiturates, Benzodiazepine-sensitive GABAA receptors are typically composed of two alphasubunits (alpha1, alpha2, alpha3 or alpha5), two beta-subunits (beta2 or beta3), and a
gamma2 subunit, whereas the delta-containing GABAA receptor which mediates tonic
inhibition is entirely insensitive to benzodiazepines, as are those containing alpha4- and
alpha6-subunits. Functionally, barbiturates increase the duration of chloride channel opening,
while benzodiazepines increase the frequency of opening
Vigabatrin is an irreversible inhibitor of the mitochondrial enzyme GABA-transaminase which is responsible for the catabolism of GABA, whereas tiagabine prevents
the removal of GABA from the synaptic cleft by blockade of GABA transport
Other antiepileptic agents, including
sodium valproate, gabapentin and topiramate have also been reported to influence GABA
turnover
linking AED mechanism to efficacy against different seizure types
One of the more surprising aspects of AED pharmacology is the apparent lack of a direct
relationship between mode of action and efficacy. It is, however, possible to make the
following broad generalisations regarding spectrum of activity. Selective sodium channel
blockers (i.e. carbamazepine, phenytoin, oxcarbazepine, eslicarbazepine acetate) and
selective HVA calcium channel blockers (i.e. gabapentin, pregabalin) tend to have efficacy
against partial and primary generalised tonic-clonic seizures alone and are generally inactive
against or can exacerbate most other generalised epilepsies. This characteristic is shared with
selective GABA turnover drugs (i.e. vigabatrin, tiagabine) but interestingly not with selective
GABAA receptor drugs (i.e. phenobarbital, benzodiazepines) which are active in several
generalised epilepsy syndromes. Any compound that exerts its effects by blockade of T-type
LVA calcium channels, either wholly (i.e. ethosuximide) or in part (i.e. zonisamide), is likely
to be effective against absence seizures and drugs with multiple mechanisms of action (i.e.
sodium valproate, topiramate, levetiracetam, zonisamide) tend to be broad spectrum with
efficacy against a wide range of seizure types and in multiple syndromes
