North West Clinical Neuroscience Partnership
The Management of Adults
with
Epilepsy
A North West Clinical Framework
August 2001
(Final version)
Contents
Section 1Introduction
The North West Clinical Frameworkpage 4
Section 2The Background
Incidence and prevalence of epilepsypage 5
Diagnosispage 5
Investigationpage 6
Classification of epilepsypage 6
Causationpage 9
Status epilepticuspage 9
Sudden deathpage 10
Medicationpage 10
Epilepsy in children and adolescentspage 11
Epilepsy in womenpage 12
Epilepsy in older peoplepage 13
Complex epilepsypage 14
Learning disabilitypage 14
Intractable epilepsypage 14
Surgical treatmentpage 15
Non-epileptic attack disorderpage 16
Impact of epilepsypage 16
Informationpage 17
Patterns of carepage 17
Summarypage 17
Section 3The Management of People with Epilepsy:
Clinical Guidelines
Introductionpage 19
Diagnosispage 19
Not epilepsypage 21
Uncertainpage 22
Epilepsypage 22
Treatmentpage 23
Starting antiepileptic drug therapypage 23
Management and follow-uppage 25
Diagnosis uncertainpage 25
Epilepsypage 26 Patient in remission with no
side effects of treatmentpage 26
Patient having side effects of
therapy or not in remissionpage 27
Refractory epilepsypage 28
Non-epileptic attackspage 29
Epilepsypage 29
Considering surgerypage 31
Status epilepticuspage 33
Section 4The Adolescent with Epilepsy
Services for children with epilepsypage 37
Adolescencepage 37
Management of epilepsypage 37
Transitionpage 38
Section 5Women with Epilepsy
Accuracy of diagnosispage 39
Selection of medicationpage 39
Contraceptionpage 39
Pre-conception counsellingpage 40
Pregnancypage 40
Hormonal changes and seizurespage 41
Section 6Epilepsy in Older People
Who are “the elderly”?page 42
Diagnosispage 42
Implicationspage 43
Who should provide the service
for older people?page 44
Section 7Epilepsy in Complex Situations
Epilepsy in the context of learning disabilitypage 45
Epilepsy with psychiatric or
behavioural problemspage 47
Refractory epilepsypage 48
Section 8The Epilepsy Service
The Epilepsy Servicepage 51
The neurophysiology servicepage 56
Section 9Summarypage 58
Appendix 1North West Clinical Neuroscience Partnershippage 63
Appendix 2Participants and Advisorspage 65
Appendix 3 Literature Surveypage 66
Referencespage 71
Section 1
Introduction
The North West Clinical Framework
This guidance has been produced by the North West Clinical Neuroscience Partnership as a Framework for the commissioning and provision of care for people with epilepsy in the North West of England and North Wales. It links all aspects of care in a single service with clearly identified roles and responsibilities.
The North West Clinical Neuroscience Partnership is a partnership between the 3 Zonal Commissioners of Specialised Services in the North West, working on behalf of the Health Authorities and Primary Care Groups and Trusts in the North West, and the 3 Specialist Neuroscience Provider Trusts. The Partnership works in close cooperation with the North West Regional Specialised Commissioning Group and also includes representation from the Specialised Health Commissioning Service in Wales. The Partners have jointly agreed to develop a series of Clinical Frameworks and to work together to develop a Region-wide commissioning process which will ensure that the 7 million people of the North West and North Wales have access to a high quality Clinical Neuroscience service. The details of the Partners and the Partnership Agreement are included as Appendix 1.
Guidelines for the management of people with epilepsy have been published by a number of organisations. The evidence based guidelines of the Scottish Royal Colleges (SIGN) were particularly helpful as a starting point in constructing the North West Framework. Many people have contributed to the development of this Framework. The process began in May 1999 with a meeting at which users and commissioners of the service were asked to address the question “What do I expect from an Epilepsy Service?” Following this a number of workshops and working groups have involved patients, carers, voluntary agencies, GPs, neurologists, neurophysiologists, neuropsychologists, neuropsychiatrists, specialists in learning disability, neurosurgeons and specialist nurses. The Partnership is extremely grateful to all those people who have given their time and expertise over the last 2 years: a list of participants in the process can be found in Appendix 2.
It is planned to begin to use the Framework as a basis for commissioning in 2002 following any modification necessary in the light of consultation. The Partnership believes that the Framework will ensure the delivery of a high quality, reliable, accessible and sensitive service for people with epilepsy wherever they live in the North West and North Wales.
August 2001
Section 2
The Background
Incidence and prevalence of epilepsy
Epilepsy is a relatively common condition characterised by a tendency to recurrent seizures. Although there are variations among ethnic (Wright et al[1]) and socio-economic (Morgan et al[2]) groups within populations there is remarkable agreement that the incidence is about 50 per 100,000 per year over-all (Zarrelli et al[3], MacDonald et al[4]). There is also agreement that prevalence in the population from age 10 to 60 remains steady at about 7 per 1000, with “active epilepsy” affecting 4-5 per 1000 (Wright et al1, MacDonald et al4).
The first seizure can occur at any age but onset is predominantly during childhood or older age. Not all children who develop epilepsy continue to have seizures into adult life and some adults enter remission. Thus in spite of a steady prevalence up to age 60 in the population the cumulative incidence continues to rise. After age 65 there is a marked rise in incidence from 85.9 per 100,000 to 135.4 in people over 85. (Wallace[5]). This means that about 30% of the population will have had epilepsy at some stage in their life (Anderson et al[6]). The General Practitioner with an average list size of about 2000 will see 1 person with a new diagnosis of epilepsy each year and will provide care for 10-20 people with epilepsy depending on the age and socio-economic profile of the practice population.
Some people have many seizures each day; others are free from seizures for months or years. Although the condition is chronic it does not always remain active throughout a person’s life. With appropriate treatment 68-86% of people with epilepsy can be free from seizures or have less than 1 seizure in 3 years (Cockerell et al[7]). Thus there are important distinctions between incidence and prevalence of epilepsy, prevalence of active epilepsy, and incidence and prevalence of the various types of epilepsy.
Diagnosis
The agreement on incidence and prevalence is more remarkable in the context of the difficulties that are encountered in defining and recognising seizures and hence reaching a diagnosis (Smith et al[8], Zaidi et al[9]). Identification of the many different kinds of seizure, and diagnosis of the types of epilepsy depend upon the clinical history. Dependence on the clinical history as the basis of diagnosis places a relatively high value on the experience of the doctor taking the history, interpreting or amplifying the given description. Hart[10] found that the time from the first seizure to diagnosis was more than 6 months in 50% of patients and over 2 years in 30%. Only about 25% of the people referred to neurology or epilepsy clinics with a possible diagnosis of Epilepsy turn out to have that diagnosis (Smith et al8). Most have syncope (Zaidi et al9), (simple faints), or psychological, emotional, or personal problems. Syncope is often complicated by brief motor or ocular phenomena (Lempert et al[11]) which lead the unwary to an erroneous diagnosis of epilepsy. There is also good evidence to show that in at least 20% of patients given a diagnosis of epilepsy, the diagnosis is wrong (Smith et al8, Zaidi et al9). The process of diagnosis is thus not easy.
Investigation
Diagnosis is rarely dependant on anything other than the history and a witness description but the prognosis and options for treatment are largely determined by the type (syndromic classification) of the epilepsy and by any identifiable cause (Semah et al[12], Brodie and Dichter[13]). In looking for a cause or trying to establish the type of epilepsy, imaging of the brain (by CT or MR scanning) and neurophysiological studies, (electroencephalography, EEG), are necessary (King et al[14]). The EEG is particulary important in the classification of some forms of epilepsy. For those people with persisting or frequent seizures, especially partial seizures, further specialised imaging, neurophysiological and neuropsychological investigation may reveal a cause or show that surgical treatment could be of help (Devinsky[15]). Some findings, eg diffuse, low-grade glioma, might have no treatment implications but can explain the difficulty in controlling the seizures (Smith et al[16]). CT scanning is probably adequate for the identification of treatable causes of focal or localised epilepsy beginning in adult life although it will not reveal all lesions. MR scanning is more informative than CT scanning or Positron Emission Tomography (PET) and is essential when surgery is being considered (Heinz et al[17], Helveston et al[18]). If MR is readily available it would be reasonable to use it rather than CT scanning in the investigation of epilepsy when imaging is considered necessary.
Classification of epilepsy (Tables 1 and 2)
Classification of the type of epilepsy is also based largely on the history and witness description of seizures but the EEG findings are very important in identifying some seizure types and syndromes. Doubts about the accuracy of descriptions (Samuel and Duncan[19], Wulf[20]) or the interpretation of the description (Rinaldi et al[21]) have led to calls for a new classification which is easier to use outside specialist centres (Manford[22], Everitt and Sander[23]). In spite of these difficulties, identification of the seizure type and syndrome is important because of the information they give on prognosis and on the most appropriate antiepileptic drug treatment (Devinsky15, Brown et al[24]). The extent of agreement on incidence and prevalence suggests that the diagnostic tools, though not perfect, are usable and helpful.
TABLE 1. (Modified from International League Against Epilepsy[25])
CLASSIFICATION OF EPILEPSIES AND EPILEPTIC
SYNDROMES.
1Localisation-related (focal, local, partial) epilepsies and syndromes
1.1. Idiopathic (with age-related onset)
a. Benign childhood epilepsy with centrotemporal spikes
b. Childhood epilepsy with occipital paroxysms
1.2. Symptomatic
1.3. Unknown as to whether the symptom is symptomatic or idiopathic
2.Generalized epilepsies
2.1. Idiopathic (with age-related onset, in order of age)
a. Benign neonatal familial convulsions
b. Benign neonatal convulsions
c. Benign myoclonic epilepsy in infancy
d. Childhood absence epilepsy (pyknoepilepsy)
e. juvenile absence epilepsy
f. Juvenile myoclonic epilepsy (impulsive petit mal)
g. Epilepsy with grand mal (generalized tonic-clonic) seizures on
awakening
2.2. Cryptogenic or symptomatic (in order of age)
a. West’s syndrome (infantile spasms)
b. Lennox-Gastaut syndrome
c. Epilepsy with myoclonic-astatic seizures
d. Epilepsy with myoclonic absence
2.3 Symptomatic
2.3.1. non-specific aetiology
a. early myoclonic encephalopathy
2.3.2. specific syndromes
a. epileptic seizures that complicate many disease states
3. Epilepsies and syndromes undetermined whether focal or
generalized
3.1. with both generalised and focal seizures
a. neonatal seizures
b. Severe myoclonic epilepsy in infancy
c. Epilepsy with continuous spike-wave during slow wave sleep
d. Acquired epileptic aphasia (Landau-Kleffner syndrome)
4. Special syndromes
a. Febrile convulsions
TABLE 2. (Modified from International League Against Epilepsy[26])
CLASSIFICATION OF EPILEPTIC SEIZURES ACCORDING TO
CLINICAL TYPE.
I.Partial (focal and local) seizures
A. Simple partial seizures (consciousness not impaired)
1. With motor symptoms
2. With somatosensory or special sensory symptoms (simple hallu-
cinations, such as tingling, light flashes, buzzing)
3. With autonomic symptoms or signs (e.g., epigastric sensation,
pallor, sweating, flushing, piloerection, and pupillary dilatation)
4. With psychic symptoms (disturbances of higher cerebral function
e.g. déjà vu, fear, distortion of time perception)
B. Complex partial seizures (impairment of consciousness and often
automatisms)
1. With simple partial onset followed by impairment of conscious-
ness
2. With impairment of consciousness at onset
C. Partial seizures evolving to secondarily generalized seizures (e.g.,
generalized tonic-clonic seizures)
1. Simple partial seizures evolving to generalized seizures
2. Complex partial seizures evolving to generalized seizures
3. Simple partial seizures evolving to complex partial seizures and
further evolving to generalized seizures
II.Generalized seizures (convulsive or nonconvulsive)
A. Absence seizures (impairment of consciousness alone or with mild
clonic, atonic, or tonic components and automatisms
- Typical
- Atypical
B. Myoclonic seizures
C. Clonic seizures
D. Tonic seizures
E. Tonic-clonic seizures
F. Atonic seizures
III. Unclassified seizures
Causation
Epilepsy can be directly inherited through a single gene or can be associated with other clearly inherited conditions. While the inheritance of disorders directly causing epilepsy is uncommon there is a genetic influence in a high proportion of people with epilepsy (Steinlein[27]). Thus the incidence in children of a parent with epilepsy is 3 times that of the general population (Ottman et al[28]). Children of affected mothers are at greater risk than children of affected fathers (Ottman et al[29]). Some of the inherited degenerative or metabolic disorders are associated with seizures as a part of the much bigger clinical picture of delayed development or motor, sensory and intellectual decline. About 1/3 of people with learning disability have seizures (Sillanpaa[30], Graydon[31]). Infection, haemorrhage or trauma to the brain can cause epilepsy in children or adults: ischaemic stroke and some brain tumours are also prone to cause seizures, particularly in adults.
Status epilepticus
Status epilepticus is said to occur in more than 15% of patients with epilepsy at some stage in their life: either there is a series of seizures without return to normal in between, or a seizure lasts for more than 30 minutes. The event is recurrent in more than 13% of people (Fountain[32]). Other authors suggest a lower incidence drawing attention to the need for a precise definition and better management of repetitive seizures (Coeytaux et al[33]).
Generalised convulsive status epilepticus
Status epilepticus is as common when it is the first indication of an acute cerebral illness eg encephalitis as it is in the context of chronic epilepsy (Starreveld and Starreveld[34]). Particular risk factors are young age, low antiepileptic drug levels, fever in children and stroke in adults. Status epilepticus constitutes a medical emergency necessitating urgent parenteral medication and admission to hospital. If control is not achieved within 15 minutes, admission to an intensive care unit is essential so that muscle relaxants and assisted ventilation can be started. This allows the use of further antiepileptic medication which would suppress respiration. Status epilepticus still has a mortality of 7.6% (Coeytaux et al33) but this is largely related to the underlying cause (Starreveld and Starreveld34).
There is general agreement that the recognition and early management of this condition is poor with misdiagnosis, delay in medication or inadequate doses contributing to the morbidity and mortality (Cascino et al[35], Walker et al[36]). Care can be improved by the use of an algorithm, and education of staff in its use, even in the setting of a specialist service (Gilbert[37]).
Nonconvulsive status epilepticus
The definition, management and outcome of nonconvulsive status are more controversial. Until recently it was thought to be a relatively benign condition characterised by prolonged periods of abnormal behaviour in older people which were reversed by the parenteral administration of benzodiazepines (Thomas[38], Krumholz[39]). At least 4 types have now been described (Thomas38) and some authors suggest that up to 8% of comatose patients without signs of seizure activity have nonconvulsive status epilepticus (Towne et al[40]). Other authors feel that this observation is based on a misunderstanding of the EEG changes associated with coma (Niedermeyer and Ribiero[41]). In this situation it is not surprising that there is confusion over treatment. Seriously ill, comatose patients with EEG activity suggesting nonconvulsive status epilepticus do not appear to do well with aggressive medication to control the EEG; elderly patients displaying episodic confusion do respond to medication (Kaplan[42]). In the light of the difficulties with clinical diagnosis and EEG interpretation neurologists with expertise in epilepsy should be involved whenever suspicion of such a diagnosis arises.
Sudden unexpected death
Over the last few years it has become recognised that unexpected, sudden death is 2 or 3 times more common in people with epilepsy than in the population as a whole. Most authors believe that the deaths are seizure related (Opeskin et al[43]). Seizure related deaths might account for up to 40% of all deaths in people with chronic epilepsy (Tomson[44]). The mechanisms remain unclear but possible risk factors include male sex, age 20-40, generalised seizures, poor seizure control and poor compliance with medication (Langan[45]).
Medication
There is now a wide range of effective anticonvulsant or antiepileptic drugs. With appropriate selection and dosage about 70% of people with epilepsy will have less than one seizure in three years (Cockerell et al7). There is, however, no perfect drug which combines efficacy with freedom from side-effects. Some people would rather experience an occasional seizure than endure persisting drowsiness or sluggishness. For others the balance is different; even an occasional seizure could risk the loss of employment or driving licence so they are willing to accept side-effects.
The selection of medication depends upon the type of epilepsy, the age and gender of the patient and the individual choice of each patient once the possibilities have been explained (Devinsky15). Treatment with a single drug is usually adequate: Almost 50% of patients experience satisfactory control with monotherapy of the first drug used. If adequate doses of one drug do not achieve control the substitution of another single drug achieves control in 27% of that group. Only when monotherapy with therapeutic doses of 2 appropriate drugs has failed should combination therapy be tried. Although there is no clinical trial based evidence, when a second drug is to be added there is a theoretical advantage in choosing a drug with a different mechanism of action (Devinsky15). About 7% of patients who continue to have seizures in spite of adequate trials of monotherapy will have control improved by the addition of a second drug (Kwan and Brodie[46]).
Side effects of antiepileptic drugs are important in determining their usefulness in practice. Individual idiosyncratic reactions, acute sensitivity reactions, interference with the metabolism of other drugs and the chronic results of very long-term treatment are all factors to be taken into consideration when prescribing these medications.
Choosing between anti-epileptic drugs (See also Pathway 2, page 23)
(a)Partial onset seizures (simple partial, complex partial, secondary generalised tonic clonic).
Carbamazepine is the conventional drug of first choice (Mattson et al[47]). Valproate is effective but concerns about cosmetic effects and teratogenic potential limit its use in women of childbearing age. Claims that Lamotrigine (Brodie et al[48]) and Oxcarbazepine (Dam et al[49]) posses similar efficacy and better tolerability than Carbamazepine are not established. There is evidence of the effectiveness of Levetiracetam as an add-on medication (Shorvon et al[50]). However, like Gabapentin (Datta and Crawford[51]) and Topiramate (Sachdeo et al[52]) these new compounds are potential first-line agents.