Epilepsies in children, young people and adults

Why the committee made the recommendations

Referral after a first seizure

The committee agreed that people presenting with a suspected first seizure should be referred urgently to ensure that a specialist is involved early in diagnosing epilepsy. Diagnosing epilepsy can be complex and involving a specialist can help avoid misdiagnosis and ensure that the person receives the right care and support.

Assessing the risk of a second seizure

The evidence suggested that adults having a first seizure who have a mental health condition are almost 3 times more likely to have a second seizure when compared with the general population. The risk was even higher for people with sepsis, who are 4.5 times more likely to have a second seizure than people who do not have sepsis. The committee agreed that these are significant risk factors that could be modified to try to prevent second seizures. Evidence for vascular risk factors did not show a difference in risk. However, based on their knowledge and experience, the committee agreed that conditions such as diabetes, hypertension and atrial fibrillation are important risk factors for seizures in adults that may also be modified.

In children, the committee acknowledged that there was a lack of clarity in the evidence for risk associated with higher or lower temperature. A seizure because of a high temperature does not necessarily predispose a child to more seizures, but they agreed that increased temperature is important to take into consideration. Febrile seizures tend not to predispose to a second afebrile seizure. However, afebrile seizures may be associated with an increased risk of a second seizure. The committee agreed that parents and carers should be given information about the potential risk and how to self-refer should the child have a second seizure. Safety guidance should also be given so that parents and carers can take precautions to minimise the risk of injury.

The risk factors identified in the studies are not the only factors that affect a person's chances of having a second seizure. For this reason, the committee decided that assessment should include identifying any potential mental, physical and social risk factors, which should then be discussed with the person and their family or carers.

The committee discussed the evidence for prediction tools for a second seizure, but did not recommend using any of these tools because they were considered to carry the potential for harm. The evidence suggested that the tools had a poor capacity to discriminate between people at low and high risk of second seizure. Therefore, the committee made a recommendation for research on developing and testing a risk prediction tool for second seizure.

How the recommendations might affect practice

These recommendations are likely to mean a change in clinical practice for how adults are cared for after a first seizure. In current practice, only about 25% of adults are fully assessed for modifiable risk factors. Assessment includes checking for underlying mental health problems, vascular risk factors and sepsis. Although the recommendations for adults will result in a change in clinical practice, the assessment does not take long and is not expected to result in a substantial resource impact. A small increase in costs is likely for additional staff time to assess people presenting with a first seizure.

The recommendation made for children reflects current practice so the committee agreed there should be no substantial resource impact.

Return to recommendations

Why the committee made the recommendations

In assessing the evidence for individual tests to diagnose epilepsy, the committee agreed that a diagnostic test would need to give the lowest possible level of false-positive and false-negative results. False-positive results may result in unnecessary treatment and anxiety, whereas false-negative results may result in people with epilepsy remaining undiagnosed and untreated. Given the seriousness of these harms, the committee agreed that a 10% rate for false negatives and a 10% rate for false positives were the highest acceptable rates (equating to a minimally acceptable value of 0.9 for both sensitivity and specificity). Most tests evaluated in the review did not meet this threshold.

Clinical history and examination provided by a specialist in epilepsy demonstrated levels of sensitivity and specificity for detecting epilepsy that were above the agreed threshold. Although the evidence was restricted to adults, the committee were confident that this could also be applied to diagnosis in children and young people. Witness reports and review of video footage were included as useful additional features of the clinical history. The evidence did not show sufficient diagnostic accuracy to warrant the use of witness reports or video footage independently, but the committee agreed that they increase the accuracy of expert clinical diagnosis.

The committee agreed, based on their knowledge and experience, that a positive electrocardiogram (ECG) can identify cardiac causes of seizure-like symptoms, and a negative ECG can support a further investigation of suspected epilepsy. Similarly, they agreed that the assessment of metabolic disturbances, such as hypoglycaemia can help to exclude alternative causes of a first seizure.

Although none of the imaging modalities were sufficiently accurate for use as diagnostic tools, the committee agreed that neuroimaging should be used to investigate potential structural causes of epilepsy.

How the recommendations might affect practice

No impact on practice is expected, because these recommendations do not substantially change current practice.

Return to recommendations

Why the committee made the recommendations

How the recommendations might affect practice

The use of neuroimaging varies in current practice, and is not routinely used in all settings. The recommendations will reduce variation in current practice. There may be an increase in the number of people who have neuroimaging. However, with the use of regionally agreed protocols, the detection of abnormalities may avoid the need for more scans in the future.

Return to recommendations

Why the committee made the recommendations

The committee agreed that a discussion with a neurologist or geneticist may be needed to advise on who to test and which type of test to use if there are uncertainties. Access to genetic testing varies but is likely to increase. The committee noted that the NHS National Genomic Test Directory for rare and inherited disease lists the genetic tests for epilepsy that are nationally commissioned by the NHS in England, and when they should be performed.

A genetic diagnosis can provide information about treatment options, related medical problems and prognosis. It can also inform genetic counselling for the person and their family members. The committee discussed the importance of genetic counselling to ensure that people, and their families or carers if appropriate, are supported to understand the possible results of testing and what they might mean for them and their family. Results can reveal information about the person who is being tested, but also their family members, which may have emotional and social consequences.

The committee agreed that a full discussion of the purpose and implications of genetic testing is needed before testing, so that the person, and their family or carers if appropriate, can make decisions about testing and give informed consent. Informed consent needs to be documented and it involves a full discussion about the benefits and possible limitations and risks of genetic testing, as well as consent to share information, where appropriate, about the results to advise the person's relatives.

The evidence did not support genetic testing for all people with epilepsy, so the committee took a pragmatic approach and agreed that testing should be considered in situations most likely to yield positive diagnostic results. The committee agreed that there was not enough evidence to recommend genetic testing at a specific point in the clinical pathway.

People whose epilepsy started at an early age (under 2 years) and people whose epilepsy is associated with developmental disorders or certain clinical features are more likely to have epilepsy with a genetic cause. Therefore, based on their knowledge and experience, the committee agreed that whole-genome sequencing would be best targeted to this population. This is in line with the eligibility criteria that accompany the NHS National Genomic Test Directory (clinical indication R59: early-onset or syndromic epilepsy).

The diagnostic yield of genetic testing in people with epilepsy of unknown cause with onset between 2 and 3 years is affected by factors such as co-occurring conditions and MRI and EEG findings. The committee noted that a genetic diagnosis in this group of people can have a significant positive impact on outcomes and management, therefore, based on their knowledge and experience and as specified by the NHS National Genomic Test Directory, they agreed that whole-genome sequencing should be carried out if agreed by a specialist multidisciplinary team.

How the recommendations might affect practice

The use of genetic testing varies in current practice, and it is not routinely carried out even when a genetic cause is suspected. The recommendations clarify when genetic testing should be considered, which will reduce variation in current practice. There may be an increase in the number of people who have a genetic test and who are referred for genetic counselling. However, with the use of whole-genome sequencing, there may also be a reduction in the number of people having unnecessary tests that do not provide additional information on their diagnosis.

Return to recommendations

Why the committee made the recommendation

The evidence for antibody testing was limited and did not support routine antibody testing for people with epilepsy. However, the committee discussed antibody testing in the context of suspected autoimmune encephalitis in people with new-onset epilepsy because it is recognised that people with autoimmune encephalitis can present with seizures or status epilepticus with encephalopathy. Although not part of the evidence review, the committee was aware that treatment guided by antibody testing (immunotherapy) may improve outcomes in these people compared with standard antiseizure medication.

There is emerging evidence on 'autoimmune epilepsy', but the committee agreed that further research is needed to assess which immunomodulation strategies are effective in people with defined autoimmune epilepsy syndromes. The committee agreed that further research is needed and developed a recommendation for research on immunomodulation strategies for people with autoimmune epilepsy syndromes to help inform future guidance.

How the recommendation might affect practice

Suspected autoimmune encephalitis is relatively rare and antibody testing is already current practice, so the recommendation reinforces current best practice.

Return to recommendation

Why the committee made the recommendations

The evidence showed that there are gaps in current practice in communication and in the information and support available to people with epilepsy.

The committee agreed that tailored information should be provided to people with epilepsy, and their parents and carers if appropriate, to enable them to be fully informed and involved in decisions about their care. The evidence reported that some people with epilepsy and some parents and carers felt that information was withheld, making it difficult to be fully involved in their care. It also showed that children and young people wanted to be involved but sometimes struggled to understand information inappropriate for their age, including information about sudden unexpected death in epilepsy (SUDEP). The committee stressed the importance of providing age-appropriate information to enable children to be involved in discussions about their care. Extra support and time in consultations for people with learning disabilities or complex needs, such as other comorbidities, were also highlighted by the committee.

The committee highlighted the important role that epilepsy specialist nurses play in information giving. This was supported by the evidence and recommendations on epilepsy specialist nurses.

The evidence showed that parents and carers struggled to find help from sources other than their doctor. The committee acknowledged that information on where and how people with epilepsy can access information and support for activities of daily living, such as local and national support groups, should be provided.

The committee stressed the importance of providing key information for self-managing epilepsy during the first appointment. Medicines adherence, mitigating epilepsy-related risk and avoiding potential provoking factors for seizures were identified as key topics from the evidence to enable the person to self-manage their epilepsy and maintain everyday activities. Based on their experience, the committee also included activities of daily living, including driving, as a key topic for discussion because it is commonly raised as a concern at the first appointment. The committee agreed that this key information should be repeated at subsequent appointments.

The committee recognised that people with epilepsy may have a range of worries and anxieties that may change over time, and that opportunities should be provided to discuss these at each appointment. They agreed on some topics that are often of concern to people with epilepsy and their families and carers, based on their experience and themes identified in the evidence, which could be used to help provide a framework for discussions.

How the recommendations might affect practice

The recommendations reflect current practice so the committee agreed there should be no substantial resource impact.

Return to recommendations

Why the committee did not make any recommendations

No evidence was found on using digital health technologies, so the committee agreed that no recommendation could be made. The committee noted that people are already using devices, such as night monitors and alarms, as self-management tools, but that evidence is lacking to support this use, and these are not currently offered by the NHS. Based on their experience, the committee acknowledged that some monitoring tools may offer benefit to people with epilepsy.

There is a trend towards the use of digital health technologies and the committee were keen to encourage more research in this emerging and potentially important area. A recommendation for research on digital health technologies was developed to help determine their clinical and cost effectiveness for people with epilepsy in the hope that these interventions could lead to improvements in self-management.

Return to recommendation for research

Why the committee made the recommendations

There was a lack of evidence on referral to specialist services, so the committee based their recommendations on their clinical experience and expertise, and also used the NHS England 2019 guidance for referral pathways to specialist services for adults as a reference.

Children, young people and adults may need access to tertiary services at certain times during their care and these services should be available to everyone who needs them through their specialist. However, the committee acknowledged that some groups may need extra tertiary support to manage their epilepsy, even if they are already receiving care from other specialists for another condition. The committee noted that people with a learning disability or mental health problem may struggle to access tertiary services and may need help to get appropriate referrals. They may also need additional support to attend appointments, such as having a family member or carer accompany them.

With the number of referrals increasing, the committee agreed that clearer and more specific criteria for referral would help to ensure that people who will benefit most from specialist services are prioritised. The proposed criteria aim to ensure that people with epilepsy that is difficult to diagnose or manage receive the specialist care and treatment they need, including consideration for clinical trials. The committee recommended, based on their knowledge and expertise, that people with epilepsy meeting these criteria should be seen within 4 weeks. The committee agreed that these are people with substantial health needs, so prompt referral could have a significant positive impact on their prognosis. This is in line with current clinical practice and consistent with the previous version of the guideline, so it is not expected to have an impact on resources. The committee agreed to retain and update the recommendation from the previous guideline because it was important to continue to guide clinicians to improve care and to reduce variation in service provision and outcomes between services.

The committee also discussed particular groups of children that should be prioritised for more urgent referral within 2 weeks. This was based on the committee's knowledge and experience, and on the probability that treatment from a tertiary paediatric epilepsy service would significantly benefit children meeting these criteria. They agreed that all children under 3 years should be referred to tertiary services without delay, because of the risk of developmental problems with some paediatric syndromes with onset before this age. Children with myoclonic seizures presenting aged up to 4 years should be referred because myoclonic seizures may start after 3 years and could indicate an underlying neurodegenerative disorder that may be treatable. Children with a unilateral structural lesion should be prioritised for immediate referral because this is likely to lead to difficulties in seizure management and surgery may need to be considered early. The presence of deterioration in the child's behaviour, speech or learning (such as loss of previously acquired developmental milestones), particularly in the absence of an established diagnosis, should also be a priority for prompt investigation in tertiary services.

How the recommendations might affect practice

The recommendations will not change current practice, but will reinforce current best practice.

Return to recommendations

Why the committee made the recommendations

These recommendations are based on the committee's informal consensus on principles for the use of antiseizure medications for treating all epilepsy syndromes and seizure types. They are based on recommendations from the previous version of the guideline and have been retained and updated because the committee agreed it was important to include them to guide clinicians to improve care.

The committee agreed on some general factors to consider once the diagnosis of epilepsy is confirmed to ensure that the best treatment is started, balancing the risks and benefits of the medicine with the lifestyle and choices of the person. For example, if a person is starting university and the most effective medication can affect cognitive performance, they may wish to choose a different option without these side effects. For some people, their seizure type may mean that a medicine that is faster-acting to reduce seizures might be a priority. These factors should be discussed with the person with epilepsy (and their families and carers if appropriate) in all settings where antiseizure medications may be prescribed and managed, including primary, secondary and tertiary care.

The committee agreed on some particular issues that should additionally be taken into account when giving antiseizure medications in older people. This group presents with specific considerations because of differences in clinical presentation and aetiology when compared with younger people, particularly those with co-occurring conditions. Treatment can be more difficult because of drug interactions, therefore other medications the person may be taking should be taken into account. Additionally, care is needed in choosing the adequate initial dose and subsequent titration. The committee noted that following the NICE guideline on multimorbidity, which covers care optimisation in people with multiple long-term conditions, may benefit people with epilepsy and 1 or more mental or physical health condition.

The committee agreed that an antiseizure medication treatment strategy should take account of these factors and the special considerations for antiseizure medications in women and girls. The committee agreed that a shared decision should be made with the person to agree their individualised antiseizure medication treatment strategy.

The committee agreed on some principles if seizures continue after monotherapy treatment, which included reviewing the diagnosis and trying monotherapy with another antiseizure medication to ensure that the most effective treatment strategy is being used. The committee agreed that, when starting an alternative antiseizure medication, the dose of the new antiseizure medication should be slowly increased while the existing antiseizure medication is tapered off because this can reduce the risk of drug-related withdrawal symptoms of the first medication and clinicians can monitor the correct dose of the second medication.

If alternative antiseizure medications prove to be unsuccessful, an add-on treatment should be considered. Because of the possible interactions between antiseizure medications, for example sodium valproate and lamotrigine, the committee agreed that add-on therapies should be carefully titrated and people should be monitored for adverse effects and their medicines reviewed frequently.

The committee highlighted the importance of using the regimen that provides the best balance in terms of effectiveness and tolerability of side effects, and that the benefits of rationalising medications (using a single medicine if possible and what to consider if this monotherapy is unsuccessful) should be discussed with the person with epilepsy (and their families and carers if appropriate) to ensure people are not taking more medicines than is necessary to reduce the impact of side effects.

How the recommendations might affect practice

The recommendations will not change current practice, but will reinforce current best practice.

Return to recommendations

Why the committee made the recommendations

These recommendations are based on the committee's informal consensus on when to start treatment with antiseizure medication for all epilepsy syndromes and seizure types.

The committee agreed that treatment with antiseizure medication after a first unprovoked seizure should not be offered routinely. However, they agreed that some clinical features should prompt early treatment after a first unprovoked seizure, such as if an examination shows the person has a neurological deficit or the EEG shows unequivocal epileptic activity, which may indicate that the risk of recurrence is high. In some circumstances, for example if there is a risk of loss of employment, further seizures may be unacceptable, so the person or their family or carers may choose to start early treatment. A structural brain abnormality indicates that the brain is damaged, therefore prompt treatment may stop further seizures.

How the recommendations might affect practice

The recommendations will not change current practice, but will reinforce current best practice.

Return to recommendations

Why the committee made the recommendations

These recommendations are based on the committee's informal consensus on safety considerations for starting antiseizure medication to treat all epilepsy syndromes and seizure types with antiseizure medication.

Antiseizure medications differ significantly in their characteristics, therefore the risk of switching between different manufacturer's products, different generic products or branded originator and generic products needs to be taken into account. The committee agreed that Medicines and Healthcare products Regulatory Agency (MHRA) advice on switching between different manufacturer's products needs to be followed.

In line with the BNF, the committee agreed clinicians should be aware of the risks of serious complications associated with phenytoin for people of Han Chinese or Thai family background, and the risks of serious complications associated with carbamazepine and potentially medicines with a similar chemical structure (such as oxcarbazepine and eslicarbazepine acetate) for people of Han Chinese, Thai, European or Japanese family background.

In line with the MHRA, the committee noted the antiseizure medicines most commonly reported to cause decreased bone mineral density and increased risk of osteomalacia. The committee agreed that appropriate supplementation should be considered for those at risk.

How the recommendations might affect practice

The recommendations will not change current practice, but will reinforce current best practice.

Return to recommendations

Why the committee made the recommendations

The guideline committee wanted to ensure that women and girls with epilepsy had access to appropriate support and information about contraception, conception, pregnancy, breastfeeding and caring for children, and menopause. They stressed the importance of having regular reviews with women and girls to ensure they had access to further information and treatment as their circumstances change.

The committee referred to the MHRA's Public Assessment Report of antiepileptic drugs: review of safety of use during pregnancy to inform the recommendations.

In the absence of evidence, the committee made consensus recommendations for women and girls with epilepsy who were breastfeeding. They agreed women and girls should be supported to breastfeed if they wish, because the benefits of breastfeeding outweigh the small risk of the drug affecting the child.

Impact of the recommendations on practice

Women and girls with epilepsy do not currently have their concerns addressed adequately. Services providing reviews and support are thought to be under-commissioned at the present time and so the recommendations are likely to have an impact on practice with an increase in regular reviews. The MHRA safety advice may encourage women who are on antiseizure medications other than lamotrigine and levetiracetam to reconsider their treatment options.

Return to recommendations

Why the committee made the recommendations

Monitoring reviews are essential to reassess the clinical management plans of people with epilepsy as their needs change. Evidence comparing regular scheduled reviews with patient-initiated ad-hoc reviews did not suggest any differences in benefit or harm between these approaches. The committee discussed that patient-initiated review could help to ensure timely management of changes in a person's needs and support their sense of ownership of managing their epilepsy. However, it would have disadvantages for people with less independence or capacity to make decisions, and could lead to loss of contact with services, with potentially serious consequences. It might also be unsuitable for people with a serious or complex condition, for whom failing to contact services could be particularly harmful.

The committee agreed that patient-initiated review should be available to all people with epilepsy, but that regular reviews should be provided to groups that are less suited to a patient-initiated approach. Based on their experience, the committee agreed that groups scheduled for regular reviews should include people with reduced capacity for decision making, people with serious or complex epilepsy, those with serious comorbidities, and children and young people. They also identified people taking antiseizure medication associated with long-term side effects or drug interactions as a priority to check for any adverse effects. Long-term side effects may include adverse effects on blood parameters or bone health, or changes in lipid metabolism. Enzyme-inducing medications in particular are associated with reduced bone density. Regular review for women and girls who are able to have children and are taking valproate or other high-risk teratogenic medication was also recommended to allow a discussion of their treatment options and any plans for pregnancy. The committee commented that regular review is current practice for children and young people, typically with 2 reviews a year and at least once every 12 months.

The evidence comparing therapeutic drug monitoring with clinical review suggested there is little difference in benefit between these approaches. Based on their experience, the committee agreed that for most people with epilepsy, therapeutic drug monitoring is unnecessary, but that certain groups might gain particular benefit from it. These groups include people who need accurate titration of their medicine levels, such as those with side effects, whose seizures are not controlled with treatment and those in whom adherence is less assured. They also include people at particular risk from their medication, because of either the intrinsic nature of the medication or the increased risks of the medication in people with comorbidities or who are pregnant (for example, to monitor for changes in lamotrigine plasma levels during pregnancy and after birth).

How the recommendations might affect practice

The recommendations may change practice because regular review is currently standard practice for all people with epilepsy. There was some concern that specialist nurse services would need to be developed to coordinate patient-initiated reviews. However, the committee agreed that demand on services is likely to be manageable, provided that regular reviews are maintained for groups that might have additional need for coordination. Restricting therapeutic drug monitoring to a few specific groups will not place any extra burden on providers, and might even slightly reduce it.

Return to recommendations

Why the committee made the recommendations

The committee acknowledged the potential importance of drug monitoring in pregnancy. However, the available evidence was limited to a single study. The committee agreed that the evidence was inconclusive, so the group based the recommendations on their own experience and guidance from the MHRA about monitoring levels of carbamazepine, lamotrigine, levetiracetam, oxcarbazepine, phenobarbital or phenytoin if used in pregnancy. The committee noted that on-site testing is often available at tertiary epilepsy centres for some antiseizure medications, including carbamazepine, phenytoin and phenobarbital. They acknowledged that phenytoin and phenobarbital are not usually taken by girls and women who are planning pregnancy. The committee also agreed that pre-conception monitoring of antiseizure medication levels should be considered in women and girls at risk of their seizures worsening during pregnancy and made a recommendation based on committee consensus. The committee highlighted the importance of obtaining pre-conception levels of antiseizure medication as a baseline level to compare and titrate against when monitoring drug levels during pregnancy.

The committee expressed the need for robust evidence in this area and therefore suggested a recommendation for research on monitoring antiseizure medications in women and girls.

How the recommendations might affect practice

The committee noted that currently there would be some women with epilepsy who are planning pregnancy or who are pregnant who are not having their antiseizure medications monitored. The recommendations are in line with MHRA safety advice on monitoring in pregnancy, and may result in some increases in drug monitoring compared with current practice.

Return to recommendations

Why the committee made the recommendations

Decisions about stopping antiseizure medication are nuanced, based on the person's preferences and their individual risk of seizure recurrence. Although there was some evidence for independent risk factors associated with seizure recurrence, the committee agreed that the recommendations should be broader than listing risk factors, which could be misleading in isolation.

Ongoing risk and benefit assessment is important to take account of the evolving needs of the person with epilepsy. Based on their knowledge and experience, the committee agreed that an individualised assessment of the risk of seizure recurrence should be carried out in those who have been 2 years without seizures. The committee stressed the importance of having a discussion with the person, and their family or carer if appropriate, about their personal preferences and the person's individual risk of seizure recurrence, in particular taking into account the type of epilepsy, so that the person is able to make an informed decision about their care. For example, stopping antiseizure medication in people with certain epileptic syndromes, such as juvenile myoclonic epilepsy, structural abnormalities or with co-existing neurodegenerative and other neurological conditions will pose a significant risk of seizure recurrence.

The committee agreed that guidance should be sought if there are doubts or concerns about the risks and benefits of discontinuing antiseizure medications. Because of the complexity and wide variation of epilepsy surgery techniques, the committee agreed that those who have undergone epilepsy surgery should have antiseizure medications discontinued under the guidance of the epilepsy surgery centre.

The committee highlighted the importance of stopping antiseizure medications slowly, especially benzodiazepines and barbiturates, because of the possibility of drug-related withdrawal symptoms. They also agreed that epilepsy specialist guidance would be needed if seizures recur.

How the recommendations might affect practice

The recommendations will not change current practice, but will reinforce current best practice.

Return to recommendations

Why the committee made the recommendations

Generalised tonic-clonic seizures rapidly involve both sides of the brain. During such seizures, consciousness is lost and muscles will stiffen before jerking rhythmically.

The evidence showed that, for time to treatment failure, no drugs performed better than sodium valproate, with sodium valproate showing clear benefits over lacosamide, phenobarbital, carbamazepine and topiramate. There was no clear difference between sodium valproate and all other drugs for remission or time to first seizure. The committee agreed that sodium valproate should be offered as first-line treatment, but because of the risks to unborn babies associated with sodium valproate use in pregnancy, they highlighted that it should not be used in women and girls who are able to have children unless other treatments are unsuccessful and the MHRA safety advice is followed.

The evidence suggested that, after sodium valproate, lamotrigine and levetiracetam had the next best time until treatment failure. For this reason, the committee recommended them as first-line monotherapy options for women and girls who can have children and second-line monotherapy options when sodium valproate is unsuccessful as first-line monotherapy.

The committee discussed the evidence on adverse events and their importance in making choices about drug treatment. However, these were reported inconsistently across the studies making comparisons between drugs difficult. The committee also agreed that for most drugs, adverse events could be managed by careful titration and dosage changes.

From the evidence, it was difficult to determine the most effective add-on drug for generalised tonic-clonic seizures that have failed to respond to monotherapy. Therefore, a number of drugs were recommended as potential first-line add-on treatments. There was evidence that lamotrigine, levetiracetam, perampanel and topiramate performed better than placebo for achieving a 50% response rate. No evidence was identified for clobazam and sodium valproate, but the committee agreed to include them based on their experience and current use in practice. There was also some evidence that levetiracetam and perampanel were more effective than placebo at achieving seizure freedom, but there was a lot of uncertainty around these results.

The evidence also suggested that brivaracetam might be more effective than placebo at achieving more than 50% reduction in seizure frequency, and lacosamide was less effective than levetiracetam for the same outcome. The committee therefore recommended both brivaracetam and lacosamide, as well as phenobarbital, primidone and zonisamide, based on their experience and knowledge of current practice, as possible second-line add-on treatments.

The committee stressed again that women and girls able to have children should not be offered sodium valproate as a first-line add-on.

The committee highlighted that clinicians should take into account that some drugs used in clinical practice can exacerbate seizures in those with absence or myoclonic seizures, including in juvenile myoclonic epilepsy.

In line with the MHRA, the committee emphasised that long-term treatment with primidone and sodium valproate can cause decreased bone mineral density and increased risk of osteomalacia. The committee noted that appropriate supplementation should be considered for those at risk.

How the recommendations might affect practice

The recommendations will reinforce current practice.

Return to recommendations

Why the committee made the recommendations

Focal-onset seizures originate in 1 area on 1 side of the brain and the person may have full or partial awareness. Symptoms vary widely depending on the area of the brain they originate from.

The evidence showed that lamotrigine and levetiracetam were continued for longer than other drugs for treating focal epilepsy, suggesting that they may be more effective and better tolerated. However, the evidence also suggested they were not more effective than other drugs in terms of remission at 6 and 12 months, and the evidence for time to first seizure suggested they were less effective than carbamazepine.

The committee discussed the evidence on adverse events and their importance in making choices about drug treatment. The evidence suggested that lamotrigine, levetiracetam and gabapentin may have more tolerable adverse events than other drugs. However, adverse events were reported inconsistently across the studies making comparisons between drugs difficult. The committee also agreed that for most drugs, adverse events could be managed by careful titration and dosage changes.

The committee agreed that lamotrigine and levetiracetam should be considered as first-line monotherapy options, and this was supported by economic modelling. They agreed that if these treatments were unsuccessful, carbamazepine, oxcarbazepine or zonisamide could be considered for second-line monotherapy. The evidence was weaker for lacosamide, so this was included as a third-line option.

From the evidence, it was difficult to determine the most effective add-on treatment for people with focal epilepsy that has failed to respond to monotherapy. The evidence showed that a number of antiseizure medications are effective compared with placebo for more than 50% reduction in seizure frequency rate: brivaracetam, carbamazepine, eslicarbazepine acetate, lacosamide, lamotrigine, levetiracetam, oxcarbazepine, perampanel, pregabalin, topiramate and zonisamide. Medications with the strongest evidence for this outcome were recommended as first-line options. The committee agreed the use of cenobamate as an add-on treatment, after at least 1 other add-on antiseizure medication has been unsuccessful, in line with NICE's technology appraisal guidance on cenobamate for treating focal onset seizures in epilepsy.

As with the evidence for monotherapy, the evidence on adverse events with add-on therapy was inconsistent and the committee were not able to use it to inform the recommendations.

Although the evidence was less clear, the committee agreed that, based on their experience, sodium valproate can also be an effective option. Because of the risks to unborn babies associated with sodium valproate use in pregnancy, the committee highlighted that it should not be used in women and girls who are able to have children unless other treatments are unsuccessful and the MHRA safety advice is followed.

There was a lack of evidence for other antiseizure medications, but based on the committee's experience, phenobarbital, phenytoin and vigabatrin were recommended only when the previous treatments are not tolerated or are unsuccessful, for example because of the risk of particular adverse effects.

The committee noted that, in line with guidance from the MHRA, clinicians should be aware of the risks of serious complications associated with phenytoin for people of Han Chinese or Thai family background, and the risks of serious complications associated with carbamazepine, and potentially medicines with a similar chemical structure (such as oxcarbazepine and eslicarbazepine acetate) for people of Han Chinese, Thai, European or Japanese family background.

In addition, in line with the MHRA, the committee emphasised that long-term treatment with carbamazepine, phenytoin and sodium valproate can cause decreased bone mineral density and increased risk of osteomalacia. The committee noted that appropriate supplementation should be considered for those at risk.

How the recommendations might affect practice

The recommendations will reinforce current practice. Previous NICE guidance recommended lamotrigine and carbamazepine for first-line monotherapy, with restrictions on the use of levetiracetam owing to costs. Levetiracetam is now significantly cheaper and widely prescribed in the NHS. These recommendations may lead to a small increase in the use of levetiracetam, but this will not lead to a significant increase in costs.

All drugs recommended as add-on treatments are already widely used. Gabapentin and clobazam are no longer recommended, which may lead to a small decrease in the use of these drugs. However, these drugs are not widely used and are likely to be continued in people who are already using them successfully.

Return to recommendations

Why the committee made the recommendations

Absence seizures are a form of generalised epileptic seizure, characterised by a lack of awareness, stopping moving or talking and staring blankly. They can occur in isolation, but can also be associated epilepsy syndromes, such as childhood absence epilepsy, juvenile absence epilepsy and juvenile myoclonic epilepsy. The evidence identified was only on children and young people; however, the committee agreed that it was appropriate to extrapolate from this evidence to the adult population because of the similar pathophysiology in children, young people and adults.

How the recommendations might affect practice

The recommendations will not change current practice, but will reinforce current best practice.

Return to recommendations

Why the committee made the recommendations

There was very limited evidence on first-line treatment for myoclonic seizures, so the committee used their clinical expertise and expert opinion to inform the recommendations. The onset of myoclonic seizures before the age of 4 years may indicate an underlying neurodegenerative disorder, therefore the committee agreed that these children should be referred to a tertiary paediatric neurologist.

Myoclonic seizures are classified as generalised seizures. Because no evidence was identified on monotherapy or first-line treatments for myoclonic seizures, the committee agreed that it was appropriate to extrapolate from the evidence reviewed on generalised tonic-clonic seizures. Based on this, the committee agreed that sodium valproate is the most effective treatment option for myoclonic seizures compared with other antiseizure medications. However, because of the risks to unborn babies associated with sodium valproate use during pregnancy, the committee agreed that it was important to make separate recommendations for women and girls who are able to have children. In addition, in line with the MHRA, the committee emphasised that long-term treatment with sodium valproate can cause decreased bone mineral density and increased risk of osteomalacia. The committee noted that appropriate supplementation should be considered for those at risk.

There was some evidence that levetiracetam, when used as an add-on treatment, is effective in controlling seizures, so the committee agreed that it should be offered as the first-line treatment in women and girls who are able to have children, and for younger girls with epilepsy likely to continue beyond puberty. Based on this evidence, the committee agreed that levetiracetam should also be offered as a second-line add-on or monotherapy treatment for men and boys if sodium valproate is unsuccessful.

The committee agreed it is important to stress that for some women and girls who are able to have children, sodium valproate may still be an option, but only if the risks and benefits have been thoroughly discussed, other treatments are unsuccessful and safety advice from the MHRA is followed. This should be a shared decision taken by the person with epilepsy and their healthcare professional.

In the absence of robust evidence, the committee discussed their experience and knowledge of other antiseizure medications for myoclonic seizures and agreed that brivaracetam, clobazam, clonazepam, lamotrigine, phenobarbital, piracetam, topiramate and zonisamide may be effective as third-line treatments if second-line monotherapy or add-on treatment is not sufficient to stop seizures. The committee noted that doctors should use their knowledge and experience to determine which treatment is most appropriate for the person with myoclonic seizures, taking into account clinical factors and the person's preferences and choice. They noted that although lamotrigine is used in clinical practice and may benefit some people, it can sometimes exacerbate myoclonic seizures.

The committee wanted to make clear that carbamazepine, gabapentin, oxcarbazepine, phenytoin, pregabalin, tiagabine and vigabatrin should not be used because they are known to increase the frequency of myoclonic seizures.

How the recommendations might affect practice

The recommendations will not change current practice, but will reinforce current best practice.

Return to recommendations

Why the committee made the recommendations

Tonic or atonic seizures are events that may cause a person to suddenly drop to the floor. These may be the result of atonic (generalised loss of tone) or tonic (sustained generalised body stiffening) seizures. Although these are characteristic of Lennox–Gastaut syndrome, they are also seen in other epilepsy syndromes and aetiologies. They often result in injury and can therefore have a significant impact on quality of life.

Because of the complexities associated with the treatment of tonic or atonic seizures and the range of syndromes of which they can be a feature, the committee agreed that a neurologist with expertise in epilepsy should assess people with these seizures in order to provide accurate diagnoses if possible and advise on further investigations as well as treatment.

Tonic or atonic seizures are classified as generalised seizures. Because no evidence was identified on monotherapy or first-line treatments for tonic or atonic seizures, the committee agreed that it was appropriate to extrapolate from the evidence on generalised tonic-clonic seizures. Based on this, the committee agreed that sodium valproate is the most effective treatment option for tonic or atonic seizures compared with other antiseizure medications. However, because of the risks to unborn babies associated with sodium valproate use during pregnancy, the committee agreed that it was important to make separate recommendations for women and girls who are able to have children. In addition, in line with the MHRA, the committee emphasised that long-term treatment with sodium valproate can cause decreased bone mineral density and increased risk of osteomalacia. The committee noted that appropriate supplementation should be considered for those at risk.

There was some evidence that lamotrigine, when used as an add-on therapy, is effective in controlling tonic and atonic seizures or drop attacks, so the committee agreed that it could be considered as first-line treatment for women and girls who are able to have children and as a second-line monotherapy or add-on treatment for boys and men, and women and girls unable to have children.

However, the committee also agreed that for some women and girls who are able to have children, sodium valproate may still be an option, but only if the risks and benefits have been thoroughly discussed, other treatments are unsuccessful and safety advice from the MHRA is followed. This should be a shared decision taken by the person with epilepsy and their healthcare professional.

The evidence indicated that clobazam, rufinamide and topiramate can also be effective in the management of tonic and atonic seizures and the committee recommended that any of these antiseizure medications could be used as a third-line monotherapy or add-on treatment. In the absence of clear cost-effectiveness evidence of superiority between the different options, the committee agreed that clinicians should use their knowledge and experience to determine which treatment is most appropriate for the person with epilepsy, taking into account clinical factors and the person's preference.

Evidence was identified for a number of other treatment options; however, the low quality and absence of direct comparisons meant that it was difficult for the committee to prioritise 1 treatment over another. The committee agreed that a ketogenic diet can be considered as an add-on treatment and, if this is unsuccessful, felbamate may also be an option as an add-on treatment. However, these treatments should only be used under the supervision of a ketogenic diet team or a neurologist with expertise in epilepsy, respectively, because of the complex nature of the epilepsy.

The committee agreed that although other antiseizure medications are used in clinical practice and may benefit some people, it should be highlighted that some can exacerbate seizures.

How the recommendations might affect practice

The recommendations are not likely to change current practice, but should reinforce best practice.

Return to recommendations

Why the committee made the recommendations

Idiopathic generalised epilepsies are 1 of the most common forms of epilepsy. These are well defined, and onset is usually in adolescence, although it can begin in childhood. Seizures will continue into middle age, after which there is some evidence that these will remit, but is not possible to predict in which people this will occur. Many have a good prognosis for seizure control with antiseizure medications and treatment goal is seizure freedom.

The evidence showed that sodium valproate is the most effective treatment for idiopathic generalised epilepsies compared with other antiseizure medications. However, because of the risks to unborn babies associated with sodium valproate use during pregnancy, the committee agreed that it was important to make separate recommendations for women and girls who are able to have children. In addition, in line with the MHRA, the committee emphasised that long-term treatment with sodium valproate can cause decreased bone mineral density and increased risk of osteomalacia. The committee noted that appropriate supplementation should be considered for those at risk.

The evidence showed that both lamotrigine and levetiracetam were effective at reducing seizures, and the committee agreed that they should be options for first-line treatment in women and girls who are able to have children, and for younger girls with epilepsy likely to continue beyond puberty.

The committee agreed it was important to stress that for some women and girls who are able to have children, sodium valproate may still be an option, but only if the risks and benefits have been thoroughly discussed, other treatments are unsuccessful and safety advice from the MHRA is followed. This should be a shared decision taken by the person with epilepsy and their healthcare professional.

There was some evidence that levetiracetam is of benefit as add-on therapy compared with placebo. The evidence also showed that lamotrigine was associated with fewer side effects leading to treatment stopping and better health-related quality of life than sodium valproate. Therefore, the committee agreed that these medications could be considered as monotherapy or add-on treatment if first-line treatment is unsuccessful.

The included studies did not show a clinically important benefit for topiramate when compared with placebo or valproate; however, the committee noted that this drug is useful in clinical practice. The evidence showed that add-on perampanel is effective for reducing seizures and therefore, based on their expert opinion and the evidence reviewed respectively, the committee agreed that these drugs should be available as a third-line add-on treatment option.

How the recommendations might affect practice

The committee agreed these recommendations will reinforce current best practice.

Return to recommendations

Why the committee made the recommendations

Onset of Dravet syndrome is usually in the first year of life. Children present with prolonged febrile seizures, which may need admission to an intensive care unit. Dravet syndrome can be difficult to diagnose in the first year of life; therefore, the committee emphasised that these recommendations only apply once the diagnosis has been confirmed and discussed with a paediatric neurologist.

Dravet syndrome is complex to treat and the response to treatment is variable. Based on their experience and expertise, the committee agreed that the involvement of a neurologist with expertise in epilepsy is needed to guide the care of people with Dravet syndrome.

There was no evidence for first-line treatments, so the committee based their recommendations on clinical experience and expert opinion. The committee agreed that sodium valproate can be effective at reducing seizures in people with Dravet syndrome because it is successfully used in current practice to treat generalised epilepsies, including Dravet syndrome. The committee acknowledged that sodium valproate should be used with caution in women and girls, only after the risks and benefits have been thoroughly discussed and in line with safety advice from the MHRA. However, they agreed that sodium valproate should be considered as first-line treatment for all people with Dravet syndrome, because there are few effective treatment options and treatment is often started at a young age (under 2 years). In line with the MHRA, the committee emphasised that long-term treatment with sodium valproate can cause decreased bone mineral density and increased risk of osteomalacia. The committee noted that appropriate supplementation should be considered for those at risk.

The evidence suggested that triple therapy with sodium valproate, clobazam and stiripentol was effective at reducing seizures in children and young people whose seizures were previously treated unsuccessfully with clobazam and sodium valproate in combination. Although the evidence was limited, the committee agreed that it supported this as first-line add-on therapy. The committee also highlighted that careful titration of doses and regular review are important because of possible adverse effects such as sedation.

The committee agreed that the NICE technology appraisal guidance on cannabidiol with clobazam for treating seizures associated with Dravet syndrome supports the use of this combination as a second-line treatment option.

There was an absence of evidence to guide further treatment if seizures continue. The committee recommended further treatment options based on their experience and expert opinion, and agreed that these should be considered only under the supervision of a neurologist with expertise in epilepsy or a ketogenic diet team, as appropriate. The use of potassium bromide is unusual in practice, but the committee noted that for some people with Dravet syndrome, it can be effective.

The committee were aware of ongoing trials, but agreed that further research on treating Dravet syndrome when first-line therapy is unsuccessful or not tolerated would be beneficial and they developed a recommendation for research on complex epilepsy syndromes to help inform future guidance.

How the recommendations might affect practice

The recommendations will not change current practice, but will reinforce best practice.

Return to recommendations

Why the committee made the recommendations

Lennox–Gastaut syndrome is a severe developmental and epileptic encephalopathy with onset in childhood. It can be difficult to diagnose, so children may be started on antiseizure medication before a final diagnosis is established.

Lennox–Gastaut syndrome is complex to treat and the response to treatment is variable. Based on their experience and expertise, the committee agreed that the involvement of a neurologist with expertise in epilepsy is needed to guide the care of people with Lennox–Gastaut syndrome.

There was no evidence for first-line treatments, so the committee based the recommendations on clinical experience and expert opinion. The committee agreed that sodium valproate can be effective in suppressing seizures in people with Lennox–Gastaut syndrome because it is successfully used in current practice to treat generalised epilepsy, including Lennox–Gastaut syndrome. They acknowledged that sodium valproate should be used with caution in women and girls, and only if the risks and benefits have been thoroughly discussed and safety advice from the MHRA is followed. However, they agreed that it should be considered as first-line treatment for all people with Lennox–Gastaut syndrome because there are few effective treatment options and treatment is often started at a young age (under 2 years). In line with the MHRA, the committee emphasised that long-term treatment with sodium valproate can cause decreased bone mineral density and increased risk of osteomalacia. The committee noted that appropriate supplementation should be considered for those at risk.

The evidence showed that when used as an add-on treatment, lamotrigine is effective for reducing seizures and drop attacks, therefore the committee agreed that it could be used as second-line therapy, as either an add-on or monotherapy treatment if treatment was not successful or first-line therapy is not tolerated. If used as an add-on therapy, the committee recommended lower initial doses and caution in titration, in line with the BNF. This is because of interactions between sodium valproate and lamotrigine.

There was some evidence that clobazam, rufinamide and topiramate were of benefit in reducing seizure frequency and drop attacks when used as add-on therapy compared with a placebo. In addition, the NICE technology appraisal guidance on cannabidiol with clobazam for treating seizures associated with Lennox–Gastaut syndrome supports the use of this combination as a further treatment option. Therefore, the committee agreed that any of these treatment options could be considered as an add-on treatment if first- and second-line therapy are not tolerated or if seizures continue.

There was an absence of robust evidence to guide further treatment if seizures continue. The committee discussed possible alternative treatment options and, based on their expert opinion and knowledge, agreed that a ketogenic diet or felbamate could be considered, but only under the supervision of a ketogenic diet team or neurologist with expertise in epilepsy, respectively.

The committee noted that although other drugs are used in clinical practice and may benefit some people with Lennox–Gastaut syndrome, it should be highlighted that they can exacerbate seizures in some people.

Given the paucity of published drug trial data in this population, the committee decided to prioritise a recommendation for research on complex epilepsy syndromes including the effectiveness of antiseizure therapies in people with Lennox–Gastaut syndrome when first-line therapy is unsuccessful or not tolerated.

How the recommendations might affect practice

The recommendations are not likely to change current practice, but should reinforce best practice.

Return to recommendations

Why the committee made the recommendations

Infantile spasms are a severe developmental and epileptic encephalopathy that need urgent care. Based on experience and expertise, the committee agreed that guidance should be sought immediately from a tertiary paediatric neurologist, followed by referral if needed. If untreated, long-term risks of infantile spasms include poor neurodevelopmental outcomes, which could be a serious safety concern. Based on their experience and expertise, the committee stressed that prompt diagnosis and treatment is associated with an improved prognosis. Based on best practice and monitoring strategies used in the studies included in the review, the committee agreed that these children should be monitored both during and after treatment for the relapse of spasms and the emergence of other seizure types, as well as for possible side effects related to treatment.

The evidence suggested that first-line treatment combining steroids with vigabatrin is more effective than either steroids or vigabatrin alone in stopping spasms. There was no clear evidence about whether oral or injectable steroids were better, but the committee agreed that oral steroids would be easier to use.

Based on their expert opinion, the committee agreed that steroids may not be suitable for all children under 2 years and that vigabatrin alone should be considered for those at high risk from the side effects of steroid treatment, such as those with neurological impairments and other comorbidities.

There was evidence that vigabatrin alone is effective for children with infantile spasms associated with tuberous sclerosis, so the committee agreed that it should be used as first-line treatment in these children, with high-dose oral prednisolone added on if vigabatrin is ineffective after 1 week.

The committee agreed that there should be a discussion between parents and carers of children under 2 years taking steroids and their healthcare professional, and that information should be given about possible side effects, such as the increased risk of infection, high blood pressure and high blood sugar levels. Information should include, for example, how to reduce exposure to infections such as chickenpox and what to do if the child may have been exposed.

The evidence showed that higher doses of both vigabatrin and oral steroids gave improved freedom from spasms, so the committee agreed that dosages should be increased in line with the guidance in the BNF for children. Based on their expert opinion, the committee agreed that it may be necessary to go above the specified doses of vigabatrin if the spasms do not stop, but this should be carried out with specialist supervision.

There was an absence of robust evidence to guide second-line treatments. The committee agreed possible options based on expert opinion and experience, which should be guided and supervised by a tertiary paediatric neurologist experienced in the care of these children.

The committee agreed to prioritise a recommendation for research on complex epilepsy syndromes including the effectiveness of antiseizure therapies for infantile spasms when first-line therapy is unsuccessful, because there is no controlled trial data to support evidence-based therapy decisions when first-line treatment fails to stop the spasms.

How the recommendations might affect practice

The recommendations will not change current practice, but will reinforce best practice.

Return to recommendations

Why the committee made the recommendations

Children will grow out of self-limited epilepsy with centrotemporal spikes by their early teens. Some only have infrequent seizures, which have little impact on wellbeing. Therefore, not all children and young people and their families will choose antiseizure medication treatment. The committee acknowledged the importance of discussing the balance of risks and benefits of treatment compared with no treatment, with the child or young person and their family or carers, and agreed on some important factors that should form part of a full discussion about treatment. They also agreed that the risk of sudden unexpected death in epilepsy (SUDEP) should be discussed, and reassurance given that this is very rare.

The committee members were confident, based on their experience and knowledge, that current practice using antiseizure medications is effective at controlling seizures in children and young people with self-limited epilepsy with centrotemporal spikes.

There was a lack of evidence on antiseizure medications for self-limited epilepsy with centrotemporal spikes, but because these children and young people usually have focal seizures, the committee agreed to use the evidence on monotherapy for treating focal seizures to inform the recommendations for first- and second-line treatments. This evidence showed that lamotrigine and levetiracetam were continued for longer than other drugs for treating focal epilepsy, suggesting that they may be more effective and better tolerated. However, the evidence also suggested they were not more effective than other drugs in terms of remission at 6 and 12 months, and the evidence for time to first seizure suggested they were less effective than carbamazepine.

The evidence on focal seizures suggested that lamotrigine, levetiracetam and gabapentin may have more tolerable adverse events than other drugs. However, adverse events were reported inconsistently across the studies making comparisons between drugs difficult. The committee also agreed that, for most drugs, adverse events could be managed by careful titration and dosage changes.

Based on the evidence for focal seizures, the committee agreed that lamotrigine and levetiracetam should be considered as first-line treatment options, and carbamazepine, oxcarbazepine or zonisamide as second-line monotherapy treatments.

The committee noted that, in line with guidance from the MHRA, clinicians should be aware of the risks of serious complications associated with carbamazepine and potentially medicines with a similar chemical structure (such as oxcarbazepine and eslicarbazepine acetate) for people of Han Chinese, Thai, European or Japanese family background.

In addition, in line with the MHRA, the committee emphasised that long-term treatment with carbamazepine can cause decreased bone mineral density and increased risk of osteomalacia. The committee noted that appropriate supplementation should be considered for those at risk.

The evidence on self-limited epilepsy with centrotemporal spikes showed that sulthiame is effective for reducing seizures, and so the committee agreed that it should also be available. However, the evidence was limited in quantity and sulthiame is not currently licensed in the UK, so the committee decided that it should be considered as a third-line treatment if licensed options are unsuccessful, and only in consultation with a tertiary paediatric neurologist.

The committee noted that in their experience, carbamazepine, oxcarbazepine and lamotrigine are sometimes associated with increased seizures or the development of another epilepsy syndrome. The committee recognised that only a small number of children are likely to be affected by these problems, but agreed that any change should prompt a sleep electroencephalogram (EEG) to exclude developmental epileptic encephalopathy with spike-wave activation in sleep, which may indicate an atypical form of self-limited epilepsy with centrotemporal spikes. The committee agreed that poor school performance should also prompt a neuropsychology assessment.

Based on their experience, the committee agreed that these children and young people will have varied needs for review, for example, depending on frequency of seizures and choice of treatment. Regular reviews are important to prevent children and young people continuing on unnecessary treatment and allow discussion of stopping treatment. The committee agreed that this should usually happen when the child has been seizure-free for 2 years or at age 14 years.

How the recommendations might affect practice

The recommendations are not likely to change current practice, but should reinforce best practice.

Return to recommendations

Why the committee made the recommendations

Epilepsy with myoclonic-atonic seizures is a rare condition in young children. Successful treatment depends on accurate diagnosis, so based on their experience and expertise, the committee agreed that a tertiary paediatric neurologist should advise on management.

No evidence was identified on treating epilepsy with myoclonic-atonic seizures. Based on their experience, the committee agreed that levetiracetam and sodium valproate should be considered as first-line treatment options because they are used effectively in current practice to treat generalised epilepsy, including epilepsy with myoclonic-atonic seizures. The committee acknowledged that sodium valproate should be used with caution in women and girls, only after the risks and benefits have been thoroughly discussed, other treatments are unsuccessful and MHRA safety advice is followed. However, they agreed that sodium valproate should be considered as a first-line treatment option for girls and women with epilepsy with myoclonic-atonic seizures, with regular review of the risks and benefits, because there are few effective treatment options available, treatment is often started at a young age and most children will outgrow their seizures by their teenage years. In line with the MHRA, the committee emphasised that long-term treatment with sodium valproate can cause decreased bone mineral density and increased risk of osteomalacia. The committee noted that appropriate supplementation should be considered for those at risk.

The committee were aware of studies that showed benefits of a ketogenic diet in these children, and based on this knowledge and their experience, agreed that this should be considered as a second-line add-on or alternative treatment, under the supervision of a ketogenic diet team.

In the absence of evidence, the committee discussed their experience and knowledge of other antiseizure medications for epilepsy with myoclonic-atonic seizures. They agreed that clobazam, ethosuximide, topiramate and zonisamide may be effective. However, these medicines are less commonly used than the first-line treatments, so the committee decided that they could be considered only if first- and second-line options are unsuccessful.

The committee wanted to make it clear that carbamazepine, gabapentin, oxcarbazepine, phenytoin, pregabalin and vigabatrin should not be used for epilepsy with myoclonic-atonic seizures, because they are known to increase the frequency of seizures in this type of epilepsy.

Children can grow out of epilepsy with myoclonic-atonic seizures, so the committee discussed discontinuing treatment. Based on their experience, they agreed that this should be considered if the child is seizure-free for 2 years.

The committee agreed that further research is needed on treating epilepsy with myoclonic-atonic seizures when first-line therapy is unsuccessful or not tolerated and developed a recommendation for research on complex epilepsy syndromes to help inform future guidance.

How the recommendations might affect practice

The recommendations are not likely to change practice.

Return to recommendations

Why the committee made the recommendations

Convulsive status epilepticus is a medical emergency that needs immediate treatment with antiseizure medication. The committee noted the importance of an agreed, individualised emergency management plan for people with epilepsy that should be followed for people experiencing status epilepticus. The management plan should include details of any emergency medicine that has been prescribed, who is trained to use it and when to give it.

The evidence showed an overall benefit for benzodiazepines, but no clear evidence to support a particular drug. The committee agreed that the speed of delivery is more important than the type of benzodiazepine, and that the route of administration is likely to depend on whether the drug is being given in the community or in a hospital. In community settings, medicines are usually given in buccal or rectal forms because intravenous access is not available. The committee discussed that intravenous lorazepam is routinely given in hospitals and agreed that it should be the first-choice treatment in this setting because of its rapid action and because it causes less respiratory depression and sedation than other drugs. Buccal midazolam is currently used in the community, and based on their experience and the evidence, the committee agreed that it should remain as the first choice, with rectal diazepam as an alternative if agreed, based on previous use or if buccal midazolam is unavailable.

The committee agreed that the evidence for further antiseizure medication, if seizures continue after 2 doses of a benzodiazepine, showed a benefit for the intravenous administration of levetiracetam, phenytoin or valproate, but did not favour 1 specific medication over the others. However, based on their experience, the committee agreed that levetiracetam can be quicker to prepare, easier to administer and may be associated with fewer adverse effects than the alternative options, so it is likely to become the preferred second-line treatment. However, because the evidence showed no difference in efficacy, the committee agreed that phenytoin or valproate can also be considered. If status epilepticus does not respond to 1 of these medications, the committee agreed that another second-line medication should be considered.

A small amount of evidence showed benefit for general anaesthesia and phenobarbital if status epilepticus continues after second-line treatment. The committee agreed that these should be considered as third-line treatment options, but cautioned that guidance should be sought from an expert in administering these drugs.

The committee discussed concerns that some causes of status epilepticus may need additional treatment and agreed that awareness of the different circumstances that can cause status epilepticus should be promoted. They also highlighted the need to differentiate non-epileptic attacks from convulsive status epilepticus.

How the recommendations might affect practice

The committee agreed that the recommendations reflect current practice and are not likely to involve a significant change in practice or have a substantial resource impact.

Return to recommendations

Why the committee made the recommendations

The committee discussed the limited evidence available for repeated or clusters of seizures. There was some evidence that benzodiazepines are effective, and the committee agreed that they should be an option. Clobazam and midazolam were given as examples, reflecting the committee's experience and knowledge of current practice. Rectal diazepam is not preferred owing to the route of administration. The committee agreed that further research using clear, consistent definitions for repeated or cluster seizures are needed and developed a recommendation for research on antiseizure medication for repeated or cluster seizures to inform future guidance.

How the recommendations might affect practice

The committee agreed that the recommendations reflect current practice and are not likely to involve a significant change in practice or have a substantial resource impact.

Return to recommendations

Why the committee made the recommendations

No evidence was found on treating prolonged convulsive seizures, defined as seizures that last less than 5 minutes but are more than 2 minutes longer than the person's usual seizures. The committee noted that the definition of prolonged convulsive seizures used to include those longer than 5 minutes, because status epilepticus was defined as seizures that persist for 30 minutes. The International League Against Epilepsy (ILAE) proposed a new definition of status epilepticus meaning that all seizures lasting longer than 5 minutes constitute status epilepticus.

The committee noted that prolonged convulsive seizures should be managed as an emergency. Based on their experience and knowledge, they agreed that benzodiazepines should be a treatment option and that midazolam is often used in current practice.

How the recommendations might affect practice

The committee agreed that the recommendations reflect current practice and are not likely to involve a significant change in practice or have a substantial resource impact.

Return to recommendations

Why the committee made the recommendation

The committee were unable to ascertain clear benefits for ketogenic diets in either adults or children with drug-resistant epilepsy from the evidence available. The committee were also mindful of the potential long-term health drawbacks of ketogenic diets, as well as the high-cost implication of providing ketogenic diets. However, from their experience and knowledge, the committee were also aware that benefits are sometimes seen in clinical practice for a small number of people with drug-resistant epilepsy, including children with certain childhood epilepsy syndromes.

The committee decided against recommending ketogenic diets routinely, but did not wish to remove them completely as an option for specialist consideration in people with few further treatment options. They highlighted the need for high-quality clinical data in this area and agreed that a recommendation for research for children and adults on the effectiveness and long-term tolerability of ketogenic diets would help to inform future guidance.

How the recommendation might affect practice

Ketogenic diets are not routinely offered and so the recommendations are unlikely to have an impact on current practice.

Return to recommendation

Why the committee made the recommendations:

The evidence on surgical interventions showed that resective epilepsy surgery is the most clinically effective treatment for children, young people and adults with drug-resistant focal epilepsy. This was based on the evidence showing better quality of life and lower rates of recurrence after surgery compared with medical care. The committee also considered the relative harms of surgery, such as higher rates of postoperative cognitive deficits and other adverse events. The benefits of surgery were agreed to outweigh these harms, because in many cases, the cognitive effects did not cause significant dysfunction in everyday life, and many of the other adverse events (perioperative infection, bleeding and postoperative changes in mood) were self-limiting. In addition, the committee noted that the risk of harm from surgery needed to be balanced against the risks of ongoing seizures, which include injury, head injury and sudden unexpected death in epilepsy (SUDEP). The committee accepted that the risk of harm may increase as the surgical complexity increases, but agreed that the overall balance in favour of a benefit is likely to apply across most types of epilepsy surgery for both children and adults.

Original health economic modelling was undertaken to assess the cost effectiveness of resective epilepsy surgery in adults. However, insufficient data were available to model cost effectiveness in children. The results of the analysis indicated that resective epilepsy surgery was cost effective in adults. Overall, the committee concluded that resective epilepsy surgery was also highly likely to be cost effective for children because they typically have better outcomes after surgery than adults, and the benefits of surgery are experienced for longer. The committee did note that the cost of assessment for resective epilepsy surgery may be higher in children, but these costs would likely be offset by later savings in the form of fewer outpatient appointments and the benefits observed from resective epilepsy surgery.

No evidence was found on the most effective criteria for referral. However, the committee agreed that because benefits from surgery would outweigh harms across all the populations considered, including improvement in seizure control, potential seizure freedom, better quality of life and reduced risk of epilepsy-related death, all people with drug-resistant epilepsy would benefit from a referral to a tertiary centre for consideration of resective surgery, including those without identified MRI abnormalities.

The committee also discussed whether there were other groups that might benefit from a referral for consideration of surgery. The committee agreed, by consensus, that people with specific MRI abnormalities that might indicate future resistance to antiseizure medication, should be referred to a tertiary centre at diagnosis, rather than waiting until treatment is unsuccessful.

In addition, the committee discussed that, in their experience, people with genetic abnormalities or learning disabilities can sometimes be excluded from referral to a tertiary centre for consideration of surgery. This may happen because it is thought that surgery is not suitable for them, or they might be erroneously considered unable to cope with surgical assessment. The committee agreed they should have treatment in the same way as other people with epilepsy and be referred if indicated.

How the recommendations might affect practice

Only a proportion of people with drug-resistant epilepsy are referred for resective surgery currently, because of relatively low levels of epilepsy surgical treatment provision, as well as lengthy waiting times for presurgical assessments. Therefore, referral of all people with drug-resistant epilepsy to surgical centres will probably lead to an increase in presurgical investigations and surgical procedures. This may necessitate the need for more epilepsy surgical training and a greater investment in epilepsy surgery programmes.

Referral of people whose epilepsy is not drug resistant might increase the burden on tertiary centres, but the number of people in these groups is likely to be relatively small.

Return to recommendations

Why the committee made the recommendations

There was no long-term robust evidence on vagus nerve stimulation in epilepsy. The committee noted that there is variation in current use, but that it tends to be offered when antiseizure medications have failed to control seizures and epilepsy surgery is not suitable or has not been successful. There was no evidence to suggest that use of vagus nerve stimulation should stop for this small group with complex needs and few management options. The committee agreed that the benefits and harms should be discussed with the person because the intervention does not work in all people and is not a risk-free procedure. NICE has published interventional procedures guidance on the use of vagus nerve stimulation in children with refractory epilepsy. The committee agreed that more research is needed in this area and drafted a recommendation for research on the effectiveness of vagus nerve stimulation.

How the recommendations might affect practice

The use of vagus nerve stimulation in practice is currently quite varied. It tends to be a palliative procedure for people with epilepsy that is resistant to treatment and who are not candidates for surgery. The recommendations are not expected to lead to a large change in current practice.

Return to recommendations

Why the committee made the recommendations

How the recommendations might affect practice

The recommendations on providing coordinated care do not reflect routine practice, and therefore will involve a change of practice for the majority of providers.

The guideline committee acknowledged the lack of psychological therapies available to people with epilepsy in current practice, especially children and young people. Although some centres have access to neuropsychology services, this varies across the country and there are often long waiting lists. An improvement in access to psychological services for people with epilepsy and depression or anxiety may result from highlighting existing NICE guidance. No significant resource impact is expected as a consequence of the recommendations on support and treatment.

Return to recommendations

Why the committee made the recommendations

How the recommendations might affect practice

The committee highlighted that there is variation in the information currently discussed about the risk of premature death including SUDEP. The recommendations will have an impact on clinical practice by focusing on specific modifiable risk factors and working with people to reduce risk and prevent premature death. The recommendations will create a framework for discussions between healthcare professionals and the person with epilepsy and their families and carers to help inform decisions.

The recommendation on night-time supervision reflects current good practice and is unlikely to change current practice.

Return to recommendations

Why the committee made the recommendations

The clinical evidence on epilepsy specialist nurses was limited in quality and quantity, and the committee acknowledged that research in this area can be challenging because of limited funding and difficulties conducting high-quality randomised studies, given that epilepsy specialist nurses are already embedded in many services. However, because there was economic evidence of cost savings both long term and within the first year, the committee agreed that people with epilepsy should have access to an epilepsy specialist nurse. Epilepsy specialist nurses are already embedded in practice and the committee agreed that they play a vital role in supporting other healthcare professionals in primary, secondary and tertiary care, as well as educational and social care settings. They are specialised in supporting children, young people and adults with all aspects of living with epilepsy, providing support with information giving, advice on administering medications, care planning, management of side effects and the impact of epilepsy on daily activities. In addition, epilepsy specialist nurses may identify problems that had been previously unnoticed, such as long-standing side effects of antiseizure medications.

The committee used the evidence to determine the common features of clinically and cost-effective epilepsy specialist nurse interventions. Based on the clinical evidence, they agreed that interventions should include emotional wellbeing and self-management strategies to support improvements to day-to-day living and health. Based on economic evidence and their own expertise, the committee agreed that people's needs for information and care planning may vary over time and more contact may be needed when seizures are ongoing. The economic evidence showed that epilepsy specialist nurse-led interventions are likely to be cost saving and cost effective when delivered twice a year to people with epilepsy and after attending an emergency department. The committee recommended this approach for people with ongoing seizures or after an emergency department visit. Although it was a wider population (approximately 300,000 people in England) to that explored in the economic analyses, they are much more likely to be in regular contact with healthcare services, with a quarter of this group making at least 1 emergency department visit per year. Nearly all will have some sort of outpatient appointment in a hospital setting, often with an epilepsy specialist nurse where these are available. Therefore, a large number of these epilepsy specialist nurse appointments will already be taking place or will replace appointments with other healthcare practitioners. Although there will be additional total healthcare appointments as a result of these recommendations, the increase would be much smaller in this group than for all people with epilepsy, and costs would be potentially regained within the first year.

How the recommendations might affect practice

The recommendations reflect current practice available in some services, but there is variation in epilepsy specialist nurses' involvement across different settings. Some services may need to make changes to practice, such as hiring epilepsy specialist nurses or changing how they work, but this should lead to a number of advantages, including improved patient satisfaction and emotional wellbeing, greater consistency in provision and care, improved access to epilepsy specialist nurses and potentially cost savings.

The involvement of an epilepsy specialist nurse is likely to result in cost savings by reducing the overall use of healthcare services especially in terms of reduced emergency department visits and the subsequent length of hospital stay.

Return to recommendations

Why the committee made the recommendations

The committee acknowledged that many of the recommendations in the NICE guideline on transition from children's to adults' services for young people using health or social care services are directly applicable to young people with epilepsy and their families or carers. So, they reviewed the evidence on identified themes specific to people with epilepsy that are not covered by that guideline.

The committee noted that a young person's transition should be tailored to their needs and that this should be recognised by both paediatric and adult services. The committee discussed the value of reviewing diagnosis and management at transition and agreed that this should involve both paediatric and adult multidisciplinary teams working together with the young person and their family or carers. The evidence found that young people who engaged with a multidisciplinary team felt more confident and able to communicate their needs, so the committee stressed that planning and decision making should be patient-centred, with young people and their families and carers (if appropriate) fully involved in discussions about their transition and ongoing care.

The committee acknowledged that transition can be a distressing time for young people with epilepsy and their families and carers. The evidence showed a lack of clarity in communication and the information given to young people and their families and carers at transition, for example they lacked clear information about the risks of SUDEP and unplanned pregnancy. The committee agreed that clear and balanced information is vital during transition and, based on their experience, listed some key areas that should be covered in discussions. In particular, stigma around epilepsy was highlighted as an important topic to raise so as to give the young person the opportunity to discuss their experiences and for support to be provided. The evidence showed that stigma can have a significant impact on young people's everyday activities, educational achievement and engagement with healthcare professionals. The evidence also suggested that repeating information at intervals during transition would help young people understand and retain key information. Based on the evidence and their experience, the committee highlighted the importance of providing information in a suitable format for the person, using language they understand and that is appropriate for their developmental age, and avoiding technical terms.

The evidence showed that young people with epilepsy and a learning disability, and their parents, often struggled with transition, and had difficulty finding information and understanding the changes in service provision. They noted that transition was often not planned or happened much later than for young people without learning disabilities. Young people with learning disabilities may have complex needs, and transition to adult services may need more planning and involve other specialties, such as a learning disabilities multidisciplinary team and child and adolescent mental health services. Based on their experience, the committee agreed that it would be beneficial to start transition planning earlier than usual for young people with complex health or social care needs, including people whose seizures were not fully controlled by their treatment or those with a learning disability, to allow time for care packages to be set up. They agreed that the timeframe would depend on individual circumstances.

How the recommendations might affect practice

The recommendations reflect current best practice so the committee agreed there should be no resource impact.

Return to recommendations