Article Text

Download PDFPDF

Neonatal seizures—part 2: Aetiology of acute symptomatic seizures, treatments and the neonatal epilepsy syndromes
  1. Anthony R Hart1,2,
  2. Elizabeth L Pilling2,
  3. James J P Alix3
  1. 1Department of Paediatric and Neonatal Neurology, Sheffield Children's Hospital NHS Foundation Trust, Ryegate Children's Centre, Sheffield, South Yorkshire, UK
  2. 2Department of Neonatology, Sheffield Teaching Hospitals NHS Foundation Trust, Sheffield, UK
  3. 3Department of Clinical Neurophysiology, Sheffield Teaching Hospitals NHS Foundation Trust, Royal Hallamshire Hospital, Sheffield, UK
  1. Correspondence to Dr Anthony Hart, Department of Paediatric and Neonatal Neurology, Sheffield Children's Hospital NHS Foundation Trust, Ryegate Children's Centre, Tapton Crescent Road, Sheffield, South Yorkshire S10 5DD, UK; anthony.hart{at}


Most neonatal epileptic seizures are provoked by an underlying condition or problem—‘acute symptomatic seizures’. However, a few neonatal epilepsy syndromes exist, and these are defined by the constellation of seizure types, EEG findings and family history seen. Making an accurate diagnosis of an epilepsy syndrome can help direct investigations, treatment options and provide prognostic information. This article discusses the investigative approach and treatments for neonatal epileptic seizures, including the neonatal epilepsy syndromes.

  • Neurology
  • Neonatology
  • Clin Neurophysiology

Statistics from


Once a clinician has diagnosed neonatal epileptic seizures and defined the seizure type, the next step is to decide whether they are acute symptomatic seizures or part of an epileptic syndrome. Most neonatal epileptic seizures are acute symptomatic;1 however, a small number of neonatal epilepsy syndromes exist. Diagnosing neonatal epilepsy syndromes helps the formulation of investigation plans, choice of antiepileptic drugs and the duration of treatment. Prognostic information can also be gathered. This article reviews neonatal acute symptomatic seizures and epilepsy syndromes, with a particular focus on aetiologies, investigations and treatments.

Acute symptomatic seizures

Acute symptomatic seizures occur at the time of, or near to, either a systemic or brain insult, including metabolic derangements.2 The timing of seizure onset may help to determine the possible aetiology.3 Table 1 summarises the typical causes of neonatal seizures by their time of appearance and includes the neonatal epilepsy syndromes for comparison. Other information gained from family, antenatal, birth and postnatal histories can also help determine aetiology. Hypoxic ischaemic encephalopathy (HIE) is the commonest cause of neonatal acute symptomatic seizures.1 Clinicians should also be aware that fetuses with metabolic or neurological disorders may decompensate during labour and delivery and experience seizures. Thus, beware assuming all encephalopathic neonates with seizures have HIE.

Table 1

Aetiology by predominant time of onset

Where the history, examination and initial screen for acute symptomatic seizures do not reveal a cause, further investigations may be warranted. We suggest some investigations in table 2. A few points on investigations are worth noting: serum glucose samples should be taken before the lumbar puncture (LP) to help identify glucose transporter type 1 deficiency syndrome, where a low cerebrospinal fluid (CSF) glucose (below 2.2 mmol/L) or CSF:plasma ratio (<0.45) is noted in the absence of infection.4 If the serum samples are taken after the LP, the plasma glucose will be falsely high because of the stress response and the ratio will subsequently be meaningless. The LP will then need to be repeated. Similarly, if an LP is being undertaken in a neonate with seizures, then paired CSF and plasma amino acids should be considered to look for glycine or serine encephalopathies. This may also prevent the need to repeat the LP at a later date.

Table 2

Suggested investigations useful in determining the cause of neonatal seizures

Investigations for vitamin-responsive epilepsies and a therapeutic trial of vitamins should be given for refractory neonatal seizures where no other cause is known. The recommended doses are given in table 3. Pyridoxine can be given intravenously but may cause apnoea, so careful observation is required. Pyridoxine was not traditionally thought to treat pyridoxal phosphate-dependent seizures, and thus recommendations advised starting enteral pyridoxal phosphate first because it treated both conditions.5 ,6 However, recent cases have highlighted cases in which pyridoxal phosphate worsened, and pyridoxine improved, seizures in individuals with mutations in the pyridoxal phosphate (PNPO) gene.7 This has confused matters considerably, and the authors’ practice is now to try two doses of intravenous pyridoxine first and then swiftly move on to enteral pyridoxal phosphate where no response is seen. In an ideal world, investigations would be completed quickly and then treatment subsequently instigated with minimum delay. Given the difficulties in achieving this, together with risk of the possible deterioration in a neonate's condition, we recommend vitamin treatments be instigated immediately and investigations completed as soon as possible. Biotin and follinic acid (calcium follinate) should be commenced at the same time as the B vitamins and subsequently stopped when investigations rule these conditions out.6

Table 3

Recommended doses for vitamin-responsive epilepsy in neonates

The neonatal epilepsy syndromes

The neonatal epileptic syndromes include

  • benign neonatal seizures (fifth day fits)

  • benign neonatal familial seizures

  • early infantile epileptic encephalopathy (EIEE) (Ohtahara syndrome)

  • early myoclonic epileptic encephalopathy (EMEE)

  • migrating partial seizures of infancy

A summary of the neonatal epilepsy syndromes, including EEG findings, is found in table 4.

Table 4

A summary of the clinical features of the neonatal epilepsy syndromes

Benign neonatal seizures are colloquially called fifth day fits. They are characterised by unilateral, bilateral or migrating clonic movements of limbs and face lasting minutes and occurring in clusters or status epilepticus.8 The seizures may also be associated with apnoea.9 They are not associated with a family history, so this epilepsy syndrome is often a diagnosis of exclusion after other causes of acute symptomatic seizures prove negative (table 2), the seizures have disappeared and the baby looks well. The acute management of acute seizures may be with phenobarbital, benzodiazepines and/or phenytoin. Long-term treatments are not necessary as the seizures disappear within a day.8 ,9

Benign familial neonatal seizures are frequent, short tonic seizures with posturing, apnoea, vocalisations, eye movement abnormalities and autonomic changes. Focal or generalised clonic seizures can also be seen alongside the tonic seizures, but pure clonic seizures are rare.10 The key to the diagnosis is the seizure type, timing of onset and the family history. The latter may not be known in the acute phase, leading to neonates being investigated for acute symptomatic causes (table 2). The acute treatment is with phenobarbital, benzodiazepines and/or phenytoin. Regular preventative medication should be considered where seizures are frequent, and options include carbamazepine, sodium valproate or levetiracetam.

Early infantile epileptic encephalopathy (Ohtahara syndrome) is an uncommon epileptic syndrome with poor prognosis. Onset of seizures may occur in utero or postnatally. Tonic seizures and epileptic spasms, occurring in isolation or clusters, are the commonest seizures and may be symmetrical or asymmetrical. They classically occur in both the awake and sleep states and are associated with bursts on the EEG with suppression between them lasting typically 3–5 s, but sometimes longer.11

Developmental brain abnormalities are the commonest cause,11 some of which may only be demonstrated at postmortem.12 ,13 A number of gene mutations are also associated with Ohtahara syndrome, including in the STXBP1, ARX and SCN2A genes.14 Investigations will include MRI, consideration of metabolic investigations (table 2) and a review by the genetics team. There are no consistently effective treatments: vigabatrin may help to some degree15 and a therapeutic trial of vitamins should be tried.16 ,17 Chloral hydrate, an old-fashioned drug now used for sedation but with antiepileptic properties, has also helped in an isolated case.18 Where a surgically resectable lesion is present, epilepsy surgery can stop or reduce seizures.19 Where no surgically resectable lesion is present, the prognosis is poor.

Early myoclonic encephalopathy (EME) is similar to Ohtahara syndrome, but the predominant seizure type is myoclonic, rather than tonic, seizures. The epileptic seizures may come in clusters or occur continuously. Seizure onset is shortly after birth. The EEG shows burst suppression, which is more apparent in sleep than wakefulness, in contrast to Ohtahara syndrome where the pattern is continuous.11

There are many causes of EME. Autosomal-recessive genetic and metabolic aetiologies are most common, with structural lesions less prevalent than Ohtahara syndrome.11 The investigative approach will initially involve metabolic tests and neuroimaging (table 2). No medications are consistently effective. Vigabatrin may worsen and the ketogenic diet improve seizure frequency caused by non-ketotic hyperglycinaemia.20 ,21 We advocate trying vitamins early. The prognosis is poor.

Migrating partial seizures in infancy is an epileptic syndrome usually presenting later in infancy, which can be seen in the neonatal period. The seizures are focal clonic and/or tonic and may be associated with autonomic features (table 4). They may migrate from one region of the body and EEG to another. The cause is unknown, but genetic aetiologies exist including in the KCNT1 potassium ion channel.22 The vast majority of antiepileptic drugs are ineffectual.23 ,24 Stiripentol and clobazam, and possibly also levetiracetam, have been described to gain complete control of seizures when given together,24 ,25 but did not improve the EEG abnormalities.24 Potassium bromide has gained complete control in one patient26 but not in others.23 Rufinamide has reduced seizure frequency by 50% in 2 of 5 patients,27 and acetazolamide has gained complete control of apnoeic seizures.23 Quinidine has been used successfully in one child.28 We have had success with a combination of phenytoin, levetiracetam and acetazolamide, with seizures recurring when any of the three drugs was withdrawn or phenytoin levels fell. A therapeutic trial of vitamins should be attempted early. The ketogenic diet has also not been effective in published cases,24 although we have cases that have responded.

Investigations for the aetiology of the neonatal epilepsy syndromes

The investigations will depend on the clinical picture. For example, a child who presents on day 5 in status epileptics is likely to have a septic screen and possible neuroimaging. But if the seizures quickly respond to treatment and the baby looks otherwise well, further investigations may not be warranted once the epilepsy syndrome is diagnosed. In contrast, the investigations for Ohtahara syndrome and EMEE will be more extensive, including metabolic investigations, neuroimaging and genetic testing.

Treatments of neonatal epileptic seizures

Acute epileptic seizures

No nationally agreed guidelines exist on the acute treatment of neonatal seizures. Phenobarbital is used as first-line treatment in many neonatal units. Animal experiments have noted it is associated with neuronal apoptosis in rats, as are benzodiazepines.29 Phenobarbital also causes electroclinical dissociation in human neonates30 and is implicated in reduced cognitive and memory abilities in young children when given long term.31 ,32 On the other hand, early administration in neonatal HIE may improve outcome.30 Without good evidence that other antiepileptic drugs are better and safer, it seems logical to retain phenobarbital for first-line use (figure 1).

Figure 1

Suggested algorithm for the treatment of neonatal epileptic seizures. BD, twice a day.

Phenytoin has the same efficacy as phenobarbital, but therapeutic-level monitoring is required and it has a worse side effect profile.30 In our centre, phenytoin is not used in neonates requiring inotropic support because it causes hypotension and cardiac arrhythmias. Benzodiazepines or lidocaine may also be of use. Midazolam can suppress the conscious level and respiratory drive, leading to ventilation and, in our experience, hypotension. Lidocaine can cause arrhythmias, and we try to avoid this in neonates who have had phenytoin.

Newer antiepileptic drugs, such as topiramate and lamotrigine, are also being used in small numbers of neonates, but data on their effectiveness are limited.30 Levetiracetam is being used increasingly, particularly because of its intravenous and enteral preparations and data showing loading doses are safe and effective with minimal side effects.33 ,34 However, the correct loading dose is not known with studies varying from 10  to 50 mg/kg.34 ,35 Disagreement also exists about whether maintenance doses should be given twice or three times a day.35

The epilepsy syndromes

The treatment of the neonatal epilepsy syndromes will be guided by the specific syndrome. For example, there is no need for long-term therapy in benign neonatal seizures, but in migrating partial seizures of infancy, for example, you may elect to try a trial of vitamins, acetazolamide for seizures with apnoea, and a combination of levetiracetam, stiripentol and clobazam for other seizure types. We would strongly advise consulting expert neurological advice in situations where a complex epilepsy syndrome is considered.


Most neonatal epileptic seizures are acute symptomatic, resulting from conditions like HIE. Investigations should be targeted, depending on what the differential diagnoses are following a full history and examination. Little evidence exists on which antiepileptic drugs are most effective for acute symptomatic seizures or have the least short-term and long-term side effects. No national consensus exists on the correct pathway to treat neonatal seizures.

A limited number of neonatal epileptic syndromes exist, some with good and some with dire prognosis. A careful history, noting the seizure type and EEG pattern, help to identify these syndromes. Where a benign syndrome is found, treatment for status epilepticus or frequent seizure clusters can be instituted, followed by a rapid withdrawal of medication. For the more severe syndromes, few treatment options are likely to be effective and prognosis is poor if no surgically resectable lesion is found. Vitamin-responsive seizures should always be considered early in refractory neonatal seizures and specialist advice sought.


View Abstract


  • Note Neonatal seizures—part 1 was published in Education & Practice, August 2015. doi:10.1136/archdischild-2014-306385.

  • Contributors ARH performed the literature review and wrote the article, providing neonatal and neurological viewpoints. JJPA reviewed and edited the article, particularly with reference to neurophysiology data. ELP reviewed the article and provided comments from a neonatal perspective.

  • Competing interests None.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.