eMedicine Specialties > Neurology > Pediatric Neurology

Benign Neonatal Convulsions

Author: Nancy Theresa Rodgers-Neame, MD, Assistant Professor, Department of Molecular Pharmacology and Physiology, University of South Florida; Director, Florida Comprehensive Epilepsy and Seizure Disorders Program
Contributor Information and Disclosures

Updated: Apr 8, 2009

Introduction

Background

Benign neonatal convulsions are defined as seizures with onset after birth through day 28 in an otherwise healthy child with no other known medical or neurological problems. Such cases may be familial or isolated. Psychomotor development should be normal for a full-term or near full-term infant with benign convulsions. Between seizures, findings on neurologic examination should be normal. Clinically, the seizures are frequent and brief, occasionally occurring many times within a day. The brief seizures are followed by a short or no postictal state. The episodes usually resolve within days but may continue for several months and have no neurological sequelae.

The occurrence of status epilepticus is common in benign idiopathic neonatal convulsions (BINC) but is uncommon in benign familial neonatal convulsions (BFNC). Because benign idiopathic neonatal convulsions is a diagnosis of exclusion, it is nearly always made in retrospect, when the seizures spontaneously resolve and the infant is found to have neurologically normal development.^1,2,3,4

At the outset, considering how broadly to define benign neonatal convulsions is important: for example, whether to include those with myoclonic or partial onset components or those with a known or treatable etiology. Certainly, multiple presentations of seizures that may have a benign long-term outcome are possible in the neonatal period.^3,7 Definite advantages exist in approaching the subject from each position. Too broad of a definition in a research situation can lead to confusion when searching for a common pathology. Too narrow of a definition in the clinical setting may result in confusion about a clinical diagnosis. Later, when the mechanisms are more well defined, broader groups not meeting the initial criteria may exist.

Clinically, the more important considerations are taking an appropriate approach to the patient and family, making the correct diagnosis, and pursuing treatment options concordant with the situation. Sometimes, the correct clinical plan may include a decision not to treat a benign condition with medications that often are not so benign. It should also be emphasized that a definitive diagnosis may take some time, given the often retrospective nature of the diagnosis.

Pathophysiology

The genetics of benign familial neonatal convulsions is currently an area of active investigation. Inheritance is autosomal dominant. Loci on chromosome arm 20q have been identified for most families. At least one family was identified to have a locus on chromosome arm 8q. Some of these loci have been identified further as specific mutations in the KCNQ2 and KCNQ3 M-type potassium channel proteins.^{8,9,10,11,12,36,37,38} The specific location of the mutation appears to vary from family to family and at least 1 family has been noted to include an increased incidence of rolandic epilepsy.⁵

Several additional genes have been associated with benign familial neonatal convulsions in single families, including KCNQ5 M-type potassium channel in one family.¹³ Another family has been noted to have abnormalities in the acetylcholine alpha-4 receptor subunit, which also has been associated with autosomal dominant nocturnal frontal lobe epilepsy.^14,15

Expression of the mutated genes in xenopus oocytes has provided some insight into how the potassium channel mutation leads to lowering of the seizure threshold. The potassium current was reduced in the channel expressed by the mutated gene to 5% of that in the channel expressed by the normal gene. However, voltage sensitivity and kinetics were not affected. The effect is therefore to impair repolarization of the neuronal cell membrane, leading to hyperexcitability of the central nervous system.^16,17

Given the severity of the impairment to the M-type potassium channel, that these seizures are difficult to treat is not surprising, since no currently used antiepileptic medications are known to increase the efficiency of the potassium channel. What is surprising is the self-remitting nature of the condition, that many individuals never have another seizure, and that the profound abnormalities of the voltage-gated potassium channel do not appear to compromise the nervous system in any other way. Possibly, some intrinsic method exists for up-regulating expression of the normal potassium channel genes, or the neurons may find other ways of normalizing the hyperexcitability, but these theories remain to be demonstrated.^18,19,20

A number of cases have been reported where benign idiopathic or familial neonatal convulsions have preceded the development of epilepsy later in life. Similarly, perhaps, febrile seizures early in life may predispose to later development of epilepsy. Given the polygenic etiology of susceptibility to epilepsy, it is not surprising that an abnormality in part of the system maintaining homeostasis within the neuron should render the neuron more prone to dysfunction.^21,22,23,24

A complicating factor is that in neonates the action of the GABA-A receptor is excitatory rather than inhibitory in the brain.^25,26,27,28

The pathophysiology of BINC has been less well defined and remains somewhat elusive. One issue is that the neonatal brain is more prone to seizures, which has been demonstrated in a number of experimental systems.^25,26,27,28

Several etiologies have been proposed as a result of isolated findings of lowered zinc level in the cerebrospinal fluid and low levels of vitamin B-6.²⁹ Both of these compounds are important cofactors in ligand-gated ion channel function. However, these findings have not been robust, and the search continues. These seizures are likely also linked in some way to ion channel dysfunction as is found in the familial seizures, but they may be caused by multiple etiologies or occur as a multigenomic entity.^30,31,32 These sorts of multifactorial etiologies are more difficult to define precisely. More research is needed in this area, and apparent monogenic diseases, such as benign familial neonatal convulsions, provide important insight into more complex etiologies.

Frequency

United States

Benign neonatal convulsions in the United States are uncommon, ie, not rare but not common, either. Underreporting is likely an issue. Seizures that resolve in the early months of life without sequelae and normal neonatal development are often lost to follow up. Exact frequencies are undetermined. Families identified with the familial form thus far have been primarily of western European origin, although one report from Japan exists.^33,34 Part of the reason for this is likely the stability of reporting resources in European countries.

International

Benign neonatal convulsions also are uncommon internationally. Families identified with the familial form thus far have been primarily of western European or Japanese origin.^33,34 This is certainly an artifact of observation rather than occurrence.

Mortality/Morbidity

The risk of seizures later in life is 11-16% in benign familial neonatal convulsions and somewhat less in benign idiopathic neonatal convulsions, perhaps as low as 2%. Other reported problems have been sporadic and within the incidence range expected for the general population.²⁹

Race

All cases to date have been reported in families of European or Japanese origin.^33,34

Sex

In benign familial neonatal convulsions, the frequency in males is equal to that in females, compatible with simple autosomal dominant inheritance.²⁹

In benign idiopathic neonatal convulsions, males are affected somewhat more frequently than females (62%) in examined cases (N=199).²⁹ With such a small number of cases reported, this may be due to reporting bias or simple sampling error, or it may represent a real difference in frequency.

Age

• In benign idiopathic neonatal convulsions, patients are aged 1-7 days at onset, with day 5 the most frequently reported day of onset. The frequent onset on the fifth day of life is responsible for the term fifth day disease or fifth day fits, which continues to be used in the pediatric literature. However, in actuality, fifth day fits are most likely seizures that were reportedly linked to the use of hexachlorophene (pHisoHex), which now has been discontinued.³⁵
• Patients may be slightly older at onset in benign familial neonatal convulsions, with some patients in previously identified families several weeks old. Characteristically, the onset of benign familial neonatal convulsions occurs when neonates are aged 2 days.^1,2,4
• Interestingly, unaffected family members of patients with benign familial neonatal convulsions have a higher-than-expected risk of developing epilepsy in later life. Presently, family studies have not clarified whether these relatives always share the genetic defect in the potassium channel.^3,29

Clinical

History

• The history should be free of suspicion of causative elements for neonatal seizures.
  • Such elements include prenatal or perinatal stroke, perinatal asphyxia, fever, persistent lethargy, pertinent maternal illness or drug abuse, signs of metabolic dysfunction, or neurological abnormalities.
  • Other criteria that have been noted but may be of limited value are a lack of neurological or genetic abnormalities in siblings or other family members or metabolic dysfunction of any cause. A 5-minute Apgar score less than 9 also has been suggested but is almost certainly too stringent to require for diagnosis.
• Diagnostic criteria have been suggested by Miles and Holmes²⁹ as well as Plouin³ for both benign familial neonatal convulsions and benign idiopathic neonatal convulsions.
  • Diagnostic criteria - Benign idiopathic neonatal convulsions^3,29
    • Infants born after 39 weeks' gestation
    • Apgar score of 9 or more at 5 minutes (lower scores should not exclude the diagnosis if other criteria are met)
    • Presence of a seizure-free interval between birth and the onset of seizures
    • Clonic and/or apneic seizures
    • Negative findings on evaluation for alternative etiology
    • Normal developmental and intellectual outcome (largely determined in retrospect)
    • Lack of seizures beyond the neonatal period
  • Diagnostic criteria - Benign familial neonatal convulsions^3,29
    • Normal neurologic examination findings
    • Negative findings on evaluation for alternative etiology
    • Normal developmental and intellectual outcome (retrospective criteria)
    • Positive family history of newborn or infantile seizures
    • Onset of seizures during neonatal period or early infancy

Physical

Physical and neurologic examination findings should be normal during the interictal period.

Causes

• Several causes have been proposed for benign idiopathic neonatal convulsions, including rotavirus infection, low CNS zinc levels, and vitamin B-12 deficiency. None of these causes has been confirmed. The more likely explanation is the presence of a self-limited malfunction in one of the ligand-gated or voltage-gated ion channels.³
• Most families in which benign familial neonatal convulsions occurs have abnormalities in the genes coding for the KCNQ2 and KCNQ3 potassium channels.^39,40
  • This defect leads to abnormal repolarization of the neuronal membrane and likely causes the neonatal seizures.
  • The real puzzle is why this profound abnormality in membrane polarization does not lead to more problems in later life or persistent seizures extending from the neonatal period. This may be evidence that the drive toward homeostasis in the brain is strong, with redundant systems capable of maintaining a seizure-free state until more than one system is affected, or that systems are affected that do not have the redundancy of the voltage-gated potassium channels. Moreover, the normal potassium channels may be up-regulated to accommodate for the deficiency in function of the abnormal channels.
  • Clearly, the immature infant brain is different electrophysiologically during early development. GABA has a seemingly paradoxical excitatory effect.²⁵ Glutamate synapses are slow to develop, and there is delayed expression of the K⁺/Cl^- cotransporter KCC2 and NKCC1.²⁶ The primary inhibition in the neonatal brain is presynaptic rather than postsynaptic. As the brain matures and expression of postsynaptic inhibitory and excitatory processes develop, the maintenance of neuronal homeostasis as well as postsynaptic EPSPs and IPSPs gradually approach the adult state. Since all of these processes are under development at the same time as neonatal convulsions appear, it is likely that neonatal seizures are affected by the normal developmental sequence of the other neurotransmitter systems.

• Search for CME/CE on This Topic »

• Neonatal Seizures (Neurology)

• Brain and Nervous System Center
• Seizures in Children Causes
• Seizures in Children Symptoms
• Seizures in Children Overview
• Seizures in Children Treatment