Febrile seizures are the most common convulsive disorder through the lifecycle. In the United States and Western Europe, about 4% of children under the age of 5 years experience a febrile seizure^47,65; the frequency has been reported to be higher in some Asian populations (8% in Japan and 14% in Guam).^58,64 Febrile seizures are characterized by their benign prognosis.

Until the mid-19th century, febrile seizures were not recognized as distinct from other seizures. Before that time, infantile convulsions were thought to be the result of irritation of the central nervous system (CNS). Presumed causes included gastrointestinal irritation, teething, improper diet, and fever. Treatment was nonspecific and focused on relieving the symptoms. By the mid-19th century, this began to change, and treatment was directed at underlying causes rather than symptoms. Beginning in the latter part of the 19th century, fever came to be regarded as a primary factor in producing infantile convulsions. This emphasis on fever may have been related to the introduction of the thermometer into clinical use around this time.²⁶

Throughout the late 19th and early 20th centuries, any infantile convulsion was considered to be serious and potentially fatal. This was because of a failure to separate febrile from neonatal and other symptomatic seizures.

By the turn of the century, as it was gradually understood that seizures were a symptom and not a disease, treatment of the underlying cause became important. Unfortunately, very few effective treatments were available.

In the mid-20th century, studies focused on febrile seizures were published by Lennox⁴⁰ and Livingston.⁴¹ These investigators found that a benign outcome was predicted by onset between 1 and 3 years, generalized seizure of short duration, single or few episodes, family history positive for febrile convulsions, male sex, and normal findings on the electroencephalogram (EEG). Increased likelihood of later epilepsy was associated with onset before 1 year of age, prolonged or partial convulsions, multiple seizures, abnormal birth history, family history of epilepsy, female sex, and abnormal EEG findings. Subsequent studies have confirmed many of these factors as predictive, albeit not very accurate predictors.

By the 1970s, three large population-based studies had been published that demonstrated a much lower risk for epilepsy following febrile seizures than previously proposed. These studies had very similar results, and formed the foundation for our current understanding of febrile convulsions. In their study of 18,500 children, van den Berg found a 2% incidence of febrile seizures with a 3% rate of later epilepsy.⁶⁵ In 1976, the National Institute of Neurological Disorders and Stroke (NINDS) Collaborative Perinatal Project, which followed 54,000 infants prospectively, found an incidence of almost 4% for febrile seizures.⁴⁷ The developmental outcome was unaffected, and the overall incidence of later epilepsy was relatively small (2%). Verity followed a British cohort of 14,676 children.⁷⁰ The incidence of at least one febrile seizure was 2.7%, and 2.3% of those with febrile seizures later developed epilepsy.

In 1980, the National Institutes of Health (NIH) Consensus Panel defined febrile seizures as “An event in infancy or early childhood, usually occurring between three months and five years of age, associated with fever but without evidence of intracranial infection or defined cause. Seizures with fever in children who have suffered a previous nonfebrile seizure are excluded. Febrile seizures are to be distinguished from epilepsy, which is characterized by recurrent nonfebrile seizures.”²¹

This definition excludes any seizure that occurs in the presence of neurologic illness, such as encephalitis, meningitis, or toxic encephalopathy. The prognosis for such seizures is different from that of febrile seizures, because the underlying illnesses can potentially cause CNS sequelae. Seizures caused by mild encephalopathies may be included within this definition, such as those associated with Shigella, a bacterium that produces a neurotoxin. Febrile seizures are difficult to define in malaria-endemic areas where all children are infected from time to time, many have elevated temperatures, and the severity of brain involvement is often measured by severity of the seizure disorder as well as level of consciousness.⁴⁸

Simple febrile seizures are those that last less than 15 minutes and are generalized. Complex febrile seizures are prolonged, multiple within 24 hours, or partial. Seizures in either of these subgroups may be accompanied by a pre-existing neurologic abnormality or a family history of either febrile or afebrile seizures.

Febrile status epilepticus follows the usual definition of 30 minutes of continuous or intermittent convulsive phenomena without the return of consciousness.

Discussion remains about the age range used in the NIH Consensus Panel definition. Many experts feel that ages between 3 and 6 months is rather young for the onset of a first febrile seizure. In fact, it may herald the onset of Dravet syndrome, a recently described epilepsy syndrome. As well, children have febrile seizures beyond the fifth birthday, and it has been suggested that the upper age limit be the sixth birthday. Another area of contention is a standard definition of “fever” and how a child’s temperature should be measured (i.e., oral, rectal, axillary, or tympanic measurement). Based on the childhood vaccine trial literature, our definition of an elevated temperature is 38°C per rectum or 38.5°C orally. However, we recognize that such accuracy is not always available.

Seizures are not considered to be “febrile seizures” when they occur with fever but in the context of exposure to proconvulsant drugs such as diphenhydramine, tricyclic antidepressants, amphetamines, and cocaine.

It has been suggested that febrile seizures should be categorized as acute symptomatic because of the coincident febrile
illness. We feel this is misleading. It is our opinion that fever is a systemic trigger that provokes the seizure, as contrasted to the underlying severe brain dysfunction found in most conditions included in the acute symptomatic category. In addition, the outcome for febrile seizures is so much better than for seizures associated with many acute symptomatic causes that it is misleading to lump them together.

In the United States, South America, and Western Europe, febrile seizures are experienced by 2% to 5% of all children before the age of 5 years.^28,47 They are reported to be even more common in certain Asian countries.^58,64

The first febrile seizure is complex (lasting more than 15 minutes, partial, or multiple within 24 hours) in about 30% to 40% of cases.⁶⁹ The peak incidence is in the second year of life, with the average age of onset between 18 and 22 months. Febrile seizures are slightly more common in boys.

Risk factors for a first febrile seizure have been studied in comparison to febrile controls without seizures (Table 1). A higher temperature was a risk factor for febrile seizures in two studies,^9,52 and a history of febrile seizures in the immediate family was noted in another study.¹⁵ Low plasma ferritin, reflecting poor iron status, also increased the risk. When both febrile and afebrile children were used as controls, in addition to a family history of febrile seizures, neonatal discharge at 28 days or later, parental report of slow development, and day care attendance were also risk factors.¹⁶

Approximately 40% of children with a first febrile seizure will experience one or more recurrences. The risk of a second seizure after a first is about the same as a third febrile seizure after a second. Most recurrences (75%) are within 1 year of the first febrile seizure. Four major risk factors for recurrence have been identified: (a) a young age (<12–18 months)^17,69; (b) a family history of febrile seizures^16,52,66; (c) a low temperature at the time of seizure¹²; and (d) a short duration of fever before the first seizure.¹² The recurrence risk for those with none of the four factors was 14%, with one factor 23%, with two 32%, with three 62%, and with all four 76%.¹³ In a meta-analysis, young age of onset and a family history of febrile convulsions were the strongest predictors of recurrence¹¹ (see Table 1). A family history of afebrile seizures has inconsistently been associated with risk for recurrence.^12,47,49,69 “Complex” febrile convulsions and neurologic dysfunction do not influence the chance of febrile recurrences.^12,47,69

There is an important genetic predisposition for febrile seizures, and genetic factors exist that may be either causative or protective. It is natural to consider the possibility of a genetic propensity for the occurrence of a febrile seizure; however, there may also be protective gene(s) in those who never experience a febrile seizure. These protective gene(s) might raise the seizure threshold.

Although the exact mode of inheritance is not known, febrile seizures tend to run in families. An autosomal recessive mode of inheritance is not likely, because there is an excess of affected parents and the risk to siblings is less than 25%. Either a polygenic mode of inheritance or dominance with incomplete penetrance is more likely.^3,63 Afebrile seizures or epilepsy may be more common in families of children with febrile seizures, but the evidence is controversial. When a child has a first febrile seizure, the risk for younger siblings is in the range of 10% to 20% and even higher if a parent reports having had a febrile convulsion.⁶

Two unique epilepsy syndromes always include seizures with fever: generalized epilepsy with febrile seizures plus (GEFS+)⁵⁴ and Dravet syndrome (severe myoclonic epilepsy of infancy).²² GEFS+ is an autosomal dominant epilepsy syndrome with 80% penetrance and is associated with heterogeneous clinical phenotypes (see Chapter 257).²⁰ In a given family, about one third of those affected have one or more febrile seizures that often persist beyond the usual age of 5 years. About one third have febrile seizures and then afebrile generalized tonic–clonic seizures in adolescence that usually remit. A final third have other epilepsy syndromes, usually, but not always, generalized syndromes (myoclonic-astatic epilepsy, typical childhood absence, and even Dravet syndrome). Several gene mutations have been identified involving voltage-gated sodium channels of SCN1A and the γ2 subunit of ligated-γ-aminobutyric acid (GABA)-2 receptor.²⁰

Dravet syndrome (Chapter 230) begins in the first year of life with prolonged, often unilateral febrile seizures precipitated by relatively low-grade fever. Development stagnates or regresses around 1 year of age, when other seizure types develop (myoclonic, atypical absence, or partial). Ataxia and important behavior problems (hyperactivity) are very common. Dravet syndrome may occur within families with GEFS+ and with the identical gene defect, but in approximately 50% of cases, no de novo missense or truncating mutations of the SCN1A subunit are present.²⁰

The complete genetic story of febrile seizures is still incomplete. More than seven chromosome linkage sites have been associated with febrile seizures, suggesting locus heterogeneity. In addition, at least five genes have been identified as causal for epilepsy syndromes that include febrile seizures, including GEFS+.⁷³ These findings may provide insights into the biology of age-limited temperature-dependent seizure susceptibility.

“Everyday” infections, such as tonsillitis, upper respiratory infections, otitis media, roseola infantum, or Shigella gastroenteritis are often implicated as the cause of fever in children with febrile seizures. Unfortunately, there are no recent studies of the epidemiology of these infections, since vaccines against Haemophilus influenzae, varicella, pneumococcus, meningococcus and influenza are in widespread use and have likely
changed the panorama of childhood febrile illness. Seizures occurring soon after immunization with whole-cell diphtheria-pertussis-tetanus (DPT) and measles vaccines should not be regarded as a direct adverse effect of the vaccine.³⁰ Such seizures are believed to be “ordinary febrile seizures” triggered by fever induced by the vaccine. Their subsequent clinical course is identical to other febrile seizures,²⁹ with no increased risk for subsequent afebrile seizures or abnormal neurologic development.⁷ The frequency of febrile seizures after DPT or measles vaccination is six to nine, and 24 to 25 per 100,000 children vaccinated, respectively. Newer acellular pertussis vaccines rarely induce a febrile reaction and, since their introduction in Canada, there has been a 79% decrease in hospital admissions for DPT vaccine related febrile seizures.³⁸