There is no consensus regarding the mode of inheritance of febrile seizures or their clinical expression. Autosomal dominant (3), autosomal recessive (4), and polygenic theories (5,6) have all been formulated.

Febrile seizures are approximately two to three times more common among family members of affected children than in the general population (3,7). Affected parents increase the risk for the occurrence of febrile seizures in siblings. The risk increases when both parents are affected and is increased further in proportion to the number of febrile seizures experienced by the proband (8). A higher incidence of afebrile epilepsy has been found in first-degree relatives of patients with febrile seizures (8,9). Conversely, the occurrence of febrile seizures in first-degree relatives is itself a risk factor for febrile seizure recurrence (10). Siblings have the greatest risk, followed by offspring, nieces, and nephews (8). Coexistence of febrile seizures and epilepsy increases the risk for both disorders in siblings (8).

The incidence of febrile seizures also varies according to geographic region and race. Parents and siblings of Asian children are at considerably higher risk for febrile seizures than are Western families. Sibling risk approaches 30% if one parent has had a febrile seizure. The difference in frequency of febrile seizures in Asian compared with European or North American families suggests a strong, genetically determined population effect (11).

Linkage studies in a number of large pedigrees have identified several mutations in sodium channel subunit genes (12). Putative febrile seizure loci include FEB 1 (chromosome 8q13-q21), FEB 2 (chromosome 19p), FEB 3 (chromosome 2q23-24), and FEB 4 (chromosome 5q14-q15) (13,14). All affected individuals present with recurrent febrile seizures by 3 years of age, with no evidence of structural brain pathology or intracranial infection. Although most individuals are predisposed to later afebrile seizures, families mapping to the FEB 3 locus have significantly higher rates of later epilepsy compared with that reported in general population studies or in families with febrile seizures (15).

Despite the identification of multiple febrile seizure loci and mutated genes, little evidence points to their direct contribution toward the majority of febrile seizures reported
in most affected individuals. This probably reflects the marked heterogeneous clinical manifestations of febrile seizures and their lack of association with known genetic loci (16). Furthermore, family pedigrees of most known febrile seizure phenotypes are atypical of “common” febrile seizures, in that mutation-specific febrile seizures often have an extended age of onset and offset, and predispose individuals to later afebrile seizures. Pal and associates (17) used a case-control study design to identify specific phenotypic subgroups of febrile seizures and reduce clinical heterogeneity. In a comparison of 83 patients with febrile seizures who had a first-degree family history and 101 control patients with febrile seizures who lacked affected family members, the investigators found that a first-degree family history of febrile seizures and the later occurrence of afebrile seizures were specifically and independently associated with an increased risk for febrile seizure recurrence.

The onset of febrile seizures generally follows a bellshaped pattern. Ninety percent of these seizures occur within the first 3 years of life (18), 4% before 6 months, and 6% after 3 years of age. Approximately 50% appear during the second year of life, with a peak incidence between 18 and 24 months (18).

Febrile seizures occurring before 6 months of age should always raise the level of suspicion of infectious causes; bacterial meningitis must be excluded by examination of the cerebrospinal fluid (CSF) in patients of this age group. Febrile seizures after 5 years of age also should be managed cautiously, because benign causes are less common in older children.

The limited age range in febrile seizures has never been satisfactorily explained. Immaturity of central neurotransmission may play a role but should affect other childhood seizure types equally. Prostaglandin E₂, but not homovanillic or 5-hydroxyindoleacetic acid, is increased in lumbar CSF following febrile seizures in humans (19,20). Hyperthermia-induced convulsions in the developing rat can alter nicotinic and muscarinic cholinergic function (21). The maximum changes occur 55 days after the last convulsion, suggesting the importance of secondary factors.

Febrile seizures typically occur relatively early in an infectious illness, usually during the rising phase of the temperature curve. Rectal temperatures at this time may exceed 39.2°C (102.6°F), and approximately one-fourth of seizures occur at temperatures above 40.2°C (104.4°F). Despite the implicit relationship between fever and seizure activation, temperature itself probably does not lower the seizure threshold. The incidence of febrile seizures does not increase in proportion to temperature elevation, and febrile seizures are generally uncommon in the later stages of a persistent illness. Moreover, children between the ages of 6 and 18 months who experience a fever higher than 40°C (104°F) have a sevenfold reduction in seizure recurrence compared with children with a fever below 40°C (104°F) (22). A brief duration of fever before the initial febrile seizure has been linked to an increased risk for seizure recurrence (23).

Febrile seizures typically are associated with common childhood illnesses, most frequently viral upper respiratory tract, middle ear, and gastrointestinal infections. Bacterial infections, including bacteremia, pneumonia, sepsis, and meningitis, are rare concomitants of febrile seizures. None of the common viral or bacterial childhood infectious illnesses appears to be uniquely capable of activating febrile seizures.

Febrile seizures in conjunction with shigellosis constitute the most frequent extraintestinal manifestation of this infection (24). A direct neurotoxic effect of the Shigella bacterium on seizure threshold has been proposed.

Immunization-related seizures also manifest with fever, usually within 48 hours of inoculation (25). Approximately one-fourth of immunization-related seizures are related to administration of diphtheria-pertussis-tetanus (DPT) vaccine, and one-fourth follow measles immunization. Data from the National Collaborative Perinatal Project indicate that age of onset, personal and family history, and clinical presentation of postimmunization seizures resemble those of febrile seizures from infectious causes (26). The risk of DPT-induced febrile seizures increases if a family member has had an afebrile seizure (27,28). These shared features suggest that infectious and immunization-related febrile seizures are expressions of a unitary condition.

Ancillary factors related to underlying illness or fever may be implicated in the pathogenesis of febrile convulsions, usually with little supportive evidence. Direct viral invasion of brain tissue has been proposed (29), but children with proven viral infections appear no more likely to experience seizure recurrence than do uninfected children (30). Electrolyte disturbances are said to lower seizure threshold, but this mechanism remains relatively unsupported (31). Transient pyridoxine deficiency seems unlikely, and the association of Shigella infection and febrile seizures has prompted a search for an epileptogenic neurotoxin.

Proinflammatory cytokines have recently been implicated in the pathogenesis of febrile seizures. Interleukin (IL)-1β, tumor necrosis factor-α, and nitrite levels are all increased in the CSF of children with a febrile seizure (32). Increased secretion of IL-6 and IL-10 by liposaccharidestimulated mononuclear cells is higher in patients with a history of previous febrile seizures (33).

Girls younger than 18 months of age have a slightly higher risk than boys of experiencing more frequent and
severe febrile seizures (34,35). Ounsted (36) proposed that an excess of boys from one-sex sibships may explain the male predominance.

Simple febrile convulsions are solitary events, lasting less than 15 minutes and lacking focality. They occur in neurologically normal children and are not associated with persistent deficits. The source of the fever is always outside the central nervous system (CNS).

Between 80% and 90% of all febrile seizures are simple episodes (18,37,38). This figure is probably an underestimate, because most published series are hospital based and thus weighted toward children with complex risk factors (39).

Despite their common occurrence, the sporadic nature and brief duration of febrile seizures make analysis difficult. Descriptions provided by parents and emergency department personnel are retrospective and probably not entirely accurate. Video-electroencephalographic studies of afebrile generalized seizures, for example, often reveal subtle atonic or myoclonic components that were omitted in the witnessed accounts. Lack of objectivity notwithstanding, febrile seizures are described as tonic, clonic, or tonic-clonic events that usually begin without warning and display upward eye deviation as consciousness is lost. Atonic forms are rare, and postictal depression is generally brief.

Electroencephalography has not been particularly useful in the evaluation of simple febrile seizures. Although paroxysmal and nonspecific electroencephalographic (EEG) abnormalities are often evident within 24 hours of seizure onset, they have little prognostic significance. Slow-wave activity occurs in up to one-third of patients (40), and is often bilateral and prominent in the posterior regions (41). Twenty percent of patients, usually older than 2.5 years of age, have generalized spike-and-wave discharges on the electroencephalogram.

In a longitudinal study of 89 patients with febrile seizures followed until puberty, Doose and associates (42) identified three patterns of EEG abnormality: rhythms of 4 to 7 Hz, generalized spike-and-wave discharges, and photosensitivity. None were specific for febrile seizures because all had been described in generalized epilepsies as well. Genetic factors probably account for the age-related expression of these EEG patterns in benign, simple febrile seizures.

Because simple febrile seizures may result from CNS infection, trauma, or electrolyte disturbance, laboratory investigation is usually warranted even when findings on the physical examination are normal. The diagnostic yield of such studies is usually well below 2%, however, and difficult to justify (43). The skull roentgenogram and lumbar puncture are even less likely to contribute useful information in healthy children (44), and the rare febrile seizure caused by electrolyte disturbance usually can be diagnosed from the patient’s history. The confirmation of viral meningitis by lumbar puncture does not alter long-term management.

The evaluation of simple febrile seizures should therefore rely primarily on careful history taking, and judicious laboratory and radiologic testing. This approach, which is particularly important in children who are normal, has been underscored in an editorial (45) stating that “children who have their first febrile convulsion need no more tests than the clinical findings dictate.” An exception is the requirement for CSF examination in all patients younger than 6 months of age who lack any of the classic signs of bacterial meningitis. The rule that all children younger than 18 months of age with a first febrile seizure should always undergo CSF examination is probably excessive, and each child should be evaluated individually. When meningitis is suspected clinically, lumbar puncture should be performed promptly in the physician’s office or emergency department.

Hospitalization is rarely necessary following a simple febrile seizure. Testing can usually be performed in an outpatient setting because risk of seizure recurrence is low. Even so, pediatricians may hospitalize patients who can be sent home safely. In 1975, 24% of practicing pediatricians routinely admitted children after a first febrile seizure; a decade later, 20% still followed this practice (46).

The concept of a “complex” febrile seizure originated with epidemiologic studies indicating that several patient- and seizure-related variables predicted higher rates of subsequent epilepsy: seizure duration longer than 15 minutes, focal seizure manifestations, seizure recurrence within 24 hours, abnormal neurologic status, and afebrile seizures in a parent or sibling (47). Six percent of patients with two or more risk factors developed afebrile epilepsy by the age of 7 years, compared with only 0.9% if risk factors were absent (47).

Studies conducted at the Mayo Clinic also reveal a less favorable prognosis for patients with complex febrile seizures (38). Seventeen percent of neurologically impaired children with complex febrile seizure manifestations developed epilepsy by the third decade, compared with 2.5% of children who lacked risk factors. The occurrence of focal, recurrent, and prolonged seizures raised the risk for afebrile episodes to nearly 50%.

Children with complex febrile convulsions may subsequently exhibit a variety of afebrile seizure patterns. The National Collaborative Perinatal Project (37) found generalized tonic-clonic seizures to be the most frequent and absence or myoclonic seizures less common. In the Mayo Clinic experience (48), 29 cases of afebrile epilepsy developed in a cohort of 666 patients with febrile seizures. Seizures were classified as focal in 16 patients and of
temporal origin in 10 patients. Generalized tonic-clonic seizures were reported in 12 patients, 3 of whom also had absence seizures. One patient had unclassifiable seizures. In a retrospective analysis of 504 children with epilepsy, Camfield and colleagues (49) found a 14.9% incidence of prior febrile seizures. Febrile seizures most often preceded generalized tonic-clonic afebrile seizures and were regarded as fundamentally indicative of reduced seizure threshold.

Complex febrile seizures must be managed more aggressively than simple episodes. Meningitis must be excluded by timely performance of CSF examination, and neuroimaging studies are indicated to detect structural lesions. In acute bacterial meningitis, focal febrile seizures may accompany cortical vein or sagittal sinus thrombosis. In North America, parasitic disease and brain abscess are uncommon causes of complex febrile seizures.

Although children with complex febrile seizures may be expected to show a higher rate of abnormal EEG recordings than normal, confirmatory data are sparse. Studies of febrile seizures rarely include EEG findings, although this type of information would enhance the value of electroencephalography in the management of patients with febrile seizures.

Although most febrile seizures are self-limited, prolonged episodes and febrile status epilepticus are not rare. The reported occurrence of epilepsy, brain damage, or death following febrile status epilepticus further underscores its serious nature. Of 1706 children with febrile seizures followed in the National Collaborative Perinatal Project, 8% experienced seizures for longer than 15 minutes, and 4% had seizures for longer than 30 minutes (37). Febrile status epilepticus accounted for approximately 25% of all cases of status epilepticus in children (27,50), and is often the initial presentation of chronic epilepsy (37).

Children with febrile status epilepticus are usually mentally and physically normal. As with simple febrile seizures, common childhood infectious diseases and immunizations are the primary cause of the fever. An association between female sex and febrile status epilepticus has been observed (51), with younger age strongly predisposing patients toward prolonged unilateral febrile seizures (52).

Postmortem studies of patients dying of febrile status epilepticus reveal widespread neuronal necrosis of the cortex, basal ganglia, thalamus, cerebellum, and temporolimbic structures (53). Rare inflammatory changes suggest that seizures and anoxia, rather than infection, are the primary causes of mortality (41,53,54).

Prospective studies reveal that the risk for death or permanent neurologic impairment following febrile status epilepticus is negligible (37,55). The tendency for febrile status epilepticus to recur is especially low in neurologically normal children (56), and mortality in this group has markedly declined. None of the 1706 patients followed in the National Collaborative Perinatal Project died as a consequence of febrile seizures, a finding confirmed by others (56).