The incidence for epileptic spasms is approximately 2 to 5 per 10,000 live births worldwide (5, 6, 7, 8, 9), with an estimated lifetime prevalence by age 10 years of 1.5 to 2 per 10,000 children (6,10). The lower prevalence rates are a result of mortality, evolution of epileptic spasms into other seizure types, and incomplete determination in population-based studies of older children (11). A genetic predisposition may exist, as infantile spasms have been reported in both monozygotic and dizygotic twins (12,13). Sex differences are inconsistent (14), although some studies suggest a moderate male predominance (5,8).

West syndrome is a triad of epileptic spasms, hypsarrhythmia, and developmental failure or regression. Over the years, salaam seizures, jackknife seizures, axial spasms, periodic spasms, and serial spasms have been used to describe events that are not epileptic spasms (15).

Age of onset is typically between 4 and 8 months (16), but epileptic spasms can occur as early as 2 weeks or as late as 18 months of age (11,17) and, rarely, can begin in adulthood. In some studies (17), late-onset spasms were cryptogenic and occurred despite previously normal development; other studies (15) found that half were associated with cortical dysplasia, hypoxic-ischemic encephalopathy, or genetic anomalies and were refractory to medications (16). Late-onset spasms have been intermixed with atonic, tonic, partial, myoclonic, or generalized tonic-clonic seizures or atypical absences. The characteristic spasms generally resolve spontaneously or evolve into Lennox-Gastaut syndrome or intractable partial seizures. Spasms may persist in 15% to 23% of all patients beyond 3 to 7 years of age (18,19).

Epileptic spasms consist of clusters of seizures involving a flexion jerk of the neck, trunk, and extremities. The phasic contraction lasts for less than 2 seconds, the ensuing tonic contraction for 2 to 10 seconds, although only the phasic contraction may be present (16). Sometimes called tonic spasms, prolonged muscle or tonic contractions are seen in intractable cases (20). The three types of spasms—flexion, extension, and mixed—are classified by the type of contraction. In flexion spasms, the trunk, arms, legs, and head flex. In extension spasms, the back arches and arms and legs extend. Mixed spasms combine extension of the legs and flexion of the neck, trunk, and arms. The mixed type accounts for 42% of all epileptic spasms; extensor spasms, the least common, comprise 23% of all epileptic spasms (16). Many children have more than one type, even in the same cluster (21), often influenced by position. If the trunk remains vertical, the resemblance is to a flexion spasm; if the patient is horizontal, what looks like an extension spasm is seen (22). The contractions themselves also vary. Spasms can range from only a subtle head drop or shoulder shrug to more violent action (21). Subtle spasms usually occur at the onset or offset of the episode (23). Although videotelemetry alone reveals no electrographic difference in the spasm, electromyography (EMG) with videotelemetry shows that the first activated muscle can vary in the same patient between different clusters or even from spasm to spasm within the same cluster. Even if the same muscle were initially activated with every spasm, the ensuing sequence or pattern of muscle involvement may differ within the same cluster (24).

Eye deviation or nystagmus may occur in two-thirds of all infantile spasms (16). Eye movements may be independent of the spasm or may precede its development by weeks. Usually an aspect of typical spasms, eye movements may coexist with the spasm and represent variability or changes in consciousness (23). Decreased responsiveness may follow motor spasms or occur independently as a second seizure type (11). Between spasms, most children cry, although this is apparently not an ictal phenomenon (16) and may be a result of surprise or pain. Up to 60% of all patients have respiratory pauses. Pulse changes occur less often. Some spasms are induced by sound or touch, rarely by photic stimulation (21).

Rarely, one arm or leg is more extended or the head deviates to one side. Spasms are usually asymmetric on the side contralateral to a unilateral lesion such as hemimegalencephaly. Symmetric spasms and a symptomatic etiology usually indicate diffuse lesions, not lateralized as in Down syndrome or neurofibromatosis (25); however, some children with focal or unilateral lesions may have only symmetric spasms (21,25). Recently, videotelemetry has allowed more frequent detection of asymmetric spasms. These patients either have consistently asymmetric spasms or alternate between asymmetric or symmetric spasms.

Spasms may be intermixed with other seizure types in one-third to one-half of patients (18,19,26, 27, 28). The muscle contraction in spasms is faster than that in tonic seizures but slower than that in myoclonic seizures (29,30). Tonic seizures can occur simultaneously with or precede spasms and may be difficult to differentiate, requiring videotelemetry to define the seizure type. Tonic seizures last longer than spasms and lack the initial phasic component. Both may be generated by a similar mechanism or have a similar origin such as the brainstem (16).

Partial seizures may occur before, during, or after a spasm and frequently precede a cluster of spasms (22). Partial seizures suggest a symptomatic cortical lesion. Prenatal etiology is implied if the partial seizure precedes the spasms (31). Partial seizures occurring simultaneously with spasms may be a result of chance. Alternatively, partial seizures may induce the appearance of infantile spasms or may stem from a critical factor, such as arousal mechanism, that simultaneously affects a mechanism generating both partial seizures and spasms (16). If the partial seizure precedes the spasms themselves, the patient usually will have asymmetric spasms with the predominant side changing to conform to that of the preceding partial seizure. In a patient with partial seizures and symmetric spasms, the spasm following the seizure would be initially asymmetric but gradually become symmetric (22,25).

Spasms occur on awakening or after feeding, and less often during sleep (16). Most occur in clusters, although single spasms, lasting for less than 1 to 5 seconds, have been documented (16). Clusters consist of 3 to 20 spasms that occur several times a day (11). The spasms decrease in intensity at the end of longer clusters (16). In a 2-week comparison, the number and type of spasms varied markedly; day-to-day variation was less common (32).

Approximately 50% of patients have abnormal neurologic findings on presentation, including blindness, hemiparesis, or microcephaly (33), that help to identify the 85% to 90% of that group who will eventually have developmental delay (14,33,34). Other studies quote mental retardation in 75% and cerebral palsy in 50% of patients (10,18,19,27,28,35,36).

In many children, infantile spasms eventually evolve into Lennox-Gastaut syndrome. Tonic seizures usually coexist with and are more marked in that syndrome. Seizure clustering is seen in West syndrome, but infrequently in Lennox-Gastaut syndrome (25). Clusters of epileptic spasms will become the single spasms of Lennox-Gastaut syndrome concurrently with the change in interictal pattern from hypsarrhythmia to diffuse slow spike and wave at 1 to 2.5 Hz (37). Conversely, if the interictal background continues to show multifocal independent spike discharges, the seizure semiology may not change (37).

Conclusive evidence as to the pathophysiology of epileptic spasms is lacking, perhaps because there are no good
experimental models (38). Some believe that infantile spasms represent a nonspecific age-dependent reaction of the immature brain to injury involving subcortical structures (11) that acts diffusely on the cortex, leading to the hypsarrhythmic electroencephalogram pattern and the generalized spasms. Individual case reports have described abnormalities in the pons and involvement of the serotonergic, noradrenergic, or cholinergic neurons in the brainstem nuclei (39, 40, 41). Brainstem origin has also been postulated on the basis of abnormalities in brainstem auditory evoked responses in patients with spasms (42) and disruptions of rapid eye movement (REM) sleep (43,44). Because hypsarrhythmia occurs mainly during sleep and the brainstem controls sleep cycles, the sleep association suggests involvement of the brainstem in infantile spasms (44,45).

The frequent intermixture of partial seizures with generalized or asymmetric spasms suggests a cortical-subcortical interaction, a hypothesis supported by the effectiveness of cortical resection in controlling generalized infantile spasms (22). In other words, the cortical lesion interacts with developing brainstem pathways, causing motor spasms that are similar to startle or cortical reflex myoclonus (37,46).

Further supporting the brainstem hypothesis are results of positron emission tomography (PET) scans in patients with epileptic spasms showing hypermetabolism of the lenticular nuclei (47). That serotonergic ([¹¹C]-methyl-L-tryptophan) and γ-aminobutyric acid (GABA)-ergic ([¹¹C]-flumazenil) tracers may be more effective than fluorodeoxyglucose PET in defining a focus in patients with epileptic spasms also suggests brainstem involvement, because the raphe-cortical or striatal projections use serotonin as a neurotransmitter and these pathways cause the diffuse hypsarrhythmic patterns on EEG (48). On ictal single-photon-emission computed tomography studies, both subcortical and cortical structures were activated (49).

The response to corticotropin suggests involvement of the hypothalamus and pituitary-adrenal axis. Baram has proposed that a nonspecific stressor releases the proconvulsant corticotropin-releasing hormone (CRH) (50), which may be the final common pathway for the multitude of etiologies of infantile spasms. CRH causes severe seizures and death in neurons associated with learning and memory, and its effects are especially important in infants because CRH receptors are most abundant during the early developmental period (51). To support the hypothesis that corticotropin inhibits the release and production of CRH through a negative feedback mechanism, Nagamitsu and colleagues (52) measured cerebrospinal fluid (CSF) levels of β-endorphin (also derived from a common precursor of corticotropin), corticotropin, and CRH (which releases both corticotropin and β-endorphin) in 20 patients with spasms. The CSF levels of β-endorphin and corticotropin were lower than in controls, as was the CRH level, although not significantly. Riikonen observed that CSF corticotropin levels were higher in infants with cryptogenic than symptomatic spasms (53).

Because infantile spasms typically begin at the time when the first immunizations are administered, the question has been whether the association is causative or coincidental. Numerous anecdotal reports have noted the appearance of infantile spasms within a few hours to a few days after a diphtheria, pertussis, tetanus vaccination, although all controlled studies to date have failed to demonstrate any association (54, 55, 56, 57). Some proposed immunologic mechanisms have been based on antibodies to brain tissue in blood samples from patients with infantile spasms (58,59), or increased numbers of activated B and T cells in the blood (60), or increased levels of HLA-DRw52 antigen (34). Finally, calcium-mediated models have been postulated, but no studies have yet been published (61).

Many disorders can give rise to spasms, and correct identification of the cause often has therapeutic and prognostic implications. When the underlying cause cannot be identified, spasms are classified as cryptogenic and in the past accounted for up to 50% of cases. Since the advent of magnetic resonance imaging (MRI) and newer sequences, however, only 10% to 15% are cryptogenic (34,62, 63, 64). In fact, some imaging studies that were normal early in life may later demonstrate lesions on MRI as a result of progressive myelination (65). PET may also increase the chance of detecting a lesion in some cryptogenic patients. Spasms are symptomatic when a disorder can be identified by history, physical and neurologic examination, neuroimaging, and metabolic or genetic testing. Symptomatic patients account for 70% to 80% of all cases (8,28, 36,38,48) and generally have a poorer prognosis than cryptogenic children. Cryptogenic patients often are products of a normal pregnancy and birth, with normal development prior to the onset of spasms and normal findings on physical examination. The spasms begin abruptly without a background of previous partial seizures. Results of neuroimaging and laboratory evaluations are normal (66). Cryptogenic patients have higher levels of CSF corticotropin, serum progesterone, CSF GABA, and CSF nerve growth factor (53), but these may reflect brain damage from the spasms. Symptomatic patients usually have more focality on neurologic examination, a history of partial seizures evolving into spasms, or lateralization on EEG (67).

Watanabe (66) proposed a subset of the cryptogenic group with “idiopathic” West syndrome. These patients have normal development, and their spasms stop completely after a short time. Developmental regression and focal interictal EEG abnormalities are absent, and hypsarrhythmia disappears between each spasm, which is symmetric. A family history of seizures is common. This group may represent from 44% to 87% of all cryptogenic cases (30,68).

Although most children with spasms have no family history, a possible X-linked transmission has been mapped to regions Xp11.4-Xpter and Xp21.3-Xp22.1 that also is associated with mental retardation (69). One of these loci is implicated in neuroaxonal processing (radixin, RDXP2) (70).

The diseases associated with symptomatic epileptic spasms are classified as prenatal, perinatal, and postnatal. Prenatal causes include congenital malformation, TORCH (toxoplasmosis, other infections, rubella, cytomegalovirus, and herpes simplex) infections, neurocutaneous disorders, chromosomal abnormalities, metabolic disorders, and congenital syndrome. Prenatal etiologies account for 30% to 45%, perhaps as many as 50%, of all cases (26,35,71,72). Tuberous sclerosis causes from 10% to 30% of prenatal spasms (27,72,73). A comparatively large tuber burden is more likely to produce spasms rather than partial seizures (74). Although partial seizures are most common with focal cortical dysplasia, epileptic spasms can occur (75); PET scans may help to identify these patients (76). Occipital lesions are associated with earlier onset of epileptic spasms than are frontal lesions (77). Neurofibromatosis type I can also cause spasms, but these usually have a better prognosis than other symptomatic causes (78). Chromosomal abnormalities, most commonly Down syndrome (72,79), represent approximately 13% of prenatal etiology; these children usually do not have a poor prognosis compared to other symptomatic cases (68).