INTRODUCTION AND TERMINOLOGY — Infantile spasms represent an age-specific epileptic disorder of infancy and early childhood. Children with infantile spasms typically exhibit epileptic spasms along with the electroencephalographic (EEG) pattern known as hypsarhythmia.
The International League Against Epilepsy has proposed the term "infantile epileptic spasms syndrome (IESS)" to encompass infants presenting with epileptic spasms, with or without fulfilling all of the criteria for West syndrome, which is a triad of epileptic spasms, arrest or regression of psychomotor development, and hypsarhythmia [1,2]. Infants with IESS may often lack one of these three criteria, but management recommendations remain the same.
The management and prognosis of infantile spasms are reviewed here. The etiology and pathogenesis, clinical features, and diagnosis of this disorder are discussed separately. (See "Infantile epileptic spasms syndrome: Etiology and pathogenesis" and "Infantile epileptic spasms syndrome: Clinical features and diagnosis".)
CHOOSING INITIAL THERAPY — The major options for treatment of infantile epileptic spasms syndrome (IESS), also known as "infantile spasms," are hormonal therapy (ie, adrenocorticotropin hormone [ACTH] in the form of corticotropin injection gel, cosyntropin, tetracosactide, and oral glucocorticoids) and/or vigabatrin. Initial combination treatment with both hormonal therapy and vigabatrin may be more effective than hormonal therapy alone and is the standard of first-line management in many centers or protocols . (See 'Combination therapy' below.)
Urgency of treatment — Treatment for infantile spasms should be started urgently (within three days or less) once the diagnosis is made. The importance of urgent treatment initiation has been well recognized, given particular evidence for improved developmental outcomes associated with early recognition and implementation of therapy .
Hormonal therapy for patients without tuberous sclerosis — The mainstay of medical treatment for most patients with infantile spasms (excepting those with tuberous sclerosis complex) is hormonal therapy with corticotropin injection gel (ACTH) or oral glucocorticoids [5-12]. (See 'Corticotropin (ACTH)' below.)
Glucocorticoid hormonal therapy regimens (eg, prednisone, prednisolone, methylprednisolone, or dexamethasone) are effective alternatives to corticotropin (ACTH) and are less costly and easier to administer than corticotropin. Historically, corticotropin gel was the preferred hormonal therapy, but high-quality data comparing oral glucocorticoids to corticotropin are lacking, and there is no clear consensus for choosing one or the other. (See 'Glucocorticoids' below.)
With the exception of vigabatrin (see 'Vigabatrin' below), antiseizure medications are not considered effective as a first-line treatment for infantile spasms. (See 'Other antiseizure medications' below.)
Vigabatrin for patients with tuberous sclerosis — For infants with tuberous sclerosis complex, vigabatrin is first-line therapy for infantile spasms. (See 'Vigabatrin' below and "Tuberous sclerosis complex: Management and prognosis", section on 'Seizure control'.)
Special patient populations — In addition to infants with tuberous sclerosis complex, we sometimes use vigabatrin as initial therapy for infantile spasms (rather than hormonal therapy) in selected infants with special considerations including the following:
●At high risk of infections
●Presence of cortical dysplasia
●Chronic infantile spasms
The ketogenic diet is used to treat infantile spasms in children with pyruvate dehydrogenase deficiency. (See "Ketogenic dietary therapies for the treatment of epilepsy", section on 'Pyruvate dehydrogenase deficiency'.)
Hormonal therapy — The mainstay of medical treatment for most infants with infantile spasms is hormonal therapy with corticotropin (ACTH) or oral glucocorticoids [5-7].
Formulation — The ACTH formulation most commonly used in the United States is a natural preparation (corticotropin injection gel), given intramuscularly (IM) or subcutaneously, and it is very expensive . This preparation is a mixture of peptides and derivatives and is measured in units.
A synthetic long-acting corticotropin preparation, cosyntropin (tetracosactide), is available in Canada, Europe, and elsewhere and is measured in milligrams. According to manufacturer data, patients receiving 40 units of natural corticotropin injection gel may be converted to 0.5 mg IM every other day of cosyntropin depot . Data comparing the natural and synthetic forms are not available, but uncontrolled observations suggest that efficacy is similar [15,16].
Dose — The optimal dose of corticotropin injection gel is often debated.
●High-dose protocol – Most centers (and the authors of this topic) utilize a high-dose protocol (table 1), as this is the best-studied dose and typically provides the best short-term response in the authors' clinical experience. The high-dose regimen begins with corticotropin injection gel 75 units/m2 IM twice daily (or 150 units/m2 IM once daily) for two weeks, followed by a taper over two to four weeks [17-20]. The dosing recommendations from the Pediatric Epilepsy Research Consortium (PERC) can be found in the table (table 1) .
●Low-dose therapy – Two small randomized trials reported that response to low-dose therapy (20 units per day) was comparable to higher doses and resulted in fewer side effects [21,22]. In the larger of the trials, 50 patients with recently diagnosed infantile spasms were randomly assigned to receive high-dose ACTH (150 units/m2 per day for three weeks, 80 units/m2 per day for two weeks, 80 units/m2 every other day for three weeks, 50 units/m2 every other day for one week, and taper to zero over three weeks) or low-dose ACTH (20 to 30 units per day for two to six weeks, then taper to zero over one week) . The groups were similar in response to treatment (cessation of spasms and disappearance of hypsarhythmia; 50 versus 58 percent) and rate of relapse. Hypertension occurred more often in the high-dose group (31 versus 4 percent); other side effects were similar between groups. Other retrospective studies also confirm the similar efficacy and fewer adverse side effects using lower-dose synthetic ACTH therapy [15,16,23].
●Step-up dosing – A strategy individualizing therapy to treatment response was evaluated in a multicenter study of 30 patients . Treatment was started with low-dose ACTH (3 units/kg per day) administered for two to three weeks. In patients who did not respond initially, the dose was increased in intervals to a maximum of 12 units/kg per day . Most patients with spasms of unknown etiology and one-half of those with spasms of known etiology responded to low-dose ACTH given over the course of two to three weeks. Side effects were related to dose. Other centers have also reported success with a "step-up" dosing protocol, with even lower initial doses .
Glucocorticoids — Given the advent of data (see 'Efficacy of hormonal therapy' below) that have suggested, but not proven, that high-dose prednisolone regimens are as effective as corticotropin (ACTH), and given considerable reduction in the cost of treatment and ease of administration with oral glucocorticoids, many centers are now routinely using high-dose oral glucocorticoids as initial therapy for infantile spasms.
A 2015 consensus document from the International League Against Epilepsy (ILAE) concluded that glucocorticoids are probably effective in the short-term control of spasms, but that the optimal preparation, dose, and duration have not been established .
Dose — We typically begin treatment with prednisolone using the regimen recommended by the PERC, which can be found in the table (table 1). Other experts use slightly different dosing from the United Kingdom Infantile Spasms Study (UKISS) .
Adverse effects of hormonal therapy — Significant adverse effects may occur with ACTH, as they did in 37 and 85 percent of treated patients in two reports [27,28]. Adverse effects include hypertension, immune suppression and infection, electrolyte imbalance, gastrointestinal disturbances, ocular opacities, hypertrophic cardiomyopathy, cerebral atrophy, growth impairment, and adrenal insufficiency [5,21,23,29-34]. Cases of anaphylaxis have been reported . A transient cushingoid appearance, hirsutism, irritability, and sleep disorders also are seen ; these changes are reversible upon discontinuation of the drug. Similar adverse effects are associated with glucocorticoid treatment. (See "Major side effects of systemic glucocorticoids".)
The risk of adverse effects with hormonal therapy is typically increased with prolonged duration of treatment. (See 'Duration of initial therapy' below.)
●Hypertension – Hypertension is the most common complication . In one series, hypertension, defined as three consecutive blood pressure measurements greater than the 95th percentile (stage 1) or greater than the 99th percentile (stage 2) for age, sex, and height (which for infants of 10 kg is typically near 100 to 110 mmHg systolic), occurred in 44 percent of patients . Intracranial hemorrhage has been associated with hypertension in this condition .
●Electrolyte imbalance – Electrolyte imbalance may depend upon the glucocorticoid dose. Hypokalemic alkalosis occurred in as much as 12 percent of patients on high-dose ACTH . By contrast, no significant electrolyte abnormalities were detected by serial laboratory monitoring during low-dose, short-term hormonal therapy .
●Irritability and sleep disturbance – Patients with infantile spasms often have increased irritability during treatment and may have disturbed sleep patterns, although these may be caused in part by the underlying disease. Compared with normal infants, total sleep time and the proportion of rapid eye movement (REM) sleep are reduced in infantile spasms . Thus, sleep disturbances may be caused by infantile spasms rather than hormonal therapy.
●Brain atrophy – Reduced brain size may result from ACTH and glucocorticoid therapy. Ventriculomegaly and increased volume of the subarachnoid space, consistent with cerebral atrophy, have been detected with neuroimaging of patients receiving hormonal treatment for infantile spasms [29,39-42]. These changes may persist after termination of therapy. In a trial comparing ACTH and prednisone, cerebral atrophy was detected by CT scan in 62 percent of patients during treatment; changes persisted in 42 percent six weeks after therapy was stopped . Subdural hematoma, resulting from brain shrinkage, may occur, even during administration of low-dose ACTH therapy . These findings are consistent with studies in animals treated with ACTH and glucocorticoids that demonstrate interference with synthesis of brain DNA (deoxyribonucleic acid), RNA (ribonucleic acid), and protein, and reduction in brain size [44,45]. The relationship of the changes in brain volume to long-term neurologic outcome of patients with infantile spasms is unknown.
●Risk for decreased bone density – Patients treated with high-dose and prolonged regimens of glucocorticoid therapy may be at risk for reduced bone density later in life ; however, there are no studies that indicate whether follow-up is needed to diagnose and treat this condition.
Monitoring hormonal therapy — Monitoring of patients on hormonal therapy with corticotropin or glucocorticoids should include measurement of blood pressure (baseline and once weekly) and serum concentrations of glucose, potassium, and sodium (baseline and every other week). More frequent monitoring and/or intervention is indicated for persistently abnormal results. Infectious contacts should be avoided, and the patient should be observed closely and treated promptly for signs of infection .
Efficacy of hormonal therapy — Several meta-analyses of randomized trials comparing the effectiveness of ACTH (corticotropin or tetracosactide) with oral glucocorticoids have found no difference between the two forms of hormonal treatment for outcomes including cessation of infantile spasms, hypsarhythmia resolution, adverse effects, relapse rate, or subsequent development of epilepsy [48-50].
Data from the National Infantile Spasms Consortium prospective multicenter cohort study also support corticotropin and oral glucocorticoids as effective first-line treatments . Among 423 infants two months to two years of age with new-onset infantile spasms, 190 were treated with corticotropin, 95 with glucocorticoids (prednisolone or prednisone), 86 with vigabatrin, and 51 with nonstandard therapies. Propensity score weighting was used to reduce the effects of selection bias. Freedom from treatment failure (defined as freedom from infantile spasms beginning within 30 days of the start of treatment, and no second treatment prescribed within 60 days) varied by treatment: ACTH, 46 percent; oral glucocorticoids, 44 percent; vigabatrin, 37 percent; nonstandard therapies, 8 percent. In the vigabatrin group, freedom from treatment failure was higher among infants with tuberous sclerosis complex (13 of 21, 62 percent) compared with other infants (19 of 65, 29 percent).
Of note, conclusions have been limited by the overall poor methodology and small size of most of the available clinical trials and studies. Lack of adherence to standardized case definitions and outcome measures is one problem with many studies. Another is that inclusion of a control group is impracticable, as the natural history of the disease is that clinical spasms subside and EEG patterns evolve without therapy, yet many clinicians would be reluctant not to treat, particularly since there are observational data suggesting that delayed therapy may worsen prognosis. As a result, questions remain regarding the optimal drug, dose, and duration of therapy.
Mechanism of hormonal therapy — The precise mechanism of action of ACTH and glucocorticoids is not known. Administration of ACTH can control spasms in patients with adrenal suppression, suggesting that the effect is independent of adrenal glucocorticoid release.
Corticotropin may have direct anticonvulsant effects, perhaps via suppression of corticotropin-releasing hormone (CRH), an endogenous neuropeptide that may provoke convulsions in immature brain. This theory was suggested by a report in which administration of high-dose ACTH to rats resulted in a reduction of the expression of the CRH gene in the amygdala [52,53]. This reduction occurred in animals with and without adrenalectomy and thus was independent of endogenous cortisol production. The effect was reproduced by a peptide fragment of ACTH without corticotrophic activity and abolished by a melanocortin receptor antagonist.
Vigabatrin — Vigabatrin is an irreversible inhibitor of gamma-aminobutyric acid (GABA)-transaminase and thus raises the concentration of GABA in the central nervous system. Concerns regarding toxicity must be balanced against its efficacy.
Dose and administration — We use the dosing regimen for vigabatrin recommended by the PERC, as found in the table (table 1) . A typical dosing escalation scale of vigabatrin in infantile spasms uses an initial dose of 50 mg/kg per day, escalating to 100 to 150 mg/kg per day as required for clinical response . Efficacy is assessed at two weeks after dose change. The 2010 consensus report suggests that patients who respond to therapy can be continued on the drug for six months, with continued evaluation for toxicity and assessment of risks and benefits . Vigabatrin should be withdrawn earlier in patients who do not respond to treatment.
However, while several variations have been studied, the optimal dose and duration of vigabatrin therapy for infantile spasms is unclear . In a review of nine prospective studies of vigabatrin in infantile spasms, doses from 18 to 200 mg/kg per day were used . The evidence was insufficient to determine if response to vigabatrin was dose dependent. In a single-masked trial that compared low-dose (18 to 36 mg/kg per day) and high-dose (100 to 148 mg/kg per day) regimens, cessation of spasms for 7 consecutive days within the first 14 days of vigabatrin therapy was significantly greater in the high-dose group (36 versus 11 percent), and the time to response was shorter . A randomized trial in 221 patients with newly diagnosed infantile spasms also suggested greater efficacy for high-dose vigabatrin . Children were randomly assigned to either high-dose (100 to 148 mg/kg per day) or low-dose (18 to 36 mg/kg per day) vigabatrin for 14 to 21 days. More patients in the high-dose group became seizure free with normalization of a video-EEG (15.9 versus 7.0 percent). The benefit appeared to be maintained in a subsequent open-label treatment phase for up to three years. Adverse events were similar in both groups.
Among the studies reviewed in the 2004 American Academy of Neurology (AAN) practice parameter, the time from initiation of therapy to cessation of spasms ranged from 12 to 35 days . In one prospective study of 19 children with infantile spasms, whose underlying cause was hypoxic-ischemic encephalopathy or not identified and who were treated for a mean of 5.1 months, withdrawal of vigabatrin over a 30- to 60-day period was not associated with relapse within a mean follow-up time of 26 months .
●Visual field constriction – The frequency of this complication in children has been difficult to estimate, in part because of the poor reliability of visual field testing in children younger than six to eight years of age and in those with cognitive impairment . However, several studies have suggested that the risk of visual field constriction is relatively low in infants [60,61]. In a retrospective review of 284 children taking vigabatrin undergoing a total of 1281 visual assessments over a 10-year period, only 2 out of 284 (0.7 percent) had evidence of ocular toxicity that was definitely attributable to vigabatrin, and four children (1.4 percent) had optic atrophy that was considered possibly attributable. This is in contrast to studies done in adults, in which the risk of mild to severe bilateral concentric visual field constriction related to vigabatrin is estimated to be 30 to 50 percent .
In one series, only 1 of 16 children with infantile spasms who were treated with vigabatrin for a mean of 21 months had visual field abnormalities at age 6 to 12 years, suggesting that young age might be protective . However, in another series that included 32 school-aged children with infantile spasms treated with vigabatrin in infancy, the prevalence of visual field defects was 34 percent, similar to that in adults . The frequency of defects increased with treatment duration and cumulative dose, ranging from 9 percent in those treated for <12 months to 63 percent in those treated for >24 months. Another observational series using electroretinograms (ERG) in infants with infantile spasms treated with vigabatrin confirmed that treatment duration is a key factor in determining likelihood of retinal damage: ERG amplitude was reduced from baseline within 6, 12, and 30 months of treatment initiation in 5, 13, and 38 percent of patients, respectively . The impact of these changes on function and quality of life has not been well documented. (See "Antiseizure medications: Mechanism of action, pharmacology, and adverse effects", section on 'Vigabatrin'.)
●Monitoring – Monitoring for vigabatrin-associated vision loss is challenging, as visual fields are difficult to assess in infancy. Ophthalmologic evaluation is recommended prior to initiation of therapy, every three to six months during therapy, and three to six months after treatment has been discontinued . In the United States, prescribers, patients, and pharmacies must participate in the Risk Evaluation and Mitigation Strategies (REMS) program.
●MRI abnormalities – Vigabatrin-associated brain abnormalities on magnetic resonance imaging (VABAM) have been reported in infants treated for infantile spasms [66-68]. These abnormalities (hyperintense lesions on T2-weighted images with restricted diffusion on diffusion-weighted imaging in the basal ganglia, thalamus, brainstem, and dentate nucleus of the cerebellum) appear to be a subclinical, dose-related phenomenon that usually resolves with discontinuation of vigabatrin. However, there are rare reports of more fulminant, symptomatic VABAM [69,70]. (See "Antiseizure medications: Mechanism of action, pharmacology, and adverse effects", section on 'Vigabatrin'.)
●Minor adverse effects – The incidence of minor adverse effects (eg, drowsiness, irritability, hypotonia, sleep disturbance, weight gain) with vigabatrin is low and usually does not necessitate discontinuation of the drug. In the three randomized controlled trials of vigabatrin, sedation was noted in 9 to 24 percent, irritability in 4 to 9 percent, insomnia in 9 percent, and hypotonia in 9 percent . Vigabatrin was discontinued in 0 to 6 percent of children in these trials because of adverse effects.
Efficacy — Vigabatrin may be effective as initial treatment for infantile spasms, although most reviews have found superior short-term outcomes with hormonal therapy [6,7,11,51,54,71]. An exception is that vigabatrin is considered to be the most effective treatment for infantile spasms in patients with tuberous sclerosis complex [11,51,72]. (See "Tuberous sclerosis complex: Management and prognosis", section on 'Seizure control'.)
In an earlier AAN practice parameter review, nine prospective studies of vigabatrin were reviewed . The time from onset of treatment to cessation of spasms ranged from 12 to 35 days. The time to EEG response ranged from 7 to 35 days, and 11 to 83 percent of children had resolution of hypsarhythmia. In the National Infantile Spasms Consortium prospective cohort study, freedom from treatment failure was higher with corticotropin and oral glucocorticoid therapy (46 percent and 44 percent respectively) compared with vigabatrin (37 percent); in the vigabatrin group, freedom from treatment failure was higher among infants with tuberous sclerosis complex (13 of 21, 62 percent) compared with other infants (19 of 65, 29 percent) .
One of the largest reports of vigabatrin as initial monotherapy comes from an uncontrolled retrospective study of 250 infants with infantile spasms in which vigabatrin was given as monotherapy . Initial suppression of spasms was noted in 68 percent. A placebo-controlled trial of vigabatrin for infantile spasms randomly assigned 40 infants to receive vigabatrin or placebo for five days, followed by open treatment with vigabatrin for at least 24 weeks . By the end of the five-day double-blind phase, spasm cessation had occurred more frequently in children treated with vigabatrin compared with placebo (35 versus 10 percent), but this result did not achieve statistical significance. At the end of open treatment, spasms had resolved in 42 percent of patients. In prospective randomized controlled trials of vigabatrin, treatment response for children with infantile spasms of known etiology ranged from 21 to 44 percent, and response for infantile spasms of unknown etiology from 27 to 57 percent [9,54,74,75]. The infantile spasms relapse rate ranged from 8 to 20 percent in those trials in which relapses were assessed [54,74,75].
Pyridoxine — Some infants with West syndrome have been reported to respond to treatment with high-dose pyridoxine (vitamin B6, 30 to 400 mg daily) or pyridoxal 5'-phosphate. Although vitamin B6 deficiency or dependency is a rare cause of the disorder, it is important not to miss it because of the remarkable response to treatment. (See "Infantile epileptic spasms syndrome: Etiology and pathogenesis", section on 'Inborn errors of metabolism'.)
Outside of pyridoxine-dependent epilepsy, high-quality data for pyridoxine in treatment of infantile spasms are lacking . The response rates to pyridoxine ranged from 13 to 29 percent in two uncontrolled, prospective, open-label studies [76,77].
MONITORING TREATMENT RESPONSE — An expert consensus panel has defined effective treatment of infantile epileptic spasms syndrome (IESS) as complete cessation of spasms and resolution of hypsarhythmia on EEG . However, hypsarhythmia is present in only 60 percent of children with epileptic spasms , and there is poor interrater reliability to diagnose hypsarhythmia on EEG . Hence, resolution of spasms is the main treatment outcome to target in this vulnerable group of patients. Per the West Delphi group consensus statement, this outcome is measured as the absence of witnessed spasms beginning within 14 days of commencement of treatment, and for ≥28 consecutive days from the last witnessed spasm .
Parents, caregivers, and trained observers may miss the occurrence of spasms, especially if they are subtle. Less commonly, they may "overcount" imitators of spasms . As a result, overnight inpatient video-EEG monitoring is ideally used to assess treatment response in children with infantile spasms. We do overnight inpatient video-EEG monitoring at two weeks from the start of therapy and as clinically indicated thereafter. However, resource limitations may preclude inpatient monitoring in many institutions. While there is no consensus on a standard duration of the post-treatment EEG needed for assessing treatment response, one study found that outpatient EEG (with median duration of 150 minutes, range 90 to 240 minutes), but not routine EEG, provided up to 85 percent sensitivity for detection of treatment failure . Therefore, outpatient EEG monitoring may be used as an alternative to overnight inpatient monitoring for follow-up at two weeks of treatment. (See "Infantile epileptic spasms syndrome: Clinical features and diagnosis".)
DURATION OF INITIAL THERAPY — In general, a successful treatment response for infantile epileptic spasms syndrome (IESS) is defined by resolution of epileptic spasms and hypsarhythmia within two weeks of treatment initiation .
●Hormonal therapy – We begin a taper of hormonal therapy after two weeks of therapy at the maximum dose, in general agreement with the 2010 consensus statement  and the Pediatric Epilepsy Research Consortium (PERC) treatment guidelines (table 1) .
The effect of adrenocorticotropic hormone (ACTH) persists when therapy is discontinued. However, the optimal duration of treatment is uncertain. Although some series suggest that hormonal therapy should be continued for many months [82,83], prolonged therapy may not be necessary and may be hazardous. As an example, in a comparison trial of low-dose ACTH and prednisone in 24 patients, response occurred within two weeks of initiation of therapy in 75 percent of patients who responded . Five patients who responded relapsed within 12 to 33 months with clinical seizures but not hypsarhythmia; four of these responded to a second course of treatment. Many side effects of hormonal therapy are associated with long-term use. (See 'Adverse effects of hormonal therapy' above.)
●Vigabatrin – We agree with the 2010 consensus report suggestion that patients who respond to vigabatrin therapy can be continued on vigabatrin for six months, with ongoing evaluation for toxicity and assessment of risks and benefits . Some protocols wean vigabatrin over one month after a three-month trial .
Vigabatrin should be withdrawn earlier in patients who do not respond to treatment.
RELAPSES — Relapses of infantile spasms in patients with infantile epileptic spasms syndrome (IESS) are not uncommon among patients who responded to an initial treatment course. No data are available to guide therapy in these cases. Often, a second course of the agent that was previously effective in obtaining control is administered .
TREATMENT FAILURE — Alternative treatments may be considered for children who fail initial therapy with hormonal therapy and vigabatrin and are therefore considered to have refractory spasms.
Sequential therapy — Lack of a successful response within two weeks should prompt a change in treatment strategy . Our general approach after failure of the first standard treatment (hormonal or vigabatrin) is to switch to the alternative standard treatment (ie, "sequential therapy"). If the alternative standard treatment is unsuccessful, other options (eg, combination therapy, ketogenic diet, surgery) are considered on a case-by-case basis, but there is no consensus approach, and good-quality data are lacking.
Combination therapy — Initial treatment with both hormonal therapy and vigabatrin may be more effective than hormonal therapy alone, but more studies on toxicity and long-term effects are needed before this strategy should be adopted routinely.
In a multicenter open-label trial, 377 children with infantile spasms were randomly assigned to receive either hormonal therapy (prednisolone minimum dose 10 mg orally four times per day or cosyntropin [tetracosactide depot] 0.5 mg intramuscular [IM] 40 international units every other day) plus vigabatrin (100 mg/kg per day) or hormonal therapy alone . By six weeks, the addition of vigabatrin resulted in a greater clinical seizure remission rate, defined as no seizures witnessed by parents or caregivers from day 14 through 42 (72 versus 57 percent; difference 15 percent, 95% CI 5.1-24.9). Adverse effects were similar in both groups, including rates of serious adverse reactions requiring hospitalization (9.1 versus 8.4 percent). Limitations of the trial included the large discrepancy between screened versus enrolled patients (766 versus 377), lack of objective measures of response and efficacy, and lack of long-term efficacy and safety endpoints. Additional studies are needed to determine whether these early results are sustained and not outweighed by increased toxicity and burdens of monitoring .
Ketogenic diet — A ketogenic diet may control spasms in cases refractory to first-line treatment, and it is the recommended treatment for infants with infantile spasms and pyruvate dehydrogenase deficiency. (See "Ketogenic dietary therapies for the treatment of epilepsy".)
A randomized controlled trial published in 2021 evaluated 91 children with epileptic spasms refractory to hormonal therapy and randomly assigned 46 to adjunctive treatment with the modified Atkins diet (a less restrictive form of the ketogenic diet) and 45 to the control group; all patients remained on their usual antiseizure medications . At four weeks, treatment with the modified Atkins diet compared with control led to a higher rate of spasm cessation (24 versus 0 percent) and an increased proportion of children with >50 percent reduction in spasms (65 versus 0 percent). Results from a small, underpowered randomized controlled trial published in 2019 suggested that the ketogenic diet has similar short-term efficacy compared with adrenocorticotropin hormone (ACTH), mainly for infants with prior vigabatrin treatment .
A 2018 systematic review identified 13 observational studies (341 patients) and no randomized trials evaluating the ketogenic diet for infantile spasms . Across studies, the median response rate (proportion achieving >50 percent reduction in spasms within six months) was 65 percent. The median spasm-free rate was 35 percent within six months of diet initiation and 10 percent at 12 to 24 months. Infantile spasms of unknown etiology were associated with increased likelihood of response to the ketogenic diet. Adverse effects and developmental outcomes were not presented.
Surgical treatment — Although many patients with infantile spasms have diffuse abnormalities, some have a focal-cortical structural or metabolic abnormality that may respond to surgical excision [89-92].
Resective surgical treatment is considered in such patients when infantile spasms are refractory to medical treatment (approximately 30 to 40 percent of patients) and when there is no evidence of diffuse brain damage or degenerative or metabolic disease . Surgery is not considered if the cortical resection would create an unacceptable new neurologic deficit. (See "Seizures and epilepsy in children: Refractory seizures", section on 'Epilepsy surgery'.)
Other antiseizure medications — Other antiseizure medications are generally not as effective as first-line treatment of infantile spasms . (See "Antiseizure medications: Mechanism of action, pharmacology, and adverse effects".)
●Valproate – Reports of treatment with valproic acid have inconsistent results [37,93-96]. Valproic acid may benefit 40 to 70 percent of patients who do not respond to ACTH [93-96]. However, the apparent response may reflect the natural history of infantile spasms rather than the effect of treatment .
High-quality data in the form of controlled trials of valproate for infantile spasms are lacking [5,37]. In one prospective open-label study, cessation of spasms and resolution of hypsarhythmia occurred in 73 and 91 percent of 22 children at six months from onset of treatment . A majority responded in two weeks, but 23 percent relapsed, and thrombocytopenia developed in one-third. In the other open-label study, cessation of spasms occurred in 72 percent of children at three months from onset of treatment .
●Zonisamide – Zonisamide is a sulfonamide derivative. Its primary mechanism of action appears to be related to blockage of voltage-dependent sodium and T-type calcium channels. In case series, zonisamide was associated with improvement in a minority of patients with infantile spasms, some of whom had failed to respond to pyridoxine or valproate [100-104].
●Topiramate – Limited data on the efficacy of topiramate for infantile spasms have been mixed. Encouraging results were reported in an open-label study in 11 patients with refractory infantile spasms [105,106]. However, in a single-center retrospective study that included 31 patients with refractory infantile spasms treated with topiramate, only three patients achieved a nonsustained clinical remission .
●Sulthiame – Sulthiame, a sulfonamide derivative, is an antiseizure medication that is available in some countries outside the United States. A randomized, placebo-controlled trial of 37 infants between 3.5 and 15 months of age with newly diagnosed West syndrome (infantile spasms plus hypsarhythmia on EEG) found that infants randomly assigned to receive sulthiame plus pyridoxine had a higher rate of complete response than those assigned to placebo plus pyridoxine (30 versus 0 percent) . Follow-up in this study was short (only nine days); the long-term efficacy of sulthiame treatment is unknown.
Overall prognosis — The overall prognosis is guarded for children with infantile epileptic spasms syndrome (IESS) (table 2) [109,110]. The mortality rate varies from 3 to 33 percent [109,111-113], with the most important risk factor being significant respiratory system comorbidity. The higher mortality rates may be attributed in some part to a period effect and are likely lower overall in the setting of modern medical care. Other adverse outcomes include epilepsy, which may be intractable, and moderate to severe neurodevelopmental disability. However, approximately 25 percent of children with infantile spasms have a favorable long-term outcome with seizure freedom and good cognitive outcomes [10,114,115].
Prognostic factors — In a 2020 review, the underlying etiology of infantile spasms was considered the most important predictor of subsequent seizure and neurodevelopmental outcomes . In studies with long-term follow-up (up to 50 years), a favorable cognitive outcome was observed in approximately 25 percent, while complete seizure freedom was achieved in approximately 33 percent. Favorable prognostic factors included early spasm recognition and treatment, short duration of hypsarhythmia, early treatment of spasm relapses or multifocal epileptic discharges on EEG, and prompt management of adverse effects.
Seizures and epilepsy
Infantile spasms and hypsarhythmia — In natural history data from historical cohorts before effective treatments were available, clinical spasms stopped by three years of age in approximately one-half of children with infantile spasms and rarely persist after five years of age [116-119]. Hypsarhythmia also tends to resolve with maturation [97,116]. (See "Infantile epileptic spasms syndrome: Clinical features and diagnosis", section on 'Clinical course'.)
Other seizure types — Fifty to 90 percent of patients with infantile spasms develop other seizure types [6,109,111,119,120]. In general, patients with infantile spasms of known etiology are more likely to develop other seizure types than those with infantile spasms of unknown etiology (formerly termed cryptogenic or idiopathic; 58 versus 35 percent) [121,122]. In one report, seizure outcomes were more favorable for patients with infantile spasms of unknown etiology as opposed to known etiology, and for those with earlier onset of spasms (<4 months versus 4 to 8 months versus 8 months) , although these relationships have not been found in other studies . In cases with identified causes, seizure outcome is better in some etiologies than in others. As examples, outcome for seizure control is relatively good in patients with infantile spasms and neurofibromatosis type 1  or Down syndrome [125-127].
Lennox-Gastaut syndrome — Some patients with infantile spasms develop a severe form of epilepsy known as Lennox-Gastaut syndrome (table 2) [6,128,129]. This syndrome includes characteristic types of seizures (typical drop attacks and axial tonic seizures) and specific EEG patterns . Children with infantile spasms of unknown etiology are less likely to develop Lennox-Gastaut syndrome compared with children who have a known etiology . (See "Lennox-Gastaut syndrome".)
Developmental outcome — Developmental impairment occurs in 75 to 85 percent of patients with infantile spasms [131,132].
●Impact of underlying disease – The long-term prognosis for developmental outcome in infantile spasms appears to depend upon the underlying disease [121,122,133-135]. In general, outcome is better for patients with infantile spasms of unknown etiology than for those with infantile spasms of known etiology [78,111,122,123,135,136]. Some infants with infantile spasms of unknown etiology do remarkably well, particularly if detected early and treated effectively. In one study of 57 infants with infantile spasms who were followed for 12 to 60 months, severe cognitive impairment (developmental score <50) occurred in 44 percent; 28 percent had mild impairment (developmental score <70) or no impairment . The mean developmental score was higher in the unknown-cause cohort compared with the group with identified cause (71.2 versus 48.4). Neurologic deficits, including motor impairment (eg, hypertonia, spasticity, and hemiparesis) and visual and/or hearing abnormalities occurred less often in the group with no known etiology (24 versus 75 percent). Another study of 95 children with infantile spasms found that the adjusted odds ratio for the later development of autism spectrum disorder was higher in the known etiology group versus the unknown etiology group (8.73 versus 1.55) . In a case series of 68 children, developmental delay was noted in 100 percent of patients with infantile spasms of known etiology .
Cognitive deficits are also common among children with infantile spasms of unknown etiology, whose spasms cease soon after onset and do not recur. In one report, neurodevelopment was assessed at four to six years of age in 15 children whose spasms resolved before one year of age . Intelligence was normal in 12 children. However, 5 of the 12 had specific cognitive deficits, including problems with attention, language, memory, and visual motor skills. In a later retrospective cohort study of 133 children with infantile spasms of unknown cause and at least six months of follow-up, developmental delay was documented in 85 percent .
●Effect of treatment – It is possible but unproven that treatment of infantile spasms improves developmental and other long-term outcomes; many observational but no randomized studies have reported that treated patients were more likely to have normal intelligence at follow-up than untreated patients [5,83,123,139,140].
Proposed mechanisms underlying improved outcome with hormonal therapy include acceleration of the physiologic rate of maturation  or deactivation of a neuroendocrine stress response. The pretreatment neurologic status may also be important; in one study, normal cognitive outcome was found in all 25 patients who had no or only mild mental deterioration at presentation but in only 3 of 12 patients who had marked or severe deterioration before treatment .
Limited data suggest developmental outcomes may be better with treatment with adrenocorticotropin hormone (ACTH) rather than vigabatrin [5,7]. In a randomized trial of 107 patients with infantile spasms that compared vigabatrin with hormonal treatment, cognitive outcomes at 12 to 14 months were better in those who had no identified underlying cause and had received hormonal treatment . When reassessed at a mean age of four years, developmental and epilepsy outcomes were similar between the two treatment groups; however, for those 77 patients with no identified etiology ("cryptogenic") of infantile spasms, cognitive outcomes remained better in those assigned to hormonal treatment .
Outcome may also be better if therapy is started soon after presentation . Among cases with no known etiology of spasms, normal outcome occurred in more patients treated within one to two months of the onset of seizures compared with those treated after one or two months (70 to 85 versus 20 to 25 percent) [82,83,133,141,143]. However, in one prospective study that used a defined treatment regimen for ACTH or prednisone and 24-hour video-EEG monitoring to determine response, the prognosis was poor for long-term outcome independent of response to therapy (cessation of spasms and resolution of hypsarhythmia), treatment agent used (ACTH or prednisone), or delay between presentation and treatment . By contrast, in the clinical trial comparing hormonal treatment with vigabatrin, an analysis of the effect of lead time to treatment on outcome found that each increase in category of lead-time duration (within 7 days, 8 to 14 days, 15 days to 1 month, 1 to 2 months, greater than 2 months) was associated with worse developmental outcome at four years .
In some reports, prolonged treatment (10 or more months) with high-dose corticotropin (80 units every two days) appeared to improve the prognosis for mental and motor development [82,83,140]. However, in a long-term follow-up study of 214 children, prognosis was not improved by larger (120 to 160 units) compared with smaller doses (20 to 40 units) of ACTH, although fewer side effects occurred in the latter .
Studies assessing the benefit of therapy on long-term outcomes are both conflicting and difficult to evaluate; most of the available studies are retrospective and lack specific treatment courses or objective criteria for response. Because of these limitations, the American Academy of Neurology (AAN) systematic review of treatment for infantile spasms concluded that successful treatment of infantile spasms possibly improves the long-term prognosis .
Outcome in adults — Little information is available on very-long-term outcomes of children with infantile spasms. Some adults with a history of infantile spasms have normal intelligence and socioeconomic status, as illustrated by one report of 214 children born between 1960 and 1976 who were followed for 20 to 35 years or until death . Death occurred in 31 percent at 3 months to 30 years of age. Intelligence was normal or mildly impaired in 24 percent of the survivors, and 33 percent had no seizures. Some patients with normal intelligence and socioeconomic status had infantile spasms with identified etiology or focal EEG findings. All patients with a good outcome had short delays between presentation and treatment and responded to ACTH therapy.
SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Seizures and epilepsy in children".)
SUMMARY AND RECOMMENDATIONS
●Choosing initial therapy
•Hormonal therapy for most infants – For most children with infantile epileptic spasms syndrome (IESS), we suggest initial treatment with hormonal therapy using corticotropin injection gel (adrenocorticotropin hormone [ACTH]) or oral glucocorticoids rather than vigabatrin (Grade 2C). This recommendation is based upon advantages of hormonal therapy in effectiveness, risk profile, and longer experience compared with vigabatrin. (See 'Choosing initial therapy' above and 'Hormonal therapy' above.)
•Vigabatrin for infants with tuberous sclerosis complex – For infants with IESS and tuberous sclerosis complex, vigabatrin is first-line therapy. (See 'Choosing initial therapy' above and 'Vigabatrin' above and "Tuberous sclerosis complex: Management and prognosis", section on 'Seizure control'.)
•Dosing – Typical dose and taper recommendations for hormonal therapy and vigabatrin are listed in the table (table 1). (See 'Corticotropin (ACTH)' above and 'Glucocorticoids' above and 'Vigabatrin' above.)
-Hormonal therapy – Adverse effects are common and include hypertension, irritability, infection, reversible cerebral atrophy, and rarely death due to sepsis. Monitoring should include measurement of blood pressure (baseline and once weekly) and serum glucose, potassium, and sodium levels (baseline and every other week). Infectious contacts should be avoided, and infections should be treated promptly. (See 'Adverse effects of hormonal therapy' above.)
-Vigabatrin – Permanent visual field constriction due to retinal toxicity is a potentially severe adverse effect of vigabatrin. Ophthalmologic evaluation and monitoring is recommended. (See 'Adverse effects and monitoring' above.)
●Monitoring treatment response – Use of overnight inpatient video-EEG to evaluate treatment response is ideal; however, outpatient EEG studies up to 240 minutes can be considered as an alternative. Clinical observation frequently misses subtle spasms, and shorter EEGs are less sensitive for detecting hypsarhythmia. (See 'Monitoring treatment response' above.)
●Duration of initial therapy – Hormonal therapy is generally given at the maximum dose for 14 days, followed by a gradual taper starting on day 15, as outlined in the table (table 1). Vigabatrin is generally continued for six months in patients who respond to therapy, with continued evaluation for toxicity. (See 'Duration of initial therapy' above.)
●Relapses – For infants who relapse after termination of initial therapy, a second course of the agent that was effective in obtaining control should be administered. (See 'Relapses' above.)
●Refractory infantile spasms – Lack of a successful response to initial therapy within two weeks should prompt a change in treatment strategy. Alternatives for children who do not respond to hormonal therapy or vigabatrin include:
•Combination therapy – Treatment with both hormonal therapy and vigabatrin may be more effective than hormonal therapy alone, and many centers and protocols utilize combination therapy as first-line management. (See 'Combination therapy' above.)
•Ketogenic diet – A ketogenic diet may control spasms in cases refractory to first-line treatment. (See 'Ketogenic diet' above and "Ketogenic dietary therapies for the treatment of epilepsy".)
•Surgery – Patients with refractory infantile spasms who have focal brain lesions and no evidence of diffuse brain damage or degenerative or metabolic disease should be evaluated for early epilepsy surgical intervention. (See 'Surgical treatment' above.)
●Outcomes – The overall prognosis for children with IESS is guarded. Mortality ranges from 3 to 33 percent, and most patients will have impaired neurodevelopmental outcome and/or epilepsy (table 2). There is insufficient evidence to conclude that successful treatment of infantile spasms improves the long-term prognosis, although that is suggested by some observational data. (See 'Outcomes' above.)
ACKNOWLEDGMENT — The editorial staff at UpToDate acknowledges Daniel G Glaze, MD, who contributed to an earlier version of this topic review.
آیا می خواهید مدیلیب را به صفحه اصلی خود اضافه کنید؟