Home Medications UK Infantile Spasms Study
UK Infantile Spasms Study PDF Print E-mail
Thursday, 05 June 2008 18:07
Citation: The Lancet, Volume 364, Issue 9447, 13 November 2004-19 November 2004, Pages 1773-1778;

The United Kingdom Infantile Spasms Study comparing Vigabatrin with prednisolone or tetracosactide at 14 days: a multi-centre, randomised controlled trial.

Andrew L Lux MD, Stuart W Edwards PhD, Elanor Hancock MD, Anthony L Johnson PhD, Colin R Kennedy MD, Richard W Newton MD, Finbar JK O'Callaghan PhD, Christopher M Verity MA, Prof John P Osborne MD and on behalf of participating investigators.



Infantile spasms, which comprise a sever infantile seizure disorder, have a high morbidity and are difficult to treat.  Hormonal treatments (adrenocorticotropic hormone or ACTH and prednisolone) have been the main therapy for decades, although little evidence supports their use.  Vigabatrin has been recorded to have a beneficial effect in this disorder.  We aimed to compare the effects of vigabatrin with those of prednisolone and tetracosactide in the treatment of infantile spasms.


The United Kingdom Infantile Spasms Study assessed these treatments in a multicentre, randomised controlled trial in 150 hospitals in the UK.  The primary outcome was cessation of spasms on days 13 and 14.  Minimum doses were vigabatrin 100mg/kg per day, oral prednisolone 40mg per day, or intramuscular tetracosactide depot 0.5mg (40IU) on alternate days.  Analysis was by intention to treat.


Of 208 infants screened and assessed, 107 were randomly assigned to vigabatrin (n=52) or hormonal treatments (prednisolone n=30, tetracosactide n=25).  None was lost to follow-up.  Proportions with no spasms on days 13 and 14 were: 40 (73%) of 55 infants assigned hormonal treatments (prednisolone 21/30 [70%], tetracosactide 19/25 [76%]) and 28 (54%) of infants assigned vigabatrin (difference 19%, 95% CI 1-36%, p=0.043).  Two infants allocated tetracosactide and one allocated vigabatrin received prednisolone.  Adverse events were reported in 30 (55%) of 55 infants on hormonal treatments and 28 (54%) of 52 infants on vigabatrin.  No deaths were recorded.


Cessation of spasms was more likely in infants given hormonal treatments than those given vigabatrin.  Adverse events were common with both treatments.

Additional Information

The following are additional excerpts from the paper that were useful to us.  I do not want to reproduce the entire paper here.


...spasms usually occur in clusters.... More than half of infants who develop infantile spasms have an underlying neurological disorder....

...The chaotic and high-voltage interictal electroencephalogram (EEG) pattern is called hypsarrhythmia, but this term might not be used because the EEG can vary with the underlying cause of the spasms and can change as the disorder progresses.  Onset of spasms is often associated with developmental arrest or regression...

...Infantile spasms are difficult to treat, because they do not respond to conventional antiepileptics.  Since 1958, the usual interventions have been hormonal treatments, with either intramuscular [injections]  of ACTH or oral  corticosteroids.  In the 1990's vigabatrin, a y-amniobutyric acid inhibitor, was introduced in Europe as an effective treatment for infantile spasms...  However since 1997, visual-field defects have been identified in 30% of adults treated with vigabatrin and have also been reported in children.  Hormonal therapy can also have severe adverse effects...


...Most serious adverse effects [of corticosteroids/tetracosactides]  have occurred after long term treatment.....

...Inclusion criteria were clinical diagnosis of infantile spasms...and a hypsarrhythmic (or similar) EEG.  Exclusion criteria were: age younger than 2 months or older than 12 months; diagnosis or high risk of tuberous sclerosis; previous treatment (with 28 days) with a contraindication to vigabatrin or hormonal treatments; a lethal or potentially lethal disorder other than infantile spasms; inability of parents or guardians to give informed signed consent or to know when spasms stop; leaving the UK within one month of randomisation; and enrollment in a concurrent trial...that might affect the outcome...

...The study treatments were prednisolone...tetracosactide....and vigabatrin... randomly allocated in the ratio of 1:1:2 respectively....

Prednisolone was given orally (10mg four times a day for 2 weeks, increasing to 20mg three times a day after 1 week if spasms continued).

Tetracosactide depot was given intramuscularly (0.5mg [40IU] on alternate days for two weeks, and increased to 0.75mg [60IU] on alternate days after 1 week if seizure control had not been achieved).

After two weeks of treatment, children allocated to prednisolone or tetracosactide  received a reducing dose of prednisolone with reductions of 10mg every 5 days or if on the higher dose of treatment, 40mg daily then 20mg, then 10mg for 5 day periods.

Vigabatrin was given orally in two divided doses per day (50mg/kg per day for the first two doses; increasing to 100mg/kg per day after 24h and, if spasms continued, to 150mg/kg per day after 96 h from the start of the treatment).

If a 48 hour period free of spasms was followed by further spasms before the end of day 14, treatment was increased to the maximum permitted by protocol.

...The primary outcome was cessation of spasms, defined as no reported spasms for at least 48 hours including days 13 and 14 after randomisation...

...For hormonal treatments, blood pressure was checked twice a day for 2 days then weekly, and urine tested for glucose at 48 hours and weekly....


...Of 208 infants assessed for eligibility, 107 (64 boys) were allocated a randomised treatment..... 83 (78%) of 107 EEGs were reported as hypsarrhythmia, and the remainder (apart from three) recorded as almost hypsarrhythmia; we reviewed their reports and believed them to be compatible with infantile spasms....
StatisticPrednisoloneTetracosactideHormonal TreatmentsVigabatrin
 (n=30)(n=25)Combined (n=55)(n=55)
Median Age (IQR) in completed
Months at onset of spasms
5 (4-6)*5 (3.5-7)**5 (4-6)*5 (4-7)
Age at randomisation (days)    
>= 2401171819
Median Age (IQR) at
Randomisation (months)
6 (4-8)6 (5-8)6 (4-8)6 (4-9)
Median duration of spasms (IQR)
at Randomisation (months)
0 (0-1.5)*1 (0-1)**1 (0-1)***1 (0-1)
Higher risk of neurodevelopmental
Chromosomal abnormality2022
Neonatal encephalopathy with seizures4598
Cerebral Palsy before spasms841210
Delayed development before spasms13122519
Underlying aetiology    
Other (uncertain classification)2246
No aetiology found13122521
Not known (cranial imaging not reported) 111
* Unknown in two infants. ** Unknown in one infant.  *** Unknown in three infants. **** Some infants had more than one risk factor.

...52 infants were allocated vigabatrin, 25 tetracosactide and 30 prednisolone... The dose was given in full as per protocol in 21 (70%) infants assigned prednisolone, 16 (64%) tetracosactide, and 29 (56%) vigabatrin.  In another 18, an appropriate dose was given according to adverse events reported (five assigned prednisolone, four tetracosactide, nine vigabatrin).  We could not provde teh dose was given as per protocol in 11 children (one on prednisolone, five on tetracosactide, five on vigabatrin), and the dose did not follow protocol in 12 (three on prednisolone, nine on vigabatrin)...

...Nine allocated prednisolone, nine tetracosactide, and 14 vigabatrin received at least one other anticonvulsant befire and throughout the trial for other seizure types.  Four infants received pyridoxine (one allocated prednisolone, three vigabatrin), and in one of these the possibilty of pyridoxine-dependant seizures could not be excluded...

...Adverse events were reported in a similar number on hormonal and vigabatrin treatments.  Infants showed increased irritability and appetite on hormonal treatments and more drowsiness on vigabatrin.  Treatment had to be stopped for one child in each hormonal group.  Blood pressure above 110/80 mm Hg occured in 11 (20%) of 55 infants allocated hormonal treatments (seven [23%] on prednisolone, four [16%] on tetracosactide) and above 120/90 mm Hg in eight (15%) of 55 (four [13%] of prednisolone, four [16%] on tetracosactide), and two infants allocated prednisolone were given diuretics.  Glycosuria was recorded in one patient allocated to tetracosactide, but no infants needed treatment for this condition or needed treatment to be stopped because of diabetes mellitus.

StatisticPrednisoloneTetracosactideHormonal Treatment subgroupVigabatrin
Number of infants with adverse events19113028
Treatment unchanged148*2219
Treatment not increased as protocol required32*57
Treatment reduced11*22
Treatment stopped11*20
Specific Adverse Events**    
Increased Appetite4371
Fluid and electrolyte (including high
blood pressure)
Neuropsychiatric (including sleep disturbance)1014
Treatment for varicella exposure1120

* Treatment actions not mutually exclusive.  ** No adverse events reported in the following categories: hypertonia, endocrine/metabolic (including excessive weight gain), opthalmic (although visual field losses would be undetectable), musculoskeletal, varicceaal infection, and immunosupression.  ** Fever, culture-positive disease, or treatment with antibiotics, but not varicalla.


...Hormonal treatment prevented significantly more infantile spasms over the 48h period  comprising days 13 and 14  after randomisation than vigabatrin...resolution of hypsarrithmia occured in significantly more infants allocated hormonal treatments than allocated vigabatrin...

...With respect to our treatment dose, little is known about the metabolism of steroids in infancy, but some evidence suggests that the increased perhiperal detoxifcation of steroids seen in newborn infants takes some time to dissapear.  Older paediatricians remember being taught that children need an adult does of prednisolone.  Asthama treatments for children are much higher than adult does on a bodyweight basis.  Most adverse events due to steroids are not dose-depenent.  We did not want to use too small a dose and miss an important effect....

...Vigevano and Cilio compared ACTH ....with vigabatrin... in a study similar but smaller than our own.  ...cessation of spasms in 74% given ACTH and 48% of those given vigabatrin (difference 26%, 95% CI - 3% to 54%).... at 21 days, and all responders were free of spasms by day 14....