Treatment of Infantile Spasms: Questions remain

An evidence-based update to the 2004 guidelines for the treatment of infantile spasms has recently been published. Important new recommendations include use of low-dose adrenocorticotropic hormone (ACTH) over high-dose ACTH or vigabatrin. A paucity of data, however, leaves several key questions unanswered

Infantile spasms represent an epileptic encephalopathy that usually occurs before 2 years of age, and is accompanied by cognitive, motor and sensory impairment. 80% of infantile spasms are classed as ‘symptomatic’, for which a history of pre-, peri- or postnatal cerebral pathology exists, and the predisposing aetiology can be identified. In the cryptogenic group (15–20%), development is normal prior to the onset of the spasms, MRI scans show no abnormalities, and the cause is unknown. The rare ‘idiopathic’ form of the disease (accounting for 5% of cases) is a purely functional cerebral abnormality, in which genetic factors play an important part.(Riikonen 2009)

Clinicians currently choose a therapy for infantile spasms on the basis of published evidence and less-objective measures, such as personal experience, costs and adverse-effect profile. The main therapeutic options are adrenocorticotropic hormone (ACTH) or the antiepileptic drug vigabatrin, but treatment of this condition varies considerably, necessitating an evidence-based review.

In the recent paper, published in Neurology,(Go 2012) the American Academy of Neurology and the Practice Committee of the Child Neurology Society provide an update of their guidelines published in 2004.(Mackay 2004). The researchers carried out database searches of MEDLINE® and Embase® for relevant studies conducted during the time period, 2002–2011, and placed their findings in the context of pre-2002 evidence. 68 articles were selected for detailed review, and 26 were included in the analysis. Recommendations were made on a four-tier classification scheme, designated levels A, B, C and U.

The aim of the guideline update (Go 2012) was to address five questions that could not be answered in 2004 owing to insufficient data.

First, are other forms of corticosteroids as effective as ACTH for short-term treatment of infantile spasms?
Evidence was insufficient to recommend the use of prednisolone, dexamethasone or methylprednisolone over ACTH (level U recommendation).

However, one large class III randomized controlled trial (RCT) found no significant difference in the response rate to synthetic ACTH and high-dose (40–60 mg/d) oral prednisolone.(Lux, 2004) This important finding needs to be confirmed because the study was underpowered to compare treatment efficacy. Many treatment centres favour oral preparations (usually prednisone, the prodrug formulation of prednisolone) over ACTH, which requires intramuscular injection.

An important caveat to this approach is that discontinuation of high-dose prednisolone can lead to suppression of the hypothalamus–pitutary–adrenal axis and, thereby, adrenal insufficiency, which can be fatal. Hydrocortisone substitution therapy should be given and continued until the adrenal axis has normalized.

The second question to be addressed was whether low-dose ACTH, as opposed to high-dose ACTH, is effective for short-term treatment of infantile spasms.
No new studies have been carried out to address this issue since the 2004 guidelines but, on the basis of four studies conducted before 2000, the current guidelines state that low-dose (20–40 IU/d) ACTH is probably as effective as high-dose (150 IU/d) ACTH for short-term treatment of infantile spasms (level B recommendation).(Go 2012)

In addition, a prospective study that began in the 1960s showed that patients given low-dose ACTH had better long-term cognitive outcomes and fewer adverse effects than did patients treated with high-dose ACTH.(Riikonen 2001) These results, however, were not included in the current analysis.

Third, the authors of the current study assessed data on the relative efficacy of ACTH versus vigabatrin for short-term treatment of infantile spasms.(Go 2012)
Combined results from two class III studies indicated that hormonal therapy leads to resolution of spasms three times faster and in more infants than does vigabatrin.(Lux, 2004) By contrast, one small study did not find a difference in efficacy between the two drugs.

The resultant recommendation was that ACTH (level B recommendation) or vigabatrin (level C recommendation) may be offered for short-term treatment of infantile spasms.(Go 2012) The fact that child development might be better following hormonal therapy compared with vigabatrin(Darke 2010) makes ACTH an even more attractive therapeutic option.

Vigabatrin was effective for short-term treatment of infantile spasms in children with tuberous sclerosis, the most important genetically defined disorder in infantile spasm syndrome, in two class III studies from the 2004 practice parameter (level C recommendation).

The fourth question was whether the ketogenic diet, novel therapies or combination therapy should be considered as treatment options for infantile spasms.
The ketogenic diet has emerged as a potential nonpharmacological therapy predominately for intractable spasms. Only one class IV study on this intervention was available for the current guideline update, however, providing insufficient evidence to recommend its use. Similarly, evidence was insufficient to recommend other therapies (valproic acid, vitamin B6, nitrazepam, levetiracetam, zonisamide, topiramate), or combination therapies such as ACTH plus magnesium sulphate.(Go 2012))

The final issue to be addressed in the guideline update was long-term outcomes in infantile spasms.
Important information came from the UK Infantile Spasms Study (UKISS; class II study). Patients with cryptogenic spasms who were initially allocated to hormonal therapy had a better cognitive outcome at 4 years of age than did those allocated to vigabatrin. Furthermore, scores reflecting neurocognitive development at 4‑year follow-up worsened with increasing lag-time from onset of spasms to initiation of treatment.

The updated guidelines make a level C recommendation for early hormonal therapy in preference to vigabatrin.(Go 2012)

Why does early hormonal treatment lead to better long-term neurodevelopmental outcome than treatment with vigabatrin?
Long duration of hypsarrhythmia (as seen with vigabatrin treatment in UKISS) might damage early synaptic development and have long-term consequences for brain development. Patients who respond poorly to treatment express low levels of insulin-like growth factor, a factor that is essential for synaptic development.(Riikonen 2010) In addition, vigabatrin therapy itself might have toxic effects on the developing brain.(Bittigau 2003)

In 2009, the US FDA approved vigabatrin for use in infantile spasms, with a warning of potential permanent visual impairment. The drug is only available through the Lundbeck programme, which calls for regular visual testing of patients receiving the drug. This programme would, however, be extremely difficult in practice, as visual field examination cannot be carried out in young infants by current methods. Electroretinogram (ERG) needs anaesthesia, which carries a risk in infants, and can influence the ERG results. Furthermore, ERG findings do not correlate significantly with visual field loss (VFL)(Moskowitz 2012).

A multicentre, international study is currently under way to evaluate the frequency of VFL in school-age children who were treated in infancy for infantile spasms.(Riikonen 2012) It remains to be seen whether retinal damage can be prevented through taurine supplementation and by avoiding light exposure.

The recent update(Go, 2012) does not provide definitive answers to the five questions presented. Only one class I study was available for inclusion in the analysis, and all recommendations were level B or weaker, highlighting the lack of adequate trials in infantile spasms and the need for further research. Important ‘take-home messages’ can, however, be found: hormonal treatment resolves spasms faster and in more infants than does vigabatrin; hormonal treatment might result in better long-term outcomes in cryptogenic cases; other treatments cannot be recommended on the basis of evidence reviewed; low doses of ACTH are as effective as high doses; and short treatment lag possibly improves the long-term outcome. In light of the evidence presented in the currently study and my own work 30 years ago,5 I strongly agree with the last two statements in particular.

Raili Riikonen
Department of Child Neurology, University of Eastern Finland, Puijonlaaksontie 2, PO Box 1627, FI‑70211 Kuopio, Finland.
raili.riikonen@kolumbus.fi

References

• Riikonen, R. in Long-term Evolution of Epileptic Encephalopathies (eds Nikanorova, M. et al.) 13–28 (John Libbey Eurotext, Esher, 2009).
• Go CY, Mackay MT, Weiss SK, Stephens D, Adams-Webber T, Ashwal S et al. (2012) Evidence-based guideline update: medical treatment of infantile spasms. Report of the Guideline Development Subcommittee of the American Academy of Neurology and the Practice Committee of the Child Neurology Society.Neurology 78 (24):1974-80. DOI: 10.1212/WNL.0b013e318259e2cf PMID: 22689735.
• Mackay, M. T. et al. Practice parameter: medical treatment of infantile spasms: report of American Academy of Neurology and Child Neurology Society. Neurology 62, 1668–1681 (2004).
• Lux, A. L. et al. The United Kingdom Infantile Spasms Study comparing vigabatrin with prednisolone or tetracosactide at 14 days: a multicentre, randomized controlled trial. Lancet 364, 1773–1778 (2004).
• Riikonen, R. Long-term outcome of patients with West syndrome. Brain Dev. 23, 683–687 (2001).
• Darke, K. et al. Developmental and epilepsy outcomes at age 4 years in the UKISS trial comparing hormonal treatments to vigabatrin for infantile spasms: a multicentre randomized trial. Arch. Dis. Child. 95, 382–386 (2010).
• Riikonen, R., Jääskeläinen, J. & Turpeinen, U. Insulin-like growth factor‑1 is associated with cognitive outcome in infantile spasms. Epilepsia 51, 1283–1289 (2010).
• Bittigau, P. Sifringer, M. & Ikonomidou, C. Antiepileptic drugs and apoptosis in the developing brain. Ann. NY Acad. Sci. 993,103–114 (2003).
• Moskowitz, A., Hansen, R. M., Eklund, S. E. & Fulton, A. B. Electroretinographic (ERG) responses in pediatric patients using vigabatrin. Doc. Ophthalmol. 124, 197–209 (2012).
• Riikonen, R. Does vigabatrin treatment for infantile spasms bring visual field defects? Presented at the Joint 12th International Child Neurology Congress and 11th Asian and Oceanian Congress of Child Neurology, http://icnapedia.org/lectures/53-icnc2012/sig-meetings/2484-vigabatrin-visual-field-defects (2012).

Article published in Nature Reviews Neurology Riikonen, R. Nat. Rev. Neurol. advance online publication 7 August 2012; doi:10.1038/nrneurol.2012.157