Response to comments on “Infantile spasms: A U.S. consensus report”


We thank Drs. Riikonen, Dulac, Fukuyama, and Galanopoulou for their thoughtful commentaries on “Infantile Spasms: A U.S. Consensus Report” (Pellock et al., 2010). Their comments amplify our effort to update the science and clinical practice regarding infantile spasms (West syndrome) since publication of the American Academy of Neurology/Child Neurology Society practice parameter (Mackay et al., 2004).

Both Dr. Riikonen and Professor Fukuyama discuss appropriate regimens and dosages based on practices in Europe/Finland and Japan, respectively. The conflicting data from studies of high- and low-dose hormonal treatment certainly underscores the need for well-controlled long-term studies at different dosages and with natural versus synthetic forms of adrenocorticotropic hormone (ACTH). We agree that the dosages used in the United States are generally higher than those used elsewhere [which is true of antiepileptic drugs (AEDs) in general as well as ACTH]. The dosages highlighted in our report are based on a randomized clinical trial demonstrating superiority of ACTH over prednisone (Baram et al., 1996), but we recognize that lower dosages have been reported in some trials. We also recognize that high-dose B6 (PAL-P or pyridoxal phosphate) is a therapeutic mainstay in Japan. We do recommend a challenge with pyridoxine to screen for pyridoxine-dependent seizures.

Throughout our report we stressed infantile spasms (IS) as a syndrome with multiple etiologies and presentations, so we agree with Dr. Dulac and colleagues that IS does not represent a single disease entity. The list of underlying conditions and etiologies continues to grow, and this complicates the appropriate evaluation of a child presenting with IS. Our Table 2 was organized to help clinicians follow a structured evaluation that would guide clinical practice, rather than a more detailed list of testing perhaps more suited to research protocols. The current grouping into cryptogenic and symptomatic IS remains reasonable and is supported by a large body of data, although subgroups such as tuberous sclerosis and trisomy 21 have already been distinguished due to their differential therapeutic responses. The new International League Against Epilepsy (ILAE) classification (Berg et al., 2010) provides a way to classify IS as a syndrome while taking into account different etiologies.

We appreciate Dr. Galanopoulou’s update and comments on the animal model section of our review, especially because the reported progress occurred in 2010, after collation of our consensus views. Because IS results from numerous etiologies and can be divided into symptomatic, cryptogenic, and idiopathic subtypes, it is reasonable that different animal models aim to recreate these subtypes. Dr. Galanopoulou’s work nicely fills the need to model symptomatic IS in the context of cortical and subcortical injuries. We are pleased that she concurs with the need to evaluate age, the presence of hypsarrhythmia, and the response to IS-ameliorating drugs in evaluating animal models and highlights the importance of animal models for a better understanding of optimal therapies for IS.

A difficult aspect of our reaching consensus on IS treatment is the lack of well-controlled long-term studies using consistent methods. Our panel agreed that a rigorous methodology requires electroencephalography (EEG) evidence of hypsarrhythmia at study onset and complete resolution of both clinical spasms and hypsarrhythmia (based on video-EEG) at study evaluation/conclusion. The preceding commentaries, along with our report, serve to focus our attention on the clinical and research work that needs to be done if we are to better understand IS and improve the lives of families affected by this disease.