Traditionally the ketogenic diet is used for children in whom many anticonvulsant drugs have been unsuccessful. In those children who have significant seizure reduction, many parents ask “Why didn’t we try this sooner?” The truth is, for several kinds of epilepsy the diet may work better than drugs, and often also works better if used sooner (those who have tried fewer medications first). So why not use the diet earlier….and why not first?

The first formal study to evaluate the ketogenic diet first was published in 2005 by Rubenstein et al. In this study, 13 children who had been started on the diet either first or after one drug were reported. The diet worked well, with about half of children having 50% or greater seizure reduction, which is similar to results seen in those with difficult-to-control epilepsy.

In September, a follow-up study was published in the journal Epilepsia that looked at our experience at Johns Hopkins Hospital using the ketogenic diet as a first-line therapy for new-onset infantile spasms. Infantile spasms is a severe form of tonic seizures that will start in babies usually about 6 months of age and is mostly treated with ACTH (injectable steroids), prednisolone (oral steroids), or vigabatrin. All of these therapies have side effects and can be quite expensive. Since the ketogenic diet works well for children with infantile spasms that have not been controlled with these medications, we thought it would work well right away, too. The study also looked at the Johns Hopkins experience with ACTH over the same time period as a comparison group.

Thirteen babies were treated with the diet first; 8 (62%) became seizure-free within a few days. The diet was well-tolerated and the recurrence rate was lower than with ACTH. ACTH, however, did normalize the EEG a bit faster than the diet, but both (when successful) eventually did. If it worked, the diet was continued usually about 6 months, which is much shorter than the traditional 2-year seizure-free period with the diet. If it didn’t work, ACTH was then started immediately and was helpful in 4 of 5 babies, showing that time wasn’t “wasted” by giving the diet a chance first.

As a result of this research, at this time the ketogenic diet is offered to most parents of babies with new-onset infantile spasms as long as there aren’t any metabolic reasons not to use the diet. Many other centers are starting to use the diet in this way as well.

Will the ketogenic (or modified Atkins) diet be used first for other epilepsies besides infantile spasms in the future? Perhaps, but as a neurologist I would suspect the only seizure disorders that might make a parent willing to try a diet over a medication (which is certainly easier to start and maintain!) is one in which seizures are daily or even more frequent from the beginning. It may be even more acceptable for young babies or those with gastrostomy tubes, since it can be given relatively easily as a formula. Regardless, it is now certainly no longer acceptable to think of the ketogenic diet as a treatment of last resort.

REFERENCES

Rubenstein JE, Kossoff EH, Pyzik PL, Vining EPG, McGrogan JR, Freeman JM. Experience in the use of the ketogenic diet as early therapy. J Child Neurol 2005;20:31-34.

Kossoff EH, Hedderick EF, Turner Z, Freeman JM. A case-control evaluation of the ketogenic diet versus ACTH for new-onset infantile spasms. Epilepsia 2008;49:1504-1509.

Topic Editor: Robert Fisher, MD, PhD
Last Reviewed: 10/28/08