Can You "Fine-Tune" the Ketogenic Diet?

Epilepsy News From: Thursday, September 03, 2009

Perhaps nothing is more upsetting to parents or frustrating for neurologists and dietitians than a child who is doing spectacularly well on the ketogenic diet, and then begins to lose seizure control. There may be a simple answer, such as a carbohydrate “indiscretion”, febrile illness, or skipped dose of anticonvulsant medication. However, many times there are no clues as to why a particular child is having increased seizures.

In order to improve seizure control, many neurologists and dietitians will then attempt to "fine-tune" the ketogenic diet. With a therapy as complex as the ketogenic diet, there are so many potential changes that can be made that it seems logical to try. Certainly, most parents are strongly in favor of making a change in the diet versus adding or increasing the dose of anticonvulsants. Being as the ketogenic diet may have been the most successful therapy for these children, there is a natural tendency to try and adjust it first, compared to medications.

What are these dietary changes? The most common adjustment is calories. One common myth regarding the ketogenic diet is that significant weight gain can lead to loss of seizure control. In 2007, Dr. Hamdy reported in Journal of Child Neurology that changing the diet to achieve an ideal body mass index (BMI) did not lead to better seizure control. Another adjustment is to the ketogenic ratio (fat: carbohydrates and protein combined). Usually neurologists will increase the ratio (e.g. from 3:1 to 4:1) to achieve higher ketosis and seizure control. Does it help? In a study by Dr. Seo and colleagues published in Epilepsia also in 2007, investigators increased the ratio from 3:1 to 4:1 in children who were not seizure-free after 3 months. 9 (75%) of 12 had further improvement in seizures, however, none became seizure-free with the change. Of course, for an individual child these changes may be very important: always consult with your neurologist and dietitian if you are considering these changes and do not consider this research to imply there is no possibility of a seizure-free outcome by a ratio or calorie change. Lastly, I have had some anecdotally good experience at times in my patients lowering the ketogenic ratio, switching to the modified Atkins diet, fasting for 12-24 hours and then restarting the diet, and eliminating all higher glycemic carbohydrates.

What else can be done? My personal suspicion is that the better choice is to add a supplement (or anticonvulsant) to the ketogenic diet in order to achieve better efficacy, rather than trying to change the diet itself. Previously described supplements include medium chain triglyceride (MCT) oil and carnitine, although there is no published evidence for their use in this manner. As there is suggestion that zonisamide may be helpful when used in combination with the ketogenic diet, as well as vagus nerve stimulation (VNS), these therapies could also be added. In an article just published last month in the Journal of Child Neurology, Dr. Evangeliou and his colleagues from Greece have another interesting suggestion. They reported results from providing a powder of up to 20 grams of branched-chain amino acids to 17 children receiving a 4:1 ketogenic diet who were not seizure-free. These branched-chain amino acids (leucine, isoleucine, and valine) were in sufficient quantities to lower the ratio to 2.5:1 (which might theoretically reduce ketosis and seizure control). The opposite happened: 3 became seizure-free and an additional 5 had a 50-90% reduction in seizures without loss of ketosis. Unfortunately, no child with a less than 50% reduction in seizures improved with this supplement, only those who had some level of prior seizure control. Although obviously a small study, it raises a provocative question as to whether these amino acids may have independent effects on seizure control outside of the diet or is it a combination effect.

In conclusion, researchers are actively looking for ways in which we can help those children who initially respond to the ketogenic diet, but then later lose seizure control. Fine tuning the diet has gone from personal experience to scientific evidence. No family wants to give up on the ketogenic diet after making the time investment in starting it, but this is especially true for the family in whom seizure control was previously present and wonderful. This research confirms my belief that improvement may come more frequently in a pill or additional food substance, rather than by changing the foods the child is already eating.


  • Evangeliou A, Spilioti M, Doulioglou V, et al. Branched Chain Amino Acids as Adjunctive Therapy to Ketogenic Diet in Epilepsy: Pilot Study and Hypothesis. J Child Neurol 2009 Aug 17 (Epub Online Early).
  • Hamdy RF, Turner Z, Pyzik PL, et al. Lack of influence of body mass index on the efficacy of the ketogenic diet. J Child Neurol 2007;22:1167-1171.
  • Seo JH, Lee YM, Lee JS, et al. Efficacy and tolerability of the ketogenic diet according to lipid:nonlipid ratios--comparison of 3:1 with 4:1 diet. Epilepsia. 2007;48:801-805.

Authored by

Eric Kossoff MD

Reviewed by

Robert Fisher MD, PhD

Reviewed Date

Wednesday, September 02, 2009

Join Our Newsletter

Want to get notified about new devices? Stay up to date with the latest epilepsy news and stories from the community.