By Ellen Feldman, MD
Altru Health System, Grand Forks, ND
Dr. Feldman reports no financial relationships relevant to this field of study.
SYNOPSIS: This retrospective study of 59 children with refractory seizures of genetic etiology investigates the impact of a strict low-carb diet (either modified Atkins or ketogenic diet) on seizure reduction.
SOURCE: Jagadish S, Payne ET, Wong-Kisiel L, et al. The ketogenic and modified Atkins diet therapy for children with refractory epilepsy of genetic etiology. Pediatr Neurol 2019;94:32-37.
- This retrospective investigation looked at decrease in seizure frequency in 59 children with seizures of genetic etiology who were treated with a low-carb diet (modified Atkins diet or ketogenic diet).
- A reduction in seizures > 50% was measured periodically, beginning at one month into the study and ending at two years (24 months).
- Response rates to this diet for refractory seizures in children are about 50%; in this population, response rates were 63% at the end of the first month and showed a gradual decline to 41% at the end of the study (24 months.)
- Adverse effects included hypoglycemia, vomiting, and refusal to feed. At study conclusion, 47% of the study group remained on the diet.
“The keto craze is hitting the mainstream,” proclaimed a banner headline on CNN in September 2018.1 “The high-fat, low-carb diet has swept Hollywood,” noted The New York Times in March 2019, pointing to weight loss and energy gain as the main attraction of this diet.2 All of the hype and recent attention may lead to the idea that the ketogenic diet (KD) is a new phenomenon.
The medical world, however, has been aware and interested in dietary control of disease since the ancient time of Hippocrates.3 More recently, medical interest in the benefits of a diet mimicking a fasting state and specifically promoting ketosis emerged in the early 1900s. Based on a theory that a toxin released in the intestines causes seizures and that a fasting state could help free the body of these poisons, most of the early 20th century studies of controlling seizures and epilepsy revolved around dietary interventions. In 1921, after ketones were noted to be produced by the liver in response to a fasting state or a low-carbohydrate and high-fat diet, Mayo Clinic physician Russell Wilder coined the term “ketogenic diet,” and was among the first to use this ketone-producing diet plan to treat epilepsy. Subsequent investigations developed an optimal formula for KD in children, which remains as a guideline today: 1 g of protein/kg of body weight daily, 10-15 g of carbohydrate daily, and the remainder of daily calories (55-60% of total calories) in fat.4,5
The low-carbohydrate intake in this diet leads to several metabolic shifts including gluconeogenesis (endogenous glucose production) and ketogenesis. Once the body's stores of glucose are depleted, ketones are available as an alternate source of energy. Notably, this metabolic condition is termed a “nutritional ketosis,” which is different from life-threatening ketoacidosis — a state in which ketone bodies are produced in much larger quantities.4,5
However, the restrictive nature and some of the adverse effects of this diet made it unappealing to many families. The advent of new antiepileptic drugs (AEDs) during the late 1900s led to a slowing of the use of KD in seizure control. However, when it became clear that about 10-20% of children remained with intractable seizures (defined as seizures that occur despite two adequate trials of appropriately dosed AEDs), the use of KD re-emerged in the early 2000s.4,5 Jagadish et al set out to define the reach and limits of KD, which currently is considered an alternative therapy for intractable seizures, by determining response rate and tolerability of the diet in children with refractory epilepsy of genetic origin.
This work is a single-center, retrospective chart review of 59 patients with refractory epilepsy of genetic origin presenting to the Mayo Clinic in Rochester, MN between 2005 and 2016. While 53 of the 59 children were started on the KD, six were started on a modified Atkins diet (MAD). The International Ketogenic Diet Study Group noted MAD as one of four recognized KDs useful in seizure control: classic KD, MAD, medium-chain triglyceride diet, and low glycemic index treatment.
The children were a median age of 6 months when seizures first were noted and 2.2 years when starting the diet. While the type of seizure varied, just over 50% were focal seizures.
According to the International Ketogenic Diet Study Group, the expected rate of seizure reduction in children with intractable seizures of any origin on the KD is about 50% patient response (based on a response of > 50% seizure reduction). This consensus group reported a specific diagnosis with a more robust response rate, but noted that evidence-based studies are needed to confirm such conclusions.6
In this evidence-based study by Jagadish et al, the overall response rates at each time point between months one and 24 are noted in Table 1. Also noted in this table is the proportion of children remaining on the chosen diet at each study follow-up point, as well as the children remaining in the study for follow-up. There was no specified control group because response was defined as a reduction in seizures > 50% from baseline prior to diet initiation.
When analyzed for differences in response according to gender, age of diet onset, age of seizure onset, and specific genetic etiology, no statistically significant differences were identified.
Tolerability of the KD was important to determine in this young population. During initiation of the chosen diet, of the 59 patients, 46 were hospitalized. Of these 46 patients, 20 (43%) experienced a complication during initiation, ranging from mild emesis to significant hypoglycemia. Ten of the children required intervention during this time; the majority of these patients were under 2 years of age. During the 12-month study, seven patients reported adverse effects, including severe reflux, severe ketoacidosis, weight loss, and persistently increased triglycerides, and stopped the diet. Eighteen patients reported a variety of adverse effects, including mild weight loss, mild abdominal pain, food refusal, mild emesis, and constipation, and did not stop the diet.
On the surface, this study appears mostly relevant to the providers, patients, and families of those with intractable seizures of genetic origin. It certainly gives validation of the KD as a treatment option for this population, and adds important data for this group. However, the impact of this study is even more widespread, and serves as a reminder of the need to investigate all interventions with the care and scientific scrutiny given to conventional pharmacologic treatments.
We know that medication dosage and indications vary according to the disorder and individual; it is only logical that these parameters vary for nonpharmacologic interventions as well. The widespread popularity of the KD for nonmedical uses as well as investigatory uses for a variety of disorders, including type 2 diabetes, obesity, Alzheimer’s disease, and autism spectrum disorders is exciting,7 but studies such as this remind us of the need for robust and scientifically rigorous studies. Results from such investigations can clarify the reach, applicability, and limits of the KD, while potentially revealing underlying mechanisms of action.
There are many theories regarding how the KD helps with seizure control, but a definitive answer remains elusive. Investigators agree that an increase in ketone bodies and a decrease in blood glucose may modify neuronal excitability, and that the research into mechanism of action is a priority in the field.7 As we further our understanding of the KD and its effect on a metabolic and cellular level, more targeted uses of this potentially powerful, nonpharmacologic tool can emerge.
Jagadish et al noted several limitations of their study, including the difficulty of determining why children were lost to follow-up. The design itself is a limitation — as with any retrospective study, cause and effect is difficult to determine with certainty. Additionally, there was no control for concurrent use of AEDs, or changes in AEDs, over the course of the study.
Maintenance of a stringent KD in a child requires strong parental involvement. The link between response rate and parental oversight is unknown, but remains a clear direction for future studies.
The take-home message for the primary care and integrative provider is two-fold. Certainly for those working with this patient population, it is important to consult with a dietician and neurologist to implement, maintain, and monitor a KD. This diet can be difficult to initiate, harder to maintain, and has potential serious side effects; working with a skilled team can help address these challenges on multiple levels. There is also utility in thinking about such a diet for adults with intractable seizures of genetic etiology, although that age group was not covered in this study.
In a broader sense, this study reminds us of the importance of diet to health and the often significant, clinically relevant impact of dietary changes.
- Wiener-Bronner D. The keto craze is hitting the mainstream. CNN website. Available at: . Published Sept. 17, 2019. Accessed July 16, 2019.
- Williams A. What’s the skinny on the keto diet? The New York Times website. Available at: . Published March 15, 2019. Accessed July 16, 2019.
- Tsiompanou E, Marketos SG. Hippocrates: timeless still. J R Soc Med 2013;106:288–292.
- Masood W, Uppaluri KR. Ketogenic Diet. StatPearls. Available at: . Updated March 21, 2019. Accessed July 10, 2019.
- Wheless JW. History and Origin of the Ketogenic Diet. In: Stafstrom CE, Rho JM, eds. Epilepsy and the Ketogenic Diet. Nutrition and Health. Totowa, NJ: Humana Press; 2004.
- Kossoff E, Zupec-Kania, Auvin S, et al. Optimal clinical management of children receiving dietary therapies for epilepsy: Updated recommendation of the International Ketogenic Diet Study Group. Epilepsia Open 2018;3:175-192.
- D’Andrea Meira I, Romão TT, Pires do Prado HJ, et al. Ketogenic diet and epilepsy: What we know so far. Front Neurosci 2019;13:5.