TL;DR: The ketogenic diet is one of the oldest and most evidence-backed dietary interventions in all of medicine, with randomized controlled trials demonstrating that roughly 50% of children with drug-resistant epilepsy achieve a 50% or greater reduction in seizure frequency. For those who find the classical protocol too restrictive, the modified Atkins diet (MAD), MCT oil diet, and low glycemic index treatment offer meaningful alternatives with strong-to-moderate evidence. The mechanisms extend well beyond simple ketosis, involving shifts in GABA/glutamate balance, adenosine signaling, BDNF expression, and gut microbiome composition. Dietary therapy for epilepsy should always be undertaken with medical supervision, but for the right patients it can be transformative.

Introduction

Epilepsy affects approximately 50 million people worldwide, making it one of the most common neurological disorders. While anti-seizure medications (ASMs) remain the first-line treatment and achieve seizure freedom in roughly two-thirds of patients, the remaining third face a difficult reality: drug-resistant epilepsy, defined as failure to achieve sustained seizure freedom after adequate trials of two appropriately chosen and tolerated ASMs.

For these patients, dietary therapy represents one of the most underutilized tools in neurology. The idea that what you eat can influence seizure activity is not a fringe concept or recent discovery. It is a clinical approach with roots stretching back to antiquity and a modern evidence base that includes randomized controlled trials, systematic reviews, and over a century of documented clinical use.

Yet dietary therapy for epilepsy remains surprisingly underemployed. A 2015 survey of neurologists in the United States found that fewer than half routinely discussed dietary options with patients who had drug-resistant epilepsy. This gap between evidence and practice is worth closing.

This article examines the major dietary approaches to epilepsy, the biological mechanisms that explain why they work, who is most likely to benefit, and how to implement these strategies safely and effectively.

The Classical Ketogenic Diet: A Century of Evidence

Historical Context

The therapeutic use of fasting for seizure control dates to at least 500 BCE, when Hippocrates described the case of a man whose seizures resolved after complete abstinence from food and drink. The modern ketogenic diet emerged in the 1920s when Dr. Russell Wilder at the Mayo Clinic proposed that a high-fat, very-low-carbohydrate diet could mimic the metabolic effects of fasting without the obvious limitation of starvation. Wilder coined the term “ketogenic diet” in 1921 and published early clinical results demonstrating efficacy in epilepsy.

The diet was widely used through the 1930s and 1940s but fell out of favor with the introduction of phenytoin and other effective anti-seizure medications. It experienced a resurgence in the 1990s, driven largely by the work of John Freeman and colleagues at Johns Hopkins Hospital, and by the story of Charlie Abrahams, whose drug-resistant epilepsy responded dramatically to the ketogenic diet, leading his father to establish the Charlie Foundation to promote dietary therapy research.

The Neal 2008 Randomized Controlled Trial

The study that moved the ketogenic diet from observational evidence to randomized trial-level support was the landmark RCT published by Elizabeth Neal and colleagues in The Lancet Neurology in 2008. This multicenter trial enrolled 145 children aged 2 to 16 years with drug-resistant epilepsy (at least daily seizures despite trials of two or more ASMs) and randomized them to either an immediate ketogenic diet group or a control group that continued standard treatment for three months before starting the diet.

The results were unequivocal. After three months, the diet group showed a mean reduction in seizure frequency of 38% compared to baseline, while the control group showed no significant change. Among those on the diet, 38% achieved a greater than 50% reduction in seizures, and 7% achieved a greater than 90% reduction. No children in the control group achieved these thresholds.

These results are particularly impressive given that the study population consisted entirely of patients who had already failed multiple medications. The Neal trial established Level 1 evidence for the ketogenic diet in pediatric drug-resistant epilepsy and prompted international consensus guidelines recommending its use.

What the Classical Ketogenic Diet Involves

The classical ketogenic diet typically uses a 4:1 or 3:1 ratio of fat to combined protein and carbohydrate by weight. In practical terms, this means that roughly 85-90% of daily calories come from fat, 6-8% from protein, and only 2-4% from carbohydrate.

A typical day might include:

  • Breakfast: Scrambled eggs cooked in butter with cream cheese, served with a small portion of berries
  • Lunch: Tuna salad made with mayonnaise and olive oil, served on a lettuce wrap with avocado
  • Dinner: Grilled salmon with butter sauce, a small serving of steamed broccoli with cream, and a fat bomb dessert made from coconut oil and cocoa

Every meal must be precisely calculated and weighed, typically with a digital gram scale. Carbohydrate intake is usually restricted to 10-20 grams per day. This level of precision is one of the reasons the classical ketogenic diet is most commonly initiated in a hospital setting under the supervision of a ketogenic diet team consisting of a neurologist, dietitian, and often a nurse specialist.

Beyond the Classical Protocol: Alternative Dietary Therapies

The strictness of the classical ketogenic diet has long been recognized as its primary limitation. Many patients — particularly adolescents and adults — find the protocol difficult to sustain. Over the past two decades, several less restrictive alternatives have been developed and studied, each with meaningful evidence behind it.

The Modified Atkins Diet (MAD)

The modified Atkins diet was developed at Johns Hopkins Hospital by Eric Kossoff and colleagues in the early 2000s as a more practical alternative to the classical ketogenic diet. Named for its similarities to the Atkins weight-loss diet, the MAD restricts carbohydrates to approximately 10-20 grams per day in children and 15-20 grams per day in adults, but does not restrict calories, fluid, or protein, and does not require precise weighing of fat intake.

Kossoff and colleagues published initial results in 2003 in Epilepsia, reporting that 6 of 20 children with drug-resistant epilepsy achieved a greater than 50% seizure reduction, with 3 achieving a greater than 90% reduction. Since then, multiple prospective studies and several randomized controlled trials have confirmed these findings.

A pivotal randomized trial by Sharma and colleagues, published in Epilepsia in 2013, compared the MAD to a standard diet in 102 children with drug-resistant epilepsy. At three months, 52% of children on the MAD achieved a greater than 50% seizure reduction, compared to 11.5% in the control group. These results are remarkably similar to those seen with the classical ketogenic diet, despite the MAD being substantially easier to follow.

The MAD has become particularly popular for adolescents and adults, populations in which the classical ketogenic diet is often considered impractical. A 2008 prospective study by Kossoff and colleagues in Epilepsy & Behavior specifically evaluated the MAD in adults with drug-resistant epilepsy and reported that 33% achieved a greater than 50% seizure reduction at six months.

The Medium-Chain Triglyceride (MCT) Diet

The MCT diet was originally developed by Peter Huttenlocher in the 1970s as a way to make the ketogenic diet more palatable. Medium-chain triglycerides — found naturally in coconut oil and available as a concentrated supplement — are more ketogenic per calorie than long-chain triglycerides, meaning that less total fat is needed to achieve ketosis. This allows a greater proportion of calories to come from carbohydrate and protein, making the diet feel less restrictive.

In the MCT protocol, approximately 30-60% of calories come from MCT oil, with the remainder from conventional foods. The trade-off is that MCT oil can cause gastrointestinal side effects — nausea, cramping, and diarrhea — particularly when introduced too quickly.

The Neal 2008 trial actually included both a classical ketogenic diet arm and an MCT diet arm. Both arms showed similar seizure reduction efficacy, with no statistically significant difference between them. This finding provided RCT-level evidence that the MCT diet is a viable alternative to the classical protocol.

The Low Glycemic Index Treatment (LGIT)

The low glycemic index treatment, developed by Elizabeth Thiele and colleagues at Massachusetts General Hospital and described in a 2005 publication in Neurology, represents the most liberal of the established dietary therapies for epilepsy (for more on the cognitive benefits of this approach, see low-glycemic eating and the brain). It restricts carbohydrate intake to approximately 40-60 grams per day — substantially more than the ketogenic diet or MAD — but specifies that all carbohydrates consumed must have a glycemic index below 50.

This means that refined grains, potatoes, sugar, and most processed foods are eliminated, while legumes, most fruits, non-starchy vegetables, and whole grains with a low glycemic index are permitted.

Retrospective and prospective case series have reported that approximately 50% of patients on the LGIT achieve a greater than 50% seizure reduction, though randomized controlled trial data are more limited compared to the ketogenic diet and MAD. A 2009 study by Muzykewicz and colleagues, published in Epilepsia, followed 76 patients on the LGIT and found that 50% achieved at least a 50% reduction in seizure frequency at one year.

The LGIT is often considered a first-line dietary option for patients who are unwilling or unable to follow stricter protocols, and as a potential long-term maintenance diet for patients who have achieved seizure control on a stricter protocol and wish to liberalize their intake.

Mechanisms: Why Diet Affects Seizures

The biological mechanisms connecting dietary therapy to seizure control are more complex and more interesting than the simple concept of “ketosis controls seizures.” Research over the past two decades has identified multiple converging pathways.

GABA and Glutamate Balance

Seizures fundamentally reflect an imbalance between excitation and inhibition in neural circuits. Glutamate is the brain’s primary excitatory neurotransmitter; gamma-aminobutyric acid (GABA) is its primary inhibitory counterpart. In epilepsy, this balance is tilted toward excitation.

Ketone bodies — particularly beta-hydroxybutyrate and acetoacetate — influence this balance through several mechanisms. They serve as alternative substrates for the tricarboxylic acid (TCA) cycle, altering the metabolism of glutamate and increasing its conversion to GABA via glutamic acid decarboxylase. Studies using magnetic resonance spectroscopy have demonstrated elevated brain GABA levels in patients on the ketogenic diet (Dahlin et al., 2005). Additionally, ketone bodies appear to directly inhibit vesicular glutamate transporters, reducing excitatory neurotransmitter release (Juge et al., 2010).

Adenosine Signaling

Adenosine is an endogenous anticonvulsant. It acts via A1 receptors to hyperpolarize neurons and reduce synaptic transmission. The ketogenic diet appears to increase adenosine signaling through at least two mechanisms: reducing the activity of adenosine kinase (the enzyme that metabolizes adenosine) and increasing extracellular adenosine levels. Work by Detlev Boison and colleagues has demonstrated that the anticonvulsant effects of the ketogenic diet are partially dependent on adenosine A1 receptor activation, and that the diet can reduce adenosine kinase expression epigenetically through DNA methylation changes (Masino et al., 2011).

BDNF and Neuroprotection

Brain-derived neurotrophic factor (BDNF) supports neuronal survival, synaptic plasticity, and the development and maintenance of inhibitory interneuron circuits (for more on this growth factor, see foods that increase BDNF). Dysregulation of BDNF signaling has been implicated in epileptogenesis. The ketogenic diet has been shown to upregulate BDNF expression in animal models, potentially contributing to both acute seizure control and longer-term neuroprotective effects (Vizuete et al., 2013). This neuroprotective dimension may help explain why some patients continue to experience seizure reduction even after discontinuing the diet — a phenomenon well-documented in the clinical literature.

Gut Microbiome

One of the most exciting recent discoveries in this field is the role of the gut microbiome. A landmark 2018 study by Olson and colleagues, published in Cell, demonstrated that the ketogenic diet’s anti-seizure effects in mouse models were dependent on specific changes in gut microbiota composition. Germ-free mice (lacking gut bacteria) did not respond to the ketogenic diet, but colonization with two specific bacterial species — Akkermansia muciniphila and Parabacteroides — restored the diet’s protective effects. The mechanism appears to involve microbial modulation of GABA and glutamate levels in the brain via the gut-brain axis.

This finding has significant implications. It suggests that the gut microbiome is not merely a bystander but an active mediator of dietary therapy’s effects on seizure control, and it opens the door to potential microbiome-targeted adjunctive therapies.

Metabolic Regulation and Mitochondrial Function

Beyond these specific pathways, the ketogenic diet broadly improves mitochondrial bioenergetics, reduces oxidative stress through upregulation of antioxidant pathways (including Nrf2), and stabilizes neuronal membrane potentials through effects on ATP-sensitive potassium channels. The net result is a brain that is metabolically more stable and less prone to the runaway excitation that characterizes seizure activity.

Who Benefits Most?

Drug-Resistant Epilepsy

The strongest evidence and the clearest indication for dietary therapy is drug-resistant epilepsy. International consensus guidelines, including those published by the International League Against Epilepsy (ILAE) in 2018, recommend that dietary therapy be considered after failure of two appropriately chosen ASMs. In practice, it is often considered alongside or instead of epilepsy surgery evaluation.

Pediatric Populations

The evidence base is strongest in children, and dietary therapy is most commonly initiated in pediatric patients. Certain epilepsy syndromes respond particularly well, including:

  • Glucose transporter type 1 (GLUT1) deficiency syndrome — The ketogenic diet is the treatment of choice, not merely adjunctive, because it bypasses the impaired glucose transport into the brain by providing ketone bodies as an alternative fuel.
  • Dravet syndrome — Multiple studies have reported significant seizure reduction with the ketogenic diet, often outperforming available medications.
  • Infantile spasms (West syndrome) — Particularly in cases that do not respond to ACTH or vigabatrin.
  • Lennox-Gastaut syndrome — A notoriously difficult-to-treat epilepsy syndrome in which dietary therapy has shown meaningful benefit.
  • Myoclonic-astatic epilepsy (Doose syndrome) — Strong responder to the ketogenic diet.

Adults

While the evidence base in adults is smaller, it is growing. A 2016 systematic review by Liu and colleagues, published in Seizure, identified 16 studies of ketogenic dietary therapies in adults with epilepsy and found that approximately 32% of adult patients achieved a greater than 50% seizure reduction. The modified Atkins diet has been the most commonly studied protocol in adults, owing to its greater practicality.

Adults face unique challenges with dietary therapy — independent food preparation, social eating, alcohol, and occupational demands — but the evidence supports its use for motivated patients with drug-resistant epilepsy who have adequate support.

Nutrient Considerations on Restrictive Diets

Any highly restrictive dietary protocol carries the risk of nutritional deficiency, and the ketogenic diet is no exception. Careful monitoring and supplementation are essential.

Common Deficiencies to Monitor

  • Calcium and vitamin D — The high-fat, low-dairy nature of many ketogenic protocols, combined with the effects of some ASMs on vitamin D metabolism, creates a significant risk of bone mineral density loss. Supplementation with calcium and vitamin D is standard practice.
  • Selenium — Selenium deficiency has been reported in children on the ketogenic diet and can cause cardiomyopathy in severe cases. Regular monitoring and supplementation are recommended.
  • B vitamins — Particularly folate and B12, which may be inadequate on very restrictive protocols.
  • Fiber — The severe carbohydrate restriction of the classical ketogenic diet often results in very low fiber intake, contributing to constipation — one of the most frequently reported side effects.
  • Carnitine — Some ASMs (particularly valproate) deplete carnitine, and the ketogenic diet increases carnitine demand. Supplementation is often recommended when the diet is used alongside valproate.

Kidney Stones and Lipid Profiles

Approximately 3-7% of children on the ketogenic diet develop kidney stones, likely due to chronic acidosis and hypercalciuria. Adequate hydration and, in some cases, oral potassium citrate supplementation help mitigate this risk. Lipid profiles typically worsen initially on the ketogenic diet, with elevations in total cholesterol and LDL, though these changes often stabilize or partially reverse over time. Long-term cardiovascular implications remain an area of ongoing study.

The Mediterranean Diet as a Maintenance Strategy

An emerging area of interest is whether less restrictive dietary patterns can provide some degree of seizure protection, either as a maintenance strategy after a period on a stricter protocol or as a stand-alone intervention for patients with less severe epilepsy.

The Mediterranean diet shares several features with the established epilepsy diets: it is relatively low in refined carbohydrates, rich in healthy fats (olive oil, nuts, fish), and supportive of gut microbiome diversity. While no randomized trial has directly tested the Mediterranean diet as an epilepsy intervention, its anti-inflammatory properties, favorable effects on BDNF expression, and support for gut microbiome health align with the known mechanisms of dietary seizure control.

Clinically, some epileptologists recommend a modified Mediterranean-style diet as a transition strategy for patients who have achieved seizure control on a ketogenic or MAD protocol and wish to move to a more sustainable long-term dietary pattern. The LGIT can serve as a bridge between the two, maintaining some degree of glycemic control while broadening food options.

Practical Implementation

Working with an Epileptologist and Ketogenic Diet Team

Dietary therapy for epilepsy is a medical intervention, not a lifestyle choice. It should be initiated and monitored by a team that includes, at minimum, a neurologist or epileptologist experienced in dietary therapy and a registered dietitian with ketogenic diet expertise. This team will:

  • Perform baseline laboratory studies (metabolic panel, lipid panel, urinalysis, carnitine levels, selenium, vitamin D, complete blood count)
  • Rule out contraindications (fatty acid oxidation disorders, pyruvate carboxylase deficiency, porphyria, severe liver disease)
  • Select the most appropriate dietary protocol based on the patient’s age, seizure type, medication regimen, and lifestyle factors
  • Calculate individualized meal plans
  • Monitor for side effects and nutritional deficiencies
  • Adjust ASM dosages if seizure control improves

Starting the Diet

The classical ketogenic diet was historically initiated with a 24-48 hour fast in a hospital setting, though more recent evidence suggests that a gradual, non-fasting initiation is equally effective and better tolerated (Bergqvist et al., 2005). The MAD and LGIT can typically be initiated on an outpatient basis.

Regardless of protocol, a three-month trial is generally recommended before assessing efficacy. Some patients respond within the first few weeks; others require the full three months. If a greater than 50% seizure reduction is achieved, the diet is typically continued for at least two years before a gradual weaning attempt.

Building Sustainability

Long-term adherence is the primary challenge. Strategies that help include:

  • Family-wide dietary changes — When the entire household adopts a compatible eating pattern, compliance improves substantially, particularly for children.
  • Meal preparation and batch cooking — Planning and preparing meals in advance reduces the daily burden of precise calculation.
  • Online communities and resources — Organizations such as the Charlie Foundation and Matthew’s Friends provide recipes, support, and educational materials specifically for families using dietary therapy for epilepsy.
  • School and social planning — Working with schools to ensure appropriate meals and snacks, and developing strategies for birthday parties, field trips, and other social eating situations.
  • Regular follow-up — Scheduled check-ins with the ketogenic diet team help maintain motivation, address challenges, and adjust the diet as needed.

Practical Takeaway

  1. Understand that dietary therapy for epilepsy is evidence-based medicine — The ketogenic diet has Level 1 evidence from randomized controlled trials. It is not alternative medicine; it is an established treatment option for drug-resistant epilepsy.
  2. Know your options — The classical ketogenic diet is the most studied, but the modified Atkins diet, MCT diet, and low glycemic index treatment offer less restrictive alternatives with meaningful efficacy. Discuss all options with your neurologist.
  3. Seek specialized care — Do not attempt a therapeutic ketogenic diet for epilepsy without medical supervision. Find an epilepsy center with a ketogenic diet program, or ask your neurologist for a referral to one.
  4. Commit to a genuine trial — Allow at least three months on the protocol before assessing whether it is working. Early side effects (fatigue, nausea, constipation) often resolve within the first few weeks.
  5. Monitor nutritional status — Restrictive diets require regular blood work and targeted supplementation. Calcium, vitamin D, selenium, and carnitine are commonly needed.
  6. Plan for sustainability — Work with your dietitian to develop meal plans that are compatible with your lifestyle, cultural food preferences, and family dynamics. The best diet is one you can actually maintain.
  7. Consider long-term dietary patterns — If seizure control is achieved, discuss transition strategies with your team. The low glycemic index treatment or a modified Mediterranean-style diet may serve as more sustainable long-term approaches.

Frequently Asked Questions

Can the ketogenic diet replace anti-seizure medications?

In some cases, yes. A subset of patients — particularly those with certain genetic epilepsy syndromes such as GLUT1 deficiency — may achieve complete seizure control on the ketogenic diet alone. For many others, the diet allows a reduction in the number or dosage of medications, which can significantly improve quality of life by reducing medication side effects. However, any medication changes should be made gradually and only under the direct supervision of a neurologist. Never stop or reduce anti-seizure medications on your own.

Is the ketogenic diet safe for children long-term?

The ketogenic diet has been used in children for over a century, and long-term follow-up studies generally confirm its safety when properly supervised. The primary concerns are growth retardation (usually modest and partially recoverable after diet discontinuation), bone mineral density reduction, kidney stones, and dyslipidemia. These risks are real but manageable with appropriate monitoring and supplementation. For children with severe, drug-resistant epilepsy, the risks of uncontrolled seizures — including injury, cognitive decline, and sudden unexpected death in epilepsy (SUDEP) — typically outweigh the risks of the diet.

How quickly does the ketogenic diet work for seizures?

Response time varies. Some patients experience a reduction in seizure frequency within the first one to two weeks, while others require up to three months. A small number of patients experience an initial increase in seizures during the adaptation period. International guidelines recommend a minimum three-month trial before concluding that the diet is ineffective. If there is no improvement at three months despite confirmed ketosis, the diet is unlikely to be beneficial for that individual.

Can adults use dietary therapy for epilepsy?

Yes. While the evidence base is larger in children, studies of the modified Atkins diet and classical ketogenic diet in adults with drug-resistant epilepsy have shown meaningful seizure reduction in approximately one-third of patients. The MAD is generally the preferred protocol for adults because of its greater flexibility. Adults should work with an epilepsy center experienced in dietary therapy and should be aware that adherence may be more challenging due to social and occupational factors.

Does the ketogenic diet help with the cognitive effects of epilepsy?

This is an active area of research. Some studies have reported improvements in attention, alertness, and cognitive function in children on the ketogenic diet, independent of seizure reduction. Whether these benefits reflect direct neuroprotective effects of ketone bodies (via BDNF upregulation, improved mitochondrial function, and reduced neuroinflammation) or are secondary to reduced seizure burden and lower medication doses remains unclear. The available evidence is encouraging but not yet definitive.

Sources

  • Bergqvist, A. G. C., et al. (2005). Fasting versus gradual initiation of the ketogenic diet: A prospective, randomized clinical trial of efficacy. Epilepsia, 46(11), 1810-1819.
  • Dahlin, M., et al. (2005). The ketogenic diet influences the levels of excitatory and inhibitory amino acids in the CSF in children with refractory epilepsy. Epilepsy Research, 64(3), 115-125.
  • Huttenlocher, P. R., et al. (1971). Medium-chain triglycerides as a therapy for intractable childhood epilepsy. Neurology, 21(11), 1097-1103.
  • Juge, N., et al. (2010). Metabolic control of vesicular glutamate transport and release. Neuron, 68(1), 99-112.
  • Kossoff, E. H., et al. (2003). Efficacy of the Atkins diet as therapy for intractable epilepsy. Neurology, 61(12), 1789-1791.
  • Kossoff, E. H., et al. (2008). Prospective study of the modified Atkins diet in combination with a ketogenic liquid supplement during the initial month. Journal of Child Neurology, 23(10), 1229-1232.
  • Kossoff, E. H., & Dorward, J. L. (2008). The modified Atkins diet. Epilepsia, 49(Suppl 8), 37-41.
  • Liu, H., et al. (2016). Ketogenic diet for treatment of intractable epilepsy in adults: A meta-analysis of observational studies. Seizure, 36, 43-47.
  • Masino, S. A., et al. (2011). A ketogenic diet suppresses seizures in mice through adenosine A1 receptors. Journal of Clinical Investigation, 121(7), 2679-2683.
  • Muzykewicz, D. A., et al. (2009). Efficacy, safety, and tolerability of the low glycemic index treatment in pediatric epilepsy. Epilepsia, 50(5), 1118-1126.
  • Neal, E. G., et al. (2008). The ketogenic diet for the treatment of childhood epilepsy: A randomised controlled trial. The Lancet Neurology, 7(6), 500-506.
  • Olson, C. A., et al. (2018). The gut microbiota mediates the anti-seizure effects of the ketogenic diet. Cell, 173(7), 1728-1741.
  • Sharma, S., et al. (2013). Use of the modified Atkins diet for treatment of refractory childhood epilepsy: A randomized controlled trial. Epilepsia, 54(3), 481-486.
  • Thiele, E. A. (2003). Assessing the efficacy of antiepileptic treatments: The ketogenic diet. Epilepsia, 44(Suppl 7), 26-29.
  • Vizuete, A. F., et al. (2013). Brain changes in BDNF and S100B induced by ketogenic diets in Wistar rats. Life Sciences, 92(17-19), 923-928.

This article is for educational purposes only and does not constitute medical advice. Dietary therapy for epilepsy is a medical intervention that requires supervision by a neurologist and a registered dietitian experienced in ketogenic dietary therapies. Never modify anti-seizure medications without direct guidance from your treating physician.