Why ketogenic diets work for epilepsy, and why that matters beyond seizures

University of Colorado Anschutz

Ketogenic diets have been used to treat epilepsy since the 1920s. For a century, clinicians have watched the approach work in patients who do not respond to medication, often dramatically reducing seizure frequency. But the scientific understanding of why it works has lagged behind the clinical observation.

A new review published in The Lancet Neurology in March 2026, led by Anna Figueroa of the University of Colorado Anschutz Skaggs School of Pharmacy, synthesizes five years of research to provide the most integrated explanation to date of the mechanisms behind ketogenic diet therapy, and to identify the critical gaps that are holding the field back.

Rewiring the brain's fuel supply

The core mechanism is a metabolic shift. A standard diet provides the brain with glucose as its primary energy source. Ketogenic diets severely restrict carbohydrates, forcing the body to produce ketones from fat as an alternative fuel. Ketones provide a steadier, more efficient energy source for neurons than glucose.

That shift has cascading effects. Overactive neurons, the kind that trigger seizures, become more stable when running on ketone-based metabolism. The brain's overall energy regulation improves. But the review makes clear that the diet does far more than just change the fuel: it reduces neuroinflammation, protects neurons from damage, and strengthens cellular energy systems in ways that most current antiseizure medications do not.

Figueroa and her co-authors, Charuta Joshi and Manisha Patel, emphasize that these are mechanistically distinct benefits. A drug that prevents seizures by blocking sodium channels does not simultaneously reduce brain inflammation or improve mitochondrial function. The ketogenic diet appears to do all of these things, which helps explain why it works in patients for whom conventional drugs fail.

Children studied, adults neglected

Most ketogenic diet research has focused on children. While some pediatric studies have compared different ketogenic diet variants and evaluated them against standard epilepsy care, even these studies are relatively limited in scale. In adults, the evidence gap is far wider. Only one randomized controlled trial comparing a ketogenic diet to standard epilepsy care in adults has been conducted in the past five years.

The review identifies this as an urgent need. Adults constitute a large proportion of the epilepsy population, and their biology introduces additional complexities. Long-term antiseizure medications can cause liver changes that affect how well adults tolerate or metabolize ketogenic diets. This may explain why early initiation in childhood appears most effective and why the authors emphasize starting the diet early when possible.

The compliance problem

Ketogenic diets are demanding. They require strict macronutrient tracking, significant dietary restriction, and ongoing medical supervision. Adherence drops over time, particularly in adolescents and adults who face social and practical pressures around food. The review acknowledges this as a fundamental barrier to wider adoption.

One of the most promising directions the review highlights is the development of diet-mimicking therapies: drugs or supplements that replicate the metabolic effects of a ketogenic diet without requiring patients to follow the diet itself. If the key mechanisms are the shift to ketone metabolism and the resulting anti-inflammatory and neuroprotective effects, then pharmacological approaches that trigger those same pathways could offer the benefits without the dietary burden.

This is not yet a reality. No diet-mimicking drug has been approved for epilepsy. But the improved mechanistic understanding described in the review provides a clearer target for drug development than existed even five years ago.

Beyond epilepsy

The mechanisms the review describes are not specific to seizure disorders. Neuroinflammation, mitochondrial dysfunction, and energy dysregulation play roles in a range of neurological conditions, including Alzheimer's disease, Parkinson's disease, and traumatic brain injury. Emerging evidence suggests ketogenic approaches may benefit some patients with these conditions, though the evidence base is far less developed than for epilepsy.

The review does not make specific claims about these broader applications but notes that metabolism-based therapies may have wider therapeutic value. If ketone-based metabolism genuinely offers neuroprotection, the implications extend well beyond seizure control.

A review of scattered evidence, not new data

This is a narrative review, not a meta-analysis or original study. It synthesizes and interprets existing research rather than generating new evidence. The authors selected and interpreted studies based on their clinical and scientific judgment, which introduces a degree of subjectivity that a systematic review would mitigate.

The mechanistic explanations draw heavily on animal models and cell culture studies. Whether these mechanisms operate identically in human brains, with their far greater complexity, is not fully established.

The review also highlights how few large, well-controlled clinical trials exist for ketogenic diets in epilepsy. The evidence base, while growing, remains thinner than what would be expected for a therapy that has been in use for a century. The authors frame this as both a limitation of their review and a call to action for the field.