The Ketogenic Diet and Drug-Resistant Epilepsy in Children — What Did the First Randomized Trial Show?

Drug-resistant epilepsy (also called treatment-resistant epilepsy) is a situation in which seizures don't stop despite correctly chosen and appropriately dosed use of at least two antiepileptic drugs, taken alone or in combination. This isn't rare — it's estimated to affect as many as 20-30% of all people with epilepsy, and among children the proportion can be even higher with certain epilepsy syndromes that begin in early childhood.

The consequences of uncontrolled, frequent seizures in a child go far beyond the seizure itself. Daily or near-daily seizures carry the risk of injury, disrupted cognitive development, learning difficulties, limitations on the whole family's daily functioning, and — in some cases — the risk of sudden unexpected death in epilepsy (SUDEP). When successive antiepileptic drugs fail, neurologists turn to third-line options, among which the ketogenic diet has long had a place.

This is exactly where the ketogenic diet's history diverges from its much broader association today with weight loss — a divergence we discuss more fully in our knowledge-base entry on the ketogenic diet. It's worth separating these two contexts from the very start, since the rest of this article deals exclusively with the medical application, strictly supervised by specialists.

The ketogenic diet as an epilepsy therapy is not a new trend borrowed from the fitness world — it's the opposite. It was developed in the 1920s by physicians at the Mayo Clinic as a way to recreate the metabolic effects of fasting, which had been observed since antiquity to ease seizures. For several decades, before modern antiepileptic drugs appeared, it was one of the few available therapeutic options for severe childhood epilepsy.

With the development of antiepileptic pharmacotherapy in the 1940s and later, the diet's clinical popularity declined — medications were more convenient and better tolerated by most patients. It wasn't until the 1990s that the ketogenic diet experienced a revival as a recognized third-line therapy specifically for that group of children for whom medications aren't enough. Only much later, from the 2000s onward, did this same eating pattern begin gaining an entirely different, much broader popularity as a weight-loss method in adults — the population, the goal, and how the intervention is conducted are fundamentally different in the two cases.

For decades, evidence for the effectiveness of the ketogenic diet in drug-resistant epilepsy came mainly from clinical observations and studies without a control group — clinically compelling, but methodologically weaker than a randomized controlled trial (RCT), regarded as the gold standard in medicine. It wasn't until 2008 that the prestigious journal Lancet Neurology published the results of the first truly randomized controlled trial testing this intervention in a scientifically rigorous way.

It's worth noting the figure of 136.9% in the control group — this isn't an error. It means that in children not receiving the diet, seizure frequency during the observation period increased on average relative to baseline, which is typical of the natural course of drug-resistant epilepsy without a change in treatment. It's only when this increase is set against the decrease in the diet group (down to 62% of baseline) that the full picture of the difference between groups emerges — hence the relative reduction of around 75%.

38% of children on the diet achieved a seizure reduction of more than half — a clinically meaningful result for the family of a child whose previous pharmacological treatment had failed, but it also means that most children (62%) didn't experience that level of improvement within those 3 months. The ketogenic diet in drug-resistant epilepsy isn't a miracle cure that works for everyone — it's one of the therapeutic options with moderate-to-good effectiveness in a population where other methods have already failed.

The proportion of children with a greater than 90% seizure reduction (7% in the diet group) is small, but for those specific children the effect can be transformative — for the family of a child who had daily seizures before starting the diet, a reduction of more than 90% means a qualitatively different life. This illustrates well why neurologists take this intervention seriously despite it not helping every patient.

The Neal et al. study also had its limitations. It wasn't blinded — neither parents nor the physicians assessing seizures could avoid knowing which child was on the diet, which is practically unavoidable with a dietary intervention and could theoretically have influenced parents' subjective reporting of seizure frequency. The observation period lasted only 3 months, so the study doesn't directly answer questions about the diet's effectiveness and safety over a span of several years, although data from later observational studies and clinical practice suggest that some children continue the therapy for much longer with sustained benefit. Complete data was obtained from 103 of the 145 randomized children — some participants dropped out of the study, which is typical for restrictive dietary interventions, but could have slightly affected the estimated effect size.

Despite these limitations, this study has, since its 2008 publication, remained one of the most frequently cited pieces of evidence for the effectiveness of the ketogenic diet in drug-resistant childhood epilepsy, and it formed the basis for subsequent clinical guidelines and meta-analyses that in later years confirmed a similar direction of effect in other populations and diet variants (including the less restrictive modified Atkins diet).

The ketogenic diet for drug-resistant epilepsy in children is a rare example of a dietary intervention with genuinely strong, randomized scientific evidence behind it — evidence rooted in a history stretching back almost a hundred years, long before this same eating pattern gained an entirely different popularity as a weight-loss method. This is exactly why it's worth distinguishing these two contexts every time the topic comes up — confusing them leads either to underestimating a serious medical therapy, or to dangerously viewing a restrictive pediatric diet as something that can be implemented independently without specialist supervision.