The ketogenic diet reduces seizure frequency by more than 50% in half of the patients who try it and by more than 90% in a third of patients. Three-quarters of children who respond do so within two weeks, though experts recommend a trial of at least three months before assuming it has been ineffective. Children with refractory epilepsy are more likely to benefit from the ketogenic diet than from trying another anticonvulsant drug. Some evidence indicates that adolescents and adults may also benefit from the diet.
First reported in 2003, the idea of using a form of the Atkins diet to treat epilepsy came about after parents and patients discovered that the induction phase of the Atkins diet controlled seizures. The ketogenic diet team at Johns Hopkins Hospital modified the Atkins diet by removing the aim of achieving weight loss, extending the induction phase indefinitely, and specifically encouraging fat consumption. Compared with the ketogenic diet, the modified Atkins diet (MAD) places no limit on calories or protein, and the lower overall ketogenic ratio (about 1:1) does not need to be consistently maintained by all meals of the day. The MAD does not begin with a fast or with a stay in hospital and requires less dietitian support than the ketogenic diet. Carbohydrates are initially limited to 10 g per day in children or 20 g per day in adults, and are increased to 20–30 g per day after a month or so, depending on the effect on seizure control or tolerance of the restrictions. Like the ketogenic diet, the MAD requires vitamin and mineral supplements and children are carefully and periodically monitored at outpatient clinics.