The Ketogenic diet is not by all means a new treatment. Claims for a dietary treatment of epilepsy are very old, (even before any anticonvulsants were available). Earlier attempts to a diet for epilepsy include: salt restriction, protein restriction, acid-ash diets, water restriction (to provoke dehydration), etc. There is even a reference to fasting as a "cure" for seizures in the Bible . 
The first scientific report on fasting as of value in the treatment of epilepsy was in France by Guelpa and Marie (1) in 1910 who reported that seizures stopped during absolute fasting. Later on, other investigators observed cessation of seizures and improvement in mental activity during starvation . These results prompted the reproduction of ketosis and acidosis by means of a high fat -low carbohydrate diet in 1921 by Wilder (2) at the Mayo Clinic who was trying to prolong the state of ketosis in diabetics. Dr John M. Freeman (3), reports that almost concurrently at the Johns Hopkins Department of Pediatrics, Howland and Gamble observed that "prayer and a water diet which involved starvation for three to four weeks" improved the seizures of the nephew of a professor of pediatrics. They decided to use the diet because prayer alone was ineffective. And so the ketogenic diet was born. Other investigators like Drs Lennox and Cobb In Harvard University started investigating the ketogenic diet concurrently.

Dr. Wheless (8) mentions that the co-discoverer of phenytoin, Dr. Houston Merritt lamented that the introduction of phenytoin stopped these vigorous ongoing investigations to study the mechanisms of the ketogenic diet, which in turn could have shed light into the neurochemistry of epilepsy.
The most enthusiastic proponent of the diet in those days, Dr Livingston (4,5). studied almost one thousand patients after extensive use of the Ketogenic Diet and reported, excellent seizure control in 1954 and 1958.

The Johns Hopkins version of the diet was found to be best tolerated in a study that compared it with the use of medium-chain triglyceride (MCT) oil in 1989 (6).
By the 1960's with the introduction of other anticonvulsants, the ketogenic diet became gradually more of a curiosity. The general feeling amongst neurologists was that it was ineffective. I remember trying it in several difficult to control patients, with poor results. In retrospect I can see that the diet was not strictly applied and that we were mostly using it as a “last resort” in intractable patients of all sorts without rational selection.

By the end of the 80's and beginning of the 90's the interest for the diet was revived by the studies of Dr Freeman who found by 1992 that the diet was followed by complete seizure control in 30 percent of children with uncontrollable seizures, and that an additional 38 percent showed marked improvement (9). 
One of the children treated successfully with the diet by the Johns Hopkins team was Charlie Abrahams. His parents in response have created the Charlie Foundation which has given widespread national publicity to the diet, in part by making available a free video tape (7).

Presently a multi-center collaborative study is being conducted by 11 centers, to clarify the controversy surrounding the reported efficacy of the diet. The results are expected to refine the use of the ketogenic diet (8).

How Does it work?

The name Ketogenic derives from the conversion of fat to ketones which is induced by the elimination of carbohydrates from the diet and substitution for high amounts of fat. In abscence of sugars as a source of energy, the body burns fat instead, and breaks it into ketones. The elevation of these substances is known as Ketosis.
The anticonvulsant effect of the ketogenic diet has been also observed in experimental animals. Several hypotesis exist about the mechanism by which the diet produces its beneficial effects, and these include: a reduction of extracellular fluid, a negative balance of sodium and potassium, a rise in plasma lipids, a compensated acidosis, changes in the lipids of the neuronal membrane as well as changes in the production of neurotransmitters. Nevertheless the studies suggest that the diet’s protection against seizures is related to the ketosis itseself, rather than other metabolic effects.

Plasma elevations of very long chain fatty acids are very common in patiens that are on the diet (Theda etc.). This is an important factor that could lead to confusion since these acids are typically elevated in patients with peroxisomal dysfunction. The diet also increases the plasma concentrations of medium chain fatty acids, (as does valproic acid). This could theoretically increase the possibility of a serious Reye-like syndrome or acute pancreatitis. It is therefore important to avoid aspirin in children taking valproic acid or on the diet. Also because of this, valproic acid invalidates the use of dipstick to monitor ketosis, which can otherwise be done by measurement of betahydroxybutirate in serum.

It is also interesting that the diet appears to be more effective in younger animals just as it is more effective in children. Also in experimental animals the reversal of ketosis by ingesting carbohydrates is followed by a reversal of anticonvulsant effects within hours.

Who needs it?

The diet is not used in children under one year of age due to difficulty in maintaining ketosis and normal blood glucose levels. It is most useful in children under 6 years of age with atypical abscence, myoclonic and atonic seizures, as in the Lennox Gastaut syndrome. This diet is well tolerated and effective in children up to 15 years old.

Implementation.

The proportion of Fats to carbohydrates is called the ratio. The usual starting ratio is 4:1. Four parts of fat for each part of carbohydrates and protein combined.
During the initial period of fasting the ketosis is established, the diet then maintains it. The body is normally capable of storing only a 24 hour supply of glucose, after which time it becomes depleted therefore this is normally the fasting time that is needed. The ketones that are produced by the breakdown of fats have a sedative effect and also interestingly an appetite supressing effect. At this time traditionally the fluids are restricted to provoke a partial dehydration, but the rationale of this procedure has been questioned (Wheless). In any case the diet in contrast with the anticonvulsant medication has an alerting effect.

The diet is tried for no less than one month before it is regarded as ineffective. If it is effective, the same ratio is continued until 12 months later when a decrease in ratio can be contemplated according to the clinical circumstances.

Results.

Although the results of the published studies are difficul to interpret, they frequently indicate seizure control in over 50% of cases. There is nevertheless a need for further clinical studies to define de role of the diet in the treatment of epilepsy. One of the most trying factors involved is the difficulty to remain on this demanding regime. In a recent study (Marcio Vasaconcelos, MD) the diet was effective in 12 of 23 patients (53%), mostly patients with generalized tonic clonic, akinetic, and myoclonic seizures. Partial seizures being less responsive. Half of the patients showed improved mood and level of activity. One patient developed severe acidosis.

Conclusion.

In summanry as old as the ketogenic diet may be we know very little about it, but the preliminary data suggests that it has a definite place in the treatment of epilepsy, and it may be perhaps understated because it lacks a big Pharmacological Company behind it investing millions of dollars in its promotion.

Sources of information.

The Epilepsy Diet Treatment. An introduction to the Ketogenic Diet.
John M. Freeman, M.D., Millicent T. Kelly, R.D., L.D., Fennifer B. Freeman.
Copies of this book may be obtained from Demos Publications:
386 Park Avenue South, New York, NY 10016.

Keto Klub Newsletter.
61557 Miami Meadows Court, South Bend, IN 46614.

Introduction to the Ketogenic Diet: A Tratment for Pediatric Epilepsy.
The Charlie Foundation To Help Cure Pediatric Epilepsy, 501 10th Street, Santa Monica, CA 90402.

Internet Ketogenic Diet World Wide Web page at Packard Children's Hospital @ Stanford University. This Web Site was created to facilitate communication between health care providers
who are interested in the ketogenic diet as a method of treatment of epilepsy.
http://www-leland.stanford.edu/group/ketodiet/

The Keto Chat Line at America On Line. First tuesday night of every month. Exchange of ideas between parents of children that are on the ketogenic diet.

References.

1. Livingston, S., Comprehensive Management of Epilepsy in Infancy, Childhood and Adolescence. Charles C. Thomas. 1972.
2. Wilder, R.M. The effect of ketonuria on the course of epilepsy. Mayo Clinic Bull, 2:307, 1921.
3. Freeman, J.M., Kelly, M.T., Freeman, J.B.: The Epilepsy Diet Treatment.: an introduction to the ketogenic diet. Demos, 1994.
4. Livingston, S., The Diagnosis and Treatment of Convulsive Disorders in Children. Springfield, Thomas, 1954.
5. Livingston, S., Convulsive disorders in Infants and Children. In Levine, S.Z. (Ed.): Advances in Pediatrics. Chicago, Year Book, 1958.

6. Schwartz, RM, Boyes S, Aynsley-Green A.: Metabolic effects of three ketogenic diets in the tratment of severe epilepsy. Dev Med Child Neurol 1989;31:152-60
7. An Introduction to the Ketogenic Diet-A treatment for Pediatric Epilepsy. Video Tape. Charlie Foundation to Help Cure Pediatric Epilepsy, 501 10th Street, Santa Monica, CA 90402. 1994.
8. Wheless, JW. The Ketogenic Diet: Fa(c)t or Fiction. J Child Neurol. 10:6 419. Nov 1995.
9. Kinsman SL. Vining EP. Quaskey SA. Mellits D. Freeman JM. Efficacy of the ketogenic diet for intractable seizure disorders: review of 58 cases. Epilepsia.33(6):1132-6, 1992 Nov-Dec.