TABLE
Special Meeting: Controversies in Epilepsy
The Ketogenic Diet
Taken from "Epilepsy Update" Issue #6 1996
The following presentations are highlights from a session organized by Gregory Homes, M.D. (Harvard Medical School, Boston). Included are presentations by Edwin Trevathan, M.D., MPH (University of Kentucky), Marcio Vasconcelos, M.D. (Childrens National Medical Center, Washington, DC), Carl Stafstrom, M.D., Ph.D. (Tufts University School of Medicine and New England Medical Center, Boston), and James Wheless, M.D. (University of Texas Medical School, Houston).
As most people who work in the field of epilepsy treatment are aware, there has been a great deal of interest recently in the use of the ketogenic diet to reduce the severity of seizures in patients with epilepsy, particularly younger patients with very severe disease. The controversy surrounds the reported efficacy of the diet, its unknown mechanism of action, and the difficulty of maintaining it - in the context that its value has not been proven in large-scale controlled clinical trials. Also, although administration of the diet requires the full participation of a dietitian or nutritionist, many dietitians are not trained to provide this high-fat and very-low- carbohydrate diet.
Description: The ketogenic diet gets its name because the high fat content of the diet results in conversion of fat-to-ketones that are utilized as an energy source in place of glucose. If carbohydrates (which are composed of sugars) are eliminated from the diet, and a diet very high in fat is substituted, the body has no dietary sources of glucose. As a result, ketones are made from the available sources and these are used as fuel instead. It is necessary to be scrupulous in the restriction of carbohydrates because even a very small amount of sugar can cause the body to shift to glucose production and use, which it prefers to ketones. For example, this restriction is such that children on the diet have to be careful to take sugarless daily multivitamins.
Background: Ketosis, or the physiologic state in which there are elevated levels of ketones in the blood, usually occurs in people in the fasting state. A state similar to fasting can be induced by a very high fat diet. Hence the ketogenic diet. A beneficial effect of ketosis on seizure activity was observed during the last century, and systematic efforts to use this effect to treat epilepsy date back to the 1920s before many of the current antiepileptic medications were available.
In "The Ketogenic Diet - Does it Work in Children?" Edwin Trevathan, M.D., MPH, who is director of the Comprehensive Epilepsy Program at the University of Kentucky, presented a review of the history of the clinical use of the ketogenic diet. He defined it as a "rigidly calculated high-fat diet, with adequate protein and carbohydrate for growth, designed to maintain ketosis". Clinical reports of the ketogenic diet may be classified as pre-1980 and post-1980. Prior to 1980, the studies were mostly retrospective or anecdotal, had mixed patient populations, did not include carbamazepine and valproate (which were unavailable for antiepileptic treatment), and used the "classic" ketogenic diet. Studies subsequent to 1980 also were mostly retrospective and small in scale, but generally used the medium-chain triglyceride (MCT) ketogenic diet and usually involved patients with severe intractable epilepsy. The MCT diet was designed to be more palatable, with 60% of total calories coming from MCT oil; it allows more protein and carbohydrate than the classic ketogenic diet. However, recent experience suggests that the classic ketogenic diet may be better tolerated than the MCT diet. Although it is difficult to interpret the results because of many of the factors mentioned above, clinical series from the 1960s and 1970s generally reported good results with the ketogenic diet. In a 1977 report of 1,000 patients, 792 of whom had intractable disease, seizures were controlled in 54% and marked improvement was seen in 26%. Nevertheless, Dr. Trevathan emphasized, it is difficult to interpret these results in a modern context when talking about a different patient population with different additional treatment options available. Unfortunately, one of the major drawbacks has been the difficulty of remaining on this rigidly calculated diet. (A child on this diet cannot even accept a cookie or candy from another child or have a special treat at a celebration or holiday gathering). This is unfortunate because the probability of sustaining an improvement in seizure activity appears to be correlated with remaining on the diet; this also may be interpreted to mean that the ability to remain on the diet influences the ability to sustain control of seizures.
While it is clear that some children with refractory epilepsy may be helped by the
ketogenic diet, the absence of good long-term data on risk and efficacy make its
use appropriate only in intractable patients. Further clinical studies are needed to
define the proper role of the ketogenic diet in clinical epilepsy practice.
Marcio Vasconcelos, M.D., who is a fellow at the Children's National Medical Center in Washington, DC, presented the results of a clinical trial of the ketogenic diet in his talk titled "Efficacy and Toxicity of the Ketogenic Diet in Children With Intractable Epilepsy". He identified three objectives: 1) to assess retrospectively the efficacy of the ketogenic diet in patients between 1992 and 1995, 2) to identify which factors influence its efficacy, and 3) to study the incidence and severity of its toxicity. The diet was started during hospitalization. Of the 29 patients, 12 received a ketogenic diet in a 3:1 ratio, meaning that the diet was 3 parts fat to 1 part protein/carbohydrate, 14 received a diet with a 4:1 ratio, 2 patients received a 5:1 ratio, and one patient received an MCT diet. Six patients discontinued the diet during the first week, 16 discontinued after the first week, and 7 remained on it for periods up to 12 months. Of the 16 who discontinued after the first week, 6 did so because of no improvement, 5 because of seizure recurrence, and 2 because of refusal to keep the diet. Results were evaluated in 23 of the patients (ie, those who did not discontinue during the first week). Of these 23, 5 (22%) became seizure-free, 7 (31%) had a 50% or greater reduction in seizures, 1 had a less-than-50% improvement, and 10 (43%) had no change. Of the 23, 11 displayed better mood or an increased level of activity, 8 had no change, and 2 were worse (irritable or aggressive).
Based on these results, Dr. Vasconcelos was able to reach the following conclusions:
Calling the ketogenic diet a "therapy in search of an explanation", Carl E. Stafstrom, M.D., Ph.D., associate professor at Tufts University School of Medicine and the New England Medical Center in Boston described research in animal models to investigate the mechanisms of action through which the ketogenic diet confers its apparent beneficial effect. Animal models are useful because it is possible to control and evaluate precisely factors such as components of the diet, types of seizures, and various bio-chemical and neurologic parameters. Although cautioning the audience that a wide variety of experimental designs makes broad conclusions difficult, Dr. Stafstrom indicated that we have learned the following from animal models: 1) the ketogenic diet appears to afford protection against acute seizures in some models, 2) the mechanism of anticonvulsant action may be related to the ketosis itself, rather than to the associated acidosis or other metabolic effects, and 3) several features of experimental responses parallel those observed clinically. For example, the diet seems to be more effective in younger animals just as it seems to be more effective in children than in adults. Also, although the onset of action is gradual, the reversal of the effect occurs rapidly. We see this clinically when it may take several days after beginning the diet for the effect to build up, but, if ketosis is interrupted by the consumption of carbohydrates, the anticonvulsant effect can reverse within hours. This is why it is so important to maintain the diet strictly, and why it is so easy to undo its effects: weeks of hard work can be undone if the child eats a couple of cookies or candies because they contain enough glucose to switch the body over to using glucose for fuel rather than the ketone bodies. (Remember that the body only uses the ketones when sugar is unavailable; as soon as the slightest amount of sugar is made available the body immediately converts it to glucose to use for fuel).
Dr. Stafstrom described several experiments in which rats given ketogenic diets
were compared with rats on normal diets as various parameters of cognition and
behavior were tested. In general, the rats on the ketogenic diet did better in several
tests, such as the "water maze", which measures the ability to learn and remember
the location of food on a platform in a pool of water through which the rat has to
swim to reach the food. "I am not willing to say that the ketogenic diet makes rats
smarter, but at least it's clear that it doesn't make them any dumber", said Dr.
Stafstrom, and this observed effect correlates with observations in children with
severe epilepsy on the ketogenic diet who experience improvements in mood,
behavior, and cognition. Unfortunately, this improvement carries with it an
increase in activity, possibly hyperactivity, in terms of awareness and exploration of
the surroundings; this may correlate with observations of increased irritability in
some children on the diet.
In his presentation titled "The Ketogenic Diet - Is It Worth The Trouble?" James Wheless, M.D., associate professor at the Texas Comprehensive Epilepsy Program, the University of Texas Medical School in Houston, described the complex approach he uses at this medical center to implement the ketogenic diet. He emphasized that two factors are extremely important: the integration of a team approach and education of the parents and the child. The team should include the epileptologist's nursing staff and other office personnel (for training and to answer questions) and a dietitian trained in the implementation of this diet. The Table shows the John Hopkins Hospital multicenter protocol used by Dr. Wheless and his group for administration of the ketogenic diet.
Dr. Wheless commented on several of the practical issues involved in implementing
the ketogenic diet. "Nutritional labeling of food has made all of this much easier",
was one of his points. Parents undergo an extensive training procedure with a
specially trained dietitian on how to prepare ketogenic diet meals and how to shop
in supermarkets. The team approach is very important, with the dietitian perhaps
being the most important member. One problem is that most dietitians have not
been trained to handle this sort of high-fat, low-carbohydrate diet. In terms of
education of the parents, Dr. Wheless said he has found it helpful to refer to this
whole process as "Ketosis therapy" because if it's thought of as a medical therapy
people have appropriate expectations about its outcome, whereas if it's thought of
as a diet people expect it to be foolproof and without adverse effects. "This is a
medical therapy, and like any medical therapy, it can have adverse effects, but we
try to balance them against the potential benefit," said Dr. Wheless.
TABLE
Many questions remain regarding the use of this diet. First, although partial fluid restriction is still part of most protocols, there does not seem to be a current rationale for its inclusion; in general, its use is based on long-standing tradition. Because very little is understood about the mechanism of the diet, it is hard to evaluate the role, if any, played by the level of fluid intake. Dr. Wheless said he hopes answers about this will come from ongoing and future clinical trials of the diet.
In addition to the overwhelming questions about its mechanism of action, there are many questions remaining about the practical application of the ketogenic diet:
These questions highlight the need for additional research. Problems with current
implementation include dietitians' unfamiliarity with the diet, parents' guilt when
these children's seizures do not improve while on the diet, parents' expectations for
cognitive/behavioral improvement, and management of side effects. This places us
in a unique and uncomfortable position, suggesting a treatment regimen that
appears to be effective in patients with severe disease who do not respond to
conventional therapy, but one that is difficult to implement and to explain. Future
clinical studies should clear up many of the urgent questions concerning the
ketogenic diet, and more clinical experience should enlighten us on its practical
daily use in our patients with severe and intractable epilepsy.
Return to COPE