After years of neglect, this method of managing hard-to-control seizures is being used more often by pediatric neurologists.
|Jump to:||Choose article section... History of the diet Nature of the diet Appropriate use of the diet Does the diet work? Outcomes of the ketogenic diet at Johns Hopkins Medical Institutions How long do children stay on the diet? Concerns about the diet Care of the ill child the pediatrician's role Give no carbohydrates Side effects of the diet How do you know if the diet is helping? Worth a try|
After years of neglect, this method of managing hard-to-control seizures is being used more often by pediatric neurologists. A working knowledge of the diet will help you support patients and parents and avoid treatment mistakes that can sabotage the diet's therapeutic effects.
The ketogenic diet is a high-fat, adequate-protein, very-low-carbohydrate diet that pediatric neurologists are using increasingly to treat children with difficult-to-control seizures. As use of the diet grows, pediatricians and family practitioners are likely to see more patients who are on it at office visits or in the emergency room. They should be familiar with the diet, its side effects, and its role in seizure management. And every physician who sees a patient on the diet should be aware of the cardinal rule of the ketogenic diet: Give NO CARBOHYDRATES.
In the early 1920s, the only effective medications available for seizures were phenobarbital and bromides. An osteopath, Hugh Conklin, believing that seizures originated in the intestines, "rested" the intestines of children with seizures by having the child fast for 10 to 20 days. Fasting often resulted in remarkable and long-lasting seizure control. The dramatic results were replicated by others, and a high-fat, adequate-protein, limited-carbohydrate diet was devised to simulate the effects of starvation.1 In the absence of sufficient carbohydrates, fats are incompletely metabolized, leaving residual ketone bodies. This ketosis is believed to produce the anticonvulsant effects of the ketogenic diet by mechanisms that remain unknown.2
The ketogenic diet was widely used during the 1930suntil the discovery of phenytoin in 1938. With the subsequent development of other anticonvulsant medications, the diet was prescribed less often. Because physicians no longer ordered the diet, dietitians were no longer trained in its use, and it was less successful when it was used. The diet was revived in the early 1990s and is once again recognized as an effective therapy for children whose seizures are difficult to control.36
The ketogenic diet is an alternative medical therapy for children with difficult-to-control seizures. It must be undertaken only under medical supervision and in collaboration with a well-trained dietitian who knows how to monitor and adjust the diet.
The diet is carefully calculated to be nutritionally adequate for the individual child. In it, 90% of the calculated calories are in the form of fat, with no less than 1 g/kg of protein and minimal carbohydrates.7 The standard ratio of fat calories to carbohydrate and protein is 4:1. Younger children and adolescents may be placed on a 3:1 ratio to allow more protein (1 to 1.5 g/kg) for adequate growth. Caloric intake is calculated at 75% of the usual daily requirement. Fluids are generally restricted to 80% of the usual amount. The diet must be supplemented with vitamins and minerals. Infants and younger children may be given more protein and, occasionally, even less carbohydrate than older children are.
The dietitian carefully and repeatedly adjusts the diet by phone, e-mail, or outpatient follow-up to achieve and maintain the child's ideal body weight for height, maximize ketosis, and achieve optimal seizure control.7 Children typically return for follow-up at three months, six months, and one year after starting the diet to assure adequate growth and nutrition and to identify and correct problems.
With rare exceptions, the ketogenic diet should not be the first treatment for a child with seizures. At the Johns Hopkins Medical Institutions, we have largely restricted the use of the diet to children who have difficult-to-control seizures of any typethat is, seizures that are not controlled by the first two anticonvulsant medications tried. Most children started on the diet at Johns Hopkins have more than two seizures per week (often, many each day) and have failed at least two anticonvulsants. These restrictions are, admittedly, arbitrary. We believe that taking pills is far easier than maintaining the diet, and because anticonvulsants easily control most seizures, they should be tried first. Atonic seizures, seizures associated with the Lennox-Gastaut syndrome, and infantile spasms are often refractory to medication, and children with these seizure types may be candidates for early use of the diet.
We are often asked to prescribe the diet "because it is more natural" than anticonvulsant medication. The diet is not more natural than medication; it is only an alternative medical therapy. It must be carefully calculated, adjusted to the patient's needs, and adhered to rigorously. We have observed that eating three or four extra macadamia nuts, for example, or even using a sunscreen lotion containing sorbitol just once can make the difference between seizure control and frequent seizures. The diet should be used ONLY under medical and nutritional supervision. It should never be started or managed by the family alone.
When the ketogenic diet was reintroduced in the 1990s, many were skeptical about its usefulness at a time when an increasing number of new anticonvulsant medications were available. The table below shows the results of recent studies of the diet at Johns Hopkins.4,5
|Time from start of the ketogenic diet*||Seizure control|
|Seizure-free N (N/150)||>90% reduction N (N/150)||50%90% reduction N (N/150)||<50% reduction N (N/150)||Patients maintained on the diet N (N/150)|
|3 mo||4 (3%)||46 (31%)||39 (26%)||36 (24%)||125 (83%)|
|6 mo||5 (3%)||43 (29%)||29 (19%)||29 (19%)||106 (71%)|
|12 mo||11 (7%)||30 (20%)||34 (23%)||8 (5%)||83 (55%)|
|36 yr||20 (13%)||21 (14%)||24 (17%)||18 (12%)||15 (10%)|
One hundred and fifty children with difficult-to-control seizures were started on the diet. They were followed initially for one year4 and subsequently for three to six years.5 The children had mixed seizure types, with different causes, and varying levels of intellectual function. Their mean age was 5.3 years. They averaged 410 seizures a month at baseline and had failed a mean of 6.2 anticonvulsant medications before starting the diet.
Most (83%) children who started the diet stayed on it for three months. Seventy-one percent remained on it for six months, and 55% remained on it for a year. The diet's effects on seizure control were independent of patients' intelligence and developmental delay. We have found that the diet is palatable and tolerable if it is effectivethat is, if it either substantially decreases the number of seizures or significantly reduces medication side effects by permitting a decrease in anticonvulsants.
The ketogenic diet has been prescribed traditionally for children older than 1 year and was long thought not to be tolerated by teenagers or adults. Recent studies, however, document that it is well tolerated and effective in children as young as 3 months of age,8 teenagers,9 and, possibly, adults.10
We ask parents to continue the diet for a minimum of three months to allow the child time to adjust to it, to modify recipes to meet the child's preferences, and to allow time for us to modify the diet to achieve maximal ketosis and optimal seizure control. During this time we often reduce or eliminate one or more of the anticonvulsants the child is taking. After three months, parents and child are free to continue or discontinue the diet.
As noted, 83% of children in the Johns Hopkins studies stayed on the diet for the minimum three-month period and 55% stayed on it for one year. The major reason for discontinuing the diet was that it was not sufficiently effective to warrant the restrictions or the bother. (Children who started the studies having failed multiple medications and having hundreds of seizures had a lower than 10% chance of achieving seizure control.) If the diet was effective, it was well tolerated. Note in the table that 21 children (14%) of the initial study population had greater than 90% seizure reduction three to six years after starting the diet. Most children discontinue the diet after two or three years. Figure 1 shows examples of meals on the diet. Some sample ketogenic recipes are shown below.
Reservations that have been expressed about the ketogenic diet include the following:
"That's not a healthy diet. They won't grow. They'll all get fat." Figure 2 shows growth on the ketogenic diet. Children's growth rate remains within normal ranges, although younger children grow somewhat more slowly (along the third percentile).12 The child's weight is closely monitored and caloric intake is adjusted to maintain ideal body weight. Calories are increased or decreased depending on the child's activities and growth.13,14 Normally active children are maintained at close to the 50th percentile of weight for height. Children with a substantial handicap are maintained at the 10th to 25th percentile.
"Everyone knows that a high-fat diet will elevate lipids and cause heart disease." Figure 3 shows changes in cholesterol and other lipids while children were on the ketogenic diet.14,15 True, after six months on the diet plasma lipids rise to levels that, if maintained, would raise concern about later atherosclerosis. Few patients remain on the diet for more than two years, however.
The mean cholesterol level becomes slightly elevated, but stabilizes at 220 mg/dL. Low-density-lipoproteins (LDLs) also stabilize, and triglycerides return to baseline levels.14,15 Because most children discontinue the ketogenic diet within two years of starting it and return to a diet containing far less fat, there is little reason to expect long-term adverse effects. This supposition has not been studied, however. Parents should be made aware of potential adverse effects of a high-fat diet so that they can weigh those risks against the risks associated with continuing seizures and taking anticonvulsant medications.
Because the ketogenic diet is best started under the care of a team that includes a pediatric neurologist and a dietitian with experience in managing children on the diet, few pediatricians have sufficient demand or support to develop expertise in initiating or managing the diet on their own. The pediatrician does play an important role as part of the management team, however. We see this role as following the child's health, growth, and nutritional status. An additional major role, because the family often lives a distance from the medical center where the ketogenic diet was prescribed, is managing intercurrent illness. There is no evidence that children on the ketogenic diet have more (or fewer) illnesses than their peers. When children with a seizure disorder have an illness or fever, however, ketosis may decrease and seizures return or increase.
When the parent of a child on the diet calls to tell us the child has had a seizure or the seizures have become worse, the first questions we ask are, "Is he ill? Does he have a fever? Is he coming down with something?"13 Children who are ill should be treated appropriately (see "Points to remember: Caring for an ill child on the ketogenic diet"). Except for using carbohydrate-free medications, treatment is the same as for children who are not on the diet.
Vomiting, diarrhea, pneumonia, and other illnesses are rarely reasons to discontinue the diet. If the sick child is not tolerating solid foods, ketogenic diet eggnogs may provide nutrition (and may be diluted if necessary). Clear, carbohydrate-free fluids may provide adequate hydration for a day or so. Diluted oral rehydration solutions (e.g., Pedialyte diluted in half because of the glucose content) may be used for a day to maintain electrolyte levels, if needed. Severe vomiting and diarrhea may cause substantial acidosis (pH less than 7.1) and dehydration. Intravenous hydration with saline solution (without carbohydrate) is usually sufficient to return the child to normal balance.
Carbohydratesincluding glucose, fructose, starches, and compounds ending in -ol (sorbitol, mannitol) or -ose (maltose, zylose)are converted completely or partially to glucose in the body, which can negate ketosis and lead to seizures. Most pediatric medications contain carbohydrates (see "Watch out for carbohydrates in medications").
The reported side effects in children on the ketogenic diet have been described by Wheless.16 The two most common effects are constipation and kidney stones. Although death, dyslipidemia, cardiomyopathy, and trace mineral deficiencies have been reported, the causal relationship of these side effects, so to speak, to the diet remains unclear.
Constipation. We have found that adding Miralax, MCT oil, or carbohydrate-free, liquid aloe vera, which parents can usually get at a local pharmacy, is the most effective way to manage constipation. Other classic remedies (if carbohydrate-free) are sometimes useful. Increasing intake of fluids may help. The point is to use whatever works for the child.7,13
Kidney stones. Kidney stones occur in 6% to 7% of children on the ketogenic diet. These are usually calcium citrate or uric acid stones caused by precipitation of these compounds in the concentrated, acidic urine.17,18 They often are accompanied by "grit" in the urine, abdominal discomfort, urinary tract infection, or an otherwise unexplained increase in the frequency of the child's seizures.
We ask parents to test the child's urine weekly for blood using test strips. When urine is positive for blood on two consecutive tests, the pediatrician performs a urinalysis to look for red blood cells. A positive microscopic urinalysis is followed by a consultation with a urologist and ultrasonography of the kidney. A spot urine specimen taken every three months is also tested for the calcium/creatinine ratio, which should be less than 0.2. An elevated ratio, or the presence of a stone, is treated by increasing fluid intake to 100% to 120% of the usual daily requirement and alkalinizing urine to pH 6.5 with polycitrate solution (a solution containing sodium and potassium citrate, which is readily available at pharmacies). Most stones dissolve, but persistent stones may need ultrasound treatment or, rarely, surgery. Once a stone has been treated, patients have remained successfully on the ketogenic diet with the addition of liberalized fluids and polycitrate.
If the seizures are gone, and the medication side effects are reduced, the diet is helping. If seizures continue to occur, although less often, and the child is still taking medication, parents may ask, "Is the diet helping?" The only way to determine whether the diet is helping is to test it by discontinuing it to see if the frequency of seizures remains the same or increases. The classic way to discontinue the diet is to taper the ratio of fat calories to protein and carbohydrate very slowly, progressing every three to six months from a 4:1 ratio to 3.5:1, then to 3:1, to 2.5:1, and to 2:1and then discontinuing the diet.
A more rapid way, which we often use but have not assessed carefully, is to maintain the same meal plans but substitute regular milk for the cream in the diet. This reduces the ratio from 4:1 to approximately 2.5:1. After one week, if the frequency of the child's seizures remains unchanged, we switch from regular milk to skimmed milk, which reduces the ratio to about 1.5:1. If the frequency still remains unchanged, we discontinue the diet. The advantage of this approach is that it maintains the restrictions of the diet during the test periodthe child is not allowed to eat foods such as candy or potato chipsso that it is not difficult for the child to resume the regular ketogenic diet if necessary.
A third way to test the diet is to allow the child to cheata little. If nothing adverse happens, gradually introduce some carbohydrates into the child's meals until he or she is no longer in ketosis.
The ketogenic diet is used more and more in children whose seizures are difficult to control.19 When it is effective, it is well-tolerated by children and parents. Pediatricians need to be aware of the dietespecially its carbohydrate limitationsand be prepared to support the child and his (or her) parents during intercurrent childhood illnesses as they struggle with difficult seizure problems.
Successful use of the diet often permits a decrease in anticonvulsant medications and, therefore, their side effects. When a child with difficult-to-control seizures becomes seizure-free or virtually so, and is able to decrease or eliminate the side effects of multiple medications, the diet is effective. If, after two years, the child can discontinue the diet, has no seizures, and requires no anticonvulsant medication, that is a very significant achievement. Even when the diet does not prove effective, most parents are grateful to have given it a chance.
1. Swink TD, Vining EPG, Freeman JM: The ketogenic diet 1996. Adv Pediatr 1997;44:297
2. Schwartzkroin PA: Mechanisms underlying the anti-epileptic efficacy of the ketogenic diet. Epilepsy Res 1999;37:171
3. LeFever F, Aronson N: Ketogenic diet for the treatment of refractory epilepsy in children: A systematic review of efficacy. Pediatrics 2000;105(4)URL: http://www.pediatrics.org/cgi/content/full/105/4/e46 .
4. Freeman JM, Vining EPG, Pillas DJ, et al: The efficacy of the ketogenic diet1998: A prospective evaluation of intervention in 150 children. Pediatrics 1998;102:1358
5. Hemingway C, Freeman JM, Pillas DJ, et al: The ketogenic diet: A 3-6-year follow-up of 150 children prospectively enrolled. Pediatrics 2001;108:898
6. Freeman JM, Vining EPG: Seizures rapidly decrease after fasting: Preliminary studies of the ketogenic diet. Arch Pediatr Adolesc Med 1999;153:946
7. Freeman JM, Freeman JB, Kelly MT: The Ketogenic Diet: A Treatment of Epilepsy, ed 3. New York, Demos Press, 2001
8. Kossoff EH, Pyzik PL, McGrogan JR, et al: Efficacy of the ketogenic diet for infantile spasms. Pediatrics 2002;109(5):780
9. Mady MA, Kossoff EH, McGregor AL, et al: The ketogenic diet: Adolescents can do it too. Epilepsia (in press)
10. Sirven J, Whedon B, Caplan D, et al: The ketogenic diet for intractable epilepsy in adults: Preliminary results. Epilepsia 1999;40(12):1721
11. Stafstrom CE, Spencer S: The ketogenic diet: A therapy in search of an explanation. Neurology 2000;54:282
12. Vining EPG, Pyzik P, McGrogan J, et al: Growth of children on the ketogenic diet. Dev Med Child Neurol 2002;44:796
13. Casey JC, McGrogan J, Pillas D, et al: The implementation and maintenance of the ketogenic diet in children. J Neurosci Nurs 1999;31:294
14. Vining EPG, Kwiterovitch P, Freeman JM, et al: The effect of the ketogenic diet on plasma cholesterol. Presented at the Annual Meeting of the American Epilepsy Society, San Francisco, December 1996. Epilepsia 1996;37(Suppl 5):107
15. Kwiterovitch P, Vining EPG, Goodman S, et al: Changing lipid profiles after institution of the ketogenic diet. (in preparation)
16. Wheless JW: The ketogenic diet: An effective medical therapy with side-effects. J Child Neurol 2001;16:633
17. Furth SL, Casey JC, Pyzik PL, et al: Risk factors for urolithiasis in children on the ketogenic diet. Pediatr Nephrol 2000;15:126
18. Kossoff EH, Pyzik PL, Furth SL, et al: Kidney stones, carbonic anhydrase inhibitors, and the ketogenic diet. Epilepsia 2002;43(10):1168
19. Hemingway C, Pyzik PL, Freeman JM: Changing physician attitudes toward the ketogenic diet: A parent-centered approach to physician education about a medication alternative. Epilepsy and Behavior 2001;2:574
Peanut butter "sandwich"
27 g egg white
13 g cottage cheese (4% fat)
26 g butter
4 g Skippy peanut butter
1/4 tsp cream of tartar
Whip egg white until peaks form. Add cottage cheese, cream of tartar, and a couple of drops of sweetener. Place two piles of the egg white mixture onto a cookie sheet sprayed with Pam. Bake at 350° for 20 to 25 minutes (until lightly browned). When the baked egg whites have cooled, mix butter and peanut butter together and spread on the egg whites to make a "sandwich." To make bagels, make holes in the egg white piles on the cookie sheet before baking.
The sandwich plus 30 g of cream as a drink makes a 345-calorie meal with a 4:1 ratio of fat to protein and carbohydrate.
2 egg whites
1 tsp cream of tartar
1 small package of sugar-free Jello
Beat egg whites until stiff. Add cream of tartar and Jello. Place on nonstick baking sheet and bake at 325° for 6 to 8 min or until brown. Makes 20 cookies.
One serving = 2 cookies.
John Freeman. What every pediatrician should know about the ketogenic diet.