Nutrition in Clinical Practice
Invited Review
An Overview of the Ketogenic Diet for
Pediatric Epilepsy
Beth A. Zupec-Kania, RD; and Emily Spellman, MS, RD
Financial disclosure: none declared.
Epilepsy is the most common serious neurological condition in
the world, with an estimated prevalence of 1% of the population.
The highest incidence occurs in childhood and in the elderly,
with lower levels in early adulthood. Traditional epilepsy man-
agement includes pharmacological treatment, epilepsy surgery,
and vagal nerve stimulation. Despite these therapies, 25% of
children continue to have uncontrolled seizures. The ketogenic
diet (KD), which has been in use since 1921, is a treatment
option for many of these children. A meta-analysis of 19 studies
with a combined sample of 1084 pediatric patients was com-
pleted in 1998 by Blue Cross Blue Shield. Estimates of the over-
all efficacy of the KD in controlling seizures were reported as
follows: 16% became seizure free, 32% had a >90% reduction in
F
asting as a method to control seizures has been rec-
ognized for centuries, perhaps since biblical times.
Deliberate fasting has been shown to control
seizures.1 The impracticality of prolonged starvation
became the drive to formulate the ketogenic diet (KD).
The diet was designed to simulate the ketosis of starva-
tion by supplying fat as the major source of calories, while
restricting protein and carbohydrate. During metabolism
of a regular diet, glucose is derived from carbohydrate and
is the main fuel source for the body and brain. Conversely,
in the absence of significant carbohydrate and the abun-
dance of dietary fat, mitochondrial β-oxidation of fatty
acids in the liver generates large quantities of the ketone
bodies acetoacetate and β-hydroxybutryate (BHB), which
the brain can readily use as energy metabolites. Excess
production of ketone bodies results in a state of ketosis.
The exact mechanisms of seizure inhibition remains
unknown; however, animal research has demonstrated
the neuroprotective effects of the diet, which no other
antiepileptic therapy has been known to provide.2,3
Ketogenic diets are categorized as either long-chain
fatty acid based (classic) or medium-chain fatty acid
From Nutritional Services, Children’s Hospital of Wisconsin,
Milwaukee, Wisconsin.
Address correspondence to: Beth A. Zupec-Kania, RD, Children’s
Hospital of Wisconsin, Nutritional Services MS#802, 9000 W.
Wisconsin Ave, Milwaukee, WI 53122; e-mail: bkania@chw.org.
589
Downloaded from ncp.sagepub.com at UCSF LIBRARY & CKM on March 16, 2015
Volume 23 Number 6
December 2008 589-596
© 2008 American Society for
Parenteral and Enteral Nutrition
10.1177/0884533608326138
http://ncp.sagepub.com
hosted at
http://online.sagepub.com
seizures, and 56% had a >50% reduction. The KD is high in fat,
moderate in protein, and low in carbohydrates. This combina-
tion of energy results in a sustained ketosis that somehow serves
to abate seizures through an unknown mechanism. Strict adher-
ence to the diet is required for it to be effective. Newer, more
liberal versions of the KDs have been recently introduced and
are being studied in children and adults. Administration of all
of these metabolically adjusted diets must be medically man-
aged as there can be adverse effects. The focus of this review
is on the pediatric application of the KD. (Nutr Clin Pract.
2008;23:589-596)
Keywords: diet therapy; ketosis; ketones; seizures; epilepsy
based. Common to both diets is the low glycemic effect
that they produce because carbohydrate is restricted and
the high fat content slows the absorption of carbohydrate.
Serum glucose levels of children on KDs are typically
between 55 and 75 mg/dL. These diets are provided with
controlled calories to prevent excessive weight gain but
with sufficient calories to support growth.
The classic diet is calculated using a ratio of the
weight of fat to the sum of protein and carbohydrate. In a
4:1 ratio, there are 4 g of fat for every 1 g of protein and
carbohydrate combined. Protein is provided to meet dietary
reference intake, which is approximately 1 g per kilogram
of body weight. Carbohydrate completes the remaining
allowance of the ratio. A 3:1 ratio is generally used for
older children who have greater body mass and therefore
higher protein needs. To achieve these ratios, foods rich
in carbohydrate are eliminated from the diet, including
sugar, pasta, bread, and grains; however, low-carbohydrate
vegetables and fruit are allowed. The main sources of fat
are heavy cream, vegetable oils, and butter. In addition
to meals, KD snacks are recommended for active children.
A typical day includes 3 meals plus 1-2 snacks (as shown
in Table 1).
An example of a 1000-cal diet with a 4:1 ratio pro-
vides the following: 100 g fat, 18 g protein, and 7 g car-
bohydrate. These values are divided evenly among the
number of daily meals and thus all meals are inter-
changeable. A mathematical algorithm is used to account
for the fat, protein, and carbohydrate content of each
590 Nutrition in Clinical Practice / Vol. 23, No. 6, December 2008
Table 1. Examples of Meals and Snack; 1100 kcal,
4:1 Ketogenic Diet (For a 3-Year-Old Child)
Breakfast
40 g 36% Heavy cream
32 g Sausage links
24 g Avocado—Hass
5 g Canola oil
Lunch
40 g 36% Heavy cream
18 g Sliced turkey breast
22 g Raw cucumber slices and celery sticks
22 g Mayonnaise
Supper
40 g 36% Heavy cream
20 g Baked cod
43 g Roasted cauliflower
23 g Butter
Snack
10 g Sliced strawberries
28 g 36% Whipped heavy cream
food and to calculate the amounts of food in grams that
are needed to achieve goal values per meal. Each meal
plan becomes a recipe to guide the caregiver on how to
prepare and then weigh each food item on a metric scale.
The calculation of a single meal can be tedious, and the
possibility for errors is significant. Computer applications
for calculating KDs, such as KetoCalculator, are available
and can save time and minimize errors.4 A sample calcu-
lation of a single meal is shown in Table 2.
A medium-chain triglyceride (MCT) diet contains
70% calories from fat, using MCT oil as the main source
(50%) and 20% from polyunsaturated sources. (MCT oil
is extracted from coconut oil by a process called fraction-
ation.) This results in a ratio of fat to nonfat of approxi-
mately 1:1, allowing more carbohydrate and protein and
significantly less fat than the classic KD. The main com-
plaints from users of this version of the diet have been
GI side effects, including diarrhea and abdominal pain.
Another disadvantage includes the high cost of MCT oil
and the difficulty in receiving reimbursement from insur-
ance companies for its use.5 A study of 6 children receiv-
ing the MCT diet who were transitioned to coconut oil
(which contains 45% MCTs) was recently presented. This
small study revealed that coconut oil was equally effective
in maintaining ketosis, was well tolerated by all 6 chil-
dren, provided the same seizure-control benefits, and was
more economical than MCT oil.6
Dietitians with extensive experience in managing
these diet therapies have resorted to combining features
of both diets. For example, adding MCT oil to the classic
KD is a strategy that is used in an attempt to improve
seizure control by improving ketosis. Another example is
reducing the diet to a ratio of 3:1 or 2:1 to provide a larger
quantity and variety of protein-rich and carbohydrate-rich
Downloaded from ncp.sagepub.com at UCSF LIBRARY & CKM on March 16, 2015
foods. These adjustments are often employed when
seizure control has been successful but the individual is
struggling to adhere to the diet. Through trial and error,
we have also learned to begin the diet with lower ratios,
such as 2:1 or 1:1, for adolescents and teenagers for
whom we believe compliance will be best achieved (simi-
lar to the newer versions of the diet described later).
Within these liberal ratios, we also use a much greater
quantity of protein than carbohydrate, going by the the-
ory that protein consumed in excess of needs will be con-
verted to glucose (gluconeogenesis) more slowly than
carbohydrate conversion to glucose (glycolysis). This may
contribute to more stable glucose levels and stronger
ketosis and may benefit seizure control; however, this the-
ory has not been tested. We also incorporate MCT oil or
coconut oil into these liberal ratios when attempting to
achieve better seizure control.
Newer Versions of the Diet
Two newer, more liberal versions of the KD have been
recently reported to have seizure-control benefits. The
low-glycemic-index treatment focuses on limiting spe-
cific types of carbohydrate. Foods that have an individual
glycemic index of 50 or greater are eliminated.7 In addi-
tion, 60% of calories are provided from fat, and protein
is provided at approximately 1 g/kg of body weight. This
diet requires measuring or estimating foods in household
measure and is offered when the classic KD is expected
to be too difficult to manage, especially for adolescents.
The second version is the modified Atkins diet, which
limits carbohydrate and provides 65% of calories from
fat.8 Both versions result in an approximate ratio of 1:1
(fat to nonfat). Early reports of both diets show encour-
aging results; however, there is no efficacy data of long-
term use of either of these diets. Medical monitoring of
patients following these diets is the same as for the
stricter versions.
Indications
The KD has been historically reserved for patients who
are poor candidates for epilepsy surgery and for whom
traditional antiseizure therapy has failed or caused unac-
ceptable side effects. In a consensus report from the
International KD Study Group, administering the diet
early in the course of epilepsy is recommended, after the
failure of the second or third antiseizure medication.9
This is exciting news for patients, caregivers, and practi-
tioners alike because “early use” also means early in a
child’s life, when diet restrictions are less problematic.
The KD has been used with infants as young as
3 months and adults as old as 58 years. Practitioners agree
that compliance is highest in children up to 5 years.
Older children are more likely to resist the strictness of
 Ketogenic Diet for Pediatric Epilepsy / Zupec-Kania, Spellman 591

Table 2. Calculation of a Single Ketogenic Meal

Food Item Grams Fat (g) Protein (g) Carbohydrate (g) Kilocalories Ratio

36% Heavy cream 40 14.4 0.8 1.2 138

Cauliflower, cooked 43 0.19 0.79 0.78 8
Cod, baked 20 0.17 4.57 0 20
Butter 23 18.66 0.2 0.01 169
Sum 33.42 6.36 1.99 335 4:1

the diet. Children with enteral feeding tubes are not Adverse Effects
affected by the compliance factor and, from this stand-
The most common adverse effect of the KD is constipation.
point, make good candidates. The diet appears to have a
The regular inclusion of fibrous vegetables, carbohydrate-
broad spectrum of efficacy among different epilepsy
free stool softeners, mild laxatives, and sufficient fluids are
syndromes. No seizure or epilepsy type has been shown to
helpful in maintaining bowel regularity. Gastroesophageal
be uniquely resistant to the KD, and several types have
reflux disease may be exacerbated by the high fat intake,
been reported to respond well to the diet, including
which slows gastric emptying. Dividing the diet into several
Doose syndrome, tuberous sclerosis complex, Rett syn-
meals along with medication management can improve or
drome, Dravet syndrome, myoclonic astatic epilepsy, and
correct this problem. Kidney stones have been reported in
infantile spasms.10-15 The KD is the treatment of choice
a small percentage of patients.22,23 The use of carbonic
for 2 metabolic disorders: glucose 1 deficiency syndrome
anyhdrase inhibitors carry a risk for nephrolithiasis in
and pyruvate dehydrogense deficiency.16,17 In both condi-
themselves; therefore, caution is advised with this class of
tions, glucose metabolism is impaired, and the KD provides
antiepileptic medication when used in combination with
ketones as an alternative fuel source.
the diet. Prophylactic use of bicarbonate solutions with
There have been many anecdotal reports of beneficial
these medications has been practiced; however, this alka-
effects of the diet on behavior and cognition; however,
line environment may impair the absorption of folic acid
there is limited research on this topic. A preliminary
in the gut and can lead to megabolastic anemia.24
report from researchers at Johns Hopkins of 34 children
Abnormal lipid levels are a concern for most care-
who received developmental testing prior to and after 1
givers and providers with this therapy. A prospective study
year on KD therapy found significant improvements in
of 141 children following the KD revealed elevations in
the children’s attention and social functioning.18 In a
low-density lipoproteins and reductions in the high-density
recent study, epileptiform activity was measured using
lipoprotein lipids but concluded that further studies were
electroencephalogram monitoring prior to the initiation
necessary to determine if these effects promoted ather-
of the KD and after 3 months on the diet. A significant
sclerosis.25 Diet adjustments for improving lipid profiles
correlation was found between the reduction in the inter-
are addressed later in this article.
ictal discharges and clinical seizures and improvement in
Adverse effects related to nutrition include growth
attention and concentration in these children.19
retardation, hypoproteinemia, and micronutrient defi-
ciencies. Additional serious adverse effects have been
Contraindications reported in a few patients and include hemolytic anemia,
Fanconi renal tubular acidosis, elevations in liver func-
The diet is contraindicated in individuals with defects in tion tests, pancreatitis, increased bruising, prolonged QT
the transport or oxidation of fatty acids, lipid myopathies, intervals, and coma.21,26,27 It is worthwhile to note that
primary or acquired carnitine deficiency, pyruvate car- these effects occurred in combination with antiepileptic
boxylase deficiency, acute intermittent porphyria, and medications and may have been an exacerbated effect of
organic acidurias.20,21 A candidate for the KD should also the medication in combination with the diet.
be nutritionally stable and be able to receive the diet, fluid
recommendations, and vitamin and mineral supplements
on a consistent basis. The diet requires special training Interdisciplinary Team
and a high degree of motivation by the individual or care- Approach to the KD
giver. Meal preparation requires weighing of specific
foods, and the recipient must consume each meal in its Medical nutrition therapy (MNT) is described by the
entirety to achieve optimal effectiveness. Because the diet American Dietetic Association as an essential component
includes the purchase and special preparation of food, the of healthcare that affects the outcome of an individual’s
caregiver should be able to grocery shop once a week and health and/or chronic medical condition. It can increase
be organized to ensure daily therapy is realized. productivity and satisfaction levels through decreased

Downloaded from ncp.sagepub.com at UCSF LIBRARY & CKM on March 16, 2015
592 Nutrition in Clinical Practice / Vol. 23, No. 6, December 2008
Table 3. Education Goals for the Ketogenic Diet
Fluid guidelines for daily hydration
Food preparation techniques
Use of gram scale and weighing technique
Vitamin and mineral supplementation requirements
Sick-day diet
Constipation prevention and management
Troubleshooting hypoglycemia and excessive ketosis
Diary for tracking ketones, weight, and seizures
doctor visits and hospitalizations and reduced prescrip-
tion drug use. The KD should be viewed as an MNT, and
one that necessitates a team approach. Of foremost
importance in establishing a KD program is the commit-
ment of an epileptologist or neurologist as a medical
director. This individual should have pediatric experience
and an active involvement in this labor-intensive regimen.
A registered nurse is essential in providing medical man-
agement of the KD. The nurse should also have a back-
ground in pediatric epilepsy care. The team may also
include a medical or psychiatric social worker to aid the
family in the challenges of this therapy. A registered phar-
macist is an important asset to the team in providing cur-
rent information regarding the carbohydrate content of
medications and nutrition supplements.
A registered dietitian with pediatric experience is vital
to the success of a KD program. The dietitian provides
nutrition management and technical manipulation of the
diet to optimize seizure control. The dietitian’s frequent
communication between caregivers and the KD team
necessitates his/her role as coordinator of the KD pro-
gram. Because of the multiple demands of this program,
it is advantageous to have a second dietitian (within the
facility) with training on the KD protocols to provide con-
tinuous care in the absence of the primary dietitian. Table 3
identifies the educational goals that should be taught to
caregivers by the dietitian for the safe management of
this therapy.
The KD is generally used for a period of up to 3 years.
Seizure control benefits are typically seen within 1-3 months
of the diet’s initiation. The international study group reports
that the diet should be trialed for at least for 3½ months
before deciding to discontinue it.8 The diet can be discon-
tinued earlier if seizures worsen beyond expectations or
adverse effects cannot be corrected. A clinical pathway for
the course of KD therapy is shown in Figure 1.
Of the children who respond positively to the KD,
some experience a complete cessation of seizures with the
onset of ketosis. Most children, however, experience grad-
ual improvements. After 1 or 2 years, the diet is gradually
tapered by lowering its fat content and increasing the car-
bohydrate and protein portion of the diet (by lowering the
ratio) until ketosis is eliminated. The exact period of time
depends on the individual patient. Most families will
Downloaded from ncp.sagepub.com at UCSF LIBRARY & CKM on March 16, 2015
Epilepsy/Neurology Clinic Neurological evaluation and referral to KD program
Biochemical studies (screening)
Dietitian consultation Nurse notified Insurance preauthorization
Social worker consultation
Pharmacist consultation
Hospital Admission for KD Laboratory studies (baseline)
KD initiation
Serum glucose checks
Urine ketones/serum β−hydroxybutryate
Fluid management education
Diet education/demonstrations
Diet management (sick days/constipation)
Posthospitalization Outpatient follow-up program
Phone support from KD team
KD team clinic follow-up at 1 month
postdischarge and every 3 months thereafter
Repeat baseline laboratory indices
Monitor growth
Review seizure history
Evaluate/adjust medications if needed
Evaluate/adjust diet if needed
Provide additional meal calculations
Discontinue Diet Rapid tapering off KD after 3½ months if ineffective
Gradual tapering off KD after 2 years if effective (ie, over 1 year)
Discontinue diet if serious complications exist or persist
Figure 1. Ketogenic diet (KD) clinical pathway.
make this determination by weighing the improvement in
seizure control and quality of life with the time invested
in managing the diet.
The planning and preparation of a KD program
begins with hospital administration. The process should
include a protocol for inpatient and outpatient care, care-
giver education materials, and a computer program for
calculating and managing the diet. In 1999, it was esti-
mated that the KD required 16 hours of dietitian time
over a period of 5 months for each patient initiated on the
diet.28 A comparison of medical costs for the management
of a patient with epilepsy on the KD indicated that chil-
dren who were successfully maintained on the KD had
lower medical costs during the course of the diet than
prior to the diet.29
To assist in the assessment of a candidate for KD
therapy, an intake form should be completed by the care-
giver. Pertinent dietary information, including food pref-
erences, intolerances, and allergies, should be collected
along with a diet diary. Current medications, supple-
ments, weight, height, bowel habits, as well as seizure
type and frequency should also be obtained. This infor-
mation will assist in the design of an individual’s diet and
assist the dietitian in counseling the caregiver regarding
expectations.
A pharmacist or nurse should review the current
medications of the patient and provide recommendations
for alternatives to minimize the carbohydrate contribu-
tion from this source. A social worker can help the family
to obtain healthcare services that may assist with the
special needs of the diet, including the appropriate food
 Ketogenic Diet for Pediatric Epilepsy / Zupec-Kania, Spellman
Table 4. Suggested Laboratory Studies for Patients
Recieving Ketogenic Diet (KD) Therapy
Pre-KD biochemical screening
Fasting lipid profile
Serum acylcarnitine profile
Anticonvulsant drug levels
Quantitative urine organic acids
Chemistries including serum levels of glucose, blood urea
nitrogen, liver function, calcium, phosphorus, proteins
Electrolytes with bicarbonate
Complete blood count with differential
Surveillance laboratory studies
Fasting lipid profile
Serum acylcarnitine profile
Anticonvulsant drug levels
Chemistries including serum levels of glucose, blood urea
nitrogen, liver function, calcium, phosphorus, proteins
Electrolytes with bicarbonate
Complete blood count with differential
25-Hydroxyvitamin D
procurement and purchase of a scale and the required
vitamin and mineral supplements. A psychiatric social
worker is beneficial in counseling the family regarding the
restraints that the diet may place on family life.
Administration of the KD
Prior to initiating the KD, the candidate should be evalu-
ated by an epileptologist or neurologist. Laboratory stud-
ies should be obtained to determine baseline nutrition
status. In addition, antiepileptic drug levels (when appro-
priate) and biochemical studies to rule out disorders in
fatty acid oxidation should be obtained (Table 4).
Fasting is no longer considered necessary to begin the
diet. Two randomized clinical trials of fasting versus
gradual initiation of the diet showed that the patients had
equivalent long-term seizure reduction at 3 months.
Hypoglycemia and vomiting were less problematic for the
groups that did not fast and therefore hospitalization
was shorter.30,31 Even without a fasting period, the child
is at risk for hypoglycemia, dehydration, acidosis, and
antiepileptic drug toxicity. Therefore, initiation is usually
provided in a hospital setting under the guidance of the
KD team. The KD diet may be introduced with a gradual
transition to the established ratio over a period of 3-4
days. After the initiation period, the child is followed in
an outpatient clinic by the members of the KD team.
Regularly scheduled appointments are necessary to monitor
the child’s seizures, growth, and nutrition status through-
out the course of therapy. Laboratory studies are performed
at these visits to monitor metabolic responses. Figure 1
outlines a clinical pathway for KD therapy.
Downloaded from ncp.sagepub.com at UCSF LIBRARY & CKM on March 16, 2015
593
Adjusting and Fine-tuning the Diet
During the course of therapy, the diet will need to be
adjusted to support normal growth. Growth disturbances
have been reported and are likely related to restricted
energy and protein intake.32 An increase in energy expendi-
ture may be experienced by children who have achieved
improvements in seizure control and have become more
active. A caloric increase should be considered in these
cases. The diet may also require adjustments to make it
more palatable. Some children enjoy consuming fat in the
form of butter whereas others prefer less butter and cream
with higher fat content (ie, 40% vs 36% cream). Some chil-
dren are unable to consume the quantity of fat required in
the 4:1 diet but will tolerate a 3:1 diet with lower fat, which
also supplies more protein and carbohydrate.
The initial goal is to ensure that 100% of the diet is
being consumed. This may take several weeks, and the
dietitian may need to offer creative meals to improve
acceptance. Feeding the child at evenly spaced intervals
throughout the day will assist with hunger management.
Ketosis suppresses appetite, and most children adapt to
the small volume of food without difficulty. Other chil-
dren require strategies to assist them in adjusting, such as
removing them from the eating area and engaging them
in a different activity directly after the meal.
The next goal is to find and maintain a state of keto-
sis that serves to abate seizures. After a few weeks of KD
therapy, a defining pattern of ketosis will emerge. Ketosis
is monitored at home with urine testing. Urine ketone
levels are a crude estimate of serum ketone levels but can
suffice for simple monitoring purposes. BHB can be
measured quantitatively in blood serum and is less influ-
enced by hydration status. According to one study, a BHB
>4 mmol/L was associated with better seizure control
than lower levels.33 Serum glucose monitoring is another
method of monitoring this therapy, because serum glu-
cose levels have been observed to occur in a narrow range
of approximately 55-75 mg/dL (at least 2 hours postpran-
dial) and strong ketosis is noted within this range.
Patients who experience improved seizure control with
low to moderate ketosis may achieve better seizure control
with higher ketosis (although there are exceptions and the
opposite may be true). One method to achieve higher keto-
sis is to increase the ratio of fat to nonfat calories of the diet.
Another approach to achieving stronger ketosis is to adjust
energy intake. If a patient is receiving too many calories, he
or she may experience lower ketosis because of excessive
energy intake and may gain weight too rapidly. In this case,
reducing the calories of the diet may improve ketosis.
Conversely, if a patient is receiving insufficient calories, he
or she may experience lower ketosis (ie, the starvation
effect) because of inadequate energy intake. In this case,
increasing the calories of the diet may improve ketosis.
594 Nutrition in Clinical Practice / Vol. 23, No. 6, December 2008
Are seizures controlled?
Yes
Continue diet
No
Adjust diet to improve intake;
Calculate new food preferences
Creative new recipes
Reduce ratio to liberalize diet
Rapid weight gain or loss?
Yes
No
Adjust calories
Increase ratio
If too ketototic, reduce ratio
Eliminate processed foods
Incorporate MCT oil into diet
If no seizure improvement after 12-14 weeks,
taper ratio to eliminate ketosis (ie, over 3 days).
Figure 2. Flowchart for adjusting the ketogenic diet (KD).
AED, anti-epileptic drug; MCT, medium chain triglyceride.
Incorporation of MCT oil or coconut oil (which contains
45% MCT) into the classic diet is an effective strategy to
enhance ketosis. The MCT also has a laxative effect, which
is helpful because constipation is the main adverse effect of
the KD. Adjusting or fine-tuning the diet should be dis-
cussed with the KD team. It is important to make only 1
change at a time, then to allow a period of time to ascertain
the effect of each change. Possible causes of temporary
breakthrough seizures may include the stressors that have
triggered the child’s seizures in the past, such as sleep dep-
rivation, an illness, or becoming ill. Seizures may occur
because of disruptions in the diet, such as a mistake made
in the preparation or weighing of food, eating or drinking a
non-KD food, or the introduction or change in a medica-
tion. Figure 2 outlines a flowchart to assist in fine-tuning
and adjusting the diet during the course of therapy.
Nutrition Adequacy of KD
Individuals with epilepsy may experience normal growth
and development; however, some may be severely
Downloaded from ncp.sagepub.com at UCSF LIBRARY & CKM on March 16, 2015
No
Is child consuming 100% of diet?
Yes
Are ketones consistently large?
Urine 80-160 mg/dL
β-Hydroxybutryate 4-10mmol/L
No Yes
Are there temporary stressors?
Illness or becoming ill
Constipation
New medication or medication with sugar
AED too strong or tapering off
Teething
Errors in diet or noncompliance
No Yes
Reduce or correct stressor
or “ride it out”
incapacitated by seizures. A nutrition assessment should
include an evaluation of the individual’s ability to con-
sume food and fluids, growth pattern, activity level, and
cognitive function. The KD can be modified in consis-
tency for safe consumption such as pureed textures or
into liquid formulations for infant or enteral feeding.
Individuals with severe feeding difficulty should be evalu-
ated by a feeding therapist prior to considering the KD. If
a child is at risk for aspiration of food or liquids, an
enteral feeding tube can be placed, which will allow safe
administration of the diet.
Vitamin and mineral supplementation is vital because
of insufficient sources supplied from the diet. A micronutri-
ent analysis of an optimally selected 4:1 KD revealed that
only 5 of the 24 known essential micronutrients were met
even when only nutrient-dense foods were included. The
water-soluble vitamins were significantly lacking except for
vitamin B12. Fiber, linolenic acid, and vitamin K were also
deficient. All the minerals except for chromium were
present in suboptimal amounts.34 Vitamin D, calcium, and
phosphorus are of particular concern because of osteoporo-
sis, fractures, and rickets reported in children receiving
 Ketogenic Diet for Pediatric Epilepsy / Zupec-Kania, Spellman
long-term antiepileptic drug therapies.35 Although it is not
known whether the KD demands certain nutrients, supple-
mentation is recommended to meet minimum needs.
Dietary reference intakes established by the Food and
Nutrition Board of the Institute of Medicine are used as ref-
erences for daily intake of micronutrients for the oral diet
and for liquid feedings such as infant and enteral formulas.
Monitoring of compliance to diet, fluids, and vitamin and
mineral supplementation are recommended to ensure
nutrition sufficiency. Monitoring of growth and laboratory
indices during the course of this therapy are recommended
outcome measures to ensure good nutrition status and to
prevent or correct the adverse effects of the diet.
Fluid restriction along with the KD has been a prac-
tice at some centers with the belief that a mild state of
dehydration enhances the effect of the diet; however,
this theory has never been proven. The diuretic effect of
carbohydrate-restricted diets is a known physiological
response. In addition, the limited volume of high-moisture
and high-fiber foods on the diet indicates a need to ensure
adequate hydration. Fluid education to the caregiver is
necessary to ensure his or her understanding of this
practice. Predictive equations for estimating fluid require-
ments may be used to determine daily needs such as the
Holida–Segar equation or dietary reference intake for
total water.
Adjusting the KD to improve abnormal lipids has not
been documented in the literature. A natural decline in
elevated lipids over time has been observed in our experi-
ence. Strategies that can be trialed to reduce persistently
high lipids include reducing the animal sources of cho-
lesterol and replacing saturated fats with monosaturated
and polyunsaturated fats. The inclusion of an ω-3 fat
source from foods or a high-quality supplemental source
may assist in increasing high-density lipoproteins. Lowering
the fat content of the diet by lowering the ratio is also an
option that may also be trialed.
There is limited evidence in the literature regarding
the use of carnitine in combination with the KD.
Carnitine is a nonessential amino acid that is involved in
the transfer of long-chain fatty acids into the mitochondria
for β-oxidation. There is a poor correlation of plasma car-
nitine levels in relation to tissues stores. The symptoms of
carnitine deficiency are difficult to discern from other
causes, that is, lethargy, anorexia, poor growth, hepatic
insufficiency, and elevated triglycerides. Secondary carni-
tine deficiencies have been reported to occur, with
patients receiving certain antiepileptic therapies including
valproates, phenobaritol, phenytoin, carbamazepine, fel-
bamate, and the KD.36,37 A consensus statement from a
panel of 9 pediatric neurologists suggested that carni-
tine supplementation is “strongly suggested” for certain
patients, including those receiving the KD who have
hypocarnitinemia.38
Downloaded from ncp.sagepub.com at UCSF LIBRARY & CKM on March 16, 2015
595
Implications
As healthcare providers, we are faced with decisions
regarding allocation of our resources. The KD is an MNT
in which dietitians have the opportunity to positively
affect patient outcome and to provide cost-effective care.
Despite the restrictiveness of this therapy, most families
are eager to try this option and their compliance is excep-
tional. Careful planning and inclusion of the appropriate
team members will help ensure a successful KD program.
The Charlie Foundation is a nonprofit organization that
exists to promote the safe use of the KD. This organiza-
tion has provided dietitians and the public with a variety
of educational tools, which are described on their Web
site.39 A surge of interest in the KD has spurned new
research and collaborative projects, which will hopefully
increase resources for healthcare professionals and pro-
vide wider availability of the KD and better outcomes to
the patients we serve.
References
1. Wheless JW, Ashwal S. The ketogenic diet. In: Swaiman KF,
Ashwal S, eds. Pediatric Neurology: Principles and Practice. Vol. 1.
3rd ed. St Louis, MO: Mosby; 1999:719-728.
2. Bough KJ, Rho JM. Anticonvulsant mechanisms of the ketogenic
diet. Epilepsia. 2007;48:43-58.
3. Todorova MT, Tandon P, Madore RA, Stafstrom CE, Seyfried TN.
The ketogenic diet inhibits epileptogenesis in EL mice: a genetic
model for idiopathic epilepsy. Epilepsia. 2000;41:933-940.
4. KetoCalculator Web site. https://www.ketocalculator.com. Accessed
July 20, 2008.
5. Huttenlocher PR, Wilbourn AJ, Signore JM. Medium-chain triglyc-
erides as a therapy for intractable childhood epilepsy. Neurology.
1971;21:1097-1103.
6. Wang HS, Hung PC, Lin KL. Coconut oil is acceptable in replac-
ing medium chain triglycerides for ketogenic diet therapy in child-
hood epilepsy. Paper presented at: International Symposium on
Dietary Therapies for Epilepsy and Other Neurological Disorders;
April 2-5, 2008; Phoenix, AZ; Abstract 23.
7. Pfeifer H, Thiele EA. Low-glycemic-index treatment: a liberalized
ketogenic diet for treatment of intractable epilepsy. Neurology.
2005;65:1810-1812.
8. Kossoff EH, McGrogan JR, Bluml RM, Pillas DJ, Rubenstein JE,
Vining EP. A modified Atkins diet is effective for the treatment of
intractable pediatric epilepsy. Epilepsia. 2006;47:421-424.
9. Kossoff EH, Zupec-Kania BA, Amark PE, et al. Optimal clinical
management of children receiving the ketogenic diet: recommen-
dations of the international ketogenic diet study group. Epilepsia.
In press.
10. Laux LC, Devonshire KA, Kelley KR, Goldstein J, Nordli DR Jr.
Efficacy of the ketogenic diet in myoclonic epilepsy of Doose.
Epilepsia. 2004;45(suppl 7):251.
11. Kossoff EH, Thiele EA, Pfeifer HH, McGrogan JR, Freeman JM.
Tuberous sclerosis complex and the ketogenic diet. Epilepsia.
2005;46:1684-1686.
12. Giampietro PF, Schowalter DB, Merchant S, Campbell LR, Swink T,
Roa BB. Widened clinical spectrum of the Q128P MECP2 mutation
in Rett syndrome. Childs Nerv Syst. 2006;22:320-324.
596 Nutrition in Clinical Practice / Vol. 23, No. 6, December 2008
13. Caraballo RH, Cersosimo RO, Sakr D, Cresta A, Escobal N,
Fejerman N. Ketogenic diet in patients with Dravet syndrome.
Epilepsia. 2005;46:1539-1544.
14. Caraballo RH, Cersosimo RO, Sakr D, Cresta A, Escobal N,
Fejerman N. Ketogenic diet in patients with myoclonic-astatic
epilepsy. Epileptic Disord. 2006;8:151-155.
15. Nordli DR Jr, Kuroda MM, Carroll J, et al. Experience with the
ketogenic diet in infants. Pediatrics. 2001;108:129-133.
16. Klepper J, Leiendecker B. GLUT1 deficiency syndrome—2007
update. Dev Med Child Neurol. 2007;49:707-716.
17. Wexler ID, Hemalatha SG, McConnel J, et al. Outcome of pyru-
vate dehydrogenase deficiency treated with ketogenic diets.
Studies in patients with identical mutations. Neurology. 2001;49:
1655-1661.
18. Pulsifer MB, Gordon JM, Brandt J, Vining EP, Freeman JM.
Effects of ketogenic diet on development and behavior: prelimi-
nary report of a prospective study. Dev Med Child Neurol. 2001;
43:301-306.
19. Hallbook T, Kohler S, Rosen I, Lundgren J. Effects of ketogenic
diet on epileptiform activity in children with therapy resistant
epilepsy. Epilepsy Res. 2007;77:134-140.
20. Wheless JW. The ketogenic diet: an effective medical therapy with
side effects. J Child Neurol. 2001;16:633-635.
21. Ballaban-Gil K, Callahan C, O’Dell C, Pappo M, Moshé S, Shinnar S.
Complications of the ketogenic diet. Epilepsia. 1998;39:744-748.
22. Furth SL, Casey JC, Pyzik PL, et al. Risk factors for urolithiasis in
children on the ketogenic diet. Pediatr Nephrol. 2000;15:125-128.
23. Kielb S, Koo HP, Bloom DA, Faerber GJ. Nephrolithiasis associ-
ated with the ketogenic diet. J Urol. 2000;164:464-466.
24. Benn A, Swan CH, Cooke WT, Blair JA, Matty AJ, Smith ME.
Effect of intralumninal pH on the absorption of pteroylmonoglu-
tamic acid. Br Med J. 1971;1:148-150.
25. Kwiterovich PO Jr, Vining EP, Pyzik P, Skolasky R Jr, Freeman JM.
Effect of a high-fat ketogenic diet on plasma levels of lipids,
lipoproteins, and apolipoproteins in children. JAMA. 2003;290:
912-920.
26. Stewart WA, Gordon K, Camfield P. Acute pancreatitis causing
death in one child on the ketogenic diet. J Child Neurol. 2001;
16:682.
Downloaded from ncp.sagepub.com at UCSF LIBRARY & CKM on March 16, 2015
27. Berry-Kravis E, Booth G, Taylor A, Valentino LA. Bruising and the
ketogenic diet: evidence for diet-induced changes in platelet func-
tion. Ann Neurol. 2001;49:98-103.
28. MacCracken KA, Scalisi JC. Development and evaluation of a
ketogenic diet program. J Am Diet Assoc. 1999;99:1554-1558.
29. Mandel A, Ballew M, Pina-Garza JE, Stalmasek V, Clemens LE.
Medical costs are reduced in children with severe epilepsy who are
successfully maintained on a ketogenic diet. Future Dimens Clin
Nutr Manag. 1998;3:12-13.
30. Bergqvist AG, Schall JI, Gallagher PR, Cnaan A, Stallings VA.
Fasting versus gradual initiation of the ketogenic diet: a prospective,
randomized clinical trial of efficacy. Epilepsia. 2005;46:1810-1819.
31. Kim DW, Kang HC, Park JC, Kim HD. Benefits of the nonfasting
ketogenic diet compared with the initial fasting ketogenic diet.
Pediatrics. 2004;114:1627-1630.
32. Vining EP, Pyzik P, McGrogan J, et al. Growth of children on the
ketogenic diet. Dev Med Child Neurol. 2002;44:796-802.
33. Zupec-Kania B. Micronutrient composition of the classic ketogenic
diet. Paper presented at: International Symposium on Dietary
Therapies for Epilepsy and Other Neurological Disorders; April 2-
5, 2008; Phoenix, AZ; Abstract 24.
34. Gilbert DL, Pyzik PL, Freeman JM. The ketogenic diet: seizure
control correlates better with serum beta-hydroxybutryate than
with urine ketones. J Child Neurol. 2000;15:787-790.
35. Bergqvist AG, Schall JI, Stallings VA. Vitamin D status in children
with intractable epilepsy and impact of the ketogenic diet. Epilepsia.
2007;48:66-71.
36. Raskind JY, El-Chaar GM. The role of carnitine supplementation
during valproic acid therapy. Ann Pharmacother. 2000;34:630-638.
37. Coppola G, Epifanio G, Auricchio G, Federico R, Resocapt G,
Pascotto A. Plasma free carnitine in epilepsy children, adolescents
and young adults treated with old and new antiepileptic drugs with
or without the ketogenic diet. Brain Dev. 2006;28:358-365.
38. De Vivo DC, Bohan TP, Coulter DL, et al. L-carnitine supplemen-
tation in childhood epilepsy: current perspectives. Epilepsia. 1998;
39:1216-1225.
39. Abrahamas J. The Charlie Foundation to Help Cure Pediatric
Epilepsy. Charlie Foundation Web site. http://www.charliefoundation
.org. Accessed July 7, 2008.