Management of Epilepsy with Ketogenic Diet

Abstract

Intractable epilepsy has been defined as epilepsy that failed to respond to 3 or more anti-epileptic drugs (AED). About 30% of children with epilepsy experience uncontrolled seizures or side effects due to AED. The ketogenic diet (KD) was found to be effective in treatment of intractable seizures. The KD consists of a high fat, low carbohydrate and adequate protein. The KD need to be tailored individually for every patient. Different protocols and variable dosing of KD have been used. The KD is prescribed by physician and dietician, thereafter side effects such as acidosis, hypoglycemia, vomiting, gastro-esophageal reflux, constipation, hyper-lipidemia, renal stone, growth failure, bone density, micro-nutrient and vitamin deficiency need to be monitored.

Full text

Cite: Jain S. Management of Epilepsy with Ketogenic Diet. Indian Journal of Psychosocial Sciences. 2017
Apr; 7(1):15-20
Management of Epilepsy with Ketogenic Diet
Shobhit Jain
Department of Psychiatry, Institute of Medical Sciences, Banaras Hindu University, Varanasi, INDIA
E-mail: dr.shobhitjain@yahoo.in
Introduction
Intractable epilepsy has been defined as epilepsy that failed to respond to 3 or more anti-epileptic drugs
(AED). About 30% of children with epilepsy experience uncontrolled seizures or side effects due to AED
(1).
The ketogenic diet (KD) was found to be effective in treatment of intractable seizures. The KD consists of
a high fat, low carbohydrate and adequate protein. The KD need to be tailored individually for every
patient. Different protocols and variable dosing of KD have been used. The KD is prescribed by physician
and dietician, thereafter side effects such as acidosis, hypoglycemia, vomiting, gastro-esophageal reflux,
constipation, hyper-lipidemia, renal stone, growth failure, bone density, micro-nutrient and vitamin
deficiency need to be monitored.
History of Ketogenic Diet in Epilepsy
Fasting has been recognized since Hippocratic era (460BC-370BC) as treatment for epilepsy. In early 20th
Century, French and American physicians like Guelpa, Marie, Conklin, and Geyelin started research on
fasting and starvation as treatment for epilepsy (1). It was shown that the fasting is more effective more
in treating children than adolescents and its efficacy decreases with increasing age. Thereafter, Lennox
and Cobb (1922) explained dehydration, acidosis, and ketosis as possible mechanism by which fasting
helps in treating epilepsy.
Dr. Wilder proposed that the diet which produces ketosis could also be used in treatment of epilepsy.
He termed such diet as “Ketogenic Diet”, which were rich in fat and deficient in carbohydrate. He found
benefit of KD over starvation in providing similar efficacy, but can be used for prolonged maintenance.
Later, in 1938 diphenylhydantoin was discovered and thereafter research focus shifted to development
of other AED, which were found convenient to administer.

Due to intractable seizures which remained uncontrolled despite of several trial of AEDs, KD regained its
popularity. In 1972, Dr. Robert C. Atkins promoted “Atkins Diet” which consisted of high fat and low
carbohydrate and produces ketosis. Thereafter, “Modified Atkins Diet” was developed in 2003 at John
Hopkins. These diets were more palatable and tolerable than previous ones.
Subsequently, low glycemic index treatment diet was proposed in 2005, which hypothesis that stable
blood glucose at a lower level would result in modulation of insulin release and other metabolic effects,
thus improving seizure.
Mechanism of Action
The exact mechanism of action of ketogenic diet is still not fully clear. However, preclinical studies have
suggested following possible mechanism as alterations in mitochondrial function, effect of ketone
bodies and fatty acids, and glucose stabilization. Ketone bodies easily cross the blood brain barrier, and
are only source of energy to brain during starvation. Substitution of glucose by ketones in brain results
in decreased glycolysis and increased krebs cycle for energy production. Ketone bodies causes γ-
aminobutyric acid (GABA) synthesis, neuronal membrane hyperpolarization, decrease release of
glutamate, increase norepinephrine and adenosine, neuroprotective and antioxidant activity, decrease
insulin-like growth factor 1 (IGF-1) and the mammalian target of rapamycin (mTOR), and increase
sirtuins and adenosine monophosphate-activated protein kinase (AMPK) in brain. The GABA is
synthesized during krebs’s cycle when glutamate is converted into GABA by enzyme glutamate
decarboxylase. Studies have shown that beta-hydroxybutyrate (ketone body) inhibits GABA-
transaminase expression thus increases GABA in brain. Ketone bodies regulates activity of these enzyme
thus increases the level of GABA in the brain. Increased level of GABA results in stimulation of chloride
receptors and ATP sensitive potassium channel leading to hyperpolarisation of neuronal membrane and
inhibition of sodium and calcium channels leading to decreased excitability. Studies have shown ketone
bodies (acetoacetate more than betahydroxybutyrate) affect VGLUT channels on presynaptic glutamate
vesicles (via chloride channels), thus leading decreased release of excitatory glutamate
neurotransmitters. Ketone bodies also cause alteration in level of biogenic amines such as increase in
nor-epinephrine and adenosine, and decrease in dopamine and serotonin. Increased adenosine causes
activation of adenosine A1 receptors resulting in decreased brain excitability. Ketone bodies have
neuroprotective role by causing mitochondrial biogenesis. It also have antioxidant role by increasing
uncoupling protein in mitochondrial electron transport chain thus reducing reactive oxygen species and

increasing level of mitochondrial reduced glutathione level thus protecting mitochondrial level from
oxidative stress. Increase in sirtuins causes increase in mitochondrial number and size, and also cause
decrease in insulin like growth factor-1. Increase in AMPK is directly related to ATP production. Ketone
bodies by inhibiting mTOR pathway have anticonvulsant action. Since ketone bodies metabolizing
enzyme (namely, monocarboxylic acid transporter) is found more abundant in infants and children,
which later decreases with increasing age and adults. Therefore the KD is more effective during infants
and childhood than in adulthood (2).
The Classical Ketogenic Diet
Most of classical ketogenic diet consist of higher fat than carbohydrate and protein, in such a ratio that
the ratio of quantity of fat and the sum of quantity of carbohydrate and protein is in range of 3:1 to 4:1.
Therefore, 90% of the total calories delivered in KD is by fat and the remaining 10% is by carbohydrate
and protein. The total calories is restricted to 80 to 90% of daily recommendation for the age and the
fluid restriction is 90% of previous dietary intake. The KD is preferably started after admitting patient in
hospital. Whereas, out-patient setting is not a contraindication, in-door setting has several advantages
as it provides better opportunity to thoroughly examine and evaluate the case for diagnosis and
contraindications of KD, provide education session for parents about KD preparation and administration,
monitoring of side effects, and also some centers initiate 4:1 preparation of classical KD after brief
fasting to provide accelerated ketosis. However, gradual initiation of KD is better tolerated, lesser side
effects, similar level of maintenance ketosis, and similar level of seizure control compared with fasting
KD initiation. Therefore, gradual initiation of KD should be preferred over fasting KD initiation. During
gradual initiation, initially 2:1 combination of KD is started which is later increased to 3:1 and 4:1 based
on tolerability (2). Most of patients respond well to one year treatment with KD, averaging about 80%
response rate. Among responders, more than 50% response is noticed in initial 3 months, and among
those with less than 50% response in initial 3 months, not much improvement is noticed thereafter. The
average time for treatment with KD is 2years after which it may be discontinued (2). As a clinical
practice, usually KD is used as a reserved therapy for intractable seizures. The International Ketogenic
Diet Study Group, the therapy should be offered after failure of two AED trial. However, KD remains
treatment of choice for GLUT1 deficiency syndrome and pyruvate dehydrogenase deficiency syndrome.
Whereas, for Dravet Syndrome (myoclonic epilepsy of infancy) and Doose Syndrome (myoclonic-atonic
epilepsy), Infantile Spasm/West Syndrome, Lennox Gestaut Syndrome, KD have been found more
effective and hence can be considered earlier. The adverse effects reported for KD are unpalatability,

taste change, nausea, vomiting, abdominal pain, constipation, diarrhea, infection, dehydration, hunger,
lack of energy, mood changes. Pancreatitis though rare but serious and found in association with
triglyceridemia among patients on KD. Other long term complications are growth retardation in children,
electrolyte, vitamins, and mineral deficiency, hyperlipidemia, nephrolithiasis (3).
The KD is contraindicated in patients with disorder of disorders of fatty acid transport and oxidation.
Therefore, developmental delay, cardiomyopathy, hypotonia, exercise intolerance, myoglobinuria, and
easy fatigability are few features that should raise high suspicion of fatty acid metabolism disorder, and
patients with such feature should undergo screening test for inborn error of metabolism before
initiation of KD therapy. The use of KD with valproate may cause liver failure. The studies on efficacy of
KD have been summarized in table 1.
Study
Findings
Vining et al. 1998
(4)
51 children with
average 230
seizures/month.
 Continued treatment for a year: 47%
 Efficacy in terms of seizure control after 1 year treatment:
>90% response – 43%; 50-90% response – 39%; <50% response – 17%
 Adverse effects: lethargy, severe dehydration, acidosis, mood changes,
increased infections, constipation, and vomiting.
 Reason for discontinuation: intolerance, difficulty of maintaining the
restrictive diet, and inadequate seizure control.
Freeman et al. 1998
(5)
150 children with
average 410
seizures/month.
 Efficacy in terms of seizure control after 1 year treatment:
100% response - 7%; >90% response - 27%; 50-90% response - 50%
 More than 50% improvement in initial 3 months responded well,
whereas less than 50% response in initial 3 months did not had much
improvement later.
Freitas et al. 2007
(6)
70 children with
refractory epilepsy
 Continued treatment for a year: 55%
 Efficacy in terms of seizure control after 1 year treatment:
>75% response - 70%; 75-25% response - 25%; <25% response – 2.5%
 Reason for discontinuation: unpalatability, nausea, and vomiting
 More effective for generalized epilepsy than for partial epilepsy
Neil et al. 2008 (7)
Randomised
 Efficacy in terms of seizure control after 3 months of treatment:
 KD Group (n = 54) had 75% reduction in seizure frequency compared

Controlled Trial on
103 children with
refractory epilepsy
to Control (AED) Group (n = 49).
 > 90% response – 7%; 50-90% response – 38%
 Adverse effect: Constipation, vomiting, lack of energy, hunger,
diarrhea, abdominal pain, and taste problems among 25% KD Group.
Table 1: Summary of studies on efficacy of Ketogenic Diet.
The Medium Chain Triglyceride (MCT) Diet
Since unpalatability was one of the major reason for discontinuation of use, MCT diet limited this
adverse effect to some extent with a greater proportion of carbohydrate and protein content than KD.
The MCT diet are more rapidly metabolized compared to long chain triglycerides in classical KD,
therefore has several advantages over classical KD. It produces more ketosis per kilocalorie of energy
and requires lesser amount of fat intake than classical KD. It also has lesser total cholesterol to high
density lipoprotein ratio compared to classical KD. Although, it has adverse effects as diarrhea, vomiting,
bloating, and abdominal pain. Therefore to reduce these adverse effects, the modified MCT diet have
been developed. Instead of providing 60% total energy with MCT diet, the modified MCT diet provides
30% of energy with MCT and 30% energy with long chain triglycerides. However, contraindications for
MCT diet are almost similar to KD.
The Modified Atkins Diet (MAD)
The MAD in composed of fat, protein , and carbohydrate is a ratio such that 60-65% calories is delivered
by fat, 30% by protein, and the remaining 10% by carbohydrate. For children, the net carbohydrate are
initially limited to 10 g/day which is later on increased to 20 g/day, whereas in adults the initial
carbohydrate dose is started at 15 g/day which is later on increased to 20-30 g/day. Goal of MAD is to
increase the ketosis without causing weight loss, therefore only carbohydrate consumption is restricted
but fat and protein consumption is encouraged. There is no fasting required, calories are not restricted,
weighing of food is not required, lesser education to family members and patient required and can be
safely started at home.
Study
Findings
Sharma et al. 2013
(8)
Randomised
 Efficacy in terms of seizure control after 3 months of treatment:
 MAD Group (n= 50) had significant reduction in seizure frequency

controlled trial
comparing MAD
and AED on 102
children and
adolescents with
refractory epilepsy.
compared to Control (AED) Group (n = 52).
 The MAD was well tolerated and did not required discontinuation,
though constipation was most common side effect.
Table 2: Summary of study on efficacy of Modified Atkins Diet.
The Low Glycemic Index Treatment (LGIT)
The low glycemic index (GI) food has lesser tendency to elevate postprandial blood sugar level. The food
with GI less than 50 is preferred, with total carbohydrate 40-60 g/day initially, however, fat and protein
is encouraged. The LGIT produces lower level of ketosis than the classical KD. The key findings of studies
on efficacy is summarized in table 3.
Study
Findings
Muzykewicz et al.
2009 (9)
76 children with
refractory epilepsy
 Efficacy in terms of seizure control after 1 year treatment: >50%
response - 66%
 Only 3 patients reported transient lethargy as side effect.
 Most common reason for discontinuation was restrictiveness.
Coppolla et al. 2011
(10)
15 children and
young adults with
refractory epilepsy
 Efficacy in terms of seizure control after 1 year of treatment:
 75-90% response - 40%; 50% response - 13%; <50% response – 47%
 No adverse effect reported.
Table 3: Summary of studies on efficacy of Low Glycemic Index Treatment.
Conclusion
Though, the classical KD is proven treatment for intractable epilepsy, the newer diet have been
developed in order to reduce the side effects and making it more palatable for prolonged maintenance
use. Despite having fewer side effects and better tolerability, the studies have found these newer diets
to be comparably effective as the classical KD (Table 4). Based on understanding about mechanism of
ketogenic diet, the recent researches are focusing to develop newer dietary alternatives eg. pyruvate,

triheptanoin, alpha-ketoglutarate, succinate, oxaloacetate supplementation. The development of future
diet for epilepsy may reduce side effects of classical KD and increase palatability, tolerability, and
compliance.
Study
Findings
Neal et al. 2009 (11)
Randomised
controlled trial
comparing MCT and
classical KD.
 No difference among efficacy of the MCT diet (n = 49) and the KD (n =
45).
Martin et al. 2016
(12)
Cochrane review of
7 Randomised
Controlled Trial
among children and
adolescents with
epilepsy.
 Efficacy of classical KD: 85% seizure reduction and 55% seizure free.
 Efficacy of MAD: 60% seizure reduction and 10% seizure free.
 Gastrointestinal symptoms (diarrhoea, constipation and vomiting) were
among most commonly reported side effects, and also one of the most
common reason for discontinuation. Whereas, cardiovascular adverse
effects were among most common long term side effects.
 Classical KD had greater side effects.
Klein et al. 2014
(13)
Review of studies
on KD and MAD
among adults with
refractory epilepsy.
 Efficacy of classical KD: >50% response – 32% and >90% response –9%.
 Efficacy of MAD: >50% response – 29% and >90% response –5%.
 Side effects of both diets are benign and similar. The most common
being weight loss and most serious being reversible hyperlipidemia.
 Refusal to participate because of diet restrictiveness and complexity is
more for KD than MAD. Also, discontinuation rate is more among KD
(51%) than MAD (42%).
Ye et al. 2015 (14)
Meta-analysis
comparing efficacy
of KD (n = 168),
MAD (n = 87), MCT
+ KD (n = 15) among
270 adults with
 The KD (52%) had significantly higher efficacy than MAD (34%).
 The overall compliance rate of KD(38%) was significantly lower than
MAD (56%).
 Due to better compliance rate MAD may be initial treatment of choice
for adults, which may be switched to classical KD if greater seizure
control is desired.

intractable epilepsy.
Table 4: Comparing of efficacy of Ketogenic Diet with newer variants.
References
1. Sampaio LP de B. Ketogenic diet for epilepsy treatment. Arq Neuropsiquiatr. 2016;74(10):842–848.
2. Lima PA de, Sampaio LP de B, Damasceno NRT. Neurobiochemical mechanisms of a ketogenic diet
in refractory epilepsy. Clinics. 2014;69(10):699–705.
3. Vaccarezza MM, Silva WH. Dietary therapy is not the best option for refractory nonsurgical
epilepsy. Epilepsia. 2015;56(9):1330–1334.
4. Vining EPG, Freeman JM, Ballaban-Gil K, Camfield CS, Camfield PR, Holmes GL, et al. A Multicenter
Study of the Efficacy of the Ketogenic Diet. Arch Neurol. 1998 Nov 1;55(11):1433–7.
5. Freeman JM, Vining EP, Pillas DJ, Pyzik PL, Casey JC, others. The efficacy of the ketogenic diet—
1998: a prospective evaluation of intervention in 150 children. Pediatrics. 1998;102(6):1358–1363.
6. Freitas A, Paz JA da, Casella EB, Marques-Dias MJ. Ketogenic diet for the treatment of refractory
epilepsy: a 10 year experience in children. Arq Neuropsiquiatr. 2007 Jun;65(2B):381–4.
7. Neal EG, Chaffe H, Schwartz RH, Lawson MS, Edwards N, Fitzsimmons G, et al. The ketogenic diet
for the treatment of childhood epilepsy: a randomised controlled trial. Lancet Neurol.
2008;7(6):500–506.
8. Sharma S, Sankhyan N, Gulati S, Agarwala A. Use of the modified Atkins diet for treatment of
refractory childhood epilepsy: a randomized controlled trial. Epilepsia. 2013;54(3):481–486.
9. Muzykewicz DA, Lyczkowski DA, Memon N, Conant KD, Pfeifer HH, Thiele EA. Efficacy, safety, and
tolerability of the low glycemic index treatment in pediatric epilepsy. Epilepsia. 2009;50(5):1118–
1126.
10. Coppola G, D’Aniello A, Messana T, Di Pasquale F, della Corte R, Pascotto A, et al. Low glycemic
index diet in children and young adults with refractory epilepsy: First Italian experience. Seizure.
2011 Sep;20(7):526–8.
11. Neal EG, Chaffe H, Schwartz RH, Lawson MS, Edwards N, Fitzsimmons G, et al. A randomized trial
of classical and medium-chain triglyceride ketogenic diets in the treatment of childhood epilepsy.
Epilepsia. 2009;50(5):1109–1117.
12. Martin K, Jackson CF, Levy RG, Cooper PN. Ketogenic diet and other dietary treatments for
epilepsy. Cochrane Libr [Internet]. 2016 [cited 2017 Feb 11]; Available from:
http://onlinelibrary.wiley.com/doi/10.1002/14651858.CD001903.pub3/pdf
13. Klein P, Tyrlikova I, Mathews GC. Dietary treatment in adults with refractory epilepsy A review.
Neurology. 2014;83(21):1978–1985.

14. Ye F, Li X-J, Jiang W-L, Sun H-B, Liu J. Efficacy of and patient compliance with a ketogenic diet in
adults with intractable epilepsy: a meta-analysis. J Clin Neurol. 2015;11(1):26–31.