e u r o p e a n j o u r n a l o f p a e d i a t r i c n e u r o l o g y 2 0 ( 2 0 1 6 ) 5 5 5 e5 5 9

Official Journal of the European Paediatric Neurology Society

Original Article

Valproate effect on ketosis in children under

ketogenic diet
Martha Spilioti a,b, Evangelos Pavlou a,b, Maria Gogou c,*,
Irene Katsanika c, Efimia Papadopoulou-Alataki c, Olga Grafakou c,
Anastasia Gkampeta a,b, Argyrios Dinopoulos d, Athanasios Evangeliou c
a

1st Department of Neurology, Aristotle University of Thessaloniki, University General Hospital AHEPA,
Thessaloniki, Greece
b
2nd Department of Pediatrics, Aristotle University of Thessaloniki, University General Hospital AHEPA,
Thessaloniki, Greece
c
4th Department of Pediatrics, Aristotle University of Thessaloniki, Papageorgiou General Hospital, Thessaloniki,
Greece
d
3rd Department of Pediatrics, University of Athens, Attikon Hospital, Athens, Greece

article info

abstract

Article history:

Introduction: Although ketogenic diet has been proven useful in the management of

Received 13 October 2015

intractable seizures, interactions with other medicines have been reported. This study

Received in revised form

reports two patients on co-administration with ketogenic diet and valproate appearing

28 March 2016

undesirable side effects after increase or decrease of valproate pharmaceutical levels.

Accepted 6 April 2016

Methods: Totally 75 patients suffering from drug-resistant epilepsy were treated with
ketogenic diet in our departments. Their age varied from 6 months to 9 years. All patients

Keywords:

were followed for at least 12 months and up to five years. Clinical and laboratory variables

Ketogenic diet

have been regularly assessed.

Valproate

Results: In 75 patients treated with ketogenic diet and valproate at the same time treatment

Children

was well tolerated. Two patients presented mild to moderate undesirable effects. In these

Side-effects

patients the removal of valproate treatment resulted in an increase of ketosis with

Epilepsy

respective clinical signs. The conversion of the diet from 4:1 to 1:1 and 2,5:1 respectively
resulted in reduction of ketosis and clinical improvement.
Conclusion: In the majority of cases co-administration of valproate and ketogenic diet seems
to be safe. In two cases, valproate appeared to have a negative effect on ketosis (and
weaning it led to over-ketosis). This interaction is worthy of future study.
2016 European Paediatric Neurology Society. Published by Elsevier Ltd. All rights
reserved.

Abbreviations: KD, ketogenic diet.

* Corresponding author. Dimitriou Nika 44, 60 100 Katerini, Greece. Tel.: 30 6978905873.
E-mail address: mariaangogou@gmail.com (M. Gogou).
http://dx.doi.org/10.1016/j.ejpn.2016.04.003
1090-3798/ 2016 European Paediatric Neurology Society. Published by Elsevier Ltd. All rights reserved.

556

1.

e u r o p e a n j o u r n a l o f p a e d i a t r i c n e u r o l o g y 2 0 ( 2 0 1 6 ) 5 5 5 e5 5 9

Introduction

Over the last decades, the Ketogenic Diet (KD) has become a
valuable weapon in the management of intractable
seizures.1e3
This treatment, however, is not free of side effects, which
mainly include complications arising directly from the diet, as
well as synergies of the diet with other medicines.4e7 One of
the most ambiguous synergies of KD is that with valproic acid.
Both valproic acid and KD affect the process of beta-oxidation.
Thus, co-administration of both could potentially result in an
overload of the beta-oxidation pathway with a subsequent
negative energy balance for the body.8e10 Although it has been
postulated that valproate and KD co-administration leads to
undesirable effects (e.g. increased toxicity of valproate,
carnitine deficiency), there are no clear clinical observations
confirming these assumptions.4,7,11e13 On the contrary, there
are clinical studies claiming that co-administration of valproate and KD is safe, as the observed hypocarnitinemia is
exclusively due to valproate and not the diet, and valproate
levels are not affected by KD.14,15 These findings are reinforced
by recent experimental work showing that acetone enhances
the anticonvulsant effects of valproate, without affecting the
pharmacokinetic and side-effect profiles.16
In the last four years we have had more than 70 patients on
KD in our Departments. The objective of this study is to
retrospectively assess the safety of KD and valproate coadministration and describe any adverse events. In the majority of cases the co-administration of KD and valproate was
safe. However, in two cases the co-administration seemed to
have some undesirable effects.

2.

Patients and methods

We conducted a retrospective study of the medical records of

children suffering from drug-resistant epilepsy, according to
ILAE criteria, co-treated with KD and valproate and followed at
the Department of Pediatric Neurology of the General Hospital
Papageorgiou (Thessaloniki), of the University General Hospital AHEPA (Thessaloniki) and the General Hospital Attikon
(Athens) during the period 1.1.2011e31.12.2013.
All patients were followed for at least 12 months and up to
five years. Initial assessment included full physical examination, laboratory work up (biochemical parameters, valproate
residual levels, plasma and urine amino acids, urine organic
acids, plasma carnitine esters) and documentation of seizure
activity as the number of seizures reported per week. All patients and parents provided consent to the use of their data in
our study and our study was approved by the Ethical Committee of the Aristotle University of Thessaloniki. Our work
has been carried out in accordance with the Code of Ethics of
the World Medical Association.
In those responding to the diet and showing a reduction of
epileptic activity at least of 70%, gradual reduction of the drugs
was initiated. The last drug to reduce or discontinue was that
with the greatest clinical benefit for the patient, according to
medical history of each patient. In every case of deterioration,
the same laboratory work up was performed. Safety was

assessed by the appearance of signs or symptoms related to

ketosis (weight loss, nausea, fatigue) and the change in lab
indices. All patients were substituted with carnitine
throughout their treatment with KD.

3.

Results

Totally 75 patients (41 males and 34 females) suffering from

drug-resistant epilepsy were treated with KD in our departments and also received valproate. Their age varied from
6 months to 9 years. At the beginning of KD application difficulties in maintaining ketosis were observed in some patients,
but no significant correlation with any of the antiepileptic
drugs used was detected. In 73 patients treated with KD and
valproate at the same time treatment was well tolerated
without any adverse events. There were, however, 2 patients
(2.6%) who presented undesirable effects. Their cases are reported below, while basic patient traits at the initiation of the
diet and their clinical course hallmarks are summarized in
Tables 1e3. Ketone levels are graphically presented in Fig. 1. It
is noteworthy that with regards to the other 73 patients, no
effects of valproate administration on ketosis were observed.

3.1.

Patient 1

A 4-year-old-girl with severe psychomotor retardation and

refractory epilepsy resulting from perinatal asphyxia was
treated with a 4:1 ketogenic diet (4 parts fat to 1 part carbohydrates and protein combined) KD. At the initiation of the diet
the patient was under valproate and zonisamide. The patient
responded positively to the diet and after six months without
seizures we started to discontinue gradually zonisamide. After
two years without seizures we started progressively to discontinue valproate. While valproate dose had been gradually
reduced by 50 mg every two weeks, no changes in fat ratio were
made and ketosis progressively increased. Levels of ketosis
were closely monitored with weekly assessments of bhydroxybutyrate. By the final removal of valproate a sharp
increase in ketosis was observed accompanied by clinical signs
(somnolence, apathy). By changing the diet ratio of 4:1 down to
2.5:1 the patient returned to her former condition.

3.2.

Patient 2

A 4-year-old-girl with drug resistant epilepsy was put on 4:1

ketogenic diet. The patient responded immediately to the diet
treatment and two weeks later she was seizure free. Three
months later we started to discontinue gradually the drugs,
firstly topiramate, then zonisamide and finally valproate.
After valproate cessation an increase in the level of ketosis
was observed with accompanying clinical signs, including dry
mouth with ketotic breath, lack of hunger, fatigue and somnolence. Therefore, in a stepwise progressive way we reduced
the diet ratio from 4:1 down to 1:1; firstly 3:1, then 2:1 and last
to a 1:1 ratio. At this point the adverse clinical signs disappeared and ketosis returned to its previous level when the
diet was combined with valproate. At present the patient has
experienced 6 months on a 1:1 KD ratio with no recurrence of
seizures.

557

e u r o p e a n j o u r n a l o f p a e d i a t r i c n e u r o l o g y 2 0 ( 2 0 1 6 ) 5 5 5 e5 5 9

Table 1 e Patients with interactions between Ketogenic diet (KD) and valproate (VPA). This table summarizes basic
epidemiological and clinical traits of our sample at the initiation of KD (TPM: topiramate, ZND: zonisamide).
Diagnosis

Age at KD
initiation (years)

Weight at KD
initiation (kg)

Treatment at KD
diet initiation

VPA dose
(mg/kg/day)

VPA levels
(mg/ml)

KD ratio at
initiation

Patient 1

Lennox-Gastaut

15

30

80

4:1

Patient 2

Complex-partial

17

ZND
VPA
TPM
ZND
VPA

35

80

4:1

Table 2 e Clinical course of patient 1 (VPA: valproate, KD: ketogenic diet, AED: antiepileptic drugs, b-OH-b: bhydroxybutyric acid).
Month

KD ratio

VPA dose (mg/kg/day)

Seizure status

Other AED

Weight (kg)

b-OH-b (mg/dL)

4:1
4:1
4:1
4:1
4:1
2.5:1

30
16
10
e
e
e

seizure free
seizure free
seizure free
seizure free
seizure free
seizure free

zonisamide
e
e
e
e
e

15
15
17
19
19
19

2.4
2.2
2.4
3
3.7
2.6

3
6
12
24
27
28

Table 3 e Clinical course of patient 2 (VPA: valproate, KD: ketogenic diet, AED: antiepileptic drugs, b-OH-b: bhydroxybutyric acid).
Month

KD ratio

VPA dose (mg/kg/day)

Seizure status

Other AED

Weight (kg)

b-OH-b (mg/dL)

4:1
4:1
4:1
4:1
4:1
4:1
2:1
1:1
1:1

35
35
35
35
e
e
e
e
e

seizure free
seizure free
seizure free
seizure free
seizure free
seizure free
seizure free
seizure free
seizure free

topiramate, zonisamide
topiramate, zonisamide
zonisamide
e
e
e
e
e
e

17
17
17
17
17
17
17
17
18

2.8
2.8
2.8
2.7
2.9
3.9
3
2.8
2.7

3
6
8
10
12
12
13
13
24

4.

Discussion

In this study we present our experience about coadministration of KD and valproate. In the vast majority this
co-administration is safe and only two of our patients presented undesirable effects. In these two patients we had

ketone levels
(mg/dL)
4
3.5
3
Paent 1
2.5

Paent 2

2
1.5
0

10

15

20

25

30
months aer
KD iniaon

Fig. 1 e Ketone levels in our 2 patients during their followup.

evidence of the inhibitory effect of valproic acid on ketone

body production.
Certainly, in these patients it is possible that the valproic
acid had some inhibitory effect on beta-oxidation and consequently on the production of ketone bodies. Removal of valproate stopped any inhibitory action on beta-oxidation and
increased ketone bodies. These indications are supported by
clinical and experimental work which suggests that a principal cause of liver failure associated with valproic acid therapy is likely the inhibition of hepatic mitochondrial betaoxidation, known to be the main source of ketone bodies.17e20
Moreover, in these patients the interesting fact is that they
had sufficient levels of ketosis at a relatively low carbohydrate
relationship protein/fat ratio. At the age of 4 years patients
usually reach satisfactory levels of ketosis at a ratio of at least
3:1 if not 4:1. With our data is difficult to find a logical explanation for this observation. Perhaps, the specific characteristics of metabolism of each individual or alternatively the
possibility of a counterbalancing overproduction of ketone
bodies in order to meet the energy deficit of the brain should
be speculated.
Our patients show putative interactions between valproic
acid and KD. In general, it is interesting to note such interactions in epileptic individuals seem to exist in separate

558

e u r o p e a n j o u r n a l o f p a e d i a t r i c n e u r o l o g y 2 0 ( 2 0 1 6 ) 5 5 5 e5 5 9

tissue systems and in different biochemical pathways. First of

all, it is recognized from experimental and clinical evidence
that valproate is a well-known inhibitor of hepatic beta
oxidation, where the ketone bodies are mainly produced.19,21
Besides, experimental and clinical data suggest that both ketone bodies and valproic acid use the monocarboxylate
transporter 1 to cross the blood brain barrier.10,22,23 Thus,
increased levels of valproate may lead to a competition with
ketone bodies to cross the blood brain barrier with a consequent decreased entry of ketone bodies into the brain. Finally,
a negative effect from valproate on energy production by ketones could also account for undesirable effects in some
cases.24 All the above mechanisms are graphically illustrated
in Fig. 2. In our 2 patients we may have interactions between
valproic acid and ketogenic diet in location 1.
The question arising is why our patients developed this
problem, which is unobserved in the majority of patients. This
is a question difficult to answer requiring intensive research.
The KD and antiepileptic medications are complex in their actions and processes and their use in combination, and perhaps
with other drugs, could be influenced by the individual characteristics of the intermediary metabolism of each person.4

LIVER

Previous studies about safety of KD in childhood report

generally low frequencies of adverse events. According to
Ballaban-Gil et al. (1998), the vast majority of children on KD
who experienced adverse effects were on simultaneous
treatment with valproate (80%) and these effects mainly
included serious and life-threatening conditions (severe
hypoproteinemia, Fanconi's renal tubular acidosis, liver steatosis).4 On the other hand, Lyczkowski et al. (2005) detected no
significant differences in adverse-events profiles between
children on KD receiving valproate and those not receiving. In
this study no life-threatening conditions were observed in the
group receiving valproate.25
The basic limitation of our study is its retrospective design,
which does not permit to draw higher evidence level conclusions. Longitudinal prospective trials on this topic could shed
more light in KD and valproate interactions. Furthermore, the
fact that only 2 patients presented adverse events did not
allow us to perform statistical comparisons and investigate
possible risk-factors. However, it should be emphasized that
only 2 patients in a total of 75 (2.6%) showed undesirable interactions between KD and valproate. The severity of the
observed adverse events varied from mild to moderate

BBB

Beta-oxidation

2
MCT1- receptors

Ketone bodies

Ketone bodies
Ketonbodies
Acetyl-CoA

TCA
cycle

MCT1- receptors

NADH+

Respiratory chain

Energy

Fig. 2 e Simplified schema with numbers representing possible sites of interaction between valproic acid and ketone bodies.
1 beta-oxidation, 2 monocarboxylate transporter 1, 3 tricarboxylic acid cycle. 1 and 2: possible locations for
competitive interaction between valproic acid and ketone bodies. 3: possible location for direct valproate action without
competition with ketone bodies; here, as a consequence of valproate action, decreased energy production through
inhibition of the TCA cycle could occur and therefore defective combustion of ketone bodies. In our 2 patients we may have
interactions between valproic acid and ketogenic diet in location 1. Abbreviations: BBB Blood brain barrier,
MCT1 monocarboxylate transporter 1, TCA tricarboxylic acid cycle.

e u r o p e a n j o u r n a l o f p a e d i a t r i c n e u r o l o g y 2 0 ( 2 0 1 6 ) 5 5 5 e5 5 9

medical complications, but did not include any lifethreatening conditions.

9.

5.

Conclusion

In conclusion, we should be aware that, although in the majority of cases the co-administration of valproate and KD
seems to be safe, in very few cases a fine tuning is required in
order to avoid any undesirable effects. More specifically, if a
patient is over-ketotic when valproate is weaned, lowering the
ratio of the diet seems to decrease ketosis to the desired levels.

Ethical standards
Our work included no experimentation involving human or/
and animals.

10.

11.

12.

13.

14.

Financial support
This research received no specific grant from any funding
agency, commercial or not-for-profit sectors.

15.

16.

Conflicts of interest
17.

None.
18.

Acknowledgments
None.

19.

references
20.
1. Kim do Y, Rho JM. The ketogenic diet and epilepsy. Curr Opin
Clin Nutr Metab Care 2008;11:113e20.
2. Cross JH, McLellan A, Neal EG, Philip S, Williams E,
Williams RE. The ketogenic diet in childhood epilepsy: where
are we now? Arch Dis Child 2010;95:550e3.
3. Freeman JM, Kossoff EH. Ketosis and the ketogenic diet, 2010:
advances in treating epilepsy and other disorders. Adv Pediatr
2010;57:315e29.

S,
4. Ballaban-Gil K, Callahan C, O'Dell C, Pappo M, Moshe
Shinnar S. Complications of the ketogenic diet. Epilepsia
1998;39:744e8.
5. Wilong AA. Complications and consequences of epilepsy
surgery, ketogenic diet, and vagus nerve stimulation. Semin
Pediatr Neurol 2007;14:201e3.
6. Kang HC, Chung DE, Kim DW, Kim HD. Early- and late-onset
complications of the ketogenic diet for intractable epilepsy.
Epilepsia 2004;45:1116e23.
7. Bough KJ, Eagles DA. Comparison of the anticonvulsant
efficacies and neurotoxic effects of valproic acid, phenytoin,
and the ketogenic diet. Epilepsia 2001;42:1345e53.
8. Silva MF, Ruiter JP, Overmars H, Bootsma AH, van Gennip AH,
Jakobs C, et al. Complete beta-oxidation of valproate: cleavage

21.

22.

23.

24.

25.

559

of 3-oxovalproyl-CoA by a mitochondrial 3-oxoacyl-CoA

thiolase. Biochem J 2002;362:755e60.
Veitch K, Draye JP, Van Hoof F. Inhibition of mitochondrial
beta-oxidation and peroxisomal stimulation in rodent livers
by valproate. Biochem Soc Trans 1989;17:1070e1.
Bentourkia M, Tremblay S, Pifferi F, Rousseau J, Lecomte R,
Cunnane S. PET study of 11C-acetoacetate kinetics in rat
brain during dietary treatments affecting ketosis. Am J Physiol
Endocrinol Metab 2009;29:796e801.
Erickson JC, Jabbari B, Difazio MP. Basal ganglia injury as a
complication of the ketogenic diet. Mov Disord
2003;18:448e51.
cklin B, Reithofer E, Benninger F, Freilinger M,
Dressler A, Sto
Hauser E, et al. Long-term outcome and tolerability of the
ketogenic diet in drug-resistant childhood epilepsy e the
Austrian experience. Seizure 2010;19:404e8.
Coppola G, Verrotti A, D'Aniello A, Arcieri S, Operto FF, Della
Corte R, et al. Valproic acid and phenobarbital blood levels
during the first month of treatment with the ketogenic diet.
Acta Neurol Scand 2010;122:303e7.
Coppola G, Epifanio G, Auricchio G, Federico RR, Resicato G,
Pascotto A. Plasma free carnitine in epilepsy children,
adolescents and young adults treated with old and new
antiepileptic drugs with or without ketogenic diet. Brain Dev
2006;28:358e65.
Dahlin MG, Beck OM, Amark PE. Plasma levels of antiepileptic
drugs in children on the ketogenic diet. Pediatr Neurol
2006;35:6e10.
Zarnowska I, Luszczki JJ, Zarnowski T, Buszewicz G, Madro R,
Czuczwar SJ, et al. Simultaneous toxicities in a child on
multiple anticonvulsants. J Child Neurol 2008;23:1054e7.
nig S. Oral valproic acid for epilepsy e
Gerstner T, Bell N, Ko
long-term experience in therapy and side effects. Expert Opin
Pharmacother 2008;9:285e92.
Ghozzi H, Hakim A, Sahnoun Z, Ben Mahmoud L,
Atheymen R, Hammami S, et al. Relationship between
plasma concentrations of valproic acid and hepatotoxicity in
patients receiving high doses. Rev Neurol (Paris)
2011;167:600e6.
Silva MF, Aires CC, Luis PB, Ruiter JP, Ijlst L, Duran M, et al.
Valproic acid metabolism and its effects on mitochondrial
fatty acid oxidation: a review. J Inherit Metab Dis
2008;31:205e16.
McPherson PA, McEneny J. The biochemistry of ketogenesis
and its role in weight management, neurological disease and
oxidative stress. J Physiol Biochem 2012;68:141e51.
Bjorge SM, Baillie TA. Studies on the beta-oxidation of
valproic acid in rat liver mitochondrial preparations. Drug
Metab Dispos 1991;19:823e9.
Fischer W, Praetor K, Metzner L, Neubert RH, Brandsch M.
Transport of valproate at intestinal epithelial (Caco-2) and
brain endothelial (RBE4) cells: mechanism and substrate
specificity. Eur J Pharm Biopharm 2008;70:486e92.
kli M, Merezhinskaya N, Tervo T,
Vellonen KS, Ha
Honkakoski P, Urtti A. Monocarboxylate transport in human
corneal epithelium and cell lines. Eur J Pharm Sci
2010;39:241e7.
Lus PB, Ruiter JP, Aires CC, Soveral G, de Almeida IT,
Duran M, et al. Valproic acid metabolites inhibit dihydrolipoyl
dehydrogenase activity leading to impaired 2-oxoglutaratedriven oxidative phosphorylation. Biochim Biophys Acta
2007;1767:1126e33.
Lyczkowski DA, Pfeifer HH, Ghosh S, Thiele EA. Safety and
tolerability of the ketogenic diet in pediatric epilepsy:
effects of valproate combination therapy. Epilepsia
2005;46:1533e8.