Epilepsia, 46(9):1539–1544, 2005
Blackwell Publishing, Inc.
C ?2005 International League Against Epilepsy
Ketogenic Diet in Patients with Dravet Syndrome
Roberto Horacio Caraballo, Ricardo Oscar Cers´ osimo, Diego Sakr, Araceli Cresta, Nidia Escobal,
and Natalio Fejerman
∗Servicio de Neurolog´ ıa, Hospital de Pediatr´ ıa “Prof Dr Juan P Garrahan,” Buenos Aires, Argentina
Summary: Purpose: The ketogenic diet (KD) has been used
as a therapeutic alternative to antiepileptic drugs (AEDs) for re-
In this retrospective study, we evaluated the efficacy and tolera-
bility of the KD in patients with diagnostic criteria of DS.
Methods: Between March 1, 1990, and August 31, 2004, 52
patients who met diagnostic criteria for DS were enrolled in a
study at our department. Twenty of them were placed on the KD
with the Hopkins protocol and followed up for a minimum of 1
12 months, four children for 2 years, four children for 3 years,
and two children for 4 years. One year after initiating the diet,
13 (65%) of the initial patients remained on the diet. Two (15%)
patients were seizure free, eight (61.7%) children had a 75–99%
decrease in seizures, and the remaining three (23%) children
diet, 10 (77%) children had achieved a >75% decrease in their
seizures. Four patients have been off the diet for >2 years; one
abandoned the diet after 2 years of adhering to it, relapsed. No
differences in seizure control when compared with age, sex, or
seizure type were found.
Conclusions: Considering the severity and intractability of
seizures in patients with DS, the fact that 10 of the 13 children
who remained on the diet had a significant reduction in number
of seizures shows that the KD is at present an interesting ther-
apeutic alternative. Even in patients in whom seizure reduction
number of AEDs was reduced to one or two. We consider that
children with DS should be offered the KD immediately after
three adequate trials of AEDs have failed. Key Words: Dravet
syndrome—Ketogenic diet—Refractory epilepsy—Seizures.
The ketogenic diet (KD) has been used as a therapeutic
alternative to antiepileptic drugs (AEDs) for refractory
epilepsy (1–3). The diet consists of an intake of 3 or 4
times as much fat as carbohydrates and protein combined
Fasting has since long been believed to be a cure for
epilepsy. Hippocrates used fasting as a specific treatment
for patients with epilepsy in the 5th century BC, and in
the Bible, Jesus suggested fasting after an epilepsy attack
(New Testament: Saint Mark 9;14–29).
In 1921, Wilder (5) was the first to formulate the KD
to induce the metabolic effects of fasting for the man-
agement of seizures. In spite of its effectiveness, the diet
was replaced by the new AEDs. Phenobarbital (PB), the
first AED, was introduced in 1912, and between 1935 and
1968, 16 additional AEDs became available. The KD was
reserved for use in selected patients and at selected treat-
ment centers. A variant of the classic diet using medium-
Accepted April 22, 2005.
Address correspondence and reprint requests to Dr. R.H. Caraballo
at Neurolog´ ıa, Hospital de Pediatr´ ıa “Prof Dr Juan P Garrahan,” Com-
bate de los Pozos 1881, C.P. 1245, Buenos Aires, Argentina. E-mail:
chain triglycerides was introduced in the 1970s (6), but
over the last decade, most of the centers have been adopt-
and carbohydrates. The group of Johns Hopkins Medical
Institutions in Baltimore was the most enthusiastic in ad-
vising the KD, and they were able to reach public recog-
nition through a movie and a very practical little book
entitled, The Epilepsy Diet Treatment (1). Furthermore,
the popularity of the Atkins diet as a weight-loss diet in
products that are “keto-compatible,” increasing the toler-
ability of the KD (7).
No mechanism of action of the diet has been defined.
Its efficacy has been ascribed to acidosis, cellular and
extracellular dehydration, the direct action of aceto ac-
etate or β-hydroxy-butyrate, and changes in the source
or utilization of energy within the brain (8). Alternative
mechanisms for the action of the KD are an increase in
brain γ-aminobutyric acid, and the potential effects of
as antiepileptic mediators (3,9).
Recent reports of series of patients after 1 year on the
diet show an overall efficacy ranging from 15% to 50% in
1540R. H. CARABALLO ET AL.
terms of patients becoming seizure free or having a 50–
90% reduction in seizures (2–4,10–18). Use of the KD is
but it is also well tolerated and efficacious in adolescents
and adults with intractable epilepsy (19–21).
Severe myoclonic epilepsy in infants or Dravet syn-
drome (DS) is one of the most malignant epileptic syn-
dromes and was first described and characterized by
Dravet et al. (22,23). Since then, DS has been recognized
as an independent epileptic syndrome with characteris-
tic features in many countries. However, diagnosis of DS
in the early stage is rather difficult because of the cur-
rent diagnostic criteria, which place more importance on
the clinical course than on the specific electroclinical fea-
tures. DS is characterized by an onset of febrile hemi-
clonic or generalized status epilepticus in the first year of
followed by the evolution of afebrile seizures including
myoclonic, absence, atonic, and partial seizures between
ages 1 and 4 years. The EEGs in the early stage usually
do not show paroxysmal discharges, but later generalized
spike–waves and polyspikes–waves and focal abnormal-
ities appear. Photosensitivity may appear early (22,23).
Developmental progress is initially normal but slows in
the second year, and thereafter the intellectual prognosis
is usually poor (22,23).
In this retrospective study, we evaluate the efficacy and
tolerability of the KD in patients who met diagnostic cri-
teria of DS.
MATERIALS AND METHODS
Between March 1, 1990, and August 31, 2004, 52 pa-
tients that met diagnostic criteria of DS were enrolled in a
study at our department. Twenty of them were placed on
the KD with the Hopkins protocol (1) and followed up for
a minimum period of 1 year. We saw ∼11,500 children
with epilepsy in our center during the same time period.
risk families were those not considered to be prepared to
follow the diet strictly for different reasons, such as very
numerous families or those with psychological problems.
As Argentina is a developing country, a large percentage
of the population lives in poverty and is not able to meet
daily food requirements. All patients except two were re-
fractory to AEDs. None of these patients had head-drops
or drop attacks secondary to myoclonic–astatic seizures.
Frequency of the seizures was registered on the basis
of daily seizure calendars kept by the parents. Electroen-
cephalograms during wakefulness and sleep were per-
formed ≥6 months before starting, while on, and after
photic stimulation (IPS). Baseline blood tests and lipid
profiles were also obtained.
Children started fasting in the hospital for 36–48 h and
were then gradually initiated on the classic KD (Johns
protein plus carbohydrate) and stayed in the hospital for
another 4 days for close monitoring. During this period,
parents were taught about the diet. They were asked to
keep the child on the diet for ≥2 months to regulate the
diet for optimal tolerance and seizure control. The ratio of
urinary ketosis and to avoid both weight gain and weight
loss. Adverse events and reasons for diet discontinuation
were recorded, as were changes in medication.
Fifty-two children with the diagnosis of DS were fol-
eight girls) were placed on the KD as add-on to the use
of one to three AEDs. Ages at initiation of the KD were
between 3 and 9.5 years (mean, 6 years).
All patients had more than one type of seizure before
starting the diet: five patients had two types, 13 had three
types, and two had more than three types of seizures. Pa-
tients with myoclonic seizures averaged 92 events per
month; those with generalized tonic–clonic, clonic, or
atonic seizures averaged 43 events per month; those with
motor focal seizures averaged 39 events per month; those
with absence or atypical absence seizures averaged 35
events per month, and those with secondarily generalized
tonic–clonic seizures averaged 21 seizures per month.
The children had previously been exposed to a mean of
6.4 different AEDs and were taking a mean of 2.2 AEDs
when the diet was begun. All patients had an IQ of <69.
Table 1 shows the electroclinical features of the pa-
tients before starting the KD, and the clinical and EEG
features of our series of patients while on the diet are de-
Duration on the diet
Three of the 20 original children stayed on the diet for
12 months; four children remained on the diet for 2 years;
four children remained on the diet for 3 years; and two
children, for 4 years.
Efficacy of the diet
One year after initiating the diet, 13 (65%) of the initial
patients remained on the diet. Two (15%) patients were
seizure free; eight (61.7%) children had a 75–99% de-
crease in seizures; and the remaining three (23%) chil-
dren had a 50–74% decrease in seizures. Thus 1 year after
starting the diet, 10 (77%) children had achieved a >75%
decrease in their seizures. Four patients have been off the
diet for >2 years: one of them is seizure free, two have
Epilepsia, Vol. 46, No. 9, 2005
KETOGENIC DIET AND DRAVET SYNDROME1541
TABLE 1. Electroclinical features of the patients before starting the KD and while on the diet
Before the diet
Age at Seizure
retardation onset (months)abnormalitiesAEDs
Daily GSW VPA-CLBMild
5 5.5Weekly VPA-TPMModerate
6 4.5WeeklyGSW-PSW LTG-CLB-PHTModerate
7 11 DailyYes GSW-FS VPA-TPM- LTGModerate
89 WeeklyGSW-MS VPA-ESMMild
196Weekly YesGSW-FS VPA-PHT-TPMModerate
While on the diet
after KD endsPatientsabnormalitiesAEDsSide-effectsResults
Discont. Adv. Effects
75–99%3 years sporadic seizures
66Monthly CLB-PHT2 years50–74%
Epilepsia, Vol. 46, No. 9, 2005
1542 R. H. CARABALLO ET AL.
TABLE 1. Continued.
While on the diet
after KD endsPatients abnormalities AEDsSide-effectsResults
2.5 sporadic seizures
3 years seizure-free
2.4 years seizure recurrent
GTCS, generalized tonic–clonic seizure; SPS, simple partial seizure; CPS, complex partial seizure; GSW, generalized spike–wave; GPSW,
generalized polyspikes–wave; MS, multifocal spike; FS, focal spike.
aMost-frequent seizure type.
sporadic seizures, and one relapsed and abandoned the
diet after 2 years of adhering to it.
paring age and sex. Neither were any differences found
when comparing the effect of the diet on seizure types.
to the diet and did not repeat the event.
The number of children on the diet and the vary-
ing levels of seizure control they achieved are shown in
Reasons for discontinuation of the diet, tolerability,
and adverse events
Seven (35%) of the 20 children who initiated the diet
discontinued within the first year. In five, the reason given
these children discontinued between 1 and 3 months and
one discontinued between 3 and 6 months after starting
Persistent and severe vomiting was the reason for dis-
continuing the diet in two children. The 13 patients who
remained on the diet for >1 year did not develop compli-
In the 13 patients who followed the KD with good re-
sponse, the EEG recordings of ≥3 months before start-
TABLE 2. Ketogenic diet in 20 patients with Dravet syndrome
because of decrease
the diet with
ing KD showed generalized, symmetric or asymmetric
polyspikes and polyspikes–waves in nine patients and
multifocal spikes and spike–waves in four.
One year after initiating the diet, the EEG abnormali-
ties had improved in all 13 patients. In the two patients
who became seizure free, the EEG recording was nor-
mal in one and showed isolated generalized polyspikes–
waves in the other. In the eight patients who achieved
a 75–99% decrease in their seizures, the EEG recording
showed occasional and isolated generalized polyspikes–
waves or spikes in seven, and in the remaining patient,
the EEG recording had become normal. In the three pa-
EEG abnormalities during sleep improved between 60%
and 80%. The changes of abnormalities on the sleep EEG
were determined according to the quantity of paroxysmal
discharges on the EEG recording. In two patients, we did
not find any changes with IPS while on the diet.
In the five patients in whom the diet proved to be in-
effective and in the two cases in whom the diet was not
tolerated, the EEG abnormalities did not change.
Decreasing and discontinuing medications
Medications were decreased and discontinued nonsys-
temically with the aim of having the patient medication
free. The different types of AEDs of the children on the
diet are shown in Table 3.
Epilepsia, Vol. 46, No. 9, 2005
KETOGENIC DIET AND DRAVET SYNDROME 1543
TABLE 3. AEDs in patients on the ketogenic diet
Our experience with the KD in 20 children with DS
shows an overall good response in terms of seizure fre-
quency and quality of life.
To our knowledge, this is the first study of treatment
with the KD related to DS, regardless of the seizure type.
In this study, no differences were found in the effect of
the diet on the seizure types. Previous reports have been
published on children with refractory seizures, including
infantile spasms, who were placed on the KD (16). We
also described children with DS and myoclonic epilep-
sies treated with the KD in earlier reports (24,25). Our
results support the general assumption that the KD is ef-
of epileptic seizures. We believe that more precise defi-
nitions of epileptic syndromes may lead to the possibility
of giving a clear prognosis in each particular case and
that this concept should be applied to all therapeutic tri-
als in epilepsy. For example, considering idiopathic and
symptomatic focal seizures in the same trial to evaluate
efficacy of a specific medication may lead to false results,
to AEDs. The length of follow-up in trials should also be
considered. Results at 3 or 6 months are of no real value
to evaluate AEDs, the KD, or even surgical treatments.
A clear decrease in seizure control after a time on the
KD has been reported (17,19). Incidentally, the signifi-
cant decrease in efficacy after 9–12 months on the KD
in patients with refractory symptomatic and cryptogenic
partial or generalized epilepsies found in one study may
have been due to the inclusion of all etiologies (19).
Selection of families that will be able to make the effort
that treatments such as the KD require in children has to
be very strict to reduce the number of failures.
It is interesting to note that seven patients who re-
sponded well to the diet received topiramate (TPM). In
children with refractory epilepsy, co-treatment with TPM
and KD may be considered, but special attention should
be paid to the combined risk for metabolic acidosis and
nephrolithiasis (26). However, we have not yet observed
these complications. According to Takeoka et al. (26),
seizure reduction did not appear to correlate with the re-
duction in HCO3/reduction levels. The reasons that co-
treatment with TPM and KD is effective are unknown.
Considering the difficulty of controlling seizures in the
studied population with medically refractory epilepsy, we
believe the observed response to be significant.
Some authors have suggested that stiripentol and bro-
mides are effective in DS (27,28). However, in Argentina,
these drugs are not available.
Considering the severity and intractability of seizures
in patients with DS, the fact that 13 of the 20 children
who stayed on the diet had a significant decrease in their
number of seizures shows that the KD is at present an in-
the reduction in seizures was not dramatic, an improve-
ment in quality of life was seen, and in all of them, the
number of AEDs was reduced to one or two. One of these
patients did not show any further mental deterioration.
The youngest child on the KD in our series of patients
with DS was 3 years old. That is partly due to the difficul-
ties we had in the past to define the diagnosis clearly and
for the necessary time to elapse while trying an AED.
We consider that children with DS should be offered
the KD immediately after three adequate trials of AEDs
prospective comparative trials including the early use of
the KD in one treatment arm.
1. Freeman JM, Kelly JB. The epilepsy diet treatment. New York: De-
2. Lefevre F, Aronson N. Ketogenic diet for the treatment of refrac-
tory epilepsy in children: a systematic review of efficacy. Pediatrics
3. Phelps SJ, Hovinga CA, Rose DF, et al. The ketogenic diet in the
pediatric epilepsy. Nutr Clin Pract 1998;13:267–82.
4. Di Mario FJ, Holland J. The ketogenic diet: a review of the ex-
perience at Connecticut Children’s Medical Center. Pediatr Neurol
5. Wilder RM. The effect of ketogenemia on the course of epilepsy.
Mayo Clin Bull 1921;2:307–14.
6. Huttenlocher PR, Wilbourn AJ, Signore JM. Medium chain triglyc-
erides as a therapy for intractable childhood epilepsy. Neurology
7. Thiele EA. Assessing the efficacy of antiepileptic treatments: The
ketogenic diet. Epilepsia 2003;44(suppl 7):30–7.
8. Vining EP. Ketogenic diet. In: Engel J, Pedley TA, eds.
Epilepsy: a comprehensive textbook. Philadelphia: Lippincott-
9. Seymour KJ, Bluml S, Sutherling W, et al. Identification of cerebral
diet. MAGNA 1999;8:33–42.
10. Nordli DR, De Vivo DC. The ketogenic diet revisited: back to the
future. Epilepsia 1997;38:743–9.
11. Schwartz RH, Eaton J, Bower BD, et al. Ketogenic diets in the
treatment of epilepsy: short-term clinical effects. Dev Med Child
ketogenic diet in the treatment of severe epilepsy. Dev Med Child
13. Kinsman SL, Vining EPG. Efficacy of ketogenic diet for intractable
seizure disorders: review of 58 cases. Epilepsia 1992;33:1132–6.
14. SwinkTD,ViningEPG,FreemanJM.Theketogenicdiet.In: Barnes
Epilepsia, Vol. 46, No. 9, 2005
1544 R. H. CARABALLO ET AL. Download full-text
of the efficacy of the ketogenic diet. Arch Neurol 1998;55:1433–7.
16. Nordli DR, Kuroda MM, Carroll J, et al. Experience with the keto-
genic diet in infants. Pediatrics 2001;108:129–33.
17. Freeman JM, Vining EPG, Pillas D, et al. The efficacy of the ke-
togenic diet, 1998: a prospective evaluation of intervention in 150
children. Pediatrics 1998;102:1358–63.
18. Caraballo R, Tr´ ıpoli J, Escobal L, et al. Ketogenic diet: effi-
cacy and tolerability in childhood intractable epilepsy. Rev Neurol
adolescents and young adults with refractory epilepsy: an Italian
multicentric experience. Epilepsy Res 2002;48:221–7.
20. Mady MA, Kossoff EH, McGregor AL. The ketogenic diet: adoles-
cents can do it, too. Epilepsia 2003;44:847–51.
21. Sirven J, Whedon B, Caplan D, et al. The ketogenic diet
for intractable epilepsy in adults: preliminary results. Epilepsia
22. Dravet Ch, Bureau M, Oguni H, et al. Severe myoclonic epilepsy
in infancy (Dravet syndrome). In: Roger J, Bureau M, Dravet Ch,
et al., eds., Epileptic syndrome in infancy, childhood and adoles-
cence, 3rd ed. Eastleigh: John Libbey, 2002:81–104.
23. Fejerman N. Severe myoclonic epilepsy in infancy (Dravet syn-
drome). In: Wallace S, Farrel K, eds. Epilepsy in children, 2nd ed.
London: Arnold, 2004:157–60.
24. Caraballo R, Cers´ osimo R, Y´ epez I, et al. Severe myoclonic
epilepsy in infancy treated with ketogenic diet [Abstract]. Epilepsia
25. Fejerman N, Caraballo R, Cers´ osimo R. Ketogenic diet in pa-
tients with Dravet syndrome and myoclonic epilepsies in infancy
and early childhood. In: Delgado Escueta AV, Guerrini R, Med-
ina MT, et al, eds. Myoclonic epilepsies, advances in neurology
series. Philadelphia: Lippincott Williams & Wilkins, 2004:299–
26. Takeoka M, Riviello J, Pfeifer H, et al. Concomitant treatment
with topiramate and ketogenic diet in pediatric epilepsy. Epilepsia
27. Chiron C, Marchand MC, Tran A, et al. Stiripentol in se-
vere myoclonic epilepsy in infancy: a randomised placebo-
controlled syndrome-dedicated trial: STICLO study group. Lancet
28. Kanazawa O, Shirane S. Can early zonisamide medication improve
the prognosis in the core and peripheral types of severe myoclonic
epilepsy in infants? Brain Dev 1999;21:503.
Epilepsia, Vol. 46, No. 9, 2005