Clinical Nutrition 31 (2012) 246e249

Contents lists available at SciVerse ScienceDirect

Clinical Nutrition
journal homepage: http://www.elsevier.com/locate/clnu

Original article

Effects of the ketogenic diet on nutritional status, resting energy expenditure,

and substrate oxidation in patients with medically refractory epilepsy: A 6-month
prospective observational study
Anna Tagliabue a, *, Simona Bertoli b, Claudia Trentani a, Paola Borrelli c, Pierangelo Veggiotti d
a

Human Nutrition and Eating Disorders Research Centre, University of Pavia, Pavia, Italy
International Centre for the Assessment of Nutritional Status (ICANS), Department of Food Science Technology and Microbiology (DISTAM), University of Milan, Milan, Italy
Department of Health Sciences, Section of Medical Statistics and Epidemiology, University of Pavia, Pavia, Italy
d
Child Neuropsychiatry Department, Casimiro Mondino Foundation, University of Pavia, Pavia, Italy
b
c

a r t i c l e i n f o

s u m m a r y

Article history:
Received 26 April 2011
Accepted 28 September 2011

Background & aims: This 6-month prospective, single-arm observational study was designed to assess the
effects of the KD on the nutritional status, resting energy expenditure (REE), and substrate oxidation in
patients with drug-resistant epilepsy.
Methods: Eighteen patients with medically refractory epilepsy underwent assessment of body composition, REE, and substrate oxidation rates before and after 6 months of KD.
Results: Compared with baseline, there were no statistically signicant differences at 6 months in terms of
height, weight, BMI z-scores, and REE. However, the respiratory quotient decreased signicantly (from
0.80
0.06 to 0.72
0.05, p < 0.001) whereas fat oxidation was signicantly increased (from 50.9
25.2 mg/
min to 97.5
25.7 mg/min, p < 0.001). Interestingly, we found that the increase in fat oxidation was the main
independent predictor of the reduction in seizure frequency (beta 0.97, t 6.3, p < 0.05).
Conclusions: Administering a KD for 6 months in patients with medically refractory epilepsy increases fat
oxidation and decreases the respiratory quotient, without appreciable changes in REE.
2011 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved.

Keywords:
Ketogenic diet
Drug-resistant epilepsy
Resting energy expenditure
Fat oxidation
Respiratory quotient

1. Introduction
The ketogenic diet (KD) is a high-fat, low-carbohydrate, and
normal-protein diet which increases the production of the ketone
bodies (i.e., b-hydroxybutyrate and acetoacetate) and has been used
for the treatment of medically refractory childhood epilepsy since
the 1920s.1,2 Although numerous studies have shown the clinical
efcacy of the KD, the mechanisms underlying its antiseizure effect
remain only partially elucidated.3,4 Multiple theories have been
proposed to explain how the KD could protect against epilepsy,
including modication of the tricarboxylic acid cycle to increase gaminobutyric acid synthesis in the brain, limit reactive oxygen
species generation, and boost energy production in brain tissue. As
a result, hyperpolarization of neurons occurs, stabilizing synaptic
function and increasing resistance to seizures throughout the brain.5
Besides its effects on the central nervous system, the KD may
predispose to nutritional decits in energy, proteins, minerals, and
Abbreviations: KD, ketogenic diet; REE, resting energy expenditure; RQ, respiratory quotient.
* Corresponding author. Tel.: 39 0382 987868; fax: 39 382 987869.
E-mail address: anna.tagliabue@unipv.it (A. Tagliabue).

vitamins and excess in lipids, saturated fat, and cholesterol.6,7 For

these reasons, the use of such an unbalanced diet requires strict
adherence to the dietary plan and particular attention to implementation and monitoring, particularly at a young age.8 Importantly, most of the side effects from the KD seem to be related to
energy and nutrient deciencies.9 However, the long-term metabolic effects of the KD have not been adequately studied.
We therefore conducted this 6-month prospective single-arm
study to assess the effects of the KD on resting energy expenditure (REE), substrate oxidation, and nutritional status in patients
with drug-resistant epilepsy. Clarication of these aspects is
clinically important to optimize the energy requirements of
patients on the KD, as well as to reduce its potential metabolic side
effects.
2. Patients and methods
2.1. Ethics
The study protocol received Institution Review Board approval
and complied with all tenets of the Helsinki declaration. All

0261-5614/$ e see front matter 2011 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved.
doi:10.1016/j.clnu.2011.09.012

A. Tagliabue et al. / Clinical Nutrition 31 (2012) 246e249

patients (or parents if the patient was <18 years of age) provided
written informed consent before the beginning of the study.
2.2. Study design
This was a 6-month, prospective, single-center, single-arm
study of the metabolic effects of the KD in patients with medically
refractory epilepsy treated with the KD. The main outcome
measures were the change from baseline in nutritional status,
resting energy expenditure, and substrate oxidation.
2.3. Participants
Patients with medically refractory epilepsy were consecutively
enrolled from the Department of Child Neuropsychiatry of the
IRCCS Casimiro Mondino Foundation (Pavia, Italy) between 2000
and 2006. All participants had tried a minimum of 2 appropriate
rst-line antiepileptic drugs in a maximally tolerated dose and still
had a minimum of 2e3 seizures per month. Patients were
excluded if they had disease states causing signicant nutritional
status impairment (neoplasia, end-stage renal failure, chronic
infection) or changes in energy metabolism (thyroid disorders).
Children requiring enteral and parenteral nutrition were also
excluded.
2.4. Ketogenic diet
We followed the Johns Hopkins Hospital protocol with few
changes as previously described in Italian patients.10 Detailed
counseling was given including a recipe demonstration. The
counseling included the scientic basis of the diet, the rigid nature
of the diet (only prescribed food could be taken), demonstration of
the urine ketone testing and the possible side effects. Initially,
patients were admitted to the Child Neuropsychiatry Department
and during a fasting phase of 12e36 h (only water being allowed),
blood sugar, vital parameters and urine ketones were checked
every 4 h. Once urine ketone levels reached 80e160 mg/dL,
participants were started on the KD on a 4:1 ketogenic ratio. The
ketogenic ratio refers to the ratio of the fat amount in grams to the
combined protein and carbohydrate amount in grams. A 3:1 ratio
was occasionally used to meet protein needs. The initial calorie
prescription was based on an average between this prediet intake
and recommendations from Johns Hopkins for energy requirements on the KD but taking into account weight and height (both
current and recent trends), and physical activity levels. Alterations
in calorie prescriptions were made as needed during the course of
follow-up. A minimum of 0.8e1 g of protein from animal sources
(e.g. eggs, milk, meat, poultry and sh) per kilogram of body weight
per day was given. All participants received sugar-free multivitamin
and mineral supplements according to the patients age and sex.
Dietary planning was carried out using a computerized system
(Dieta ragionata 7.0 e ESI informatica, San Donato Milanese, Milan,
1997) developed using published Italian food composition tables.
Families were instructed to check urine ketones daily and to
appropriately report unexpected changes.

247

2.6. Anthropometry
Standard anthropometry was carried out by one operator using
conventional criteria and measuring procedures. A digital scale was
used to measure body weight to the nearest 100 g. Subjects were
weighed without shoes, in light clothing. Body height was
measured using supine length to the nearest 0.5 cm. Body mass
index (BMI) was calculated as body weight in kilograms divided by
height in meters squared. The reference standards published by the
Centers for Disease Control and Prevention (2000)11 were used to
calculate the age- and sex-specic z-scores for weight, height and
BMI.
2.7. Body composition assessment
Fat mass (FM) and fat free mass (FFM) were determined by dual
energy X-ray absorptiometry (DXA) using a Lunar DPX-IQ scanner
(Lunar Corp, Madison, WI, USA). Total body scans were performed
on all subjects in the supine position. The entire body of each
subject was scanned, beginning at the top of the head and moving
in a rectilinear pattern down the body to the feet. The mean
exposure time was 15 min, and the dose was less than 7 mSv.
Calibration of the machines was performed daily, and quality
assurance was checked by measuring the manufacturers spine
phantom at daily intervals. The coefcients of variation were less
than 1% for all measurements. A single operator performed all scans
and analyses.
2.8. Outcome measures
REE was measured with an open-circuit ventilated-hood
system (Sensor Medics 29, Anaheim, CA, USA). All measurements
were performed in the post-absorptive state (12e14 h fast) in
a thermoneutral environment (ambient temperature 24e26  C)
in the absence of external stimuli. The calorimeter was calibrated
at the beginning of each test using two reference gas mixtures
(26% O2 and 74% N2; 16% O2, 4.09% CO2 and 79.91% N2, respectively). Data were collected for 30 min. A 10-min run-in time was
allowed for stabilization and for participants to acclimatize to the
canopy and instrument noise.12 The average of the last 20 min
was used to determine 24 h REE using the abbreviated Weir
equation13:

REEKcal=day 3:941 VO2 mL=min

1:106 VCO2 mL=min  1:44
The respiratory quotient (RQ) was calculated from the ratio:
RQ CO2 produced/O2 consumed. The oxidation of carbohydrate
and fat was determined from the measurements of O2 consumption, CO2 production according to the following equations14:
Carbohydrates (mg/min) 4.585 V CO2  3.2255 V O2
Fat (mg/min) 1.7012 V CO2 1.6946 V O2

2.9. Data analysis

2.5. Dietary intake assessment
Dietary intake before initiation of the KD (baseline) and then at
6 months while on the KD was assessed with the use of 7d weighed food records. Families were provided a calibrated food
scale accurate to 0.1 g and instructions for weighing and recording
everything the child consumed. Dietary intake data were analyzed
with the use of the Dieta ragionata 7.0 software.

The KolmogoroveSmirnov test was performed in all continuous

variables to dene the presence of normality. Variables are
expressed as means
standard deviations or medians and interquartile ranges, whereas categorical variables are expressed as
counts and percentages. Because of the paired data, comparisons of
continuous variables before and after 6 months of KD were performed by the paired t-test or the Wilcoxon signed-rank test, as

248

A. Tagliabue et al. / Clinical Nutrition 31 (2012) 246e249

appropriate. The Spearmans correlation coefcient was used to

investigate correlation between variables. All skewed variables
were log-transformed for further analysis. Multivariable stepwise
regression was used to estimate the association between metabolic
changes and the reduction in seizure frequency. All calculations
were performed using SPSS version 17.0 for Windows (SPSS, Inc.,
Chicago, IL, USA). A value of p < 0.05 (two-sided) was considered
statistically signicant.
3. Results
A total of 18 Caucasian Italian patients with medically refractory
epilepsy (8 males and 10 females, mean age: 12.4
5.6 years) were
enrolled in this study. All participants completed the 6-month
protocol. Table 1 shows the daily dietary intake before and after
the beginning of the KD. The total energy intake did not differ
signicantly before and after the completion of the study. Adherence
to the KD protocol was documented by constant ketonuria in all
subjects. All of the children tolerated the diet well, and there were no
adverse events. The seizure frequency decreased from a median of 5
per day (interquartile range: 1e20) at baseline to 2 per day (median:
1e12) after 6 months of KD (p < 0.001; Wilcoxon signed-rank test).
3.1. Main outcome measures
Table 2 shows the nutritional status, resting energy expenditure,
and substrate oxidation of the study participants before and after 6
months of KD. Compared with baseline values, there were no
differences at 6 months in terms of height, weight, and BMI zscores. Both predicted and measured REE did not differ signicantly
before and after 6 months of KD. However, the respiratory quotient
decreased signicantly (from 0.80
0.06 to 0.72
0.05; p < 0.001)
while fat oxidation was signicantly increased (from
50.9
25.2 mg/min to 97.5
25.7 mg/min; p < 0.001). As expected,
carbohydrate
oxidation
decreased
signicantly
(from
72.5
54.1 mg/min to 21.5
48.2 mg/min, p < 0.001) after the
completion of the study. We found no correlation between fat
oxidation and fat free mass, either at baseline or after completion of
the study. The results of multivariable stepwise regression analysis
demonstrated that the increase in fat oxidation was the main
independent predictor of the reduction in seizure frequency
(beta 0.97, t 6.3, p < 0.05).
4. Discussion
The results of our pilot study indicate that administering a KD
for 6 months increases fat oxidation and decreases the respiratory
quotient, without appreciable changes in REE.
Table 1
Daily dietary intake before and after the beginning of the ketogenic diet.
Baseline

6 months

p value

Mean Standard Mean Standard

deviation
deviation
Energy intake (kcal/24 h)
Energy intake (kcal/kg)
Fat (g/kg)
Fat (% energy)
Saturated fat (% energy)
Monounsaturated fat (% energy)
Polyunsaturated fat (% energy)
Protein (g/kg)
Protein (% energy)
Carbohydrates (g/kg)
Carbohydrates (% energy)

1487
59
2.5
37
10
13
3
2.5
16
6.9
47

ns indicates nonsignicant differences.

448
23
1.0
5
4
2
1
1.1
3
3.0
8

1512
58
5.9
89
21
27
9
1.0
8
0.5
3

331
21
1.8
2
5
4
6
0.3
2
0.2
1

ns
ns
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001

Table 2
Nutritional status, resting energy expenditure, and substrate oxidation at baseline
and after 6 months of KD.
Baseline

Body height z-score

Body weight z-score
BMI z-score
REE (predicted, kcal)
REE (measured, kcal)
REE (%, measured versus
predicted)
REE/body weight (kcal/kg)
REE/fat free mass (kcal/kg)
Respiratory quotient
Fat oxidation (mg/min)
Carbohydrate oxidation
(mg/min)

6 months

p value

Mean

Standard
deviation

Mean

Standard
deviation

0.72
0.81
1.33
1277
1107
16.4

1.70
2.42
2.17
258
277
12.1

0.76
0.86
1.10
1233
1081
15.7

1.73
2.47
2.11
224
237
12.6

ns
ns
ns
ns
ns
ns

33.3
48.5
0.80
50.9
72.5

14.3
21.3
0.06
25.2
54.1

32.4
43.5
0.72
97.5
21.5

11.1
16.2
0.05
25.7
48.2

ns
ns
< 0.001
< 0.001
< 0.001

Despite intensive investigation on the clinical benets of the

KD for the treatment of difcult-to-control seizures,1e5 changes in
energy expenditure and metabolism have been poorly studied. In
the present report, height, weight, and BMI z-scores after 6
months of the KD remained roughly constant compared with
baseline values. This is important and reassuring, as it clearly
suggests that a tradition KD can provide an adequate calorie intake
to ensure a proper growth during childhood and adolescence.
Accordingly, the KD was carefully calculated in an effort to achieve
the childs ideal body height, weight, and BMI besides improving
seizure control. This is clinically relevant as Hamdy et al.15 have
shown that stable body mass index or ideal body mass index-forage do not correlate per se with the efcacy with the traditional
KD. Therefore, attributing changes in seizure control to a rapid
weight gain or loss may be unjustied. This assumption was
conrmed in our study, as seizures control improved despite
stable height, weight, and BMI z-scores over time. Our results
differ from those of Neal et al.16 who showed that weight z-scores
decreased signicantly between baseline and 3, 6, and 12 months
of KD, while height z-scores decreased signicantly by 6 and 12
months. This may possibly result, at least in part, from different
KD plannings. Of note, our results are in line with those by Vining
et al.17 who showed that the KD generally provides sufcient
nutrition to maintain growth within normal parameters over
a dened period. However, it is known that very young children
may grow poorly on the KD and should be followed-up carefully
over long periods of use.8
Resting energy expenditure represents the largest fraction
(50e70%) of an individuals total daily energy expenditure.18 Our
data clearly indicate that there was no change in resting energy
expenditure associated with the KD. Previous studies have shown
that the body has a great capacity to adjust substrate oxidation to
substrate intake after approximately 1 week for carbohydrate and
fat.19 As expected, fat oxidation increased in our study as an
adaptation to the high fat intake typical of the KD. The consequence of an isoenergetic exchange of fat for carbohydrate is that
the results can also be interpreted as being an adaptation to a low
carbohydrate intake.20 In particular, fat oxidation can be raised on
high-fat diets by maintaining glycogen concentrations in a lower
range. Although the very low carbohydrate intake must have
resulted in reduction of the glycogen stores, this was not, however,
detectable in terms of body weight and BMI. It therefore seems
that the glycogen content of the body decreased until a new
concentration was reached in which fat oxidation was sufciently
elevated to become in equilibrium with the elevated fat intake.19 It
is also possible that fat oxidation was elevated because of

A. Tagliabue et al. / Clinical Nutrition 31 (2012) 246e249

increased enzymatic capacity for fat oxidation, which occurred

because of the exposure to the KD.19 Interestingly, we found that
the increase in fat oxidation was the main independent predictor
of the reduction in seizure frequency. This preliminary result
seems to suggest that certain fat oxidation products may be
directly or indirectly involved in the reduction of seizure
frequency induced by the KD. In this regard, an experimental
study by Harney et al.21 has shown that acute inhibition of fatty
acid oxidation in KD-fed animals resulted in a shorter latency to
experimentally-induced seizures, which was accompanied with
a drop in plasma b-hydroxybutyrate levels. Further studies are
needed to shed more light on the antiseizure potential of bhydroxybutyrate in humans.
Several caveats of this study merit comments. Our report was
designed as an exploratory pilot project. Although the main effects
of 6 months of the KD on nutritional status, resting energy
expenditure, and substrate oxidation were clear, our data need to
be conrmed in a randomized double-stranded study comparing
the KD with a comparison diet. Our ndings clearly provide
a strong rationale for such a study. Future studies should also
include a longer follow-up period, because a 6-month observational period may not be sufcient to determine the impact of the
KD on weight and height. Another limitation is that we have no
data on bone mineralization in the study participants. To our
knowledge, only one previous 15-month study was specically
designed to address this issue.22 Of note, the results demonstrated
that the KD may lead to progressive loss of bone mineral content,
albeit the exact mechanisms underlying this adverse effect remain
unclear.22 It can be also argued that an intermediate evaluation at 3
months would have been useful to study the dynamic evaluation of
the main outcome measures. Future ad hoc studies are needed to
specically address this important issue. Finally, this study included
only subjects of Italian nationality, so that results cannot be
extrapolated to populations with different ethnic background.
In summary, the major nding of this pilot study is that
administering a KD for 6 months increases fat oxidation and
decreases the respiratory quotient, without appreciable changes in
REE. Intriguingly, our preliminary data seem to suggest that the
increase in fat oxidation was the main independent predictor of the
reduction in seizure frequency. Although the small sample size
limits our ability to draw denite conclusions, this report may
prompt larger studies on the metabolic effects of the KD in children
with drug-resistant epilepsy.
Statement of authorship
All authors have made substantial contributions and nal
approval of the conceptions, drafting, and nal version. All authors
have met the criteria for authorship as established by the International Committee of Medical Journals Editors, believe that the
paper represents honest work, and are able to verify the validity of
the results reported. Authors contribution: AT: conception and
design, data collection and interpretation, general supervision and
nal approval for submission; SB: conception and design, data
collection and interpretation; CT: clinical data collection and
interpretation; PB: data analysis; PV: clinical data collection and
general supervision and nal approval for submission.

249

Conict of interest
There are no conicts of interest to disclose.
Acknowledgments
The authors received no specic nancial support for this
research. We are grateful to Enzo Emanuele, MD, PhD, (Living
Research s.a.s., Robbio, Italy) for expert editorial assistance.
References
1. Veggiotti P, Burlina A, Coppola G, Cusmai R, De Giorgis V, Guerrini R, et al. The
ketogenic diet for Dravet syndrome and other epileptic encephalopathies: an
Italian consensus. Epilepsia 2011;52(Suppl. 2):83e9.
 ski M. The ketogenic diet for epilepsy therapy in children: Quo
2. Pluta R, Jab1on
vadis? Nutrition 2011;27:615e6.
3. Freeman JM, Kossoff EH. Ketosis and the ketogenic diet, 2010: advances in
treating epilepsy and other disorders. Adv Pediatr 2010;57:315e29.
4. 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.
5. Nylen K, Likhodii S, Burnham WM. The ketogenic diet: proposed mechanisms
of action. Neurotherapeutics 2009;6:402e5.
6. Park S, Kim da S, Kang S, Daily 3rd JW. A ketogenic diet impairs energy and
glucose homeostasis by the attenuation of hypothalamic leptin signaling and
hepatic insulin signaling in a rat model of non-obese type 2 diabetes. Exp Biol
Med (Maywood) 2011;236:194e204.
7. Lord K, Magrath G. Use of the ketogenic diet and dietary practices in the UK.
J Hum Nutr Diet 2010;23:126e32.
8. Zupec-Kania B, Zupanc ML. Long-term management of the ketogenic diet:
seizure monitoring, nutrition, and supplementation. Epilepsia 2008;49(Suppl.
8):23e6.
9. Rogovik AL, Goldman RD. Ketogenic diet for treatment of epilepsy. Can Fam
Physician 2010;56:540e2.
10. Tagliabue A, Bertoli S, Trentani C, Lanzola E, Manfredi L, Veggiotti PA, et al.
Drug-resistant epilepsy treated with ketogenic diet. Recenti Prog Med
1997;88:77e9.
11. Kuczmarski R, Ogden C, Grummer-Strawn L, Flegal KM, Guo SS, Wei R, et al.
CDC growth charts: United States. Adv Data 2000;314:1e27.
12. Isbell TR, Klesges RC, Meyers AW, Klesges LM. Measurement reliability and
reactivity using repeated measurements of resting energy expenditure with
a face mask, outhpiece, and ventilated canopy. J Parenter Enteral Nutr
1991;15:165e8.
13. Weir JB. New methods for calculating metabolic rate with special reference to
protein metabolism. Nutrition 1990;6:213e21.
14. Peronnet F, Massicotte D. Table of non-protein respiratory quotient; an update.
Can J Sport Sci 1991;16:23e9.
15. Hamdy RF, Turner Z, Pyzik PL, Kossoff EH. Lack of inuence of body mass index
on the efcacy of the ketogenic diet. J Child Neurol 2007;22:1167e71.
16. Neal EG, Chaffe HM, Edwards N, Lawson MS, Schwartz RH, Cross JH. Growth of
children on classical and medium-chain triglyceride ketogenic diets. Pediatrics
2008;122:e334e40.
17. Vining EP, Pyzik P, McGrogan J, Hladky H, Anand A, Kriegler S, et al.
Growth of children on the ketogenic diet. Dev Med Child Neurol 2002;
44:796e802.
18. Levine JA, Kotz CM. NEATenon-exercise activity thermogenesiseegocentric &
geocentric environmental factors vs. biological regulation. Acta Physiol Scand
2005;184:309e18.
19. Schrauwen P, van Marken Lichtenbelt WD, Saris WH, Westerterp KR. The
adaptation of nutrient oxidation to nutrient intake on a high-fat diet.
Z Ernahrungswiss 1997;36:306e9.
20. Schrauwen P, van Marken Lichtenbelt WD, Westerterp KR. Fat and carbohydrate balances during adaptation to a high-fat diet. Am J Clin Nutr
2000;72:1239e44.
21. Harney JP, Madara J, Madara J, IAnson H. Effects of acute inhibition of fatty acid
oxidation on latency to seizure and concentrations of beta hydroxybutyrate in
plasma of rats maintained on calorie restriction and/or the ketogenic diet.
Epilepsy Res 2002;49:239e46.
22. Bergqvist AG, Schall JI, Stallings VA, Zemel BS. Progressive bone mineral
content loss in children with intractable epilepsy treated with the ketogenic
diet. Am J Clin Nutr 2008;88:1678e84.