Epilepsia, 48(1):6671, 2007

Blackwell Publishing, Inc.


C 2006 International League Against Epilepsy

Vitamin D Status in Children with Intractable Epilepsy, and

Impact of the Ketogenic Diet
A.G. Christina Bergqvist, Joan I. Schall, and Virginia A. Stallings
Divisions of Neurology and Gastroenterology, Hepatology, and Nutrition, The Childrens Hospital of Philadelphia, Department of
Pediatrics and Neurology, University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania, U.S.A.

Summary: Purpose: The aim of this study was to describe

vitamin D status in children with intractable epilepsy prescribed
newer antiepileptic drugs (AEDs) before initiation of and during
15-month treatment with the ketogenic diet (KD).
Methods: Serum vitamin D (25-OHD and 1,25-OHD) and
parathyroid hormone (PTH) were assessed in prepubertal children with intractable epilepsy before initiation of and during KD
therapy. Three-day weighed dietary records including KD and
vitamin and mineral supplementation were obtained at baseline
and at 1 month.
Results: Forty-five children (aged 5.1 2.7 years) were enrolled. Before KD therapy, 4% had deficient and 51% had insufficient serum 25-OHD levels. Vitamin D intake was less than

recommended in 47%. Adequate vitamin D intake, fewer AEDs,

and generalized seizures were associated with higher serum 25OHD levels (p < 0.01). After 3 months on the KD, 25-OHD levels increased (p < 0.001), and PTH declined (p < 0.001). Over
the next 12-month period, 25-OHD levels steadily declined (p
< 0.001), and PTH did not significantly change.
Conclusions: Children with intractable epilepsy treated with
newer AEDs had poor vitamin D status. Their status improved
over the first 3 months of KD therapy with vitamin D supplementation and slowly declined thereafter. Key Words: Vitamin
DParathyroid hormoneCalciumIntractable epilepsy
Ketogenic dietChildren.

Osteoporosis and epilepsy are both recognized as national health problems. Children with intractable epilepsy
are exposed to many factors that contribute to increased
risk of poor bone health, including falls, which directly
increase risk for fracture (Dent et al., 1970; Vestergaard et
al., 1999). Dietary calcium and vitamin D may be inadequate, and seizure treatment may negatively affect nutrient
metabolism and bone health (Fitzpatrick, 2004).
Older antiepileptic drugs (AEDs) such as phenobarbital (PB) and phenytoin (PHT) have been known to cause
adverse effects on both vitamin D and calcium (Gough
et al., 1986). Multiple AEDs are often required for intractable epilepsy, and polytherapy increases the risk for
fractures (Eadie, 1987; Devinsky et al., 1999). The effects
of newer AEDs on bone health have not been well studied. Cross-sectional and casecontrol studies indicate that
these AEDs may have a negative effect on bone health
(Pack et al., 2005). Alternative treatments for intractable
epilepsy such as the ketogenic diet (KD) may also negatively affect bone health (Hahn et al., 1979). The aims

of this study were to describe vitamin D status in a contemporary cohort of children with intractable epilepsy and
to evaluate changes over a 15-month period of treatment
with the KD.
METHODS
Subjects and protocol
Prepubertal children with intractable epilepsy aged 1
to 14 years were recruited for a longitudinal study of the
KD and growth and nutritional status (Bergqvist et al.,
2005). This protocol was approved by the Institutional
Review Board, and informed consent was obtained from
the parent or guardian, and assent was given by children
cognitively able. Measurements were taken at baseline (1
week before beginning the KD) and then every 3 months
for the 15-month study.
Ketogenic diet
Initiation of the KD was a 6-day inpatient hospitalization, and details were previously described (Bergqvist
et al., 2005). Supplementation included one sugar-free
multivitamin (10 g vitamin D and 100 mg elemental
calcium) and 5001,000 mg additional elemental calcium
adjusted for intake, based on dietary records (Institute of
Medicine, 1997). Children were maintained on their AEDs

Accepted May 22, 2006.

Address correspondence and reprint requests to Dr. C. Bergqvist at
Departments of Neurology and Pediatrics, Division of Neurology, The
Childrens Hospital of Philadelphia, 3535 Market Street, Philadelphia,
PA 19104, U.S.A. E-mail: bergqvist@email.chop.edu
doi: 10.1111/j.1528-1167.2006.00803.x

66

VITAMIN D AND THE KETOGENIC DIET

for the first 3 months. Weight and height were obtained by
using standard research techniques (Lohman et al., 1988;
Bergqvist et al., 2005).
Serum vitamin D, PTH, and electrolytes
A fasting blood sample was drawn between 7:30
and 10:00 a.m. for serum 25-hydroxyvitamin D (25OHD), 1,25-dihydroxyvitamin D (1,25-OHD), calcium,
phosphorus, magnesium, and intact parathyroid hormone
(PTH). Vitamin D levels were determined by radioiodinated tracer techniques (Hollis et al., 1993). The lower
limit of detection of 25-OHD was 2.5 nM (1 ng/ml) with
a reference range of 11 to 100 ng/ml, and 1,25-OHD had
a reference range of 15.955.6 pg/ml. Intact PTH levels
were determined by using the radioiodinated tracer (Diasorin, Stillwater, MN, U.S.A.) with a reference range of
13 to 65 pg/ml. Serum calcium, magnesium, and phosphorus were obtained simultaneously and analyzed by using
standard techniques and laboratory reference values.
Dietary intake assessment
A 3-day weighed diet record was obtained at baseline
and after 1 month. A research dietician provided instructions, and food was weighed with 0.1-g accuracy. The
record included 2 weekdays and 1 weekend day, and supplements were included (Minnesota Nutrition Data Systems, NCC Food and Nutrient Database, version 2005).
Energy intake was compared with estimated energy requirements (EERs) for children with low-active and sedentary activity levels, and expressed as a percentage of EER
(%EER) (Institute of Medicine, 2002). Intakes of vitamin
D, calcium, magnesium, and phosphorus were compared
with the Dietary Reference Intake (Institute of Medicine,
1997) and expressed as percentage of adequate intake
(%AI) for vitamin D and calcium and percentage of recommended dietary allowance (%RDA) for magnesium and
phosphorus.
Statistical analysis
Data analysis occurred in two phases. Phase I consisted
of both descriptive statistics of the sample and inferential
statistics to determine significant predictors of serum vitamin D status at baseline. Intake after 1 month on the KD
of energy, vitamin D, calcium, and other selected nutrients was described. Subjects were grouped according to
serum 25-OHD concentrations: deficient vitamin D status, defined as <11 ng/ml (27.5 nM); insufficient as 11
ng/ml to 31.9 ng/ml (27.5 to 79.9 nM); and sufficient as
32 ng/ml (80 nM) (Heaney, 2004; Hollis, 2005). Subjects were also grouped by PTH concentrations: PTH >65
pg/ml was considered elevated, and <13 pg/ml was considered suppressed (Cioffi et al., 2000).
Multiple-regression models were used to determine significant predictors of 25-OHD, 1,25-OHD, and PTH at
baseline. Subjects were categorized by vitamin D and cal-

67

cium intake at baseline, with adequate intake defined as

100% AI, and inadequate intake as <100% AI (reference group). Seasonal effects on serum vitamin D levels
were tested, with the season of the blood draw categorized
as Summer (JuneAugust), Fall (SeptemberNovember),
Winter (DecemberFebruary), and Spring (MarchMay),
with summer as the reference group.
In Phase II, longitudinal mixed-effects models were
used to determine predictors of 25-OHD, 1,25-OHD, and
PTH both in the short term (baseline to 3 months) for
the whole sample, and in the long term (at 3, 6, 9, 12,
and 15 months) for those subjects who responded to
the KD treatment with significant seizure reduction at
3 months and continued an additional 12 months. In all
models, independent predictors of vitamin D status, in
addition to time, were age, gender, season, number of
AEDs taken at KD initiation, type of seizure (generalized, partial), and adequacy of vitamin D and calcium
intake. The level of significance was set to = 0.01, for
each of six related models, after adjustment for multiple,
moderately related outcomes. All statistical analyses were
performed with STATA 7.0 (STATA, College Station, TX,
U.S.A.).
RESULTS
Baseline sample characteristics
Serum vitamin D 25-OHD and 1,25-OHD concentrations along with clinical, biochemical, and dietary characteristics of the 45 children enrolled are presented in
Table 1. The distribution of serum 25-OHD and PTH at
baseline is presented in Fig. 1. Before the initiation of the
KD, 4% had deficient (<11 ng/ml), and 55% had insufficient (<32 ng/ml) 25-OHD levels. No children had elevated PTH levels; however, 42% of children had low PTH.
Serum 25-OHD and PTH were not significantly correlated
at baseline. All children had 1,25-OHD levels within the
reference range. Mean serum calcium, magnesium, and
phosphorus were within reference ranges for the group as
a whole. Although the mean for vitamin D and calcium
dietary intake was adequate, a large proportion of individual children had inadequate vitamin D (47%) and calcium
(38%) intake. The children had suboptimal growth status,
and estimated energy requirements best correlated with a
low-active or sedentary status (see Table 1).
Predictors of serum vitamin D status at baseline
Results from multiple regression analyses for factors
predicting serum 25-OHD at baseline are presented in
Table 2. At baseline, after adjusting for seasonal effects
(25-OHD was lower in winter and spring), 25-OHD levels were higher (13 ng/ml higher; p = 0.004) in subjects
with adequate vitamin D intake (100% AI; 53%) than
those with inadequate intake (<100% AI; 47%). Serum

Epilepsia, Vol. 48, No. 1, 2007

68

A. G. C. BERGQVIST ET AL.

TABLE 1. Clinical, biochemical, and dietary characteristics

of children with intractable epilepsy at baseline
N
Age (yr)
Female/male (%)
AED failed before KD
AED used when KD initiated
Seizures per week
Age at first seizure (yr)
Duration of seizures before KD (yr)
Generalized seizures (%)
Partial seizures (%)
Cerebral palsy (%)
Nonambulatory (%)
Mental retardation (%)
Status epilepticus (%)
Anthropometry
Weight (kg)
WAZ (Z-score)
HAZ (Z-score)
BMIZ (Z-score)
Serum variables
Vitamin D-25-OH (ng/ml)
Vitamin D-1,25-OH (pg/ml)
Intact parathyroid hormone (pg/ml)
Calcium (mg/dl)
Phosphorus (mg/dl)
Magnesium (mg/dl)
Dietary variables
Kilocalories per day
Estimated energy requirements: sedentary (%)
Estimated energy requirements: low active (%)
Protein (% kcal per day)
Carbohydrate (% kcal per day)
Fat (% kcal per day)
Vitamin D (%AI)
Calcium (%AI)
Phosphorus (%RDA)
Magnesium (%RDA)

45
5.1 2.7
27/73
7.7 2.6
2.3 0.8
9.6 (0.2, 5,206)
0.9 (0,10.0)
2.5 (0.5, 8.7)
58
42
42
38
73
40
19.6 11.1
0.4 1.6
0.2 1.3
0.3 2.1
31.4 14
27.2 8.7
15.8 6.3
9.6 0.4
4.7 0.5
2.1 0.2
1,283 474
104 30
93 26
14 3
52 9
35 8
118 87
120 60
175 60
160 61

Values reported as mean SD, median (range) or %.

AI, Adequate intake; RDA, recommended daily allowance; AED,
antiepileptic drug; WAZ, weight for age Z-score; HaZ, height for age
Z-score; BMIZ, body mass index for age Z-score.

25-OHD levels were also higher in those with generalized

seizures (10 ng/ml higher; p = 0.009). The number of
AEDs taken at KD initiation was negatively associated
with 25-OHD levels (p = 0.006), with a decrease of 7
ng/ml for each additional AED. No age or gender effects
on 25-OHD levels were found and no contribution of adequate calcium intake after vitamin D intake was taken
into account. Together, these variables explained 38% of
the variance in 25-OHD (p = 0.002). Further analysis revealed no interactions between number of AEDs, generalized versus partial seizures, or vitamin D intake in these
models. In similar analyses, significant predictors of PTH
levels at baseline were age (older children had higher levels) and number of AEDs, with higher levels in children
using more AEDs explaining 2% of the variance in PTH
(p = 0.019). No significant associations were found for
1,25-OHD. Ambulatory status, cerebral palsy, and mental
retardation were not associated with vitamin D status.
Epilepsia, Vol. 48, No. 1, 2007

Dietary intake on the KD

Dietary intake was assessed after 1 month of the KD.
Mean energy intake was 90 11% EER for low-active
children, with 83% of individual children <100% EER.
Vitamin D intake was 216 75% AI, a significant increase
from 118 87% at baseline (Institute of Medicine, 1997).
Calcium intake, was 99 45%, a decrease from the 120
60% AI at baseline. KD food and beverages contained
minimal vitamin D and calcium and were provided almost
exclusively from supplementation. Phosphorus intake was
111 37% RDA, and magnesium intake was 61 29%
RDA, both considerably lower than baseline.
Vitamin D status after KD exposure
Figure 2 presents 25-OHD, 1,25-OHD, and PTH levels
over time. Serum 25-OHD increased and PTH declined
significantly over the first 3 months of the KD. After 3
months of the KD, 21% had insufficient levels of 25-OHD.
In those children who responded to the KD, serum 25OHD showed a steady and significant decline from 3 to 15
months, whereas PTH remained stable. After 15 months
on the KD, 2% had deficient 25-OHD level, and 30% had
insufficient levels. No changes over time were noted for
1,25-OHD.
Results from longitudinal mixed-effects models describing predictors of both short-term and long-term
changes in 25-OHD are presented in Table 2. Serum 25OHD increased over time from baseline to 3 months by
8 ng/ml (p < 0.001). Children with generalized seizures
had higher 25-OHD, and seasonal effects were evident.
Together, these variables explained 33% of the variance
in 25-OHD. Gender, age, and number of AEDs at baseline did not predict 25-OHD levels. PTH levels declined 6
pg/ml (p < 0.001), and age, season, and number of AEDs
did not predict PTH levels. No significant associations
were found with 1,25-OHD levels.
From 3 to 15 months, 25-OHD declined by 0.5 ng/ml
per month (p = 0.003). Aside from the continuing seasonal
effect, no significant associations of 25-OHD with type of
epilepsy, age, or gender over this time period. Time and
seasonal effects explained 19% of the variance in serum
25-OHD. PTH and 1,25-OHD did not change over time
from 3 to 15 months, and no associations were found with
other variables.
DISCUSSION
Vitamin D is an essential nutrient and is a regulator of
calcium and phosphorus homeostasis, bone formation, and
maintenance. Vitamin D may also have important central
nervous system functions, such as regulating cellular differentiation, proliferation, immune function, and genomic
stability. Therefore this neurosteroid is of interest within
the neurological community.
Osteoporosis is increasingly recognized in patients with
epilepsy, and the causes are likely multifactorial. Older

VITAMIN D AND THE KETOGENIC DIET

69

Serum parathyroid hormone (pg/ml)

40

30

20
PTH
within range

10

deficient
25-OHD

sufficient
25-OHD

insufficient
25-OHD

low PTH

0
0

10

20

30

40

50

60

70

44
42
40
38
36
34
32
30
28
26
24
22
20
18
16
14
12
10
8
6
4
2
0

c
a

Vitamin D 25-OHD

Vitamin D 1,25-OHD

PTH

12

44
42
40
38
36
34
32
30
28
26
24
22
20
18
16
14
12
10
8
6
4
2
0

Serum parathyroid hormone (pg/ml)

Serum 25-OHD (ng/ml) and 1,25-OHD (pg/ml)

Serum 25-OHD (ng/ml)

FIG. 1. A: Serum 25-OH vitamin D versus parathyroid hormone levels in children

with intractable epilepsy. B: Serum vitamin
D (25-OHD and 1,25-OHD) and parathyroid hormone (PTH) status in children with
intractable epilepsy at baseline and over
15 months of treatment with the ketogenic
diet, presented as means (boxes) and
standard errors (whiskers). Serum levels
were significantly different for time points
a and b at p < 0.001 and from time points
b and c at p = 0.003.

15

Months on Ketogenic Diet

studies linked P450 enzymeinducing AEDs with poor vitamin D status. The P450 induction increased conversion
of vitamin D to inactive metabolites, directly inhibiting
intestinal calcium absorption and increasing bone
turnover. Previous studies were conducted on institutionalized individuals who were not exposed to sunlight, an
important source of vitamin D. Baer et al. (1997) investigated vitamin D, calcium, and bone status in 338 children
with developmental disabilities living at home by ambulatory and AED status. They found poor vitamin D and
calcium intake in all groups: 70% and 56%, respectively,
consumed less than the recommended intake. Vitamin D
intake was positively associated with serum calcium and
25-OHD. In addition, nonambulatory status and AED use
were negatively associated with 25-OHD and calcium levels. The authors raised concerns for poor bone health and
suggested vitamin D and calcium supplementation.
Today an awareness exists of poor intake of vitamin
D and calcium among all children (Black et al., 2002).
In our study, 47% of the children had a baseline intake of vitamin D that was less than recommended for
healthy children. Four percent had 25-OHD in the defi-

cient range, and more than half had levels in the insufficient range, placing them at risk for osteoporosis and other
sequelae.
What has changed since the study by Baer et al. reported
in 1997? Most of the subjects (72%) in the older study
were using PHT or PB, and the severity of the epilepsy
was not described. In our study, only 4% and 8% were
treated with PHT and PB, respectively. The remaining
subjects were using the newer AEDs. This is the first study
investigating vitamin D status of children with intractable
epilepsy treated with the newer AEDs.
Currently no nutritional guidelines exist for children
with epilepsy. Most vitamin D and calcium intake in children is provided in vitamin Dfortified dairy products. To
meet calcium and vitamin D needs in a nondairy diet is
impossible, unless the diet is carefully planned and individualized supplements are provided. As neurologists and
pediatricians, we must consider the diets of these children,
encourage fortified dairy consumption, and consider supplementation.
Hahn et al. examined vitamin D status in a small number of children on the KD, and suggested that the KD had

Epilepsia, Vol. 48, No. 1, 2007

70

A. G. C. BERGQVIST ET AL.
TABLE 2.

Predictors of Serum Vitamin D-25-OHD

Coefficient

Multiple regressions
Baseline
Age
Gender (male)
Fall
Winter
Spring
Number of AEDs
Generalized epilepsy
Vitamin D intake >100%
Calcium intake >100%
Constant
Longitudinal mixed-effects models
Baseline to 3 mo
3-mo interval (B3 mo)a
Gender (male)
Age (yr)
Fall
Winter
Spring
Number of AEDs
Generalized epilepsy
Constant
3 to 15 mo
1-mo interval (315 mo)b
Gender (male)
Age (yr)
Fall
Winter
Spring
Generalized epilepsy
Constant

SE

Obs

Adjusted R2

p Value

1.19
4.99
0.72
7.79
10.83
6.85
9.72
12.99
7.57
53.36

0.82
3.73
5.54
5.52
5.50
2.34
3.51
4.26
5.36
9.08

0.158
0.189
0.898
0.167
0.057
0.006
0.009
0.004
0.167
0.000

44

0.38

0.002

8.04
0.28
0.05
5.83
12.84
7.54
2.16
9.15
38.18

2.21
3.34
0.63
3.78
3.84
3.66
2.04
3.20
6.88

0.000
0.932
0.933
0.125
0.001
0.039
0.290
0.004
0.000

84

0.33

<0.0001

0.48
0.36
0.96
1.10
7.74
0.06
4.41
46.99

0.16
3.33
0.70
2.00
2.07
2.06
3.35
6.04

0.003
0.915
0.168
0.583
0.000
0.977
0.188
0.000

138

0.19

<0.0001

a For the baseline to 3-mo model, the change in serum vitamin D 25-OHD (ng/ml) is presented as the change for the
entire 3-mo interval.
b For the 3-mo to 15-mo model, the change in serum vitamin D 25-OHD is presented as the ng/ml change per month.

negative vitamin D and bone-health effects (Hahn et al.,

1979). They studied five children on the KD and AEDs for
an average of 2.5 years and documented deficient vitamin
D, elevated PTH, and low bone mineral density by using
an older technology, single-photon absorptiometry. They
postulated that the KD resulted in a negative effect on
vitamin D status because of chronic acidosis placing demands on the bone mineral content for buffering capacity,
and that acidosis further interfered with renal conversion
of 25-OHD to 1,25-OHD.
In our study, vitamin D status improved on the KD,
including supplementation with only 18% with vitamin
D intake less than recommended. Improved 25-OHD
blood levels confirmed the improved dietary intake. By
3 months, 25-OHD increased, and the PTH declined. After this initial improvement, the 25-OHD levels declined
over the next 12 months. The etiology for this decline is unknown. Possibilities include a change in vitamin D absorption or metabolism after months of exposure to an almost
all-fat diet, or a decrease in adherence to supplements.
Noncompliance with the diet may result in an immediate
adverse effect (i.e., return of seizures), while forgetting

Epilepsia, Vol. 48, No. 1, 2007

a supplement will have no immediate detectable adverse

effect.
In summary, children with intractable epilepsy treated
with newer AEDs had poor vitamin D intake and blood
levels. Vitamin D supplementation included in the KD
treatment improved the vitamin D status over the first 3
months. Over the next 12 months of KD, a decline in the
25-OHD level was observed. Future research is needed to
investigate the relation between vitamin D status and bone
health in children on the KD.
Acknowledgment: We are grateful to all the children, parents, and other care providers for their participation and cooperation in this research study. We
thank the CHOP Ketogenic Diet Team, General Clinical
Research Center, and Nutrition and Growth Laboratory
staff for their dedication and commitment to this project.
We also thank Paul Gallagher for statistical analysis
support.
This study was supported in part by RRK-23 16074 and
General Clinical Research Center (MO1RR00240), the
Nutrition Center of the Childrens Hospital of Philadelphia, and the Catharine Brown Foundation.

VITAMIN D AND THE KETOGENIC DIET

REFERENCES
Baer MT, Kozlowski BW, Blyler EM, Trahms CM, Taylor ML, Hogan
MP. (1997) Vitamin D, calcium, and bone status in children with
developmental delay in relation to anticonvulsant use and ambulatory
status. American Journal of Clinical Nutrition 65:10421051.
Bergqvist AG, Schall JI, Gallagher PR, Cnaan A, Stallings VA. (2005)
Fasting versus gradual initiation of the ketogenic diet: a prospective, randomized clinical trial of efficacy. Epilepsia 46:1810
1819.
Black RE, Williams SM, Jones IE, Goulding A. (2002) Children who
avoid drinking cow milk have low dietary calcium intakes and poor
bone health.[see comment]. American Journal of Clinical Nutrition
76:675680.
Cioffi M, Corradino M, Gazzerro P, Vietri MT, Di Macchia C, Contursi
A, Colicigno R, Catalano T, Molinari AM. (2000) Serum concentrations of intact parathyroid hormone in healthy children. Clinical
Chemistry 46:863864.
Dent CE, Richens A, Rowe DJ, Stamp TC. (1970) Osteomalacia with
long-term anticonvulsant therapy in epilepsy. British Medical Journal 4:6972.
Devinsky O, Westbrook L, Cramer J, Glassman M, Perrine K, Camfield C. (1999) Risk factors for poor health-related quality of life in
adolescents with epilepsy. Epilepsia 40:17151720.
Eadie MJ. (1987) The risk-benefit ratio of anticonvulsant drugs. Medical
Toxicology & Adverse Drug Experience 2:324337.
Fitzpatrick LA. (2004) Pathophysiology of bone loss in patients receiving anticonvulsant therapy. Epilepsy & Behavior 5(suppl 2):S3
S15.
Gough H, Goggin T, Bissessar A, Baker M, Crowley M, Callaghan N.
(1986) A comparative study of the relative influence of different

71

anticonvulsant drugs, UV exposure and diet on vitamin D and calcium metabolism in out-patients with epilepsy. Quarterly Journal of
Medicine 59:569577.
Hahn TJ, Halstead LR, Devivo DC. (1979) Disordered mineral
metabolism produced by ketogenic diet therapy. Calcified Tissue
International 28:1722.
Heaney RP. (2004) Functional indices of vitamin D status and ramifications of vitamin D deficiency. American Journal of Clinical Nutrition
80:1706S1709S.
Hollis BW. (2005) Circulating 25-hydroxyvitamin D levels indicative of
vitamin D sufficiency: implications for establishing a new effective
dietary intake recommendation for vitamin D. Journal of Nutrition
135:317322.
Hollis BW, Kamerud JQ, Selvaag SR, Lorenz JD, Napoli JL. (1993)
Determination of vitamin D status by radioimmunoassay with a 125 Ilabeled tracer. Clinical Chemistry 39:529533.
Lohman TG, Roche AF, Martorell R. (1988) Anthropometric standardization reference manual. Human Kinetics Books, Champaign, IL.
Institute of Medicine. (1997) Dietary reference intakes for calcium,
phosphorus, magnesium, vitamin D and fluoride. National Academy
Press, Washington, DC.
Institute of Medicine. (2002) Dietary reference intake for energy: carbohydrate, fiber, fat, fatty acids, cholesterol, protein, and amino acids.
National Academy Press, Washington, DC.
Pack AM, Morrell MJ, Marcus R, Holloway L, Flaster E, Done S, Randall
A, Seale C, Shane E. (2005) Bone mass and turnover in women with
epilepsy on antiepileptic drug monotherapy. Annals of Neurology
57:252257.
Vestergaard P, Tigaran S, Rejnmark L, Tigaran C, Dam M, Mosekilde
L. (1999) Fracture risk is increased in epilepsy. Acta Neurologica
Scandinavica 99:269275.

Epilepsia, Vol. 48, No. 1, 2007