Download Lipids, Lipoproteins, Apolipoproteins, Selected Trace Elements and

C Basic & Clinical Pharmacology & Toxicology 2006, 98, 599–603.
Printed in Denmark . All rights reserved
Copyright C
ISSN 1742-7835
Lipids, Lipoproteins, Apolipoproteins, Selected Trace
Elements and Minerals in the Serum of Children on Valproic
Acid Monotherapy
George A. Karikas1, Kleopatra H. Schulpis2, Anastasia Bartzeliotou3, Theodore Karakonstantakis3, Sophia Georgala4,
Ino Kanavaki3, Elizabeth Demetriou3 and Ioannis Papassotiriou3
Parenteral Nutrition and Pharmacokinetics Unit, ‘‘Aghia Sophia’’ Children Hospital, 2Institute of Child Health,
‘‘Aghia Sophia’’ Children’s Hospital, 3Department of Clinical Biochemistry, ‘‘Aghia Sophia’’ Children’s Hospital,
Athens, and 4Dermatological Clinic, Athens University Medical School, Athens, Greece
1
(Received November 10, 2005; Accepted January 4, 2006)
Abstract: We evaluated the serum levels of lipids, lipoproteins, apolipoproteins, along with a number of minerals and
trace elements such as Ca, Mg, Cu and Zn in a group of children after 6 months of valproic acid monotherapy. Thirty
patients with seizures, mean age, 9.8∫2.6 years and 79 healthy children (controls), mean age, 10.9∫3.2 years, formed the
two styd groups. The patient group was treated with valproic acid (27.9∫14.8 mg/kg/24 hr). Patients underwent clinical
and laboratory evaluations including liver function tests, NH3, lipid, mineral and selected trace element levels before and
after six months on valproic acid treatment, whereas controls only one evaluation. Liver function data and NH3 levels
were found to be elevated in the group of patients, whereas albumin level was reduced. Triglycerides, total cholesterol,
HDL-C, apolipoprotein (ApoA)-1, Apo B and Ca concentrations were found relative to control values, LDL-C, VLDLC, Mg, Cu, Zn, were measured significantly altered (P⬍0.0001) compared to controls. The ratios ApoA-1/ApoB, HDLC/ApoA-1, LDL-C/Apo B, which were closely related to the size of LDL particles, where correlated with Zn/Cu
(P⬍0.001). Serum lipid profile, especially LDL size, indirectly evaluated for the first time and metal levels were found to
be significantly changed, after six months on valproic acid monotherapy, suggesting a possible risk of developing coronary
heart disease. Since valproic acid is a long-term treatment, it could be recommended that the incorporation of measurements of lipids, lipoproteins, apolipoproteins and trace elements in the ‘‘follow up’’ laboratory testing could be a preventive
measure.
Until the exciting discovery of the anticonvulsant properties
of valproic acid it was initially used as an organic solvent.
Since its introduction into clinical use in the late 1960s, it
has been proved effective as both as a component of polytherapy and as a sole medication in the simple and complex
epileptic seizures treatment (Rettie et al. 1988). Preclinical
studies have been carried out during the past four decades
to investigate the different mechanisms of valproic acid actions. The mechanisms of valproic acid, which seem to be of
clinical importance, include increased GABAergic activity,
reduction in excitatory neurotransmission and modification
of monoamines. Valproic acid has many side effects when
compared with other antiepileptic drugs (Ceniz et al. 2000).
The more common dose-related phenomena specific to valproic acid include weight gain, tremor, skin rash, and hair
loss (alopecia), and do not usually abate with continued
treatment, but may respond to a weaning of the dosage or
to a change in the dosing regimen (Wallace 1996). Fatal
hepatoxicity is a rare specific adverse reaction, not related
Author for correspondence: Ioannis Papassotiriou, Department of
Clinical Biochemistry, ‘‘Aghia Sophia’’ Children’s Hospital, 115 27
Athens, Greece (fax π30 210 7467171, e-mail biochem/paidon-ag
iasofia.gr, jpapasotiriou/ath.forthnet.gr).
to dose that has been reported to coincide with valproic
acid treatment. Nevertheless, valproic acid seems to impair
liver mitochondrial function, resulting in a low biotinidase
activity (Schulpis et al. 2001).
The mechanisms of epileptogenesis are not yet well established. The membrane lipid peroxidation due to increase in
free radicals and the decrease in activities of antioxidant
defense mechanisms have been suggested to be partially involved in some forms of epilepsy (Jesberger & Richardson
1991). Kurekci et al (1995) have demonstrated that valproic
acid administration seems to enhance the clearance of selenium (Se), copper (Cu) and zinc (Zn), thereby decreasing
the synthesis of free radicals scavenging enzymes.
Dreon et al. (1994 & 2000) found that the LDL-C/Apo
B ratio correlates well with LDL size. It has been used as a
marker of atherogenicity of LDL (Campos et al. 1992),
while variations in HDL levels are partially due to differences in fractional catabolic rates. Large HDL particles
with high HDL-C/HDL-Apo AI plus Apo AII ratio exhibit
low catabolic rates (Brinton et al. 1991).
Furthermore, according to the Klevay hypothesis (Klevay 1975) coronary heart disease is predominately a disease
of Zn and Cu metabolic imbalance caused by a high Zn/Cu
ratio, and principally due to Cu deficiency. In fact, the latter
deficiency increases total serum cholesterol levels through
600
GEORGE A. KARIKAS ET AL.
Table 1.
Clinical findings, drug serum levels and biochemical data in patients
(before and after therapy) on valproic acid (VPA) controls group.
Patients (NΩ30)
Age (years)
Therapy (months)
Doseπ (mg/kg/24 hr)
VPAππ (mg/ml)
SGOT (U/l)
SGPT (U/l)
ALP (U/l)
NH3 (mg/dl)
Albumin (g/l)
gGT (U/l)
Before
After
Controls
(NΩ79)
25.0∫10.0
23.1∫9.0
58.2∫10.1
25.3∫12.2
4.0∫0.5
12.6∫3.5
9.8∫2.6
6.5∫1.8
27.9∫14.8
62.9∫19.3
39.8∫8.0*
39.6∫15.0*
86.9∫27.0*
47.3∫3.8*
3.6∫0.6*
17.3∫7.0*
10.9∫3.2
–
–
–
23.0∫9.0
21.0∫8.9
60.0∫15.0
29.8∫9.0
4.2∫0.4*
12.0∫4.0
Values are expressed as: mean∫S.D.
* P⬍0.001, versus before therapy as well as versus controls.
π
Range: 8.2–52.5 mg/kg/24 hr, ππ Range: 32.8–109.3 mg/ml.
Abbreviations: SGOT: serum glutamate oxaloacetate transaminase;
SGPT: serum glutamate pyruvate transaminase; NH3: ammonia;
gGT: g-glutamyltransferase.
modulating the control biosynthesis and both metals regulate superoxide dismutase (Yount et al. 1990).
Finally, Cu and iron (Fe) have been shown to be prooxidants and to play an important part in lipoprotein peroxidation (Reunanen et al. 1996). For these reasons, a number
of studies investigated whether serum lipids, lipoproteins,
apolipoproteins and trace elements are affected in one way
or another after antiepileptic therapy (Calandre et al. 1991;
Franzoni et al. 1992; Kaji et al. 1992; Kuzuya et al. 1993 &
2002; Reunanen et al. 1996; Sozuer et al. 1997; Verroti et
al. 1997 & 2002). Although the experimental data available
so far regarding a number of antiepileptic drugs such as
carbamazepine, phenobarbital, phenytoin are well documented, the effects of valproic acid on lipids, lipoproteins,
apolipoproteins and trace elements levels are still to a certain extent rather poor and controversial (Reddy 1985; Verroti et al. 1998).
In an attempt to clarify the valproic acid effects regarding
the above specific biochemical parameters, we evaluated
serum valproic acid, hepatic enzymes, lipids, lipoproteins,
apolipoproteins, Ca, Mg, Cu, and Zn concentrations along
with the ratios HDL/ApoA-1, LDL/Apo B indirectly correlated with size of LDL particles as well as Ca/Mg, Zn/Cu
in patients after six months on valproic acid treatment.
Materials and Methods
The study was approved by the Greek Ethical Committee. Thirty
valproic acid-treated patients with seizures, mean age, 9.8∫2.6, and
79 healthy school children of comparable mean age 10.9∫3.2 (control group) participated in the present study. The patients were
treated with valproic acid (27.9∫14.8 mg/kg/24 hr), for six months.
Valproic acid serum levels were mostly measured within therapeutic
range (50–100 mg/ml serum). Biochemical analyses including liver
function tests, lipids, selected trace elements and minerals were performed in patients before and 6 months after entering valproic acid
treatment, whereas in controls only once.
Blood chemistry and drug analysis. All patients and controls underwent clinical examination and blood tests, such as full blood counts,
liver function tests, including SGOT, SGPT, NH3, alkaline phosphatase (ALP), gGT, and serum albumin using routine methods. Serum
total cholesterol, triglycerides, HDL-C, LDL-C and VLDL-C were
measured using the ADVIA-1650 Chemistry System (Bayer Corporation, Tarrytown, NY, USA), while Apo-A1 and Apo B were determined by latex-particle-enhanced immunonephelometric assays on
the BN ProSpec nephelometer (Dade Behring, Liederbach, Germany). Quality control has previously been indicated (Srinivasan &
Berenson 1996). Interassay variation coefficients for total cholesterol, triglycerides, HDL-C, Apo AI, and Apo B were 3.5%, 3.7%,
5.1%, 5.0% and 4.5%, respectively. Duplicate serum samples were
analyzed for Ca, Mg, Cu and Zn by atomic absorption spectrometry using the Perkin Elmer A-Analyst 800 atomic absorption
spectrometer. Lanthanium chloride (Merck Darmstadt, Germany)
was added to avoid interferences. Aliquots of human serum were
used as an internal control to assess precision in the mineral determinations. The interassay relative standard deviation was 2.1%,
4.5%, 4.0%, 3.0% for Ca, Mg, Cu, and Zn, respectively. Certified
reference materials (CRM 63, CRM 185, Community Bureau of
Reference Brussels) were used to assess accuracy. All laboratory
equipment employed in the trace elements analysis was washed with
nitric acid to avoid contamination. Lanthanium chloride 0.1% and
nitric acid 1% (AAS grade) were used for the preparation of dilutions (1:50 for Ca and Mg, 1:5 for Zn and 1:2 for Cu and standards of the mineral/trace element analysis).
Valproic acid level determination was performed by fluorescence
polarization immunoassay (FIA) method, using the ABBOTT
(TDXR system). In FIA the antigen label is usually fluoroscein, a
substance that emits light at 520 nm, when excited by light at a
wavelength of 490nm, although reaction end-product fluorescence
may also be detected. The sampling time was immediately before
the next dose (trough) (Cobb & Gother 1982).
Statistical analysis. Data were analyzed with ANOVA followed by
Student’s t-test and pair observations t-test or Dunnet test. Corre-
Table 2.
Lipids, lipoproteins, apolipoproteins, minerals and trace elements
in patients on valproic acid (VPA) (before and after therapy) versus
control group.
Patients (NΩ30)
Triglycerides (mmol/l)
Total cholesterol (mmol/l)
LDL-C (mmol/l)
HDL-C (mmol/l)
VLDL-C (mmol/l)
Total cholesterol/HDL-C
Apo A-1 (g/l)
Apo B (g/l)
Apo A-1/Apo B
HDL-C/Apo A-1π
LDL-C/Apo Bπ
Ca (mmol/l)
Mg (mmol/l)
Cu (mmol/l)
Zn (mmol/l)
Ca/Mg
Zn/Cu
Before
After
Controls
(NΩ79)
1.00∫0.50
4.30∫0.78
2.40∫0.50
1.22∫0.50
0.59∫0.20
3.49∫0.80
1.48∫0.40
0.78∫0.30
2.32∫0.80
0.31∫010
1.30∫0.18
2.35∫0.20
0.80∫0.10
16.5∫2.50
12.5∫1.60
2.80∫0.40
0.75∫0.12
0.97∫0.42
4.50∫0.84
2.95∫0.75*
1.09∫0.24
0.45∫0.20**
4.12∫0.30*
1.18∫0.23
0.77∫0.18
1.53∫0.20**
0.92∫0.10*
3.88∫0.33*
2.41∫0.17
0.94∫0.17*
14.8∫4.10*
15.7∫3.6**
2.63∫0.37
1.13∫0.33**
1.10∫0.50
4.19∫0.80
2.41∫0.50
1.20∫0.50
0.61∫0.30
3.56∫0.90
1.50∫0.30
0.80∫0.30
2.28∫0.80
0.28∫0.05
1.28∫0.15
2.30∫0.17
0.82∫0.08
16.9∫3.0
12.8∫1.5
2.96∫0.30
0.72∫0.15
Values are expressed: mean∫S.D.
* P⬍0.001 and ** P⬍0.0001 versus before therapy and versus control.
π
Calculated from data expressed as mg/dl.
601
VALPROIC ACID EFFECT ON LIPIDS AND MINERALS
Table 3.
Pearson correlations (r) between lipids, lipoproteins (Apo) and minerals and trace elements in serum of patients on valproic acid (VPA)
therapy (NΩ30) and control group (NΩ79).
Groups
A
Ca
Mg
Ca/Mg
Cu
Zn
Zn/Cu
Triglycerides
A
B
C
A
B
C
A
B
C
A
B
C
A
B
C
A
B
C
A
B
C
A
B
C
A
B
C
A
B
C
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
0.43**
–
–
–
–
–
0.45**
0.42*
0.44**
0.44**
–
–
0.48**
0.45**
0.45*
0.49**
–
–
–
–
–
–
–
–
0.52**
–
–
–
–
–
ª0.46*
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
ª0.42*
–
–
0.51**
–
–
ª0.52**
–
–
–
–
–
–
–
–
–
–
–
ª0.51***
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
ª0.41*
ª0.45**
ª0.41*
–
–
–
ª0.52**
ª0.44*
ª0.46**
–
–
–
–
–
–
–
–
–
–
–
–
ª0.45**
ª0.42**
ª0.40*
0.54***
–
–
0.49***
0.50***
0.46**
–
–
–
–
–
–
–
–
–
–
–
–
0.52***
–
–
0.44*
–
–
0.64***
–
–
Total cholesterol
LDL-C
HDL-C
Total cholesterol/HDL
Apo AI
Apo B
Apo A I/Apo B
HDL-C/Apo AI
LDL-C/Apo B
P⬍0.05, **P⬍0.01, ***P⬍0.001.
A: Patients after 6 months on VPA therapy, B: Patients before entering VPA therapy and C: control group.
lations were determined by Pearson’s test. P⬍0.05 was considered
statistically significant.
Results
Table 1 presents the mean age of patients and controls, duration and dose of therapy. Valproic acid serum levels were
62.9∫ 19.3 mg/ml. Statistically significant elevations of the
liver enzymes and NH3 concentrations were found in the
group of patients after treatment as compared to the pretreatment values as well as to controls. Albumin levels were
found to be reduced in the same group of patients.
As shown in table 2, LDL-C, VLDL-C, the ratios total
cholesterol/HDL-C, Apo A-1/Apo B, HDL-C/Apo A-1,
LDL-C/Apo B, as well as Mg, Cu, Zn, and the ratio Zn/Cu
differed significantly in the patient group after treatment,
as compared to the control group. Other biochemical data
such as total cholesterol, triglycerides, HDL-C, Apo A-1,
Apo B, as well as Ca, and the ratio Ca/Mg were similar
among all groups.
Table 3 presents positive correlations between Mg and
HDL-C and Apo A-1 in all the groups whereas the mineral
correlated with total cholesterol, Apo B and the ratio LDLC/Apo B only in patients on valproic acid. The trace element Cu negatively correlated with triglycerides, LDL-C
and Apo B and positively with total cholesterol in group A.
Additionally, Zn was negatively correlated with HDL-C and
Apo A-1 in all the studied groups and the ratio Zn/Cu with
triglycerides and LDL-C in all the groups. On the contrary,
the above ratio of trace elements positively correlated with
the ratios Apo A-1/Apo B and LDL/Apo B after start of
valproic acid treatment.
Discussion
Many studies on side-effects of antiepileptic drugs, including valproic acid, have been conducted over the last years,
studying either the alteration of lipid profile, as found in
our patients, or changes in trace elements in blood. The
administration of the acid enhances the clearance of Se, Cu,
and Zn and thereby decreases the synthesis of free radical
scavenging enzymes. Very recently we have reported that
valproic acid therapeutic and/or toxic levels induced DNA
oxidative damage in children due to their low antioxidant
capacity (Schulpis et al. 2006). Nevertheless, current knowledge about the impact of antiepileptic drugs on plasma
trace elements and antioxidant enzymes is partial and
mainly controversial (Kurekci et al. 1995).
The minor elevations of transaminases (SGOT and
602
GEORGE A. KARIKAS ET AL.
SGPT) which are frequent and appeared to be valproic acid
dose-related (Wallace 1996), were also observed in the present study. A number of common antiepileptic drugs, such
as carbamazepine and phenobarbital, have affected serum
lipid, lipoprotein and apolipoprotein levels (Sozuer et al.
1997; Eiris et al. 2000; Verroti et al. 2002). In contrast, similar studies on valproic acid have shown uncertain results.
Furthermore, the duration the treatment, short or long, has
also produced conflicting findings (Verroti et al. 2002).
It is already well established that plasma lipoprotein abnormalities may be essential for the common occurrence of
atherosclerotic vascular diseases. These abnormalities include elevated concentrations of LDLs and VLDLs and reduced concentrations of HDLs, as generally estimated from
measurements of plasma total cholesterol, triglycerides and
HDL-C. However, other studies indicate that the most pronounced lipoprotein abnormalities in patients with early
onset of coronary heart diesease are high triglyceride and
low HDL-C with fewer elevations of LDL-C. In these individuals, the LDL particles are somewhat smaller and denser
than in those with a more favourable lipoprotein profile
(Havel 2000).
Although the effects on the hepatic enzyme-inducing
antiepileptics, such as phenobarbital and carbamazepine,
are well established, the effect of valproic acid on lipid profile remains so far pretty unclear. Some articles claim no
effect whatsoever on plasma concentration of total cholesterol, HDL, or its components (Franzoni et al. 1992; Kurekci et al. 1995), whereas others have demonstrated significant changes in lipids, lipoproteins and apolipoproteins
(Calandre et al. 1991; Aynaci et al. 2001; Verroti et al. 2002).
In a recent study patients treated with antiepileptic drugs
demonstrated significant changes in lipid and lipoproteins,
but when the authors re-evaluated the same three groups
(carbamazepine, phenobarbital, valproic acid) of children 1
year after the end of treatment, a complete return to normal
of all parameters was observed (Verroti et al. 2002). These
results indicate that the duration of treatment seems to play
an important role in lipid serum concentration. For these
reasons we selected a certain uninvestigated six-month
period of treatment. In our study, a six month valproic acid
monotherapy induced a significantly increase in LDL-C
levels, while VLDL-C levels were decreased, compared to
control group as well as before entering treatment. The
ratio Total Chol/HDL-C was found significantly higher,
while the triglycerides, total cholesterol and HDL-C levels
were unaltered by valproic acid.
Interestingly, although the ApoA-1 and Apo B in
children treated with valproic acid were observed to be
within the normal range, the ratios Apo A-1/Apo B, HDLC/Apo A-1, and LDL-C/Apo B were significantly higher,
compared to control values. Low Cu level, as found in our
patients, was related to the presence of larger LDL particles
(LDL/Apo B) (Dreon et al. 1994). According to previous
studies (Jesberger & Richardson 1991; Schulpis et al.
2001 & 2006) a high LDL-C/Apo B ratio was related to the
presence of larger and less atherogenic LDL particles, as
was found in our patients. Zn levels were observed to be
elevated. Cu, Zn alterations have been observed in a previous study (Kuzuya et al. 1993), whereas recent experiments in rats have found that a high excretion of Cu into
the bile induced by valproic acid could upset the homeostatic balance of Cu and cause abnormalities in serum concentration (Speich et al. 1984).
The present findings show that the elevated Zn/Cu ratio
in the patients on valproic acid therapy clearly appeared to
be related not only to the reduced Cu levels but also to
some lipid and lipoprotein alterations. An interesting hypothesis can be drawn from our data, as the Zn/Cu ratio
correlates with LDL-C/Apo B ratio which was also found
to be statistically higher in comparison with the control
group. Magnesium plays an important role in lipoprotein
metabolism which is essential for fatty acid and protein synthesis and critical in energy-requiring metabolic processes
as an adenosine triphosphate co-factor (Speich et al. 1984).
In the present study Mg was found to be significantly elevated after valproic acid treatment, whereas Ca and Ca/Mg
ratio was unaltered. Ca/Mg ratio appeared to have no correlation to the alterations in the lipid, lipoprotein and apoliprotein profile. This may be due to the homeostatic control
of calciotropic hormones, which permits a little variation
of Ca serum levels (Al-Chadi et al. 1994).
Finally, the present study confirms all the interesting correlations of previous studies, as the Zn/Cu ratio was positively correlated with LDL-C and with the LDL-C/Apo B
ratio (LDL atherogenicity), but negatively correlated with
triglycerides and total cholesterol levels (Schulpis et al.
2002 & 2003).
In conclusion, our data suggests that mineral and trace
element alterations, such as Mg, Cu, Zn, observed after six
months on valproic acid may play a substantial role in the
lipoprotein metabolism as found in our study group. These
metals are involved in the metabolism and especially in the
size of LDL particles, and the latter was indirectly evaluated
for the first time in patients on valproic acid monotherapy.
High lipid levels, trace element and mineral alterations in
children after 6 months on valproic acid therapy, suggest
that the evaluation of lipids, lipoproteins, apolipoproteins
along with minerals and trace elements should be incorporated in routine periodical laboratory tests as a useful preventive measure against a possible development of coronary
heart disease, since valproic acid administration is a longlasting and very often life-long therapy.
References
Al-Chadi, S. M. G., E. C. Cameron & R. A. L. Sutton: Magnesium
deficiency: pathophysiologic and clinical overview. Amer. J. Kidney. Dis. 1994, 24, 737–752.
Aynaci, F. M., F. Orhan, A. Orem, S. Yildirmis & Y. Gedik: Effect
of antiepileptic drugs on plasma lipoprotein(a) and other lipid
levels in childhood. J. Child Neur. 2001, 16, 367–369.
Brinton, E. A., S. Eiserberg & J. L. Breslow: Increase Apo AI and
Apo AII fractional catabolic rate in patients with low HDL-C
with or without hypertriglyceridemia. J. Clin. Invest. 1991, 87,
536–544.
VALPROIC ACID EFFECT ON LIPIDS AND MINERALS
Calandre, E. P., C. Rodriquez-Lopez, A. Blazquez & D. Cano:
Serum lipids, lipoproteins and apolipoproteins A and B in epileptic patients treated with valproic acid, carbamazepine or phenobarbital. Acta neurol. scand. 1991, 83, 250–253.
Campos, H., J. Genest, E. Blijlevens, J. R. MacNamara, J. L. Jenner, J. M. Ordovas, P. W. Wilson & E. J. Schaeffer: Low density
lipoprotein size and coronary artery disease. Arterioscler.
Thromb. 1992, 12, 187–195.
Cengiz, M., A. Yukcel & M. Seven: The effects of carbamazepine
and valproic acid on the erythrocyte glutathione, glutathione peroxidase, superoxide dismutase and serum lipid peroxidation in
epileptic children. Pharmacol. Res. 2000, 41, 423–425.
Cobb, M. & S. Gotcher: Fluorescence immunoassay in the clinical
laboratory. Am. J. Med. Technol. 1982, 48, 671–677.
Cengiz Dreon, D. M., H. Fernstorm, D. Miller, et al.: Low density
lipoprotein subclass patterns and lipoprotein response to a reduced fat diet in men. FASEB J. 1994, 8, 121–126.
Dreon, D. M., H. Fernstorm, P. Williams, et al.: Reduced LDL
particle size in children consuming a very low fat diet is related
to parental LDL-subclass patterns. Amer. J. Clin. Nutr. 2000, 71,
1611–1616.
Eiris, J., M. I. Novo-Rodriguez, M. Del Rio, et al.: The effects on
lipid and apolipoprotein serum levels of long term carbamazepine, valproic acid and phenobarbital therapy in children with epilepsy. Epilepsy Res. 2000, 41, 1–7.
Franzoni, E., M. Govoni, S. D’Addato, et al.: Total cholesterol,
high-density lipoprotein cholesterol, and triglycerides in children
receiving antiepileptic drugs. Epilepsia 1992, 33, 932–935.
Havel, R.: Genetic underpinnings of LDL size and density:a role
for hepatic lipase. Amer. J. Clin. Nutr. 2000, 71, 1390–1391.
Jesberger, J. A. & J. S. Richardson: Oxygen free radicals and brain
dysfunction. Int. J. Neurosci. 1991, 57, 1–17.
Kaji, M., M. Ito, T. Okuno, et al.: Serum copper and zinc levels in
epileptic children with valproic acid treatment. Epilepsia 1992,
33, 555–557.
Klevay, L. M.: Coronary heart disease: the zinc/copper hypothesis.
Amer. J. Clin. Nutr. 1975, 28, 764–70.
Kurekci, A. E., F. Alpay, S. Tanindi, et al.: Plasma trace elements,
plasma glutathione peroxidase levels in epileptic children, receiving antiepileptic drug therapy. Epilepsia 1995, 36, 600–604.
Kuzuya, T., T. Hasegawa, K. Shimizu, et al.: Effect of antiepileptic
drugs on serum zinc and copper concentrations in epileptic patients. Int. J. Clin. Pharmacol. Therap. Toxicol. 1993, 31, 61–65.
Kuzuya, T., K. Amioka & T. Nabeshima: Valproic acid increases
biliary copper excretion in the rat. Epilepsy Res. 2002, 51, 279–
285.
Reddy, M. N.: Effect of anticonvulsant drus on plasma total chol-
603
esterol, high density lipoprotein cholesterol, and apoliprotein A
and B in children with epilepsy. Proc. Soc. Exp. Biol. Med. 1985,
180, 359–363.
Rettie, A. E., M. Bobert, A. W. Rettenmeier, et al.: Cytochrome P450-catalyzed desaturation of valproic acid in vitro. J. Biol.
Chem. 1988, 263, 13733–13738.
Reunanen, A., P. Knekt, J. Marniemi, et al.: Serum calcium, magnesium, copper and zinc and risk of cardiovascular death. Eur.
J. Clin. Nutr. 1996, 50, 431–437.
Schulpis, K. H., G. A. Karikas, J. Tjamouranis, et al.: Low serum
biotinidase activity in children with valproic acid monotherapy.
Epilepsia 2001, 42, 1359–1362.
Schulpis, K. H., G. A. Karikas & E. D. Papakonstantinou: Homocysteine and other vascular risk factors in patients with phenylketonuria on diet. Acta paediatr. 2002, 91, 1–5.
Schulpis, K. H., S. Tsakiris, G. A. Karikas, M. Moukas & P.
Behrakis: Effect of diet on total antioxidant status in PKU patients. Eur. J. Clin. Nutr. 2003, 57, 383–387.
Schulpis, K., C. Lazaropoulou, S. Regoutas, G. Karikas, A. Margeli, S. Tsakiris & I. Papassotiriou: Valproic acid monotherapy induces DNA oxidative damage. Toxicology 2006, 217, 228–232.
Sozuer, D. T., D. Atakil, O. Dogu, et al.: Serum lipids in epileptic
children treated with carbamazepine and valproic acid. Eur. J.
Pediatr. 1997, 156, 565–567.
Speich, M. S., S. Gelot, R. Arnaud, et al.: Multiple and simple
correlations between magnesium, calcium, zinc, potassium, total
and HDL-C in 111 reference subjects. Magn. Bull. 1984, 4, 137–
141.
Srinivasan, S. R. & C. S. Berenson: In: Serum lipoproteins in children
and methods for study, vol. III. Ed.: L. A. Lewis. CBC Press Boca
Raton, 1996, p. 185–196.
Verroti, A., S. Domizio, B. Angelozzi, et al.: Changes in serum lipids
and lipoproteins in epileptic children treated with anticonvulsants. J. Paediatr. Child Health 1997, 33, 242–245.
Verrotti, A., F. Basciani, S. Domizio, et al.: Serum lipids and lipoproteins in patients treated with antiepileptic drugs. Pediatr. Neurol. 1998, 19, 364–367.
Verrotti, A., F. Basciani, D. Trotta, et al.: Serum copper, zinc, selenium, glutathione peroxidase and superoxide dismutase levels in
epileptic children before and after 1 year of sodium valproic acid
and carbamazepine therapy. Epilepsy Res. 2002, 48, 71– 75.
Wallace, S. J.: A comparative review of the adverse effects of anticonvulsant in children with epilepsy. Drug Safe 1996, 15, 378–
393.
Yount, N. Y., D. J. McNamara, A. A. Al-Ohtman, et al.: The effect
of copper deficiency on rat hepatic 3-hydroxy-3-methylglutaryl
co-enzyme A reductase activity. J. Nutr. Biochem. 1990, 1, 27–33.