Serum Hormones in Male Epileptic Patients Receiving Anticonvulsant Medication Jouko I. T. Isoj\l=a"\rvi,MD, PhD; Arto J. Pakarinen, MD, Pekka J. Ylipalosaari, MD; Vilho V. Myllyl\l=a"\,MD, PhD
\s=b\ Circulating sex and thyroid hormones, well as the pituitary function, were assessed in 63 male patients with epilepsy receiving either a single medication of as
carbamazepine, phenytoin, or valproate or a combination of carbamazepine plus phenytoin or carbamazepine plus valproate. All therapeutic regimens, including carbamazepine and/or phenytoin were associated with low levels of circulating thyroxine (T4), free thyroxine (FT4), and dehydroepiandrosterone sulfate, and with low values for the free androgen index, and phenytoin and carbamazepine plus phenytoin were associated with high serum
concentrations of
sex
hormone-bind-
ing globulin. These hormone parameters were unaffected by valproate monotherapy. It seems probable that accelerated hormone metabolism is responsible for the hormonal changes found in patients treated with carbamazepine and/or phenytoin. However, every drug regimen studied also had depressant and/or stimulatory effects on the function of the hypothalamic-pituitary axis. The diverse endocrine effects of different antiepileptic drug regimens should be considered when starting antiepileptic drug therapy. (Arch Neurol. 1990;47:670-676) t is well known that in
patients with concentrations of several hormones may be different from those of the average population of similar age.15 Moreover, epilepsy patients often suffer from symptoms epilepsy the
serum
Accepted for publication September 29, 1989. From the Departments of Neurology (Drs Isoj\l=a"\rvi,Ylipalosaari, and Myllyl\l=a"\)and Clinical Chemistry (Dr Pakarinen), University of Oulu (Finland). Reprint requests to Department of Neurology, University of Oulu, SF-90220, Oulu, Finland (Dr Isoj\l=a"\rvi).
PhD;
that may be caused by changes in the hormonal balance.6"9 However, it is not clear whether it is epilepsy per se or the antiepileptic drugs (AEDs), or both these together, that cause changes in serum hormone levels and/ or the function of the hypothalamicpituitary axis ( ).1012 Furthermore, the level at which the AEDs affect the endocrine system is not certain. It has been proposed that the liver enzyme-inducing property of most AEDs accounts for the changes in serum hormone levels by accelerating the metabolism of steroid and thyroid hormones.1*313 However, a disturbance of the central regulation of the also seems probable,1416 and the direct effects of AEDs on, eg, the thyroid gland, are not ruled out either.17 The effects of single AEDs on thy¬ roid function are well document¬ ed. 1·2·5·18·19 However, knowledge of the effects of AEDs on serum sex hormone levels is mainly based on studies in pa¬ tients with AED polypharmacy,3·4·20·21 although reports on serum sex hor¬ mone levels in epilepsy patients on AED monotherapy have also been published.2224 Knowledge about the effects of AEDs on the function of the is sparse and controver¬ sial.1416·2426 Indeed, little is known about the possible differences between various AEDs and AED combinations with regard to their effects on the en¬ docrine system as a whole. The aim of this study was to evalu¬ ate the effects of carbamazepine, phe¬ nytoin, and valproate treatment, given as monotherapy, on the endocrine function of male patients with epi¬ lepsy. Furthermore, it proved impor¬ tant to compare the effects of different drug combinations (carbamazepine *
plus phenytoin, carbamazepine plus valproate) with the effects of different monotherapies on the endocrine pa¬ rameters measured.
SUBJECTS AND METHODS
The study was carried out in the Depart¬ ments of Neurology and Clinical Chemistry,
Oulu (Finland) University Central Hospi¬ tal. The approval of the ethical committee of the medical faculty of the University of Oulu was obtained. Eighty-two male outpatients with epi¬ lepsy participated in the study after giving their informed consent. The diagnosis of epilepsy was made after careful clinical ex¬ aminations. Cerebral computed tomogra¬ phy and electroencephalography were per¬ formed on each patient. The seizure type was classified according to the Interna¬ tional League Against Epilepsy Classi¬ fication.27 The therapeutic regimen of these patients had been unchanged for at least 1 year before the study. None of the patients received any medication besides AEDs, or showed any obvious symptoms or signs of illnesses, other than epilepsy, known to af¬ fect the endocrine system. More detailed information about the pa¬ tients participating in the study is given in Table 1, Twenty patients were receiving carbamazepine monotherapy, the mean daily dose was 520 ± 101 mg, and the mean serum level of carbamazepine was 26.4 ± 5.3 /¿mol/L (in our laboratory, the therapeutic range of serum carbamazepine is 20.0 to 40.0 Mmol/L). Most of these 20 pa¬ tients had well-controlled epilepsy: only two of them experienced more than one sei¬ zure per month during the previous months. Seventeen patients received phenytoin monotherapy: the mean daily phenytoin dose was 297 ± 67 mg, and the mean serum level of phenytoin was 43.3 ± 28.0 Mmol/L (in our laboratory, the therapeutic range of serum phenytoin is 40.0 to 80.0 ^mol/L). All 17 patients who were receiving phenytoin monotherapy had well-controlled epilepsy (less than one seizure per month). Eleven
Downloaded From: http://archneur.jamanetwork.com/ by a Karolinska Institutet University Library User on 05/24/2015
Table
1.—Demographics of Study Patients Receiving Antiepileptic Drugs* No. of Patients by Mean Duration of
No. of
Patients
Mean Age, y
(Range) (21.5-50.2) 34.3 (19.0-49.8) 32.3 (23.3-44.7)
Anticonvulsive Medication
Carbamazepine Phenytoin Carbamazepine plus phenytoin 22.2 (17.3-33.4) Valproate 32.5 (18.1-48.3) Carbamazepine plus valproate primary generalized; P, partial: IP, idiopathic;
Carbamazepine 3.3 (1.0-7.5)
33.1
PG indicates
patients received a combination of carba¬ mazepine plus phenytoin: 7 patients had well-controlled epilepsy, while 4 experi¬
enced more than one seizure per month. In this group the mean carbamazepine dose was 691 ± 230 mg/d, and the mean pheny¬ toin dose was 255 ± 52 mg/d. The mean se¬ rum levels of carbamazepine and phenytoin were 20.4 ± 5.4 Mmol/L, and 36.6 ± 18.4 Mmol/L, respectively. Seven patients were receiving valproate monotherapy; the mean daily dose was 986 ± 285 mg, and the mean serum level of valproate was 0.37 ± 0.14 mmol/L (in our laboratory, the therapeutic range for serum valproate is 0.28 to 0.70 mmol/L). Only one of these seven patients experienced more than one seizure monthly, the rest having well-controlled epilepsy with a few or no seizures per year. At the time most of the male patients receiving valproate monotherapy in our
hospital
ceived
were
young.
Eight patients
re¬
combination therapy of carba¬ mazepine plus valproate; the mean daily dose was 700 ± 160 mg of carbamazepine and 1325 ± 255 mg of valproate. The mean serum levels of carbamazepine and val¬ proate were 28.5 ± 4.9 Mmol/L and 0.36 ± 0.10 mmol/L, respectively. Five of these patients had well-controlled epilepsy, while three of them experienced at least one seizure per month. A group of 19 untreated male patients with recently diagnosed epilepsy served as control subjects for the studies of pituitary responsiveness. The mean age of these pa¬ tients was 27.5 (16.0 to 49.7) years. Thirteen of them had experienced two or three pri¬ mary generalized seizures and six of them two partial seizures, with or without sec¬ ondary generalization, during the previous year. They were studied before they had re¬ ceived any anticonvulsant medication. Sev¬ enteen healthy male volunteers with a mean age of 33.4 (23.6 to 51.3) years served as control subjects for serum basal hor¬ mone and sex hormone-binding globulin (SHBG) levels. Venous blood samples were drawn at 8 am after an overnight fast and after a seizurefree period of at least 1 week. The serum samples were stored at —20° C until as¬ sayed for hormones, AEDs, and SHBG. Immediately after the blood sample for baseline values was drawn, an intravenous injection of 100 Mg of luteinizing hormonereleasing hormone (Hoechst AG, Frankfurt am Main, West Germany), and 200 Mg of thyrotropin-releasing hormone (Hoffa
Phenytoin 10.4
11.7
(2.5-21.0)
10.3
(5.0-21.0)
Medication,
11.6
PG
(2.8-25.0) (1.0-21.0)
10
8.4
(1.3-11.0) (3.5-16.0)
S, symptomatic.
mann-La Roche, Basel, Switzerland) was given to the patients in order to assess the pituitary responsiveness. A bolus of 10 mg of metoclopramide (Metopram, Leiras, Turku, Finland) was injected intravenously 60 minutes later to stimulate pituitary pro¬ lactin secretion. Blood samples for prolac¬ tin, luteinizing hormone, follicle stimulat¬ ing hormone, and thyrotropin assays were obtained 20, 60 (before metoclopramide in¬ jection), 80, and 120 minutes after the first injection. The control subjects were studied only for
Valproate
5.4
and
Type of Seizures
y
serum
basal hormone and SHBG
con¬
centrations, and the untreated patients only for pituitary responsiveness to lutein¬ izing hormone-releasing hormone, thyrotropin-releasing hormone and metaclopramide.
ASSAYS
Serum thyroxine (T4), triiodothyronine (T3), thyrotropin, testosterone, estradiol (E2), progesterone, luteinizing hormone, and follicle-stimulating hormone concen¬ trations were analyzed by radioimmunoassays using reagent kits from Farmos Diag¬ nostica (Turku and Oulunsalo, Finland). The sensitivity of the T4 assay was 5 nmol/L, the coefficient of intra-assay variation was 4.5%, and the coefficient of interassay vari¬ ation was 5.7%. The respective values were 0.1 nmol/L, 5.1 %, and 6.2% for the T3 assay; 0.5 mU/L, 3.4%, and 5.4% for the thyrotro¬ pin assay;0.26 nmol/L, 7.0%, and 12.2% for the testosterone assay; 0.03 nmol/L, 8.0%, and 9.7% for the E2 assay; 0.9 nmol/L, 7.4%, and 8.4% for the progesterone assay; 1.0 U/L, 3.2%, and 6.3% for the luteinizing hormone assay; and 0.8 U/L, 4.5%, and 7.6% for the follicle-stimulating hormone assay. The concentrations of serum free
thyroxine (FT4), free testosterone, dehydroepiandrosterone sulfate, and prolactin were measured using radioimmunoassays obtained from Diagnostic Products Corp (Los Angeles, Calif). The sensitivity of the FT4 assay was 0.13 pmol/L, the intra-assay variability was 4.4%, and the interassay variability was 5.0%. The respective values were 0.52 pmol/L, 4.4%, and 5.5% for the free testosterone assay; 0.06 mol/L, 4.5%, and 5.6% for the dehydroepiandrosterone sulfate assay; and 1.4 Mg/L, 3.2%, and 7.5% for the prolactin assay. The concentrations of serum SHBG were determined by an immunoradiometric method using reagent kits obtained from Farmos Diagnostica. The sensitivity of this assay was 0.5
nmol/L, the intra-assay variation
3.2%, and the interassay variation
was
was
5.0%. Serum carbamazepine, phenytoin, and valproate concentrations were assayed by a fluorescence polarization immunoassay system using an analyzer (TDX, Abbott Diagnostic Division, Irving, Tex). The sen¬ sitivity of the carbamazepine assay was 2.1 Mmol/L, the intra-assay variation was 1.5%, and the interassay variation was 2.5%; and, respectively, 1.3 Mmol/L, 2.0%, and 2.6% for the phenytoin assay; and 0.005 mmol/L, 2.3%, and 3.1% for the valproate assay.
The free androgen index (FAI) was cal¬ culated from the formula FAI 100X Se¬ rum Testosterone (nanomoles per liter)/ Serum SHBG (nanomoles per liter). The results were evaluated statistically by us¬ ing analysis of variance and the Bonferroni method. Linear regression analysis was used for correlations. =
RESULTS
The results are presented in Tables through 4. Serum T4 concentrations were lower in the patients treated with carbamazepine, phenytoin, carbamaz¬ epine plus phenytoin, and carbamaz¬ epine plus valproate (P < .001) than in the control subjects (Table 2). Serum 2
FT4 levels were lower in pa¬ tients treated with carbamazepine
(P26.7
± ± ±
Follicle-stimulating hormone, U/L After 20
Valproate
± ± ±
±
4.0 ± 2.5 2.6 ± 1.7
± ± ±
±
± ±
±
91.8 £
4.3 ± 2.2 3.3 ± 1.9
±
± ±
± ± ±
±
Untreated
8)_(n
=
21.7_28.5 6.0_7.0 32.7_76.3 34.74_45.5
19)
± 14.7 ± 6.9
± 46.1 ± 25.6
11.3_17.3 6.9_12.8 8.3+_10.2 6.7_7.4
± 6.6
3.7_3.3 2.3_3.2 4.2_3.2 4.5+_2.5
± 2.5
3.5_9.7 1.9_5.5
± 5.1
± 5.4 ± 3.8 ± 4.9
± 2.2
± 2.2 ± 1.8
± 2.9
4.3 ± 1.1
4.7 ± 2.8
2.8 ± 0.7
3.0 ± 2.3
SDs. untreated
untreated
patients. patients.
(%) of Patients With Serum Hormone
or Sex Hormone-Binding Globulin Concentrations Values Outside Normal Range
or
Free Androgen Index
Patient group
Carba¬
Testosterone Sex hormone-binding globulin
Dehydroepiandrosterone Luteinizing hormone Free androgen index
sulfate
Carba¬
mazepine Plus Valproate (n 8) 8 (100)
mazepine (n 20) 13 (65)
Phenytoin (n 17)
7(41.2)*
7
5(25)*
(23.5) 3(17.6)+ 12 (70.6)+
3
-
Thyroxine Free thyroxine
Carba¬
mazepine Plus Phenytoin (n 11)
7
(35) +
3(15)*
=
4
=
2
(63.6)* (27.3)*
Valproate (n 7) =
1(14.3)*
=
5
(18.2)+ (63.6)+
(62.5)*
0
3
(37.5)+
3(27.3)* 3 (27.3)*
0
1
(12.5)*
8
(35.3)* (47.1)*
6(85.7)
4
(50) "
1
(5.9)
0
o
2
(25)*
6
7
*
Below normal range. +Above normal range.
monotherapy, or as a combi¬ carbamazepine plus pheny¬ toin, or carbamazepine plus valproate, given
as
nation of
had some kind of effect on the endo¬ crine parameters that were measured. However, the effects were clearly less marked in patients receiving valproate monotherapy. The combination of car¬ bamazepine plus valproate was the therapeutic regimen with the most marked effect on serum thyroid hor¬ mone levels and the FAI. Serum T4 and FT4 levels were low in every patient group except the group receiving valproate monotherapy. Se¬ rum T3 concentrations were unaf¬ fected in each group. Moreover, serum thyrotropin levels were high only in patients receiving a combination of
carbamazepine plus valproate. The finding of low serum T4 and FT4 con¬ centrations in epilepsy patients receiv¬ ing carbamazepine, phenytoin, or a combination of these drugs is consis¬ tent with earlier reports.1·2·5·13·18·19·28"34 Previously slightly low or unchanged serum
T3 levels have been found in ep¬
ilepsy patients receiving these drugs.1·2·13·1819·2933 In agreement with most earlier reports,1·2·19·2932 serum thy¬ rotropin concentrations were not high despite low serum thyroid hormone levels. So far, reports on the effect of valproate on the thyroid hormone bal¬ ance are conflicting, and both low1·31 and unchanged25·33 serum T4 and FT4 levels have been found in patients re¬ ceiving valproate monotherapy. We
found that serum T4 and FT4 levels were lowest in the patient group re¬ ceiving both carbamazepine and val¬ proate. These patients also had higher baseline thyrotropin serum levels. As far as we know, no reports have been published on the effect of the combined use of carbamazepine plus valproate on serum thyroid hormones. In every patient group receiving AEDs, except the one receiving phe¬ nytoin monotherapy, serum thyrotro¬ pin responses to thyrotropin-releasing
hormone were similar to the responses in the untreated patient group. The normal pituitary thyrotropin re¬ sponses to thyrotropin-releasing hor¬ mone in patients receiving carba¬
mazepine
Downloaded From: http://archneur.jamanetwork.com/ by a Karolinska Institutet University Library User on 05/24/2015
or
valproate monotherapy
consistent with most previous findings,1·2·18·32 but we have also re¬ ported slightly high thyrotropin re¬ sponses to thyrotropin-releasing hor¬ mone in patients receiving carba¬ mazepine monotherapy.19 However, a low thyrotropin response to thyrotro¬ pin-releasing hormone has also been reported in six carbamazepine-treated patients with affective illness15 and, in accordance with our results, in four nonepileptic volunteers receiving phenytoin.35 No data exists on the are
effects of
a
combination of carba¬
mazepine plus phenytoin, or carba¬ mazepine plus valproate on thyro¬ tropin-releasing hormone-stimulated thyrotropin responses. It has been postulated that the se¬ rum T4 and FT4 levels are low in epi¬ leptic patients receiving carbamaz¬ epine and/or phenytoin treatment be¬
of accelerated metabolism of thyroid hormones in the liver.32·36·37 Furthermore, an increased peripheral conversion of T4 to T3 during carba¬ mazepine and/or phenytoin therapy has also been suggested2 as an expla¬ nation for the only slightly changed or unchanged serum T3 levels. The possi¬ bility has also been raised3038 of the competitive binding of carbamazepine and phenytoin on thyroxin-binding globulin. The results of the present study support the hypothesis that in¬ creased metabolism is the main reason for decreased T4 and FT4 serum levels because carbamazepine and phenytoin are well known inducers of the microsomal enzyme system of the liver cause
metabolizing thyroid hormones, while valproate does not seem to have a sim¬ ilar enzyme-inducing effect.39 How¬ ever, the
serum
T4 and FT4 levels were
lowest in
patients receiving both car¬ bamazepine and valproate. Both car¬ bamazepine and valproate are highly bound to serum proteins.40·41 It seems possible that valproate could displace T4 from its binding sites on plasma proteins, and thereby lead to a larger amount of T4 liable to increased me¬ tabolism caused by the liver enzyme-
inducing properties of carbamazepine. Basal serum thyrotropin was not in¬ creased in patients who were receiving carbamazepine, phenytoin, or a com¬ bination of these two drugs, despite
the low serum levels of T4 and FT4. It is possible that the unchanged serum T3 levels are sufficient to inhibit in¬ creased secretion of thyrotropin. How¬ ever, an inhibitory effect of AEDs on the hypothalamic and/or anterior pi¬ tuitary level also seems probable. Be¬ cause the thyrotropin-releasing hor¬ mone-stimulated thyrotropin re¬ sponses were unaltered by these drugs
(except the response after 20 minutes in the phenytoin monotherapy group), a hypothalamic interference would seem more likely. The high levels of serum basal thyrotropin in patients receiving a combination therapy of carbamazepine plus valproate may be explained by the -aminobutyric acid ergic properties of valproate,42·43 since it is known that 7-aminobutyric acid is inhibitory of somatostatin release, and sumatostatin inhibits thyrotropin secretion.44 Epileptic seizures can manifest in association with thyrotoxicosis,45 and T4 may induce status epilepticus.46 It seems possible, therefore, that the lowering effect of AEDs on serum T4 and FT4 levels might contribute to the anticonvulsive properties of these
drugs.
Patients who are receiving pheny¬ toin monotherapy, or who were receiv¬ ing a combination therapy with car¬ bamazepine plus phenytoin had high levels of serum testosterone and SHBG, but patients who were receiv¬ ing other therapeutic regimens did not. Serum free thyroxine levels were unchanged in every patient group, while FAI values were low in all pa¬ tient groups, except the one group that was
receiving valproate monotherapy.
The lowest FAI values were found in patients receiving a combination ther¬ apy of carbamazepine plus valproate. Serum testosterone levels have been found to be mostly normal in male pa¬ tients with epilepsy who were receiv¬ ing AED polytherapy,4·20·22·24 but ele¬ vated serum concentrations of tes¬ tosterone have also been found.3 A few reports exist on normal serum testos¬ terone levels during carbamazepine and valproate monotherapy,13·23·24 and one report on a slightly increased se¬ rum testosterone level during pheny¬ toin monotherapy.13 Furthermore, high serum levels of SHBG have been found in male epilepsy patients receiv¬
ing phenytoin, carbamazepine, or AED polypharmacy.3·20·2224 In this study, the highest serum SHBG concentrations were found associated with phenytoin monotherapy. It seems possible that the high serum testosterone concen¬ trations in patients receiving pheny¬ toin medication are explained by these high SHBG levels. It appears reason¬ able to assume that the rise of serum SHBG concentrations is caused by in¬ creased synthesis due to liver enzyme induction by AEDs, as suggested before.23·47 Previously low serum free testoster¬ one levels have been reported in male patients receiving a combination of various AEDs.3·20 Recently, an un-
changed serum free testosterone level and decreased FAI were reported in male epileptic patients receiving car¬ bamazepine monotherapy.2348 Calcu¬
lated free testosterone fraction has also been found decreased in male pa¬ tients on receiving carbamazepine monotherapy, while phenytoin or val¬ proate did not have a similar effect.24 No previous data exists on serum free testosterone levels or FAI values in male epileptic patients receiving a combination of carbamazepine plus phenytoin, or carbamazepine plus val¬ proate. It is not yet clear whether the low FAI values have clinical signifi¬ cance in male patients receiving AEDs. The discrepancy between low FAI values and unchanged serum free testosterone concentrations may be explained by displacement of tes¬ tosterone from serum albumin by AEDs, as suggested before.23·48 Tes¬ tosterone binds with high affinity to SHBG, and loosely to albumin.49 In the present study patients receiv¬ ing phenytoin and valproate mono¬ therapy had slightly higher levels of serum E2 than the control subjects, but patients receiving other therapeu¬ tic regimens did not. The significance of this finding is not clear. However, serum progesterone levels were unaf¬ fected in every patient group. Serum dehydroepiandrosterone sul¬ fate levels were low in patients receiv¬
ing carbamazepine or phenytoin monotherapy, or receiving a combina¬ tion of carbamazepine plus phenytoin, or carbamazepine plus valproate. The levels were lowest in patients receiving phenytoin monotherapy, and un¬ changed in patients receiving val¬ proate monotherapy. The low serum levels of dehydroepiandrosterone sul¬ fate in male epileptic patients on re¬ ceiving carbamazepine, phenytoin, or both of these drugs confirm earlier findings.23·24·47·48·50 Previous reports on serum dehydroepiandrosterone sul¬ fate concentrations in patients receiv¬ ing a combination of carbamazepine plus valproate are lacking. Since, in the present study, all therapeutic reg¬ imens, including liver enzyme-induc¬ ing drugs, carbamazepine and/or phenytoin, were associated with low serum levels of dehydroepiandroster¬ one sulfate, while serum dehydroepi¬
androsterone sulfate concentrations unchanged in patients receiving valproate monotherapy, it seems likely that increased metabolism of dehy¬ droepiandrosterone sulfate is the main reason for the decreased serum dehy¬ droepiandrosterone sulfate levels that were found. Serum basal prolactin concentrawere
Downloaded From: http://archneur.jamanetwork.com/ by a Karolinska Institutet University Library User on 05/24/2015
were low in patients receiving phenytoin monotherapy, but other therapeutic regimens were associated with unchanged serum basal prolactin concentrations. However, the prolac¬ tin responses to thyrotropin-releasing hormone were slightly higher in pa¬ tients receiving carbamazepine, phe¬ nytoin, or the combination of these two drugs, but prolactin responses to me¬ toclopramide were higher in patients receiving phenytoin, a combination of carbamazepine plus phenytoin, or car¬ bamazepine plus valproate. Slightly higher or unchanged prolactin re¬ sponses to thyrotropin-releasing hor¬ mone and metaclopramide have been reported in patients receiving carba¬ mazepine, phenytoin, valproate, or AED polypharmacy.14'23-24'26 The reports on serum basal prolactin levels in male epileptic patients treated with car¬ bamazepine, phenytoin, or valproate are conflicting, but the levels have mostly been normal.14·16'23·24'26 All therapeutic regimens except car¬ bamazepine monotherapy were asso¬
with valproate in female patients with epilepsy, it being assumed that this was caused by interference of val¬ proate at the level of the .52 It is known that epileptic seizures can in¬ crease pituitary secretion of luteiniz¬ ing hormones53 and that a long history of seizures of temporal lobe origin may be associated with changes in the func¬ tion of .1011 Therefore, it may be assumed that the high serum levels of luteinizing hormone reported earlier in patients receiving AED polytherapy could be explained by effects of epi¬ lepsy per se, because it seems that the patients studied in many instances were institutionalized and suffered from severe epilepsy with frequently occurring seizures for many years. The patients in the present study had
epine, phenytoin, or valproate mono¬ therapy reported so far have been mostly normal,16·23·24·48 though a high basal luteinizing hormone serum level
slightly elevated, a hypothalamic rather than pituitary interference by these drugs seems likely. It has also been suggested that 7-aminobutyric acid-containing neurons may be in¬ volved in the regulation of gonadotro¬ pa secretion.55 This could explain the mechanism through which valproate has effects on serum luteinizing hor¬ mone concentrations. The regulation of gonadotropin secretion is complex,
tions
ciated with low basal serum luteiniz¬ ing hormone levels. Previously high serum luteinizing hormone levels have been reported in association with AED polytherapy.3·4·24 The luteinizing hor¬ mone serum levels in male patients with epilepsy receiving carbamaz¬
has also been found in association with carbamazepine24·26 and phenytoin26 monotherapy. Moreover, a low basal luteinizing hormone level has been ob¬ served in male patients receiving phé¬ nobarbital monotherapy,51 and amenorrhea following initiation of therapy
mostly well-controlled epilepsy. It has been previously suggested that the high serum luteinizing hor¬ mone levels in male epileptic patients are caused by a feedback mechanism due to low serum free testosterone levels.54 However, the results of the present study indicating low basal se¬ rum luteinizing hormone levels in most patient groups suggest a direct . effect of AEDs on the function of Because luteinizing hormone re¬ sponses to luteinizing hormonereleasing hormone are unaffected or
and the mechanisms through which other AEDs, eg, phenytoin, have ef¬ fects on serum luteinizing hormone levels are not clear. Despite the low serum luteinizing hormone levels in most patient groups, the basal and stimulated serum folli¬ cle-stimulating hormone were un¬ changed in most of the patient groups receiving AEDs. The results of the present study in¬ dicate that the serum hormone con¬ centrations and pituitary responses to various stimuli in male epileptic pa¬ tients receiving AEDs are different from those of control subjects. The therapeutic regimen with least effects on the hormone balance was valproate monotherapy, while combined therapy of carbamazepine and valproate had the most marked effect on both the thyroid hormone balance and the lev¬ els of FAI. It remains to be established how se¬ vere the clinical consequences of the hormonal changes found in this study really are, with regard to, eg, impo¬ tence, and a possible latent hypothyrosis in these patients. Nevertheless, these results are indicatory of marked differences between various AED reg¬ imens with regard to their effects on the hormone balance. This should be borne in mind when an AED regimen is considered for a patient with re¬
cently diagnosed epilepsy.
This study was supported with grants from the Finnish Neurology Foundation, Helsinki, Fin¬ land, and from the Research and Science Founda¬ tion of Farmos, Turku, Finland. The authors thank the personnel of the Depart¬ ment of Neurology and the Laboratory of Oulu (Finland) University Central Hospital for assis¬ tance in collecting samples and in laboratory work.
References
1. Fichsel H, Kn\l=o"\pfleG. Effects of anticonvuldrugs on thyroid hormones in epileptic children. Epilepsia. 1978;19:323-336. 2. Liewendahl K, Majuri H, Helenius T. Thyroid function tests in patients on long-term treatment with various anticonvulsant drugs. Clin Endocrinol. 1978;8:185-191. 3. Toone BK, Edeh J, Fenwick P, et al. Hormonal and behavioral changes in male epileptics. In: Porter RJ, Ward AA Jr, Mattson RH, Dam M, eds. Advances in Epileptology: XVth Epilepsy International Symposium. New York, NY: Raven sant
Press;
1984:283-289. E, Subramanian MG, Gilroy J. Investigation of sex hormones in male epileptic pa4. Rodin
Epilepsia. 1984;25:690-694. Haidukewych D, Rodin EA. Chronic antiepileptic drug therapy: classification by medication regimen and incidence of decreases in serum thyroxine and free thyroxine index. Ther Drug Monit.
tients. 5.
1987;9:392-398.
6. Aanderud S, Strandjord RE. Hypothyroidism induced by anti-epileptic therapy. Acta Neurol Scand. 1980;61:330-332.
7. Fenwick PBC, Toone BK, Wheeler MJ, Nanjee MN, Grant R, Brown D. Sexual behaviour in a centre for epilepsy. Acta Neurol Scand.
1985;71:428-435. 8. Mattson RH, Cramer JA, Collins JF,
et al.
Comparison of carbamazepine, phenobarbital, phenytoin, and primidone in partial and secondarily generalized tonic-clonic seizures. N Engl J
Med. 1985;313:145-151. 9. Webber MP, Hauser WA, Ottman R, Annegers JF. Fertility in persons with epilepsy: 1935-1974. Epilepsia. 1986;27:746-752. 10. Herzog AG, Seibel MM, Schomer DL, Vaitukaitis JL, Geschwind N. Reproductive disorders in women with partial seizures of temporal lobe origin. Arch Neurol. 1986;43:341-346. 11. Herzog AG, Seibel MM, Schomer DL, Vaitukaitis JL, Geschwind N. Reproductive endocrine disorders in men with partial seizures of temporal lobe origin. Arch Neurol. 1986;43:347-350. 12. Bilo L, Meo R, Nappi C, et al. Reproductive endocrine disorders in women with primary generalized epilepsy. Epilepsia. 1988;29:612-619. 13. L\l=u"\hdorfK, Christiansen P, Hansen JM,
Lund M. The influence of phenytoin and carbamazepine on endocrine function: preliminary results. In: Penry JK, ed. Epilepsy. The Eighth International Symposium. New York, NY: Raven Press; 1977:209-213. 14. Franceschi M, Perego L, Cavagnini F, et al. Effects of long-term antiepileptic therapy on the hypothalamic-pituitary axis in man. Epilepsia.
1984;25:46-52. 15. Joffe RT, Post RM, Ballenger JC, Rebar R,
Gold PW. Neuroendocrine effects of carbamazepine in patients with affective illness. Epi-
lepsia. 1986;27:156-160. 16. Isoj\l=a"\rviJIT, Myllyl\l=a"\VV, Pakarinen AJ. Effects of carbamezepine on pituitary responsiveness to luteinizing hormone-releasing hormone, thyrotropin-releasing hormone, and metoclopramide in epileptic patients. Epilepsia. 1989;30:50\x=req-\ 56. 17. Villa
SM, Alexander NM. Carbamazepine (Tegretol) inhibits in vivo iodide uptake and hormone synthesis in rat thyroid glands. Endocr Res. 1987;13:385-397. 18. M\l=o/\lholmHansen J, Skovsted L, Birk Lau-
Downloaded From: http://archneur.jamanetwork.com/ by a Karolinska Institutet University Library User on 05/24/2015
ridsen U, Kirkegaard C, Siersbaek-Nielsen K. The effect of diphenylhydantoin on thyroid function. J Clin Endocrinol Metab. 1974;39:785-789. 19. Isoj\l=a"\rviJIT, Pakarinen AJ, Myllyl\l=a"\VV: Thyroid function in epileptic patients treated with carbamazepine. Arch Neurol. 1989;46:1175\x=req-\ 1178. 20. Dana-Haeri J, Oxley J, Richens A. Reduction of free testosterone by antiepileptic drugs. Br Med J. 1982;284:85-86. 21. Rodin E, Subramanian MG, Schmaltz S, Gilroy J. Testosterone levels in adult male epileptic patients. Neurology. 1987;37:706-708. 22. Dana-Haeri J, Richens A. Effect of antiepileptic drugs on the hypothalamic-pituitary axis in epileptic patients. In: Dam M, Gram L, Penry JK, eds. Advances in Epileptology: The XIIth Epilepsy International Symposium. New York, NY: Raven Press; 1981:469-475. 23. Isoj\l=a"\rviJIT, Pakarinen AJ, Myllyl\l=a"\VV. Effects of carbamazepine therapy on serum sex hormone levels in male patients with epilepsy.
Epilepsia. 1988;29:781-786. 24. Macphee GJA, Larkin JG, Butler E, Beastall GH, Brodie MJ. Circulating hormones and pituitary responsiveness in young epileptic men receiving long-term antiepileptic medication. Epilepsia. 1988;29:468-475. 25. London DR, Loizou LA, Butt WR, Rovei V, Bianchetti G, Morselli PL. The effect of anti-convulsant drugs (AED) on hormonal responses in normal volunteers. In: Johannesen SI, Morselli PL, Pippenger CE, Richens A, Schmidt D, Meinardi H, eds. Antiepileptic Therapy: Advances in Drug Monitoring. New York, NY: Raven Press;
1980:405-411. 26. Dana-Haeri J, Oxley J, Richens A. Pituitary responsiveness to gonadotrophin-releasing and thyrotrophin-releasing hormones in epileptic patients receiving carbamazepine or phenytoin. Clin. Endocrinol. 1984;20:163-168. 27. Dreifuss FE, Bancaud J, Henriksen O, Rubio-Donnadieu F, Penry JK, Seino M. Proposal for revised clinical and electroencephalographic classification of epileptic seizures. Epilepsia. 1981; 22:489-501. 28. Chin W, Schussler GC. Decreased serum free thyroxine concentration in patients treated with diphenylhydantoin. J Clin Endocrinol Metab. 1968;28:181-186. 29. Yeo PPB, Bates D, Howe JD, et al. Anticonvulsants and thyroid function. Br Med J. 1978; 1:1581-1583. 30. Rootwelt K, Ganes T, Johannesen SI. Effect of carbamazepine, phenytoin and phenobarbital on serum levels of thyroid hormones and thyro-
tropin in humans. Scand Clin Lab Invest. 1978; 38:731-736. 31. Bentsen KD, Gram L, Veje A. Serum thyroid hormones and blood folic acid during monotherapy with carbamazepine or valproate. Acta Neurol Scand. 1983;67:235-241. 32. Connell JMC, Rapeport WG, Gordon S, Brodie MJ. Changes in circulating thyroid hormones during short-term hepatic enzyme induction with carbamazepine. Eur J Clin Pharmacol.
1984;26:453-456.
33. Tanaka K, Kodama S, Yokoyama S, et al. Thyroid function in children with long-term anticonvulsant treatment. Pediatr Neurosci. 1987;
13:90-94. 34. Liewendahl K, Majuri H. Thyroxine, triiodothyronine, and thyrotropin in serum during long-term diphenylhydantoin therapy. Scand J Clin Lab Invest. 1976;36:141-144. 35. Surks MI, Ordene KW, Mann DN, KumaraSiri MH. Diphenylhydantoin inhibits the thyrotropin response to thyrotropin-releasing hormone in man and rat. J Clin Endocrinol Metab. 1983; 56:940-945. 36. Larsen PR, Atkinson AJ Jr, Wellman HN, Goldsmith RE. The effect of diphenylhydantoin on thyroxine metabolism in man. J Clin Invest.
1970;49:1266-1279. 37. Mendoza DM, Flock EV, Owen CA Jr, Paris
J. Effect of 5,5\m='\-diphenylhydantoin on the metabolism of L-thyroxine-131I in the rat. Endocrinology.
1966;79:106-118.
38. Oppenheimer JH, Tavernetti RR: Studies the thyroxine-diphenylhydantoin interaction: effect of 5,5\m='\-diphenylhydantoinon the displacement of L-thyroxine from thyroxine-binding globulin (TBG). Endocrinology. 1962;71:496-504. 39. Perucca E, Hedges A, Makki KA, Ruprah M, Wilson JF, Richens A. A comparative study of the relative enzyme-inducing properties of anticonvulsant drugs in epileptic patients. Br J Clin Pharm. 1984;18:401-410. 40. Mattson RH, Cramer JA. Valproic acid (Depakane, Valontin). In: Browne TR, Feldman RG, eds. Epilepsy. Diagnosis and Management. Boston, Mass: Little Brown & Co; 1983:225-233. 41. Morselli PL, Gerna M, de Maio D, Zanda G, Viani F, Garattini S. Pharmacokinetic studies on carbamazepine in volunteers and in epileptic patients. In: Schneider H, Janz D, Gardner-Thorpe C, Meinardi H, Sherwin AL, eds. Clinical Pharmacology ofAntiepileptic Drugs. New York, NY: Springer-Verlag NY Inc; 1975:166-180. 42. L\l=o"\scherW, Schmidt D. Increase of human plasma GABA by sodium valproate. Epilepsia. on
1980;21:611-615.
43. L\l=o"\scherW. Effect of inhibitors of GABA\x=req-\ transaminase on the synthesis, binding, uptake and metabolism of GABA. J Neurochem. 1980; 34:1603-1608. 44. Foord SM, Peters JR, Dieguez C, et al. Thyroid stimulating hormone. In: Lightman SL, Everitt BJ, eds. Neuroendocrinology. Boston, Mass: Blackwell Scientific Publications; 1986:450-471. 45. Jabbari B, Huott AD. Seizures in thyrotoxicosis. Epilepsia. 1980;21:91-96. 46. Sundaram MBM, Hill A, Lowry N. Thyroxine-induced petit mal status epilepticus. Neurology.
1985;35:1792-1793.
47. Connell JMC, Rapeport WG, Beastall LGH, Brodie MJ. Changes in circulating androgens during short-term carbamazepine therapy. Br J Clin Pharm. 1984;17:347-351. 48. Isoj\l=a"\rviJIT, Pakarinen AJ, Myllyl\l=a"\VV. Effects of carbamazepine medication on the hypothalamic-pituitary-gonadal axis in male patients with epilepsy: a prospective study. Epilepsia. 1989;30:446-452. 49. S\l=o"\derg\l=a%o\rdR, B\l=a"\ckstr\l=o"\mT, Shanbhag V, Carstensen H. Calculation of free and bound fractions of testosterone and estradiol-17 to human plasma proteins at body temperature. J Steroid Biochem. 1982;16:801-810. 50. Levesque LA, Herzog Ag, Seibel MM. The effect of phenytoin and carbamazepine on serum dehydroepiandrosterone sulfate in men and women who have partial seizures with temporal lobe involvement. J Clin Endocrinol Metab. 1986; 63:243-245. 51. Murialdo G, Manni R, De Maria A, Bonura ML, Pollen A, Tartara A. Luteinizing hormone pulsatile secretion and pituitary response to gonadotropin releasing hormone and to thyrotropin releasing hormone in male epileptic subjects on chronic phenobarbital treatment. J Endocrinol Invest. 1987;10:27-31. 52. Margraf JW, Dreifuss FE. Amenorrhea following initiation of therapy with valproic acid.
Neurology. 1981;31:159.
53. Trimble MR, Dana-Haeri J, Oxley J, Baylis PH. Some neuroendocrine consequences of seizures. In: Porter RJ, Ward AA Jr, Mattson RH, Dam M, eds. Advances in Epileptology: XVth Epilepsy International Symposium. New York, NY: Raven Press; 1984:201-208. 54. Mattson RH, Cramer JA. Epilepsy, sex hormones, and antiepileptic drugs. Epilepsia.
1985;26:S40-S51. 55. Everitt BJ, Keverne EB. Reproduction. In: Lightman SL, Everitt BJ, eds. Neuroendocrinology. Boston, Mass: Blackwell Scientific Publications; 1986:472-537.
Downloaded From: http://archneur.jamanetwork.com/ by a Karolinska Institutet University Library User on 05/24/2015
\s=b\ Circulating sex and thyroid hormones, well as the pituitary function, were assessed in 63 male patients with epilepsy receiving either a single medication of as
carbamazepine, phenytoin, or valproate or a combination of carbamazepine plus phenytoin or carbamazepine plus valproate. All therapeutic regimens, including carbamazepine and/or phenytoin were associated with low levels of circulating thyroxine (T4), free thyroxine (FT4), and dehydroepiandrosterone sulfate, and with low values for the free androgen index, and phenytoin and carbamazepine plus phenytoin were associated with high serum
concentrations of
sex
hormone-bind-
ing globulin. These hormone parameters were unaffected by valproate monotherapy. It seems probable that accelerated hormone metabolism is responsible for the hormonal changes found in patients treated with carbamazepine and/or phenytoin. However, every drug regimen studied also had depressant and/or stimulatory effects on the function of the hypothalamic-pituitary axis. The diverse endocrine effects of different antiepileptic drug regimens should be considered when starting antiepileptic drug therapy. (Arch Neurol. 1990;47:670-676) t is well known that in
patients with concentrations of several hormones may be different from those of the average population of similar age.15 Moreover, epilepsy patients often suffer from symptoms epilepsy the
serum
Accepted for publication September 29, 1989. From the Departments of Neurology (Drs Isoj\l=a"\rvi,Ylipalosaari, and Myllyl\l=a"\)and Clinical Chemistry (Dr Pakarinen), University of Oulu (Finland). Reprint requests to Department of Neurology, University of Oulu, SF-90220, Oulu, Finland (Dr Isoj\l=a"\rvi).
PhD;
that may be caused by changes in the hormonal balance.6"9 However, it is not clear whether it is epilepsy per se or the antiepileptic drugs (AEDs), or both these together, that cause changes in serum hormone levels and/ or the function of the hypothalamicpituitary axis ( ).1012 Furthermore, the level at which the AEDs affect the endocrine system is not certain. It has been proposed that the liver enzyme-inducing property of most AEDs accounts for the changes in serum hormone levels by accelerating the metabolism of steroid and thyroid hormones.1*313 However, a disturbance of the central regulation of the also seems probable,1416 and the direct effects of AEDs on, eg, the thyroid gland, are not ruled out either.17 The effects of single AEDs on thy¬ roid function are well document¬ ed. 1·2·5·18·19 However, knowledge of the effects of AEDs on serum sex hormone levels is mainly based on studies in pa¬ tients with AED polypharmacy,3·4·20·21 although reports on serum sex hor¬ mone levels in epilepsy patients on AED monotherapy have also been published.2224 Knowledge about the effects of AEDs on the function of the is sparse and controver¬ sial.1416·2426 Indeed, little is known about the possible differences between various AEDs and AED combinations with regard to their effects on the en¬ docrine system as a whole. The aim of this study was to evalu¬ ate the effects of carbamazepine, phe¬ nytoin, and valproate treatment, given as monotherapy, on the endocrine function of male patients with epi¬ lepsy. Furthermore, it proved impor¬ tant to compare the effects of different drug combinations (carbamazepine *
plus phenytoin, carbamazepine plus valproate) with the effects of different monotherapies on the endocrine pa¬ rameters measured.
SUBJECTS AND METHODS
The study was carried out in the Depart¬ ments of Neurology and Clinical Chemistry,
Oulu (Finland) University Central Hospi¬ tal. The approval of the ethical committee of the medical faculty of the University of Oulu was obtained. Eighty-two male outpatients with epi¬ lepsy participated in the study after giving their informed consent. The diagnosis of epilepsy was made after careful clinical ex¬ aminations. Cerebral computed tomogra¬ phy and electroencephalography were per¬ formed on each patient. The seizure type was classified according to the Interna¬ tional League Against Epilepsy Classi¬ fication.27 The therapeutic regimen of these patients had been unchanged for at least 1 year before the study. None of the patients received any medication besides AEDs, or showed any obvious symptoms or signs of illnesses, other than epilepsy, known to af¬ fect the endocrine system. More detailed information about the pa¬ tients participating in the study is given in Table 1, Twenty patients were receiving carbamazepine monotherapy, the mean daily dose was 520 ± 101 mg, and the mean serum level of carbamazepine was 26.4 ± 5.3 /¿mol/L (in our laboratory, the therapeutic range of serum carbamazepine is 20.0 to 40.0 Mmol/L). Most of these 20 pa¬ tients had well-controlled epilepsy: only two of them experienced more than one sei¬ zure per month during the previous months. Seventeen patients received phenytoin monotherapy: the mean daily phenytoin dose was 297 ± 67 mg, and the mean serum level of phenytoin was 43.3 ± 28.0 Mmol/L (in our laboratory, the therapeutic range of serum phenytoin is 40.0 to 80.0 ^mol/L). All 17 patients who were receiving phenytoin monotherapy had well-controlled epilepsy (less than one seizure per month). Eleven
Downloaded From: http://archneur.jamanetwork.com/ by a Karolinska Institutet University Library User on 05/24/2015
Table
1.—Demographics of Study Patients Receiving Antiepileptic Drugs* No. of Patients by Mean Duration of
No. of
Patients
Mean Age, y
(Range) (21.5-50.2) 34.3 (19.0-49.8) 32.3 (23.3-44.7)
Anticonvulsive Medication
Carbamazepine Phenytoin Carbamazepine plus phenytoin 22.2 (17.3-33.4) Valproate 32.5 (18.1-48.3) Carbamazepine plus valproate primary generalized; P, partial: IP, idiopathic;
Carbamazepine 3.3 (1.0-7.5)
33.1
PG indicates
patients received a combination of carba¬ mazepine plus phenytoin: 7 patients had well-controlled epilepsy, while 4 experi¬
enced more than one seizure per month. In this group the mean carbamazepine dose was 691 ± 230 mg/d, and the mean pheny¬ toin dose was 255 ± 52 mg/d. The mean se¬ rum levels of carbamazepine and phenytoin were 20.4 ± 5.4 Mmol/L, and 36.6 ± 18.4 Mmol/L, respectively. Seven patients were receiving valproate monotherapy; the mean daily dose was 986 ± 285 mg, and the mean serum level of valproate was 0.37 ± 0.14 mmol/L (in our laboratory, the therapeutic range for serum valproate is 0.28 to 0.70 mmol/L). Only one of these seven patients experienced more than one seizure monthly, the rest having well-controlled epilepsy with a few or no seizures per year. At the time most of the male patients receiving valproate monotherapy in our
hospital
ceived
were
young.
Eight patients
re¬
combination therapy of carba¬ mazepine plus valproate; the mean daily dose was 700 ± 160 mg of carbamazepine and 1325 ± 255 mg of valproate. The mean serum levels of carbamazepine and val¬ proate were 28.5 ± 4.9 Mmol/L and 0.36 ± 0.10 mmol/L, respectively. Five of these patients had well-controlled epilepsy, while three of them experienced at least one seizure per month. A group of 19 untreated male patients with recently diagnosed epilepsy served as control subjects for the studies of pituitary responsiveness. The mean age of these pa¬ tients was 27.5 (16.0 to 49.7) years. Thirteen of them had experienced two or three pri¬ mary generalized seizures and six of them two partial seizures, with or without sec¬ ondary generalization, during the previous year. They were studied before they had re¬ ceived any anticonvulsant medication. Sev¬ enteen healthy male volunteers with a mean age of 33.4 (23.6 to 51.3) years served as control subjects for serum basal hor¬ mone and sex hormone-binding globulin (SHBG) levels. Venous blood samples were drawn at 8 am after an overnight fast and after a seizurefree period of at least 1 week. The serum samples were stored at —20° C until as¬ sayed for hormones, AEDs, and SHBG. Immediately after the blood sample for baseline values was drawn, an intravenous injection of 100 Mg of luteinizing hormonereleasing hormone (Hoechst AG, Frankfurt am Main, West Germany), and 200 Mg of thyrotropin-releasing hormone (Hoffa
Phenytoin 10.4
11.7
(2.5-21.0)
10.3
(5.0-21.0)
Medication,
11.6
PG
(2.8-25.0) (1.0-21.0)
10
8.4
(1.3-11.0) (3.5-16.0)
S, symptomatic.
mann-La Roche, Basel, Switzerland) was given to the patients in order to assess the pituitary responsiveness. A bolus of 10 mg of metoclopramide (Metopram, Leiras, Turku, Finland) was injected intravenously 60 minutes later to stimulate pituitary pro¬ lactin secretion. Blood samples for prolac¬ tin, luteinizing hormone, follicle stimulat¬ ing hormone, and thyrotropin assays were obtained 20, 60 (before metoclopramide in¬ jection), 80, and 120 minutes after the first injection. The control subjects were studied only for
Valproate
5.4
and
Type of Seizures
y
serum
basal hormone and SHBG
con¬
centrations, and the untreated patients only for pituitary responsiveness to lutein¬ izing hormone-releasing hormone, thyrotropin-releasing hormone and metaclopramide.
ASSAYS
Serum thyroxine (T4), triiodothyronine (T3), thyrotropin, testosterone, estradiol (E2), progesterone, luteinizing hormone, and follicle-stimulating hormone concen¬ trations were analyzed by radioimmunoassays using reagent kits from Farmos Diag¬ nostica (Turku and Oulunsalo, Finland). The sensitivity of the T4 assay was 5 nmol/L, the coefficient of intra-assay variation was 4.5%, and the coefficient of interassay vari¬ ation was 5.7%. The respective values were 0.1 nmol/L, 5.1 %, and 6.2% for the T3 assay; 0.5 mU/L, 3.4%, and 5.4% for the thyrotro¬ pin assay;0.26 nmol/L, 7.0%, and 12.2% for the testosterone assay; 0.03 nmol/L, 8.0%, and 9.7% for the E2 assay; 0.9 nmol/L, 7.4%, and 8.4% for the progesterone assay; 1.0 U/L, 3.2%, and 6.3% for the luteinizing hormone assay; and 0.8 U/L, 4.5%, and 7.6% for the follicle-stimulating hormone assay. The concentrations of serum free
thyroxine (FT4), free testosterone, dehydroepiandrosterone sulfate, and prolactin were measured using radioimmunoassays obtained from Diagnostic Products Corp (Los Angeles, Calif). The sensitivity of the FT4 assay was 0.13 pmol/L, the intra-assay variability was 4.4%, and the interassay variability was 5.0%. The respective values were 0.52 pmol/L, 4.4%, and 5.5% for the free testosterone assay; 0.06 mol/L, 4.5%, and 5.6% for the dehydroepiandrosterone sulfate assay; and 1.4 Mg/L, 3.2%, and 7.5% for the prolactin assay. The concentrations of serum SHBG were determined by an immunoradiometric method using reagent kits obtained from Farmos Diagnostica. The sensitivity of this assay was 0.5
nmol/L, the intra-assay variation
3.2%, and the interassay variation
was
was
5.0%. Serum carbamazepine, phenytoin, and valproate concentrations were assayed by a fluorescence polarization immunoassay system using an analyzer (TDX, Abbott Diagnostic Division, Irving, Tex). The sen¬ sitivity of the carbamazepine assay was 2.1 Mmol/L, the intra-assay variation was 1.5%, and the interassay variation was 2.5%; and, respectively, 1.3 Mmol/L, 2.0%, and 2.6% for the phenytoin assay; and 0.005 mmol/L, 2.3%, and 3.1% for the valproate assay.
The free androgen index (FAI) was cal¬ culated from the formula FAI 100X Se¬ rum Testosterone (nanomoles per liter)/ Serum SHBG (nanomoles per liter). The results were evaluated statistically by us¬ ing analysis of variance and the Bonferroni method. Linear regression analysis was used for correlations. =
RESULTS
The results are presented in Tables through 4. Serum T4 concentrations were lower in the patients treated with carbamazepine, phenytoin, carbamaz¬ epine plus phenytoin, and carbamaz¬ epine plus valproate (P < .001) than in the control subjects (Table 2). Serum 2
FT4 levels were lower in pa¬ tients treated with carbamazepine
(P26.7
± ± ±
Follicle-stimulating hormone, U/L After 20
Valproate
± ± ±
±
4.0 ± 2.5 2.6 ± 1.7
± ± ±
±
± ±
±
91.8 £
4.3 ± 2.2 3.3 ± 1.9
±
± ±
± ± ±
±
Untreated
8)_(n
=
21.7_28.5 6.0_7.0 32.7_76.3 34.74_45.5
19)
± 14.7 ± 6.9
± 46.1 ± 25.6
11.3_17.3 6.9_12.8 8.3+_10.2 6.7_7.4
± 6.6
3.7_3.3 2.3_3.2 4.2_3.2 4.5+_2.5
± 2.5
3.5_9.7 1.9_5.5
± 5.1
± 5.4 ± 3.8 ± 4.9
± 2.2
± 2.2 ± 1.8
± 2.9
4.3 ± 1.1
4.7 ± 2.8
2.8 ± 0.7
3.0 ± 2.3
SDs. untreated
untreated
patients. patients.
(%) of Patients With Serum Hormone
or Sex Hormone-Binding Globulin Concentrations Values Outside Normal Range
or
Free Androgen Index
Patient group
Carba¬
Testosterone Sex hormone-binding globulin
Dehydroepiandrosterone Luteinizing hormone Free androgen index
sulfate
Carba¬
mazepine Plus Valproate (n 8) 8 (100)
mazepine (n 20) 13 (65)
Phenytoin (n 17)
7(41.2)*
7
5(25)*
(23.5) 3(17.6)+ 12 (70.6)+
3
-
Thyroxine Free thyroxine
Carba¬
mazepine Plus Phenytoin (n 11)
7
(35) +
3(15)*
=
4
=
2
(63.6)* (27.3)*
Valproate (n 7) =
1(14.3)*
=
5
(18.2)+ (63.6)+
(62.5)*
0
3
(37.5)+
3(27.3)* 3 (27.3)*
0
1
(12.5)*
8
(35.3)* (47.1)*
6(85.7)
4
(50) "
1
(5.9)
0
o
2
(25)*
6
7
*
Below normal range. +Above normal range.
monotherapy, or as a combi¬ carbamazepine plus pheny¬ toin, or carbamazepine plus valproate, given
as
nation of
had some kind of effect on the endo¬ crine parameters that were measured. However, the effects were clearly less marked in patients receiving valproate monotherapy. The combination of car¬ bamazepine plus valproate was the therapeutic regimen with the most marked effect on serum thyroid hor¬ mone levels and the FAI. Serum T4 and FT4 levels were low in every patient group except the group receiving valproate monotherapy. Se¬ rum T3 concentrations were unaf¬ fected in each group. Moreover, serum thyrotropin levels were high only in patients receiving a combination of
carbamazepine plus valproate. The finding of low serum T4 and FT4 con¬ centrations in epilepsy patients receiv¬ ing carbamazepine, phenytoin, or a combination of these drugs is consis¬ tent with earlier reports.1·2·5·13·18·19·28"34 Previously slightly low or unchanged serum
T3 levels have been found in ep¬
ilepsy patients receiving these drugs.1·2·13·1819·2933 In agreement with most earlier reports,1·2·19·2932 serum thy¬ rotropin concentrations were not high despite low serum thyroid hormone levels. So far, reports on the effect of valproate on the thyroid hormone bal¬ ance are conflicting, and both low1·31 and unchanged25·33 serum T4 and FT4 levels have been found in patients re¬ ceiving valproate monotherapy. We
found that serum T4 and FT4 levels were lowest in the patient group re¬ ceiving both carbamazepine and val¬ proate. These patients also had higher baseline thyrotropin serum levels. As far as we know, no reports have been published on the effect of the combined use of carbamazepine plus valproate on serum thyroid hormones. In every patient group receiving AEDs, except the one receiving phe¬ nytoin monotherapy, serum thyrotro¬ pin responses to thyrotropin-releasing
hormone were similar to the responses in the untreated patient group. The normal pituitary thyrotropin re¬ sponses to thyrotropin-releasing hor¬ mone in patients receiving carba¬
mazepine
Downloaded From: http://archneur.jamanetwork.com/ by a Karolinska Institutet University Library User on 05/24/2015
or
valproate monotherapy
consistent with most previous findings,1·2·18·32 but we have also re¬ ported slightly high thyrotropin re¬ sponses to thyrotropin-releasing hor¬ mone in patients receiving carba¬ mazepine monotherapy.19 However, a low thyrotropin response to thyrotro¬ pin-releasing hormone has also been reported in six carbamazepine-treated patients with affective illness15 and, in accordance with our results, in four nonepileptic volunteers receiving phenytoin.35 No data exists on the are
effects of
a
combination of carba¬
mazepine plus phenytoin, or carba¬ mazepine plus valproate on thyro¬ tropin-releasing hormone-stimulated thyrotropin responses. It has been postulated that the se¬ rum T4 and FT4 levels are low in epi¬ leptic patients receiving carbamaz¬ epine and/or phenytoin treatment be¬
of accelerated metabolism of thyroid hormones in the liver.32·36·37 Furthermore, an increased peripheral conversion of T4 to T3 during carba¬ mazepine and/or phenytoin therapy has also been suggested2 as an expla¬ nation for the only slightly changed or unchanged serum T3 levels. The possi¬ bility has also been raised3038 of the competitive binding of carbamazepine and phenytoin on thyroxin-binding globulin. The results of the present study support the hypothesis that in¬ creased metabolism is the main reason for decreased T4 and FT4 serum levels because carbamazepine and phenytoin are well known inducers of the microsomal enzyme system of the liver cause
metabolizing thyroid hormones, while valproate does not seem to have a sim¬ ilar enzyme-inducing effect.39 How¬ ever, the
serum
T4 and FT4 levels were
lowest in
patients receiving both car¬ bamazepine and valproate. Both car¬ bamazepine and valproate are highly bound to serum proteins.40·41 It seems possible that valproate could displace T4 from its binding sites on plasma proteins, and thereby lead to a larger amount of T4 liable to increased me¬ tabolism caused by the liver enzyme-
inducing properties of carbamazepine. Basal serum thyrotropin was not in¬ creased in patients who were receiving carbamazepine, phenytoin, or a com¬ bination of these two drugs, despite
the low serum levels of T4 and FT4. It is possible that the unchanged serum T3 levels are sufficient to inhibit in¬ creased secretion of thyrotropin. How¬ ever, an inhibitory effect of AEDs on the hypothalamic and/or anterior pi¬ tuitary level also seems probable. Be¬ cause the thyrotropin-releasing hor¬ mone-stimulated thyrotropin re¬ sponses were unaltered by these drugs
(except the response after 20 minutes in the phenytoin monotherapy group), a hypothalamic interference would seem more likely. The high levels of serum basal thyrotropin in patients receiving a combination therapy of carbamazepine plus valproate may be explained by the -aminobutyric acid ergic properties of valproate,42·43 since it is known that 7-aminobutyric acid is inhibitory of somatostatin release, and sumatostatin inhibits thyrotropin secretion.44 Epileptic seizures can manifest in association with thyrotoxicosis,45 and T4 may induce status epilepticus.46 It seems possible, therefore, that the lowering effect of AEDs on serum T4 and FT4 levels might contribute to the anticonvulsive properties of these
drugs.
Patients who are receiving pheny¬ toin monotherapy, or who were receiv¬ ing a combination therapy with car¬ bamazepine plus phenytoin had high levels of serum testosterone and SHBG, but patients who were receiv¬ ing other therapeutic regimens did not. Serum free thyroxine levels were unchanged in every patient group, while FAI values were low in all pa¬ tient groups, except the one group that was
receiving valproate monotherapy.
The lowest FAI values were found in patients receiving a combination ther¬ apy of carbamazepine plus valproate. Serum testosterone levels have been found to be mostly normal in male pa¬ tients with epilepsy who were receiv¬ ing AED polytherapy,4·20·22·24 but ele¬ vated serum concentrations of tes¬ tosterone have also been found.3 A few reports exist on normal serum testos¬ terone levels during carbamazepine and valproate monotherapy,13·23·24 and one report on a slightly increased se¬ rum testosterone level during pheny¬ toin monotherapy.13 Furthermore, high serum levels of SHBG have been found in male epilepsy patients receiv¬
ing phenytoin, carbamazepine, or AED polypharmacy.3·20·2224 In this study, the highest serum SHBG concentrations were found associated with phenytoin monotherapy. It seems possible that the high serum testosterone concen¬ trations in patients receiving pheny¬ toin medication are explained by these high SHBG levels. It appears reason¬ able to assume that the rise of serum SHBG concentrations is caused by in¬ creased synthesis due to liver enzyme induction by AEDs, as suggested before.23·47 Previously low serum free testoster¬ one levels have been reported in male patients receiving a combination of various AEDs.3·20 Recently, an un-
changed serum free testosterone level and decreased FAI were reported in male epileptic patients receiving car¬ bamazepine monotherapy.2348 Calcu¬
lated free testosterone fraction has also been found decreased in male pa¬ tients on receiving carbamazepine monotherapy, while phenytoin or val¬ proate did not have a similar effect.24 No previous data exists on serum free testosterone levels or FAI values in male epileptic patients receiving a combination of carbamazepine plus phenytoin, or carbamazepine plus val¬ proate. It is not yet clear whether the low FAI values have clinical signifi¬ cance in male patients receiving AEDs. The discrepancy between low FAI values and unchanged serum free testosterone concentrations may be explained by displacement of tes¬ tosterone from serum albumin by AEDs, as suggested before.23·48 Tes¬ tosterone binds with high affinity to SHBG, and loosely to albumin.49 In the present study patients receiv¬ ing phenytoin and valproate mono¬ therapy had slightly higher levels of serum E2 than the control subjects, but patients receiving other therapeu¬ tic regimens did not. The significance of this finding is not clear. However, serum progesterone levels were unaf¬ fected in every patient group. Serum dehydroepiandrosterone sul¬ fate levels were low in patients receiv¬
ing carbamazepine or phenytoin monotherapy, or receiving a combina¬ tion of carbamazepine plus phenytoin, or carbamazepine plus valproate. The levels were lowest in patients receiving phenytoin monotherapy, and un¬ changed in patients receiving val¬ proate monotherapy. The low serum levels of dehydroepiandrosterone sul¬ fate in male epileptic patients on re¬ ceiving carbamazepine, phenytoin, or both of these drugs confirm earlier findings.23·24·47·48·50 Previous reports on serum dehydroepiandrosterone sul¬ fate concentrations in patients receiv¬ ing a combination of carbamazepine plus valproate are lacking. Since, in the present study, all therapeutic reg¬ imens, including liver enzyme-induc¬ ing drugs, carbamazepine and/or phenytoin, were associated with low serum levels of dehydroepiandroster¬ one sulfate, while serum dehydroepi¬
androsterone sulfate concentrations unchanged in patients receiving valproate monotherapy, it seems likely that increased metabolism of dehy¬ droepiandrosterone sulfate is the main reason for the decreased serum dehy¬ droepiandrosterone sulfate levels that were found. Serum basal prolactin concentrawere
Downloaded From: http://archneur.jamanetwork.com/ by a Karolinska Institutet University Library User on 05/24/2015
were low in patients receiving phenytoin monotherapy, but other therapeutic regimens were associated with unchanged serum basal prolactin concentrations. However, the prolac¬ tin responses to thyrotropin-releasing hormone were slightly higher in pa¬ tients receiving carbamazepine, phe¬ nytoin, or the combination of these two drugs, but prolactin responses to me¬ toclopramide were higher in patients receiving phenytoin, a combination of carbamazepine plus phenytoin, or car¬ bamazepine plus valproate. Slightly higher or unchanged prolactin re¬ sponses to thyrotropin-releasing hor¬ mone and metaclopramide have been reported in patients receiving carba¬ mazepine, phenytoin, valproate, or AED polypharmacy.14'23-24'26 The reports on serum basal prolactin levels in male epileptic patients treated with car¬ bamazepine, phenytoin, or valproate are conflicting, but the levels have mostly been normal.14·16'23·24'26 All therapeutic regimens except car¬ bamazepine monotherapy were asso¬
with valproate in female patients with epilepsy, it being assumed that this was caused by interference of val¬ proate at the level of the .52 It is known that epileptic seizures can in¬ crease pituitary secretion of luteiniz¬ ing hormones53 and that a long history of seizures of temporal lobe origin may be associated with changes in the func¬ tion of .1011 Therefore, it may be assumed that the high serum levels of luteinizing hormone reported earlier in patients receiving AED polytherapy could be explained by effects of epi¬ lepsy per se, because it seems that the patients studied in many instances were institutionalized and suffered from severe epilepsy with frequently occurring seizures for many years. The patients in the present study had
epine, phenytoin, or valproate mono¬ therapy reported so far have been mostly normal,16·23·24·48 though a high basal luteinizing hormone serum level
slightly elevated, a hypothalamic rather than pituitary interference by these drugs seems likely. It has also been suggested that 7-aminobutyric acid-containing neurons may be in¬ volved in the regulation of gonadotro¬ pa secretion.55 This could explain the mechanism through which valproate has effects on serum luteinizing hor¬ mone concentrations. The regulation of gonadotropin secretion is complex,
tions
ciated with low basal serum luteiniz¬ ing hormone levels. Previously high serum luteinizing hormone levels have been reported in association with AED polytherapy.3·4·24 The luteinizing hor¬ mone serum levels in male patients with epilepsy receiving carbamaz¬
has also been found in association with carbamazepine24·26 and phenytoin26 monotherapy. Moreover, a low basal luteinizing hormone level has been ob¬ served in male patients receiving phé¬ nobarbital monotherapy,51 and amenorrhea following initiation of therapy
mostly well-controlled epilepsy. It has been previously suggested that the high serum luteinizing hor¬ mone levels in male epileptic patients are caused by a feedback mechanism due to low serum free testosterone levels.54 However, the results of the present study indicating low basal se¬ rum luteinizing hormone levels in most patient groups suggest a direct . effect of AEDs on the function of Because luteinizing hormone re¬ sponses to luteinizing hormonereleasing hormone are unaffected or
and the mechanisms through which other AEDs, eg, phenytoin, have ef¬ fects on serum luteinizing hormone levels are not clear. Despite the low serum luteinizing hormone levels in most patient groups, the basal and stimulated serum folli¬ cle-stimulating hormone were un¬ changed in most of the patient groups receiving AEDs. The results of the present study in¬ dicate that the serum hormone con¬ centrations and pituitary responses to various stimuli in male epileptic pa¬ tients receiving AEDs are different from those of control subjects. The therapeutic regimen with least effects on the hormone balance was valproate monotherapy, while combined therapy of carbamazepine and valproate had the most marked effect on both the thyroid hormone balance and the lev¬ els of FAI. It remains to be established how se¬ vere the clinical consequences of the hormonal changes found in this study really are, with regard to, eg, impo¬ tence, and a possible latent hypothyrosis in these patients. Nevertheless, these results are indicatory of marked differences between various AED reg¬ imens with regard to their effects on the hormone balance. This should be borne in mind when an AED regimen is considered for a patient with re¬
cently diagnosed epilepsy.
This study was supported with grants from the Finnish Neurology Foundation, Helsinki, Fin¬ land, and from the Research and Science Founda¬ tion of Farmos, Turku, Finland. The authors thank the personnel of the Depart¬ ment of Neurology and the Laboratory of Oulu (Finland) University Central Hospital for assis¬ tance in collecting samples and in laboratory work.
References
1. Fichsel H, Kn\l=o"\pfleG. Effects of anticonvuldrugs on thyroid hormones in epileptic children. Epilepsia. 1978;19:323-336. 2. Liewendahl K, Majuri H, Helenius T. Thyroid function tests in patients on long-term treatment with various anticonvulsant drugs. Clin Endocrinol. 1978;8:185-191. 3. Toone BK, Edeh J, Fenwick P, et al. Hormonal and behavioral changes in male epileptics. In: Porter RJ, Ward AA Jr, Mattson RH, Dam M, eds. Advances in Epileptology: XVth Epilepsy International Symposium. New York, NY: Raven sant
Press;
1984:283-289. E, Subramanian MG, Gilroy J. Investigation of sex hormones in male epileptic pa4. Rodin
Epilepsia. 1984;25:690-694. Haidukewych D, Rodin EA. Chronic antiepileptic drug therapy: classification by medication regimen and incidence of decreases in serum thyroxine and free thyroxine index. Ther Drug Monit.
tients. 5.
1987;9:392-398.
6. Aanderud S, Strandjord RE. Hypothyroidism induced by anti-epileptic therapy. Acta Neurol Scand. 1980;61:330-332.
7. Fenwick PBC, Toone BK, Wheeler MJ, Nanjee MN, Grant R, Brown D. Sexual behaviour in a centre for epilepsy. Acta Neurol Scand.
1985;71:428-435. 8. Mattson RH, Cramer JA, Collins JF,
et al.
Comparison of carbamazepine, phenobarbital, phenytoin, and primidone in partial and secondarily generalized tonic-clonic seizures. N Engl J
Med. 1985;313:145-151. 9. Webber MP, Hauser WA, Ottman R, Annegers JF. Fertility in persons with epilepsy: 1935-1974. Epilepsia. 1986;27:746-752. 10. Herzog AG, Seibel MM, Schomer DL, Vaitukaitis JL, Geschwind N. Reproductive disorders in women with partial seizures of temporal lobe origin. Arch Neurol. 1986;43:341-346. 11. Herzog AG, Seibel MM, Schomer DL, Vaitukaitis JL, Geschwind N. Reproductive endocrine disorders in men with partial seizures of temporal lobe origin. Arch Neurol. 1986;43:347-350. 12. Bilo L, Meo R, Nappi C, et al. Reproductive endocrine disorders in women with primary generalized epilepsy. Epilepsia. 1988;29:612-619. 13. L\l=u"\hdorfK, Christiansen P, Hansen JM,
Lund M. The influence of phenytoin and carbamazepine on endocrine function: preliminary results. In: Penry JK, ed. Epilepsy. The Eighth International Symposium. New York, NY: Raven Press; 1977:209-213. 14. Franceschi M, Perego L, Cavagnini F, et al. Effects of long-term antiepileptic therapy on the hypothalamic-pituitary axis in man. Epilepsia.
1984;25:46-52. 15. Joffe RT, Post RM, Ballenger JC, Rebar R,
Gold PW. Neuroendocrine effects of carbamazepine in patients with affective illness. Epi-
lepsia. 1986;27:156-160. 16. Isoj\l=a"\rviJIT, Myllyl\l=a"\VV, Pakarinen AJ. Effects of carbamezepine on pituitary responsiveness to luteinizing hormone-releasing hormone, thyrotropin-releasing hormone, and metoclopramide in epileptic patients. Epilepsia. 1989;30:50\x=req-\ 56. 17. Villa
SM, Alexander NM. Carbamazepine (Tegretol) inhibits in vivo iodide uptake and hormone synthesis in rat thyroid glands. Endocr Res. 1987;13:385-397. 18. M\l=o/\lholmHansen J, Skovsted L, Birk Lau-
Downloaded From: http://archneur.jamanetwork.com/ by a Karolinska Institutet University Library User on 05/24/2015
ridsen U, Kirkegaard C, Siersbaek-Nielsen K. The effect of diphenylhydantoin on thyroid function. J Clin Endocrinol Metab. 1974;39:785-789. 19. Isoj\l=a"\rviJIT, Pakarinen AJ, Myllyl\l=a"\VV: Thyroid function in epileptic patients treated with carbamazepine. Arch Neurol. 1989;46:1175\x=req-\ 1178. 20. Dana-Haeri J, Oxley J, Richens A. Reduction of free testosterone by antiepileptic drugs. Br Med J. 1982;284:85-86. 21. Rodin E, Subramanian MG, Schmaltz S, Gilroy J. Testosterone levels in adult male epileptic patients. Neurology. 1987;37:706-708. 22. Dana-Haeri J, Richens A. Effect of antiepileptic drugs on the hypothalamic-pituitary axis in epileptic patients. In: Dam M, Gram L, Penry JK, eds. Advances in Epileptology: The XIIth Epilepsy International Symposium. New York, NY: Raven Press; 1981:469-475. 23. Isoj\l=a"\rviJIT, Pakarinen AJ, Myllyl\l=a"\VV. Effects of carbamazepine therapy on serum sex hormone levels in male patients with epilepsy.
Epilepsia. 1988;29:781-786. 24. Macphee GJA, Larkin JG, Butler E, Beastall GH, Brodie MJ. Circulating hormones and pituitary responsiveness in young epileptic men receiving long-term antiepileptic medication. Epilepsia. 1988;29:468-475. 25. London DR, Loizou LA, Butt WR, Rovei V, Bianchetti G, Morselli PL. The effect of anti-convulsant drugs (AED) on hormonal responses in normal volunteers. In: Johannesen SI, Morselli PL, Pippenger CE, Richens A, Schmidt D, Meinardi H, eds. Antiepileptic Therapy: Advances in Drug Monitoring. New York, NY: Raven Press;
1980:405-411. 26. Dana-Haeri J, Oxley J, Richens A. Pituitary responsiveness to gonadotrophin-releasing and thyrotrophin-releasing hormones in epileptic patients receiving carbamazepine or phenytoin. Clin. Endocrinol. 1984;20:163-168. 27. Dreifuss FE, Bancaud J, Henriksen O, Rubio-Donnadieu F, Penry JK, Seino M. Proposal for revised clinical and electroencephalographic classification of epileptic seizures. Epilepsia. 1981; 22:489-501. 28. Chin W, Schussler GC. Decreased serum free thyroxine concentration in patients treated with diphenylhydantoin. J Clin Endocrinol Metab. 1968;28:181-186. 29. Yeo PPB, Bates D, Howe JD, et al. Anticonvulsants and thyroid function. Br Med J. 1978; 1:1581-1583. 30. Rootwelt K, Ganes T, Johannesen SI. Effect of carbamazepine, phenytoin and phenobarbital on serum levels of thyroid hormones and thyro-
tropin in humans. Scand Clin Lab Invest. 1978; 38:731-736. 31. Bentsen KD, Gram L, Veje A. Serum thyroid hormones and blood folic acid during monotherapy with carbamazepine or valproate. Acta Neurol Scand. 1983;67:235-241. 32. Connell JMC, Rapeport WG, Gordon S, Brodie MJ. Changes in circulating thyroid hormones during short-term hepatic enzyme induction with carbamazepine. Eur J Clin Pharmacol.
1984;26:453-456.
33. Tanaka K, Kodama S, Yokoyama S, et al. Thyroid function in children with long-term anticonvulsant treatment. Pediatr Neurosci. 1987;
13:90-94. 34. Liewendahl K, Majuri H. Thyroxine, triiodothyronine, and thyrotropin in serum during long-term diphenylhydantoin therapy. Scand J Clin Lab Invest. 1976;36:141-144. 35. Surks MI, Ordene KW, Mann DN, KumaraSiri MH. Diphenylhydantoin inhibits the thyrotropin response to thyrotropin-releasing hormone in man and rat. J Clin Endocrinol Metab. 1983; 56:940-945. 36. Larsen PR, Atkinson AJ Jr, Wellman HN, Goldsmith RE. The effect of diphenylhydantoin on thyroxine metabolism in man. J Clin Invest.
1970;49:1266-1279. 37. Mendoza DM, Flock EV, Owen CA Jr, Paris
J. Effect of 5,5\m='\-diphenylhydantoin on the metabolism of L-thyroxine-131I in the rat. Endocrinology.
1966;79:106-118.
38. Oppenheimer JH, Tavernetti RR: Studies the thyroxine-diphenylhydantoin interaction: effect of 5,5\m='\-diphenylhydantoinon the displacement of L-thyroxine from thyroxine-binding globulin (TBG). Endocrinology. 1962;71:496-504. 39. Perucca E, Hedges A, Makki KA, Ruprah M, Wilson JF, Richens A. A comparative study of the relative enzyme-inducing properties of anticonvulsant drugs in epileptic patients. Br J Clin Pharm. 1984;18:401-410. 40. Mattson RH, Cramer JA. Valproic acid (Depakane, Valontin). In: Browne TR, Feldman RG, eds. Epilepsy. Diagnosis and Management. Boston, Mass: Little Brown & Co; 1983:225-233. 41. Morselli PL, Gerna M, de Maio D, Zanda G, Viani F, Garattini S. Pharmacokinetic studies on carbamazepine in volunteers and in epileptic patients. In: Schneider H, Janz D, Gardner-Thorpe C, Meinardi H, Sherwin AL, eds. Clinical Pharmacology ofAntiepileptic Drugs. New York, NY: Springer-Verlag NY Inc; 1975:166-180. 42. L\l=o"\scherW, Schmidt D. Increase of human plasma GABA by sodium valproate. Epilepsia. on
1980;21:611-615.
43. L\l=o"\scherW. Effect of inhibitors of GABA\x=req-\ transaminase on the synthesis, binding, uptake and metabolism of GABA. J Neurochem. 1980; 34:1603-1608. 44. Foord SM, Peters JR, Dieguez C, et al. Thyroid stimulating hormone. In: Lightman SL, Everitt BJ, eds. Neuroendocrinology. Boston, Mass: Blackwell Scientific Publications; 1986:450-471. 45. Jabbari B, Huott AD. Seizures in thyrotoxicosis. Epilepsia. 1980;21:91-96. 46. Sundaram MBM, Hill A, Lowry N. Thyroxine-induced petit mal status epilepticus. Neurology.
1985;35:1792-1793.
47. Connell JMC, Rapeport WG, Beastall LGH, Brodie MJ. Changes in circulating androgens during short-term carbamazepine therapy. Br J Clin Pharm. 1984;17:347-351. 48. Isoj\l=a"\rviJIT, Pakarinen AJ, Myllyl\l=a"\VV. Effects of carbamazepine medication on the hypothalamic-pituitary-gonadal axis in male patients with epilepsy: a prospective study. Epilepsia. 1989;30:446-452. 49. S\l=o"\derg\l=a%o\rdR, B\l=a"\ckstr\l=o"\mT, Shanbhag V, Carstensen H. Calculation of free and bound fractions of testosterone and estradiol-17 to human plasma proteins at body temperature. J Steroid Biochem. 1982;16:801-810. 50. Levesque LA, Herzog Ag, Seibel MM. The effect of phenytoin and carbamazepine on serum dehydroepiandrosterone sulfate in men and women who have partial seizures with temporal lobe involvement. J Clin Endocrinol Metab. 1986; 63:243-245. 51. Murialdo G, Manni R, De Maria A, Bonura ML, Pollen A, Tartara A. Luteinizing hormone pulsatile secretion and pituitary response to gonadotropin releasing hormone and to thyrotropin releasing hormone in male epileptic subjects on chronic phenobarbital treatment. J Endocrinol Invest. 1987;10:27-31. 52. Margraf JW, Dreifuss FE. Amenorrhea following initiation of therapy with valproic acid.
Neurology. 1981;31:159.
53. Trimble MR, Dana-Haeri J, Oxley J, Baylis PH. Some neuroendocrine consequences of seizures. In: Porter RJ, Ward AA Jr, Mattson RH, Dam M, eds. Advances in Epileptology: XVth Epilepsy International Symposium. New York, NY: Raven Press; 1984:201-208. 54. Mattson RH, Cramer JA. Epilepsy, sex hormones, and antiepileptic drugs. Epilepsia.
1985;26:S40-S51. 55. Everitt BJ, Keverne EB. Reproduction. In: Lightman SL, Everitt BJ, eds. Neuroendocrinology. Boston, Mass: Blackwell Scientific Publications; 1986:472-537.
Downloaded From: http://archneur.jamanetwork.com/ by a Karolinska Institutet University Library User on 05/24/2015
