Indmedica Home | About Indmedica | Medical Jobs | Advertise On Indmedica
Search Indmedica Web
Indmedica - India's premier medical portal

Current Neurobiology

Studies on adverse metabolic effects of antiepileptics and their correlation with blood components

Author(s): Manisha Naithani, Sunny Chopra, B L Somani, R K Singh

Vol. 1, No. 2 (2010-10 - 2011-03)

Manisha Naithani, Sunny Chopra, B L Somani, R K Singh

Departments of Biochemistry, Shri Guru Ram Rai Institute of Medical and Health Sciences, Patel Nagar, Dehradun and Armed Forces Medical College, Pune, India.


Epilepsy is a neurological disorder knowing no geographical or social boundaries. Majority of patients can be treated with conventional drugs like phenytoin, phenobarbitone and carbamazepine. The present study was planned to correlate the metabolic side effects with the serum drug levels to bridge the existing gap between clinical and biochemical evaluation. Correlation of serum concentration of antiepileptics with thyroid profile, lipid profile, liver enzymes and electrolytes was carried out in 130 known patients of epilepsy on monotherapy or combination therapy. Phenytoin, carbamazepine and phenobarbitone estimation was done using High Pressure Liquid Chromatography. All the groups showed an increase in mean alkaline phosphatase and thyroid-stimulating hormone concentration as compared to healthy age and sex matched individuals. Significant positive correlation was found between serum carbamazepine and serum aspartate transaminase and alanine transaminase concen-tration (p<0.005). It was concluded that metabolic alterations are mostly mild and clinically insignificant and do not justify routine testing, except in those known to have a coexisting or recently developed hepatic abnormality.

Key Words: Epilepsy, Therapeutic drug monitoring, High Pressure Liquid Chromatography
Accepted July 22 2010


Epilepsy is the most common neurological disorder which knows no geographical, racial or social boundaries [1]. Approximately, 70-80% of the patients who develop epilepsy may expect to have their seizures controlled with optimal antiepileptic therapy.

Metabolic side effects of antiepileptic medications have been the cause of debate, whether these drugs require monitoring to assess and interventions to rectify the deranged metabolic markers. Antiepileptics may cause mild increase in liver function tests that tend to resolve over time [2]. Antiepileptic drugs also cause significant reduction in total thyroxine, free thyroxine, free thyroxine index, total triiodothyronine, free triiodothyronine and free triiodothyronine index [3]. These drugs also have effects on lipid profiles [4].

In view of the above-mentioned side effects, therapeutic drug monitoring becomes imperative in treatment of epilepsy. In the present study, we have used Micellar liquid chromatography to separate antiepileptics and tried to correlate the various metabolic side effects with serum levels of phenobarbitone, phenytoin and carbamazepine.

Material and Methods

The present study was carried out at the Department of Biochemistry, Armed Forces Medical College, Pune be-tween June, 2003 to Dec, 2005. One hundred known and twenty three freshly diagnosed patients of epilepsy attend-ing OPD were the subjects for the study. All patients were informed and consent was taken to participate in the study. The present study was cleared by the institutional ethical committee. 5 ml blood samples were collected in the morning during OPD hours in vacutainers. Serum was separated and was analyzed for drug levels using High Pressure Liquid Chromatography (HPLC), estimation of total cholesterol, alanine transaminase (ALT), aspartate transaminase (AST) and alkaline phosphatase (ALP) was carried out using Erba- Chem 5, biochemistry analyzer (Transasia Bio Medicals Ltd, Mumbai, India). Analyzer was cali-brated every alternate day for the various serum tests ac-cording to the manufacturers guideline that included total cholesterol using Cholesterol-DES kit (Source: Erba Manhiem, India), AST & ALT by UV kinetic, International Federation of Clinical Chemistry (IFCC) method using Autopak kit (Source: Bayers Diagnostics; India), ALP by pNPP-AMP (IFCC) kinetic assay using Autospan kit (Source: Bayers Diagnostics; India), IFCC denotes IFCC recommended tests. Thyroxine (T4), triiodothyron-ine (T3) and thyroid-stimulating hormone (TSH) were estimated by a Radioimmunoassay kits (Supplied by Board of Radiation and Isotope Technology, Mumbai).

Estimations of drug levels

The estimation of Anticonvulsant drugs was done by Micellar HPLC with direct injection of serum samples (5). Samples were diluted with 0.9% saline, filtered and used for HPLC analysis. Carbamazepine and phenytoin were procured from Sigma for the preparation of standards. For standard preparation of Phenobarbital injection Fenobarb (200mg/L) was used. The reagents used to prepare the Micellar mobile phase were Sodium Dodecyl Sulphate (Sigma), 1-butanol (Qualigens), disodium hydrogen phosphate, potassium dihydrogen phosphate and Hydrochloric acid (Merck). Millipore water from Milli Q was used through out the analysis.

The chromatograph (Merck HITACHI) was equipped with a quaternary pump (Lachrom L-7100) and an ultra-violet–visible detector (DAD L- 7455). The flow rate, injection volume, and detection wavelength were 1.0 mL/min, 20 μL, and 220 nm, respectively. Analytical column packed with RP silica gel (Lichrocart C18) was used. The software, HPLC System Manager (Hitachi Chroma-tography Data Software Station) was used for data handling.

The mobile phase selected as optimum was 0.05 mol/L Sodium Dodecyl Sulphate plus 56.7 mL/L 1-butanol, which gave excellent resolution and allowed the analysis time to be 20 min. All mobile phases contained phosphate buffer at pH 6.88. Calibration curves were constructed for each antiepileptic drug and quantification was also done using internal standard and serum calibration standard procured from Chromsystems.


Subjects for study were divided into the following groups:
Group 1: Patients on monotherapy (phenobarbitone/ car-bamazepine/ phenytoin)
Group 2: Patients on polytherapy (combination of pheno-barbitone and/ or carbamazepine and/ or phenytoin)
Group 3: Patients on polytherapy (combination of pheno-barbitone and/ or carbamazepine and/ or phenytoin with clobazam and/ or valproate)
60 healthy age and sex matched individuals were included in the study to compare the various metabolic parameters (Table 1).

Table 1. Mean values of various metabolic parameters in healthy volunteers.

Metabolic parameters Mean ± SD
Total Cholesterol 175± 32.8 mg/dl
ALT 21.50± 8.43 IU/L
AST 26.75± 9.19 IU/L
ALP 109.37± 20.0IU/L
TSH 2.12± 0.929 μg/dl
T3 1.27± 0.42 IU/ml
T4 10.57± 2.61 ng/dl

Table 2. Correlation between serum values and various parameters in Group 1

Parameters Correlation Coefficient
Phenytoin Carbamazepine
Total Cholesterol 0.102 -0.094
AST -0.078 0.566 ٭٭
ALT -0.023 0.693#
ALP -0.094 0.225
TSH 0.030 0.041
T3 -0.009 -0.162

٭٭ Significant positive correlation between serum Car-bamazepine and serum AST concentration (p<0.005).

>Significant positive correlation between serum Car-bamazepine and serum ALT concentration (p<0.005).

A higher mean of total cholesterol in group 2 (186.9±53.75 mg/dl) as compared to healthy volunteers (175.85±32.8 mg/dl) was obtained. This could not be attributed to one drug as patients were on polytherapy. While group 1 and 3 had decrease levels as compared to healthy volunteers.

Mean AST levels (28.64 IU/L) and ALT levels (21.96 IU/L ) in group 1 were very similar to healthy volunteers but both showed a statistically significant correlation with serum carbamazepine levels (p< 0.05). All the groups showed an increase in mean ALP concentration as compared to control (Table 3).

All groups exhibited increase in mean values of TSH compared to healthy volunteers with highest mean in group 3 (3.28±1.74μg/dl). All of the groups showed a statistically insignificant decrease in the mean value of T4 as compared to healthy volunteers. Group 1 and 2 showed a decrease in the mean value of T3 as compared to healthy volunteers while group 3 showed a statistically insignificant increase in mean concentration.

All the groups exhibited an increase in mean ALP concentration as compared to healthy volunteers. Group 1 patient on carbamazepine monotherapy showed positive correlation with serum levels.

Table 3. Serum Alkaline Phosphatase in Group 1, 2, 3 and control

Group No. of
Mean ± SD
Control 60 109.37±20.0
Group 1
n = 84
84 187.55±115.38
Group 2
n = 7
7 205.84±115.89
Group 3
n = 32
32 186.86±148.04


The present study showed only a nonsignificant increase in total cholesterol in those receiving combination therapies while those on monotherapy did not show such in-crease. There are studies supporting this observation [6-10].

There was an increase in liver enzymes in group 1 which is in agreement with previous studies. This is in confor-mity with previous studies where serum activities of liver enzymes in patients receiving a long-term anticonvulsant monotherapy were examined retrospectively and it was found that there was a predominant elevation of serum gamma glutamyl transferase (GGT) and ALP and that all enzymes evaluated were more often raised and attained higher values in those receiving phenytoin rather than carbamazepine [11]. Similar enzyme elevations were re-ported in other studies [12, 13].

There is controversy regarding the exact mechanism for increased enzyme activities. Some studies conclude that increase occurs due to enzyme induction along with liver cell damage [14], while other studies maintain that increase is due to enzyme induction and is mostly mild and clinically insignificant [15]. In our study none of the sub-jects suffered from liver disease, thus, mild increase found in enzyme levels may only reflect enzyme induc-tion and not hepatocellular damage.

Our observations regarding thyroid profile have also been supported by a number of previous studies. Rousso et al (1984) found that phenobarbital receiving pateints had low serum T4 levels and free T4 indexes but had no changes in TSH concentrations [16]. Similarly, Tiihonen et al (1995) reported that patients receiving anticonvulsant drugs chronically are eumetabolic and do not need thyrox-ine supplementation [17].

The exact cause for such changes in profile of thyroid function tests is unknown but the probable mechanisms have been studied intensively. Hamada et al (1979) studied effects of diphenylhydantoin on thyroid-hormone binding and concluded that drug induced inhibition of the thyroid hormone binding was probably responsible. While many other studies maintained that increase in conversion and metabolism of the thyroid hormones could explain this effect [18]. The results observed in our study on thyroid profile can thus be explained by interference of drugs with thyroid hormone binding to thyroxine binding globulin and enzyme-induced increased metabolic clearance rate of thyroid hormones.

It could be concluded that metabolic alterations are mostly mild and clinically insignificant and do not justify routine testing, except in those known to have a coexisting hepatic abnormality and those who develop symptoms of hepatic involvement while receiving antiepileptics or develop symptoms of hypothyroidism.


  1. Carpio A and Hauser W A. Epilepsy in developing world. Current Neurology Neuroscience Reports 2009; 9: 319-326.
  2. Warner A, Privitera M, Bates D. Standards of labora-tory practice: antiepileptic drug monitoring. Clinical Chemistry 1998;44: 1085-1095.
  3. Kirimi E, Karasalihogouglus, Boz A. Thyroid function in children under long-term administration of antiepi-leptic drugs. Eastern Journal Of Medicine 1999; 4(1): 23-26.
  4. Branswig S, Karkseiek A, Sudhop T, Luers C, Berg-mann K V and Berthold H K. Carbamazepine increases atherogenic lipoproteins: mechanism of action in male adults. Am J Physiol. Feb 2002; 282(2): 704-716.
  5. Armstrong DW, Nome F. Partitioning behavior of sol-utes eluted with micellar mobile phases in liquid chro-matography. Anal Chem 1981; 53:1662-1666.
  6. Pylvanen V, Knip M, Pakarinen AJ, Turkka J, Kotila M, Rattya J, Myllyla VV, Isojarvi JI. Fasting serum in-sulin and lipid levels in men with epilepsy. Neurology. 2003 Feb; 60(4): 571-4.
  7. Deda G, Caksen H, Icagasioglu D. Effect of long-term carbamazepine therapy on serum lipids, vitamin B12 and folic acid levels in children. J Pediatr Endocrinol Metab. 2003 Feb; 16(2): 193-6.
  8. Yalcin E, Hassanzadeh A, Mawlud K. The effects of long-term anticonvulsive treatment on serum lipid pro-file. Acta Paediatr Jpn. 1997 Jun; 39(3): 342-5. Naithani/Chopra/ Somani/ Singh Current 120 Neurobiology Volume 1 Issue 2
  9. Calandre EP, Rodriquez-Lopez C, Blazquez A, Cano D. Serum lipids, lipoproteins and apolipoproteins A and B in epileptic patients treated with valproic acid, carbamazepine or Phenobarbital. Acta Neurol Scand. 1991 Apr; 83(4): 250-3.
  10. Eiris JM, Lojo S, Del Rio MC, Novo I, Bravo M, Pavon P, Castro-Gago M. Effects of long-term treat-ment with antiepileptic drugs on serum lipid levels in children with epilepsy. Neurology. 1995 Jun; 45(6): 1155-7.
  11. Aldenhovel HG. The influence of long-term anticon-vulsant therapy with diphenylhydantoin and car-bamazepine on serum gamma-glutamyltransferase, as-partate aminotransferase, alanine aminotransferase and alkaline phosphatase. Eur Arch Psychiatry Neurol Sci. 1988; 237(5): 312-6.
  12. Mendis GP, Gibberd FB, Hunt HA. Plasma activities of hepatic enzymes in patients on anticonvulsant therapy. Seizure. 1993 Dec; 2(4): 319-23.
  13. Rao ML, Stefan H, Scheid C, Kuttler AD, Froscher W. Serum amino acids, liver status, and antiepileptic drug therapy in epilepsy. Epilepsia. 1993 Mar-Apr; 34(2): 347-54.
  14. Deutsch J, Fritsch G, Golles J, Semmelrock HJ. Effects of anticonvulsive drugs on the activity of gammaglu-tamyltransferase and aminotransferases in serum. : J Pediatr Gastroenterol Nutr. 1986 Jul-Aug; 5(4): 542-8.
  15. Verma NP, Haidukewych D. Differential but infrequent alterations of hepatic enzyme levels and thyroid hor-mone levels by anticonvulsant drugs. Arch Neurol. 1994 Apr; 51(4): 381-4.
  16. Rousso I., Pharmakiotis A., Gatzola M., Karatza E.,Tourkantonis A. Effects of phenobarbital, diphenyl-hydantoin and carbamazepine on thyroid function in epileptic children. Acta Endocrinologia 1984; Suppl. 265:48-49.
  17. Tiihonen M, Liewendahl K, Waltimo O, Ojala M, Valimaki M. Thyroid status of patients receiving long-term anticonvulsant therapy assessed by peripheral pa-rameters: a placebo-controlled thyroxine therapy trial. Epilepsia. 1995 Nov; 36(11): 1118-25.
  18. Strandjord RE, Aanderud S, Myking OL, Johannessen SI. Influence of carbamazepine on serum thyroxine and triiodothyronine in patients with epilepsy. Acta Neurol Scand. 1981; 63 (2):111-21.

Manisha Naithani,

Department of Biochemistry
Shri Guru Ram Rai Institute of Medical and Health Sciences
Patel Nagar, Dehradun 248001,Uttrakhand, India
E mail: manishasimalti(at)

Adverse metabolic effects of antiepileptics in correlation with blood components
Current Neurobiology Volume 1 Issue 2 121

Access free medical resources from Wiley-Blackwell now!

About Indmedica - Conditions of Usage - Advertise On Indmedica - Contact Us

Copyright © 2005 Indmedica