Effects of antiepileptic monotherapy on hematological and

1

Effects of antiepileptic monotherapy on

hematological and biochemical parameters

Yuri Yoshimura1, Keiko Hara2,3, Miho Akaza3, Kaseya Ohta4, Yuki Sumi3, Motoki Inaji1, Eriko Yanagisawa5,

Taketoshi Maehara1

Purpose: Severe cytopenia and liver dysfunction are character ized as antiepileptic drug (AED)

adverse effects dependent upon an idiosyncrasy. However, in clinical practice, we often find hemato-

logical and biochemical changes during AED treatment with causes other than an idiosyncrasy. This

study aims to investigate the effect of antiepileptic monotherapy on hematological and biochemical

parameters.

Methods: We retrospectively recruited 480 patients untreated with AED at baseline. Changes in

hematological and biochemical parameters before and after initiation of medication were investigated,

and correlation with plasma concentrations of AED was analyzed.

Results: Sixty-six of 480 patients treated with carbamazepine (CBZ: n = 27), sodium valproate (VPA:

n = 19) or levetiracetam (LEV: n = 20) monotherapy were eventually selected for analysis. After CBZ

treatment, decreased white blood cell (WBC) count and increased gamma-glutamyltransferase (GGT)

and alkaline phosphatase (ALP) activities were recorded at high frequencies. Decreased WBC count

tended to correlate with elevated serum CBZ level. Elevated GGT activity was observed in all patients

treated with CBZ. In patients treated with VPA, platelet (PLT) counts decreased. In patients treated

with LEV, there were no significant differences in the measured parameters before and after medica-

tion.

Discussion: We considered that the reduction in WBC count might be dose-dependently related to

AEDs. Elevated GGT activity was observed in all patients treated with CBZ, but the average increase

in GGT activity was 35.19 ± 33.08 U/L. In patients undergoing VPA treatment, decreased PLT counts

were also observed at high frequency. Thus, hematological and biochemical parameters should be

closely monitored in patients receiving AED, especially in patients treated with high doses of AEDs.

Corresponding author: Keiko Hara

Department of Respiratory and Nervous System Science, Graduate School of Medical and Dental Sciences, Tokyo Medical and

Dental University, 1-5-45 Yushima, Bunkyo-ku, Tokyo 113-8519, Japan

Tel: +81-3-5803-5077; Fax: +81-3-5803-5077; E-mail: [email protected]

Epilepsy & Seizure Journal of Japan Epilepsy Society

Vol. 11 No. 1 (2019) pp. 1-13

1Department of Neurosurgery, Graduate School of Medical and Dental Sciences, Tokyo Medical and Dental

University 2Hara Clinic 3Department of Respiratory and Nervous System Science, Graduate School of Medical and Dental Sciences,

Tokyo Medical and Dental University 4Onda-Daini Hospital 5Medical Technology, School of Health Care Sciences, Faculty of Medicine, Tokyo Medical and Dental

University

Abstract

Received: August 19, 2018 ; Accepted: December 10, 2018

Key words: Antiepileptic drug; Hematological; Biochemical; effect; Monotherapy; Leukocyte reduction

Original Article

mailto:[email protected]

2

Introduction Antiepileptic drugs (AEDs) are frequently

used for several conditions including epilep-

sy, psychiatric disorders, and neuropathic

pain. In particular, AEDs are essential for the

treatment of patients with epilepsy. The goal

of AED treatment is to optimize seizure con-

trol and quality of life while minimizing

treatment toxicity [1]. Appropriate pharmaco-

logical management can result in freedom

from seizures in 60%–70% of the patients,

with more than 90% of them being controlled

by monotherapy [2].

However, various AED side effects have

been reported. Commonly occurring side ef-

fects of AEDs are drowsiness, dizziness,

weight changes, nausea, memory problems,

drug eruption, tremors, impaired liver func-

tion, pancytopenia, gastrointestinal symp-

toms, osteoporosis, depression, among other

symptoms. In 10% of the patients treated

with carbamazepine, adverse reactions are

noted, including allergic rashes or leukopenia

[3]. AEDs are associated with 8.3% of reports

of drug-induced liver injury, which is the ma-

jor reason for their withdrawal from thera-

peutic use [4]. The “Patients with Epilepsy

Practice Guideline 2018 in Japan” describes

the mechanism of AED side effects as usually

divided into three types: depending on an idi-

osyncrasy, dose-dependence, and long-term

pharmacotherapy [5]. Cytopenia and liver

dysfunction are classified as adverse effects

depending on an idiosyncrasy. However, in

clinical practice, hematological and biochem-

ical changes caused by AEDs are encoun-

tered more often than reported. Several previ-

ous reports have described the hematological

and biochemical changes caused by AEDs,

but these reports included patients on AED

polytherapy and compared the hematological

and biochemical parameters between control

patients and patients with epilepsy. There are

few reports that analyzed hematological and

biochemical changes caused by AEDs includ-

ing only patients who were initially untreated

with AED and then treated with AED mono-

therapy. In our longitudinal study, we ana-

lyzed the effects of AEDs on hematological

and biochemical parameters of patients un-

treated with AEDs at baseline.

Patients and Methods Participants

This study was approved by the Ethics

Committee of Tokyo Medical and Dental

University (#556). We retrospectively recruit-

ed 480 patients untreated with AED at base-

line in the Department of Neurosurgery, To-

kyo Medical and Dental University between

April 2012, and December 2017, and in the

Hara Clinic between January 2000 and De-

cember 2017. Both facilities were accredited

by The Japan Epilepsy Society. All the pa-

tients selected for this study were untreated

with any AED at the first medical examina-

tion blood sampling. In six patients, a mean

of 1945 ± 3135 (73–8269) days had elapsed

since the discontinuation of AED medication;

the remaining patients were never exposed to

AEDs previously. All enrolled patients were

subsequently treated with AED monotherapy.

Data

We longitudinally investigated hematological

and biochemical parameters twice in each

patient receiving AED monotherapy. Initial-

ly, we collected data of a patient within 3

Effects of antiepileptic monotherapy Yuri Yoshimura et al.

3

months before AED initiation. From the same

patient, we collected hematological and bio-

chemical parameters and AED plasma con-

centrations from 2 weeks to 3 months after

AED initiation. Hematological parameters

included white blood cell count (WBC), red

blood cell count (RBC), hemoglobin (HGB),

hematocrit (HCT), and platelet count (PLT).

Patients taking folic acid were excluded from

hematological analysis because folic acid is

associated with the production of blood cells.

Biochemical parameters included measure-

ments of aspartate aminotransferase activity

(AST), alanine aminotransferase activity

(ALT), gamma-glutamyltransferase activity

(GGT), alkaline phosphatase activity (ALP),

lactate dehydrogenase level (LDH), blood

urea nitrogen concentration (BUN), and cre-

atinine (CRE). Patients aged 18 years and

under were excluded from liver function

evaluation of AST, ALT, GGT, ALP and

LDH analysis because these reference values

differ between adults and children. Hemato-

logical and biochemical parameters were

compared between before and after admin-

istration of AED. For the hematological and

biochemical parameters that showed signifi-

cant differences, the rates of change (post-/

pre-medication) were calculated, and we ana-

lyzed whether these differences correlated

with AED plasma concentrations.

Statistics

Data were analyzed using SPSS version 25

(IBM Corp., Armonk, NY, USA). Paired t-

test was used for the analyses of continuous

data, comparing pre- with post-medication

data. Correlation between the rates of change

in parameters and AED plasma concentra-

tions was assessed using Pearson’s correla-

tion coefficient. For non-normally distributed

data, Wilcoxon signed rank test, and

Spearman’s correlation coefficient were used.

A two-sided P < 0.05 was considered statisti-

cally significant.

Epilepsy & Seizure Vol. 11 No. 1 (2019)

Figure 1 Patient flow char t. AED: antiepileptic drug.

4

Results A flow chart of patient selection is pre-

sented in Figure 1. A total of 480 patients

were recruited, who were not treated with

AED at baseline and began AED monothera-

py at the beginning of this study period.

Three hundred of the 480 patients (63%)

were excluded because of comorbidities com-

prising cancer (n = 130), traumatic brain inju-

ries (n = 128), cerebrovascular disease (n =

11), convulsive attacks (n = 11), cerebral

fracture (n = 5), cerebral abscess (n = 5), liver

or renal dysfunction (n = 4) and other comor-

bidities (n = 6). The reason for exclusion is

that blood cell counts or liver enzyme levels

may be influenced by chemotherapy, immu-

nosuppressive drugs, acute stress, surgical

wound infection, or co-administered drugs.

Thirteen of the 480 patients (3%) were fur-

ther excluded because of discontinuation of

AED or change in AED within two weeks

after initiation of the first AED: carbamaze-

pine (CBZ, eight patients), sodium valproate

(VPA, three patients), levetiracetam (LEV,

one patient) and lacosamide (LCM, one pa-

tient). Six of 13 patients discontinued AED

treatment for reasons other than side effects,

such as discontinuation of hospital visit and

poor adherence. The remaining seven patients

changed to another AED because of side ef-

fects. Hypacusis (n = 1), drowsiness (n = 1),

or dizziness (n = 3) were reported in five pa-

tients treated with CBZ, while drug eruption

(n = 1) and decreased WBC count (n = 1)

were reported in two patients initiated with

VPA. The blood sampling time points before

and after AED medication were designated

“pre-medication” when sampled within three

months from initiation of AED medication

and “post-medication” when sampled from 2

weeks to 3 months after initiation of AED

therapy. The case of increasing AED dosage

was regarded as an exception in this criterion

related to the blood sampling time points.

Ninety-one of 480 patients (19%) were ex-

cluded because of no blood sampling at the

designated time points. Four patients were

excluded due to other factors such as preg-

nancy (n = 2), hemolysis (n = 1), and abnor-

mal AST level before AED medication. (n =

1). Seventy-two of 480 patients (15%) re-

mained for further analysis. Eventually, we

selected 66 patients treated with CBZ, VPA

or LEV monotherapy for final analysis (CBZ:

n = 27, VPA: n = 19, LEV: n = 20) (Table 1).

Six patients were excluded from analysis be-

cause of insufficient patient numbers for

AED monotherapy (phenytoin: n = 1,

lamotrigine: n = 3, zonisamide: n = 2). Of 66

patients, 36 patients had localization-related

epilepsy, 18 exhibited generalized epilepsy,

five patients were unknown, one patient was

diagnosed with bipolar disorder, and six pa-

tients were treated with AED for convulsion

prevention. According to patient medical rec-

ords, 28 of 66 patients were prescribed medi-

cations other than AED, which included med-

ications administered only occasionally. All

patients included were deemed otherwise

healthy at blood sampling.

CBZ

One patient taking folic acid was excluded

from analysis. Finally, hematological param-

eters were analyzed in 26 patients. Post-

medication WBC counts were significantly

lower compared with pre-medication WBC

counts (pre-medication 6.24 ± 1.90×103/µL,

post-medication 5.28 ± 1.56×103/µL, P <

Yuri Yoshimura et al. Effects of antiepileptic monotherapy

5

0.001) (Table 2). In 18 of 26 patients treated

with CBZ (69%), WBC counts decreased fol-

lowing treatment compared with pre-

medication counts. The decrease in WBC

count was 1.48 ± 1.10×103/µL in these 18

patients. None of the patients had WBC count

decreased below the lower limit of reference

values (3.5×103 to 9.7×103/µL). Six patients

aged 18 years or younger were excluded from

analysis of the following parameters: AST,

ALT, GGT, ALP, and LDH, and these pa-

rameters were analyzed in 21 patients. Post-

medication GGT and ALP activities were sig-

nificantly higher compared with pre-

medication levels (GGT: pre-medication

28.38 ± 17.94 U/L, post-medication 63.57 ±


Total CBZ VPA LEV

Number of patients 66 27 19 20

Gender, n (%)

Male 37 (56) 16 (59) 11 (58) 10 (50)

Female 29 (44) 11 (41) 8 (42) 10 (50)

Age of drug initiation, year

mean ± SD 37.1 ± 19.0 39.3 ± 21.1 30.2± 14.3 40.8± 19.0

range 10-78 10-78 12-59 16-74

≤18 years, n (%) 12 (18) 6 (21) 3 (16) 3 (15)

Duration from AED initiation to post-

medication blood-draw, days

mean ± SD 83.4±83.9 95.1±94.6 72.9±88.5 77.7±64.1

range 14-381 14-381 14-342 14-259

Number of patients treated with folic acid, n (%) 5 (7) 1 (4) 3 (16) 1 (5)

Table 1 Demographic character istics of patients analyzed.

AED: antiepileptic drug, CBZ: carbamazepine, VPA: sodium valproate, LEV: levetiracetam

Table 2 Compar ison of pre- and post-medication values of hematological and biochemical parameters in patients treated with carbamazepine (CBZ).

Parameter n Pre Post P-value T value Z value

WBC (×103/μL) 26 6.24±1.90 5.28±1.56 <0.001* 3.83

RBC (×104/μL) 26 448.58±59.50 449.27±47.87 0.920 -0.10

HGB (g/dL) 26 13.71±1.55 13.75±1.22 0.845 -0.20

HCT (%) 26 41.89±4.54 42.22±3.83 0.607 -0.52

PLT (×104/μL) 26 24.16±5.89 23.46±5.84 0.451 0.77

AST (U/L) 21 21.71±5.05 30.29±26.13 0.130 -1.58

ALT (U/L) 21 19.48±8.30 28.95±27.23 0.154 -1.48

GGT (U/L) 21 28.38±17.94 63.57±42.49 <0.001* -4.02

ALP (U/L) 20 237.50±110.50 267.35±107.57 0.006* -3.06

LDH (U/L) 20 230.30±74.36 247.50±103.89 0.279 -1.11

BUN (mg/dL) 25 13.62±3.44 14.54±4.57 0.135 -1.55

CRE (mg/dL) 25 0.74±0.15 0.75±0.19 0.210 -1.25

WBC: white blood cell, RBC: red blood cell, HGB: hemoglobin, HCT: hematocrit, PLT: platelet, AST: aspartate aminotransferase, ALT: alanine aminotransferase, GGT: gamma-glutamyltransferase, ALP: alka-line phosphatase, LDH: lactate dehydrogenase, BUN: blood urea nitrogen, CRE: creatinine Data are presented as mean ± SD. *: P < 0.05

6

42.49 U/L, P < 0.001; ALP: pre-medication

237.50 ± 111.50 U/L, post-medication 267.35

± 107.57 U/L, P = 0.006). In all patients treat-

ed with CBZ (100%), GGT activities were

increased on average by 35.19 ± 33.08 U/L

following CBZ commencement. In seven of

21 patients (33%), GGT levels were higher

than the upper limit of the reference values

(reference: females ≤ 48 years, males ≤ 79

years; females: n = 4, pre-medication 23.75 ±

3.59 U/L, post-medication 90.75 ± 47.52 U/

L; males: n = 3, pre-medication 68.67 ± 6.43

U/L, post-medication 131.33 ± 34.44 U/L). In

17 of 20 patients treated with CBZ (85%),

ALP levels were higher after CBZ therapy

was initiated. The increase in ALP activity

averaged 40.47 ± 37.62 U/L in 17 patients

after CBZ was prescribed. Two of 17 patients

(12%) had ALP levels higher than the upper

limit of the reference values (reference: 104

to 338 U/L; pre-medication 463.50 ± 284.96

U/L, post-medication 532.00 ± 192.33 U/L).

The mean post-medication plasma CBZ con-

centration was 5.05 ± 1.71 µg/mL in 26 pa-

tients, not including those taking folic acid.

The rates of change in WBC count tended to


Figure 2 Cor relation between CBZ plasma concentration and WBC reduction. CBZ: carbam-azepine, WBC: white blood cell


WBC (×103/μL) 16 6.39±1.67 5.82±1.63 0.141 1.55

RBC (×104/μL) 16 486.75±58.92 473.56±68.71 0.139 1.56

HGB (g/dL) 16 14.90±1.80 14.69±2.12 0.347 0.97

HCT (%) 16 45.58±5.14 14.69±2.12 0.237 1.23

PLT (×104/μL) 16 24.17±6.87 20.71±6.31 0.006* 3.17

AST (U/L) 16 23.50±7.06 21.88±8.52 0.249 1.20

ALT (U/L) 16 21.13±9.35 20.44±15.37 0.774 0.29

GGT (U/L) 16 31.88±30.89 33.81±57.17 0.232 -1.20

ALP (U/L) 12 234.75±57.46 215.33±86.34 0.152 1.54

LDH (U/L) 13 190.38±35.93 190.77±35.83 0.953 -0.06

BUN (mg/dL) 17 13.51±3.13 14.15±2.77 0.406 -0.85

CRE (mg/dL) 17 0.72±0.16 0.77±0.18 0.133 -1.50


Table 3 Compar ison of pre- and post-medication values of hematological and biochemical parameters in patients treated with sodium valproate (VPA).

7

correlate with post-medication plasma con-

centration, but this effect was not significant,

as shown in Figure 2 (Pearson’s correlation

coefficient, R = -0.38, P = 0.056). As CBZ

plasma concentration increased, WBC count

tended to decrease. On the other hand, the

rates of GGT and ALP changes did not corre-

late with post-medication plasma CBZ con-

centration.

VPA

Post-medication PLT count was signifi-

cantly lower (pre-medication 24.17 ±

6.87×104/µL; post-medication 20.71 ±

6.31×104/µL, P = 0.006) (Table 3). In 14 of

16 patients treated with VPA (88%), PLT

counts were lower after VPA treatment was

initiated. The decrease in PLT count averaged

4.21 ± 4.14×104/µL in 14 patients. Two of the

14 patients (14%) had PLT counts lower than

the lower limit of reference values (reference:

15.8×104 to 34.8×104/µL; pre-medication

16.85 ± 1.20×104/µL, and post-medication

11.25 ± 2.62×104/µL). Mean post-medication

plasma VPA concentration was 36.29 ± 25.90

µg/mL. The rate of PLT change did not cor-

relate with post-medication plasma concen-

tration. There were no other significant

changes between pre- and post-medication

parameters.

LEV

There were no significant differences in the

measured parameters between pre- and post-

medication values (Table 4) following LEV

treatment. The mean post-medication plasma

concentration was 10.68 ± 10.16 µg/mL (n =

16).

Discussion This study is retrospective. Therefore, the

number of patients varied in different param-

eters measured and the number of patients


Table 4 Compar ison of pre- and post-medication values of hematological and biochemical parameters in patients treated with levetiracetam (LEV).


WBC (×103/μL) 18 6.85±2.62 6.06±1.50 0.140 1.55

RBC (×104/μL) 18 467.89±37.93 464.94±40.65 0.649 0.46

HGB (g/dL) 18 14.16±1.14 14.10±1.44 0.741 0.34

HCT (%) 18 43.65±3.37 43.35±3.67 0.626 0.50

PLT (×104/μL) 18 25.69±4.57 24.36±4.07 0.132 1.58

AST (U/L) 17 21.24±9.35 19.82±4.26 0.501 0.69

ALT (U/L) 17 15.18±4.85 15.41±8.22 0.892 -0.14

GGT (U/L) 17 36.71±37.78 39.71±47.42 0.836 -0.21

ALP (U/L) 12 207.42±61.04 200.42±46.61 0.584 0.56

LDH (U/L) 15 191.20±29.84 183.73±30.20 0.323 1.02

BUN (mg/dL) 19 11.54±1.96 11.00±2.63 0.416 0.83

CRE (mg/dL) 19 0.69±0.14 0.69±0.15 0.952 -0.06


8

studied was not large. However, the subjects

included in this study were AED-untreated

patients who initiated AED monotherapy, and

did not possess complicating factors that

would affect liver function or blood cell

counts.

In our study, WBC counts significantly

decreased after CBZ treatment in 69% of the

patients. An observed WBC reduction in

3.7% of patients has been previously reported

[6]. However, this figure might indicate pa-

tients who had lower WBC counts than the

reference value. In this study, WBC count

decreased at a higher frequency than previ-

ously reported and none of the patients had

WBC count lower than the lower reference

limit (3.5×103 to 9.7×103/µL) after initiation

of CBZ treatment.

Bachmann et al. [7] showed a statistically

significant increase in WBC count in women

administered CBZ (P = 0.039) compared with

controls. Conversely, our study evaluated the

hematological changes before and after medi-

cation in identical patients. Moreover, we re-

vealed that the rate of WBC change tended to

correlate with post-medication plasma CBZ

concentration. Also, because we excluded

patients who discontinued AED within two

weeks after AED initiation, patients with side

effects depending on an idiosyncrasy may

have been excluded from our analysis. Our

results suggest a potential mechanism other

than allergy or drug toxicity to account for

these observations.

Huang et al. [8] suggested that the serum

levels of folate and vitamin B12 following

AED treatment were significantly lower than

those before AED treatment. It is well known

that folate and vitamin B12 are necessary for

DNA synthesis. In our study, hematological

changes other than blood cell counts were not

revealed. Also, the absence of measured fo-

late and vitamin B12 data impeded us from

investigating those changes. However, there

is a possibility that low serum levels of folate

and vitamin B12 caused by AED resulted in

the observed high frequency of WBC count

reduction. We found that decreased WBC

counts were frequently induced by CBZ treat-

ment. Because WBC subsets were not inves-

tigated, we should prospectively monitor

WBC subsets of patients treated with CBZ in

future studies.

CBZ has been described as a potent induc-

er of microsomal enzymes in the liver [9].

Liver enzymes induced by CBZ have been

previously documented [10]. However, there

are few reports of the relationship between

enzyme induction and a concrete upper limit

of elevated GGT activity. In our report, a pa-

tient with GGT level of 171 U/L showed no

clinical symptoms of liver dysfunction.

The frequency of elevated GGT has been

widely reported to be 34%-100% in patients

undergoing CBZ treatment [11-13]. Strolin et

al. [10] reported that increase in serum GGT

was generally observed in 75%–95% of pa-

tients treated chronically with enzyme-

inducing agents. In our study, GGT activities

were higher in all patients treated with CBZ

compared with pre-medication levels. Ac-

cording to Callaghan et al. [14], GGT and

ALP should not be regarded as indicators of

hepatocellular damage in patients taking anti-

convulsant drugs, because increased levels of

these enzymes may reflect enzyme induction

rather than lesions of the cells. We consid-

ered that the increased GGT and ALP activi-


9

ties observed in our study resulted from liver

enzyme induction by CBZ treatment.

Isojärvi et al. [15] has reported that pro-

gressive increase in serum GGT activity dur-

ing the first five years of CBZ medication is

common. They also demonstrated that after

two months of CBZ treatment, significantly

elevated GGT activity was observed com-

pared with that measured before treatment,

which is consistent with our result. However,

they did not report whether GGT activity was

elevated to above normal value. In our study,

GGT activity was elevated in all patients

treated with CBZ, and GGT activity exceeded

the reference values in 33% of the patients

treated with CBZ. Isojärvi et al. [15] also re-

vealed that longer CBZ treatment resulted in

higher GGT activity. Therefore, patients un-

dergoing long-term treatment with CBZ re-

quire follow-up.

Nijhawan et al. [16] reported that among

13 patients on AED therapy (PHT, PB or

CBZ) with increased serum ALP activity, 12

had increased liver ALP isoenzyme activity,

but nine had normal bone ALP isoenzyme

activity. According to the authors, the induc-

tion of liver microsomal enzymes by AEDs

could include liver ALP, but not bone ALP

[16]. However, patients administered CBZ

exhibited significantly elevated ALP levels

accompanied by increased bone and liver iso-

enzyme activities compared with controls

[17]. Thus, in our study, elevated ALP may

occur not only in the liver but also in bone.

Previous studies found no correlation be-

tween dosage of CBZ and changes in bio-

chemical parameters [18], which is consistent

with the present finding. Rehimdel et al. [18]

also revealed that the duration of CBZ treat-

ment correlated with increased ALP activity.

However, in our study, the relation between

the duration of CBZ use and the changes in

biochemical parameters was not evaluated.

As patients with epilepsy often undergo long-

term AED therapy, we should regularly mon-

itor biochemical parameters of these patients

even in those prescribed a low dosage of

AED. In our study, 14 of 16 patients treated

with VPA (88%) demonstrated decreased

PLT counts after VPA therapy was initiated.

However, PLT counts were lowered to

13×104/µL in only one of 16 patients (6%). In

a previous study, thrombocytopenia was pre-

sent (PLT count ≤ 13×104/µL) in 12 of 60

patients (20%) [19]. Another report suggested

that 17.7% of patients experienced thrombo-

cytopenia (PLT count ≤ 10×104/µL) after ex-

posure to divalproex sodium [20]. Mean plas-

ma VPA concentration in this study was 36.3

µg/mL, which was lower compared with the

concentration of 79.6 µg/ml in a previous re-

port [20]. The PLT count may infrequently

decrease to lower than 13×104/µL because of

low VPA plasma concentration. However, a

study reported that prominent hematologic

abnormalities including thrombocytopenia,

macrocytosis, anemia, and leukopenia in-

creased to 55% in 22 patients with VPA lev-

els > 100 µg/mL compared with 33% of the

total number of patients treated with VPA

[19]. Thus plasma VPA concentration should

be carefully monitored particularly in patients

treated with high doses of VPA .

According to the literature, elevated serum

vitamin B12 and no change or elevated folate

concentrations occur after VPA treatment [21

-23]. Hauser et al. [21] reported thrombocyto-

penia, macrocytosis, increased serum vitamin

B12 and a significant downward trend in red

blood cell count after initiation of VPA thera-


10

py, and suggested the possibility that direct

toxic effect on hematopoietic precursor or

stem cell was responsible for these effects.

Acharya and Bussel [24] also reported that

VPA caused bone marrow suppression, re-

sulting in hematologic toxicity. However, in

our study, there was no other hematological

change apart from decreased PLT count, pos-

sibly because of the small number of patients

administered VPA.

In this study, there were no changes in

measured biochemical parameters in patients

after VPA treatment. There are several re-

ports of alterations in liver enzymes in pa-

tients treated with VPA. Cepelak et al. [12]

described increased AST, ALT, and GGT

activities in patients administered VPA com-

pared with healthy children. In vitro VPA

treatment has been demonstrated to induce

enzyme production [25]. In contrast, Hauser

et al. [21] reported no significant upward

trends in mean AST, ALT, GGT, and ALP

levels, but a significant decline in ALT at 3

and 6 months after VPA treatment. In the pre-

sent study, there were no changes in these

levels following VPA treatment, but this

might be due to the relatively small sample

size.

LEV is a drug widely used for treating pa-

tients with epilepsy because of its superior

tolerability and efficacy. Whether LEV caus-

es hematological changes is controversial.

Neutropenia, decreased lymphocyte count, or

thrombocytopenia induced by LEV has been

reported [26-30], while Dinopoulos et al. [26]

reported decreased lymphocyte count with no

decrease in total WBC count. In our study,

there were no significant differences between

pre- and post-treatment values in all the he-

matological parameters examined in patients

administered LEV. A larger study is required

to clarify the effects of LEV on hematologi-

cal parameters.

The major metabolic pathway of LEV

does not depend on the hepatic cytochrome

P450 system, and LEV does not induce he-

patic enzymes [31]. Therefore, LEV does not

influence liver function. In our study, there

were no changes in liver enzymes, consistent

with previous reports.

Conclusions

After CBZ treatment, WBC count de-

creased in 69% of patients while GGT and

ALP activities increased in 85%. Decreased

WBC count tended to correlate with elevated

serum CBZ level. Therefore, our results sug-

gest that hematological changes are dose-

dependently related to AED concentrations.

The high frequencies of elevated GGT and

ALP activities may be caused by induction of

liver microsomal enzymes by AEDs. De-

creased PLT count was observed in 88% of

patients who initiated VPA. Hematological

and biochemical parameters should be care-

fully monitored in patients undergoing AED

treatment, especially in those treated with

high AED doses.

Acknowledgements We thank all the participants and staff in-

volved in this study.

Conflicts of interest The authors declare that they have no con-

flicts of interest.


11

References [1] Glauser T, Ben-Menachem E, Bour-

geois B, Cnaan A, Chadwick D, Guer-

reiro C, Kalviainen R, Mattson R, Pe-

rucca E, Tomson T. ILAE treatment

guidelines: evidence-based analysis of

antiepileptic drug efficacy and effec-

tiveness as initial monotherapy for epi-

leptic seizures and syndromes. Epilep-

sia 2006; 47: 1094– 1120.

[2] Stephen LJ, Brodie MJ. Antiepileptic

drug monotherapy versus polytherapy:

pursuing seizure freedom and tolerabil-

ity in adults. Current Opinion in Neu-

rology 2012; 25: 164- 172.

[3] Killian JM. Tegretol in trigeminal neu-

ralgia with special reference to hemato-

poietic side effects. Headache 1969; 9:

58- 63.

[4] Björnsson E. Hepatotoxicity associated

with antiepileptic drugs. Acta Neurol

Scand 2008; 118: 281- 290.

[5] Japanese Society of Neurology. Guide-

line of diagnosis and treatment for epi-

lepsy 2018; 2018: Japan.

[6] Nihoniyakuhinsyu Forum. Drugs In

Japan: Ethical Drugs 2018. 2017; Ja-

pan.

[7] Bachmann T, Bertheussen KH, Sval-

heim S, Rauchenzauner M, Luef G,

Gjerstad L, Taubøll E, Haematological

side effects of antiepileptic drug treat-

ment in patients with epilepsy. Acta

Neurol. Scand 2011; Suppl. 191: 23–

27.

[8] Huang HL, Zhou H, Wang N, Yu CY.

Effects of antiepileptic drugs on the se-

rum folate and vitamin B12 in various

epileptic patients. Biomed Rep 2016; 5:

413- 416.

[9] Braide SA, Davies TJ. Factors that af-

fect the induction of gamma glutamyl-

transferase in epileptic patients receiv-

ing anticonvulsant drugs. Ann. Clin.

Biochem 1987; 24: 391- 399.

[10] Strolin Benedetti, M, Ruty, B, Baltes,

E. Induction of endogenous pathways

by antiepileptics and clinical implica-

tions. Fundam Clin Pharmacol 2005;

19: 511- 529.

[11] Hoshino M, Heise CO, Puglia P, Al-

meida AB, Cukiert A. Hepatic en-

zyme’s level during chronic use of anti-

convulsant drugs. Arq Neuropsiquiatr

1995; 53: 719- 723.

[12] Cepelak I, Zanić Grubisić T, Mandusić

A, Rekić B, Lenicek J. Valproate and

carbamazepine comedication changes

hepatic enzyme activities in sera of epi-

leptic children. Clin Chim Acta 1998;

276: 121- 127.

[13] Hadzagic-Catibusic F, Hasanbegovic E,

Melunovic M, Zubcevic S, Uzicanin S.

Effects of carbamazepine and valproate

on serum aspartate aminotransferase,

alanine aminotransferase and gamma -

glutamyltransferase in children. Med

Arch 2017; 71: 239- 242.

[14] Callaghan N, Majeed T, O’Connell A,

Oliveria DB. A comparative study of

serum F protein and other liver function

tests as an index of hepatocellular dam-

age in epileptic patients. Acta Neurol

Scand 1994; 89: 237– 241.

[15] Isojärvi JI, Pakarinen AJ, Myllylä VV.

Basic haematological parameters, se-

rum gamma-glutamyl-transferase activ-

ity, and erythrocyte folate and serum

vitamin B12 levels during carbamaze-


12

pine and oxcarbazepine therapy. Sei-

zure 1997; 6: 207- 211.

[16] Nijhawan R, Wierzbicki AS, Tozer R,

Lascelles PT, Patsalos PN. Antiepilep-

tic drugs, hepatic enzyme induction and

raised serum alkaline phosphatase iso-

enzymes. Int J Clin Pharmacol Res

1990; 10: 319– 323.

[17] Voudris K, Moustaki M, Zeis PM,

Dimou S, Vagiakou E, Tsagris B,

Skardoutsou A. Alkaline phosphatase

and its isoenzyme activity for the evalu-

ation of bone metabolism in children

receiving anticonvulsant monotherapy.

Seizure 2002; 11: 377- 380.

[18] Rahimdel A, Dehghan A, Moghadam

MA, Ardekani AM. Relationship be-

tween bone density and biochemical

markers of bone among two groups tak-

ing carbamazepine and sodium

valproate for epilepsy in comparison

with healthy individuals in Yazd. Elec-

tron Physician 2016; 8: 3257- 3265.

[19] May RB, Sunder TR. Hematologic

manifestations of long-term valproate

therapy. Epilepsia 1993; 34: 1098-

1101.

[20] Nasreddine W, Beydoun A. Valproate-

induced thrombocytopenia: A prospec-

tive monotherapy study. Epilepsia

2008; 49: 438– 445.

[21] Hauser E, Seidl R, Freilinger M, Male

C, Herkner K. Hematologic manifesta-

tions and impaired liver synthetic func-

tion during valproate monotherapy.

Brain Dev 1996; 18: 105– 109.

[22] Linnebank M, Moskau S, Semmler A,

Widman G, Stoffel-Wagner B, Weller

M, Elger CE. Antiepileptic drugs inter-

act with folate and vitamin B12 serum

levels. Ann Neurol 2011; 69: 352– 359.

[23] Attilakos A, Papakonstantinou E,

Schulpis K, Voudris K, Katsarou E,

Mastroyianni, S, Garoufi A. Early ef-

fect of sodium valproate and carbamaz-

epine monotherapy on homocysteine

metabolism in children with epilepsy.

Epilepsy Res 2006; 71: 229- 232.

[24] Acharya S, Bussel JB. Hematologic

toxicity of sodium valproate. J Pediatr

Hematol Oncol 2000; 22: 62- 65.

[25] Rogiers V, Callaerts A, Vercruysse A,

Akrawi M, Shephard E, Phillips I. Ef-

fects of valproate on xenobiotic bio-

transformation in rat liver. In vivo and

in vitro experiments. Pharm Weekbl Sci

1992; 14: 127– 131.

[26] Dinopoulos A, Attilakos A, Paschali-

dou M, Tsirouda M, Garoufi A,

Moustaki M, Siafakas N, Papaevange-

lou V. Short-term effect of levetirace-

tam monotherapy on haematological

parameters in children with epilepsy: A

prospective study. Epilepsy Res 2014;

108: 820- 823.

[27] Bunnell K, Pucci F. Levetiracetam-

induced neutropenia following traumat-

ic brain injury. Brain Inj 2015; 29: 115-

117.

[28] Taberner Bonastre MT, Peralta Muñoz

S, Boza FM, Gumà I Padró J. Neutro-

penia secondary to exposure to le-

vetiracetam. Tumori 2015; 101: 145-

146.

[29] Oh SJ, Kwon HI, Moon SH, Ro YS, Ko

JY. Toxic epidermal necrolysis with

isolated neutropenia related to the use

of levetiracetam. J. Dermatol 2016; 43:


13

969-971.

[30] Kimland E, Höjeberg B, von Euler M.

Levetiracetam-induced thrombocytope-

nia. Epilepsia 2004; 45: 877- 878.

[31] Patsalos PN. Pharmacokinetic profile of

levetiracetam: toward ideal characteris-

tics. Pharmacol Ther 2000; 85: 77- 85.


Documents

Effects of antiepileptic monotherapy on hematological and