SEARCH

SEARCH BY CITATION

Keywords:

  • Antiepileptic drugs;
  • Atherosclerosis;
  • Vascular risk factors;
  • Monotherapy;
  • Intima media thickness

Summary

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure
  8. References

Purpose:  Long-term therapy with antiepileptic drugs (AEDs) has been associated with metabolic consequences that lead to an increase in risk of atherosclerosis in patients with epilepsy. We compared the long-term effects of monotherapy using different categories of AEDs on markers of vascular risk and the atherosclerotic process.

Methods:  One hundred sixty adult patients who were receiving AED monotherapy, including two enzyme-inducers (carbamazepine, CBZ; and phenytoin, PHT), an enzyme-inhibitor (valproic acid, VPA), and a noninducer (lamotrigine, LTG) for more than 2 years, and 60 controls were enrolled in this study. All study participants received measurement of common carotid artery (CCA) intima media thickness (IMT) by B-mode ultrasonography to assess the extent of atherosclerosis. Other measurements included body mass index, and serum lipid profile or levels of total homocysteine (tHcy), folate, uric acid, fasting blood sugar, high sensitivity C-reactive protein (hs-CRP), or thiobarbituric acid reactive substances (TBARS).

Key Findings:  Long-term monotherapy with older-generation AEDs, including CBZ, PHT, and VPA, caused significantly increased CCA IMT in patients with epilepsy. After adjustment for the confounding effects of age and gender, the CCA IMT was found to be positively correlated with the duration of AED therapy. Patients with epilepsy who were taking enzyme-inducing AED monotherapy (CBZ, PHT) manifested disturbances of cholesterol, tHcy or folate metabolism, and elevation of the inflammation marker, hs-CRP. On the other hand, patients on enzyme-inhibiting AED monotherapy (VPA) exhibited an increase in the levels of uric acid and tHcy, and elevation of the oxidative marker, TBARS. However, no significant alterations in the markers of vascular risk or CCA IMT were observed in patients who received long-term LTG monotherapy.

Significance:  Patients with epilepsy who were receiving long-term monotherapy with CBZ, PHT, or VPA exhibited altered circulatory markers of vascular risk that may contribute to the acceleration of the atherosclerotic process, which is significantly associated the duration of AED monotherapy. This information offers a guide for the choice of drug in patients with epilepsy who require long-term AED therapy, particularly in aged and high-risk individuals.

Although the prognosis for a majority of patients with epilepsy is good, >30% of patients do not have remission despite appropriate antiepileptic drug (AED) therapy (Kwan & Brodie, 2000). Long-term or lifelong AED therapy is usually required for those patients with refractory epilepsy. Of note is that prolonged AED therapy is often associated with a wide range of chronic adverse effects, including metabolic and endocrine disturbances, behavioral or psychiatric problems, idiosyncratic reactions, negative cognitive effects, and drug interactions (Greenwood, 2000; Aldenkamp & Bodde, 2005; Mintzer, 2010). In particular, growing evidence suggests that the older-generation AEDs that are commonly used for treatment of epilepsy, including phenytoin (PHT), carbamazepine (CBZ), phenobarbital (PB), and valproic acid (VPA), exert prominent effects on the hepatic enzyme system and may alter metabolic pathways that are related to increased vascular risks (Hamed et al., 2007; Mintzer & Mattson, 2009; Tan et al., 2009a; Lopinto-Khoury & Mintzer, 2010).

Heightened atherosclerotic risks may account for the higher mortality and morbidity arising from cerebrovascular disease or atherosclerosis-related heart disease in patients with epilepsy who received prolonged AED therapy (Cockerell et al., 1994; Mohanraj et al., 2006). We and others have identified a number of circulatory biomarkers that are related to risks of cerebrovascular and cardiovascular diseases, including lipid profiles, lipoproteins, C-reactive protein (CRP), total homocysteine (tHcy), and uric acid from patients with epilepsy who are receiving AED therapy (Nikolaos et al., 2004; Sener et al., 2006; Hamed et al., 2007; Mintzer et al., 2009; Tan et al., 2009a; Belcastro et al., 2010; Svalheim et al., 2010; Yuen et al., 2010). In addition, based on measurement of common carotid artery (CCA) intima-media thickness (IMT), a well-established surrogate marker for both stroke and myocardial infarction (Bots et al., 1997), we further demonstrated that age, gender, and duration of AED therapy are important independent factors of CCA IMT (Tan et al., 2009a). Our results implied that cumulative effects of long-term exposure to AEDs play a pivotal role in the pathogenesis of atherosclerosis.

Another important issue is that individual AEDs may have differential effects on the markers of vascular risk. Recent studies (Mintzer et al., 2009; Belcastro et al., 2010) indicated that PHT and CBZ are potent inducers of the cytochrome P450 (CYP450) system, which exerts strong effects on serum lipid profiles, CRP, and tHcy. It follows that those enzyme-inducing drugs may substantially increase the risk of atherosclerosis. However, studies that evaluated the relationship between metabolic consequences and atherosclerotic changes in patients with epilepsy who are under long-term AED monotherapy are wanting, as are different effects of individual classes of AEDs. In our previous study (Tan et al., 2009a), we have shown that CCA IMT in patient with epilepsy appears to be positively correlated with the duration of AED therapy. The present study extended those experiences (Tan et al., 2009a) to evaluate the long-term drug-specific effects of AED monotherapy, including nonenzyme inducers (lamotrigine, LTG), enzyme inducers (CBZ or PHT), or enzyme inhibitors (VPA) on serologic biomarkers and quantified CCA IMT in patients with epilepsy.

Methods

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure
  8. References

Study design

This is a single-center, cross-sectional study. The study hospital, Kaohsiung Chang Gung Memorial Hospital, is a medical center and a main referral hospital that serves an area with 3 million inhabitants in southern Taiwan. The institutional ethics committee approved the study protocol. The intended goal and processes of this study were explained before the study began, with the consent form signed by all subjects.

Subjects

From July 2006 to December 2010, 160 patients (aged between 18 and 65 years) who were seen at the Epilepsy Outpatient Clinic of Kaohsiung Chang Gung Memorial Hospital were enrolled in this study. All patients have been receiving AED monotherapy for more than 2 years. The AEDs used included LTG, CBZ, PHT, and VPA. Sixty healthy volunteers who received annual physical checkups were recruited as controls. Patients who have discontinued medication for more than 2 weeks, or who have combined other AEDs for more than 2 weeks within the two previous years, were excluded from this study. Other exclusion criteria for patient and control participants were based on our previous report (Tan et al., 2009a,b), and included stroke, ischemic heart diseases, hypertension, diabetes mellitus (fasting blood sugar ≥6.93 mm), tobacco use, hematologic diseases, endocrine disorders, or autoimmune diseases. Also excluded were subjects with documented diagnosis of gout or clinical features of arthritis, tophi, or nephrolithiasis related to hyperuricemia. Participants who took vitamin supplement, antihypertensive agents, oral hypoglycemic agents, or who received medication that could affect lipid or uric acid metabolism were also excluded.

All subjects received physical examinations and interviews, and their body mass index (BMI) was calculated. The clinical records of patients were reviewed, and the age of onset, type of seizure, average seizure frequency per month during the previous year, etiology of seizure, and duration of AED therapy were registered. All patients received brain magnetic resonance imaging (MRI) to rule out the confounding high risk for atherosclerosis in participants with symptomatic epilepsy that arises from cerebrovascular diseases.

Assessment of atherosclerosis

Common carotid artery IMT was measured by ultrasonography to assess the extent of atherosclerosis (Tan et al., 2009a,b). Images were obtained from a B-mode ultrasound system (Philips HDI 5000 System; ATL-Philips, Bothell, WA, U.S.A.) equipped with a 4–10 MHz linear array transducer. Measurements were carried out according to standardized protocol by an experienced ultrasound technologist who was blinded to the clinical history of the study participants. We routinely scanned both the left and right CCAs, defined as the 1-cm vascular wall segment of the carotid artery immediately proximal to the dilation of the bifurcation plane. The images were optimized so that only the far wall was visualized in a single longitudinal view. The obtained images were transferred to a workstation, and the IMT, the interface between the lumen intima and media adventitia, was automatically measured using a computer software program (Q-LAB; ATL-Philips). These measurements were again performed in a single-blinded fashion, and mean CCA IMT was defined as the average of measurements obtained from the left and right CCAs.

Analysis of circulating biochemical markers and lipid peroxidation

Blood samples were taken between 8 and 10 a.m. after overnight fasting, and they were analyzed by the central laboratory of Kaohsiung Chang Gung Memorial Hospital for serum levels of triglycerides, total cholesterol, high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), blood sugar, uric acid, folate, tHcy, and high sensitivity CRP (hs-CRP). The blood samples were centrifuged at 1,800 g for 10 min to obtain serum samples, and all circulating biochemical markers were immediately analyzed as reported previously (Tan et al., 2009a,b). Folate level was determined by a SimulTRAC-SNB radioassay kit (MP Biomedicals, Orangeburg, NY, U.S.A.) using an automatic gamma counter (Wizard 1470; PerkinElmer, Boston, MA, U.S.A). A fluorescence polarization immunoassay was used to measure plasma tHcy level, in conjunction with an AxSYM analyzer (Abbott Laboratories, Abbott Park, IL, U.S.A.). In the central laboratory of Kaohsiung Chang Gung Memorial Hospital, intraassay coefficient of variations for folate and tHcy analysis was 4.1% and 4.5%, respectively. Lipid peroxidation as an indicator of oxidative damage was determined by measuring the plasma concentration of thiobarbituric acid reactive substances (TBARS) using the method reported by Ohkawa et al. (1979). A standard curve for TBARS was obtained by hydrolysis of 1,1,3,3-tetraethoxypropane.

Statistical analysis

Patients were divided into four groups according to AED used (LTG, n = 26; CBZ, n = 41; PHT, n = 39; and VPA, n = 54). Statistical analyses were performed using the Statistical Package for Social Science (version 11.0 for Windows; SPSS, Chicago, IL, U.S.A.). To compare demographic data between patients with epilepsy and healthy controls, categorical variables were assessed using chi-square or Fisher’s exact test, and continuous variables were compared using Student’s t-test. Continuous variables among the five subject groups (control, LTG, CBZ, PHT, and VPA groups) were compared using one-way analysis of variance (ANOVA) for parametric data, followed by the Scheffé multiple comparison for post hoc test of significant pairwise differences. Correlation analysis was used to evaluate the relationship between mean CCA IMT and variables in patients with epilepsy, including age, duration of AED therapy, seizure frequency, cholesterol-profile, triglyceride, blood sugar, uric acid, hs-CRP, tHcy, folate, or TBARS concentration. Finally, general linear regression models were performed separately to assess the impact of independent variables on mean CCA IMT between patient group with the control and among the four treatment groups. p < 0.05 was taken to indicate statistical significance.

Results

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure
  8. References

Results of the present study were based on 160 patients with epilepsy who received long-term AED monotherapy (78 female and 82 male) and 60 healthy control subjects (30 female and 30 male).

Clinical characteristics of patients with epilepsy

Table 1 presents the baseline characteristics of the 160 patients with epilepsy. A total of 63.1% of patients had idiopathic or cryptogenic epilepsy, and 36.9% had symptomatic epilepsy that included central nervous system (CNS) infection (n = 4), head injury (n = 27), perinatal brain damage (n = 11), cerebral malformation (n = 14), and neoplasm (n = 3). The age at onset of epilepsy, duration of epilepsy, and duration of AED therapy at the time of study among the four patient groups were listed. The dosage of monotherapy in the LTG group was 100–500 mg/day; CBZ, 400–1,500 mg/day; PHT, 200–400 mg/day; and VPA, 250–2,500 mg/day. Most of the patients (75%) had partial epilepsy and 25% had generalized epilepsy. All four AEDs were used in treatment for partial epilepsy; most of the patients with generalized epilepsy were treated with VPA. A majority of our patients (76.9%) were seizure free for more than 12 months under AED monotherapy.

Table 1.   Demographic data of patients with epilepsy
 LTG (n = 26)CBZ (n = 41)PHT (n = 39)VPA (n = 54)
  1. LTG, lamotrigine; CBZ, carbamazepine; PHT, phenytoin; VPA, valproic acid; AED, antiepileptic drug.

  2. Values expressed as mean ± standard deviation (SD) or amedian (interquartile range, IQR).

  3. bDefined as seizure free for >12 months under AED therapy.

Age (year)33.0 ± 11.634.1 ± 9.537.8 ± 8.933.8 ± 11.2
Gender (female; male)18; 821; 2017; 2222; 32
Age at onset (year)21.4 ± 11.516.1 ± 7.425.1 ± 10.720.3 ± 11.7
Duration of epilepsy (year)8.9 ± 4.618.5 ± 9.612.4 ± 10.013.3 ± 8.9
Duration of AED therapy (year)5.5 ± 3.113.4 ± 6.310.7 ± 8.28.7 ± 5.2
Dosage of AED (mg/day)a300 (188, 400)800 (600, 950)300 (300, 300)1,000 (750, 1,000)
Type of seizures (n)
 Generalized43627
 Partial22383327
Etiology (n)
 Idiopathic/cryptogenic20241641
 Symptomatic6172313
Seizure control (n)
 Seizure freeb19303242
 Non–seizure-free711712

The long-term effect of individual AED on vascular risk factors

The changes of vascular risk factors in the four AED groups and controls are summarized in Table 2. The results of post hoc test were also listed. The difference between age (p = 0.457) or gender (p = 0.106) among the four patient groups and controls was insignificant. The body mass index (BMI) was significantly higher in the PHT and VPA groups when compared with controls, but insignificant in the LTG and CBZ groups. Although the level of HDL-C was not significantly different among the patient groups and controls, CBZ and PHT exerted significant effects on the lipid profile, particularly total cholesterol and LDL-C levels. However, monotherapy with VPA and LTG effected insignificant changes in total cholesterol and LDL-C levels. Significant increase in tHcy and decrease in folate were noted in patients under CBZ, PHT, or VPA therapy. However, LTG group did not exhibit significant changes in tHcy (p = 1.000) and folate (p = 0.983). Patients with long-term VPA therapy manifested significantly elevated levels of uric acid (p < 0.0001), but not hs-CRP levels (p = 1.000). Moreover, the serum level of TBARS was significantly increased by PHT and VPA therapy.

Table 2.   Vascular risk factors and IMT in patients with epilepsy
 Patient groupsControls (n = 60)p-Value
LTG (n = 26)CBZ (n = 41)PHT (n = 39)VPA (n = 54)
  1. Values expressed as mean ± SD.

  2. LTG, lamotrigine; CBZ, carbamazepine; PHT, phenytoin; VPA, valproic acid; C, control; LDL-C, low-density lipoprotein cholesterol; tHcy, total homocysteine; hs-CRP, high sensitivity C-reactive protein; TBARS, thiobarbituric acid reactive substances; CCA, common carotid artery; IMT, intima media thickness.

  3. Results of Scheffe multiple comparisons for post hoc test:

  4. aPHT versus C, p = 0.001; VPA versus C, p = 0.045; PHT versus LTG, p = 0.027.

  5. bVPA versus C, p < 0.0001; VPA versus LTG, p < 0.0001; VPA versus CBZ, p < 0.0001; VPA versus PHT, p < 0.0001.

  6. cCBZ versus C, p < 0.0001; PHT versus C, p < 0.0001; CBZ versus LTG, p = 0.009; PHT versus LTG, p < 0.0001; CBZ versus VPA, p = 0.003; PHT versus VPA, p < 0.0001.

  7. dPHT versus C; P = 0.002; PHT versus LTG, p = 0.022; PHT versus VPA, p = 0.003.

  8. eCBZ versus C, p = 0.006; PHT versus C, p < 0.0001; VPA versus C, p < 0.0001; CBZ versus LTG, p = 0.048; PHT versus LTG, p = 0.006; VPA versus LTG, p = 0.013.

  9. fCBZ versus C, p < 0.0001; PHT versus C, p < 0.0001; VPA versus C, p < 0.0001; CBZ versus LTG, p = 0.042; PHT versus LTG, p = 0.014; VPA versus LTG, p = 0.035.

  10. gCBZ versus C, p = 0.019; PHT versus C, p < 0.0001; PHT versus LTG, p = 0.013; PHT versus CBZ; p = 0.02; PHT versus VPA, p < 0.0001.

  11. hPHT versus C, p = 0.04; VPA versus C, p = 0.032.

  12. iCBZ versus C, p = 0.008; PHT versus C, p < 0.0001; VPA versus C, p = 0.020; PHT versus LTG, p = 0.006.

  13. jPHT versus C, p < 0.0001; VPA versus C, p = 0.029; PHT versus LTG, p = 0.001.

  14. kCBZ versus C, p = 0.017; PHT versus C, p < 0.0001; VPA versus C, p = 0.009; PHT versus LTG, p = 0.001.

Age (year)33.0 ± 11.634.1 ± 9.537.8 ± 8.933.8 ± 11.234.5 ± 10.50.457
Gender (female; male)18; 821; 2017; 2222; 3230; 300.106
Body mass index (kg/m2)22.78 ± 4.6023.70 ± 4.5925.99 ± 5.1224.98 ± 3.5122.64 ± 3.240.001a
Fasting blood sugar (mm)4.75 ± 1.114.94 ± 0.455.14 ± 0.485 ± 0.954.97 ± 0.430.225
Uric acid (μm)296.91 ± 77.35264.78 ± 83.90305.83 ± 65.45384.97 ± 93.42311.19 ± 71.99<0.0001b
Cholesterol (mm)
 Total4.85 ± 0.625.67 ± 0.975.88 ± 0.954.94 ± 0.914.84 ± 0.84<0.0001c
 HDL-C1.55 ± 0.471.64 ± 0.551.61 ± 0.421.46 ± 0.421.55 ± 0.350.355
 LDL-C2.83 ± 0.633.29 ± 0.863.51 ± 0.92.84 ± 0.82.83 ± 0.72<0.0001d
Triglyceride (mm)1.14 ± 0.581.17 ± 0.741.48 ± 0.871.47 ± 0.951.06 ± 0.740.022
tHcy (μm)9.46 ± 2.7013.31 ± 8.4114.50 ± 5.6013.84 ± 4.299.41 ± 2.65<0.0001e
Folate (nm)23.13 ± 11.4415.25 ± 9.1014.12 ± 9.1015.46 ± 7.0124.58 ± 12.14<0.0001f
hs-CRP (mg/L)1.39 ± 1.263.66 ± 5.094.94 ± 6.791.26 ± 1.660.96 ± 1.04<0.0001g
TBARS (μm)0.83 ± 0.451.00 ± 0.511.13 ± 0.501.11 ± 0.480.84 ± 0.410.003h
CCA IMT (mm)
 Right CCA IMT0.521 ± 0.0600.586 ± 0.1030.643 ± 0.2130.571 ± 0.1290.490 ± 0.064<0.0001i
 Left CCA IMT0.514 ± 0.0580.574 ± 0.0700.635 ± 0.1310.587 ± 0.1500.519 ± 0.081<0.0001j
 Mean CCA IMT0.518 ± 0.0530.580 ± 0.0770.639 ± 0.1510.571 ± 0.1290.505 ± 0.066<0.0001k

The long-term effect of individual AED on CCA IMT

Our previous study revealed that the CCA IMT on either side or the mean CCA IMT was significantly increased in patients with long-term AED therapy (Tan et al., 2009a). We found in the present study that the mean CCA IMT was also significantly increased in monotherapy with CBZ, PHT, or VPA (Table 2); changes in the LTG monotherapy group were not statistically significant (p = 0.992). In addition, the mean CCA IMT in PHT monotherapy group was significantly increased when compared with the LTG group (p < 0.0001). Importantly, multiple linear regression analysis between the control and treatment groups revealed that PHT, VPA, and CBZ treatment (Table 3, model 1), in descending order of efficacy, exerted detrimental effects on average CCA IMT thickness. The LTG group, in contrast, did not reach statistical significance. Subgroup analysis between three treatment groups and LTG (Table 3, model 2) on the effect of average CCA IMT thickness showed that PHT group still reached statistical significance, whereas the CBZ or VPA group was not different from the LTG group.

Table 3.   Multiple regression analysis of the effects of AED monotherapy and duration of AED therapy on mean CCA IMT
 Model 1 (n = 220)Model 2 (n = 160)
BSEp-ValueBSEp-Value
  1. Dependent variable: mean CCA IMT. Model 1 included all patients and controls (n = 220), and model 2 included the patients only (n = 160).

  2. AED, antiepileptic drug; LTG, lamotrigine; CBZ, carbamazepine; PHT, phenytoin; VPA, valproic acid; CCA IMT, common carotid artery intimal medial thickness.

  3. aThe duration of AED therapy is defined as 0 in control group.

AEDs
 CBZ versus control0.04970.02430.0419   
 PHT versus control0.09470.0222<0.0001   
 VPA versus control0.05530.01970.0054   
 LTG versus control0.01690.02190.4398   
  CBZ versus LTG   0.03250.02700.2292
  PHT versus LTG   0.07490.02610.0047
  VPA versus LTG   0.03590.02420.14
AED durationa0.00720.0012<0.00010.00730.0014<0.0001
Age0.00510.0006<0.00010.00540.0008<0.0001
Gender (male)0.04220.01200.00050.05270.01590.0011
Intercept0.3420.024<0.00010.3350.029<0.0001

Correlation analysis of the effects of candidate vascular risk factors on CCA IMT in patients with epilepsy

Our previous report showed that age, gender, and duration of AED therapy are important independent factors on CCA IMT (Tan et al., 2009a). Results from correlation analysis in this study again demonstrated that age (p < 0.0001) and duration of AED therapy (p < 0.0001) were positively correlated with CCA IMT. Furthermore, frequency of seizures (p = 0.049), BMI (p = 0.007), and levels of uric acid (p = 0.011), fasting blood sugar (p = 0.004), total cholesterol (p = 0.007), LDL-C (p = 0.005), and TBARS (p = 0.008) were significantly correlated with the mean CCA IMT. On the other hand, tHcy (p = 0.686), folate (p = 0.560), hs-CRP (p = 0.067), and triglyceride (p = 0.051) showed insignificant correlation.

We further employed multiple linear regression analysis to identify from the aforementioned significant variables the crucial determining factors that underlie the augmented CCA IMT in patients with epilepsy. The results showed that duration of AED therapy, age, and gender were significant determinants in CCA IMT thickness (Table 3). Based on multiple regression analysis on 160 patients (Table 3, model 2), a multiple linear regression formula obtained was: mean CCA IMT = 0.335 + 0.0054 × (age) 0.073 × (duration of AED therapy) + 0.0527 (male) + effects of AED monotherapy, which was similar to that reported in our previous study (Tan et al., 2009a). Note that in this formula, the effects of AED monotherapy were denoted by the regression coefficients.

Duration of PHT, CBZ, and VPA monotherapy is significantly associated with increased CCA IMT

To exclude the important confounding effects of age and gender, we also adjusted CCA IMT to equal age and gender. Linear regression models on the mean CCA IMT after this adjustment revealed positive correlation with AED duration in PHT (r = 0.362, p = 0.024), CBZ (r = 0.415, p = 0.007), and VPA (r = 0.368, p = 0.006) groups; the LTG (r = 0.324, p = 0.107) group was not statistically significant (Fig. 1).

image

Figure 1.   Linear regression models on mean common carotid artery (CCA) intima media thickness (IMT) after adjustment for age and gender revealed positive association with AED duration in individual AED treatment groups: (A) carbamazepine (CBZ) group, (B) phenytoin (PHT) group, (C) valproic acid (VPA) group, and (D) lamotrigine (LTG) group.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure
  8. References

The present study provided evaluations on vascular risk factors in patients with epilepsy who are under AED monotherapy. We also validated the advantage of direct quantification of atherosclerotic changes in vessel wall by B-mode ultrasound system in those patients. Based on CCA IMT measurements, we demonstrated that the duration of monotherapy with CBZ, PHT, or VPA is significantly associated with acceleration of atherosclerosis in patients with epilepsy, albeit via different underlying mechanisms.

Dyslipidemia has long been known to be an important risk factor for atherosclerosis (Kullo & Ballantyne, 2005). LDL-C plays an important role in the atherosclerotic process by increasing endothelial permeability, retention of lipoproteins within the intima of blood vessels, recruitment of inflammatory cells, and formation of foam cells (Stocker & Keaney, 2004; Kullo & Ballantyne, 2005). Emerging evidence further showed that treatment with enzyme-inducing AEDs, such as CBZ and PHT, is significantly associated with increased blood levels of total cholesterol, atherogenic (non-HDL) cholesterol, triglycerides, and tHcy (Mintzer et al., 2009; Belcastro et al., 2010; Svalheim et al., 2010). It is, therefore, of interest that our present results showed that the increase in thickness of CCA IMT in monotherapy with PHT or CBZ may be related to total cholesterol and LDL-C.

The effects of VPA on changes in lipid profiles and lipoproteins remains controversial (Eirís et al., 1995; Geda et al., 2002; Pylvänen et al., 2006; Lopinto-Khoury & Mintzer, 2010). Other studies (Mintzer et al., 2009; Svalheim et al., 2010) showed that newer AEDs, including LTG, do not affect blood lipid profiles. In addition, switching patients from enzyme-inducing (CBZ or PHT) to noninducing (LTG) AEDs resulted in amelioration of serologic markers of vascular risk, including lipids, tHcy, and CRP (Mintzer et al., 2009). By demonstrating that changes in the levels of total cholesterol and LDL-C were insignificant in the VPA and LTG monotherapy groups, the present study revealed that VPA and LTG exert minimal effects on blood lipids in patients with epilepsy who are on chronic therapy.

Another vascular marker of interest is tHcy, a nonessential amino acid with prothrombotic properties. High serum level of tHcy was found to be associated with increased mortality from ischemic stroke, coronary heart disease, and other cardiovascular disease (Cui et al., 2008). An increase in tHcy concentration may promote overproduction of reactive oxygen species, enhancement of platelet aggregation, inhibition of protein C, activation of nuclear factor-κB, and increase in the release of inflammatory mediators, all of which further create a prothrombotic environment (Temple et al., 2000; Kullo & Ballantyne, 2005). Chronic AED therapy may lead to hyperhomocysteinemia and lowered level of folate (Tamura et al., 2000; Verrotti et al., 2000; Tan et al., 2009a). Recent studies (Mintzer et al., 2009; Belcastro et al., 2010) showed significantly increased tHcy levels in patients who received enzyme-inducing AEDs, although tHcy level remained stable in those who received the enzyme-inhibiting VPA (Belcastro et al., 2010). In partial agreement with those findings, we observed that monotherapy with PHT, CBZ, and VPA significantly increased tHcy level and decreased folate level in our patients, whereas LTG was devoid of both effects. One possible explanation for the disruption of tHcy and folate metabolism in the VPA group may be related to its long-term use (13.3 ± 8.9 years) in our patients. Folate is a cofactor in methylation of tHcy, the primary means for regulating tHcy concentration (Temple et al., 2000). The deficiency in folate in patients under long-term VPA may also contribute to hyperhomocysteinemia. Whereas further studies are needed to substantiate this explanation, we noted that reports on the association between VPA and disruption of tHcy metabolism are conflicting (Gidal et al., 2005; Attilakos et al., 2006a; Belcastro et al., 2010).

Hyperuricemia has been reported to be related to cardiovascular and cerebrovascular morbidity and mortality (Bos et al., 2006; Chen et al., 2009). Higher levels of serum uric acid may be associated with the development of atherosclerosis that is independent of other atherosclerotic risk factors (Chen et al., 2009). Uric acid is the major end-product of purine metabolism (Rao et al., 1991). Hyperuricemia may increase superoxide production, stimulate vascular smooth cell proliferation, and upregulate the expression of platelet-derived growth factor or monocyte chemoattractant protein-1, leading to damages of endothelial cells that cause atherosclerosis (Rao et al., 1991; Bos et al., 2006). Although the effect of AEDs on serum uric acid concentrations is controversial (Ring et al., 1991; Attilakos et al., 2006b; Aycicek & Iscan, 2007), our results showed that uric acid level was significantly higher in patients with VPA monotherapy. On the other hand, PHT, CBZ, or LTG did not elicit significant effect on the metabolism of uric acid. As such, long-term therapy with VPA in patients with epilepsy may increase the levels of uric acid that may contribute to atherosclerotic risk.

Long-term AED therapy may result in low-grade systemic inflammation and increase in oxidative stress, as manifested by higher concentrations of hs-CRP and TBARS (Mintzer et al., 2009; Tan et al., 2009a). Chronic production of reactive oxygen species may exceed the capacity of cellular antioxidants, resulting in oxidative modification of LDL-C, promotion of proinflammatory responses, recruitment of macrophage, and development of atherosclerotic lesion (Stocker & Keaney, 2004). Moreover, hs-CRP has been found to induce the expression of cytokines and cell adhesion molecules, which are recognized as activators of the extrinsic pathway of the coagulation system (Ridker, 1998; Kullo & Ballantyne, 2005). Whereas the hs-CRP level was significantly increased in the PHT or CBZ treatment group in our study, the plasma concentration of TBARS remained relatively stable. VPA induced opposite effects, with a significant increase in TBARS and insignificant change in hs-CRP level. Otherwise, long-term monotherapy with LTG has no effects on hs-CRP and TBARS. Therefore, the risk of atherosclerosis in CBZ and PHT groups may be related to inflammatory mechanisms and the VPA group may be associated with oxidative mechanisms.

Prospective and retrospective incidence cohort studies have established that patients with epilepsy carry a significantly higher mortality rate than the general population (Lhatoo et al., 2001; Jallon, 2004; Mohanraj et al., 2006). Whereas cardiovascular disease is not considered a contributing factor (Lhatoo et al., 2001; Mohanraj et al., 2006), a number of studies reported an elevated standardized mortality ratio for patients with epilepsy to die of cerebrovascular diseases that are related to atherosclerosis (Cockerell et al., 1994; Jallon, 2004; Mohanraj et al., 2006). However, few data exist regarding the long-term effects of specific AEDs on vascular events. Our previous study revealed that the IMT of CCA was significantly increased in patients with long-term AED therapy (Tan et al., 2009a). An important observation in the present study is the long-term effect of AED monotherapy on the degree of increment in CCA IMT. Based on multiple linear regression analysis, we also revealed that long-term treatment with PHT, VPA, and CBZ had detrimental effects on average CCA IMT thickness, which was not shown for the LTG monotherapy group. In addition, the more detrimental effect on increment of CCA IMT was noted in the PHT monotherapy group.

The duration of AED therapy is significantly associated with the acceleration of atherosclerosis in patients with epilepsy, alongside independent contributions of age and gender to the atherosclerotic process (Tan et al., 2009a). In the present study, our linear regression models on mean CCA IMT after adjustment for age and gender also revealed positive correlation with AED duration in the PHT, CBZ, and VPA treatment groups, whereas the LTG group was not significant. Therefore, it is reasonable to suggest that the duration of monotherapy with the older-generation AEDs—including CBZ, PHT, and VPA—is at least one of the important and contributing risk factors to the atherosclerotic process. We are aware that the duration of evaluation for our LTG group was shorter than that for the other groups because LTG is a newer AED. As such, it is plausible that the shorter duration of LTG monotherapy may be a contributing factor for its lack of effect on CCA IMT. Although we are unable to rule out this factor, we reason that because this group of patients did not manifest significant changes in the markers for vascular risk, the possibility that LTG contributes to atherosclerosis by damaging vascular endothelial cells is deemed minimal. However, further studies are needed to evaluate more long-term effects of the newer-generation AEDs, including LTG on CCA IMT.

The present study provided insights into AED monotherapy by showing that prolonged use of CBZ, PHT, or VPA significantly changed serologic biomarkers of vascular risk that are associated with accelerated progression of atherosclerosis in patients with epilepsy. However, our results demonstrated that the mechanisms that underlie the progression of atherosclerosis in patients under long-term monotherapy with different AEDs might be different. The augmented CCA IMT observed in patients under long-term therapy with the CYP450 enzyme inducers, CBZ and PHT may be related to disturbances of cholesterol, tHcy, and folate metabolism, alongside increased inflammation. However, an elevation in the levels of uric acid and tHcy, together with oxidative stress, may contribute to atherosclerotic risk in patients under long-term therapy with VPA. On the other hand, patients under long-term monotherapy with the nonenzyme-inducer, LTG did not manifest significant changes in markers for vascular risk or augmented CCA IMT. Our results also suggest that long-term use of older-generation AEDs with prominent effects on the enzyme system, including CBZ, PHT, and VPA, may contribute to the progression of atherosclerosis in patients with epilepsy. To minimize metabolic disturbances and vascular risks and to reduce atherosclerosis-related diseases caused by long-term AED therapy, newer-generation AEDs such as LTG that exert no or weak effect on the enzyme system might be the most favorable choice. This information offers a guide for the choice of drug for patients with epilepsy who require long-term AED therapy, particularly in aged and high-risk individuals.

Acknowledgments

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure
  8. References

We thank the Morris-Coole Trust and the International League Against Epilepsy for awarding the Morris-Coole Prize 2010 to YCC for our previous work that forms the foundation of the present study.

Disclosure

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure
  8. References

The authors have no conflicts of interest to declare in relation to this study. We confirm that we have read the Journal’s position on issues involved in ethical publication and affirm that this report is consistent with those guidelines

References

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure
  8. References