• Antiepileptic drugs;
  • Cognitive function;
  • Lamotrigine;
  • Topiramate;
  • Oxcarbazepine


  1. Top of page
  2. Abstract

Summary:  Although the causes of cognitive impairment in patients with epilepsy have not been completely elucidated, three factors are clearly involved: the underlying etiology of epilepsy, the effects of seizures themselves, and the central nervous system effects of antiepileptic drugs (AEDs). All commonly used AEDS have some effect on cognitive function, and the effect may be substantial when crucial functions are involved, such as learning in children or driving ability in adults, or when already-vulnerable functions are involved, such as memory in elderly patients. The available evidence is insufficient to support definite conclusions about the cognitive effects of three of the newer AEDs, tiagabine, gabapentin, and levetiracetam. Better evidence is available for lamotrigine (LTG), topiramate (TPM), and, to a lesser degree, oxcarbazepine (OXC). OXC appears not to affect cognitive function in healthy volunteers or adults with newly diagnosed epilepsy, but its cognitive effects in children and adolescents have not been systematically studied. A relatively large number of studies are available for LTG, which has demonstrated a favorable cognitive profile overall, both in volunteers and in patients with epilepsy. Although dose and titration speed may be confounding factors in some of the studies of TPM, there is clear evidence that this agent does affect cognitive function, with specific effects on attention and verbal function. For LTG, attempts have been made to correlate cognitive effects with what is known of the drug's mechanism of action; this is an area of research that deserves further exploration with regard to other AEDs as well, especially TPM.

Cognitive impairment is a frequently occurring secondary consequence of epilepsy (1,2).Cognitive function is higher-order behavior involving the capacity of the brain—specifically of the cortical structures—to program adaptive behavior, to solve problems, to memorize information, and to focus attention (3). Epilepsy is a result of ictal and interictal cortical dysfunction, and the possibility that cognitive impairment develops as a secondary symptom is thus obvious. Memory impairments, mental slowing, and attentional deficits are the most frequently reported cognitive disorders (4,5). Sometimes, such consequences are more debilitating for the individual patient than the seizures; thus, it is worthwhile to explore the factors that lead to cognitive impairment. The exact cause of cognitive impairment in epilepsy has not been explored fully, but three factors clearly are involved: etiology, the seizures, and the “central” side effects of drug treatment (6). Here we concentrate on the unwanted effects of antiepileptic medication on cognitive function. When evaluating this factor separately, it is imperative to realize that in clinical practice most cognitive problems have a multifactorial origin and that, for the most part, the three aforementioned factors, combined, are responsible for the “makeup” of a cognitive problem in an individual patient. Moreover, the factors are related, which causes therapeutic dilemmas in some patients when seizure control can only be achieved with treatments that are associated with cognitive side effects.


  1. Top of page
  2. Abstract

The interest in the cognitive side effects of antiepileptic drug (AED) treatment is of recent origin. The first studies are from the early 1970s (7,8) and were probably stimulated by the widening range of possibilities for drug treatment during that period [i.e., the introduction of carbamazepine (CBZ) and valproate (VPA)]. Since then, a plethora of studies have been published, the majority on the commonly used AEDs: VPA, CBZ, and phenytoin (PHT). These studies have been evaluated in a meta-analysis (9) that resulted in the following conclusions:

  • • 
    Polypharmacy shows a relatively severe impact on cognitive function when compared with monotherapy, irrespective of the type of AEDs included. Two drugs that individually have mild cognitive effects may induce serious cognitive impairment when used together, possibly because of potentiation of tolerability problems (10).
  • • 
    All established AEDs have “absolute” cognitive side effects (i.e., all of the investigated drugs have cognitive effects when compared with no treatment in the same subjects). These effects are larger for phenobarbital (PB) and PHT than for CBZ or VPA. But even the latter drugs, which are generally considered to be drugs with a safe cognitive profile, have cognitive effects, mostly resulting in a mild, general psychomotor slowing (11).
  • • 
    The differences between the three most investigated AEDs (PHT, CBZ, and VPA) can be considered relatively small when studied within a normal therapeutic dose. An exception is PB, which has been shown to have a dramatic impact on cognition (12).

Possibly the most remarkable finding is that, although the severity of cognitive side effects is generally considered to be mild to moderate for most AEDs (9), all commonly used AEDs have some impact on cognitive function. Such mild impact may be amplified in specific conditions and may become substantial in some patients when crucial functions are involved, such as learning in children (5) or driving capacities in adults (often requiring millisecond precision), or when functions are impaired that are already vulnerable, such as memory function in the elderly (10). Moreover, the cognitive side effects represent the long-term outcome of AED therapy; therefore, the effects may increase with prolonged therapy, which contributes to the impact on daily life functioning in refractory epilepsies (13). It is therefore worthwhile to evaluate the cognitive effects of the newer drugs, as this may provide us with new possibilities in clinical practice to treat patients without inducing cognitive impairment.


  1. Top of page
  2. Abstract

In the last decade, several new AEDs have been introduced (14). Although it is claimed that these drugs have different efficacy profiles (15) and that some drugs are particularly efficacious in specific syndromes [e.g., vigabatrin (VGB)], head-to-head comparisons between the new AEDs and between the newer drugs and the commonly used drugs (such as CBZ and VPA) are rare. Nonetheless, meta-analyses such as the influential Cochrane reviews (16,17) do not show significant differences in efficacy between the newer drugs or between newer and commonly used drugs. Also, studies analyzing long-term retention do not show differences between the drugs (18,19). Several studies have shown retention rate to be the best parameter of the long-term clinical usefulness of a particular drug (20). Retention rate is considered to be a composite of drug efficacy and drug safety and expresses the willingness of patients to continue drug treatment. It is therefore the best standard for evaluating the clinical relevance of side effects. The 1-year retention rate is reported to be not higher than 55% for topiramate (TPM) (21), 60% for lamotrigine (LTG), 58% for VGB, and 45% for gabapentin (GBP) (22). Long-term (mostly 3-year) retention is ∼35% for all newer AEDs (23). Side effects appear to be the major factor affecting long-term retention for most drugs (24). In clinical practice, tolerability is a major issue and the choice of a certain AED is at least partially based on comparison of tolerability profiles of the drugs. Also, the tolerability profiles of the newer drugs have become a more important issue in drug development, stimulated by the interest of regulatory agencies (25). Cognitive side effects are particularly important tolerability problems in chronic AED treatment.


  1. Top of page
  2. Abstract

In evaluating studies of the cognitive effects of the newer AEDs, we will follow an evidence-based approach, as has been done for the most commonly used drugs (9). Randomized clinical trials with monotherapy in patients with newly diagnosed epilepsy represent the most accurate procedure for assessing the cognitive impact of AEDs (25). These studies are not clouded by the effect of concurrent or previous AED use and permit the accurate collection of nondrug baseline data that is required for determining whether a particular treatment affects cognitive processing (i.e., to isolate drug-induced impairments from those due to other sources such as the seizures). Data from such studies can be supplemented with information from studies using add-on or polytherapy designs. In these studies, the use of two AEDs makes identifying the components of the treatment that are responsible for the observed effects more complex. However, in many cases patients with epilepsy require dual AED therapy before adequate seizure control is obtained; therefore, data from add-on studies does warrant consideration. Also, data from healthy volunteers should be treated with caution. In general, the power of such studies is limited by small sample sizes, and drug-exposure periods are typically brief. It is possible that chronic treatment results in entirely different types of cognitive impairment that cannot be observed during short-term treatment. For example, such differences in side-effect profile between acute and long-term administration have been found with PHT (26). Finally, the differing cerebral substrate in patients with epilepsy and healthy volunteers suggests that cognitive responses to AEDs may be different in these populations. Nonetheless, volunteer studies may provide an early insight into the cognitive effects of an AED and therefore provide a foundation for further studies in patients with epilepsy (see reference 9 for a discussion of methodological aspects of cognitive drug trials in epilepsy).


  1. Top of page
  2. Abstract

We will analyze studies for the following newer AEDs: oxcarbazepine (OXC), LTG, TPM, tiagabine (TGB), GBP, and levetiracetam (LEV).


OXC is a keto homologue of CBZ with a completely different metabolic profile. In humans, the keto group is rapidly and quantitatively reduced to form a monohydroxy derivative that is the main active agent during OXC therapy. Oxcarbazepine was approved in the European Union in 1999 and is indicated for use as monotherapy or adjunctive therapy for partial seizures with or without secondarily generalized tonic-clonic seizures in patients ≥6 years of age.

The effects of OXC on cognitive function have been evaluated in one study in healthy volunteers and in four studies in patients with epilepsy. A double-blind, placebo-controlled, crossover study was conducted in 12 healthy volunteers (27). The effects of two doses of OXC (300 and 600 mg/day) and placebo on cognitive function and psychomotor performance were assessed. The treatment duration for each condition was 2 weeks. Cognitive function tests were administered before treatment initiation and 4 h after the morning doses on days 1, 8, and 15. In this study, OXC improved performance on a focused attention task, increased manual writing speed, and had no effect on long-term memory processes.

In patients with epilepsy, four monotherapy comparative studies evaluated the effects of OXC on cognitive functions in adult patients with newly diagnosed epilepsy (28–31). The first study was a double-blind, active-control study evaluating the effects of CBZ and OXC on memory and attention in 41 patients with newly diagnosed epilepsy (28). The treatment duration was 1 year. Cognitive function and intelligence tests were administered before treatment initiation and after 1 year of treatment. The results indicated no deterioration of memory or attention with either CBZ or OXC. The second study was an active-control study that evaluated the effects of CBZ, VPA, and OXC on intelligence, learning, and memory, attention, psychomotor speed, verbal span, and visuospatial construction in 32 patients with newly diagnosed epilepsy (29). The treatment duration was 4 months. Cognitive function and intelligence tests were administered before treatment initiation and after 4 months of treatment. The results indicated no deterioration of cognitive function in any treatment group. Significant improvements in learning and memory tests were found for the CBZ- and OXC-treated patients. Improvements were also found in attention and psychomotor speed tests for the VPA-treated patients and partly for the CBZ-treated patients. The third study was a double-blind, randomized, active-control study that evaluated the effects of PHT and OXC on memory, attention, and psychomotor speed in 29 patients with newly diagnosed epilepsy (30). The treatment duration was 1 year. Cognitive function tests were administered before treatment initiation and after 6 and 12 months of treatment. The results indicated no significant differential cognitive effects between PHT and OXC during the first year of treatment in patients with newly diagnosed epilepsy who achieved adequate seizure control. In the fourth study (31), three groups of 12 patients taking either CBZ, VPA, or PHT took a single 600-mg dose of OXC, followed 7 days later by 3 weeks of treatment with OXC 300 mg three times daily and matched placebo in random order. Seven untreated patients, acting as controls, were prescribed the single OXC dose and 3 weeks of active treatment only. There were no important changes in cognitive function test results during administration of OXC compared with placebo.

In summary, the results of these studies indicate that OXC does not affect cognitive function in healthy volunteers and adult patients with newly diagnosed epilepsy. However, the effects of OXC on cognitive function have not been systematically studied in children and adolescents. In accordance with the latest revision of the Committee for Proprietary Medicinal Products (CPMP) Note of Guidance (CPMP EWP/566/98 rev 1, dated November 16, 2000, Sections 2.5 and 5.2), a study has recently been launched (Protocol #: CTRI476E2337) to investigate the effects of OXC on cognitive function (i.e., psychomotor speed and alertness, mental information processing speed and attention, memory, and learning) in children and adolescents aged 6 to <17 years with partial seizures.


TPM is a sulfamate-substituted monosaccharide that has multiple mechanisms of action (32). TPM has proved to be effective in patients with refractory chronic partial epilepsies (15,33).

During the initial add-on clinical trials, central nervous system (CNS)–related “cognitive” subjective complaints were frequently reported, including mental slowing, attentional deficits, speech problems, and memory difficulties (15). It should be mentioned, however, that higher target doses and faster titration schedules were used than are now used in clinical practice (see references 34 and 35 for a discussion of dose and titration speed). Recent studies with TPM-treated patients have confirmed high levels of adverse cognitive effects based on subjective complaints (36,37). A recent follow-up study (Bootsma HPR, Coolen F, Aldenkamp AP, et al., unpublished data) showed long-term retention of 30% for a 4-year follow-up. For about half of the 70% of patients who discontinued treatment, side effects were the major reason, with cognitive side effects being most frequently mentioned. Only a few studies have psychometrically measured cognitive changes using neuropsychological tests.

A study by Martin et al. in six normal volunteers (38)used an acute dose of 2.8 mg/kg (∼200 mg/day) followed by a titration to 5.7 mg/kg (∼400 mg/day) in 4 weeks, resulting in weekly dose escalations of ∼100 mg. The rate at which TPM was escalated in this study was very similar to the dose escalation used in the initial TPM adjunctive-therapy trials, in which escalating the TPM dose to 200 or 400 mg/day over 2–3 weeks was associated with somnolence, psychomotor slowing, speech disorders, and concentration and memory difficulties (15,37; Bootsma HPR, Coolen F, Aldenkamp AP, et al., unpublished data). Martin et al. showed neuropsychometric changes commensurate with these CNS effects. The cognitive effects of the acute starting dose of 200 mg/day were impairments of verbal function (word finding and verbal fluency) of ∼2 standard deviations (which represents very serious impairment) and of sustained attention. Titration to 400 mg/day in 4 weeks resulted in impairments of verbal memory and mental speed of >2 standard deviations.

Four studies involving patients with epilepsy are available. In a study by Meador (39) with 155 patients with epilepsy, the effects of the gradual introduction of TPM as add-on (a 50-mg starting dose, followed by increments of 50 mg/week over 8 weeks) were compared with those of more rapid dose escalation (initial dose of 100 mg, followed by two consecutive weekly increments of 100 and 200 mg). In a test battery of 23 variables representing selective attention, word fluency, and visuomotor speed, the subjects who were on a slow-titration schedule and treated with one background AED displayed TPM-associated score changes of more than one third but <1 standard deviation (39). A study by Aldenkamp et al. (35) was specifically designed to compare cognitive effects of TPM and VPA added to therapeutic dosages of CBZ in 59 patients with epilepsy. In this study, a slow titration speed was used with a starting dose of 25 mg/day TPM and weekly increments of 25 mg. Moreover, the average achieved dose (∼250 mg) was relatively low. Neuropsychometric testing was conducted 8 weeks after the last dosage increase (20 weeks after the start of TPM therapy). The study therefore used optimal conditions (i.e., slow titration, relatively low dose, and a longer treatment period), allowing for patient habituation to the effects of TPM therapy. Nonetheless, cognitive impairment was found for verbal memory function both during titration and at end point. In a study by Burton and Harden (40), attention was assessed weekly in 10 subjects receiving TPM over a 3-month period. Four of nine subjects showed significant correlations between TPM dosage and forward digit span measured weekly, such that higher dosage was associated with poorer attention. In a retrospective study by Thompson et al. (41), the neuropsychological test scores of 18 patients obtained before and after the introduction of treatment with TPM (median dose 300 mg) were compared with changes in test performance of 18 patients who had undergone repeat neuropsychological assessments at the same time intervals. In those patients taking TPM, a significant deterioration in many domains was found. The largest changes were for verbal IQ, verbal fluency, and verbal learning.

In summary, there is clear clinical evidence for TPM-induced cognitive impairment. Not all studies are comparable because of the confusion about dose and titration speed (see reference 34 for a discussion). Moreover, the complete lack of controlled studies is remarkable.


LTG blocks voltage-dependent sodium channels, thereby preventing excitatory neurotransmitter release. Clinical evidence indicates that LTG is effective against partial and secondarily generalized tonic-clonic seizures, as well as idiopathic (primary) generalized epilepsy. LTG was introduced in Europe in 1991 and in the United States in 1994.

A large number of cognitive studies are available for LTG (see reference 42 for an overview). Five volunteer studies have been conducted with LTG. Doses of 120 and 240 mg did not produce a significant change in cognitive function compared with baseline when administered to 12 normal volunteers in an acute study of 1 day (43). Similarly, five volunteers received LTG (acute dose 3.5 mg/kg and then titrated to a maximum of 7.1 mg/kg) in a single-blind manner and were assessed for change in cognitive function after 2 and 4 weeks (38). There was no significant change in any of the neurocognitive measures relative to baseline performance. LTG and CBZ have been compared in 12 healthy male volunteers and associations were made between the observed cognitive effects and plasma concentrations of these drugs (44). The effects of these drugs were examined using adaptive tracking, which assesses eye–hand coordination and effects of attention, and eye movement tests. LTG treatment was not significantly different from placebo, but increased CBZ saliva concentrations were significantly associated with impaired adaptive tracking and smooth and saccadic eye movements. The long-term effects of LTG and CBZ were compared in 23 volunteers in a 10-week crossover study (45). The neuropsychological battery in this study consisted of 19 instruments yielding 40 variables, including both subjective and objective measures. LTG showed better performance or fewer side effects in 17 (42%) of the variables, while no statistically significant differences were seen in the remaining variables. Finally, a study by Aldenkamp et al. (46) in 30 volunteers (12 days of treatment, using a daily dose of 50 mg of LTG) showed evidence for a selective positive effect of LTG on cognitive activation, relative to both placebo and VPA. Although the results of these volunteer studies provide us with preliminary insight into the impact of LTG on cognition, the generalizability of the results from these studies to patients with epilepsy receiving long-term AED treatment is limited.

The effects of LTG on cognitive function have been compared with those of CBZ in patients with newly diagnosed epilepsy. Patients completed tests of verbal learning and memory, attention, and mental flexibility at baseline and then periodically for up to 48 weeks. Significant differences favoring LTG over CBZ were observed with semantic processing, verbal learning, and attention. Brodie et al. (47) concluded that LTG may have a favorable long-term effect on cognitive function when compared with CBZ. Other studies have reported positive cognitive effects of LTG used as adjunctive therapy. Two independent double-blind, randomized, crossover studies have examined the cognitive effects of LTG used as add-on therapy (48,49). Both studies included patients with a history of partial seizures (at least once weekly during the preceding 3 months) who had received no more than two other AEDs or VPA monotherapy. Both studies also used two treatment periods (12 and 18 weeks), which were separated by a washout period (4 and 6 weeks). Despite the similarity in trial design and patients, there is some inconsistency between the findings of these two studies. One study showed a marginal reduction in general “cerebral efficiency” (an indirect measure of cognitive function) after LTG treatment (49). Conversely, significant improvements were reported in the second study (48). In an uncontrolled add-on study (50) using CBZ as baseline drug, no deterioration on any of the cognitive tests was found after introducing LTG (200 mg). LTG therapy in seven patients with epilepsy and mental retardation caused both positive and negative psychotropic effects (51). These findings were based on the observations of parents and supervising staff. Positive effects included reduced irritability and increased compliance with simple instructions, while negative effects included behavioral deterioration with temper tantrums, restlessness, and hyperactivity. Similarly, a second study in 67 patients with mental retardation showed that after adjunctive treatment with LTG, social functioning was stable or improved in 90% of patients (52).

In addition to clinical studies that have assessed the impact of LTG on cognitive function, further evidence can be obtained from examining the effect of LTG on electroencephalographic (EEG) parameters. Overt EEG discharges can occur without any visible clinical correlate in many patients with epilepsy. These epileptiform episodes may be associated with transient deterioration in cognitive function (53,54). Data from several studies indicate that LTG may reduce spontaneous epileptiform discharges, which may partially explain the favorable cognitive profile of LTG. In five patients displaying spontaneous EEG discharges, a single dose of LTG (120 or 240 mg in addition to existing medication) resulted in a substantial reduction in spontaneous interictal discharges within a 24-h period (55). The long-term effects of LTG on paroxysmal abnormalities have also been monitored with a computer-based analysis system (56). Twenty-one patients with intractable epilepsy (20 of whom were receiving multiple AED therapy) were evaluated before and after LTG treatment for EEG ictal events and number of spikes in a 10-min period. Before LTG treatment, patients typically showed discharges characterized by diffuse spike-wave complexes. However, after a 4-month treatment period with LTG, ictal discharges disappeared and diffuse slow wave activity was seen with no adverse effect on background activity. Nineteen of the 21 patients also showed a reduction in seizure frequency.

The effect of LTG add-on therapy in 11 patients with refractory partial seizures with or without secondary generalization has also been reported (57). LTG was added to existing therapy consisting of CBZ with at least one additional AED. EEG recordings were made at rest with eyes closed, during an attentive task (blocking reaction induced by several episodes of eyes open lasting 8–9 s), during cognitive tasks, and while performing mental arithmetic. In addition, a battery of neuropsychological tests was carried out. Before LTG treatment, EEG data revealed a decrease in fast activity at rest and a reduction in alpha and beta bands during attentive and cognitive tasks. LTG treatment resulted in a selective increase in alpha reactivity and beta power during the attentive tasks with no other detectable changes. During cortical activation, subtle changes were observed that were taken as indicative of a slight improvement in attention. Neuropsychological evaluation revealed that after 3 months of LTG therapy, no deterioration in cognitive function had occurred.

LTG also shows a promising cognitive profile in elderly patients suffering from age-associated memory impairment (58). A neuropsychological test battery, in combination with auditory event-related potentials (ERPs), was used to measure the impact of LTG on cognitive function. LTG treatment caused a reduction in amplitude of the P300 component of the ERP, and a corresponding improvement in immediate and delayed visual memory and delayed logical memory. LTG may therefore improve simple memory functions in a memory-impaired elderly population.


Levetiracetam (LEV) is a new AED, structurally and mechanistically dissimilar to other marketed AEDs. It is effective in reducing partial seizures in patients with epilepsy, both as adjunctive treatment and as monotherapy. LEV has many therapeutic advantages for patients with epilepsy. It has favorable pharmacokinetic characteristics (good bioavailability, linear pharmacokinetics, insignificant protein binding, lack of hepatic metabolism, and rapid achievement of steady-state concentrations) and a low potential for drug interactions. It is licensed for use as adjunctive treatment for partial seizures, with or without secondary generalization, in people aged over 16 years. For its impact on cognitive function, we only have data from a small pilot study that does not allow definite conclusions (59).


TGB is a γ-aminobutyric acid (GABA) uptake inhibitor that is structurally related to nipecotic acid but has an improved ability to cross the blood–brain barrier. Clinical trials have shown that TGB is effective as add-on therapy in the management of patients with refractory partial epilepsy. Three cognitive studies are available.

Dodrill et al. (60) included 162 patients who received the following treatments: placebo (n = 57), 16 mg/day TGB (n = 34), 32 mg/day TGB (n = 45), or 56 mg/day TGB (n = 26) at a fixed dose for 12 weeks after a 4-week dose titration period. Eight cognitive tests and three measures of mood and adjustment were administered during the baseline period and again during the double-blind period near the end of treatment (or at the time of dropout). The results showed no cognitive effects of monotherapy with TGB at a low or high dose, but there was some evidence for mood effects of add-on treatment with TGB at higher dosing, possibly related to titration speed. In the add-on polytherapy study by Kälviäinen et al. (61), 37 patients with partial epilepsy were included. The study protocol consisted of a randomized, double-blind, placebo-controlled, parallel-group add-on study and an open-label extension study. During the 3-month double-blind phase at low doses (30 mg/day), TGB treatment did not cause any cognitive changes as compared with placebo. TGB treatment also did not cause deterioration in cognitive performance during longer follow-up with successful treatment on higher doses after 6–12 months (mean 65.7 mg/day, range 30–80 mg/day) and after 18–24 months (mean dose 67.6 mg/day, range 24–80 mg/day). Finally, a study by Sveinbjornsdottir et al. (62) was an open trial of 22 adult patients with refractory partial epilepsy followed by a double-blind, placebo-controlled, crossover trial in 12 subjects. Nineteen patients completed the initial open titration and fixed-dose phase of the study and 11 patients completed the double-blind phase. The median daily TGB dose was 32 mg during the open fixed dose and 24 mg during the double-blind period. Neuropsychological evaluation did not show any significant effect on cognitive function in the open or double-blind phase.


GBP is a novel AED, currently used as add-on therapy in patients with partial seizures. GBP apparently has a completely novel type of action, probably involving potentiation of GABA-mediated inhibition and possibly inactivation of sodium channels. Two volunteer studies and one clinical study are available to interpret the cognitive effects.

Martin et al. (38) used an acute dose and rapid titration in six volunteers and did not find cognitive effects of GBP. Meador et al. (63) compared the cognitive effects of GBP and CBZ in 35 healthy subjects using a double-blind, randomized, crossover design with two 5-week treatment periods. During each treatment condition, subjects received either GBP 2,400 mg/day or CBZ (mean 731 mg/day). Subjects were tested at the end of each AED treatment period and in four drug-free conditions [two pretreatment baselines and two posttreatment washout periods (1 month after each AED)]. The neuropsychological test battery included 17 measures yielding 31 total variables. Significantly better performance on eight variables was found for GBP, but on no variables for CBZ. Comparison of CBZ and GBP with the nondrug average revealed significant statistical differences for 15 (48%) of 31 variables. Leach et al. (64) studied GBP in 21 patients in an add-on polytherapy study after 4 weeks of adjunctive therapy and found no change in psychomotor and memory tests. Drowsiness was more often found in higher dosing (2,400 mg). Mortimore et al. (65) did not find a difference between continued polytherapy and an add-on with GBP in measures of quality of life.


  1. Top of page
  2. Abstract

Table 1 shows the available studies and evidence concerning AEDs. The following conclusions can be drawn:

Table 1. Cognitive effects according to AED and type of study
 Acute volunteer studiesChronic (>10 days) volunteer studiesControlled studies in patients with newly diagnosed epilepsyAdd-on clinical studies in patients with epilepsyEEG dataData on specific groups: children, elderly, mentally retardedData on possible working mechanisms of cognitive effects
  1. AED, antiepileptic drug; EEG, electroencephalogram; LTG, lamotrigine; TPM, topiramate; OXC, oxcarbazepine; TGB, tiagabine; GBP, gabapentin; LEV, levetiracetam; ±, no impairment; −, impairment (− mild, − moderate, − severe); +, improvement.

LTGCohen et al., 1985 (43): ±Martin et al., 1999 (38): ±Martin et al., 1999 (38): ±Hamilton et al., 1993 (44): ±Meador et al., 2000 (45): ±Aldenkamp et al., 2002 (46): + (cognitive activation)Gillham et al., 2000 (47) +(vs. CBZ)Smith et al., 1993 (48): +Banks and Beran, 1991(49): − (cerebral efficiency)Aldenkamp et al., 1997 (50): ±Binnie et al., 1986 (55): ± (EEG)Marciani et al., 1996 (56): + (EEG)Marciani et al., 1998 (57): + (EEG)Ettinger et al., 1998 (51): − (hyperactivity in mentally retarded)Earl et al., 2000 (52): ±(mentally retarded)Mervaala et al., 1995 (58): + (memory in elderly)Available
TPMMartin et al., 1999 (38): − (language/ attention)Martin et al., 1999 (38): − (attention/verbal memory) Meador, 1997 (39): – (attention/verbal fluency)Aldenkamp et al., 2000(35): − (verbal memory)Burton and Harden, 1997 (40): − (attention)Thompson et al., 2000 (41): − (verbal intelligence)   
OXC Curran and Java, 1993 (27): +(attention/memory)Laaksonen et al., 1985(28): ±Sabers et al., 1995 (29): +(learning/memory)Äikiä et al., 1992 (30): ±McKee et al., 1994 (31): ±  Ongoing study, Protocol #: CTRI476E2337 (children) 
TGB  Dodrill et al., 1997 (60): ±Kälviäinen et al., 1996(61): ±Sveinbjornsdottir et al., 1994 (62): ±   
GBPMartin et al., 1999 (38): ±Martin et al., 1999 (38): ± Leach et al., 1997 (64): ±   
LEV   Neyens et al., 1995 (59): ±   
  • • 
    A disappointing number of controlled studies are available to interpret the cognitive effects of the newer AEDs. In fact, no more than 23 different studies have been found in the literature for the six drugs, also counting the noncontrolled clinical studies and the volunteer studies. Also, no more than six controlled epilepsy studies are available, four of these on OXC. For GBP, LEV, and TPM, no controlled epilepsy studies are available. It is clear that much more attention is warranted in drug development and in postmarketing evaluation to establish the effect of drugs on behavior.
  • • 
    The best data for effects on cognitive function exist for LTG, TPM, and to a lesser extent OXC and GBP. Additional studies are needed for TGB and LEV. Even for LTG, TPM, and OXC, conclusions must be drawn with caution, as we did not fully consider all methodological aspects of the studies involved; in particular, the statistical power of some of the studies may be below critical standards.
  • • 
    The case for LTG is particularly strong. In general, LTG has a favorable cognitive profile, both in volunteers and in patients with epilepsy. There are some remarkable studies using EEG data, and some studies show improvement of cognitive activation after introducing LTG treatment. Only in mentally retarded patients is there clinical anecdotal information revealing side effects that match the general “activating” profile: hyperactivity and restlessness.
  • • 
    OXC also has a favorable cognitive profile, although the evidence is less strong than for LTG. There is an unconfirmed report about positive effects on memory.
  • • 
    TPM does induce cognitive impairment. There is an effect on attention, but also a very specific effect on verbal function and language. This is in line with studies in clinical practice (see reference 37) reporting dysphasia. It is clear that dose and titration speed have an effect but do not completely explain the development of cognitive impairments. Particularly worrisome is the complete absence of planned cognitive studies that may further elucidate the specific risk factors associated with this drug.


  1. Top of page
  2. Abstract
  • 1
    AldenkampAP, DodsonWE, eds. Epilepsy and education; cognitive factors in learning behavior. Epilepsia 1990;31(suppl 4):S920.
  • 2
    Dodson WE, Pellock JM. Pediatric epilepsy: diagnosis and treatment. New York : Demos Publications, 1993.
  • 3
    Rapin I. Children with brain dysfunction: neurology, cognition, language and behavior. New York : Raven Press, 1982.
  • 4
    Dodson WE, Trimble MR. Epilepsy and quality of life. New York : Raven Press, 1994.
  • 5
    Aldenkamp AP, Dreifuss FE, Renier WO, Suumeijer PBM. Epilepsy in children and adolescents. Boca Raton, FL: CRC Press, 1995.
  • 6
    Aldenkamp AP. Antiepileptic drug treatment and epileptic seizures–effects on cognitive function. In: TrimbleM, Schmitz B, eds. The neuropsychiatry of epilepsy. New York : Cambridge University Press, 2002: 25667.
  • 7
    Ideström CM, Schalling D, Carlquist U, Sjoqvist F. Acute effects of diphenylhydantoin in relation to plasma levels. Behavioral and psychological studies. Psychol Med 1972;2: 11120.
  • 8
    Dodrill CB, Troupin AS. Psychotropic effects of carbamazepine in epilepsy: a double-blind comparison with phenytoin. Neurology 1977;27: 10238.
  • 9
    Vermeulen J, Aldenkamp AP. Cognitive side-effects of chronic antiepileptic drug treatment: a review of 25 years of research. Epilepsy Res 1995;22: 6595.
  • 10
    Trimble MR. Anticonvulsant drugs and cognitive function: a review of the literature. Epilepsia 1987;28(suppl 3):3745.
  • 11
    Aldenkamp AP, Alpherts WCJ, Blennow G, et al. Withdrawal of antiepileptic medication—effects on cognitive function in children: The Multicentre Holmfrid Study. Neurology 1993;43: 4150.
  • 12
    Meador KJM, Loring DW, Huh K, Gallagher BB, King DW. Comparative cognitive effects of anticonvulsants. Neurology 1990;40: 3914.
  • 13
    American Academy of Pediatrics. Behavioral and cognitive effects of anticonvulsant therapy. Committee on Drugs. Pediatrics 1985;76: 6447.
  • 14
    Loescher W, Schmidt D. New horizons in the development of antiepileptic drugs. Epilepsy Res 2002;50: 316.
  • 15
    Privitera M, Fincham R, Penry J, et al. Topiramate placebo-controlled dose-ranging trial in refractory partial epilepsy using 600-, 800-, and 1000-mg daily dosages. Topiramate YE Study Group. Neurology 1996;46: 167883.
  • 16
    Jette NJ, Marson AG, Hutton JL. Topiramate add-on for drug-resistant partial epilepsy. Cochrane Database Syst Rev 2002;(3):CD001417.
  • 17
    Marson AG, Kadir ZA, Hutton JL, Chadwick DW. The new antiepileptic drugs: a systematic review of their efficacy and tolerability. Epilepsia 1997;38: 85980.
  • 18
    Wong IC. New antiepileptic drugs. Study suggests that under a quarter of patients will still be taking the new drugs after six years. Br M J 1997;314: 6034.
  • 19
    Lhatoo SD, Wong ICK, Sander JW. Prognostic factors affecting long-term retention of topiramate in patients with chronic epilepsy. Epilepsia 2000;41: 33841.
  • 20
    StefanH, KrämerG, Mamoli, B, eds. Challenge epilepsy—new antiepileptic drugs. Berlin : Blackwell Science, 1998.
  • 21
    Kellet MW, Smith DF, Stockton PA, Chadwick DW. Topiramate in clinical practice: first year's postlicensing experience in a specialist epilepsy clinic. J Neurol Neurosurg Psychiatry 1999;66: 75963.
  • 22
    Marson AG, Kadir ZA, Hutton JL, Chadwick DW. Gabapentin for drug-resistant partial epilepsy. Cochrane Database Syst Rev 2000;(2):CD001415.
  • 23
    Marson AG, Hutton JL, Leach JP, et al. Levetiracetam, oxcarbazepine, remacemide and zonisamide for drug resistant localization-related epilepsy: a systematic review. Epilepsy Res 2001;46: 25970.
  • 24
    Chadwick DW, Marson T, Kadir Z. Clinical administration of new antiepileptic drugs: an overview of safety and efficacy. Epilepsia 1996;37(suppl 6):S1722.
  • 25
    Aldenkamp AP. Cognitive and behavioural assessment in clinical trials: when should they be done Epilepsy Res 2001;45: 1559.
  • 26
    Gallassi R, Morreale A, Di Sarro R, Marra M, Lugaresi E, Baruzzi A. Cognitive effects of antiepileptic drug discontinuation. Epilepsia 1992;33(suppl 6):S414.
  • 27
    Curran HV, Java R. Memory and psychomotor effects of oxcarbazepine in healthy human volunteers. Eur J Clin Pharmacol 1993;44: 52933.
  • 28
    Laaksonen R, Kaimola K, Grahn-Teräväinen E, Waltimo O. A controlled clinical trial of the effects of carbamazepine and oxcarbazepine on memory and attention [Abstract]. 16th International Epilepsy Congress, Hamburg, 1985.
  • 29
    Sabers A, Moller A, Dam M, et al. Cognitive function and anticonvulsant therapy: effect of monotherapy in epilepsy. Acta Neurol Scand 1995;92: 1927.
  • 30
    Äikiä M, Kälviäinen R, Sivenius J, Halonen T, Riekkinen RJ. Cognitive effects of oxcarbazepine and phenytoin monotherapy in newly diagnosed epilepsy: one year follow-up. Epilepsy Res 1992;11: 199203.
  • 31
    McKee PJ, Blacklaw J, Forrest G, et al. A double-blind, placebo-controlled interaction study between oxcarbazepine and carbamazepine, sodium valproate and phenytoin in epileptic patients. Br J Clin Pharmacol 1994;37: 2732.
  • 32
    White HS. Clinical significance of animal seizure models and mechanism of action studies of potential antiepileptic drugs. Epilepsia 1997;38(suppl 1):S917.
  • 33
    Faught E, Wilder BJ, Ramsay RE, et al. Topiramate placebo-controlled dose-ranging trial in refractory partial epilepsy using 200-, 400-, and 600-mg daily dosages. Neurology 1996;46: 168490.
  • 34
    Aldenkamp AP. Cognitive effects of topiramate, gabapentin and lamotrigine in healthy young adults. Neurology 2000;54: 2702.
  • 35
    Aldenkamp AP, Baker G, Mulder OG, et al. A multicentre randomized clinical study to evaluate the effect on cognitive function of topiramate compared with valproate as add-on therapy to carbamazepine in patients with partial-onset seizures. Epilepsia 2000;41: 116778.
  • 36
    Ketter TA, Post RM, Theodore WH. Positive and negative psychiatric effects of antiepileptic drugs in patients with seizure disorders. Neurology 1999;53(5 suppl 2):5367.
  • 37
    Tatum WO, French JA, Faught E, et al. Postmarketing experience with topiramate and cognition. Epilepsia 2001;42: 113440.
  • 38
    Martin R, Kuzniecky R, Ho S, et al. Cognitive effects of topiramate, gabapentin, and lamotrigine in healthy young adults. Neurology 1999;52: 3217.
  • 39
    Meador KJ. Assessing cognitive effects of a new AED without the bias of practice effects [Abstract]. Epilepsia 1997;38(suppl 3):60.
  • 40
    Burton LA, Harden C. Effect of topiramate on attention. Epilepsy Res 1997;27: 2932.
  • 41
    Thompson PJ, Baxendale SA, Duncan JS, Sander JW. Effects of topiramate on cognitive function. J Neurol Neurosurg Psychiatry 2000;69: 63641.
  • 42
    Aldenkamp AP, Baker G. A systematic review of the effects of lamotrigine on cognitive function and quality of life. Epilepsy Behav 2001;2: 8591.
  • 43
    Cohen AF, Ashby L, Crowley D, Land G, Peck AW, Miller AA. Lamotrigine (BW430C), a potential anticonvulsant. Effects on the central nervous system in comparison with phenytoin and diazepam. Br J Clin Pharmacol 1985;20: 61929.
  • 44
    Hamilton MJ, Cohen AF, Yuen AW, et al. Carbamazepine and lamotrigine in healthy volunteers: relevance to early tolerance and clinical trial dosage. Epilepsia 1993;34: 16673.
  • 45
    Meador KJ, Loring DW, Ray PG, Perrine KR, Bazquez BR, Kalbosa T. Differential effects of carbamazepine and lamotrigine [Abstract]. Neurology 2000;54(suppl 3):A84.
  • 46
    Aldenkamp AP, Arends J, Bootsma HP, et al. Randomized, double-blind parallel-group study comparing cognitive effects of a low-dose lamotrigine with valproate and placebo in healthy volunteers. Epilepsia 2002;43: 1926.
  • 47
    Gillham R, Kane K, Bryant-Comstock L, Brodie MJ. A double-blind comparison of lamotrigine and carbamazepine in newly diagnosed epilepsy with health-related quality of life as an outcome measure. Seizure 2000;9: 3759.
  • 48
    Smith D, Baker G, Davies G, Dewey M, Chadwick DW. Outcomes of add-on treatment with lamotrigine in partial epilepsy. Epilepsia 1993;34: 31222.
  • 49
    Banks GK, Beran RG. Neuropsychological assessment in lamotrigine treated epileptic patients. Clin Exp Neurol 1991;28: 2307.
  • 50
    Aldenkamp AP, Mulder OG, Overweg J. Cognitive effects of lamotrigine as first line add-on in patients with localized related (partial) epilepsy. J Epilepsy 1997;10: 11721.
  • 51
    Ettinger AB, Weisbrot DM, Saracco J, Dhoon A, Kanner A, Devinsky O. Positive and negative psychotropic effects of lamotrigine in patients with epilepsy and mental retardation. Epilepsia 1998;39: 8747.
  • 52
    Earl N, McKee JR, Sunder TR, et al. Lamotrigine adjunctive therapy in patients with refractory epilepsy and mental retardation [Abstract]. Epilepsia 2000;41(suppl 1):72.
  • 53
    Aarts JH, Binnie CD, Smit AM, Wilkins AJ. Selective cognitive impairment during focal and generalized epileptiform EEG activity. Brain 1984;107: 293308.
  • 54
    Aldenkamp AP, Arends J, Overweg-Plandsoen TC, et al. Acute cognitive effects of nonconvulsive difficult-to-detect epileptic seizures and epileptiform electroencephalographic discharges. J Child Neurol 2001;16: 11923.
  • 55
    Binnie CD, Emde BW, Kasteleijn-Nolste-Trenite DG, et al. Acute effects of lamotrigine (BW430C) in persons with epilepsy. Epilepsia 1986;27: 24854.
  • 56
    Marciani MG, Spanedda F, Bassetti MA, et al. Effect of lamotrigine on EEG paroxysmal abnormalities and background activity: a computerized analysis. Br J Clin Pharmacol 1996;42: 6217.
  • 57
    Marciani MG, Stanzione P, Mattia D, et al. Lamotrigine add-on therapy in focal epilepsy: electroencephalographic and neuropsychological evaluation. Clin Neuropharmacol 1998;21: 417.
  • 58
    Mervaala E, Koivista K, Hanninen T, et al. Electrophysiological and neuropsychological profiles of lamotrigine in young and age-associated memory impairment (AAMI) subjects [Abstract]. Neurology 1995;45(suppl 4):A259.
  • 59
    Neyens LGJ, Alpherts WCJ, Aldenkamp AP. Cognitive effects of a new pyrrolidine derivative (levetiracetam) in patients with epilepsy. Prog Neuropsychopharmacol Biol Psychiatry 1995;19: 4119.
  • 60
    Dodrill CB, Arnett JL, Sommerville KW, Shu V. Cognitive and quality of life effects of differing dosages of tiagabine in epilepsy. Neurology 1997;48: 102531.
  • 61
    Kälviäinen R, Äikiä M, Mervaala E, Saukkonen AM, Pitkanen A, Riekkinen PJ Sr. Long-term cognitive and EEG effects of tiagabine in drug-resistant partial epilepsy. Epilepsy Res 1996;25: 2917.
  • 62
    Sveinbjornsdottir S, Sander JW, Patsalos PN, Upton D, Thompson PJ, Duncan JS. Neuropsychological effects of tiagabine, a potential new antiepileptic drug. Seizure 1994;3: 2935.
  • 63
    Meador KJ, Loring DW, Ray PG, et al. Differential cognitive effects of carbamazepine and gabapentin. Epilepsia 1999;40: 127985.
  • 64
    Leach JP, Girvan J, Paul A, Brodie MJ. Gabapentin and cognition: a double blind, dose ranging, placebo controlled study in refractory epilepsy. J Neurol Neurosurg Psychiatry 1997;62: 3726.
  • 65
    Mortimore C, Trimble M, Emmers E. Effects of gabapentin on cognition and quality of life in patients with epilepsy. Seizure 1998;7: 35964.