Introduction

Traumatic brain injury (TBI) can result in cognitive and behavioral deficits as well as seizures. Anti-seizure medications (ASMs) are the mainstay of therapy for seizure management. Thus, the effects of ASMs on cognition are vital to the practice of any healthcare provider who provides treatment for patients with TBI and seizures. ASMs exert their effects on cognition through modulating the different modalities of cognitive processing and also through affecting behavior and psychology. The most common effects of ASMs on cognitive functions include reductions in psychomotor speed, attention (particularly sustained or complex attention), dual processing, memory and naming or word finding.¹ ASMs can affect behavior by producing positive effects such as mood stabilization, or by producing adverse effects such as depression, irritability, agitation or psychosis.

Selecting the appropriate ASM for each patient can be done by weighing the positive and negative cognitive effects of each medication. Research has shown that benzodiazepines, phenobarbital and topiramate have higher associations with adverse cognitive effects compared to other ASMs. Positive behavioral modulation has been shown with carbamazepine, lamotrigine and valproate while negative behavioral and mood effects have been seen with levetiracetam, phenobarbital and topiramate. Further attention must be given to therapeutic ranges of drugs, blood levels, rates of drug titration, polypharmacy and the age of the patient. Higher doses and higher blood levels may result in more cognitive side effects. There is an increased risk of cognitive side effects when ASMs are up titrated quickly, and insufficient time is given in the first few weeks for habituation. Decreasing the number of ASMs may improve cognition and even decrease seizures. The elderly have increased vulnerability to the cognitive side effects of medications.¹ The most favorable outcomes are reached when medications are selected with the specific patient and their cognitive reserve in mind. The following sections will examine research related to many ASMs and their effects on cognition.

Carbamazepine, Phenobarbital, Phenytoin and Valproate

Carbamazepine and phenytoin have similar mechanisms of action, preventing repetitive and sustained firing of neurons via action on voltage-gated sodium channels. In addition, they also share similar cognitive side effect profiles. They were compared in a double-blind, randomized, crossover, monotherapy design study that controlled for anticonvulsant blood levels, and the results showed that there is no difference in the cognitive effects of the two medications.^{2, 3} The only significant differences found in this study included decreased performance on the finger tapping test for phenytoin in comparison to carbamazepine and decreased performance on the Stroop test for carbamazepine in comparison with phenytoin. Both valproate and phenobarbital have multiple mechanisms of action, but both exert effects by making the inhibitory neurotransmitter GABA either more available or more effective. Valproate also blocks voltage-gated sodium channels, and phenobarbital also causes decreased efficacy of the excitatory neurotransmitter glutamate. Phenobarbital was found to have worse cognitive effects than both phenytoin and carbamazepine.² A placebo-controlled, prospective, randomized, double-blind, parallel group study of anticonvulsant withdrawal in a patient population taking a single ASM with completely controlled seizures found improved cognitive function on neuropsychiatric tests once the ASM was withdrawn compared to patients in whom ASM was not withdrawn.⁴ It is important to note that being on any of these ASMs, when compared to being on no ASMs, resulted in significantly poorer performance on a battery of neuropsychiatric tests (performance was worse in 52% of tested variables).

Brivaracetam

Brivaracetam was approved for clinical use in patients with focal onset seizures in February 2016 and is among the newest ASMs. It is an analog of levetiracetam and has highly selective binding for a novel brain-specific binding site synaptic vesicle protein 2 A (SV2A).⁵ A randomized, double-blind, placebo-controlled, four-way crossover design study in healthy volunteers comparing acute single doses of brivaracetam, levetiracetam, lorazepam and placebo found no significant difference in cognitive testing between placebo, levetiracetam and brivaracetam; however, it did find a significant decline in cognitive tasks with lorazepam.⁶ In a study of brivaracetam, 36 patients who were suffering from adverse side effects of levetiracetam were switched to brivaracetam: 24 of these patients (66.67%) experienced a clinically meaningful reduction in side effects after the switch.⁷ Brivaracetam was discontinued in 26/93 patients (28%). In the ten who switched back to levetiracetam, seven had a reduction in adverse effects. However, no randomized, double-blind comparisons of brivaracetam and levetiracetam behavioral effects have been conducted.

Eslicarbazepine

Eslicarbazepine is a new generation ASM chemically related to carbamazepine and oxcarbazepine but has structural and metabolic differences. In a randomized, double-blind, crossover design comparing eslicarbazepine with immediate-release carbamazepine in healthy volunteers, eslicarbazepine demonstrated fewer adverse neuropsychological effects than carbamazepine.⁸

Gabapentin

Gabapentin was approved by the FDA in 1993 as an adjuvant for treatment of focal onset seizures. Studies examining the cognitive effects of gabapentin have found few cognitive side effects from gabapentin when used in the treatment of epilepsy. The effect of gabapentin on cognition was assessed at doses of 1,200 mg, 1,800 mg and 2,400 mg per day in a double-blind, placebo-controlled, add-on crossover study and reported no substantial effect on composite psychomotor/memory scores, nor any alteration in any self-assessment subscore.⁹ A direct comparison of gabapentin, carbamazepine and placebo in a double-blind, randomized, crossover study of healthy volunteers found significantly better performance with gabapentin than carbamazepine on 26% of variables in cognitive tests.¹⁰ A randomized, double-blind, parallel group study of carbamazepine and gabapentin in healthy volunteers found both ASMs caused EEG slowing and worsening function on cognitive testing but with no significant difference between the two drugs.¹¹

Lacosamide

Lacosamide was approved in 2008 by the FDA for adjunctive treatment of focal onset seizures. Unlike other sodium channel ASMs, lacosamide enhances the slow inactivation of voltage-gated sodium channels without affecting their fast inactivation. The effects of lacosamide and carbamazepine were compared in a randomized, double-blind, crossover study in healthy subjects, which found that carbamazepine had worse scores than lacosamide for the primary composite neuropsychological outcome and for the composite EEG score. Lacosamide was better than carbamazepine on 36% of the neuropsychological tests.¹²

Lamotrigine

Lamotrigine was approved for the treatment of epilepsy in 1994 and acts on the voltage gated sodium channels. Since its approval, there have been many studies that show a superior cognitive profile for lamotrigine in comparison to other ASMs including topiramate, phenytoin, carbamazepine and diazepam.^13–18 When compared to carbamazepine, patients on lamotrigine performed better on 19/40 (48%) variables in cognitive tests.¹³ Direct comparison of lamotrigine to topiramate in healthy volunteers yielded better performance for patients on lamotrigine in 33/41 (80%) variables without adjustment for drug levels and 19/41 (46%) with adjustment for drug levels.¹⁴ Lamotrigine was compared to valproate and placebo in healthy adults, and results showed that patients on lamotrigine performed better compared to placebo on three of four tests of reaction time. Patients on lamotrigine performed better compared to valproate on two of four tests of reaction time and had significant reduction in drug-related cognitive complaints. Patients on lamotrigine had improvement in five of six mood scales while patients on valproate had worsening on four of six mood scales in the same study.¹⁵ A separate study found significant unsteadiness in patients given phenytoin, drowsiness in patients given diazepam and found no significant side effects in patients taking lamotrigine.¹⁶ Lamotrigine’s preferable cognitive profile as well as its effectiveness as an ASM and a mood stabilizer is often of great benefit in epilepsy patients who frequently require management of mood and cognition in addition to seizure control.¹⁹

Levetiracetam

Levetiracetam was approved by the FDA for the treatment of epilepsy in 1999. The mechanism by which it exerts its antiseizure properties are not completely understood, but it binds to SV2A, a synaptic vesicle glycoprotein, inhibits presynaptic calcium channels and reduces neurotransmitter release. Levetiracetam is often a first choice in the treatment of epilepsy due to its lack of interaction with other medications and its favorable cognitive profile. The cognitive effects of levetiracetam in comparison to carbamazepine were compared in a randomized, double-blind, two-period crossover design study¹⁹ – the study found that levetiracetam had significantly fewer cognitive effects on 44% of tested variables. This data, when considered in combination with studies comparing carbamazepine to older ASMs like phenytoin, phenobarbital and valproate, indicates that levetiracetam also has fewer cognitive side effects than the listed older ASMs. Although the cognitive profile of levetiracetam is favorable, it may lead to unfavorable psychiatric side effects in the form of worsening depression, irritation or agitation. For example, the behavioral complications of levetiracetam in comparison to gabapentin and lamotrigine have been evaluated, and it has become apparent that levetiracetam carries a higher negative behavioral side effect profile.^20–21 Pyridoxine has been examined as an add-on treatment to decrease the behavioral side effects of levetiracetam. A randomized, single-blind, case-control trial in children looked at the effect of adding pyridoxine for patients who had been taking levetiracetam for 1 month and were experiencing negative side effects. The study found that 92% of patients who initiated pyridoxine after 1 month of LEV treatment did not need to change or suspend levetiracetam, and behavioral adverse effects of levetiracetam improved after 6–12 days of treatment.²² This study supports prior case reports, but an additional randomized double-blind investigation is needed.

Oxcarbazepine

Oxcarbazepine is a structural derivative of carbamazepine and was formulated to decrease the side effect profile of carbamazepine. An open-label, randomized, active-control, three-arm, parallel-group, 6-month study compared the cognitive side effects of oxcarbazepine to carbamazepine and valproate in patients 6 to <17 years of age and disclosed no significant difference.²³ The cognitive effects of oxcarbazpine have been compared to phenytoin in adult patients with epilepsy²⁴ and in healthy subjects,²⁵ and no significant difference has been found.

Perampanel

Perampanel was approved in 2012 and works as a non-competitive antagonist of AMPA receptors. It carries a black box warning for serious psychiatric and behavioral changes such as aggression, hostility and homicidal ideation. A total of 133 patients with poorly controlled focal onset seizures were randomized to perampanel or placebo in a randomized, double-blind, placebo-controlled, adjunctive study and the cognitive effects of perampanel were evaluated.²⁶ No difference from placebo was seen for perampanel for global cognitive score, two of five subdomains and four other cognitive measures. Patients on perampanel performed worse than placebo in two domains: continuity of attention and speed of memory.²⁶ Similar results were found in a long-term, open-label follow up study on 112 patients aged 12–18.²⁷ This study followed patients for up to 52 weeks and found no significant cognitive effects for cognitive drug research system global cognition score, continuity of attention, quality of episodic memory, quality of working memory or speed of memory, but did find a significant decline in power of attention at end of treatment compared with baseline.²⁷

Rufinamide

Rufinamide was approved for use in seizures related to Lennox–Gastaut syndrome in 2008 and is used in treatment of focal onset seizures. It is thought act through limiting the frequency of sodium dependent neuronal action potentials. A multicenter, multinational double-blind, randomized, placebo-controlled, parallel-study looked at the cognitive effects of rufinamide when given as an add-on medication for patients with poorly controlled seizures.²⁸ Four doses of rufinamide were administered (200 mg/day, 400 mg/day, 800 mg/day and 1,600 mg/day), and cognitive testing was performed before start of rufinamide and after 3 months of treatment. Cognitive functions (psychomotor speed and alertness, mental information processing speed and attention and working memory) showed no significant difference before and after treatment with rufinamide.

Tiagibine

Tiagabine was approved for use in epilepsy in 1998 by the FDA and works by blocking the uptake of GABA and thus reducing excitation and likelihood of seizures. Its effects on cognition were examined in a double-blind, add-on, placebo-controlled, parallel, multicenter, dose-response efficacy study in patients with focal epilepsy whose complex partial seizures were difficult to control. A total of 162 patients were randomized to placebo or tiagibine 16 mg/day, 32 mg/day or 56 mg/day. Their performance on eight cognitive tests and three measures of mood and adjustment showed no significant difference before initiation of tiagabine compared to post 12 weeks of treatment.²⁹

Topiramate

Topiramate has been approved for use in the treatment of seizures since 1996. It has been shown to produce state-dependent blocking of sodium channels, enhance GABA transmission, reduce action of excitatory neurotransmitters via kainate and AMPA receptors and decrease the action of carbonic anhydrase. Patients report cognitive side effects such as speech problems or problems with word finding, psychomotor slowing and memory impairment.^30–33 These effects are more common with higher doses or rapid up titration. A multicenter, randomized, observer-blind trial looked at the cognitive effects of topiramate in comparison to valproate when added on to carbamazepine for treatment of seizures. It found a statistically significant worsening in short-term verbal memory with topiramate in comparison to valproate and placebo. This study also found that gradual introduction of topiramate (introduced at 25 mg and increased with weekly 25 mg/d increments) can decrease the extent of cognitive impairment compared to rapid introduction at higher doses (100–200 mg/day introductory dose).³⁴ A similar study found that patients taking topiramate performed worse than those taking valproate on two cognitive tests: the Symbol Digit Modalities Test and the Controlled Oral Word Association Test.³⁵ The cognitive effects of topiramate and gabapentin were compared in a double-blind, placebo-controlled study of healthy volunteers and showed that patients taking topiramate performed worse than those taking gabapentin on 12 of 24 cognitive measures.³⁶ The cognitive effects of topiramate were compared to lamotrigine in healthy volunteers¹⁴ and in patients with epilepsy,¹⁸ and significantly greater cognitive impairment was seen on multiple measures in subjects taking topiramate in both studies. The dose-dependent effects of topiramate were studied in a double-blind, placebo-controlled study on cognitively normal adults and showed that a greater number of subjects developed cognitive impairment at higher doses of topiramate, especially at doses >200 mg/day.³⁷

Vigabatrin

Vigabatrin was approved for use in 2009 and works by irreversibly inhibiting the breakdown of GABA. Severe behavioral disturbances such as depression and psychosis have been reported in 3.4% of adult patients and 6% of pediatric patients started on Vigabatrin.³⁸ The effects of vigabatrin on cognition and quality of life were examined in patients with epilepsy in a dose-dependent fashion, and little impact was seen in cognition or quality of life with placebo versus 1 g, 3 g or 6 g of Vigabatrin.³⁹ A separate study randomized 45 patients to vigabatrin or placebo for a double-blinded period of 20 weeks and then followed them for an 18-month open label period. Patients on vigabatrin had a significant reduction in a measure of motor speed and overall score on a design learning test in the first 20 weeks of treatment. Of the patients who continued follow-up in the open label phase, these side effects either stayed the same with treatment (did not worsen) or improved.⁴⁰