Withdrawing antiepileptic drugs in seizure-free patients: what are the cognitive benefits?

Background All major antiepileptic drugs (AEDs) have cognitive side effects, but data on the extent of the problem are conflicting. Objective To evaluate the effect of AED withdrawal on attention, reaction time, and speed of information processing. Design and intervention This randomized, double-blind, placebo-controlled study, conducted in Norway, assessed 139 adult patients with epilepsy (aged 18–66 years) who had been seizure-free for at least 2 years. Exclusion criteria included juvenile myoclonic epilepsy, paroxysmal epileptiform activity on electroencephalography in patients with primary generalized epilepsy, and mental retardation. All participants were on AED monotherapy, most with carbamazepine (64%) or valproate (19%). AED treatment was discontinued over a period of 12 weeks (switch to placebo) in 64 patients, whereas AED treatment was continued in 75 patients. Neuropsychological assessments were conducted at baseline and at 7 months after the intervention (treatment discontinuation or no discontinuation) using the California Computerized Assessment Package (CalCAP®; Miller EN, Los Angeles, CA, USA). This test battery comprises four subtests of simple reaction time (RT) and six subtests that assess more-complex aspects of attention, choice RT, psychomotor speed and rapid information processing. The investigators determined the mean RTs for all subtests, as well as the total numbers of true-positive and false-positive responses for the six complex subtests (22 items tested in total). To adjust for multiple comparisons, the significance level of independent t-tests was set at P <0.01. Outcome measures The outcomes were the changes in CalCAP results (mean RT, true-positive responses, true-negative responses) between baseline and 7 months after the intervention. Results With regard to the mean RT, AED withdrawal significantly improved results on a choice RT task and on a task evaluating rapid language discrimination, compared with AED continuation (difference between baseline and 7 months: -24 ms vs 4 ms, P <0.001 for choice RT; -17 ms vs 7 ms, P = 0.003 for language discrimination). Improvements after AED withdrawal were nearly significant (P-values in the range 0.013–0.067) for three other subtests assessing complex cognitive processing and nonsignificant for all subtests evaluating simpler forms of attention and RT. With regard to true-positive responses, near-significant (P 0.05) improvements were achieved for form discrimination and language discrimination. There were no significant differences between the two groups in terms of false-positive responses. Patients receiving carbamazepine or valproate monotherapy had results that were comparable to those observed for the whole study group, but valproate withdrawal did not result in significant improvements over valproate continuation in the choice RT and language discrimination subtests. Conclusion Withdrawal of AED monotherapy in patients with seizure-free epilepsy can have a positive impact on the performance of tasks that require divided attention or fast information processing.