Background: Attention-deficit/hyperactivity disorder (ADHD) affects 10–30% of children with epilepsy, making it one of the most common comorbidities in epilepsy. Stimulant medications are first line pharmacological treatment of ADHD, yet there areconcerns regarding the safety of stimulant treatment in patients with comorbid ADHD and epilepsy. This is due to the long held view that stimulants may lower the seizure threshold and increase seizure frequency [1]. Evidence for such an effect are however inconsistent and largely based on studies with small sample sizes, highly selected patient populations and observational studies that have not sufficiently addressed issues of confounding [2]. The aim of this pharmacoepidemiological register based study wastherefore to estimate the risk of seizures in relation to ADHD medication use in a population based cohort of individuals with a history of seizures.

Methods: Using Swedish national registers, we identified a cohort of 62,361 individuals (48% female) born in Sweden between 1960 and 2004, with at least one seizure episode according to ICD codes. Each individual was followed from January 1st 2006, their first seizure or age five, up until December 31st 2013 or death, whichever came first. We identified periods of ADHD medication use (methylphenidate, amphetamine, dexamphetamine, lisdexamfetamine, and atomoxetine) from the Swedish National Prescribed Drug Register. A period was defined as on-medication when two consecutive prescriptions where no more than 183 days apart, and off-medication if more than 183 daysapart. We obtained information on medical visit for unplanned seizures events from the Swedish National Patient register using ICD codes. We estimated the population level association between ADHD medications and the rate of seizures during medicated and non-medication periods using a cox proportional hazards regression model. To adjust for individual-specific confounding that may influence both seizure risk and the likelihood of receiving ADHD medication, we used a stratified Cox regression model to estimate the rate of seizures during medicated and non-medicated periods, within the same individual.

Preliminary Result: A total of 59,749 seizure events occurred during 361,501 person years of follow-up. ADHD medications were not associated with the rate of seizures at the population level (HR = 1.06, 95%CI 0.91–1.23). In the within-individual analysis, ADHD medication periods were associated with a reduced rate of seizures (HR = 0.70, 95%CI 0.62–0.79), compared to non-medicated periods. Estimates did not differ across sex, nor in age restricted analyses including only ages 5 and 20 years. All analyses were adjusted for age as a time-varying covariate. Population level analyses were additionally adjusted for sex.

Conclusions: Our findings suggest that ADHD medications are not associated with an increased risk of seizures. Rather, results from the within-individual analysis, which adjusts for factors that are constant within the individual, such as genetic factors and underlying disorder severity, suggest a protective effect of ADHD medication treatment on seizure rates. Next, we will study the effect of concurrent antiepileptic medication use and whether the observed effect differs by stimulant and non-stimulant ADHD medications. We will also further investigate possible mechanisms contributing to the observed protective effect of ADHD medications on seizure rates.

References

[1] Williams, A.E., Giust, J.M., Kronenberger, W.G., Dunn, D.W., 2016. Epilepsy and attention-deficit hyperactivity disorder: links, risks, and challenges. Neuropsychiatr Dis Treat 12, 287–296.

[2] Ravi, M., Ickowicz, A., 2016. Epilepsy, Attention-Deficit/Hyperactivity Disorder and Methylphenidate: Critical Examination of Guiding Evidence. Journal of the Canadian Academy of Child and Adolescent Psychiatry 25 (1), 50.