This transcript has been edited for clarity.
Andrew N. Wilner, MD: Welcome to Medscape. I’m Dr Andrew Wilner, reporting on the American Epilepsy Society meeting. With me today is Dr Olga Taraschenko, director of the Comprehensive Epilepsy Program at the University of Nebraska Medical Center. We will discuss her paper where she reviewed the association between antiseizure medications and movement disorders. Welcome, Dr Taraschenko.
Olga Taraschenko, MD, PhD: Thank you for having me.
Wilner: Thanks for joining us. I picked your paper because I thought it was pretty interesting. I remember way back, seeing patients on phenytoin who also had nystagmus and ataxia. We always look for that at every visit in the epilepsy clinic, but it seems that you found a whole lot more of an association with antiseizure drugs. Tell us how you got interested in this and what you found.
Taraschenko: The idea of conducting the search in the literature came from one of our residents, Dr Dan Zhou, who rotated a lot with the epilepsy team. He was curious about the frequent side effects that we observe both in the clinic and in the epilepsy monitoring unit — new-onset abnormal movements. So we decided to conduct a systematic review of literature in collaboration with our librarian, and we did a comprehensive search. We searched all available databases from inception until the last year. We came up with about 5000 records. And then we applied our inclusion criteria. Specifically, we were interested in new-onset movement disorders, which were linked to particular antiseizure medications. We included all antiseizure medications available on the market. We also reported the data on average daily dose whenever it was available, as well as serum levels if they were measured, and also remediation steps that were taken to alleviate those side effects. As a result of this application of inclusion criteria, we came up with over 250 papers, which were mostly case reports and small case series with five or fewer patients, as well as 45 manuscripts, which were focused on comprehensive randomized controlled studies and cross-sectional studies.
Wilner: How do you know that these movement disorders were side effects from the antiseizure drugs and not just some concomitant finding?
Taraschenko: In this project, by design, we had to rely on the authors’ determination that there actually was a temporal relationship between the initiation of an antiseizure drug and also resolution, or at least improvement of the symptoms, with discontinuation of the drug. We took note of other common concomitant conditions that could present with abnormal movements. However, we only had a handful of patients with underlying Huntington’s chorea and Parkinson’s disease. There were not too many patients. And we also looked through the papers just to make sure that we agreed with the determination of the authors. But of course, it’s a retrospective study, a systematic review. So potentially there may be room for bias.
Wilner: So I’m imagining a gigantic Excel spreadsheet here. We’ve talked about phenytoin. Was that the number-one drug that was associated with abnormal movements?
Taraschenko: That’s exactly right; phenytoin came up most frequently in the literature. At first we thought that it was because phenytoin is an older drug and has been on the market for a long time. But even among older first-generation anticonvulsants, phenytoin clearly took the lead. And we found primarily chorea and athetosis, as well as eye movement abnormalities, like you mentioned — nystagmus and ophthalmoplegia, and even opsoclonus and myoclonus — in association with phenytoin. We also, as you said, in that large spreadsheet had carbamazepine and valproate. Valproate frequently came in association with parkinsonism. But even newer antiseizure medications that we think of as being safe and well tolerated came up in the search. We saw a lot of eye movement abnormalities with lamotrigine as well as or facial dyskinesia and restless leg syndrome with topiramate.
Taraschenko: Pretty much, based on the bulk of the literature that was available. But we were also surprised by the high number of reports on myoclonus in association with gabapentin and pregabalin. And interestingly, about half of the patients who presented with this, for whom the levels were available, and also some other labs were available, had abnormally high creatinine. So the clearance of the medication was clearly affected. And so in the settings they presented, it was myoclonus. That was a frequent flier as well.
Wilner: Yes, the first case I saw like that was in the dialysis unit. Many of these patients are on gabapentin for peripheral neuropathy or something like that, and this patient had really dramatic myoclonus. So I had to look it up. And now, I guess every few months, there’s another patient in the dialysis unit that has myoclonus. That’s kind of a pretty standard neurology consult now. I think this really helps if a patient who is taking a seizure drug develops a movement disorder; your first reflex shouldn’t be to do MRI, right? It should be to think about the medication and see if there’s some temporal relationship. And was it usually dose related? You mentioned that with gabapentin, it’s associated with a high creatinine. Was there usually a dose relationship?
Taraschenko: Not so much. The doses that were actually listed were quite modest. For phenytoin I think the average daily dose was around 400 mg, which is quite modest in the epilepsy world. The doses of lamotrigine were close to 200 mg daily, which is actually a tiny dose for somebody who has refractory seizures. What we’ve noticed, at least for phenytoin, is that the serum levels mattered. So for some reason, half of the patients had supratherapeutic levels, and that seemed to somehow be related to the development of abnormal movements.
Wilner: Was there any sort of inverse relationship? For example, with alcohol, can suppress tremor. Did you see any sort of effect like that with antiseizure drugs, that they’re actually reducing preexisting abnormal movements or being beneficial and in some way?
Taraschenko: Well, we didn’t particularly look at that, but we were surprised to see that even medications that we used traditionally used to suppress abnormal movements, such as benzodiazepines and phenobarbital, were named in several reports in association with abnormal movements. So that was quite a revelation. And clearly, the authors weren’t convinced that the drug was the culprit and not something else.
Wilner: So you’re making an argument, I think, for personalized medicine, because patients may respond quite differently to the same antiseizure drug. Is that right?
Taraschenko: I think it’s a very interesting idea. We clearly don’t know about the pathophysiology for the majority of these side effects. We do know that there is valproate-induced parkinsonism that interferes with their GABAergic neurotransmission in the basal ganglia. And some of the receptors are being affected by gabapentin and that results in myoclonus. We clearly don’t know what’s causing it. I think it does make a lot of sense to pay attention to individual comorbidities and also for coadministered medications. As we know, some other drugs, such as the antihistamines or antidepressants, can precipitate that independently.
Wilner: You mentioned that parkinsonism could be worsened by valproic acid. Are there any other clear-cut relationships where you might want to avoid a particular epilepsy drug?
Taraschenko: We didn’t see any other clear comorbidities; we didn’t take note of them in this particular study. There were a few patients with Huntington’s chorea and Parkinson’s disease. But I think it would make sense to avoid drugs that frequently came up with dyskinesia or myoclonus in patients already suffering from abnormal movements in their extremities that interfere with their daily activities — for example, hand myoclonus and gabapentin, or even restless leg syndrome in association with topiramate.
Wilner: One of the risks of using antipsychotic drugs is tardive dyskinesia, which can be permanent. Were these movement disorders related to antiseizure drugs permanent, or were they just temporally related to the dose?
Taraschenko: The majority of side effects resolved following either cessation or adjustment of the dose, with the two exceptions being phenytoin and carbamazepine. We’ve seen with phenytoin that only about 80% resolved completely, but only 20% improved. And it did take quite a while for phenytoin. Most patients had to wait about 3 weeks to get better, which was quite interesting. The literature was not so impressive. We just didn’t find too many reports or newer medications. But for the large majority of the side effects, they were reversible with discontinuation of the drug.
Wilner: One more question, and this may not have been reflected in your study. Were you able to associate any of these movement disorders with seizure or epilepsy type? For example, patients with generalized epilepsy or focal epilepsy or temporal epilepsy — were they more or less susceptible to a particular drug?
Taraschenko: No, we didn’t take note of the epilepsy syndrome. As you can imagine, sifting through cases and case reports is a rather tedious task. We were happy to extract the information that we were aiming to get. We did not separate them, but one of the future projects that we are thinking of is to focus on abnormal movements in patients with refractory epilepsy, presurgical, or even surgical epilepsy. We would look at patients at our center and retrospectively look at their charts and see if there is anything about polytherapy, perhaps separating the syndromes that way.
Wilner: Pre and postsurgical cases would be interesting. Also, if the movement disorder disappears when the damaged brain is extracted, that would certainly be of interest. Well, Dr Tarasenko, this has been great. Thank you for sharing the results of your paper that was presented at the American Epilepsy Society meeting.
Taraschenko: Thank you for inviting me.