Abstracts

Sleep disruption is common in patients with epilepsy, and anticonvulsant drugs (AEDs) may have beneficial or detrimental effects on sleep. Insomnia is a common problem with diverse etiologies, however can be caused by the use of some AEDs. This study evaluates a large population of patients in order to correlate insomnia with specific AEDs. The Columbia AED database includes detailed information on patients taking virtually all marketed AEDs, alone or in combination. Data have been compiled on all outpatients seen from January 1, 2000 to January 1, 2004. All 1190 patients age 16 and over were included. For this study, we specifically examined the incidence of insomnia as reported by patients or physicians during routine office visits and correlated with use of individual AEDs. Insomnia resulting in AED dose reduction or discontinuation was also examined. Due to multiple comparisons, p [lt] .005 was considered significant. We also looked at predictors of insomnia indepedent of AED use, including gender; epilepsy classification; presence of cognitive, psychiatric, or sedative side effects; and history of epilepsy surgery. A total of 818 patients were included who had been newly started on one of 14 AEDs, either as monotherapy or in combination with other AEDs. Insomnia was seen most commonly with felbamate (18%); this was significantly more common when compared with all other drugs. Insomnia was less commonly seen with lamotrigine and zonisamide (3%), topiramate (2%), and levetiracetam and gabapentin (1%) however these were not significant. Drugs not showing insomnia were clobazam, carbamazepine, oxcarbazepine, phenobarbital, phenytoin, tiagabine, vigabatrin, and valproate. When analysis was performed only on patients taking each drug in monotherapy, results were similar (no patients took felbamate in monotherapy). Only felbamate was significantly associated with insomnia causing a dose change (14%), and no AED was associated with insomnia resulting in drug discontinuation. Using multivariate analysis, insomnia (occurring with any drug) was associated with psychiatric side effects and sedation and occurred more commonly in patients who had epilepsy surgery or juvenile myoclonic epilepsy. Insomnia is significantly associated with felbamate use, and is seen rarely with several other AEDs. Except with felbamate, insomnia very rarely is a reason for dose adjustment or AED change. Association of insomnia with sedation may be directly related to reduced sleep; previous investigations also show sedation with drugs that do not cause insomnia (clobazam, phenytoin and phenobarbital). It is not surprising that psychiatric adverse effects correlated with insomnia, as depression and anxiety are commonly seen in this population and are known to be associated with sleep disruption. Further studies will be needed to determine optimal treatment for patients with insomnia related to AED use. (Supported by GlaxoSmithKline, Novartis, OrthoMcNeil, Pfizer, Elan, UCB Pharma.)