Abstracts

Rationale: Forty five to 55% of subjects with epilepsy have insomnia, with sleep maintenance insomnia being more predominant than sleep onset insomnia. Data characterizing insomnia patterns are lacking in children with epilepsy. Hence, we performed this study to identify the factors associated with the severity of insomnia. Methods: Database search identified children with insomnia and epilepsy who had polysomnography performed. Charts were reviewed for sleep and epilepsy characteristics. Sleep efficiency (SE) and wakefulness after sleep onset (WASO) on polysomnography were used as outcomes to assess severity of insomnia. Wilcoxon Rank Sum test was performed to identify differences in sleep outcomes between focal and generalized groups, and groups based on presence or absence of electroencephalogram (EEG) and magnetic resonance imaging (MRI) abnormalities, developmental delay, hypnotic medication use and seizure control. Spearman correlations were performed for outcomes with EEG spike frequency and number of antiepileptic drugs (AEDs). Multiple regression analyses, both traditional and robust, were also performed to compare the WASO and SE, individually, with age, developmental state, seizure control, EEG results and MRI results. Results: Forty six subjects met the study criteria, with a mean age 10.2 (4.5) (standard deviation) and 27 (59%) females. Twenty one (45%) subjects had focal epilepsy. Thirty five (76%) subjects had developmental delay. EEG and MRI abnormalities were present in 23 (50%) and 17 (37%) subjects respectively. Twenty six subjects were on one AED, fifteen on 2 AEDs, three on 3 and one was on 4 AEDs. Only one subject was not on any AEDs. Twenty nine subjects (63%) were taking hypnotic or similar medications, most common was melatonin in 25 (86%) subjects. The others were on clonidine, trazodone, eszopiclone, zolpidem and zaleplon. Table 1 shows sleep parameters. The children with generalized epilepsy had lower SE and higher WASO as compared to focal epilepsy [mean difference 9.9, standard deviation (SD) 11.7, p=0.019; mean difference 49.7, SD 58.8, p= 0.025)]. Similarly, children with developmental delay had lower SE and higher WASO as compared to without developmental delay [mean difference 8.5, SD 11.9, p=0.024; mean difference 33.8, SD 56.1, p=0.020], respectively. No significant differences were seen in outcomes for groups based on seizure control, use of hypnotics, and EEG or MRI abnormalities. No correlation was identified between outcomes and the number of AEDs or EEG spike frequency. Multiple regression analysis showed a trend towards association between developmental delay and SE. When controlled for age, developmental delay was significantly associated with lower SE and higher WASO. Conclusions: In our cohort, children with generalized epilepsy had higher severity of insomnia as well as subjects with developmental delay. Seizure control, EEG and MRI abnormalities, and the number of AEDs were not contributors for severity of insomnia. Larger prospective studies are needed to identify the association between the severity of insomnia and seizure control. Funding: none