Article Text
Abstract
Aim To define prevalence of sleep-related breathing disorder (SRBD) as measured by Paediatric Sleep Questionnaire (PSQ) and Excessive Daytime Sleepiness (EDS) as measured by modified Epworth sleepiness scale (ESS) in a cohort of children with epilepsy compared with healthy controls.
Methods Consecutive epileptic children attending for overnight video telemetry or outpatient clinics were recruited along with non-epileptic controls from paediatric surgical clinics in a tertiary children hospital setting. Additional information was collected on use of anti-epileptic drugs (AED). Those with additional neurological or respiratory problems were excluded. Parents were asked to complete both PSQ and modified ESS on behalf of their children. A score of >0.33 on the PSQ-SRBD scale was considered abnormal, and scores of >10 on the ESS were considered significant. Data were entered into Microsoft Excel database and analysed using SPSS v19.0. Non-parametric statistical comparison was made between groups by Mann-Whitney U-test (MWU).
Results Seventy-five children were studied, 33 of whom had a diagnosis of epilepsy median (IQR) age 9 (5–12) years, along with 42 controls of median (IQR) age 6 (4–8.5) years. 49% epileptic subjects had an abnormal PSQ-SRBD score compared with 5% controls. The median (IQR) PSQ-SRBD score in the epilepsy group was 0.36 (0.22 to 0.46) versus 0.17 (0.09 to 0.23) in controls (p < 0.001, MWU). PSQ sub-scale scores were also significantly higher for snoring (p = 0.04), daytime sleepiness (p < 0.001), and inattention/behaviour (p = 0.007) in the epilepsy group compared with controls. Additionally, 30% epileptic subjects had an ESS >10 compared with 5% controls. Median (IQR) ESS score in epileptic subjects was 5 (1 to 10) versus 1 (0–3.25) in controls (p = 0.001, MWU). Within epileptic subjects, PSQ-SRDB (but not sub-scale scores) and ESS were higher in those on AED (p = 0.001 and p = 0.48 respectively).
Conclusion This study suggests a higher prevalence of symptoms of SRBD and EDS in children with epilepsy compared with healthy children. Anti-convulsant therapy may be a confounding factor, but does not alone account for the associations seen. Further studies which should include polysomnography to verify the presence (rather than suggestion by questionnaire) of SRBD are warranted.