Abstracts

Rationale: Small series suggest that sleep apnea increases the risk of seizures in people with epilepsy; whereas its treatment improves seizure control. The prevalence of obstructive sleep apnea (OSA) measured by polysomnography (PSG) is higher in people with epilepsy, especially medically refractory cases, than the general population (Malow et al., 2000; Manni et al., 2003). However, studies involving a more diverse epilepsy population have not been performed and the role of traditional OSA predictors (age, gender, body mass index [BMI]) has not been thoroughly investigated. Methods: This is a cross-sectional study involving adult epilepsy patients recruited from a tertiary care epilepsy clinic. Consecutive patients without prior sleep disorder diagnoses were invited to participate. Subjects completed a series of questionnaires including the Epworth Sleepiness Scale (ESS), maintained sleep diaries and underwent PSG with expanded EEG. OSA was defined as an apnea-hypopnea index (AHI) > 5. The association between presence of OSA and AHI with variables including age, gender, BMI, neck circumference, ESS, mean monthly seizure frequency and number of antiepileptic drugs (AEDs) were analyzed using unpaired t-tests and chi-square tests depending on data distribution. Regression models were fit to test the independent effects of these variables on predicting the presence of OSA and AHI. Results: 132 treated subjects (mean age 38.7 13.2 yrs; BMI 28.9 7.3; male gender 44%) were included. Subjects took a mean of 1.6 (1-4) AEDs and 48.5% were treated with monotherapy. OSA was found in 55 (41.7%) subjects, with 26.5% having mild (AHI 5-<15) and 15.1% having moderate to severe OSA (AHI >15). Age-specific prevalence of OSA was 17.1%, 45.9% and 66.7% in the <30, 30-50 and >50 year-old groups, respectively. No difference was found in overall OSA prevalence in female (39.2%) and male (44.8%). subjects. In subjects over 50 years, OSA prevalence was higher in females (72.2% vs. 58.3%; Figure 1). The distribution of OSA severity was comparable between genders (Figure 2). The prevalence was doubled in obese (BMI >30) compared to non-obese (BMI<30) groups (60.4 vs. 31%) overall, although for subjects over 50 years of age, the prevalence was comparable (68.8 vs. 64.3%). Multiple logistic regression analysis identified age and BMI as significant predictors for OSA. Conclusions: Our study demonstrates a high prevalence of OSA in a diverse group of epilepsy patients, markedly exceeding that of general population. As in the general population, the prevalence increases with age and BMI. However, in contrast to general population studies, women with epilepsy appear to be at a similar risk as men. Further studies are needed to elucidate the mechanisms underlying the increased risk of OSA in epilepsy patients. These findings support the routine screening for OSA in all people with epilepsy given its potential impact on seizure control and quality of life.