Objective: To assess the prevalence of parasomnias in relation to presence and severity of obstructive sleep apnea (OSA). We hypothesized higher parasomnia prevalence with higher OSA severity.Methods: The sample comprised 4,372 patients referred to a Norwegian university hospital with suspicion of OSA (mean age 49.1 years, 69.8% males). OSA was diagnosed and categorized by standard respiratory polygraphy (type 3 portable monitor). The patients completed a comprehensive questionnaire prior to the sleep study, including questions about different parasomnias during the last 3 months. Pearson chi-square tests explored differences according to the presence and severity of OSA. Furthermore, logistic regression analyses with the parasomnias as dependent variables and OSA severity as predictor were conducted (adjusted for sex, age, marital status, smoking, and alcohol consumption).Results: In all, 34.7% had apnea-hypopnea index (AHI) <5 (no OSA), 32.5% had AHI 5-14.9 (mild OSA), 17.4% had AHI 15-29.9 (moderate OSA), and 15.3% had AHI ≥30 (severe OSA). The overall prevalence of parasomnias was 3.3% (sleepwalking), 2.5% (sleep-related violence), 3.1% (sexual acts during sleep), 1.7% (sleep-related eating), and 43.8% (nightmares). The overall parasomnia prevalence was highest in the no OSA group. In the chi-square analyses, including all OSA groups, the prevalence of sleep-related violence and nightmares were inversely associated with OSA severity, whereas none of the other parasomnias were significantly associated with OSA severity. In adjusted logistic regression analyses the odds of sleepwalking was significantly higher in severe compared to mild OSA (OR = 2.0, 95% CI = 1.12–3.55). The other parasomnias, including sleep-related violence and nightmares, were not associated with OSA presence or severity when adjusting for sex and age.Conclusions: We found no increase in parasomnias in patients with OSA compared to those not having OSA. With the exception of sleepwalking, the parasomnias were not associated with OSA severity.
Read full abstract