Sleeping position during unattended home polysomnography compared to habitual sleeping position and the potential impact on measured sleep apnea severity.

Introduction: Obstructive sleep apnoea (OSA) commonly manifests with greater severity in the supine position. Over 60% of patients with OSA have supine predominant OSA (supine apnoea-hypopnoea index [AHI] greater than twice of non-supine AHI) and 25% have supine isolated OSA (non-supine AHI< 5 events/hour). Therefore, for the majority of patients, the proportion of total sleep time in the supine body position necessarily impacts on the severity of OSA assessed by polysomnography (PSG). We previously reported that the proportion of time in the supine position is significantly higher during laboratory PSG compared to habitual sleep. Here we report the interim results of our current study comparing sleeping position during unattended home PSG to habitual sleeping position. Material(s) and Method(s): This is a prospective observational study of patients referred for unattended PSG through our tertiary academic sleep unit. We use the Night ShiftTM positional therapy device to record body position during the unattended PSG and subsequently for up to three additional nights to sample habitual sleeping position. The Night ShiftTM was programmed to record only; the feedback intervention function was disabled. A paired t-test was used to compare the mean difference in proportion of sleep time in the supine position during unattended PSG compared to habitual sleep. We then calculated a modified AHI "corrected" for habitual sleeping position. Night-to-night variability in habitual sleeping position was assessed using a correlation matrix, and a repeated measures one-way ANOVA. Result(s): To date, fifty-two patients have been recruited, out of a target of 79 patients. 38.5% female, age 47 +/- 12.8 years, body mass index 30.9 (range: 27.5-37.5) kg/m2. Median total AHI 19 (range: 3.2-27.9) events/hour, supine AHI 22 (0.9-45.3) events/hour, non-supine AHI 9.4 (1.3-24.9) events/hour. Fifteen patients (29.4%) had supine predominant OSA and 1 (2%) had supine isolated OSA. Proportion o