Evidence for OSA-Dependent Vulnerability to Sleep Restriction and Alcohol: Simulated Driving Performance [Oral Presentation]

Introduction: Driver drowsiness is an important under-recognised factor contributing up to 33% of motor vehicle accidents (MVAs) withan annual cost of up to $38 billion in the US. OSA is an important factor,with patients showing a 2–7 fold increased risk of MVA. Sleepiness isincreased by sleep restriction and exacerbated by alcohol resulting inpoor driving performance in healthy young drivers. However, vulnerability to these insults in middle-aged moderate-severe OSA patients hasnot been investigated. We compared simulated driving performance inthe mid-afternoon in moderate to severe, middle aged OSA patients andhealthy age matched controls following sleep restriction and low-dosealcohol. We hypothesised that OSA patients would show greater decrements in performance under these conditions.Methods: All participants completed a 90-minute mid-afternoonmonotonous simulated driving task (AusEd) following baseline (8h),restricted (4h) sleep, and alcohol (BAC 0.05 g/dL) conditions. Drivingsimulator data from 22 OSA patients (mean[sd]: age = 52.8[12.5]y;BMI = 36.3[9.2]kg/m2; RDI = 56.5[20.7]/hour), and 17 healthy controls (age = 49.7[13.3]; BMI = 24.3[2.5]; RDI = 8.2[4.1]/hour) wereexamined using a mixed model analysis.Results: There were significant group (P < 0.001), condition(P < 0.001), time (P < 0.001) and group ¥ condition interaction effects(p = 0.012) in steering deviation. Decrements in performance becamemore apparent after 30–40 minutes on task. More patients than controlscrashed following sleep restriction (p = 0.009) and alcohol (p = 0.02).Compared to baseline condition, breaking reaction time was slower inboth groups following sleep restriction and alcohol (p = 0.016) but wasnot different between groups.Conclusions: Compared to controls, OSA patients showed increasedvulnerability to sleep restriction and alcohol on their ability to maintainlane position. Under these conditions a greater number of patients hada complete performance failure (crash) compared to controls. Brakingreaction time (although slower following sleep loss and alcohol) was notdifferent between the two groups, suggesting that OSA patients are morevulnerable to sleep restriction and alcohol effects during sustained,vigilance-demanding, but not reaction tasks. Our findings support theneed to advise untreated moderate-severe OSA patients to avoid furthersleep loss and alcohol when driving for extended periods.Support: NH&MRC project grant #390400, AusEd driving simulatordevelopers.