Relationship between autonomic arousals, EEG arousals, respiratory events and subjective sleepiness in patients undergoing investigation for obstructive sleep apnoea

Daytime hypersomnolence is an important symptom of the sleep apnoea/hypopnoea syndrome and the most compelling reason for treatment. Measuring sleep fragmentation is thus critical in a respiratory sleep study. Sleep fragmentation can be estimated from respiratory events (apnoeas, hypopnoeas, dips in oxygen saturation), EEG arousals, and autonomic changes (heart rate and blood pressure rises). We have compared these different approaches in 40 patients having polysomnography for possible sleep apnoea, as well as their correlation with subjective sleepiness (Epworth Sleepiness Score, ESS). Respiratory signals were scored for apnoeas / hypopnoeas (AHI) and dips in arterial oxygen saturation of >4% (Dip rate); EEG was scored by the American Sleep Disorders Association criteria for 3 second arousals (ASDA); autonomic measures were heart rate rises of > 10 bpm (HR) and indirect blood pressure rises (pulse transit time, PTT, Clin Sci 1994;87:269). The number of events/hour was used to calculate the correlation between each of these and their correlation with ESS (P < 0.05 for all correlations displayed). Data analysis is not yet complete, n > 21 for all correlations. HR ASDA AHI Dip rate ESS PTT r = 0.87 r = 0.91 r = 0.72 r = 0.87 r = 0.38 HR r = 0.73 r = 0.53 r = 0.80 ASDA r = 0.92 r = 0.92 r = 0.39 AHI r = 0.88 Dip rate r = 0.49 Respiratory events, EEG arousals and autonomic arousals are closely correlated. Dip rate is the best predictor of ESS. PTT measured blood pressure rises may provide a useful automated alternative to manual scoring of EEG arousals as an index of sleep fragmentation.