The estimation of excessive daytime sleepiness in post-stroke patients - a polysomnographic study

Excessive daytime sleepiness (EDS) has been reported in stroke patients. EDS in acute stroke was studied repeatedly, but there is a modest amount of data in post-stroke patients. The aim of this study was to assess the frequency of EDS and characterize sleep architecture in patients >3 months after stroke and identify factors which may affect EDS. 66 patients were enrolled, of which 33 had experienced stroke. All underwent a standardized overnight, diagnostic single night polysomnography, including electrocephalogram (EEG) leads, electrooculograms (EOG), chin electromyogram (EMG), and electrocardiogram (ECG). Epworth Sleepiness Scale (ESS) was used to measure subjects' level of daytime sleepiness. We observed similar total ESS score, total sleep time (TST), sleep efficiency, as well as respiratory disturbance index /apnea-hypopnea index (RDI/AHI), oxygen desaturation index (ODI) and mean heart rate in both groups. We observed positive linear correlation between EDS and mean heart rate in the stroke group (r = 0.46, p < 0.05) as well as between EDS and REM duration (r = 0.23, p < 0.05). In the non-stroke group EDS didn't correlate with the heart rate or with the REM duration. In the non-stroke group EDS correlated positively with RDI/AHI and ODI index (r = 0.46; p = <0.05 r = 0.41, p < 0.05 and maximal desaturation (r = 0.55, p < 0.05), this correlation was not observed in post-stroke group. In the both groups we observed negative linear correlation between BMI and saturation (stroke group - mean as well as minimal saturation (r = -0.458, p < 0.05; r = -0.578, p < 0.05), non stroke group - minimal but not mean saturation rate (r = -0.544, p < 0.05). We also noticed in both groups positive correlation between BMI and both AHI and ODI index (stroke group respectively: r = 0.430, p < 0.05; r = 0.451, p < 0.05; non-stroke group - BMI and ODI: r = 0.405, p < 0.05). The positive BMI correlation with RDI/AHI was not significant in the non-stroke group. We noticed that BMI correlates with non-REM sleep N1 (r = 0.760, p < 0.05) in post-stroke group, while it correlates negatively with REM sleep (r = -0.709, p < 0.05). There was no such correlation in the non-stroke group. In summary, in stroke patients subjective daytime sleepiness is associated with heart rate, but not with the severity of OSA. Thus ESS (Epworth Scale Score) may be not as useful as a marker of obstructive sleep apnea (OSA) presence or severity in post stroke patient as in the general population.