Changes in sleep architecture following ischemic stroke have been poorly investigated. Our objective was to explore changes of sleep structure in patients with ischemic stroke or transient ischemic attack in order to verify a possible predictive value of sleep with respect to clinical outcome.
Patients recruited in the prospective SAS-CARE study received two polysomnographies (PSG) in the acute and chronic phases after stroke/TIA. Sleep parameters were compared between the two time-points and matched with a non-stroke population randomly selected from the HypnoLaus cohort.
Of the 169 patients investigated with PSG in the acute phase, 104 were again studied 3 months after stroke symptom onset and compared with 162 controls. The acute phase of stroke/TIA was associated with sleep disruption, which significantly improved in the chronic phase, but remained worse than controls (total sleep time improve from 318.8 ± 90.8 to 348.4 ± 81.5 min, compared to 388.2 ± 71.3 in controls, sleep latency from 49.9 ± 58.4 to 27.9 min, compared to 20.2 ± 22 in controls, sleep efficiency from 58.2 ± 18.1% to 27.9 ± 36.4 min, compared to 83.4 ± 10.3% in controls, wakefulness after sleep onset percentage from 36.5 ± 17.3 to 29.3 ± 15.6, compared to 13.2 ± 9.2 in controls). The percentage of REM sleep was negatively associated with stroke severity, whereas stroke topography did not correlate with sleep parameters.
This study confirmed a severe sleep disruption in the acute phase of stroke. Although a significant improvement of sleep quality was observed during the three months after stroke, sleep architecture did not normalize. In particular, sleep efficiency and REM sleep seem to be particularly affected by stroke in the acute phase, with a relative preservation of NREM sleep. We suggest that these sleep architecture changes represent a persistent marker of brain damage due to stroke. Further studies are needed to assess the relationship with stroke topographic and outcome.