Propofol-Facilitated Recovery Sleep Modulates Cognitive Function via Slow Oscillation-Spindle Coupling After Acute...

Sleep deprivation (SD) is common but underrecognized in intensive care unit (ICU) patients and contributes to delirium and long-term neurocognitive deficits. Propofol is widely used for titratable ICU sedation and has potential neuroprotective effects, but the mechanisms underlying its cognitive benefits during recovery sleep remain unclear. Evidence shows that coupling between slow oscillations (SO) and spindles during sleep is associated with memory consolidation processes. We hypothesized that propofol-facilitated recovery sleep after acute SD would strengthen SO-spindle coupling aligned to the SO up state and thereby improve cognition.

Male Sprague-Dawley rats (8-12 weeks, total n = 67) were assigned to control (CON), SD without intervention (SDN), spontaneous recovery (SDS), or propofol-supported recovery (SDP). All SD groups underwent 48 h of total deprivation. During recovery, SDP received propofol (20 mg·kg^-1·h^-1, 6 h), whereas SDS recovered naturally. Spatial learning and memory were tested by the Barnes maze (BM); recognition memory by 2-h and 24-h novel object recognition (NOR) for short- and long-term recognition memory. Sleep electroencephalography (EEG) and electromyography (EMG) were recorded for 12 h during recovery in SDS and SDP. We quantified SO and spindle parameters and SO-spindle phase coupling. Separate animals were analyzed for cortical Na⁺-K⁺-2Cl^- cotransporter 1 (NKCC1)/K⁺-Cl^- cotransporter 2 (KCC2) expression ratio.

Sleep deprivation impaired BM and NOR performance. Propofol during recovery restored spatial performance and 24-h recognition retention. Propofol also increased SO density compared with SDS (11.13 ± 0.14 vs. 9.49 ± 1.26 events/min, p = 0.040) and improved phase-accurate SO-spindle alignment, with higher phase-vector projection (Vproj; 0.30 ± 0.11 vs. -0.33 ± 0.37, p = 0.036), increased up-phase events (0.38 ± 0.01 vs. 0.16 ± 0.10, p = 0.036), and significant up-state preference only in SDP (V-test, p = 0.035). Molecularly, propofol decreased the NKCC1/KCC2 ratio by reducing NKCC1 and increasing KCC2.

Propofol-facilitated recovery sleep supports spatial and recognition memory, accompanied by up-state-precise SO-spindle coupling and reduced NKCC1/KCC2. These findings suggest SO-spindle alignment may serve as a candidate EEG marker of recovery sleep and highlight chloride-transport-related pathways as therapeutic targets for SD-related cognitive impairment.