Associations Between Obstructive Sleep Apnea, Macro-Scale Glymphatic Coupling, and Episodic Memory Impairment in
Qianqian Gao1, Boyu Zhang2, Haixia Mao1
1Department of Radiology, The Affiliated Wuxi People's Hospital of Nanjing Medical University, Wuxi People's Hospital, Wuxi Medical Center, Nanjing Medical University, Wuxi, People's Republic of China.
Background:
Sleep disorders are highly prevalent among middle-aged and older adults and represent a crucial modifiable risk factor for cognitive decline. Impaired brain waste clearance is hypothesized to underlie this association, but clinical evidence remains limited. We utilized global blood-oxygen-level-dependent and cerebrospinal fluid (gBOLD-CSF) coupling as a non-invasive macro-scale surrogate of fluid clearance dynamics to investigate its associations across distinct clinical phenotypes, including obstructive sleep apnea (OSA), insomnia, and comorbid insomnia and sleep apnea (COMISsA) and to delineate how objective respiratory metrics and subjective insomnia severity differentially relate to domain-specific cognitive impairment.
Methods:
A total of 143 adults were enrolled and categosrized into four groups: healthy controls (HCs, n = 30), obstructive sleep apnea (OSA, n = 46), insomnia (n = 32), and comorbid insomnia and sleep apnea (COMISA, n = 35). Participants underwent home sleep apnea testing, comprehensive neuropsychological assessments, and 3.0 T magnetic resonance imaging. gBOLD-CSF coupling strength was quantified by computing the peak negative cross-correlation between the global cortical BOLD signal and the signal extracted from the CSF regions at the bottom slice of the fMRI acquisition.
Results:
Compared with HCs, individuals with sleep disorders exhibited significantly reduced gBOLD-CSF coupling strength (all Bonferroni-corrected p < 0.05). This attenuation correlated with advancing age (r = 0.185, p = 0.027) and greater global sleep disturbances (as indexed by higher PSQI total scores, r = 0.261, FDR-corrected p = 0.012). Using the entropy balancing (EBAL) weighted dataset, structural equation modeling revealed that in participants with sleep apnea (OSA and COMISA), reduced gBOLD-CSF coupling strength statistically mediated the relationship between nocturnal hypoxia and episodic memory deficits. Specifically, elevated apnea-hypopnea index (AHI), oxygen desaturation index (ODI) and the percentage of total sleep time with oxygen saturation below 90% (T90) were associated with a positive shift in coupling values toward zero (all a coefficients > 0, p < 0.05), representing weakened glymphatic coupling strength. This diminished coupling, in turn, predicted poorer delayed episodic memory (AVLT-20 min delayed recall; indirect effects: AHI, -0.029; ODI, -0.079; T90, -0.111). Notably, the direct effects of these hypoxic parameters on memory consolidation were non-significant. In contrast, no significant glymphatic mediation pathways were identified for the insomnia-executive function model, even after adjusting for depressive burden.
Conclusion:
In summary, gBOLD-CSF coupling characteristically tracks distinct sleep disorder subtypes, showing significant reductions in middle-aged and older adults with OSA, insomnia, and COMISA. Our findings suggest that respiratory event-related hypoxic stress is associated with increased susceptibility to delayed episodic memory deficits in relation to glymphatic impairment, highlighting that cross-disciplinary sleep interventions may represent a promising strategy for preserving brain clearance and mitigating cognitive aging.
