Association Among Melatonin Dysregulation, Sleep Disorders, and Sepsis-Associated Delirium: A Prospective Cohort...

Sepsis-associated delirium (SAD) is a life-threatening complication in the intensive care unit (ICU). Although sleep disturbances are common in sepsis, their relationship with SAD remains poorly understood. We investigated the link between sleep disturbances, melatonin dysregulation, and SAD.

In this prospective cohort study, we longitudinally assessed 99 patients with sepsis. Sleep quality was evaluated using the Richards-Campbell Sleep Questionnaire (RCSQ) and melatonin concentration, and multivariable logistic analysis, propensity score matching (PSM), inverse probability weighting (IPW), mixed-effects models, Bayesian, and mediation analysis to adjust for confounding factors to evaluate the relationship of melatonin with sleep disorder and SAD.

Sleep disorders preceded SAD onset with a duration-dependent effect (P < 0.001) and were associated with a 1.45-7.69-fold higher risk of delirium (P < 0.05) across multivariable logistic analysis, PSM, IPW and Bayesian analysis. Lower melatonin concentrations were correlated with worse multidimensional sleep impairment (β = 7.65-13.10, all P < 0.05), particularly impaired sleep continuity (β = 9.81) and poorer overall sleep quality (β = 13.10). Patients with SAD exhibited biphasic melatonin suppression: initial decline coinciding with sleep disturbances (P < 0.001), followed by further reduction during active delirium episodes (P < 0.005), independent of sedation exposure (P > 0.05). Mediation analysis demonstrated substantial mediating effects of melatonin concentrations measured at sleep disturbance onset (ADE = 0.60, P < 0.001), and melatonin reduction occurring alongside sleep disturbances was associated with SAD development [average causal mediation effects (ACME) =  - 0.110, P = 0.024].

This study provides preliminary evidence that disrupted sleep patterns and altered melatonin secretion may contribute to the development of SAD in ICU patients. While our findings support melatonin not only as a risk factor for sleep disruption and SAD, but also as a biologically plausible candidate linking sleep disruption to SAD, the potential influence of other factors underscores the need for further research to confirm this relationship.