Presurgical Sleep-Wake Disturbance and Postoperative Delirium: A Prospective Cohort Study.

OBJECTIVES: Prior studies on the relationship between presurgical sleep-wake disturbance and postoperative delirium have been restricted to subjective or indirect sleep measures. This study evaluated whether objective measures of presurgical sleep-wake disturbance are associated with postoperative delirium. DESIGN: Prospective study. SETTING: Single site, tertiary care medical center. PARTICIPANTS: Adults aged 40 years or older having surgical aortic valve replacement. MEASUREMENTS: Before surgery, we assessed sleep using self-report, ambulatory polysomnography, and actigraphy. Other measures include sociodemographics, medical-surgical characteristics, and neuropsychological battery. After surgery, we assessed delirium and delirium severity. RESULTS: The 100 participants had a mean age of 66.9 years (64% male). Twenty-seven subjects developed delirium. After adjusting for age, sex, comorbidity, and surgery duration, we found that each 10% decrease in presurgical sleep efficiency was associated with more than double the odds of delirium (OR 2.16, 95% CI [1.10, 3.71]), as was each 10% decrease in presurgical relative amplitude of rest-activity rhythms (2.59 [1.34, 5.23]). Lower sleep efficiency and relative amplitude were also associated with greater delirium severity (standardized |β| = 0.39 and 0.42, respectively). Apnea-hypopnea index ≥15 events/h was associated with higher odds of delirium (OR 3.4 [1.01, 11.7]) but not delirium severity. Mild cognitive impairment and lower global cognitive function were also associated with higher odds of delirium and greater delirium severity. CONCLUSIONS: Lower presurgical sleep efficiency, apnea-hypopnea index ≥15 events/h, and weaker rest-activity rhythms were associated with a higher odd of postoperative delirium suggesting a potential role for targeting these sleep-wake disturbances to mitigate delirium.