Preoperative Sleep Disruption and Postoperative Delirium

Study Objectives: To describe preoperative and postoperative sleep disruption and its relationship to postoperative delirium. Design: Prospective cohort study with 6 time points (3 nights pre-hospitalization and 3 nights post-surgery). Setting: University medical center. Patients: The sample consisted of 50 English-speaking patients >= 40 years of age scheduled for major non-cardiac surgery, with an anticipated hospital stay >= 3 days. Interventions: None. Measurements and results: Sleep was measured before and after surgery for a total of 6 days using a wrist actigraph to quantify movement in a continuous fashion. Postoperative delirium was measured by a structured interview using the Confusion Assessment Method. Sleep variables for patients with (n = 7) and without (n = 43) postoperative delirium were compared using the unpaired Student t-tests or chi(2) tests. Repeated measures analysis of variance for the 6 days was used to examine within-subject changes over time and between group differences. The mean age of the patients was 66 +/- 11 years (range 43-91 years), and it was not associated with sleep variables or postoperative delirium. The incidence of postoperative delirium observed during any of the 3 postoperative days was 14%. For the 7 patients who subsequently developed postoperative delirium, wake after sleep onset (WASO) as a percentage of total sleep time was significantly higher (44% +/- 22%) during the night before surgery compared to the patients who did not subsequently developed delirium (21% +/- 20%, p = 0.012). This sleep disruption continued postoperatively, and to a greater extent, for the first 2 nights after surgery. Patients with WASO < 10% did not experience postoperative delirium. Self-reported sleep disturbance did not differ between patients with vs. without postoperative delirium. Conclusions: In this pilot study of adults over 40 years of age, sleep disruption was more severe before surgery in the patients who experienced postoperative delirium. A future larger study is necessary to confirm our results and determine if poor sleep is associated with delirium in larger samples and what specific sleep problems best predict postoperative delirium in older surgical patients.