Circadian Rhythm in End-Of-Life Delirium: A Secondary Analysis of Two Randomized Controlled Trials.

Abstract

The circadian pattern of restlessness in end-of-life delirium is understudied and poorly understood. To examine the timing of breakthrough restlessness in patients with advanced cancer and end-of-life delirium admitted to an acute palliative care unit. This is a secondary analysis of two randomized clinical trials that examined the effect of lorazepam (MAD trial) and neuroleptics (CHAD trial) in end-of-life delirium. In this study we examined the frequency of restlessness in 8-hour intervals (7 AM-3 PM, 3-11 PM, 11 PM-7 AM). Breakthrough restlessness was measured based on 1) rescue medications (neuroleptics or benzodiazepines) administered for breakthrough restlessness and 2) a Richmond Agitation-Sedation Scale (RASS) score ≥+1 (collected every 2-4 hours). This study included 128 patients (58 from MAD trial, 70 from CHAD trial); the mean age (SD) was 64 (12.5), and 57 (44.5%) were women. We found that 3-11 PM was significantly associated with greater rescue medication use in univariate analysis for both trials (MAD: Estimate: 0.35, 95% CI: 0.23-0.48, P < 0.001; CHAD: Estimate: 0.1, 95% CI: 0.07-0.12, P < 0.001). This association remained significant in multivariate analysis for CHAD (Estimate: 0.1, 95% CI: 0.07-0.12, P < 0.001). About 3-11 PM was also associated with greater episodes of RASS ≥+1 in MAD in univariate and multivariate analysis (Estimate:0.31, 95 % CI: 0.21-0.42, P < 0.001). Delirious patients were more restless between 3 PM and 11 PM. This observation of "sundowning" may help clinicians to better anticipate this symptom, schedule monitoring and treatments, and educate patients and caregivers.