A METHOD FOR OBJECTIVE ASSESSMENT OF REST ACTIVITY PATTERNS AND LIGHT LEVELS IN THE ICU

Abstract

TOPIC: Critical Care TYPE: Original Investigations PURPOSE: Patients in the intensive care unit (ICU), experience disturbed sleep due to environmental factors, as well as critical illness itself. Insufficient and fragmented sleep is linked to increased morbidity, mortality, and ICU length of stay. Despite these correlations, objectively measured sleep and circadian patterns are not routinely examined in ICU patients, and assessment methods are not standardized. In this study, we devised a method to use wrist actigraphy to examine rest-activity patterns and light in a medical ICU. METHODS: As part of an ongoing randomized placebo-controlled trial testing the effect of a sleep medication in preventing ICU delirium, we enrolled patients admitted to the medical ICU at Rhode Island Hospital. Patients were monitored with continuous actigraphy while they were in the ICU. We estimated sleep and wake durations using algorithms validated with polysomnography as well as maximum and average light levels. We summarized data for 2 defined periods across 24-hours based on timing of the investigational sleep aid and ICU routines in our hospital. “Subjective night” was defined from the time of study drug administration to 5:59 AM, and “subjective day” was defined from 6:00 AM to the next study drug treatment time. We used Wilcoxon Signed Ranks Tests to compare sleep and light measured between subjective night and day. RESULTS: Fifty-eight patients are included in these analyses, 45% of which are women. The mean age was 61 years, and the mean APACHE II score was 21. The mean actigraphy-recording days was 4.3. Median duration of subjective night was 513 minutes (IQR=472-526), and of subjective day was 708 minutes, (IQR=537-793; Z=-5.922, p<.0001). Patients spent a higher percentage of the subjective night asleep compared to subjective day (53.7% (IQR=32.8-72.4) vs. 32.9% (IQR=15.4-50.0), Z=-5.767, p<.0001), but estimated sleep duration did not differ between subjective night and day (median 280 minutes (IQR=175-362) vs. 216 minutes (IQR=82-342), Z=-0.664, p=.519). Mean and maximum illuminance were significantly lower during subjective night than day (median average lux 7.3 (IQR=3.0-14.0) vs. 53.8 (IQR=28.2-87.6), Z=-5.692, p<.0001; median maximum lux 116.8 (IQR=53.5-254.2) vs. 445.0 (311.8-643.4), Z=-4.854, p<.0001). CONCLUSIONS: In our study, estimated sleep percentages were higher during subjective night than day, but patients obtained similar amounts of sleep during the subjective day and night, indicating poor sleep consolidation. While patients experienced low light levels during the night, they also experienced low daytime light levels, potentially contributing to higher risk of circadian dysregulation and nighttime sleep disturbance. CLINICAL IMPLICATIONS: Our results provide evidence of disrupted rest-activity patterns in critically ill patients and of low daytime light levels in the ICU, highlighting the need for continued investigation into interventions to improve sleep quality in the ICU. DISCLOSURES: No relevant relationships by David Baek, source=Web Response No relevant relationships by Debasree Banerjee, source=Web Response No relevant relationships by Shi-Yo Chen, source=Web Response No relevant relationships by Jisoo Lee, source=Web Response No relevant relationships by Mitchell Levy, source=Web Response No relevant relationships by Amy Mathew, source=Web Responseauthor royalties relationship with UpToDate Please note: 2016-present Added 04/29/2021 by Katherine Sharkey, source=Web Response, value=Royalty