Relationship between subjective and actigraphy-measured sleep in 237 patients with metastatic colorectal cancer

Abstract

Objective: Patients with cancers frequently experience sleep and circadian dysfunction. To date, only a few studies have used both a questionnaire and actigraphy for concomitant evaluation of sleep and circadian function in patients with cancer. We sought to evaluate objective sleep and circadian parameters in metastatic colon cancer (MCC) patients and their associations with symptoms and quality of life (QOL).

Methods: Patients reported subjective sleep problems on the EORTC QLQ-C30. Sleep and circadian parameters were calculated using a wrist-actigraph that patients wore for 72 h.

Results: 237 Patients with MCC (mean age: 60.4 years; range: 20.7-77.6; Male/Female ratio: 1.66) participated in this cross-sectional study. Subjective sleep problems were reported by 63.4% of patients (S+). No differences in any sleep parameters (sleep efficiency, sleep latency, total sleep time, total time in bed, wake after sleep onset, activity bathyphase) were observed between S+ and S- patients. However, S+ patients displayed a significantly worse circadian function than S- patients (96.4 vs 98.1%; p = 0.005). The presence of poor subjective sleep and objective circadian dysfunction negatively affected symptoms and QOL domains (p = 0.038).

Conclusions: Subjective report of sleep problems was not associated with worse objectively measured sleep parameters in patients with MCC although it was associated with disrupted circadian rest-activity rhythm and poorer QOL. These findings coincide with prior research in cancer patients in that an inconsistent relationship exists between subjective and objective sleep measurements on some sleep domains. This study supports the value of coupled evaluation of self-reported and objective measures of sleep and circadian function in cancer patients.

Keywords: Actigraphy; Cancer; Circadian; Metastatic colorectal cancer; Oncology; Sleep.

Figure 1

Distribution of 4 objective actigraphy sleep parameters according to the answer to the question of the quality of life questionnaire about sleep trouble: gray, yes; white, no. Boxes represent 25th and 75th percentiles, whiskers the range, and the middle black line the median. All p values were > 0.05.

Figure 2

Distribution of the circadian parameter I

Figure 3

Mean (+ SD) values for pertinent quality of life domains or symptoms derived from the EORTC QLQ-C30 questionnaire, according to the presence or absence of subjective sleep complaints and/or objective circadian disruption. P values are derived from nonparametric Kruskal-Wallis H test.