Associations of Rest-Activity Rhythm Disturbances With Stroke Risk and Poststroke Adverse Outcomes

Abstract

Background: Many disease processes are influenced by circadian clocks and display ~24-hour rhythms. Whether disruptions to these rhythms increase stroke risk is unclear. We evaluated the association between 24-hour rest-activity rhythms, stroke risk, and major poststroke adverse outcomes.

Methods and results: We examined ~100 000 participants from the UK Biobank (aged 44-79 years; ~57% women) assessed with actigraphy (6-7 days) and 5-year median follow-up. We derived (1) most active 10-hour activity counts across the 24-hour cycle and the timing of its midpoint timing; (2) the least active 5-hour count and its midpoint; (3) relative amplitude; (4) interdaily stability; and (5) intradaily variability, for stability and fragmentation of the rhythm. Cox proportional hazard models were constructed for time to (1) incident stroke (n=1652) and (2) poststroke adverse outcomes (dementia, depression, disability, or death). Suppressed relative amplitude (lowest quartile [quartile 1] versus the top quartile [quartile 4]) was associated with stroke risk (hazard ratio [HR], 1.61 [95% CI, 1.35-1.92]; P<0.001) after adjusting for demographics. Later most active 10-hour activity count midpoint timing (14:00-15:26; HR, 1.26 [95% CI, 1.07-1.49]; P=0.007) also had higher stroke risk than earlier (12:17-13:10) participants. A fragmented rhythm (intradaily variability) was also associated with higher stroke risk (quartile 4 versus quartile 1; HR, 1.26 [95% CI, 1.06-1.49]; P=0.008). Suppressed relative amplitude was associated with risk for poststroke adverse outcomes (quartile 1 versus quartile 4; HR, 2.02 [95% CI, 1.46-2.48]; P<0.001). All associations were independent of age, sex, race, obesity, sleep disorders, cardiovascular diseases or risks, and other comorbidity burdens.

Conclusions: Suppressed 24-hour rest-activity rhythm may be a risk factor for stroke and an early indicator of major poststroke adverse outcomes.

Keywords: actigraphy; daily activity rhythms; dementia; depression; disability; mortality; stroke.

Figure 1. Flowchart of participants in the study.

The primary analysis included 1101 incident strokes. A total of 1189 stroke cases without dementia, depression, or disability were included in the secondary analysis. TIA indicates transient ischemic attack.

Figure 2. Rest‐activity rhythm and risk for stroke.

Hazard ratio for incident stroke risk by relative amplitude (RA) quartiles with the fourth quartile (the lowest amplitude) as the reference (A), activity level during the most active 10 hours (M10) quartiles with the fourth quartile as the reference (B), and activity level during the least active 5 hours (L5) quartiles with the first quartile (the highest level) as the reference (C). Results were obtained for the first occurrence of stroke adjusted for age, sex, education, ethnicity, and Townsend Deprivation Index.

Figure 3. Timing of rest‐activity rhythm and incident stroke risk.

Hazard ratios for stroke risk for activity level during the most active 10 hours (M10) midpoint quartiles with the “earlier” quartile (12:17–13:10) as the reference (A) and activity level during the least active 5 hours (L5) midpoint quartiles with the “earlier” quartile (2:04–2:56) as the reference (B). P values represent results from global tests.