Self-Reported and Objective Sleep Measures in Stroke Survivors With Incomplete Motor Recovery at the Chronic Stage

 With all these sleep problems post stroke your doctor needs EXACT STROKE PROTOCOLS. And ask if sleeping pills count as valid sleep.

Self-Reported and Objective Sleep Measures in Stroke Survivors With Incomplete Motor Recovery at the Chronic Stage

Show all authors

Melanie K. Fleming, PhD, Tom Smejka, MSc, David Henderson Slater, MRCP,

First Published July 1, 2021 Research Article Find in PubMed 

https://doi.org/10.1177/15459683211029889

Article information

 

Abstract

Background. 

Stroke survivors commonly complain of difficulty sleeping. Poor sleep is associated with reduced quality of life and more understanding of long-term consequences of stroke on sleep is needed.  

Objective. 

The primary aims were to (1) compare sleep measures between chronic stroke survivors and healthy controls and (2) test for a relationship between motor impairment, time since stroke and sleep. Secondary aims were to explore mood and inactivity as potential correlates of sleep and test the correlation between self-reported and objective sleep measures. 

Methods. 

Cross-sectional sleep measures were obtained for 69 chronic stroke survivors (mean 65 months post-stroke, 63 years old, 24 female) and 63 healthy controls (mean 61 years old, 27 female). Self-reported sleep was assessed with the sleep condition indicator (SCI) and sleep diary ratings, objective sleep with 7-nights actigraphy and mood with the Hospital Anxiety and Depression Scale. Upper extremity motor impairment was assessed with the Fugl-Meyer assessment.  

Results. 

Stroke survivors had significantly poorer SCI score (P < .001) and higher wake after sleep onset (P = .005) than controls. Neither motor impairment, nor time since stroke, explained significant variance in sleep measures for the stroke group. For all participants together, greater depression was associated with poorer SCI score (R²_adj = .197, P < .001) and higher age with more fragmented sleep (R²_adj = .108, P < .001). There were weak correlations between nightly sleep ratings and actigraphy sleep measures (r_s = .15–.24).  

Conclusions. 

 Sleep disturbance is present long-term after stroke. Depressive symptoms may present a modifiable factor which should be investigated alongside techniques to improve sleep in this population.

Keywords

actigraphy, self-report, motor impairment, sleep disruption, mood

Introduction

Many stroke survivors report a major change in their sleeping habits since having a stroke.¹ Although there is some evidence for improvements in sleep parameters from the acute to the chronic stage of stroke,^2,3 systematic reviews report a high prevalence of sleep disorders, such as insomnia and sleep disordered breathing, after stroke.^4,5 Sleep disorders are reported to be more common in stroke survivors compared to normative values or to healthy control groups.⁶ Increases in estimated sleep time (per 24 hour period) from pre-stroke to post-stroke have been found to correlate negatively with the ability to engage in activities of daily living at the chronic stage² and reduced quality of life has also been demonstrated for stroke survivors reporting insomnia.⁷

Although some research on sleep after stroke focuses on diagnosable sleep disorders, such as sleep disordered breathing, restless leg syndrome and primary insomnia,^5,8 there is also research directly comparing between people with chronic stroke and age-matched controls with sleep measures as a continuum rather than categorising participants as having a sleep disorder or not. This is important, as alongside other post-stroke complications, sub-clinical sleep disruption has the potential for debilitating long-term consequences. Recently, evidence has begun to emerge that treatments, such as cognitive behavioural therapy for insomnia, may also be useful for people with sub-clinical insomnia symptoms.⁹ The studies that have analysed sleep measures as a continuum report longer sleep latency, greater wake after sleep onset (WASO) and more fragmented sleep, as well as lower subjective sleep quality for stroke survivors compared to controls.^10-13 However, these studies typically have modest sample sizes (20–35 per group). Therefore, confirmation of these findings with a larger sample size is desirable to enable us to further understand this long-term condition.

There is some indication that sleep quality may relate to stroke outcomes, using broad measures of independence in activities of daily living¹⁴ or disability.^15,16 We previously demonstrated that sleep disruption during inpatient rehabilitation from stroke and brain injury is associated with poorer motor outcomes¹⁷ and hypothesise that this may be, at least in part, due to impaired consolidation of motor learning underlying motor recovery.¹⁸ If this is the case, and if sleep disruption persists long-term, then recovery after discharge from rehabilitation may also be limited. Additionally, factors that accompany motor impairment, such as spasticity and pain, may directly affect ability to initiate and maintain sleep. Finally, limitations in movement of the upper or lower limb can affect mobility and therefore physical activity which may indirectly affect sleep. These factors may, to some extent, depend on how long the person has been living with motor impairment. Alternatively, relationships between motor impairment and sleep quality might be expected due to effects of the stroke on cortico-subcortical circuits involved both in the control of sleep and in the control of movement. For example, Gottlieb et al¹⁹ demonstrated that stroke survivors with poor sleep efficiency had altered brain volume in the thalamus, hippocampus and caudate in comparison with controls with normal sleep efficiency. However, to our knowledge, there are no studies at the chronic stage to consider the relationship between motor impairment and sleep quality specifically.

Mood disorders are also common post-stroke complications.^20,21 Depression and anxiety have been found to relate to poor self-reported sleep in stroke survivors and older adults without stroke,^16,22-25 with greater insomnia symptoms present in stroke survivors with depression or anxiety than without.⁴ However, to our knowledge, there are few studies examining whether depression and anxiety relate to objective measures of sleep in this population. Pajediene et al²⁶ found a correlation between polysomnography variables reflecting poor sleep and more depressed mood (from the Hospital Anxiety and Depression Scale) in a sample of 13 acute stroke patients, though the magnitude of this correlation is not reported. In contrast, Bakken et al²⁷ found no significant correlation between the Beck depression Inventory Score and actigraphy variables WASO or number of awakenings. Further research is therefore needed to elucidate whether relationships between self-report measures of mood and sleep are also seen when sleep is measured with actigraphy.

It is currently unknown whether subjective reports of sleep quality are reflective of objective measures of sleep in this population. Ouellet and Morin²⁸ reported that people with traumatic brain injury subjectively reported worse sleep than control participants, but that this was not detected using polysomnography (PSG) suggesting a mismatch between objective and subjective sleep quality. In stroke survivors, 6 months post-stroke, Bakken et al³ report some correspondence between Pittsburgh Sleep Quality Index (PSQI) ratings and time spent asleep, but no meaningful correlations between PSQI and measures of sleep disruption from actigraphy. It is important to understand the relationship between self-reported and objective measures of sleep problems in this population, in order to best tailor interventions aimed at improving aspects of sleep quality.

The primary aims of this study were therefore to compare both objective and self-reported sleep measures between community dwelling chronic stroke survivors and age- and sex-matched healthy controls and to investigate whether variance in sleep measures in stroke survivors could be explained by variance in upper limb motor impairment or time since stroke. Additionally, we sought to explore potential correlates of sleep measures across both groups. We were particularly interested in the chronic stage of stroke, as this is the time when intensive rehabilitation efforts are likely to have completed and the long-term impact of stroke can be understood.

We hypothesised that stroke survivors would demonstrate poorer self-reported sleep, more fragmented sleep and longer time awake overnight than healthy controls. We also hypothesised that stroke survivors with a poorer motor outcome (worse upper limb impairment) would have more sleep disruption than those with good functional outcomes.

Finally, we aimed to test for differences in the agreement across subjective and objective sleep quality measures between stroke survivors and controls. We anticipated that there would be less correspondence between objective and subjective sleep measures for stroke survivors.

More at link.