Abstract
Objective:
Sleep disturbance and depression are common in stroke patients, however, little is known about the role of sleep in post-stroke depression. This study examined the association between pre-stroke sleep duration and depression at 90 days post-stroke in a population-based bi-ethnic sample.
Methods:
The study included 1,369 stroke patients from the Brain Attack Surveillance in Corpus Christi project who survived 90 days post-stroke. Depression at 90 days post-stroke was assessed by the 8-item Patient Health Questionnaire, and pre-stroke sleep duration was self-reported shortly after stroke in reference to the pre-stroke state. Multiple imputation and inverse probability weighting were used to deal with missing data and attrition. Weighted logistic regression models were fit to examine the association between pre-stroke sleep duration and post-stroke depression.
Results:
The mean age was 68.2 years, and 63.6% were Mexican American. The prevalence of post-stroke depression was highest among participants reporting less than 6 hours (52.4%, 95% confidence interval=45.7%-59.0%). Compared with participants reporting 7-8 hours of sleep before stroke, those with short sleep duration had significantly increased odds for post-stroke depression (odds ratio=1.96; 95% confidence interval =1.38-2.79), after adjustment for sociodemographic, stroke and pre-stroke characteristics including pre-stroke depression.
Conclusions:
Pre-stroke short sleep duration may be an independent risk factor for post-stroke depression.
Keywords: stroke, sleep, depression
INTRODUCTION
Disturbances in sleep-related breathing and sleep-wake patterns occur in more than half of stroke patients [1], and commonly exist before stroke [1, 2]. Pre-existing sleep disturbance is an established risk factor for stroke and depression [1, 3–5], may be exacerbated after stroke onset [3], and has adverse impacts on stroke outcomes [3, 6]. Particularly, insomnia or hypersomnia is a core diagnostic criterion of depression, which affects approximately one-third of stroke survivors [7]. Despite the clinical and public health significance, the role of sleep disorders in the development and persistence of post-stroke depression (PSD) remains understudied. The objective of this study was to explore the association between pre-stroke sleep duration and depression at 90 days post-stroke in a population-based sample of stroke survivors.
METHODS
Data are from the Brain Attack Surveillance in Corpus Christi (BASIC) project, an ongoing population-based stroke surveillance study designed to capture all stroke cases among residents aged ≥45 years in a bi-ethnic population in south Texas [8]. Stroke cases are identified by active and passive surveillance, and validated by stroke fellowship trained physicians. Patients with validated stroke are invited to participate in a baseline interview shortly after stroke onset and an outcome interview approximately 90 days after the index stroke. If a stroke patients is not consentable, interviews are completed by a surrogate. Details of the BASIC project have been described elsewhere [8].
Data for the present study were drawn from 3,084 ischemic or hemorrhagic stroke cases occurring between January, 2011 and December, 2016. The study sample consisted of 1,369 non-Hispanic white or Mexican American patients who were eligible for the study, survived 90 days after stroke, and participated in both baseline and outcome interviews.
Measures
Depressive symptoms at 90 days post-stroke were assessed during the in-person outcome interview by the 8-item Patient Health Questionnaire (PHQ) [9], an abbreviated version of the 9-item PHQ, which has been validated in stroke survivors [10]. Participants scoring ≥10 on the 24-point scale were classified as having depression [9].
Sleep duration in reference to the pre-stroke period was assessed during the baseline interview using the question, “How many hours of sleep do you usually get at night (or when you usually sleep)?” We categorized participants into four groups: less than 6 hours (short sleep duration), 6 hours, 7-8 hours (reference), and greater than 8 hours (long sleep duration).
Covariates evaluated as potential confounders were ascertained at baseline from interviews or medical records, including socio-demographic, stroke and pre-stroke characteristics. Sociodemographic characteristics were age (45-64, 65-74, ≥75), sex (male, female), race/ethnicity (non-Hispanic white, Mexican American), education (below high school, high school, above high school), and marital status (married/partnered, other). Stroke characteristics included stroke type (ischemic stroke, intracerebral hemorrhage) and stroke severity measured by the National Institutes of Health Stroke Scale. Pre-stroke characteristics included pre-stroke sleep apnea risk (low, high), as assessed by the Berlin questionnaire; self-reported pre-stroke depression status (no history of depression, history of depression, on medication for depression at stroke onset); pre-stroke cognitive function measured by the Informant Questionnaire on Cognitive Decline in the Elderly (normal: ≤3, cognitive impairment no dementia: >3 and <3.44, dementia: ≥3.44); pre-stroke disability measured by the modified Rankin Scale (no symptoms/disability: 0-1, slight/moderate disability: 2-3, moderately severe/severe disability: 4-5); and number of medical conditions (0-2, 3-4, 5+).
Statistical Analysis
To maintain statistical power and minimize selection bias, we imputed missing values of the outcome and covariates based on socio-demographic, stroke and pre-stroke characteristics, and stroke outcomes at 90 days; and applied inverse probability weights in the analyses to account for potential differential attrition [11]. We examined baseline characteristics by pre-stroke sleep duration using Pearson’s Chi-squared tests for categorical variables and Kruskal-Wallis tests for continuous variables. To examine the association between pre-stroke sleep duration and PSD, we fit a series of weighted logistic regression models by sequentially adjusting for sociodemographic and stroke characteristics, pre-stroke sleep apnea risk, and other pre-stroke characteristics. Statistical analyses were completed with Stata version 14.2 (StataCorp) and SAS version 9.4 (SAS Institute). The BASIC project was approved by the Institutional Review Boards at the University of Michigan and the local hospital systems. Written informed consent was provided by patients or their surrogate.
RESULTS
Baseline sample characteristics are presented in Table 1. The mean age was 68.2 years, women and men were equally distributed, 63.6% were Mexican American, 89.5% had ischemic stroke, and 59.6% were at high risk for sleep apnea before stroke. The mean sleep duration before stroke was 7.2 hours, and the median was 8 hours. Approximately 18.4% of the study sample had sleep duration less than 6 hours, and 18.0% greater than 8 hours. Compared with participants with 7-8 hours of sleep, those with shorter sleep duration were younger, and more likely to have depression history before stroke. Participants with longer than 8 hours of sleep duration, in comparison to those with 7-8 hours of sleep, were older, more functionally and cognitively impaired, and had more medical conditions before stroke.
Table 1.
Baseline sample characteristics by pre-stroke sleep duration, Brain Attack Surveillance in Corpus Christi project, United States, 2011-2016
| Pre-stroke sleep duration* |
||||||
|---|---|---|---|---|---|---|
| Total (n=1362) | <6 hours (n=252) | 6 hours (n=217) | 7-8 hours (n=647) | >8 hours (n=246) | P value | |
| Socio-demographic characteristics | ||||||
| Age, mean (SD) | 68.2 (12.1) | 64.7 (11.5) | 66.1 (11.7) | 68.6 (11.9) | 72.6 (12.3) | <0.001 |
| Age, N (%) | <0.001 | |||||
| 45-64 | 570 (41.9) | 131 (52.0) | 105 (48.4) | 268 (41.4) | 66 (26.8) | |
| 65-74 | 365 (26.8) | 65 (25.8) | 55 (25.4) | 180 (27.8) | 65 (26.4) | |
| ≥75 | 427 (31.4) | 56 (22.2) | 57 (26.3) | 199 (30.8) | 115 (46.8) | |
| Sex, N (%) | 0.359 | |||||
| Male | 677 (49.7) | 119 (47.2) | 107 (49.3) | 337 (52.1) | 114 (46.3) | |
| Female | 685 (50.3) | 133 (52.8) | 110 (50.7) | 310 (47.9) | 132 (53.7) | |
| Race/ethnicity, N (%) | 0.310 | |||||
| Non-Hispanic white | 496 (36.4) | 79 (31.4) | 84 (38.7) | 241 (37.3) | 92 (37.4) | |
| Mexican American | 866 (63.6) | 173 (68.7) | 133 (61.3) | 406 (62.8) | 154 (62.6) | |
| Education*, N (%) | 0.050 | |||||
| Below high school | 459 (34.0) | 95 (38.0) | 63 (29.3) | 211 (32.9) | 90 (37.0) | |
| High school | 380 (28.2) | 56 (22.4) | 62 (28.8) | 203 (31.6) | 59 (24.3) | |
| Above high school | 511 (37.9) | 99 (39.6) | 90 (41.9) | 228 (35.5) | 94 (38.7) | |
| Marital status*, N (%) | 0.503 | |||||
| Married/partnered | 653 (48.2) | 114 (45.6) | 110 (50.9) | 318 (49.3) | 111 (45.5) | |
| Other | 702 (51.8) | 136 (54.4) | 106 (49.1) | 327 (50.7) | 133 (54.5) | |
| Stroke characteristics | ||||||
| Stroke type, N (%) | 0.037 | |||||
| Ischemic stroke | 1219 (89.5) | 218 (86.5) | 203 (93.6) | 572 (88.4) | 226 (91.9) | |
| Intracerebral hemorrhage stroke | 143 (10.5) | 34 (13.5) | 14 (6.5) | 75 (11.6) | 20 (8.1) | |
| Stroke severity (NIHSS)*, mean (SD) | 5.4 (6.2) | 5.2 (5.3) | 4.9 (5.9) | 5.6 (6.6) | 5.9 (6.0) | 0.251 |
| Pre-stroke characteristics | ||||||
| Pre-stroke sleep apnea risk, N (%) | 0.048 | |||||
| Low | 550 (40.4) | 89 (35.3) | 82 (37.8) | 286 (44.2) | 93 (37.8) | |
| High | 812 (59.6) | 163 (64.7) | 135 (62.2) | 361 (55.8) | 153 (62.2) | |
| Pre-stroke depression status*, N (%) | <0.001 | |||||
| No history of depression | 659 (48.4) | 119 (47.2) | 106 (48.9) | 339 (52.4) | 95 (38.6) | |
| History of depression | 163 (12.0) | 37 (14.7) | 32 (14.8) | 74 (11.4) | 20 (8.1) | |
| On medication for depression at stroke onset | 211 (15.5) | 54 (21.4) | 39 (18.0) | 71 (11.0) | 47 (19.1) | |
| Missing | 329 (24.2) | 42 (16.7) | 40 (18.4) | 163 (25.2) | 84 (34.2) | |
| Pre-stroke disability (mRS)*, N (%) | <0.001 | |||||
| No symptoms/disability | 557 (42.1) | 89 (36.6) | 96 (45.3) | 301 (47.8) | 71 (29.7) | |
| Slight/moderate disability | 610 (46.1) | 127 (52.3) | 87 (41.0) | 271 (43.0) | 125 (52.3) | |
| Moderately severe/severe disability | 157 (11.9) | 27 (11.1) | 29 (13.7) | 58 (9.2) | 43 (18.0) | |
| Pre-stroke cognitive function (IQCODE)*, N (%) | 0.002 | |||||
| Normal | 603 (44.3) | 109 (43.3) | 92 (42.4) | 313 (48.4) | 89 (36.2) | |
| CIND | 386 (28.3) | 72 (28.6) | 66 (30.4) | 170 (26.3) | 78 (31.7) | |
| Dementia | 200 (14.7) | 31 (12.3) | 32 (14.8) | 81 (12.5) | 56 (22.8) | |
| Missing | 173 (12.7) | 40 (15.9) | 27 (12.4) | 83 (12.8) | 23 (9.4) | |
| Number of medical conditions, N (%) | 0.002 | |||||
| 0-2 | 639 (46.9) | 109 (43.3) | 113 (52.1) | 326 (50.4) | 91 (37.0) | |
| 3-4 | 563 (41.3) | 111 (44.1) | 84 (38.7) | 256 (39.6) | 112 (45.5) | |
| 5+ | 160 (11.8) | 32 (12.7) | 20 (9.2) | 65 (10.1) | 43 (17.5) | |
Note. CIND, cognitive impairment no dementia; IQCODE, Informant Questionnaire on Cognitive Decline in the Elderly; mRS, modified Rankin Scale; NIHSS, National Institutes of Health Stroke Scale; SD, standard deviation.
Variables with missing data. The numbers of missing data were 7 for pre-stroke sleep duration, 12 for education, 7 for marital status, 6 for stroke severity, 329 for pre-stroke depression status, 38 for pre-stroke disability and 173 for pre-stroke cognitive function.
After accounting for missing data and attrition, the prevalence of depression at 90 days post-stroke was highest among participants with less than 6 hours of sleep before stroke (52.4%, 95% CI=45.7%-59.0%), and lowest among those with 7-8 hours of sleep (32.7%, 95% CI=28.6%-36.8%). Compared with participants with 7-8 hours of sleep, the unadjusted association between pre-stroke sleep duration and PSD was J-shaped with odds ratios of 2.27 (95% CI=1.65-3.11) for those with <6 hours of sleep, 1.26 (95% CI=0.88-1.81) for those with 6 hours of sleep, and 1.37 (95% CI=0.97-1.94) for those with >8 hours of sleep. The association between short sleep duration and PSD remained statistically significant (OR=1.96; 95% CI=1.38-2.79), after adjustment for sociodemographic, stroke and pre-stroke characteristics including pre-stroke depression status (Table 2).
Table 2.
Results from weighted logistic regression models of the association between pre-stroke sleep duration and depression at 90 days post-stroke, Brain Attack Surveillance in Corpus Christi project, United States, 2011-2016
| Model 1 OR (95% CI) |
Model 2 OR (95% CI) |
Model 3 OR (95% CI) |
Model 4 OR (95% CI) |
|
|---|---|---|---|---|
| Pre-stroke sleep duration | ||||
| <6 hours | 2.27 (1.65, 3.11) | 2.20 (1.58, 3.06) | 2.12 (1.52, 2.97) | 1.96 (1.38, 2.79) |
| 6 hours | 1.26 (0.88, 1.81) | 1.31 (0.90, 1.90) | 1.26 (0.86, 1.84) | 1.10 (0.73, 1.67) |
| 7-8 hours | Reference | Reference | Reference | Reference |
| >8 hours | 1.37 (0.97, 1.94) | 1.43 (1.00, 2.05) | 1.37 (0.95, 1.97) | 1.08 (0.74, 1.57) |
| Age | ||||
| 45-64 | Reference | Reference | Reference | |
| 65-74 | 0.54 (0.40, 0.75) | 0.56 (0.40, 0.77) | 0.47 (0.33, 0.66) | |
| ≥75 | 0.54 (0.39, 0.75) | 0.56 (0.41, 0.78) | 0.49 (0.34, 0.70) | |
| Sex | ||||
| Male | Reference | Reference | Reference | |
| Female | 1.11 (0.85, 1.44) | 1.15 (0.88, 1.49) | 0.88 (0.65, 1.20) | |
| Race/ethnicity | ||||
| Non-Hispanic White | Reference | Reference | Reference | |
| Mexican American | 1.17 (0.87, 1.57) | 1.13 (0.84, 1.53) | 1.13 (0.81, 1.58) | |
| Education | ||||
| Below high school | Reference | Reference | Reference | |
| High school | 0.67 (0.48, 0.93) | 0.68 (0.49, 0.94) | 0.72 (0.50, 1.03) | |
| Above high school | 0.57 (0.41, 0.79) | 0.58 (0.42, 0.80) | 0.61 (0.43, 0.87) | |
| Marital status | ||||
| Married/partnered | Reference | Reference | Reference | |
| Other | 1.25 (0.96, 1.63) | 1.29 (0.99, 1.69) | 1.13 (0.85, 1.51) | |
| Stroke type | ||||
| Ischemic stroke | Reference | Reference | Reference | |
| Intracerebral hemorrhage stroke | 0.87 (0.53, 1.40) | 0.90 (0.55, 1.48) | 0.98 (0.57, 1.68) | |
| Stroke severity (log-transformed NIHSS) | ||||
| Linear term | 1.35 (0.88, 2.07) | 1.38 (0.89, 2.13) | 1.30 (0.80, 2.09) | |
| Quadratic term | 1.02 (0.89, 1.18) | 1.03 (0.89, 1.18) | 1.04 (0.89, 1.22) | |
| Pre-stroke sleep apnea risk | ||||
| Low | Reference | Reference | ||
| High | 1.86 (1.40, 2.46) | 1.53 (1.13, 2.08) | ||
| Pre-stroke depression status | ||||
| No history of depression | Reference | |||
| History of depression | 2.14 (1.40, 3.25) | |||
| On medication for depression at stroke onset | 3.48 (2.21, 5.49) | |||
| Pre-stroke disability (mRS) | ||||
| No symptoms/disability | Reference | |||
| Slight/moderate disability | 1.45 (1.06, 1.98) | |||
| Moderately severe/severe disability | 1.80 (0.86, 3.77) | |||
| Pre-stroke cognitive function (IQCODE) | ||||
| Normal | Reference | |||
| CIND | 1.24 (0.87, 1.76) | |||
| Dementia | 1.86 (1.05, 3.29) | |||
| Number of medical conditions | ||||
| 0-2 | Reference | |||
| 3-4 | 1.05 (0.77, 1.43) | |||
| 5+ | 1.30 (0.78, 2.16) | |||
Note. CI, confidence interval; CIND, cognitive impairment no dementia; IQCODE, Informant Questionnaire on Cognitive Decline in the Elderly; mRS, modified Rankin Scale; NIHSS, National Institutes of Health Stroke Scale; OR, odds ratio. The sample size of all three models was 1369.
DISCUSSION
We found that short sleep duration before stroke was associated with increased odds for depression at 90 days post-stroke, after adjustment for covariates including pre-stroke depression and sleep apnea risk. This finding suggests that sleep-related risk factors for the development and persistence of PSD may exist before stroke onset, which may have possible preventive and therapeutic implications for stroke recovery. Incorporating pre-stroke sleep characteristics into PSD risk assessment may improve case identification. Moreover, treatment for sleep disorders has shown favorable effects on depression outcomes in the general population [12], which may hold promise for stroke patients as well.
To the best of our knowledge, this study is among the first to explore the relationship between pre-stroke sleep characteristics and PSD. Sleep disorders are risk factors for depression [4, 5], which may occur before or after stroke. Our results are consistent with existing findings in middle-aged and older adults that short sleep duration is associated with the onset and recurrence of depression [13, 14]. Sleep disturbance causes neuroendocrine and circadian rhythm abnormalities that increase risk for depression [15]. Individuals at elevated risk may develop depression triggered by stroke through neurobiological mechanisms, such as inflammation, hypoperfusion and neurogenesis inhibition [7]. Short sleep duration that persists or exacerbates after stroke may further interfere with functional recovery [16], and therefore increase the risk of PSD through psychosocial mechanisms. Another possibility is that short sleep duration before stroke is a symptom of pre-stroke depression, a well-recognized risk factor for PSD [7], and therefore non-causally associated with PSD, as shown by the slight attenuation in the association after adjustment for pre-stroke depression status. Future research should investigate potential causal pathways of specific sleep disorders, such as sleep apnea and insomnia, to PSD.
This study is hypothesis-generating, and the results should be interpreted with consideration of several limitations. First, sleep duration was assessed by one single question in reference to the pre-stroke time period. Self-reported sleep duration is subject to recall bias and potential misclassifications due to individual variations in interpretation of the survey question. For example, periods of wakefulness after sleep may have been included in total reported sleep duration by some subjects, but not others. Moreover, we did not ask any questions about sleep quality. Future research should further investigate the association using objective measures of sleep duration and sleep quality. Second, PSD was assessed by the PHQ-8 rather than a diagnostic interview, thus we cannot identify false-positive cases due to the complexity of stroke. However, the PHQ has good sensitivity and specificity in stroke patients [10]. Third, because data on post-stroke sleep duration were not available in the study, we cannot examine how self-reported sleep duration has changed after stroke, and its relationship with PSD. Fourth, pre-stroke depression status and sleep apnea risk were based on self-report, which are also subject to measurement errors. Fifth, because the study sample was drawn from a non-immigrant bi-ethnic population, the results may not be generalizable to immigrants and other racial/ethnic groups.
In conclusion, this investigation suggests that short sleep duration before stroke may be an independent risk factor for PSD. Studies with objective sleep measures should replicate the finding and examine whether screening and treating people at high risk for post-stroke sleep disturbance lessens the burden of PSD.
Supplementary Material
Highlights.
Pre-stroke short sleep duration may be a risk factor for post-stroke depression
Sleep-related risk factors for post-stroke depression may exist before stroke onset
Treatment for sleep disorders may hold promise for stroke patients with depression
Acknowledgements
This study was performed in the Corpus Christi Medical Center and CHRISTUS Spohn Hospitals, CHRISTUS Health system, in Corpus Christi, Texas, USA.
Funding
This work was supported by the National Institutes of Health (grant numbers R01NS38916, R01NS070941 and U24NS107214).
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
REFERENCES
- [1].Khot SP, Morgenstem LB. Sleep and Stroke. Stroke. 2019;50:1612–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [2].Leppävuori A, Pohjasvaara T, Vataja R, Kaste M, Erkinjuntti T. Insomnia in ischemic stroke patients. Cerebrovascular diseases. 2002;14:90–7. [DOI] [PubMed] [Google Scholar]
- [3].Wallace DM, Ramos AR, Rundek T. Sleep disorders and stroke. International Journal of Stroke. 2012;7:231–42. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [4].Baglioni C, Battagliese G, Feige B, Spiegelhalder K, Nissen C, Voderholzer U, et al. Insomnia as a predictor of depression: a meta-analytic evaluation of longitudinal epidemiological studies. Journal of affective disorders. 2011;135:10–9. [DOI] [PubMed] [Google Scholar]
- [5].Zhai L, Zhang H, Zhang D. Sleep duration and depression among adults: A meta-analysis of prospective studies. Depression and anxiety. 2015;32:664–70. [DOI] [PubMed] [Google Scholar]
- [6].Hermann DM, Bassetti CL. Role of sleep-disordered breathing and sleep-wake disturbances for stroke and stroke recovery. Neurology. 2016;87:1407–16. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [7].Robinson RG, Jorge RE. Post-stroke depression: a review. American Journal of Psychiatry. 2015;173:221–31. [DOI] [PubMed] [Google Scholar]
- [8].Smith MA, Risser JM, Moyé LA, Garcia N, Akiwumi O, Uchino K, et al. Designing multi-ethnic stroke studies: the Brain Attack Surveillance in Corpus Christi (BASIC) project. Ethnicity and Disease. 2004;14:520–6. [PubMed] [Google Scholar]
- [9].Kroenke K, Strine TW, Spitzer RL, Williams JB, Berry JT, Mokdad AH. The PHQ-8 as a measure of current depression in the general population. Journal of affective disorders. 2009;114:163–73. [DOI] [PubMed] [Google Scholar]
- [10].Williams LS, Brizendine EJ, Plue L, Bakas T, Tu W, Hendrie H, et al. Performance of the PHQ-9 as a screening tool for depression after stroke. stroke. 2005;36:635–8. [DOI] [PubMed] [Google Scholar]
- [11].Weuve J, Tchetgen EJT, Glymour MM, Beck TL, Aggarwal NT, Wilson RS, et al. Accounting for bias due to selective attrition: the example of smoking and cognitive decline. Epidemiology (Cambridge, Mass). 2012;23:119. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [12].Franzen PL, Buysse DJ. Sleep disturbances and depression: risk relationships for subsequent depression and therapeutic implications. Dialogues in clinical neuroscience. 2008; 10:473. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [13].Lippman S, Gardener H, Rundek T, Seixas A, Elkind MS, Sacco RL, et al. Short sleep is associated with more depressive symptoms in a multi-ethnic cohort of older adults. Sleep medicine. 2017;40:58–62. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [14].Sun Y, Shi L, Bao Y, Sun Y, Shi J, Lu L. The bidirectional relationship between sleep duration and depression in community-dwelling middle-aged and elderly individuals: evidence from a longitudinal study. Sleep medicine. 2018;52:221–9. [DOI] [PubMed] [Google Scholar]
- [15].Tsuno N, Besset A, Ritchie K. Sleep and depression. The Journal of clinical psychiatry. 2005, 66(10), 1254–1269. [DOI] [PubMed] [Google Scholar]
- [16].Duss SB, Seiler A, Schmidt MH, Pace M, Adamantidis A, Muri RM, et al. The role of sleep in recovery following ischemic stroke: a review of human and animal data. Neurobiology of sleep and circadian rhythms. 2017;2:94–105. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.