Objective Delirium, a common complication after stroke, is often overlooked, and long-term consequences are poorly understood. This study aims to explore whether delirium in the acute phase of stroke predicts cognitive and psychiatric symptoms three, 18 and 36 months later. Method As part of the Norwegian Cognitive Impairment After Stroke Study (Nor-COAST), 139 hospitalized stroke patients (49% women, mean (SD) age: 71.4 (13.4) years; mean (SD) National Institutes of Health Stroke Scale (NIHSS) 3.0 (4.0)) were screened for delirium with the Confusion Assessment Method (CAM). Global cognition was measured with the Montreal Cognitive Assessment (MoCA), while psychiatric symptoms were measured using the Hospital Anxiety and Depression Scale (HADS) and the Neuropsychiatric Inventory-Questionnaire (NPI-Q). Data was analyzed using mixed-model linear regression, adjusting for age, gender, education, NIHSS score at baseline and premorbid dementia. Results Thirteen patients met the criteria for delirium. Patients with delirium had lower MoCA scores compared to non-delirious patients, with the largest between-group difference found at 18 months (Mean (SE): 20.8 (1.4) versus (25.1 (0.4)). Delirium was associated with higher NPI-Q scores at 3 months (Mean (SE): 2.4 (0.6) versus 0.8 (0.1)), and higher HADS anxiety scores at 18 and 36 months, with the largest difference found at 36 months (Mean (SE): 6.2 (1.3) versus 2.2 (0.3)). Conclusions Suffering a delirium in the acute phase of stroke predicted more cognitive and psychiatric symptoms at follow-up, compared to non-delirious patients. Preventing and treating delirium may be important for decreasing the burden of post-stroke disability.

Objective: Findings on the relationship of psychiatric symptoms with performance-based and self-reported cognitive function post-stroke are inconclusive. We aimed to (1) study the relation of depression and anxiety to performance-based cognitive function and (2) explore a broader spectrum of psychiatric symptoms and their association with performance-based versus self-reported cognitive function. Method: Individuals with supratentorial ischemic stroke performed neuropsychological examination 3 months after stroke. For primary analyses, composite scores for memory and attention/executive function were calculated based on selected neuropsychological tests, and the Hospital Anxiety and Depression Scale (HADS) was used. Psychiatric symptoms and self-reported cognitive function for secondary aims were assessed using the Symptom-Checklist-90 – Revised (SCL-90-R). Results: In a sample of 86 patients [mean (M) age: 64.6 ± 9.2; Mini-Mental State Examination (MMSE), 3–7 days post-stroke: M = 28.4 ± 1.7; National Institutes of Health Stroke Scale (NIHSS) after 3 months: M = 0.7 ± 1.6] depressive symptoms (HADS) were associated with poorer memory performance after controlling for age, sex, and education (p ≤ .01). In a subsample (n = 41; Age: M = 65.7 ± 8.1; MMSE: M = 28.4 ± 1.8; NIHSS: M = 1.0 ± 1.9), symptoms of phobic anxiety (SCL-90-R) were associated with poorer performance-based memory and attention/executive function, and symptoms of anxiety (SCL-90-R) with lower attention/executive function. Higher levels of self-reported cognitive difficulties were associated with higher scores in all psychiatric domains (p ≤ .05). Conclusion: Even in relatively well-functioning stroke patients, depressive symptoms are associated with poorer memory. The results also suggest that various psychiatric symptoms are more related to self-reported rather than to performance-based cognitive function. Screening for self-reported cognitive difficulties may not only help to identify patients with cognitive impairment, but also those who need psychological treatment.

IntroductionPost-stroke fatigue and increased need for daytime sleep are multidimensional and insufficiently understood sequelae. Our aim was to study the relationships of self-reported cognitive and psychiatric symptoms at 3 months with fatigue and daytime sleep at 12 months post-stroke.MethodsIschemic stroke patients without reported history of dementia or depression completed postal surveys 3- and 12-months post-stroke. At 3 months, psychiatric symptoms were assessed with the Hospital Anxiety and Depression Scale (HADS), and self-reported changes in cognitive symptoms (concentration and memory) compared to pre-stroke were assessed using single-item measures. At 12 months, single-item questions about changes in self-reported difficulties sleeping at night, fatigue and daytime sleep were included. First, we studied whether self-reported cognitive and/or psychiatric symptoms at 3 months were associated with daytime sleep and fatigue at 12 months using multiple logistic regression. Second, we fitted 2 structural equation models (SEMs) predicting fatigue and 2 models predicting daytime sleep. We compared a model where only age, sex, stroke severity (National Institutes of Health Stroke Scale; NIHSS), and difficulties sleeping at night predicted fatigue and daytime sleep at 12 months to a model where mental distress (i.e., a latent variable built of cognitive and psychiatric symptoms) was included as an additional predictor of fatigue and daytime sleep at 12 months.ResultsOf 156 patients (NIHSS within 24 hours after admission (mean ± SD) = 3.6 ± 4.3, age = 73.0 ± 10.8, 41% female) 37.9% reported increased daytime sleep and 50.0% fatigue at 12 months. Increased psychiatric symptoms and worsened cognitive symptoms were associated with fatigue and daytime sleep at 12 months, after controlling for NIHSS, age, sex, and difficulties sleeping at night. SEM models including mental distress as predictor showed adequate model fit across 3 fit measures (highest RMSEA = 0.063, lowest CFI and TLI, both 0.975). Models without mental distress were not supported.ConclusionSelf-reported cognitive and psychiatric symptoms at 3 months predict increased daytime sleep and fatigue at 12 months. This highlights the relevance of monitoring cognitive and psychiatric symptoms in the subacute phase post-stroke. However, future research using validated measures of self-reported symptoms are needed to further explore these relationships.