Abstract
Objective:
To compare changes in cognitive trajectories after stroke between younger (18-64) and older (65+) adults, accounting for pre-stroke cognitive trajectories.
Materials and methods:
Pooled cohort study using individual participant data from 3 US cohorts (1971-2019), the Atherosclerosis Risk In Communities Study (ARIC), Framingham Offspring Study (FOS), and REasons for Geographic And Racial Differences in Stroke Study (REGARDS). Linear mixed effect models evaluated the association between age and the initial change (intercept) and rate of change (slope) in cognition after compared to before stroke. Outcomes were global cognition (primary), memory and executive function.
Results:
We included 1,292 participants with stroke; 197 younger (47.2 % female, 32.5 % Black race) and 1,095 older (50.2 % female, 46.4 % Black race). Median (IQR) age at stroke was 59.7 (56.6-61.7) (younger group) and 75.2 (70.5-80.2) years (older group). Compared to the young, older participants had greater declines in global cognition (−1.69 point [95 % CI, −2.82 to −0.55] greater), memory (−1.05 point [95 % CI, −1.92 to −0.17] greater), and executive function (−3.72 point [95 % CI, −5.23 to −2.21] greater) initially after stroke. Older age was associated with faster declines in global cognition (−0.18 points per year [95 % CI, −0.36 to −0.01] faster) and executive function (−0.16 [95 % CI, −0.26 to −0.06] points per year for every 10 years of higher age), but not memory (−0.006 [95 % CI, −0.15 to 0.14]), after compared to before stroke.
Conclusion:
Older age was associated with greater post-stroke cognitive declines, accounting for differences in pre-stroke cognitive trajectories between the old and the young.
Keywords: Aging, Cognition, Stroke, Ischemic stroke, Hemorrhagic stroke, Cognitive decline
Introduction
Stroke is associated with an initial decline in cognitive function and accelerated cognitive declines over the long-term.1 How younger (age 18-64) compare to older stroke survivors (≥65 years) in the initial change in cognition after stroke and post-stroke cognitive decline over time is unclear. While older age has been associated with increased risk for cognitive decline after stroke, most studies have consisted of stroke survivors over the age of 65.2–4
Young stroke survivors can have cognitive impairments at 3 months5 and even 5 or 11 years6,7 after stroke. Less is known whether a similar trajectory of cognitive decline will be observed in young stroke survivors after controlling for pre-stroke cognitive trajectories. Accumulation of cerebrovascular disease with age may make older adults less able than younger adults to compensate for the effect of an acute stroke lesion on cognition. We sought to characterize and compare the trajectories of cognitive decline after stroke in younger (age 18-64) and older adults (age≥65), leveraging a pooled cohort with participants from three well-characterized US longitudinal cohorts, expert-adjudicated stroke, and harmonized repeated cognitive outcomes before and after stroke.
Methods
Study design, participants, and measurements
We performed a pooled cohort study using individual participant data of the STROKE COG (the effect of vascular risk factors on risk of Alzheimer’s disease and related dementias after stroke) cohort, which includes the longitudinal cohorts: Atherosclerosis Risk In Communities (ARIC), Framingham Offspring Study (FOS), and REasons for Geographic and Racial Differences in Stroke (REGARDS) (1971-2019) (Supplemental Methods).8 Incident stroke (ischemic or hemorrhagic) was adjudicated by physicians in each cohort. STROKE COG participants were eligible for this analysis if they were age≥18 years, self-reported Black or Non-Hispanic White race, and had ≥1 cognitive assessment before and ≥1 cognitive assessment after stroke. We excluded participants who self-reported other races due to their small number and participants with pre-stroke dementia. We included people with a self-reported history of stroke at cohort baseline because we have found no difference in the association between stroke and cognitive decline including vs. excluding these participants.8 The index stroke was the stroke that occurred during cohort follow-up.
The study was approved by the University of Michigan Institutional Review Board (IRB). Individual cohorts received IRB approval from their respective institutions. Participants provided written informed consent. BC had full access to the data and takes responsibility for the data’s integrity and analysis.
Cognitive function outcomes
Trained cohort staff administered cognitive tests by telephone (REGARDS)9 or in person (ARIC, FOS),10, 11. Each cohort administered a different set of tests that had some overlap (Supplemental Methods). To derive common cognitive outcome measures, we used methods based in item response theory to statistically cocalibrate summary factors for cognitive performance using all available cognitive test items, both those unique to a cohort as well as those common across cohorts. We generated factors for global cognitive performance (primary outcome), memory, and executive function (complex and/or speeded cognitive functions).12–14. In this approach, confirmatory factor analysis models for a given domain were estimated in a reference study (e.g., ARIC), then subsequent models in other cohorts were estimated in a serial fashion, fixing item parameters (e.g., loadings and thresholds) to their values in the reference cohort. A final model with all cohorts and no freely estimated parameters was used to generate factor scores.14–16 Global cognitive performance includes all test items including memory and executive function. In this approach, cognitive outcomes are represented as a t-score (mean 50, standard deviation [SD] of 10) at a participant’s first cognitive assessment; thus, a 1-point difference is equivalent to a 0.1 SD difference in the distribution of cognition across cohorts. Higher scores indicate better cognitive performance.
Measurement of Age
Age was measured at time of stroke. We examined age as a categorical variable defined as young (age 18-64) and older (age≥65) adults. We defined the younger age group as age 18-64, because of the expected small number of events if we had restricted the younger group to age 18-44 given that stroke in this age group is rare. We also evaluated age as a continuous variable.
Covariates
Covariates were selected based on the literature, cross-cohort data availability, and because they are potential confounders of the relationship between age and post-stroke cognitive decline. Covariates were harmonized across cohorts based on common categories and units. Self-reported race, sex, and education were measured at cohort baseline. Education categories were less than high school education, high school graduate, some college, and college graduate or above. Pre-stroke glucose is the arithmetic mean of all pre-stroke fasting glucose values. Pre-stroke systolic blood pressure (SBP) is the arithmetic mean of all pre-stroke SBP values. APOE4 is the number of E4 alleles of the APOE4 gene.
Statistical analysis
We performed descriptive statistics by age category and by cohort. Linear regression was used for continuous variables, Pearson’s chi square for categorical variables, and the Kruskal-Wallis rank test to compare the number of pre-stroke and post-stroke cognitive assessments by age category.
We used segmented linear mixed-effect models with a change point at the time of stroke to assess the effect of age on the initial change in cognition after stroke (intercept) and on the rate of change in cognition during the post-stroke period (slope) for each cognitive outcome separately. The initial change in cognition after stroke is measured at the time of the first post-stroke cognitive assessment. The model also uses the pre-stroke cognitive measurements to estimate the initial change in cognition at the time of stroke. Random effects for intercept and slope were included to account for within-subject correlation. We performed a complete case analysis.
Models included the pre-stroke slope (the change in cognition over time before stroke), a stroke indicator representing the post-stroke period (coded as 0 [pre-stroke] and as 1 [post-stroke]), and the post-stroke change in slope. Age and its interactions with the stroke variable and the change in slope post-stroke were used to assess how age influenced the initial decline in cognition level and change in the cognitive trajectory after stroke. We further adjusted models for race, education, sex, cohort, pre-stroke fasting glucose level, and pre-stroke SBP level. Because age at stroke differed significantly between race groups, we included a race × pre-stroke slope interaction term (adjusting for race differences in pre-stroke cognitive slope) and a race × post-stroke slope change (adjusting for race differences in the change in cognitive slope post-stroke) interaction term in the model to account for the potential confounding effect of race.
We calculated marginal values of the post-stroke slope in the young and older age groups.
We explored a potential non-linear age effect on post-stroke cognitive decline using a cubic splines function and log-likelihood tests, but did not find an effect. We checked model diagnostic statistics to ensure model assumptions were met.
Using the fitted model, we calculated participant-specific predicted values for the global cognition score of exemplar participants at the median of the younger age group (60 years old) and at the median of the older age group (75 years old). The covariate values for the exemplar participants were male sex, Black race, REGARDS cohort, less than a high school education, mean pre-stroke SBP of 145 mmHg (the 75th percentile), and mean pre-stroke glucose of 114 mg/dL (the 75th percentile).
We repeated the above analyses for the primary outcome on the subset of participants with APOE4 data. Given the known association between APOE4 genotype and cognitive decline,17 including post-stroke cognitive decline,8 we included interactions between APOE4 genotype and pre-stroke cognitive slope and between APOE4 genotype and post-stroke cognitive slope change in the model.
Cognitive outcomes were censored at the time of death, loss to follow-up, or end of follow-up, whichever came first.
Analyses were performed using SAS software, version 9.4.
Data availability
De-identified participant data can be provided upon cohort and principal investigator approval and with signed data use agreements with cohorts and all relevant institutions.
Results
There were 3,447 participants with stroke during the cohort period. Of these, there were 2,890 who were age ≥ 65 and 557 who were 18-64 years old. All of the excluded cohort participants who were missing pre-stroke cognitive assessments (n=319) attended at least 1 pre-stroke cohort visit but did not have a cognitive assessment. Of the 1366 excluded cohort participants who were missing post-stroke cognitive assessments, there were 1062 who died and 16 who attended at least 1 post-stroke cohort visit before death but did not have a post-stroke cognitive assessment. Of the surviving excluded cohort participants who were missing post-stroke cognitive assessments, there were 220 whose stroke occurred after the last cohort exam and there were 68 who attended at least 1 post-stroke cohort visit but did not have a post-stroke cognitive assessment. Missing data on covariates among excluded cohort participants may have been due to attrition or missed study visits. Among the stroke participants aged ≥65, there were 1095 who met inclusion criteria. Among stroke participants aged 18-64, there were 197 who met inclusion criteria (Fig. 1).
Fig. 1.
Derivation of the participant cohort.
Table 1 presents baseline characteristics. The median (IQR) age of the younger group was 59.7 (56.6-61.7) years old and of the older group was 75.2 (70.5-80.2) years old. The older group had a higher proportion of Black individuals than the younger group (46.4 % vs. 32.5 %, respectively). The younger group had a higher pre-stroke median (IQR) fasting glucose than the older group (101.5 mg/dL [92.0-121.0] vs. 97.5 mg/dL [89.5-110.0]). Younger and older stroke participants did not significantly differ in educational attainment, sex, or pre-stroke SBP.
Table 1.
Participant characteristics.
| Young Stroke Patients (N=197) (<65 years old) | Older Stroke Patients (N=1095) (≥65 years old) | P | |
|---|---|---|---|
| Age, (median, IQR), years | 59.7 (56.6-61.7) | 75.2 (70.5-80.2) | <0.001 |
| Age, (min, max), years* | 44.1, 64.0 | 64.01, 95.8 | |
| Education, No. (%) | |||
| Less than high school | 31 (15.7 %) | 145 (13.2 %) | 0.65 |
| High school graduate | 56 (28.4 %) | 340 (31.1 %) | |
| Some college | 42 (21.3 %) | 255 (23.3 %) | |
| College graduate or more | 68 (34.5 %) | 355 (32.4 %) | |
| Sex | |||
| Female, No. (%) | 93 (47.2 %) | 550 (50.2 %) | 0.44 |
| Race | |||
| Black, No. (%) | 64 (32.5 %) | 508 (46.4 %) | <0.001 |
| Cohort, No. (%) | |||
| ARIC | 63 (32.0 %) | 244 (22.3 %) | 0.0002 |
| FOS | 28 (14.2 %) | 117 (10.7 %) | |
| REGARDS | 106 (53.8 %) | 734 (67.0 %) | |
| Pre-stroke systolic BP, (median, IQR), mmHg | 135.8 (124.0-148.0) | 134.0 (123.5-145.0) | 0.28 |
| Pre-stroke fasting glucose (median, IQR), mg/dL | 101.5 (92.0-121.0) | 97.5 (89.5-110.0) | <0.001 |
| Number of pre-stroke cognitive assessments per individual, median (Q1,Q3) | |||
| Global cognitive performance | 3.0 (1.0-5.0) | 5.0 (3.0-10.0) | <0.001 |
| Executive function | 1.0 (0.0-2.0) | 2.0 (1.0-3.0) | <0.001 |
| Memory | 2.0 (1.0-3.0) | 4.0 (2.0-7.0) | <0.001 |
| Number of post-stroke cognitive assessments per individual, median (Q1,Q3) | |||
| Global cognitive performance | 2.0 (1.0-5.0) | 4.0 (2.0-9.0) | <0.001 |
| Executive function | 1.0 (1.0-3.0) | 1.0 (1.0-2.0) | 0.006 |
| Memory | 3.0 (1.0-5.0) | 2.0 (1.0-4.0) | 0.03 |
| The last cognitive score before stroke, median (IQR) | |||
| Global cognitive performance | 56.3 (49.2-58.3) | 53.2 (47.9-58.5) | 0.21 |
| Executive function | 52.8 (44.9-58.0) | 49.6 (43.4-54.7) | 0.017 |
| Memory | 57.4 (49.4-57.4) | 57.4 (49.1-57.4) | 0.004 |
| Cognitive scores at first post-stroke cognitive assessment, median (Q1,Q3) | |||
| Global cognitive performance | 52.8 (47.6-58.5) | 51.8 (42.3-58.3) | <0.001 |
| Executive function | 51.9 (41.3-56.1) | 44.6 (38.0-52.6) | <0.001 |
| Memory | 57.4 (49.1-57.4) | 50.0 (45.8-57.4) | 0.022 |
| Time (in years) from stroke to the 1st post-stroke cognitive score, median (Q1,Q3) | |||
| Global cognitive performance | 0.8 (0.4-2.6) | 0.6 (0.3-1.4) | <0.001 |
| Executive function | 1.8 (0.6-4.2) | 1.4 (0.6-2.8) | <0.001 |
| Memory | 1.0 (0.5-2.5) | 0.7 (0.4-1.5) | <0.001 |
| Time (in years) from last pre-stroke cognitive score to stroke, median (Q1,Q3) | |||
| Global cognitive performance | 1.1 (0.5-3.1) | 0.7 (0.3-2.0) | 0.30 |
| Executive function | 2.4 (1.1-4.3) | 1.6 (0.9-3.6) | 0.38 |
| Memory | 1.2 (0.6-3.1) | 0.8 (0.4-2.3) | 0.22 |
Since age at time of stroke was calculated by participant age at cohort baseline + days from cohort baseline visit at time of stroke / 365.25, participants with an age greater than 64 were included in the older age group.
Table 1 shows the number of pre- and post-stroke cognitive assessments, median cognitive score for the first cognitive assessment after stroke and the time from stroke to the first post-stroke cognitive assessment.
Demographic and clinical characteristics differed between the cohorts (Supplemental Table 1).
Association between age and global cognition after stroke
Older age was associated with a greater decline in global cognition initially after stroke and faster decline in global cognition in the years after stroke. Compared to young adults, older age was associated with a −1.69 point (95 % CI, −2.82 to −0.55) greater decline in global cognition initially after stroke. Older adults also had a −0.18 points per year (95 % CI, −0.36 to −0.01) larger change in rate of decline in global cognition, which led to an even larger gap in rate of decline between young and old after stroke compared to before stroke. Every 10 years of higher age was associated with a −1.09 point (95 % CI, −1.52 to −0.65) greater decline in global cognition initially after stroke and a −0.13 points per year (95 % CI, −0.22 to −0.05) larger change in the rate of decline in global cognition over the years following stroke compared to before stroke (Table 2).
Table 2.
Association between age and post-stroke global cognition, memory, and executive function.
| Coefficient | Global Cognitive Performance (N=1,292) |
Memory (N=1,292) |
Executive Function (N=1,222) |
|||
|---|---|---|---|---|---|---|
| Estimate (95 % CI) | P | Estimate (95 % CI) | P | Estimate (95 % CI) | P | |
| Effect of age (older vs young adults) on acute change in cognition (intercept) after stroke | −1.69(−2.82 to −0.55) | 0.004 | −1.05(−1.92 to −0.17) | 0.019 | −3.72(−5.23 to −2.21) | <0.001 |
| Effect of age at stroke (per 10-year increase) on acute change in cognition (intercept) after stroke | −1.09(−1.52 to −0.65) | <0.001 | −0.71(−1.05 to −0.37) | <0.001 | −1.38(−1.95 to −0.81) | <0.001 |
| Effect of age (older vs young adults) on post-stroke cognitive slope, per year | −0.18(−0.36 to −0.01) | 0.03 | −0.006(−0.15 to 0.14) | 0.94 | −0.14(−0.33 to 0.05) | 0.16 |
| Effect of age at stroke (per 10-year increase) on post-stroke cognitive slope, per year | −0.13(−0.22 to −0.05) | 0.003 | −0.04(−0.12 to 0.03) | 0.26 | −0.16(−0.26 to −0.06) | 0.002 |
Young adults are 18-64 years old. Older adults are ≥65 years old. Linear mixed-effect models include pre-stroke cognitive slope, age at time of stroke, age at time of stroke × time since stroke, sex, race × pre-stroke slope, race × post-stroke slope change, cohort, education category, pre-stroke systolic blood pressure, and pre-stroke glucose.
For the 60-year-old male Black exemplar participant with less than a high school education, a mean pre-stroke SBP of 145 mmHg, and mean pre-stroke glucose of 114 mg/dL, the predicted global cognition score decreases by 0.8 points from stroke onset to 2.5 years after stroke (Fig. 2). For the 75-year-old male Black participant with the same education, pre-stroke SBP, and pre-stroke glucose, the predicted global cognition score decreases by 3 points from stroke onset to 2.5 years after stroke (Fig. 2).
Fig. 2.
Change in Global cognition over time after incident stroke.
Initial change in global cognition at time of stroke and change in global cognition in the years after stroke in a stroke survivor at the median age of the younger age group (age=60) and a stroke survivor at the median age of the older age group (age=75).
Association between age and memory after stroke
The older group (65+) was associated with a −1.05 point (95 % CI, −1.92 to −0.17) greater decline in memory initially after stroke compared to the younger group. Younger adults had a −0.18 points per year (95 % CI, −0.31 to −0.05) faster decline in memory after stroke than before stroke. Older adults had a −0.19 points per year (95 % CI, −0.28 to −0.095) faster decline in memory after stroke than before stroke. The change in the slope of memory after stroke, compared to before stroke, did not differ between the young and the old (−0.006 points [95 % CI, −0.15 to 0.14]; p=0.94). Every 10 years of higher age was associated with a −0.71 point (95 % CI, −1.05 to −0.37) greater decline in memory initially after stroke (Table 2). The decline in memory over the years after stroke, compared to before stroke, did not significantly differ by age (−0.04 points [95 % CI, −0.12 to 0.03]; p=0.26).
Association between age and executive function after stroke
The older group (65+) was associated with a −3.72 point (95 % CI, −5.23 to −2.21) greater decline in executive function initially after stroke compared to the younger group, but the rate of decline in executive function after stroke did not differ between the young and the old (−0.19 points per year [95 % CI, −0.35 to −0.03] in the young compared to −0.33 points per year [95 % CI, −0.46 to −0.19] in the old). Every 10 years of higher age was associated with a −1.38 (95 % CI, −1.95 to −0.81) greater decline in executive function initially after stroke and a −0.16 (95 % CI, −0.26 to −0.06) points per year greater decline in executive function after stroke compared to before stroke (Table 2).
Pre-specified subgroup analysis (APOE4)
Having 1 compared to 0 APOE4 alleles was associated with a −0.40 points per year (95 % CI, −0.69 to −0.10) faster decline in global cognition in the years after stroke compared to before stroke. After adjusting for the number of APOE4 alleles, the initial decline in global cognition after stroke did not differ significantly between the binary age groups of the old vs. the young. However, every 10 years of higher age (continuous variable) was associated with a −0.86 points per year (95 % CI, −1.70 to −0.02) greater initial decline in global cognition after stroke. Older age, whether binary or as a continuous variable, was significantly associated with a faster decline in global cognition in the years after stroke compared to before stroke (Table 3).
Table 3.
Sensitivity analysis of the relationship between global cognitive performance and cognitive decline among participants with APOE4 information (N=919).
| Coefficient | Model with Continuous Age |
Model with Categorical Age |
||
|---|---|---|---|---|
| Estimate (95 % CI) | P | Estimate (95 % CI) | P | |
| Effect of age at stroke (older vs young adults) on acute change in cognition (intercept) after stroke | N/A | N/A | −1.49(−3.44 to 0.46) | 0.13 |
| Effect of age at stroke (per 10-year increase) on acute change in cognition (intercept) after stroke | −0.86(−1.70 to −0.02) | 0.05 | N/A | N/A |
| Effect of age at stroke (older vs young adults) on post- stroke cognitive slope, per year | N/A | N/A | −0.38(−0.65 to −0.11) | 0.006 |
| Effect of age at stroke (per 10-year increase) on post- stroke cognitive slope, per year | −0.24(−0.38 to −0.09) | 0.001 | N/A | N/A |
| Effect of 1 APOE4 allele vs. 0 APOE4 alleles on post-stroke cognitive slope | −0.39(−0.68 to −0.09) | 0.01 | −0.40(−0.69 to −0.10) | 0.01 |
| Effect of 2 APOE4 alleles vs. 0 APOE4 alleles on post-stroke cognitive slope | −0.73(−1.54 to 0.07) | 0.07 | −0.70(−1.51 to 0.11) | 0.09 |
Linear mixed-effect models include pre-stroke cognitive slope, age at time of stroke, age at time of stroke × time since stroke, sex, race × pre-stroke slope, race × post-stroke slope change, cohort, education category, pre-stroke systolic blood pressure, pre-stroke glucose, APOE x pre-stroke slope, and APOE x post-stroke slope change.
Discussion
We found that older age was associated with a greater decline in global cognition, memory, and executive function initially after stroke and a faster decline in global cognition and executive function in the years after stroke. Older age was not associated with change in memory in the years after stroke. The associations between older age with initial decline and long-term decline over time in global cognition after stroke were present even after adjusting for APOE4 genotype.
Our findings are consistent with other studies showing older age increases risk for cognitive decline in the years after stroke.2, 3 We extend prior research by showing that older age also increases risk for cognitive decline initially after stroke, even after accounting for the long-term pre-stroke cognitive trajectories (slope). In a smaller study in the REGARDS cohort, older stroke participants had faster decline in global cognition and executive function in the years following stroke than younger stroke participants. In that study, the change in cognition initially after stroke did not differ by age.2 By contrast, we found an association between older age and decline in global cognition, memory, and executive function initially after stroke. The discrepant findings between studies, despite the partially overlapping sample, might be due to our inclusion of a large cohort of younger stroke participants and harmonized cognitive outcomes, which enhanced our ability to detect differences in cognition by age.
Our study was not designed to identify the mechanism by which older age is associated with greater decline in cognition initially after stroke and faster cognitive decline in the years following stroke. However, white matter hyperintensities of presumed vascular origin increase with age and are associated with cognitive impairment.18,19 Perhaps older adults are less able than younger adults to compensate for the effect of the acute stroke lesion on cognition due to older adults’ higher burden of white matter hyperintensities, which increases their risk for cognitive decline and impairment. That APOE4 genotype did not affect the association between older age and global cognition after stroke further suggests that vascular, non-Alzheimer’s disease pathways may be responsible.
We did not find an association between older age and memory decline in the years after stroke. Characteristics of the acute stroke lesion itself might be a greater predictor of memory decline after stroke than age. Left hemispheric, subcortical, and large volume lesions have been associated with poor verbal memory performance 1 year after stroke.20 Future research should investigate whether stroke lesion location is associated with memory function over time and moderation by age.
Strengths of the study are inclusion of a large cohort of younger and older stroke participants, repeated measures of cognition, and pre-stroke cognitive trajectories. Among limitations, we did not have information on stroke severity, stroke location, functional impact of the stroke, or baseline white matter disease, which are factors that have been associated with cognitive impairment after stroke.21, 22 Different cognitive test items assessing memory or executive function may vary in precision. Our estimation of post-stroke cognitive trajectories is limited by the fixed interval of cohort follow-up visits and the variable timing of the strokes themselves. We evaluated cognitive trajectories after stroke, but did not have information on dementia or mild cognitive impairment diagnoses. Prior research has shown an association between age and cognitive impairment and dementia after stroke.21 The initial change in cognition was often measured in the early-mid stage recovery period, which might underestimate initial cognitive decline. There were no participants aged 18–44 years in our cohort. There is limited representation of the younger working age population. Since stroke is rare among those aged 18-44 years, we believe that our younger age group is representative of a young stroke population. Demographic and clinical characteristics of participants varied among the 3 cohorts, but the small sample size of younger participants in each cohort precluded us from performing within-cohort comparisons of cognitive trajectories between the younger and older groups. However, cohort was included as a variable in the model to control for cohort effects. More cognitively impaired stroke participants might be more likely to drop out of longitudinal cohorts; however, such stroke participants are likely older in age and selective attrition of this group would therefore bias our findings towards the null.
Our findings suggest that older age is associated with a greater decline in global cognition, memory, and executive function initially after stroke and faster decline in global cognition and executive function in the years after stroke. Change in memory in the years after stroke did not differ by age. Future research should design interventions to slow cognitive decline after stroke and investigate whether there are differential effects of the intervention by age.
Supplementary Material
Sources of funding
NIH RF1 AG068410 funded this project.
Declaration of competing interest
Authors report funding from the NIH (Springer, Briceno, Burke, Aparicio), Agency for Healthcare Research and Quality (Burke), Alzheimer’s Association (Aparicio), American Academy of Neurology (Aparicio), and the National Institute of Neurological Disorders and Stroke Intramural Research Program (Gottesman).
Footnotes
CRediT authorship contribution statement
Mellanie V. Springer: Writing – review & editing, Writing – original draft, Supervision, Methodology, Conceptualization. Bingxin Chen: Writing – review & editing, Visualization, Validation, Formal analysis, Data curation. Rachael T. Whitney: Writing – review & editing, Data curation. Emily M. Briceño: Writing – review & editing. Alden L. Gross: Writing – review & editing. Hugo J. Aparicio: Writing – review & editing. Alexa S. Beiser: Writing – review & editing. James F. Burke: Writing – review & editing. Bruno Giordani: Writing – review & editing. Rebecca F. Gottesman: Writing – review & editing. Rodney A. Hayward: Writing – review & editing. Virginia J. Howard: Writing – review & editing. Silvia Koton: Writing – review & editing. Ronald M. Lazar: Writing – review & editing. Jeremy B. Sussman: Writing – review & editing. Wen Ye: Writing – review & editing, Visualization, Methodology. Deborah A. Levine: Writing – review & editing, Supervision, Funding acquisition.
Supplementary materials
Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.jstrokecerebrovasdis.2024.108087.
References
- 1.Levine DA, Galecki AT, Langa KM, Unverzagt FW, Kabeto MU, Giordani B, et al. Trajectory of cognitive decline after incident stroke. JAMA. 2015;314(1):41–51. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Levine DA, Wadley VG, Langa KM, Unverzagt FW, Kabeto MU, Giordani B, et al. Risk factors for poststroke cognitive decline: the regards study (reasons for geographic and racial differences in stroke). Stroke. 2018;49(4):987–994. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Lo JW, Crawford JD, Desmond DW, Bae H-J, Lim J-S, Godefroy O, et al. Long-term cognitive decline after stroke: an individual participant data meta-analysis. Stroke. 2022;53(4):1318–1327. [DOI] [PubMed] [Google Scholar]
- 4.Boutros CF, Khazaal W, Taliani M, Sadier NS, Salameh P, Hosseini H. Factors associated with cognitive impairment at 3, 6, and 12 months after the first stroke among Lebanese survivors. Brain Behav. 2023;13(1). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Pinter D, Enzinger C, Gattringer T, Eppinger S, Niederkorn K, Horner S, et al. Prevalence and short-term changes of cognitive dysfunction in young ischaemic stroke patients. Eur J Neurol. 2019;26(5):727–732. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.de Bruijn MA, Synhaeve NE, van Rijsbergen MW, de Leeuw FE, Jansen BP, de Kort PL. Long-term cognitive outcome of ischaemic stroke in young adults. Cerebrovasc Dis. 2014;37(5):376–381. [DOI] [PubMed] [Google Scholar]
- 7.Schaapsmeerders P, Maaijwee NA, van Dijk EJ, Rutten-Jacobs LC, Arntz RM, Schoonderwaldt HC, et al. Long-term cognitive impairment after first-ever ischemic stroke in young adults. Stroke. 2013;44(6):1621–1628. [DOI] [PubMed] [Google Scholar]
- 8.Levine DA, Chen B, Galecki AT, Gross AL, Briceño EM, Whitney RT, et al. Associations between vascular risk factor levels and cognitive decline among stroke survivors. JAMA Netw Open. 2023;6(5), e2313879. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Howard VJ, Cushman M, Pulley L, Gomez CR, Go RC, Prineas RJ, et al. The reasons for geographic and racial differences in stroke study: objectives and design. Neuroepidemiology. 2005;25(3):135–143. [DOI] [PubMed] [Google Scholar]
- 10.Wright JD, Folsom AR, Coresh J, Sharrett AR, Couper D, Wagenknecht LE, et al. The ARIC (Atherosclerosis Risk In Communities) Study: JACC focus seminar 3/8. J Am Coll Cardiol. 2021;77(23):2939–2959. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Feinleib M, Kannel WB, Garrison RJ, McNamara PM, Castelli WP. The framingham offspring study. design and preliminary data. Prev Med. 1975;4(4):518–525. [DOI] [PubMed] [Google Scholar]
- 12.Samejima F. Estimation of latent ability using a response pattern of graded scores. Richmond, VA: Psychometric Society; 1969. [Google Scholar]
- 13.Cho TC, Yu X, Gross AL, Zhang YS, Lee J, Langa KM, et al. Negative wealth shocks in later life and subsequent cognitive function in older adults in China, England, Mexico, and the USA, 2012-18: a population-based, cross-nationally harmonised, longitudinal study. Lancet Healthy Longev. 2023;4(9). e461–e9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Gross AL, Li C, Briceño EM, Arce Rentería M, Jones RN, Langa KM, et al. Harmonisation of later-life cognitive function across national contexts: results from the harmonized cognitive assessment protocols. Lancet Healthy Longev. 2023;4(10): e573–ee83. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Vonk JMJ, Gross AL, Zammit AR, Bertola L, Avila JF, Jutten RJ, et al. Cross-national harmonization of cognitive measures across HRS HCAP (USA) and LASI-DAD (India). PLoS One. 2022;17(2), e0264166. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Briceño EM, Arce Rentería M, Gross AL, Jones RN, Gonzalez C, Wong R, et al. A cultural neuropsychological approach to harmonization of cognitive data across culturally and linguistically diverse older adult populations. Neuropsychol. 2023;37 (3):247–257. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Caselli RJ, Reiman EM, Locke DE, Hutton ML, Hentz JG, Hoffman-Snyder C, et al. Cognitive domain decline in healthy apolipoprotein E epsilon4 homozygotes before the diagnosis of mild cognitive impairment. Arch Neurol. 2007;64(9):1306–1311. [DOI] [PubMed] [Google Scholar]
- 18.Hu HY, Ou YN, Shen XN, Qu Y, Ma YH, Wang ZT, et al. White matter hyperintensities and risks of cognitive impairment and dementia: A systematic review and meta-analysis of 36 prospective studies. Neurosci Biobehav Rev. 2021; 120:16–27. [DOI] [PubMed] [Google Scholar]
- 19.Garnier-Crussard A, Bougacha S, Wirth M, André C, Delarue M, Landeau B, et al. White matter hyperintensities across the adult lifespan: relation to age, Aβ load, and cognition. Alzheimer’s Res Therapy. 2020;12(1). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Schouten EA, Schiemanck SK, Brand N, Post MW. Long-term deficits in episodic memory after ischemic stroke: evaluation and prediction of verbal and visual memory performance based on lesion characteristics. J Stroke Cerebrovasc Dis. 2009; 18(2):128–138. [DOI] [PubMed] [Google Scholar]
- 21.Pendlebury ST, Rothwell PM. Prevalence, incidence, and factors associated with pre-stroke and post-stroke dementia: a systematic review and meta-analysis. Lancet Neurol. 2009;8(11):1006–1018. [DOI] [PubMed] [Google Scholar]
- 22.Koton S, Pike JR, Johansen M, Knopman DS, Lakshminarayan K, Mosley T, et al. Association of ischemic stroke incidence, severity, and recurrence with dementia in the atherosclerosis risk in communities cohort study. JAMA Neurol. 2022;79(3):271. [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.
Supplementary Materials
Data Availability Statement
De-identified participant data can be provided upon cohort and principal investigator approval and with signed data use agreements with cohorts and all relevant institutions.