Predictors of Cognitive Functions After Stroke Assessed Using the Wechsler Adult Intelligence Scale: A Retrospective Study

Abstract

Background:

The mechanism(s) of cognitive impairment remains complex, making it difficult to confirm the factors influencing poststroke cognitive impairment (PSCI).

Objective:

This study quantitatively investigated the degree of influence and interactions of clinical indicators of PSCI.

Methods:

Information from 270 patients with PSCI and their Wechsler Adult Intelligence Scale (WAIS-RC) scores, totaling 18 indicators, were retrospectively collected. Correlations between the indicators and WAIS scores were calculated. Multiple linear regression model(MLR), genetic algorithm modified Back-Propagation neural network(GA-BP), logistic regression model (LR), XGBoost model (XGB), and structural equation model were used to analyze the degree of influence of factors on the WAIS and their mediating effects.

Results:

Seven indicators were significantly correlated with the WAIS scores: education, lesion side, aphasia, frontal lobe, temporal lobe, diffuse lesions, and disease course. The MLR showed significant effect of education, lesion side, aphasia, diffuse lesions, and frontal lobe on the WAIS. The GA-BP included five factors: education, aphasia, frontal lobe, temporal lobe, and diffuse lesions. LR predicted that the lesion side contributed more to mild cognitive impairment, while education, lesion side, aphasia, and course of the disease contributed more to severe cognitive impairment. XGB showed that education, side of the lesion, aphasia, and diffuse lesions contributed the most to PSCI. Aphasia plays a significant mediating role in patients with severe PSCI.

Conclusions:

Education, lesion side, aphasia, frontal lobe, and diffuse lesions significantly affected PSCI. Aphasia is a mediating variable between clinical information and the WAIS in patients with severe PSCI.

Keywords

Alzheimer’s disease cognitive impairment predictor stroke WAIS-R

INTRODUCTION

Post-stroke cognitive impairment (PSCI) refers to the cognitive impairment that occurs within a period after a stroke event and is a subtype of vascular cognitive impairment that includes the subgroups of post-stroke no dementia and post-stroke dementia [1, 2]. Huang et al. summarized previous studies and found that the incidence of PSCI was 38%, with an 18.4% incidence of post-stroke dementia one year after stroke [3]. Additionally, a meta-analysis suggested that one-third of survivors are affected by PSCI [4] and that it is possibly underestimated [5]. The incidence of PSCI, especially post-stroke dementia, increases significantly with time [1 , 6]. At present, it is believed that a definite PSCI diagnosis should be made six months after a stroke, before establishing a rehabilitation scheme [1]. With increasing stroke incidence, PSCI has gradually become a serious public health concern. Compared with the age- and sex-matched population, severe and mild stroke patients have advanced impairment of cognition by 25 years and at least 4 years, respectively [7], thus enduring a seriously reduced quality of life.

Cognitive impairment influences cognitive function as well as motor function and stroke recurrence. Lin et al. found that PSCI may directly affect limb training owing to poor cognition [8]. Aphasia is a PSCI dysfunction, usually associated with dominant hemisphere lesions and limb apraxia [9]. Kwon et al. showed that PSCI is a high-risk factor for ischemic stroke recurrence [10], and Oksalaet al. found that PSCI is an important influencing factor for long-term low survival rates after ischemic stroke [11].

Two research groups performed lesion-symptom mapping studies, and Weaver et al. found a strong association of the left frontotemporal lobe, left thalamus, and right parietal lobe with PSCI [12], and Zhao et al. [13] identified the left angular gyrus, left basal ganglia, and white matter around the left basal ganglia as strategic structures for global cognitive impairment after stroke. Additionally, age [14 –16], female sex [4 , 17], educational level [4 , 16], temporal lobe atrophy [4, 7], family history of dementia [4], previous stroke or TIA [18], multiple infarcts, diabetes [15], a trial fibrillation [4, 18], hypertension/blood pressure variability [19], social support [16], and global cerebral atrophy [7] were highly correlated with PSCI. The risk of PSCI could increase in the decades following a stroke [20, 21]. The PSCI risk factors are not sustained during the rehabilitation process, and it is imperative to refine the understanding of PSCI risk factors and translate these findings to improved outcomes [21].

Few studies have focused on the degree of influence of risk factors or used the Wechsler Adult Intelligence Score-Revised for China (WAIS-RC), which maybe an adequate neuropsychological measurement of cognition after stroke [22] and is considered more of a proxy measure of brain reserve than the Montreal Cognitive Assessment [23].

The degree of influence differs for PSCI factors, and the varying degrees to which each factor relates to PSCI should be confirmed for further research on controlling variables and matching demographic information. However, only a few studies have addressed this issue. This study adopted the WAIS-RC score as the dependent variable and used four models based on two data types, including a continuous regression model and a classification model, to explore and quantify the factors affecting cognitive function.

MATERIALS AND METHODS

Data source

This study retrospectively collected the data of post-stroke patients at the China Rehabilitation Research Center from November 2018 to September 2022. This study was approved by the Medical Ethics Committee of China Rehabilitation Research Center (2021-146-1).

Participants

According to the International Statistical Classification of Diseases and Related Health Problems-Tenth Edition (ICD-10), the diagnoses of patients in this study included Sequelae of subarachnoid hemorrhage (I69.0), Sequelae of intracerebral hemorrhage (I69.1), Sequelae of other nontraumatic intracranial hemorrhage (I69.2), Sequelae of cerebral infarction (I69.3), Sequelae of stroke, not specified as hemorrhage or infarction (I69.4), and Sequelae of other and unspecified cerebrovascular diseases (I69.8).

In this study, the inclusion criteria were as follows: 1) aged 20–75 years and a disease course of at least 2 weeks; 2) a definitive diagnosis of stroke; 3) complete availability of required data; and 4) completion of the WAIS-RC within 10 days of hospitalization. The exclusion criteria were as follows: 1) those with disorders of consciousness and 2) those who did not cooperate for other reasons. We identified 270 patients based on the inclusion and exclusion criteria.

Verbal informed consent was obtained before the interview.

Predictors and outcome

The indicators related to cognitive function after stroke included three sections (overall 18 indicators): demographic information, underlying disease, and stroke-related indicators (Supplementary File 1, Part 1).

The outcome indicator was the aggregate score (also called intelligence quotient, IQ) assessed using the WAIS-RC, comprising two dimensions: verbal intelligence quotient and performance intelligence quotient (Supplementary File 1, Part 2). A therapist with experience of >3 years assessed all patients according to the handbook using the WAIS-RC in a fixed office.

Statistical analysis

Data scoring

Age, disease course, and education were transferred to categorical data. Age ranged from 1–3 for 18–44 years, 45–60 years, and >60 years. Education was categorized into three groups, ranging from 1–3 (<12 years, 12–16 years, and >16 years); similarly, disease course ranged from 1–5 for <1 month, 1–3 months, 3–6 months, 6–12 months, and >12 months.

For logistic regression (LR) and XGBoost models (XGB), WAIS score transferred to three categories of > =90, 70–90, and <70, respectively defined as normal, lower than normal, and cognitive disorder levels. The data transformation rules are listed in Supplementary File 2.

Correlation analysis

A correlation analysis between every factor and the WAIS score was first performed for better modeling. All independent variables were categorical, and indicators with a significant correlation (p < 0.05) with the WAIS scores were included in the model using Spearman correlation analysis.

Model construction

In this study, four models were created to determine the degree of contribution of each factor to cognitive function, including a multiple linear regression model (MLR), genetic algorithm modified Back-Propagation neural network (GA-BP), LR, and XGB models. The first two models require a continuous dependent variable, while the other two require categorical variables. The modeling methods are described in Supplementary File 2.

To further explore the interaction of factors, we analyzed the mediating effect, in which we set dysfunctions as mediating variables, other factors as independent variables, and the WAIS IQ as the dependent variable.

SPSS (version 26.0) was used to construct the MLR and LR, MATLAB 2021b to build the GA-BP neural network, Python 3.11 to construct the XGB model, and Amos 26.0 to perform the mediating effect analysis.

RESULTS

Participants

Of the 496 stroke participants, 270 participants who met the inclusion and exclusion criteria were enrolled (Fig. 1). Demographic information, underlying diseases, and complications of 270 patients are listed in Table 1.

Fig. 1

Research data source and population. We collected data from 496 patients who underwent the WAIS, of whom 294 were post-stroke patients. Demographic or other required information was missing in 17 patients, and seven patients were out of the age range. A total of 270 patients were enrolled in this study.

Table 1

Information of 270 patients and correlation coefficients of all factors to WAIS score

Factors	Total (n = 270)	Correlation	p
		Coefficients
Gender
Male	223(82.59%)	−0.116	0.057
Female	47(17.41%)
Age (y)
20–44	73(27.08%)	0.072	0.241
45–60	113(40.79%)
>60	84(32.13%)
Season
Spring	65(24.07%)	0.09	0.139
Summer	58(21.48%)
Autumn	74(27.41%)
Winter	73(27.04%)
Education (y)
<12	116(43.32%)	0.119	0.05^*
12–16	132(48.38%)
>16	22(8.30%)
Hypertension
None	92(33.21%)	0.032	0.601
<1 y	19(6.86%)
>1 y	162(59.93%)
Hyperlipidemia
None	200(72.20%)	0.11	0.072
<1 y	2(0.72%)
>1 y	73(27.08%)
Diabetes mellitus
None	204(73.65%)	0.073	0.231
<1 y	9(3.61%)
>1 y	61(22.74%)
Course of disease (mo)
<1	53(20.58%)	−0.124	0.042^*
1–3	126(46.57%)
3–6	67(24.19%)
6–12	14(5.05%)
>12	10(3.61%)
Frontal lobe
Not affected	225(83.03%)	−0.154	0.011^*
Affected	45(16.97%)
Temporal lobe
Not affected	223(82.31%)	−0.122	0.045^*
Affected	47(17.69%)
Parietal lobe
Not affected	229(84.12%)	−0.097	0.111
Affected	41(15.88%)
Basal ganglia/Thalamus
Not affected	96(38.63%)	−0.043	0.481
Affected	174(61.37%)
Posterior circulation
Not affected	223(82.67%)	0.064	0.295
Affected	47(17.33%)
Diffuse lesions
Not affected	225(93.14%)	−0.190	0.002^*
Affected	45(6.86%)
Diagnosis
Infarction	151(56.32%)	0.105	0.085
Hemorrhage	119(43.68%)
Side of lesion
Left	128(46.21%)	0.261	0^*
Right	116(41.88%)
Bilateral	21(7.58%)
Others (brain stem and et al)	12(4.33%)
Aphasia
None	166(62.09%)	−0.399	0^*
Complicated	104(37.91%)
Emotional disorders (depression or anxiety)
None	106(40.43%)	0	0.997
Complicated	164(59.57%)
Sleep disorders
None	213(78.34%)	0.022	0.714
Complicated	57(21.66%)
WAIS-RC score
> =90 points	62(22.74%)
70–89 points	85(32.13%)
<70 points	123(45.13%)

^aPosterior circulation: brain stem, cerebellum, and occipital lobe. ^bDiffuse lesions: including subarachnoid hemorrhage and other diffuse lesions. ^*p < 0.05.

Correlation analysis

After correlation analysis, seven factors (education, lesion side, aphasia, frontal lobe, temporal lobe, diffuse lesions, and disease course) significantly correlated with the IQ score (Table 1) and were included in the models.

MLR

After correlation analysis, the independent variable in the MLR included seven items, and the dependent variable was the WAIS score. The homogeneity of variance between independent and dependent variables was verified. The observations were independent of each other (Durbin-Watson = 1.813). The regression tolerances were all greater than 0.1 with no multicollinearity. The regression model was statistically significant (F = 12.053, p < 0.01, adjusted R² = 0.223). Of the seven independent variables included in the model, the effects of education, lesion side, aphasia, diffuse lesions, and frontal lobe on the WAIS score were statistically significant (p < 0.05) (Table 2).

Table 2

Correlation coefficient of variables to WAIS score in MLR model

Variable	coefficient	Standard	Standardized
		Error	Coefficient
Intercept	99.101	7.621^**
Education	4.867	1.700^**	0.155
Lesion side	4.400	1.468^**	0.178
Aphasia	−12.020	2.347^**	−0.298
Frontal lobe	−5.451	2.892^*	−0.106
Temporal lobe	−2.049	2.931	−0.040
Diffuse lesions	−10.420	2.992^**	−0.198
Course of disease	−1.547	1.112	−0.075

^**p < 0.05, ^*p < 0.1.

GA-BP neural network

Data were included in the model for training with 15 neurons and fixed seed values to obtain the influence degree values of the seven independent variables on cognitive function (Table 3). Education, aphasia, frontal lobe, temporal lobe, and diffuse lesions had the greatest impact on the WAIS score and the weights of each part (Supplementary File 1, Part 3) (Training Model R² = 0.404).

Table 3

Importance of seven variables to WAIS score in GA-BP neural network and XGBoost

Variables	Importance
	GA-BP	XGBoost
Education	1.493^*	0.181^*
Lesion side	0.223^*	0.174^*
Aphasia	0.977^*	0.318^*
Frontal lobe	0.801^*	0.05
Temporal lobe	0.203^*	0.045
Diffuse lesions	0.905^*	0.159^*
Course of disease	0.169	0.074

^*p < 0.05.

LR

Two binary LRs in this study were used to assess the effects of seven independent variables on cognitive function. Both models used the normal WAIS score (level 1) as a reference for comparing the contribution of each indicator at levels 2 and 3. In the model comparing levels 1 and 2 (χ² = 47.168, p < 0.001), education and lesions substantially influenced WAIS scores. In the model comparing levels 1 and 3 (χ² = 82.155, p < 0.001), education, lesion side, aphasia, and disease course influenced the WAIS scores primarily (Table 4).

Table 4

OR values of variables

Model 1	OR (95% CI)	p
Education (y)
<12	1.000	<0.001
12–16	0.161(0.062–0.417)	0.069
>16	0.261(0.061–1.109)
Lesion side
Left	1.000	0.002
Right	0.217(0.081–0.577)	0.017
Bilateral	0.060(0.006–0.609)
Model 2
Education (y)
<12	1.000	<0.001
12–16	0.117(0.045–0.305)	0.007
>16	0.109(0.022–0.541)
Lesion side
Left	1.000	0.003
Right	0.224(0.084–0.599)	0.021
Bilateral	0.182(0.043–0.773)
Aphasia
None	1.000	<0.001
Complicated	7.251(2.703–19.451)
Course of disease (mo)		0.021
<1	1.000
3–6	4.046(1.240–13.204)

Model 1 correctly classified 79.5% of participants. The sensitivity and specificity were respectively 84.7% and 59.7%, the positive predictive value was 74.2% and the negative predictive values was 74.0%. Model 2 correctly classified 79.5% of the study participants. The sensitivity and specificity were respectively 85.4% and 67.7%, the positive predictive value was 84.0% and the negative predictive value was 70.0%.

XGB

With all the data included in the training model, the importance of the respective variables was obtained from the model, as shown in Table 3. Education, lesion side, aphasia, and diffuse lesions were the variables that most influenced WAIS scores.

Mediating effect

A structural equation was built to analyze the mediating effect, in which aphasia was set as the mediating variable, the WAIS score as the dependent variable, and the other factors as the independent variables (Fig. 2). There are significant indirect and total effects in this equation. To further explore the mediating effects of factors on WAIS, all patients were divided into two groups: mild to moderate cognitive impairment (WAIS> = 70) group (G1) and severe cognitive impairment (WAIS < 70) group (G2). In G1, only a significant total effect was observed in the equation, and the mediating effect of aphasia was unclear. In G2, the mediating effect of aphasia was significant; however, the indirect and total effects were unclear (Table 5 and Supplementary File 1, Part 4).

Fig. 2

Structure equation to analyze mediating effect. Mediating variable: aphasia; independent variables: education, lesion side, frontal lobe, temporal lobe, diffuse lesions, and disease course ‘a,’ ‘b,’ and ‘c’ in the equation represent the effects from factors on aphasia or WAIS score. Indirect effect = a * b; total effect = a * b + c; rate = indirect / total effect.

Table 5

Indirect and total effects of factors on WAIS score

	Estimate	Lower	Upper	p
Indirect effect	−18.211	−254.049	−8.145	0.001
Total effect	−53.633	−643.467	−25.220	<0.001
rate	0.340	0.087	1.246	0.023
G1
Indirect effect	1.794	−11.477	245.254	0.645
Total effect	−53.391	−512.247	−12.428	0.008
rate	−0.034	−1.900	0.348	0.684
G2
Indirect effect	2.863	−0.358	∞	0.092
Total effect	5.901	−8.900	∞	0.307
rate	0.485	0.023	∞	0.038

DISCUSSION

Based on the WAIS, this study used four common prediction models to analyze the prognostic factors of PSCI. The factors significantly impacting WAIS scores in the MLR and LR models, those with more than 0.2 impacting points in the GA-BP neural network and more than 0.1 impacting points in the XGB model, were defined as the influencing factors. Finally, the following four factors considerably influenced the outcome: education, stroke side, aphasia, and diffuse lesions. Furthermore, the frontal lobe and disease course significantly influenced the WAIS score. A high education level, lesions in the right hemisphere, short disease course, no aphasia, lesions not in the frontal or temporal lobes, or diffuse lesions are protective factors against cognitive impairment.

Education [24 –29], stroke side [13 , 31], aphasia [4 , 33], and diffuse brain injury [34, 35] were all indicators that significantly affected the outcomes. In addition, frontal and temporal lobe injuries have a greater impact on outcomes [12, 36]. Cognitive reserve has a significant impact on cognitive impairment after stroke, and the cognitive reserve of patients with high educational levels is relatively good [24 , 28]. Weaver et al. showed that left frontal-temporal lobe infarction is a significant risk factor for PSCI, demonstrating the influence of laterality and location on cognitive function [12]. Except for the factors above, lesions in the thalamus, hypertension, diabetes, previous stroke, and other relevant medical histories are also risk factors for PSCI [37]. According to this study, some factors may contribute to PSCI; however, no significant differences were found, and further exploration is needed.

In this study, we used multinomial logistic regression models with 3-categorical variables to analyze the predictive factors, but the p-value of the Test of Parallel Lines was 0.002, indicating that the 3-category regression models were not parallel. The influencing factors of different models may differ substantially. Therefore, the authors divided the model into two groups for analysis: normal mild cognitive impairment (level 1 versus level 2) and normal severe cognitive impairment (level 1 versus level 3). Aphasia, closely related to the lesion location and laterality, is a prominent factor in patients with severe cognitive impairment [32, 33]. Aphasia is a functional impairment after stroke, which significantly impacts cognitive function and assessment [4 , 33] and thus may have a masking effect. In clinical practice, differentiating between sensory aphasia and cognitive impairment poses a problem for rehabilitation doctors [4 , 21]. Therefore, it is suggested that aphasia should be assessed first in patients with obvious aphasia, followed by a cognitive function assessment and evaluation of their interactions.

The mechanisms of occurrence and rehabilitation of PSCI are complex. The PSCI includes attention, memory, visual space, language, and executive control [1, 9]. Each function can realize a cognitive process by closely cooperating with the others, but the mechanism is unclear [9 , 21], which is different from other functional disorders, such as motor disorders, with relatively distinct neural pathways. Analyzing the prognosis of cognitive impairment and establishing a predictive model requires further work.

Limitation

This study had several limitations: 1) Although the WAIS-RC scale has a high degree of discrimination and accuracy in assessing overall cognitive function with overall cognitive function as the outcome of this study, it is not perfect in each subtest. For example, the total score has age correction, but each sub-test does not, and the influencing factors of various dimensions of cognition are not well considered, so there may be deviations; it is suggested to select a specific scale for evaluating a cognitive dimension for influencing factor analysis; 2) the sample size of this study is not enough, and subsequent research should be based on an in-depth analysis of large sample size; 3) approximately 10% of patients had cognitive impairment before the first stroke [4], necessitating the inclusion of a healthy control group in the clinical study; 4) Although prognostic factors can be obtained, the contribution of these factors to the WAIS score is low (R² = 0.223 in the MLR model). We believe more influential factors for cognitive function prognosis exist, such as the disease severity. Pendlebury et al. confirmed that the incidence of dementia in patients with severe stroke (NIHSS score > 10) was significantly higher than that in patients with mild stroke (NIHSS score < 3) and TIA one year after onset [7]. In the next step, we need a larger sample size, including evaluation indicators of various dimensions and more detailed patient data to clarify the influencing factors of cognitive impairment.

AUTHOR CONTRIBUTIONS

Wenlong Su (Conceptualization; Investigation; Formal analysis; Writing – Original Draft, Review & Editing; Funding acquisition); Hui Li (Conceptualization; Investigation; Formal analysis; Writing – Review & Editing); Hui Dang (Conceptualization; Investigation; Writing – Review & Editing); Kaiyue Han (Conceptualization; Investigation; Writing –Review & Editing); Jiajie Liu (Conceptualization; Formal analysis; Writing – Review & Editing); Tianhao Liu (Conceptualization; Writing – Original Draft, Review & Editing); Ying Liu (Conceptualization; Supervision; Writing – Review & Editing); Zhiqing Tang (Conceptualization; Writing – Review & Editing); Haitao Lu (Conceptualization; Investigation; Funding acquisition; Writing – Review & Editing); Hao Zhang (Conceptualization; Funding acquisition; Supervision).

Footnotes

ACKNOWLEDGMENTS

We would like to thank the Research Program of the China Rehabilitation Research Center, Joint Doctoral Program of the University of Health and Rehabilitation Sciences, Capital Medical University, and Capital’s Funds for Health Improvement and Research. We thank the clinicians and other staff in the Department of Neurological Rehabilitation and Department of Psychology at the China Rehabilitation Research Center for their support with data collection.

FUNDING

This study was funded by the Joint Doctoral Program of the University of Health and Rehabilitation Sciences and Capital Medical University (2020kfdx-010), Research Program of China Rehabilitation Research Center (Grant No. 2021zx-19), and Capital’s Funds for Health Improvement and Research (Grant No. 2020-1-6011).

CONFLICT OF INTEREST

The authors have no conflict of interest to report.

DATA AVAILABILITY

Data supporting the findings of this study are available at [] at [DOI: 10.6084/m9.figshare.22736645].

The code of BP neural network and XGBoot model can be found in .

The supplementary material is available in the electronic version of this article: .

References

Mijajlović

, Pavlović

, Brainin

, Heiss

W-D

, Quinn

, Ihle-Hansen

, Hermann

, Assayag

, Richard

, Thiel

, Kliper

, Shin

Y-I

, Kim

Y-H

, Choi

, Jung

, Lee

Y-B

, Sinanović

, Levine

, Schlesinger

, Mead

, Milošević

, Leys

, Hagberg

, Ursin

, Teuschl

, Prokopenko

, Mozheyko

, Bezdenezhnykh

, Matz

, Aleksić

, Muresanu

, Korczyn

, Bornstein

(2017) Post-stroke dementia –a comprehensive review, BMC Med 15, 11.

Rundek

, Tolea

, Ariko

, Fagerli

, Camargo

(2022) Vascular cognitive impairment (VCI), Neurotherapeutics 19, 68–88.

Huang

Y-Y

, Chen

S-D

, Leng

X-Y

, Kuo

, Wang

Z-T

, Cui

, Tan

, Wang

, Dong

, Yu

J-T

(2022) Post-stroke cognitive impairment: Epidemiology, risk factors, and management, J Alzheimers Dis 86, 983–999.

Pendlebury

, Rothwell

(2009) Prevalence, incidence, and factors associated with pre-stroke and post-stroke dementia: A systematic review and meta-analysis, Lancet Neurol 8, 1006–1018.

Pendlebury

, Chen

P-J

, Welch

SJV

, Cuthbertson

, Wharton

, Mehta

, Rothwell

(2015) Methodological factors in determining risk of dementia after transient ischemic attack and stroke: (II) Effect of attrition on follow-up, Stroke 46, 1494–1500.

Levine

, Galecki

, Langa

, Unverzagt

, Kabeto

, Giordani

, Wadley

(2015) Trajectory of cognitive decline after incident stroke, JAMA 314, 41–51.

Pendlebury

, Rothwell

(2019) Incidence and prevalence of dementia associated with transient ischaemic attack and stroke: Analysis of the population-based Oxford Vascular Study, Lancet Neurol 18, 248–258.

Lin

, Erler

, Snider

, Bonkhoff

, DiCarlo

, Lam

, Ranford

, Parlman

, Cohen

, Freeburn

, Finklestein

, Schwamm

, Hochberg

, Cramer

(2021) Cognitive demands influence upper extremity motor performance during recovery from acute stroke, Neurology 96, e2576–e2586.

Cramer

, Richards

, Bernhardt

, Duncan

(2023) Cognitive deficits after stroke, Stroke 54, 5–9.

10.

Kwon

, Lee

, Yu

, Lim

J-S

, Yu

K-H

, Oh

, Lee

J-S

, Hong

K-S

, Lee

E-J

, Kang

D-W

, Kwon

(2020) Post-stroke cognitive impairment as an independent predictor of ischemic stroke recurrence: PICASSO sub-study, J Neurol 267, 688–693.

11.

Oksala

NKJ

, Jokinen

, Melkas

, Oksala

, Pohjasvaara

, Hietanen

, Vataja

, Kaste

, Karhunen

, Erkinjuntti

(2009) Cognitive impairment predicts poststroke death in long-term follow-up, J Neurol Neurosurg Psychiatry 80, 1230–1235.

12.

Weaver

, Kuijf

, Aben

, Abrigo

, Bae

H-J

, Barbay

, Best

, Bordet

, Chappell

, Chen

CPLH

, Dondaine

, van der Giessen

, Godefroy

, Gyanwali

, Hamilton

OKL

, Hilal

, HuengesWajer

IMC

, Kang

, Kappelle

, Kim

, Köhler

, Kort de

PLM

, Koudstaal

, Kuchcinski

, Lam

BYK

, Lee

B-C

, Lee

K-J

, Lim

J-S

, Lopes

, Makin

SDJ

, Mendyk

A-M

, Mok

VCT

, Oh

, van Oostenbrugge

, Roussel

, Shi

, Staals

, Del Valdés-Hernández

, Venketasubramanian

, Verhey

FRJ

, Wardlaw

, Werring

, Xin

, Yu

K-H

, van Zandvoort

MJE

, Zhao

, Biesbroek

, Biessels

(2021) Strategic infarct locations for post-stroke cognitive impairment: A pooled analysis of individual patient data from 12 acute ischaemic stroke cohorts, Lancet Neurol 20, 448–459.

13.

Zhao

, Biesbroek

, Shi

, Liu

, Kuijf

, Chu

, Abrigo

, Lee

, Leung

, Lau

, Biessels

, Mok

, Wong

(2018) Strategic infarct location for post-stroke cognitive impairment: A multivariate lesion-symptom mapping study, J Cereb Blood Flow Metab 38, 1299–1311.

14.

Aben

, Munter

L de

, Reijmer

, Spikman

, Visser-Meily

JMA

, Biessels

, Kort

PLM de

(2021) Prediction of cognitive recovery after stroke: The value of diffusion-weighted imaging-based measures of brain connectivity, Stroke 52, 1983–1992.

15.

Ding

M-Y

, Xu

, Wang

Y-Z

, Li

P-X

, Mao

Y-T

, Yu

J-T

, Cui

, Dong

(2019) Predictors of cognitive impairment after stroke: A prospective stroke cohort study, J Alzheimers Dis 71, 1139–1151.

16.

O’Sullivan

, Li

, Galligan

, Pendlebury

(2023) Cognitive recovery after stroke: Memory, Stroke 54, 44–54.

17.

Pavlovic

, Pekmezovic

, Mijajlovic

, Tomic

, ZidvercTrajkovic

(2022) Is the female sex associated with an increased risk for long-term cognitive decline after the first-ever lacunar stroke? Prospective study on small vessel disease cohort, Front Neurol 13, 1052401.

18.

Pendlebury

(2009) Stroke-related dementia: Rates, risk factors and implications for future research, Maturitas 64, 165–171.

19.

Lee

K-P

, Chang

AYW

, Sung

P-S

(2021) Association between blood pressure, blood pressure variability, and post-stroke cognitive impairment, Biomedicines 9, 773.

20.

Aam

, Gynnild

, Munthe-Kaas

, Saltvedt

, Lydersen

, Knapskog

A-B

, Ihle-Hansen

, Ellekjær

, Eldholm

, Fure

(2021) The impact of vascular risk factors on post-stroke cognitive impairment: The Nor-COAST Study, Front Neurol 12, 678794.

21.

Rost

, Brodtmann

, Pase

, van Veluw

, Biffi

, Duering

, Hinman

, Dichgans

(2022) Post-stroke cognitive impairment and dementia, Circ Res 130, 1252–1271.

22.

, Lu

, Li

, Tang

, Dang

, Han

, Li

, Liu

, Zhang

(2023) Characteristics of cognition impairment in patients after stroke based on the Wechsler Adult Intelligence Scale-Revised in China, Appl Neuropsychol Adult. doi:10.1080/23279095.2023.2205023.

23.

Elkana

, Soffer

, Eisikovits

, Oren

, Bezalel

, Ash

(2020) WAIS Information Subtest as an indicator of crystallized cognitive abilities and brain reserve among highly educated older adults: A three-year longitudinal study, Appl Neuropsychol Adult 27, 525–531.

24.

Umarova

, Sperber

, Kaller

, Schmidt

CSM

, Urbach

, Klöppel

, Weiller

, Karnath

H-O

(2019) Cognitive reserve impacts on disability and cognitive deficits in acute stroke, J Neurol 266, 2495–2504.

25.

Contador

, Alzola

, Stern

, La Torre-Luque de

, Bermejo-Pareja

, Fernández-Calvo

(2023) Is cognitive reserve associated with the prevention of cognitive decline after stroke? A systematic review and meta-analysis, Ageing Res Rev 84, 101814.

26.

Hamilton

OKL

, Backhouse

, Janssen

, Jochems

ACC

, Maher

, Ritakari

, Stevenson

, Xia

, Deary

, Wardlaw

(2021) Cognitive impairment in sporadic cerebral small vessel disease: A systematic review and meta-analysis, Alzheimers Dement 17, 665–685.

27.

Ojala-Oksala

, Jokinen

, Kopsi

, Lehtonen

, Luukkonen

, Paukkunen

, Seeck

, Melkas

, Pohjasvaara

, Karhunen

, Hietanen

, Erkinjuntti

, Oksala

(2012) Educational history is an independent predictor of cognitive deficits and long-term survival in postacute patients with mild to moderate ischemic stroke, Stroke 43, 2931–2935.

28.

Shin

, Sohn

, Lee

, Kim

, Lee

S-G

, Shin

Y-I

, Oh

G-J

, Lee

Y-S

, Joo

, Han

, Ahn

, Chang

, Shin

, Choi

, Kang

, Kim

Y-H

(2020) Effect of cognitive reserve on risk of cognitive impairment and recovery after stroke: The KOSCO Study, Stroke 51, 99–107.

29.

Baccaro

, Wang

Y-P

, Brunoni

, Candido

, Conforto

, Da Costa Leite

, Lotufo

, Benseñor

, Goulart

(2019) Does stroke laterality predict major depression and cognitive impairment after stroke? Two-year prospective evaluation in the EMMA study, Prog Neuropsychopharmacol Biol Psychiatry 94, 109639.

30.

Gong

, Gu

, Yu

, Wang

, Zhu

, Dong

, Xu

, Zhao

, Liu

(2020) Prognostic factors for cognitive recovery beyond early poststroke cognitive impairment (PSCI): A prospective cohort study of spontaneous intracerebral hemorrhage, Front Neurol 11, 278.

31.

Aamodt

, Lydersen

, Alnæs

, Schellhorn

, Saltvedt

, Beyer

, Håberg

(2022) Longitudinal brain changes after stroke and the association with cognitive decline, Front Neurol 13, 856919.

32.

Cumming

, Marshall

, Lazar

(2013) Stroke, cognitive deficits, and rehabilitation: Still an incomplete picture, Int J Stroke 8, 38–45.

33.

Lee

, Pyun

S-B

(2014) Characteristics of cognitive impairment in patients with post-stroke aphasia, Ann Rehabil Med 38, 759–765.

34.

Del Brutto

, Mera

, Zambrano

, Lama

(2015) Cognitive performance in community-dwelling older adults with stroke and the contribution of age and diffuse subcortical damage: A population-based study in rural Ecuador (The Atahualpa Project), Aging Clin Exp Res 27, 647–652.

35.

Man

, Fu

, Wong

, Chan

, Lam

, Hui

ACF

, Leung

, Mok

, Wong

(2011) Cognitive and functional impairments in ischemic stroke patients with concurrent small vessel and large artery disease, Clin Neurol Neurosurg 113, 612–616.

36.

Kandiah

, Wiryasaputra

, Narasimhalu

, Karandikar

, Marmin

, Chua

, Sitoh

Y-Y

(2011) Frontal subcortical ischemia is crucial for post stroke cognitive impairment, J Neurol Sci 309, 92–95.

37.

Zhang

, Wang

, Li

, Pan

, Wang

, Liao

, Shi

, Wang

(2022) Infarct location and cognitive change in patients after acute ischemic stroke: The ICONS study, J Neurol Sci 438, 120276.