Hindawi

Complexity
Volume 2021, Article ID 5569346, 12 pages
https://doi.org/10.1155/2021/5569346

Review Article
Meta-Analysis on Cognitive Benefit of Exercise after Stroke

Bo Yang and Shuming Wang

College of Physical Education & Health, East China Normal University, Shanghai 200241, China

Correspondence should be addressed to Shuming Wang; [email protected]

Received 19 January 2021; Revised 18 March 2021; Accepted 23 March 2021; Published 7 April 2021

Academic Editor: Zhihan Lv

Copyright © 2021 Bo Yang and Shuming Wang. This is an open access article distributed under the Creative Commons
Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is
properly cited.

ObjectiveThe objective of this paper is to evaluate the cognitive benefit of exercise after stroke, so as to provide more accurate and
reliable guidance for targeted exercise intervention. Methods. Randomized controlled trials of the relationship between exercise
and cognition after stroke were identified in Cochrane Library and PubMed. Methodological quality was assessed using the
Cochrane tool of bias. SMD and 95% confidence intervals were calculated, and Chi-squared test (Q) was adopted to estimate the
heterogeneity. Results. (a) Twenty-three studies met inclusion criteria, including 1528 participants. Heterogeneity was from low to
high such as attention (I2 � 0.00%), executive function (I2 � 0.00%), cognition (I2 � 64%), and working memory (I2 � 77%). (b) The
overall effect on cognition was small (SMD � 0.16 [0.04, 0.28]) but significant and there is a difference between cognitive domains
in attention (SMD � −0.35 [−0.57, −0.14]), executive function (SMD � −0.24 [−0.40, −0.08]), and working memory (SMD � 0.36
[0.20, 0.53]). (c) Exercise training was less effective before the 18th month after stroke. Higher benefits on cognition were found
after combined therapy compared with other exercise programs, and the older the stroke survivors, the less the cognitive benefit of
exercise. Conclusion. Small-to-moderate effect of exercise on cognitive benefit after stroke was found, and the effect was moderated
by treatment protocols and sample characteristics.

1. Introduction neurotransmission are mainly adopted for functional

recovery after stroke. Serotonin reuptake inhibitors, for
Cognitive and motor dysfunction is a common symptom example, can facilitate synaptic plasticity and functional
after stroke; more than 80% of the patients have different connections between synapses by increasing the con-
degrees of cognitive and limb motor impairment during 3 to centration of serotonin transporters between synapses of
12 months after stroke [1], which is associated with memory neurons, thus laying a physiological foundation for the
decline, attention deficits, and executive dysfunction [2]. recovery of sensorimotor and long-term memory function
Although survivors can get timely treatment, sedentary [5]. On the other hand, Amphetamine-like agents, py
behavior caused by long recovery cycle of limb function can promoting the release of epinephrine, norepinephrine,
further lead to anxiety, depression, and other bad emotional and dopamine from the presynaptic terminal, accelerate
states, which in turn hinder the recovery process, such as the process of functional recovery after central nervous
poor adherence to rehabilitation plan, high recurrence rate system injury [6]. However, studies have pointed out that
of stroke, and high mortality involved with suicide [3]. [7] long-term use of Amphetamine can also cause adverse
Blood supply disorders associated with ischemic or side effects, such as increased blood pressure and heart
hemorrhagic stroke can provoke hypoxic death of nerve rate, and it is controversial for adverse symptoms such as
cells, altering neuronal structures and causing partial or tension, insomnia, addiction, or anorexia after rehabili-
complete loss of neuronal function [4]. Accordingly, tation. Other pharmacotherapies like antipsychotics and
timely treatment during the critical period after stroke will antiepileptic drugs not only aggravate muscle stiffness but
greatly affect the neural plasticity and functional recovery also cause metabolic abnormalities and unpredictable
of the brain. At present, drugs that modulate toxic effects [8].
2 Complexity

Given that there is a lack of ideal pharmacotherapy to corresponding author. The keywords of the search are as
reverse the decline of cognitive function after stroke, it is follows: (physical activity OR TaiChi OR physiotherapy OR
imperative to find a cheap and reliable alternative. As the Ba duan jin OR fitness OR aerobic exercise OR dance OR
cognitive benefit of exercise intervention has also been resistance training) AND (cognitive function OR cognition
confirmed in animal models [9] and elderly people with OR attention OR memory OR executive function OR
cognitive dysfunction [10], physical exercise and cognitive neuropsychological test) AND (stroke OR cerebrovascular
training are considered the primary treatments for im- accident OR brain ischemia OR poststroke OR post stroke).
proving age-related cognitive decline [11]. Early studies have
shown that exercise intervention after stroke can effectively
avoid further cognitive function decline [12]; a recent study 2.2. Study Selection. Selection criteria are as follows: (1)
also showed that moderate-to-high-intensity exercise will stroke survivors (e.g., ischemic or hemorrhagic stroke) older
effectively promote cognitive function recovery after stroke, than 18 years of age; (2) randomized controlled trials, in-
improving the level of cardiopulmonary adaptation [13] and cluding exercise type, frequency, intensity, session duration,
facilitating the speed and endurance of physical activity, and intervention length, mean age, and course of disease; (3)
then will reduce risk of recurrence of stroke and cardio- existence of baseline and postintervention reports prior to
vascular events [14]. In fact, the cognitive benefit of exercise the experiment (for unmonitored training studies (i.e.,
was mainly induced by increased cerebral blood flow and family interventions), changes in body function before and
neurogrowth factor secretion (BDNF、IGF-1), so as to after exercise training must be observed, such as physical
facilitate structural and functional recovery of damaged fitness improvement); and (4) excluding conference papers,
parts of the brain [15] then restore cognitive function while research reviews, non-RCT research, inconsistent research
raising the level of brain arousal, and ultimately strengthen content, duplication of literature, and book chapters. Finally,
the ability of the brain in limbs coordination and control and 23 articles, including 24 independent studies, were included
ease anxiety and depression caused by cognitive and motor [2, 14, 18–38] (Figure 1).
loss [16]. Interestingly, the latest meta-analysis showed that
[17] although exercise is helpful in cognitive function re- 2.3. Data Collection and Extraction. One author performed
covery after stroke, the improvement was not reflected in all the initial articles search according to the title and abstract of
aspects; only attention and cognitive processing speed were the article. Two authors extracted information indepen-
significantly improved. Working memory and executive dently, including (a) demographic information (number of
dysfunction were not included. Besides, cognitive im- total participants in experimental and control groups, sex,
provement did not occur in all types of exercise intervention, and the mean age and age range of participants in each
and only the combination of aerobic exercise and physio- group); (b) exercise intervention features (exercise type in
therapy was shown to be effective. This completely con- each group, exercise frequency in each group, intensity of
tradicts the conclusion of a recent research [18]. In addition, physical exercise for two groups, session duration, inter-
since previous studies were based on small samples size and vention length, course of disease, drugs taken during trials
restricted to specific exercise types and sample character- and tests, adherence, dropouts, and adverse effect); (c)
istics, an updated quantitative exercise recommendation for outcome evaluation tools, including Mini-Mental State
cognitive promotion in stroke survivor is needed. Examination, Wechsler Memory Scale, Functional Inde-
Here, we provide a meta-analysis on cognitive benefit of pendence Measure, Montreal Cognitive Assessment, Trial
exercise in stroke survivors, examining the specific cognitive Making Part B, Digit Span, Stroop Test, Addenbrooke’s
benefit on executive function, attention, and working Cognitive Examination Revised, Stroke Impact Scale, Go/
memory. Additionally, sample characteristics such as session No-go, and Dual Task; and (d) the results of each cognitive
duration, frequency, intervention length, type of exercise, test in pre- and postexercise intervention which were cat-
and the time from stroke onset to initiation of the inter- egorized in the light of cognitive function they tested, in-
vention, which may moderate the exercise-cognition rela- cluding the mean scores or the mean change of the overall
tionship, have also been rated, so as to identify the optimal cognitive function, working memory, attention. and exec-
exercise training dose parameters and then maximize cog- utive function. For some studies which only show the result
nitive benefit of exercise after stroke. of standard deviation or standard error of these values, we
calculate it by specific formula. Besides, the numbers of
2. Methods participants in experiment and control groups were also
abstracted. Finally, e-mail will be sent to the author when
2.1. Search Criteria. Based on existing meta-analysis (from there is no sufficient information in all included studies to
July 2001 to 2017), our initial literature search retrieved 405 calculate effect sizes. Disagreements were resolved by dis-
records from the Cochrane Library and PubMed (from July cussion and consensus with corresponding author of the
2017 to January 2021) under the condition of independent team.
double-blind experiment, where 183 remained after deleting
inconsistent researches. There were 23 articles that fulfilled
all eligibility criteria and they were included after excluding 2.4. Quality Assessment. Two authors assessed the meth-
55 duplicate articles, 63 reviews, 10 meta-analyses, and 22 odological quality of included trials by the Cochrane risk of
articles with incomplete data after contact with the bias tool. The quality assessment was performed by the
Complexity 3

Identification
Identified through Identified through
Cochrane and PubMed latest meta-analysis
N = 405 (2017-2021) N = 14 (2001-2017)

Irrelevant researches
419 records screened
Screening
excluded N = 236

183 full-text articles assessed for eligibility

55 duplicates removed 29 studies

without results excluded
Eligibility

63 systematic
10 meta-analyses excluded
reviews excluded

Studies included in qualitative synthesis (N = 23)

Inclusion

Studies included in qualitative synthesis

(meta-analysis) (N = 23)

Figure 1: PRISMA flow diagram of included studies.

software Review Manager 5.3, which sets six criteria frequency (2 to 3 times per week and 5 times per week),
according to the responses “low risk,” “high risk,” and intensity (moderate or acute), session duration (≤30 min,
”unclear risk” to the following domains: random allocation, 45–60 min, 75–90 min, and ≥90 Min), and intervention
allocation concealment, blinding of participants, blinding of length (＜4 weeks, 6–10 weeks, 12–14 weeks, and ≥14
outcome assessment, incomplete outcome data, and selec- weeks) were conducted. In addition, three separate meta-
tive reporting. analyses on subdomains of cognition were performed:
working memory, attention, and executive functions.
Last, funnel plot asymmetry was used to detect whether
2.5. Data Analysis. Statistical analysis was performed using there was publication bias.
Review Manager 5.3 and Stata 15.1. Exercise effects of each
study were represented by standardized mean differences
3. Results
(SMD) and confidence intervals (CI) in condition of fixed-
effects or mixed-effects model, and only the baseline data 3.1. Characteristics of Included Studies. After systematic
and the data at the end of the intervention were used for retrieval, 23 studies met all inclusion criteria, including
effect sizes calculation. Effect sizes were interpreted as small 1528 participants with mean age of 62.2 years (SD: 6.2
(≤0.2), moderate (0.5), or large (≥0.8). We calculated years). The mean trial length was 11.3 weeks (SD: 5.6
standard deviation (SD) from the standard error (SE; weeks), and exercise frequency was divided into 2 to 3 times
SD � SE√n) if it is not available in the article. Q statistics and per week and 5 times per week. The mean duration of each
I2 were calculated for heterogeneity evaluation between session was about 69.3 minutes (SD: 45.6 minutes). To
studies. assess the influence of time from stroke to intervention, the
Using a fixed-effects model, we performed an overall course of disease before intervention was also investigated,
meta-analysis on cognitive benefit of exercise; as for trials which was about 20.8 months (SD: 22.1 months). Re-
with multiple cognitive outcomes, effect sizes were cal- garding types of exercise, three types of exercise inter-
culated separately for each domain. Studies were distin- vention were adopted: aerobic exercise, physiotherapy, and
guished in terms of sample characteristics, intervention combination of aerobic exercise/physiotherapy/cognition
details, and outcome measures; and subgroup analysis was therapy. The control group was treated with general care,
conducted for several characteristics. First, two subgroup walking, passive stretching, and social communication
analyses on subdomains of sample characteristics such as (Table 1).
age group (49–55 Y, 56–60 Y, 60–65 Y, and ＞65 Y) and
course of disease (＜3 months after stroke, 3–6 months
after stroke, 10–18 months after stroke, and ≥18 months 3.2. Quality Assessment. All studies included were RCT
after stroke) were conducted. Second, five subgroup an- studies, and two studies reached the criteria of high meth-
alyses on subdomains of study design including exercise odological quality assessments in six aspects. Three studies
type (aerobic exercise, physiotherapy, and combination of were designed by independent double-blind experiment; one
aerobic exercise/physiotherapy/cognition therapy), was nonblind design and another one was unknown as
4 Complexity

Table 1: Characteristics of the included studies.

Time from
Age Trial Session Cognitive
Study Number stroke to Frequency Exercise mode Control type
(mean) length duration outcomes
int.
Nilsson Ex: 36; Ex: 54.0; Track walking with
＜1 M 9.5 W 30 min 5t/W Aerobic ③
[19] Con: 37 Con: 56.0 physiotherapist
Ozdemir Ex: 30; Ex: 59.1; Conventional PA at
1.28 M 9.0 W 120 min 5t/W Physiotherapy ①
[20] Con: 30 Con: 61.8 home
Ex: 78; Ex: 65.6; Conventional PA at
Fang [21] ≤0.23 M 4.0 W 45 min 5t/W Physiotherapy ①
Con: 78 Con: 61.8 home
Studenski Ex: 44; Ex: 68.5; Conventional PA at
2.53 M 1 2W 30 min 3t/W Physiotherapy ③
[22] Con: 49 Con: 70.4 home
Ex: 25; Ex: 66.2; Conventional PA at
Chen [23] ＜6 M 1 2W 45–60 min 2t/W Physiotherapy ①
Con: 20 Con: 67.3 home
Ex: 32; Ex: 72.0; Progressive muscle
Mead [24] 5.67 M 1 2W 75 min 3t/W Combined ③
Con: 34 Con: 71.7 relaxation
Quaney Ex: 19; Ex: 64.1; Home-based
58.8 M 8.0 W 45 min 3t/W Aerobic ②⑤⑦
[25] Con; 19 Con: 59.0 stretching exercises
Tamawy Ex: 15; Ex: 48.4; Home-based
3–18 M 8.0 W 75 min 3t/W Combined ⑧
[26] Con: 15 Con: 49.7 stretching exercises
Immink Ex: 11; Ex: 56.1; Conventional PA at
52.5 M 1 0W 40–90 min 1t/W Yoga ⑨
[27] Con: 11 Con: 63.2 home
Ex: 20; Ex: 68.0;
Moore [28] 18.5 M 1 9W 45–60 min 3t/W Combined Stretching training ⑧
Con: 20 Con: 70.0
Ex: 11; Ex: 62.9; Conventional PA at
Liu [29] 32.4 M 2 4W 60 min 2t/W Combined ⑦⑤⑥
Con: 14 Con: 66.9 home
Schachten Ex: 7.0; Ex: 55.1;
44.0 M 1 0W 60 min 2t/W Golf training Standard care ⑩
[30] Con: 7.0 Con: 53.1
Gonzalo Ex: 16; Ex: 61.2; Resistance Conventional PA at
46.92 M 1 2W ＜4 min 2t/W ⑥⑦⑤
[31] Con: 16 Con: 65.7 training home
Ex: 22; Ex: 65.9; High-intensity Stretching, balance,
Tang [32] >12 M 2 4W 60 min 3t/W ⑥⑦⑤
Con: 25 Con: 66.9 exercise strength training
Ex: 14; Ex: 50.7; Stretching, balance,
Kim [33] 12.24 M 6.0 W 90 min 5t/W Combined ④
Con: 15 Con: 51.9 strength training
Meester Ex: 26; Ex: 60.9; Conventional PA at
43.64 M 1 0W 30 min 2t/W Aerobic ⑪
[34] Con: 24 Con: 62.3 home
Wang a Ex: 42; Ex: 65.1; Conventional PA at
＜6 M 1 2W 50 min 3t/W Aerobic ⑤⑦⑥
[35] Con: 47 Con: 64.4 home
Wang B Ex: 44; Ex: 66.7; Conventional PA at
＜6 M 1 2W 110 min 3t/W Combined ⑤⑦⑥
[36] Con: 47 Con: 64.4 home
Ex: 19; Ex: 59.0; Conventional PA at
Nagy [37] 11.4 M 20D 60 min 5t/W Combined ③⑥
Con: 16 Con: 62.0 home
Hansen Ex; 108; Ex: 71.4; Conventional PA at
3M 1 8M 75–90 min 2-3 t/W Aerobic ①
[38] Con: 101 Con: 72.0 home
Khattab Ex: 25; Ex: 65.9;
36 M 6.0 M 60 min 3t/W Aerobic Balance training ⑤⑦⑥
[39] Con: 25 Con: 66.9
Ex: 15; Ex: 50.6; 12–18
Yeh [40] 71.15 M 60 min 2-3 t/W Combined Aerobic ④②
Con: 15 Con: 60.2 W
Ex: 86; Ex: 59.0; Conventional PA at
Koch [41] 5M 1 2W 80-100 min 3t/W Combined ④
Con: 45 Con: 58.0 home
Ex: 37; Ex: 61.2; Conventional PA at
Swank [42] 1M 4.0 W 240 min 5/W Physiotherapy ③
Con: 36 Con: 61.3 home
① MMSE: Mini-Mental State Examination； ② WCST： Wechsler Memory Scale； ③ FIM: Functional Independence Measure; ④ MOCA: Montreal
Cognitive Assessment； ⑤ TMT B： Trial Making Part B； ⑥ DS： Digit Span; ⑦ Stroop Test； ⑧ ACE-R： Addenbrooke’s Cognitive Examination
Revised； ⑨ SIS： Stroke Impact Scale； ⑩ Go/No-go； ⑪ Dual Task； W: week; M: month; D: day.

regards blinding of participants and outcome assessment; and 3.3. Summary of Results (Meta-Analysis)
since the participates needed to sign informed consent in
other experiments, the remaining experiments were single- 3.3.1. Heterogeneity Test. Heterogeneity tests showed that
blind trials, the risk of incomplete outcome indicators in there is no significant heterogeneity in attention and ex-
eighteen articles was unknown, and the overall quality as- ecutive function, but significant heterogeneity was found in
sessment of the included literature was high (Figure 2). working memory and total cognitive function. This indicates
Complexity 5

Random sequence generation (selection bias)

Allocation concealment (selection bias)

Blinding of participants and personnel (performance bias)

Blinding of outcome assessment (detection bias)

Incomplete outcome data (attrition bias)

Selective reporting (reporting bias)

Other bias

0 25 50 75 100
(%)

Low risk of bias

Unclear risk of bias
High risk of bias

1. Methodological quality of included studies

Schachten 2015
Studenski 2005

Ozdemir 2001

Gonzalo 2016
Tamawy 2014

Immink 2014
Wang 2018 A

Meester 2018

Khattab 2019
Quaney 2009
Wang 2018 B

Hansen 2019
Nilsson 2001

Moore 2014
Swank 2020

Mead 2007

Chen 2006
Koch 2020
Nagy 2019
Tang 2016

Fang 2003
Kim 2017
Yeh 2019

Liu 2015

Random sequence generation (selection bias)

+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+

Allocation concealment (selection bias)

+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+

Blinding of participants and personnel (performance bias)

+

+
–
–
–
–

–
–
–
–
–
–
–
–
–

–
–
–
–
–

Blinding of outcome assessment (detection bias)

+
+
+
+
+

+
+
+
+
+
+
+
+
+
+

+
+
+
+
+
+
–

Incomplete outcome data (attrition bias)

+

+
+
+

+
+
–
–
–

–
–
–

+
–
–

–
–

–
–
–

–
?

Selective reporting (reporting bias)

+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+

Other bias
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+

2. The distribution of the methodological quality of included studies

Figure 2: Quality assessment of included studies in the meta-analysis.

that there are potential unknown factors moderating the Table 2: Heterogeneity analysis.
effect of exercise on the recovery of cognitive function after
Cognitive domains Q P I2 (%) Z P
stroke, so the potential moderating factors will be explored
through subgroup analysis (Table 2). Attention 4.09 0.54 0.00 3.18 ≤0.001
Executive function 6.53 0.59 0.00 2.87 ≤0.001
Working memory 57.39 ≤0.001 77.00 4.46 ≤0.001
3.3.2. Effects of PA Training on Cognition. Of all included Cognitive function 36.01 ≤0.001 64.00 2.59 ≤0.001
studies, fourteen trials examined overall cognitive function,
fourteen examined working memory, six examined atten-
tion, and nine examined executive function. The meta- analysis shows that cognitive benefit of exercise training was
analysis on cognitive benefit of exercise showed that there most significant at age of 49 to 55 years in stroke survivors
was a significant and positive effect of physical activity on (SMD � 0.59; CI: 0.05–0.45; P < 0.05), followed by 60 to 65 years
cognitive function after stroke (SMD � 0.16; CI: 0.04–0.28; (SMD � 0.25; CI: 0.25–0.93; P < 0.05), whereas patients aged 56
P < 0.05) with moderate heterogeneity (Q [df] � 36.01 [13]; to 60 years and above showed poor cognitive gain (Figure 4).
P < 0.001; I2 � 64%) within the group (Figure 3), indicating Mixed-effect analysis of stroke course showed that there
that some variables may moderate the effect of exercise; then was no significant difference in cognitive gain after exercise
mixed-effect models were used for moderating analyses for in patients with different course of disease (QB � 7.09;
they assume that variability between studies may be at- P < 0.05), which means that earlier exercise intervention
tributable to fixed and random components, as well as after stroke was not associated with greater cognitive gain.
subject-level sampling error. The cognitive benefit of exercise intervention was maxi-
mum at 18 months since stroke onset (SMD � 0.59; CI:
0.21–0.98; P < 0.05), while the cognitive gain for survivors
3.3.3. Subgroup Analyses. Mixed-effect analysis showed that that initiated exercise intervention within 18 months after
cognitive benefit of exercise varied significantly (QB � 10.42; stroke was not significantly changed compared with con-
P < 0.05) among different age groups. Moderating-effect trols (Figure 4).
6 Complexity

Experimental Control Weight Std. mean difference Std. mean difference

Study Mean SD Total Mean SD Total (%) IV, fixed, 95% Cl IV, fixed, 95% Cl
Cognitive function
Yeh 2019 24. 93 1. 18 15 20. 79 1. 60 15 1. 30 2. 87 [1. 81 , 3. 92]
Tamawy 2014 81. 07 6. 16 16 75. 93 4. 90 15 2. 50 0. 90 [0. 14 , 1. 65]
Kim 2017 23. 42 6. 85 14 22. 30 4. 97 15 2. 60 0. 18 [–0. 55 , 0.91]
Nagy 2019 35. 00 0. 37 19 35. 00 0. 19 16 3. 20 0. 00 [–0. 67 , 0. 67]
Moore 2014 92. 00 5. 00 20 91. 00 8. 00 20 3. 70 0. 15 [–0. 47 , 0. 77]
Chen 2006 26. 60 4. 60 25 22. 90 6. 70 20 3. 90 0. 65 [0. 04 , 1. 25]
Meester 2018 12. 61 0. 82 26 12. 37 0. 86 24 4. 50 0. 28 [–0. 28 , 0.84]
Nilsson 2001 32. 00 5. 00 28 31. 50 6. 60 32 5. 50 0. 08 [–0. 42 , 0. 59]
Ozdemir 2001 18. 03 8. 45 30 17. 50 6. 07 30 5. 50 0. 07 [–0. 44 , 0. 58]
Swank 2020 26. 40 3. 80 37 25. 20 4. 00 36 6. 60 0. 30 [–0. 16 , 0. 77]
Studenski 2005 33. 30 1. 40 44 33. 20 1. 90 49 8. 50 0. 06 [–0. 35 , 0.47]
Koch 2020 20. 70 5. 60 86 21. 10 6. 00 45 10. 80 –0. 07 [–0. 43 , 0. 29]
Fang 2003 23. 41 7. 70 78 21. 87 9. 80 78 14. 20 0. 17 [–0. 14 , 0. 49]
Hansen 2019 27. 50 3. 60 143 27. 50 3. 70 156 27. 30 0. 00 [–0. 23 , 0. 23]
Subtotal (95%) 580 551 100.0 0. 16 [0. 04 , 0. 28]
Heterogeneity: chi2 = 36.01, df = 13 (P = 0.0006); I2 = 64%
Test for overall effect: Z = 2.59 (P = 0.010) –2 –1 0 1 2
Favours (experimental) Favours (control)

Figure 3: Meta-analysis on the cognitive benefit of exercise.

Model N n SMD 95% CI Z P-value Heterogeneity (Q) I2 Q(P-value) Q QB Std.mean difference

B
(%) (P-value) IV, fixed, 95%CI
Overall 14 1131 0.16⁎ [0.04 , 0.28] 2.65 0.00 38.17 65.90 0.00
Age 14 10.42 0.02
45-55 Years Old 4 149 0.59⁎ [0.25 , 0.93] 2.65 0.01 24.97 88.0 0.00
56-60 Years Old 2 166 –0.05 [–0.37 , 0.26] 0.33 0.74 0.03 0.00 0.86
60-65 Years Old 5 384 0.25⁎ [0.05 , 0.45] 2.45 0.01 2.52 0.00 0.64
>65 Years Old 3 432 0.03 [–0.16 , 0.22] 0.28 0.78 0.23 0.00 0.89
Course of Disease 14 7.09 0.07
<3 Months 5 442 0.15 [–0.04 , 0.33] 1.53 0.13 0.82 0.00 0.93
3-6 Months 3 475 0.04 [–0.14 , 0.23] 0.45 0.65 4.49 55.50 0.11
10-18 Months 3 94 0.33 [–0.08 , 0.75] 1.59 0.11 3.46 42.20 0.18
>18 Months 3 120 0.59⁎ [0.21 , 0.98] 3.00 0.00 22.32 91.00 0.00
Training Length 14 2.62 0.45
<4 Weeks 3 264 0.19 [–0.05 , 0.43] 1.52 0.13 0.57 0.00 0.75
6-10 Weeks 5 229 0.24 [–0.02 , 0.50] 1.79 0.07 3.97 0.00 0.41
12-14 Weeks 4 299 0.25⁎ [0.01 , 0.49] 2.02 0.04 30.81 90.30 0.00
>14 Weeks 2 339 0.02 [–0.20 , 0.23] 0.16 0.87 0.20 0.00 0.66
Session Duration 14 4.38 0.22
<30 Minutes 3 203 0.12 [–0.15 , 0.40] 0.87 0.39 0.44 0.00 0.80
45-60 Minutes 5 306 0.36⁎ [–0.12 , 0.59] 3.01 0.00 27.02 85.20 0.00
75-90 Minutes 4 489 0.05 [0.13 , 0.23] 3.51 0.61 5.88 49.00 0.12
>90 Minutes 2 133 0.20 [–0.14 , 0.54] 1.15 0.25 0.45 0.00 0.50
Exercise Frequency 14 0.00 0.96
2-3 Times Per Week 8 718 0.16⁎ [0.01 , 0.31] 2.12 0.03 37.34 81.30 0.00
5 Times Per Week 6 413 0.16 [–0.04 , 0.35] 1.58 0.11 0.83 0.00 0.98
Exercise Intensity 14 1.76 0.18
Moderate Intensity 11 742 0.22⁎ [0.07 , 0.37] 2.92 0.00 35.44 71.80 0.00
High Intensity 3 389 0.05 [–0.15 , 0.25] 0.51 0.61 0.97 0.00 0.62
Exercise Mode 14 2.33 0.31
Aerobic Exercise 3 409 0.05 [–0.15 , 0.24] 0.47 0.64 0.89 0.00 0.64
Physiotherapy 6 467 0.20⁎ [0.02 , 0.38] 2.16 0.03 3.17 0.00 0.67
Combined Therapy 5 255 0.28⁎ [0.02 , 0.55] 2.11 0.04 31.78 87.40 0.00
Cognitive Domain 29 38.00 0.00
Working Memory 14 642 0.36⁎ [0.20 , 0.53] 4.46 0.00 57.39 77.00 0.00
Attention 6 331 –0.35⁎ [–0.57 , –0.14] 3.18 0.00 4.09 0.00 0.54
Executive Function 9 586 –0.24⁎ [–0.40 , –0.08] 2.87 0.00 6.53 0.00 0.59

–2 –1 0 1 2
Favours (experimental) Favours (control)

Figure 4: Effect sizes for each subgroup within the included studies.
Complexity 7

To assess the influence of trial length, studies were 0.20–0.53; P < 0.05) and executive function (SMD � 0.36; CI:
stratified based on trial length, and the mixed-effect analysis 0.20–0.53; P < 0.05) (Figure 5).
showed that longer intervention length did not yield better
cognitive gain (QB � 2.62; P > 0.05). Moderating-effect
analysis revealed that 12 to 14 weeks’ exercise intervention 3.4. Publication Bias. We created funnel plots by plotting the
was associated with the greater magnitude of cognitive gains cognitive benefit of exercise training against the reciprocal of
(SMD � 0.25; CI: 0.01–0.49; P < 0.05), but longer or shorter the standard error of the exercise efficacy. The funnel plot of
intervention showed less effectiveness on improving cog- overall cognitive function, attention, and executive function
nitive function (P < 0.05). was symmetrical around the mean effect size line. For
As for the influence of exercise frequency, mixed-effect working memory, the funnel plot suggests the occurrence of
analysis of intervention frequency showed that higher-fre- publication bias. The asymmetry of graphic plot might be
quency intervention did not associate with the better cog- owing to the fact that smaller sample studies showing less
nitive gains (QB � 0.00; P > 0.05). While moderating-effect effectiveness were reported in the literature, and the
analysis demonstrated that 2 or 3 times per week may omission of the unpublished trials may also contribute to the
maximize cognitive gains (SMD � 0.16; CI: 0.01–0.31; biased estimation (Figure 6).
P < 0.05), excessive exercise did not have a significant effect
on cognitive gains (SMD � 0.16; CI: −0.04–0.35; P < 0.05) 4. Discussion
(Figure 4).
Of the studies that examined the effects of session du- Impairment of cognitive and motor function after stroke
ration, no significant difference was found between different seriously affects the quality of life, while physical activity is
durations (QB � 4.38; P > 0.05). While moderating-effect highly praised for its high practicability in cognitive and
analysis showed that the cognitive gains were not increased motor function recovery after stroke. These meta-analytical
with longer duration, 40–60 minutes’ training for each findings indicated a small but significant positive effect of
session had favourable effects on cognitive function exercise on cognition in survivors after stroke relative to
(SMD � 0.36; CI: 0.12–0.59; P < 0.05). However, once the controls, and moderating-effect analysis showed that cog-
session duration beyond this range could not produce sig- nitive benefits are affected by different moderators by which
nificant cognitive benefit (Figure 4). exercise-related benefits for cognition can be optimized.
We also examined whether the magnitude of cognitive Mixed-effects analysis showed that the cognitive benefits of
gains differed depending on exercise intensity. Mixed- exercise appear to be selective rather than general, as effect
effect analysis demonstrated that high-intensity exercise sizes differed significantly between the age groups. More-
was not associated with better cognitive gains (QB � 1.76; over, our subgroup analysis further indicated that the lower
P > 0.05). On the contrary, moderate-intensity exercise the age of stroke survivors (49–55 years), the better the
training was most likely to maximize cognitive gains cognitive recovery after exercise training. However, cogni-
(SMD � 0.22; CI: 0.07–0.37), while the pooled effect size tive benefits of exercise decrease with age (60–65 years),
for trials using high-intensity exercise was not ideal in especially when the stroke survivors are above 65 years of
cognitive improvement (SMD � 0.05; CI: −0.15–0.25; age; it is difficult to obtain any effective cognitive im-
P > 0.05) (Figure 4). provement from exercise training. The reason may be that
The type of exercise training was also assessed to ex- the volume of hippocampus and the thickness of white
amine the influence of different exercise training on cog- matter and gray matter in prefrontal cortex which related to
nitive function involving (1) aerobic exercise, (2) cognitive function decrease with age. At the same time, due
physiotherapy, or (3) combination of aerobic exercise/ to the narrowing of blood vessels, the blood flow through the
physiotherapy/cognitive training. However, no significant brain decreases, which hinders the formation of new
difference was found between three physical activities in capillaries, thus leading to the loss of existing nerve cells and
cognitive improvement (QB � 2.33; P > 0.05). Moderating- synapses and further degeneration of myelin sheath,
effect analysis shows that the cognitive improvement from blocking the functional connection between nerve synapses,
aerobic exercise alone was not significant (SMD � 0.05; CI: and ultimately affecting cognitive function [39]. However,
−0.15–0.24; P > 0.05). On the contrary, physiotherapy with exercise training exerts greater effects on many of the
strength/endurance/flexibility/balance may generate rela- aforementioned processes. For instance, a certain degree of
tively ideal cognitive gains (SMD � 0.20; CI: 0.02–0.38; exercise training can promote the normalization of growth
P < 0.05), while combined therapy generates the largest factor secretion (BDNF, IGF-1, NGF, and VEGF) [40], re-
cognitive benefits (SMD � 0.28; CI: 0.02–0.55; P < 0.05) versing the atrophy of hippocampal volume and prefrontal
(Figure 4). cortex effectively and then delaying the decline of cognitive
We also investigated the specific benefit and difference function [41]. However, due to the motor dysfunction after
among three cognitive domains induced by exercise train- stroke and the inability of elderly survivors to adapt to high-
ing, and significant differences were found in cognitive gains intensity physical activities, it is difficult for exercise training
between three domains (QB � 38.0; P < 0.05). Of the three to play its due role.
domains, exercise training generated the greatest and most Regarding the influence of time from stroke onset to
significant effect on working memory (SMD � 0.36; CI: initiation of the intervention, our analysis demonstrated that
0.20–0.53; P < 0.05), followed by attention (SMD � 0.36; CI: neither exercise training soon after stroke within three
8 Complexity

0.4 0.36(0.08) QB = 38.0; P = 0.00

apoptotic cell death [44]. Nonetheless, the existing animal
Cl(confidence interval) model studies support the priority of early exercise inter-
N(number of studies) vention (1–7 days) in reducing lesion volume in damaged
Effect size (SMD(95% CI))

0.2 brain regions, preventing further damage of surrounding

tissues from inflammation and oxidation [40, 45], and then
promoting nerve regeneration. However, there is still in-
0.0
sufficient evidence to support that the best sensitive period in
animal models applies equally to human rehabilitation [46].
–0.2 The effect size of exercise varied in different exercise
–0.24(0.08)
training lengths. Moderating-effect analysis revealed that
exercise training length neither less than 4 weeks nor from 6
–0.35(0.11)
–0.4 to 10 weeks can generate insufficient effect accumulation.
Working Attention(n = 6) Executive However, cognitive function improved significantly only in
memory(n = 14) function(n = 9) trial with exercise period of 12 to 14 weeks, while once the
Figure 5: Mean effect of PA on cognitive domain. Significant training cycle extended beyond 14 weeks, the cognitive
difference (P < 0.05) was observed between three domains. benefit disappeared again, which is consistent with the prior
studies [47]. At present, the cycle of exercise intervention
after stroke is mainly controlled within 12 weeks. So far, no
0 clear optimal cycle has been proposed in existing study. But
known studies have shown that [48] aerobic training after
0.2
stroke lasts at least eight weeks to have a positive effect on
cardiopulmonary and cognitive function, and a recent study
further demonstrated [49] that 12-week aerobic exercise not
0.4
SE (SMD)

only has good patient tolerance but also produces appro-

priate long-term potentiation on physical and mental health.
0.6
Our analysis also found that session duration for each
exercise intervention should be controlled within an ap-
0.8 propriate range to produce the ideal cognitive gain, and extra
time cannot generate additional cognitive improvement.
1 Among them, cognitive benefits of exercise training may not
–2 –1 0 1 2 be significantly improved if the exercise training time was
SMD
less than 30 minutes for each session duration. Similarly, if
Subgroups the session duration was beyond one hour for each time, the
Cognitive function Attention cognitive function cannot be significantly improved as well.
Working memory Executive function In contrast, only when the intervention time is controlled
Figure 6: Funnel plot of cognitive benefit of exercise. within 45 to 60 minutes can the cognitive benefit reach a
significant level. The reason why the cognitive benefits of this
exercise duration reach the maximum level in 45 to 60
months nor that from month three to month eighteen after minutes may be because the exercise intervention of this
stroke exerts fewer positive effects on cognition. On the duration can obviously improve the glucose tolerance [27]
contrary, exercise training after 18th month of stroke onset and insulin sensitivity [50] of patients, while restoring motor
can effectively promote the recovery of impaired cognitive and balance ability [51], reversing the decline of cardio-
function, and this conclusion is consistent with the existing pulmonary function caused by impaired motor function,
studies [17]. The unsatisfactory effect among trials intro- and gradually improving self-perception [52] of the body
ducing exercise training within 3 months after stroke may, in and cognitive function [53].
part, be a consequence of conflicts with the short-lived As for exercise frequency, our subgroup analysis indi-
plasticity window after stroke, for most spontaneous re- cated that there was no significant difference in cognitive
covery phase of endogenous repair occurring in the first benefit between different exercise training frequencies for
three months after stroke, which also is the best sensitive each week, and higher-frequency training did not signifi-
period of brain plasticity after stroke [42]. At this moment, cantly improve cognitive function. Instead, maintaining
the primary task is to recuperate the body to restore the body exercise training 2 to 3 times per week is more conducive to
function and immunity. Meanwhile, psychological guidance the improvement of cognitive function. This is in line with
is also needed for stroke survivors to shape an optimistic and the latest international stroke exercise care guidelines [53].
positive attitude in the long-term rehabilitation process. The cognitive benefit of aerobic exercise is a cumulative
Introducing exercise training in this time may hinder the process, and it is mainly determined by the interaction of
potential plasticity change of the damaged brain area and frequency, intensity, and duration. At present, the exercise
aggravate the trauma [43]. Other studies show that exercise training frequency in the known studies mainly ranges from
training within 24 hours after stroke increases the con- 2 to 5 times per week but more often 3 times per week
centration of hyperglycolytic related markers and the risk of according to the baseline fitness levels at the initial stage of
Complexity 9

the intervention [54], the status of cardiopulmonary func- function get equal improvement, among which exercise has
tion and nervous system injury, the existence of comor- the greatest effect on improving working memory, followed
bidity, and so forth. This may be because exercise by attention and finally executive function. These results are
intervention should ensure the maximum cognitive benefit more specific than those of the previous meta-analysis [17].
and higher compliance of the subjects after each session. First of all, due to the insufficient power (n � 5) in detecting
However, studies have shown that [55] only when the the effectiveness of exercise training on working memory,
training frequency is kept at 2 to 3 times per week can the the results are suspected of overestimation. A total of 14
survivors show the highest compliance of 23%, and once the relative studies on working memory were included in our
exercise frequency is increased to 4 times per week, the analysis to avoid errors caused by small samples. Secondly,
compliance is slightly reduced to 19%. However, when the the included studies on executive function are also increased
frequency reaches 5 times per week, the compliance of compared with the previous meta-analysis, and the corre-
subjects is only 9%. Therefore, after considering the cog- sponding cognitive benefit has also reached a significant
nitive gain of exercise and the participants’ compliance level among the experimental groups, relative to controls.
comprehensively, it is considered that 2 to 3 times per week However, analysis on attention performance has not
of exercise can maximize the cognitive gain. changed compared with the previous meta-analysis, which
Moreover, we found that cognitive benefit was not re- to some extent makes up for the shortcomings of previous
lated to the higher exercise intensity. Moderating-effect studies.
analysis further shows that high-intensity exercise does not
generate higher cognitive gain. Instead, exercise training at 5. Summary
moderate intensity is more conducive to the improvement of
cognitive function. This may be because the rehabilitation In summary, results from our quantitative synthesis and
programs after stroke are based on health status and physical meta-analysis support a small but significant improvement
response to exercise training, by optimizing the dose- of cognitive function after stroke after aerobic exercise
response relationship among exercise frequency, duration, training. Our findings indicated that cognitive benefits can
and intensity with cognitive function, to avoid muscle be maximized in specific exercise regimen; all types of ex-
soreness and fatigue during training and alleviate further ercise training seem to be effective, but combined exercise
cognitive damage [56]. Other studies also show that [57, 58] therapy promises more pronounced cognitive benefits
low-intensity physical activity is insufficient to generate compared with aerobic exercise alone and physiotherapy.
enough cognitive benefit similar to moderate-intensity The positive efficacy of exercise on cognitive function cannot
aerobic exercise. Therefore, it is necessary to ensure suffi- be maximized by roughly increasing the total dose. Instead,
cient exercise dose/intensity to produce cognitive im- under the premise of moderate-intensity exercise, we rec-
provement effect while avoiding adverse physical reactions ommend the 2 to 3 times a week, 45 to 60 min for each
caused by excessive stress. Therefore, higher intensity within session duration and lasting for 12 to 14 weeks to maximize
a reasonable range can generate greater improvement in their cognitive benefits. Additionally, exercise training is less
cardiopulmonary and cognitive functions. However, before effective in the earlier stage after stroke; 18 months after
the best cognitive gain is achieved, safety and feasibility must stroke is optimal for the initiation of exercise interventions;
be considered. Therefore, at the beginning of rehabilitation exercise training generated the most favourable effects for
exercise program, participants should start with moderate survivors aged 49 to 55 years. Beyond recommendations for
intensity with lower risk of abnormal reaction [59]. optimizing cognitive performance from different exercise
We also investigated differences in efficacy as a function regimen, our findings have further implications for future
of type of exercise regimen, and the result indicated that study, including identifying the effects of cognitive level at
exercise training involving only aerobic exercise did not the initiation of the training and its interaction with exercise
yield significant cognitive benefits. While physiotherapy regimen, avoiding the interference of baseline cognition
consisting of stretching/balance/strength training achieved level to the overall effect of the exercise training. Moreover,
significant effects in cognitive improvement, combined the optimal training parameters also need to be considered
therapy (aerobic exercise/physiotherapy/cognitive training) carefully to realize the precision of exercise intervention
generated the largest cognitive benefit. Although the neg- effect.
ative results of aerobic exercise only here may be due to However, some limitations still remain in this study. For
insufficient statistical test, there are still empirical studies example, because of the stricter inclusion criteria, a limited
pointing out the limitations of aerobic exercise alone [60]. number of studies were included; therefore significant
Moreover, there is ample evidence that multiway combi- heterogeneity remained after subgroup analysis. Further-
nation therapy can not only avoid monotonicity and im- more, as all data analyzed in this study were abstracted from
prove participants’ compliance [61] but also improve published literature results instead of raw data, the au-
efficiency of cognitive recovery and shorten exercise training thenticity of data cannot be guaranteed, and the publishing
length [62]. bias is therefore unavoidable. In addition, the included
Our final analysis of exercise efficacy on cognitive do- studies are all from English-based journals, which may ig-
mains showed that there were significant differences in the nore the potential differences between countries. Based on
improvement of three cognitive domains by exercise the above risks, larger trials are needed to evaluate the
training, indicating that not all domains of cognitive cognitive benefit of exercise, in order to better understand
10 Complexity

the moderator effect of session duration, exercise intensity, function after stroke: a systematic review,” International
and training length. Finally, future research should be Psychogeriatrics, vol. 24, no. 4, pp. 557–567, 2012.
carried out to evaluate the effect of long-term adherence to [13] D. H. Saunders, C. A. Greig, and G. E. Mead, “Physical activity
exercise after stroke. and exercise after stroke,” Stroke, vol. 45, no. 12,
pp. 3742–3747, 2014.
[14] S. Koch, “Randomized trial of combined aerobic, resistance,
Data Availability and cognitive training to improve recovery from stroke:
feasibility and safety,” Journal of the American Heart Asso-
The data used to support the findings of this study are
ciation, vol. 9, no. 10, Article ID e015377, 2020.
available from the corresponding author upon request. [15] A. J. Alfini, L. R. Weiss, K. A. Nielson, M. D. Verber, and
J. C. Smith, “Resting cerebral blood flow after exercise training
Conflicts of Interest in mild cognitive impairment,” Journal of Alzheimer’s Disease,
vol. 67, no. 2, pp. 671–684, 2019.
The authors declare that they have no conflicts of interest [16] L. Luo, C. Li, X. Du et al., “Effect of aerobic exercise on BDNF/
reported in this paper. proBDNF expression in the ischemic hippocampus and de-
pression recovery of rats after stroke,” Behavioural Brain
References Research, vol. 362, no. 19, pp. 323–331, 2019.
[17] L. E. Oberlin, A. M. Waiwood, T. B. Cumming,
[1] M. Patel, C. Coshall, A. G. Rudd, and C. D. Wolfe, “Natural A. L. Marsland, J. Bernhardt, and K. I. Erickson, “Effects of
history of cognitive impairment after stroke and factors as- physical activity on poststroke cognitive function,” Stroke,
sociated with its recovery,” Clinical Rehabilitation, vol. 17, vol. 48, no. 11, pp. 3093–3100, 2017.
no. 2, pp. 158–166, 2003. [18] D. Nagy, “The effect of low-intensity aerobic training on
[2] T.-T. Yeh, K.-C. Chang, and C.-Y. Wu, “The active ingredient cognitive functions of severely deconditioned subacute and
of cognitive restoration: a multicenter randomized controlled chronic stroke patients: a randomized, controlled pilot study,”
trial of sequential combination of aerobic exercise and International Journal of Rehabilitation Research, vol. 42, no. 3,
computer-based cognitive training in stroke survivors with pp. 275–279, 2019.
cognitive decline,” Archives of Physical Medicine and Reha- [19] L. Nilsson, J. Carlsson, A. Danielsson et al., “Walking training
bilitation, vol. 100, no. 5, pp. 821–827, 2019.
of patients with hemiparesis at an early stage after stroke: a
[3] L. Deng, S. Qiu, Y. Yang et al., “Efficacy and tolerability of
comparison of walking training on a treadmill with body
pharmacotherapy for post-stroke depression: a network meta-
weight support and walking training on the ground,” Clinical
analysis,” Oncotarget, vol. 9, no. 34, pp. 23718–23728, 2018.
[4] L. Viale, N. P. Catoira, G. Di Girolamo, and C. D. González, Rehabilitation, vol. 15, no. 5, pp. 515–527, 2001.
“Pharmacotherapy and motor recovery after stroke,” Expert [20] F. Özdemir, M. Birtane, R. Tabatabaei, S. Kokino, and
Review of Neurotherapeutics, vol. 18, no. 1, pp. 65–82, 2018. G. Ekuklu, “Comparing stroke rehabilitation outcomes be-
[5] K.-P. Lesch and J. Waider, “Serotonin in the modulation of tween acute inpatient and nonintense home settings,” Ar-
neural plasticity and networks: implications for neuro- chives of Physical Medicine and Rehabilitation, vol. 82, no. 10,
developmental disorders,” Neuron, vol. 76, no. 1, pp. 175–191, pp. 1375–1379, 2001.
2012. [21] Y. Fang, X. Chen, H. Li, J. Lin, R. Huang, and i. Zeng, “A study
[6] D. Sulzer, “How addictive drugs disrupt presynaptic dopa- on additional early physiotherapy after stroke and factors
mine neurotransmission,” Neuron, vol. 69, no. 4, pp. 628–649, affecting functional recovery,” Clinical Rehabilitation, vol. 17,
2011. no. 6, pp. 608–617, 2003.
[7] L Martinsson, H Hårdemark, and S Eksborg, “Amphetamines [22] S. Studenski, P. W. Duncan, S. Perera, D. Reker, S. M. Lai, and
for improving recovery after stroke,” The Cochrane Database L. Richards, “Daily functioning and quality of life in a ran-
of Systematic Reviews, no. 1, p. CD002090, 2007. domized controlled trial of therapeutic exercise for subacute
[8] P. N. Patsalos, D. J. Berry, B. F. D. Bourgeois et al., “Anti- stroke survivors,” Stroke, vol. 36, no. 8, pp. 1764–1770, 2005.
epileptic drugsbest practice guidelines for therapeutic drug [23] X. Chen, “Effect of community-based-rehabilitation on activities
monitoring: a position paper by the subcommission on of daily life and cognitive function in stroke patients,” Chinese
therapeutic drug monitoring, ILAE Commission on Thera- Journal of Clinical Rehabilitation, vol. 10, no. 32, pp. 4–6, 2006.
peutic Strategies,” Epilepsia, vol. 49, no. 7, pp. 1239–1276, [24] G. E. Mead, C. A. Greig, I. Cunningham et al., “Stroke: a
2008.
randomized trial of exercise or relaxation,” Journal of the
[9] K. D. Langdon and D. Corbett, “Improved working memory
American Geriatrics Society, vol. 55, no. 6, pp. 892–899, 2007.
following novel combinations of physical and cognitive ac-
[25] B. M. Quaney, L. A. Boyd, J. M. McDowd et al., “Aerobic
tivity,” Neurorehabilitation and Neural Repair, vol. 26, no. 5,
exercise improves cognition and motor function poststroke,”
pp. 523–532, 2012.
Neurorehabilitation and Neural Repair, vol. 23, no. 9,
[10] L. L. F. Law, F. Barnett, M. K. Yau, and M. A. Gray, “Effects of
combined cognitive and exercise interventions on cognition pp. 879–885, 2009.
in older adults with and without cognitive impairment: a [26] M. A. Immink, S. Hillier, and J. Petkov, “Randomized con-
systematic review,” Ageing Research Reviews, vol. 15, no. 5, trolled trial of yoga for chronic poststroke hemiparesis: motor
pp. 61–75, 2014. function, mental health, and quality of life outcomes,” Topics
[11] National Academies of Sciences, Preventing Cognitive Decline in Stroke Rehabilitation, vol. 21, no. 3, pp. 256–271, 2014.
and Dementia: A Way Forward, National Academies Press [27] S. A. Moore, K. Hallsworth, D. G. Jakovljevic et al., “Effects of
(US), Washington, DC, USA, 2017. community exercise therapy on metabolic, brain, physical,
[12] T. B. Cumming, K. Tyedin, L. Churilov, M. E. Morris, and and cognitive function following stroke,” Neurorehabilitation
J. Bernhardt, “The effect of physical activity on cognitive and Neural Repair, vol. 29, no. 7, pp. 623–635, 2014.
Complexity 11

[28] M. S. Tamawy, “Aerobic exercises enhance cognitive functions training early after stroke in a mouse model,” Stroke, vol. 46,
and brain derived neurotrophic factor in ischemic stroke pa- no. 10, pp. 2951–2960, 2015.
tients,” NeuroRehabilitation, vol. 34, no. 1, pp. 209–213, 2014. [43] F. Li, J. T. Pendy, J. N. Ding et al., “Exercise rehabilitation
[29] A. Liu, “Exercise training and recreational activities to pro- immediately following ischemic stroke exacerbates inflam-
mote executive functions in chronic stroke: a proof-of- matory injury,” Neurological Research, vol. 39, no. 6,
concept study,” Journal of Stroke and Cerebrovascular pp. 530–537, 2017.
Diseases, vol. 24, no. 1, pp. 130–137, 2015. [44] J. Shen, M. Huber, E. Y. Zhao et al., “Early rehabilitation
[30] T. Schachten and P. Jansen, “The effects of golf training in aggravates brain damage after stroke via enhanced activation
patients with stroke: a pilot study,” International Psycho- of nicotinamide adenine dinucleotide phosphate oxidase
geriatrics, vol. 27, no. 5, pp. 865–873, 2015. (NOX),” Brain Research, vol. 1648, no. A, pp. 266–276,
[31] M. Gonzalo, “Functional and cognitive adaptations after 2016.
flywheel resistance training in stroke patients: a pilot ran- [45] M. W. Austin, M. Ploughman, L. Glynn, and D. Corbett,
domized controlled trial,” Journal of NeuroEngineering and “Aerobic exercise effects on neuroprotection and brain repair
Rehabilitation, vol. 13, no. 1, p. 37, 2016. following stroke: a systematic review and perspective,”
[32] A. Tang, J. Eng, A. Krassioukov, T. Tsang, and T. Liu- Neuroscience Research, vol. 87, no. 1, pp. 8–15, 2014.
Ambrose, “High- and low-intensity exercise do not improve [46] S. R. Zeiler, “Should we care about early post-stroke reha-
cognitive function after stroke: a randomized controlled trial,” bilitation? Not yet, but soon,” Current Neurology and Neu-
Journal of Rehabilitation Medicine, vol. 48, no. 10, pp. 841– roscience Reports, vol. 19, no. 3, p. 13, 2019.
846, 2016. [47] F. Cechetti, P. V. Worm, V. R. Elsner et al., “Forced treadmill
[33] J. Kim, “Effects of an exercise protocol for improving exercise prevents oxidative stress and memory deficits fol-
handgrip strength and walking speed on cognitive function in lowing chronic cerebral hypoperfusion in the rat,” Neurobi-
patients with chronic stroke,” Medical Science Monitor, ology of Learning and Memory, vol. 97, no. 1, pp. 90–96, 2012.
vol. 23, pp. 5402–5409, 2017, review groups � (Dementia and [48] P. Boyne, K. Dunning, D. Carl et al., “High-intensity interval
Cognitive Improvement; Stroke). training and moderate-intensity continuous training in am-
[34] B. Wang, “Effects of combined intervention of physical ex- bulatory chronic stroke: feasibility study,” Physical Therapy,
ercise and cognitive training on cognitive function in stroke vol. 96, no. 10, pp. 1533–1544, 2016.
survivors with vascular cognitive impairment: a randomized [49] S. A. F. D. Brito, L. T. Aguiar, L. N. Garcia, P. D. C. Peniche,
controlled trial,” Clinical Rehabilitation, vol. 33, no. 1, M. T. F. D. Reis, and C. D. C. D. M. Faria, “Cardiopulmonary
pp. 54–63, 2018. exercise testing and aerobic treadmill training after stroke:
[35] H. Hansen, “A physical activity intervention to prevent feasibility of a controlled trial,” Journal of Stroke and Cere-
cognitive decline after stroke: secondary results from the Life brovascular Diseases, vol. 29, no. 7, Article ID 104854, 2020.
after STroke study, an 18-month randomized controlled trial,” [50] F. M. Ivey, A. S. Ryan, C. E. Hafer-Macko, A. P. Goldberg, and
Journal of Rehabilitation Medicine, vol. 51, no. 9, pp. 646–651, R. F. Macko, “Treadmill aerobic training improves glucose
2019. tolerance and indices of insulin sensitivity in disabled stroke
[36] D. Meester, E. Al-Yahya, A. Dennis et al., “A randomized survivors,” Stroke, vol. 38, no. 10, pp. 2752–2758, 2007.
controlled trial of a walking training with simultaneous [51] A. B. Rosenfeldt, S. M. Linder, S. Davidson et al., “Combined
cognitive demand (dual-task) in chronic stroke,” European aerobic exercise and task practice improve health-related
Journal of Neurology, vol. 26, no. 3, pp. 435–441, 2019. quality of life poststroke: a preliminary analysis,” Archives of
[37] S. Khattab, J. Eng, T. Liu-Ambrose, J. Richardson, Physical Medicine and Rehabilitation, vol. 100, no. 5,
J. MacDermid, and A. Tang, “Sex differences in the effects of pp. 923–930, 2019.
exercise on cognition post-stroke: secondary analysis of a [52] F. M. Ivey, S. J. Prior, C. E. Hafer-Macko, L. I. Katzel,
randomized controlled trial,” Journal of Rehabilitation R. F. Macko, and A. S. Ryan, “Strength training for skeletal
Medicine, vol. 52, no. 1, pp. 1–8, 2020. muscle endurance after stroke,” Journal of Stroke and Cere-
[38] C. Swank, M. Trammell, L. Callender et al., “The impact of a brovascular Diseases, vol. 26, no. 4, pp. 787–794, 2017.
patient-directed activity program on functional outcomes and [53] M. Gunnes, B. Indredavik, B. Langhammer et al., “Associa-
activity participation after stroke during inpatient rehabili- tions between adherence to the physical activity and exercise
tation-a randomized controlled trial,” Clinical Rehabilitation, program applied in the LAST study and functional recovery
vol. 34, no. 4, pp. 504–514, 2020. after stroke,” Archives of Physical Medicine and Rehabilitation,
[39] S. Snigdha and G. A. Prieto, “Chapter 16 - exercise enhances vol. 100, no. 12, pp. 2251–2259, 2019.
cognitive capacity in the aging brain,” in Physical Activity and [54] S. Billinger et al., “Does aerobic exercise and the FITT
the Aging Brain, R. R. Watson and R. R. Watson, Eds., , principle fit into stroke recovery?” Current Neurology and
pp. 161–172, Academic Press, 2017. Neuroscience Reports, vol. 15, no. 2, p. 519, 2015.
[40] M. Ploughman, M. W. Austin, L. Glynn, and D. Corbett, “The [55] K. K. Miller, R. E. Porter, E. DeBaun-Sprague,
effects of poststroke aerobic exercise on neuroplasticity: a M. Van Puymbroeck, and A. A. Schmid, “Exercise after stroke:
systematic review of animal and clinical studies,” Transla- patient Adherence and beliefs after discharge from rehabili-
tional Stroke Research, vol. 6, no. 1, pp. 13–28, 2015. tation,” Topics in Stroke Rehabilitation, vol. 24, no. 2,
[41] A. Quigley et al., “Effects of exercise on cognitive performance pp. 142–148, 2017.
in older adults: a narrative review of the evidence, possible [56] D. L. Marsden, A. Dunn, R. Callister, P. McElduff, C. R. Levi,
biological mechanisms, and recommendations for exercise and N. J. Spratt, “A home- and community-based physical
prescription,” Journal of Aging Research, vol. 2020, no. 1, activity program can improve the cardiorespiratory fitness
15 pages, Article ID 1407896, 2020. and walking capacity of stroke survivors,” Journal of Stroke
[42] K. L. Ng, E. M. Gibson, R. Hubbard et al., “Fluoxetine and Cerebrovascular Diseases, vol. 25, no. 10, pp. 2386–2398,
maintains a state of heightened responsiveness to motor 2016.
12 Complexity

[57] P. Boyne, J. Welge, B. Kissela, and K. Dunning, “Factors

influencing the efficacy of aerobic exercise for improving
fitness and walking capacity after stroke,” Archives of
Physical Medicine and Rehabilitation, vol. 98, no. 3,
pp. 581–595, 2017.
[58] J. Crozier et al., “High-intensity interval training after stroke:
an opportunity to promote functional recovery, cardiovas-
cular health, and neuroplasticity,” Neurorehabilitation and
Neural Repair, vol. 32, no. 6-7, pp. 543–556, 2018.
[59] D. Riebe, ACSM’s Guidelines for Exercise Testing and Pre-
scription, Wolters Kluwer, Alphen aan den Rijn, The Neth-
erlands, 2018.
[60] L. Lai, H. Bruce, L. Bherer, M. Lussier, and K. Z. H. Li,
“Comparing the transfer effects of simultaneously and se-
quentially combined aerobic exercise and cognitive training in
older adults,” Journal of Cognitive Enhancement, vol. 1, no. 4,
pp. 478–490, 2017.
[61] Y. K. Chang, J. D. Labban, J. I. Gapin, and J. L. Etnier, “The
effects of acute exercise on cognitive performance: a meta-
analysis,” Brain Research, vol. 1453, no. 9, pp. 87–101, 2012.
[62] X. Zhu, S. Yin, M. Lang, R. He, and J. Li, “The more the better?
A meta-analysis on effects of combined cognitive and physical
intervention on cognition in healthy older adults,” Ageing
Research Reviews, vol. 31, no. 10, pp. 67–79, 2016.