Combined cognitive and motor training improves the outcome in the early phase after stroke and prevents a decline of executive functions: A pilot study

Abstract

BACKGROUND:

The negative impact of cognitive dysfunction on motor rehabilitation as a relearning-process is well known in stroke patients. However, evidence for combined cognitive and motor training (CMT) is lacking.

OBJECTIVE:

To evaluate the effects of combined CMT in early stroke rehabilitation.

METHODS:

In a controlled pilot study, 29 moderately affected stroke patients with low-level motor performance and cognitive impairment received motor therapy plus either cognitive (experimental group, EG) or low-frequency ergometer training (control group, CG) for eight days.

RESULTS:

Both groups improved their motor functioning significantly. After training, between-group comparison revealed significant differences for cognitive flexibility and trends for set-shifting, working memory, and reaction control in favor of the EG. Within-group effects showed improvement across all cognitive domains in the EG, which correlated with gains in bed-mobility, while the CG showed no significant improvement in cognition. Rather, a trend towards reaction control decline was observed, which correlated with less functional progression and recovery. Furthermore, a decline in cognitive flexibility, set-shifting, and working memory was descriptively observed.

CONCLUSIONS:

Combined CMT may enhance cognition and motor relearning early after stroke and is superior to single motor training. Further studies are needed to replicate these results and investigate long-term benefits.

Keywords

Cerebrovascular event rehabilitation physiotherapy cognition recovery of function

1 Introduction

Even though medical standards of care (e.g., thrombolysis, thrombectomy) have improved over the last decades and stroke-related mortality was reduced (Feigin et al., 2014), stroke-related morbidity has increased as a result of higher stroke-survival rates (WHO, 2008). In the acute phase, 20–80% of patients suffer from cognitive (Sun et al., 2014; Rist et al., 2013; Nys et al., 2007) or motor deficits (Kwakkel et al., 2004; Feigin et al., 2014). Even in the long-term, many patients do not fully recover, making stroke the leading cause for acquired disability in adulthood with a massive impact on the patients’ quality of life and an increased risk to develop dementia (Farokhi-Sisakht et al., 2019; McDonald et al., 2019; Rohde et al., 2019; Rist et al., 2013). After tissue death, no current medical intervention can revert tissue loss (Mehrholz, 2008). Therefore, restorative therapeutic interventions are today’s standard of care (Barnett & Muzaffar, 2014).

Rehabilitation after stroke is described as a process of (motor) relearning (Carr & Shepherd, 2006; Nudo, 2013), aiming at restoration of skilled movement through training providing impulses for neuroplastic changes (Khan et al., 2017; Nudo, 2013; Shepherd & Carr, 2005; Kwakkel et al., 2004; Kleim, 2011). Nevertheless, motor recovery most often remains incomplete when applying today’s standards of care (Brewer et al., 2013). In the long-term, 30% of cognitive (Monsund et al., 2019; Douiri et al., 2013; Ilhe- Hansen et al., 2011; Nys et al., 2005) and 15–30% of motor deficits persist (Zorowitz et al., 2002), indicating an urgent need for novel treatment strategies.

A possible explanation for the observed limitations is that cognition and motor function are commonly assessed and addressed as two separate entities (Hostenbach et al., 1998; Chen et al., 2013). Relearning, especially in the early phase after stroke, is, however, a cognitively directed learning-process (Nudo, 2013; Hostenbach & Mulder, 1999). Unfortunately, cognitive resources are frequently limited after stroke due to disturbances in cognitive networks (Kalaria et al., 2016; Pasi et al., 2012; Tatemichi et al., 1994). Thus, considering only one function (or both but independently of each other) is likely to result in limited relearning. As the integrity and availability of cognitive resources are inevitably associated with successful motor relearning, future treatment concepts may benefit from addressing this relationship explicitly.

Combined cognitive and motor therapy approaches have been successfully applied in healthy older adults (Rahe et al., 2015a) as well as in stroke patients to improve cognition (Kalra et al., 1997; Carter et al., 1988) and motor functions (Theill et al., 2013; Kalra et al., 1997; Carter et al., 1988). The benefits of cognitive training were more significant when it targeted multiple domains (Rahe 2015a; 2015b; Petrelli et al., 2014; Carter et al., 1983; 1988) rather than one domain only (Cicerone et al., 2000; das Nair et al., 2016). Furthermore, cognitive training in the acute phase post stroke is more promising for long-term outcome (Carter et al., 1988; Nys et al., 2007) than interventions starting in the late phase (Nys et al., 2007; Cumming et al., 2013; Cicerone et al. 2000; 2005; 2011) or no intervention at all (Carter et al., 1988; Nys et al., 2007; das Nair et al., 2016; Cumming et al., 2013; Cicerone et al. 2000; 2005; 2011). However, to the best of our knowledge, no study so far systematically applied combined cognitive and motor training in daily early stroke rehabilitation.

Therefore the present study aimed at evaluating the feasibility and effects of combined cognitive and motor training on the functional outcome in early stroke rehabilitation in comparison to pure motor training. For this purpose, in a non-randomized (consecutive recruitment) controlled study design, stroke patients received 20 minutes of cognitive training (CT), serving as a primer to the learning process, followed by 45 minutes of standard motor therapy (MT) or pure MT with the same intensity and frequency (65 minutes) for eight consecutive days. We hypothesized, that (1) the study protocol is feasible in a clinical routine setting, i.e., that patients will attend therapy and not show adverse events caused by the training sessions, (2) CT plus MT improves cognition and (3) motor functions (compared to an equivalent amount of control training), and (4) that the changes in cognition are associated with improved motor functions in the experimental group (EG) greater than in the control group (CG).

2 Methods

2.1 Patients

Patients were recruited from the Early Rehabilitation Unit in the Department of Neurology of the University Hospital Cologne between May 2017 and June 2018.

Study inclusion criteria were: (1) first-ever stroke, (2) time elapsed since symptom onset less than ten days, (3) no previous or current other neurological or psychiatric disorder, (4) the ability to understand and follow therapeutic instructions, and (5) the ability to provide informed written consent before the baseline assessment. There were no limitations regarding age or gender.

The ethics committee of the Medical Faculty of Cologne University approved the study (docket number: 17-075) that followed the declaration of Helsinki (version 2013, Fortaleza). The study was retrospectively registered in the German Clinical Trials Register (DRKS 00011876).

2.2 Study design and procedure

Due to logistical constraints resulting from the clinical routine setting (e.g., time constraints of LB as one of the CT-instructors, intervening concurrent studies), randomized recruitment was not feasible. Consecutive recruitment was performed (i.e., the first patients were assigned to the EG; after completion of the EG, the patients were assigned to the CG to minimize the risk for selection bias).

For each patient, the study protocol (Fig. 1) started the day after the group assignment.

Fig. 1

Overview of the study protocol, training, and assessment. EG: experimental group, CG: control group.

On the first day, patients underwent a baseline assessment evaluating their cognitive and motor abilities. Furthermore, sociodemographic (gender, age, education) and disease-related factors (handedness: Edinburg Handedness Inventory, days since stroke onset, lesion side, stroke type, and depression: Hospital Anxiety and Depression Scale, German version) were assessed. The day after the baseline assessment, the combined training started. The EG received 20 minutes of CT plus subsequent 45 minutes of MT, while CG received 20 minutes of low-frequency ergometer training (ET) plus subsequent 45 minutes of MT. The intervention-length of 20 minutes per session was chosen based on considerations including the practicability in daily rehab routines as well as patients’ typical training endurance. Both groups received the training for eight consecutive days as the total early rehab length of stay is typically 14 days, and days for the pre- and post- assessments and “back-up” days for days in which patients were unavailable e.g. due to additional examinations or medical issues were also calculated. After the intervention phase, the post-assessment was conducted. For the DemTect for which a parallel version exists, this was used post-intervention. For all other instruments, the same version was used for baseline and post-test. The study was carried out in addition to the standard early rehabilitation program (105 minutes therapy/ day: 30 minutes MT, 45 minutes speech therapy, and 30 minutes therapeutic care plus nursing care, and medical interventions). The combined therapy (EG, CG) was scheduled every morning together with the multidisciplinary team to integrate the study protocol into the daily routine.

2.3 Trainings

2.3.1 Cognitive training

For the CT in the EG, the paper-based individual program NEUROvitalis HOME (Baller et al., 2017), which originated from the group-training NEUROvitalis (Baller et al., 2010), was used. NEUROvitalis has been successfully applied to improve cognitive functioning in healthy older adults (Rahe et al., 2015a), patients with amnestic mild cognitive impairment (Rahe et al., 2015b), and Parkinson patients (Petrelli et al., 2014). NEUROvitalis HOME was chosen, because (i) it is scientifically based, (ii) covers a broad range of cognitive domains including memory, executive functions, attention, visuocognition, and language, which are typically disturbed in stroke-patients, and (iii) it is suitable for usage in eventually bed-bound patients and exercises can be selected according to individual needs. Although feasibility and effects of the program selected have not been investigated in stroke-patients, yet, it fulfills all our requested criteria.

NEUROvitalis HOME offers 180 exercises with two difficulty levels allowing adaptation to the individual patient’s needs and progress. In the one-to-one training sessions, individual exercises from the program, according to the needs identified in the baseline-assessment, were chosen as recommended by Carter and colleagues (1988). The number of exercises depended on the individual patient’s potential and, therefore, varied between patients. Nonetheless, all patients in EG received precisely 20 minutes of this training (3–8 exercises depending on daily performance).

Although NEUROvitalis HOME was originally designed for patients to train themselves at home, for study purposes, the staff (one physiotherapist, one nurse; MG, LB) who supervised the CT were trained. Furthermore, they were encouraged to use positive reinforcement and to provide feedback.

2.3.2 Ergometer training

For ET, Reck’s MOTOMED viva 2 or MOTOMED letto were used. The advantage of these devices is that patients can train sitting on a regular chair or wheelchair (MOTOmed viva 2) as well as in an adapted laying position in bed (MOTOmed letto), allowing even non-ambulatory or bed-bound patients to participate. There were no complete breaks during the training sessions; however, if patients felt tired, they continued cycling passively. A physical therapist or rehabilitation nurse set up the bike and supervised the patients during the training. Trainers were trained to observe the patient for eventual changes in medical condition and to ensure training-continuity.

2.3.3 Motor therapy

MT lasted 45 minutes, conducted either by a physical or occupational therapist. During MT, patients and therapists worked on aspects of bed mobility (e.g., turning in bed), sitting (e.g., in bed, wheelchair, chair), sitting activities (e.g. eating, working), hand function, transfers, standing and standing activities, walking, self-care, and adaptation to aids.

2.4 Outcomes

2.4.1 Cognitive outcome parameters

Cognition was screened with the DemTect (Kalbe et al., 2004). To reduce learning effects, the parallel versions A and B (Kessler et al., 2010) were randomized for baseline and post-training assessment. Verbal short and long-term memory were examined with the DemTect, subtests immediate recall wordlist and delayed recall wordlist, respectively, and working memory was assessed with the DemTect subtest digit span reverse.

For executive functions, cognitive flexibility (DemTect: transcoding), paced set-shifting (Trail Making Test B; Reitan & Wolfson, 1985), and verbal fluency (DemTect: verbal fluency) were assessed. Attention was examined by assessing processing speed (Trail Making Test A), alertness (Test of Attentional Performance mobility, TAP-m; Zimmermann & Fimm, 2012, subtest alertness), and reaction control (TAP-M subtest Go/ NoGo).

2.4.2 Motor-functional outcome parameters

A physical therapist or rehabilitation nurse assessed bed mobility (Trunk Control Test, TCT; Fischer, 2014; Collin & Wade, 1990) as a basal motor function, balance (Berg Balance Scale, BBS; Berg et al., 1989) and hand function (Fist Closure Frequency, FCF; Grefkes et al., 2010; Sunderland et al., 1989) as measures of higher motor function, and functional independence (Functional Independence Measure; FIM; Keith et al., 1987). Patients’ neurological condition in terms of stroke severity was assessed with the National Institute of Health Stroke Scale (NIH-SS; Criddle et al., 2003) by the rehabilitation physicians.

2.5 Statistics

The statistical analysis was performed with SPSS 25.0 (IBM, Armonk, NY, USA).

As testing for normal distribution with the Kolmogorov-Smirnov-Test indicated that most variables were not normally distributed, non-parametrical analyses were performed.

Frequencies are reported for gender, education, handedness, lesion side, stroke type, and depression. For age and days since stroke onset, means and standard deviations were calculated. Results from cognitive and motor-functional testing are reported with medians and ranges. The significance level was set at p≤0.05, values of p≤0.1 were considered as statistical trends. Bonferroni-corrections were applied to correct for multiple testing. For the cognitive domains memory, executive function, and attention, as well as for the higher motor functions, the adjusted α-level was set at p≤0.016. Corrected p-values are reported.

To evaluate within-group effects of training, the Wilcoxon-Rank-Sum test for repeated testing (baseline x post-assessment) within related samples (EG; CG) was performed.

Between-group analyses for “baseline” and “change post training” (i.e., difference scores post minus pre-test) were performed with the Mann-Whitney-U test for independent samples (EG x CG).

Pearson’s (r) was calculated as effect sizes and interpreted according to Cohen (1992): r = 0.10 weak, r = 0.30 moderate, r = 0.50 strong.

Significant within-group cognitive changes were correlated with motor-functional parameters applying Kendall’s Tau-B (τ _B ) to examine associations between cognitive improvement and motor-functional parameters.

3 Results

Twenty-nine patients (mean age 71.86±17.99; 15 women) were included in the study, 16 were in the EG, and 13 were in the CG; no patient dropped out of the study after signing the informed consent. An overview of sociodemographic and disease-related characteristics of patients is given in Table 1. No significant between-group differences for these parameters were noticed at baseline.

Table 1
Overview of sociodemographic and disease-related parameters

EG (n = 16) CG (n = 13) p

Gender 1.000

Female: male 8: 8 7: 6

Age 74 (±21.63) 77 (±11.4) 0.368

Education 0.237

Technical college (min. 12 y.) 9 10

High school (min. 13y.) 3 2

University degree (>16y.) 4 1

Handedness 1.000

Right 14 12

Bilateral 2 1

Days since stroke onset 6 (±1.48) 6 (±2.18) 0.772

Lesion side 0.319

Right 7 9

Left 9 2

Bilateral 0 2

Stroke type 1.000

Ischemia 15 12

Intracerebral hemorrhage 1 1

Depression 0.632

No/ normal mood 14 10

Mildly depressed 1 3

	EG (n = 16)	CG (n = 13)	p
Gender			1.000
Female: male	8: 8	7: 6
Age	74 (±21.63)	77 (±11.4)	0.368
Education			0.237
Technical college (min. 12 y.)	9	10
High school (min. 13y.)	3	2
University degree (>16y.)	4	1
Handedness			1.000
Right	14	12
Bilateral	2	1
Days since stroke onset	6 (±1.48)	6 (±2.18)	0.772
Lesion side			0.319
Right	7	9
Left	9	2
Bilateral	0	2
Stroke type			1.000
Ischemia	15	12
Intracerebral hemorrhage	1	1
Depression			0.632
No/ normal mood	14	10
Mildly depressed	1	3

EG: experimental group, CG: control group; p: significance level.

3.1 Feasibility and adverse effects

All patients participating in the study tolerated the additional 20 minutes of training well and were highly motivated. No adverse effects were reported by patients, patients’ relatives, instructors, or members of the multidisciplinary team.

3.2 Baseline assessment

3.2.1 Cognitive outcome parameters

Cognitive impairment in one or multiple domains was observed in all patients (Table 2). In the EG 75% of the patients and in the CG 69.2% of the patients demonstrated cognitive impairment (EG: 19% MCI, 56% severe CI; CG: 23% MCI, 46% severe CI). The CG showed slightly better cognitive performance, even though group differences were not significant.

Table 2
Results for cognitive tests at baseline and post-training within and between both groups

Cognitive domains Baseline EG Post-training EG p Baseline CG Post-training CG p Changes between- groups; p

Overall cognition 8 (2;14) 12 (6;17) 0.003 ^* 9 (1;15) 12 (2;18) 0.213 0.122

Memory

Short-term memory 1.5 (0;3) 2 (0;3) 0.021 ^t 1 (0;3) 2 (0;4) 0.453 0.412

Long-term memory 1 (0;4) 2 (0;5) 0.057 ^t 2 (0;5) 2 (0;5) 0.549 0.517

Working memory 2 (1;3) 3 (0;3) 0.727 2 (1;3) 2 (0;3) 0.289 0.097 ^t

Executive function

Cognitive flexibility 1.5 (0;3) 3 (0;3) 0.008 ^* 3 (0;3) 2 (0;3) 0.453 0.008 ^*

Paced set-shifting 252.3 (110.1;475.8) 156.6 (67.7;326.4) 0.016 ^* 133.5 (85.3;282.4) 151.0 (78.5;193.8) 0.563 0.081 ^t

Verbal fluency 2 (0;4) 2 (0;4) 0.754 2 (0;4) 2 (0;4) 0.125 0.361

Attention

Processing speed 103.4 (35.8;310) 69.1 (24.9;310.2) 0.002 ^* 98.8 (42.7;271.3) 80.4 (34.1; 157.8) 0.160 0.283

Alertness 8 (0;8) 0 (0;4) 0.375 2.5 (0;11) 0 (0;12) 0.688 0.883

Reaction control 5 (0;21) 3 (0;14) 0.418 1 (0;17) 10 (4;14) 0.094 ^t 0.073 ^t

Cognitive domains	Baseline EG	Post-training EG	p	Baseline CG	Post-training CG	p	Changes between- groups; p
Overall cognition	8 (2;14)	12 (6;17)	0.003 ^*	9 (1;15)	12 (2;18)	0.213	0.122
Memory
Short-term memory	1.5 (0;3)	2 (0;3)	0.021 ^t	1 (0;3)	2 (0;4)	0.453	0.412
Long-term memory	1 (0;4)	2 (0;5)	0.057 ^t	2 (0;5)	2 (0;5)	0.549	0.517
Working memory	2 (1;3)	3 (0;3)	0.727	2 (1;3)	2 (0;3)	0.289	0.097 ^t
Executive function
Cognitive flexibility	1.5 (0;3)	3 (0;3)	0.008 ^*	3 (0;3)	2 (0;3)	0.453	0.008 ^*
Paced set-shifting	252.3 (110.1;475.8)	156.6 (67.7;326.4)	0.016 ^*	133.5 (85.3;282.4)	151.0 (78.5;193.8)	0.563	0.081 ^t
Verbal fluency	2 (0;4)	2 (0;4)	0.754	2 (0;4)	2 (0;4)	0.125	0.361
Attention
Processing speed	103.4 (35.8;310)	69.1 (24.9;310.2)	0.002 ^*	98.8 (42.7;271.3)	80.4 (34.1; 157.8)	0.160	0.283
Alertness	8 (0;8)	0 (0;4)	0.375	2.5 (0;11)	0 (0;12)	0.688	0.883
Reaction control	5 (0;21)	3 (0;14)	0.418	1 (0;17)	10 (4;14)	0.094 ^t	0.073 ^t

EG: experimental group, CG: control group; p: significance level; ^*: significant result; t: statistic trend; Assessments: overall cognition: DemTect; short-term memory: DemTect, subtest immediate recall wordlist; long-term memory: DemTect subtest delayed recall wordlist; working memory: DemTect subtest digit span reverse; cognitive flexibility: DemTect subtest transcoding; paced set-shifting: Trail Making Test part B; verbal fluency: DemTect subtest “supermarket”; processing speed: Trail Making Test part A; reaction control: Test of Attentional Performance (TAP) subtest Go/NoGo.

3.2.2 Motor-functional performance

At baseline, groups were comparable concerning most parameters (Table 3). More precisely, most patients showed impairments in bed mobility (TCT; EG: 42.5 (12; 100); CG: 24 (12; 100)). All patients had disturbed balance (BBS; EG: 0.5 (0; 49); CG: 0 (0; 42)) and hand function of the non-paretic (FCF; EG: 2.15 (1.1; 3.4); CG: 1.3 (0.6; 2.4)) and the paretic hand (FCF; EG: 0.47 (0; 2.9); CG: 0.1 (0; 1.6)) as measures of higher motor function. All patients were moderately to maximally dependent in ADLs according to the FIM (EG: 46.5 (25; 90); CG: 35 (18; 83)) and suffered from minor to moderate strokes (NIH-SS; EG: 7.5 (1; 13); CG: 8.0 (0; 13)).

Table 3
Results of motor-functional tests at baseline and post-training within and between both groups

Motor domains Baseline EG Post-training EG p Baseline CG Post-training CG p Changes between- groups; p

Basic motor function

Bed mobility 42.5 (12;100) 64 (12;100) 0.012 ^* 24 (12;100) 49 (12;100) 0.012 ^* 0.589

Higher motor function

Balance 0.5 (0;49) 7.5 (0;55) 0.001 ^* 0 (0;42) 6 (0;34) 0.057 ^t 0.171

Hand-function, non-paretic hand 2.15 (1.1;3.4) 2.25 (1.1;3.6) 0.790 1.3 (0.6;2.4) 1.9 (0.7;8) 0.013 ^* 0.080 ^t

Hand-function, paretic hand 0.47 (0;2.9) 1.1 (0;3.6) 0.011 ^* 0.1 (0;1.6) 0.7 (0;2.4) 0.008 ^* 0.311

Functional independence 46.5 (25;90) 78 (34;97) 0.002 ^* 35 (18;83) 52 (26;99) <0.001 ^* 0.357

Neurological condition 7.5 (1;13) 3,5 (0;9) <0.001 ^* 8 (0;13) 5 (0;12) 0.004 ^* 0.090 ^t

Motor domains	Baseline EG	Post-training EG	p	Baseline CG	Post-training CG	p	Changes between- groups; p
Basic motor function
Bed mobility	42.5 (12;100)	64 (12;100)	0.012 ^*	24 (12;100)	49 (12;100)	0.012 ^*	0.589
Higher motor function
Balance	0.5 (0;49)	7.5 (0;55)	0.001 ^*	0 (0;42)	6 (0;34)	0.057 ^t	0.171
Hand-function, non-paretic hand	2.15 (1.1;3.4)	2.25 (1.1;3.6)	0.790	1.3 (0.6;2.4)	1.9 (0.7;8)	0.013 ^*	0.080 ^t
Hand-function, paretic hand	0.47 (0;2.9)	1.1 (0;3.6)	0.011 ^*	0.1 (0;1.6)	0.7 (0;2.4)	0.008 ^*	0.311
Functional independence	46.5 (25;90)	78 (34;97)	0.002 ^*	35 (18;83)	52 (26;99)	<0.001 ^*	0.357
Neurological condition	7.5 (1;13)	3,5 (0;9)	<0.001 ^*	8 (0;13)	5 (0;12)	0.004 ^*	0.090 ^t

EG: experimental group, CG: control group; p: significance level; ^*: significant result; t: statistic trend. Assessments: bed mobility: Trunk Control Test; balance: Berg Balance Scale; non-paretic hand function: Fist Closure Frequency; paretic hand function: Fist Closure Frequency; functional independence: Functional Independence Measure; stroke-induced impairment assessed by the National Institute of Health Stroke Scale.

There were significant group differences for the paced hand coordination of the less affected hand (U = 57.500, p = 0.041, r = 0.38) and a trend for bed mobility (TCT; U = 65.000; p = 0.083) in favour of the EG.

3.3 Effects from baseline to post training

3.3.1 Cognition

Between group differences of change scores were significant for cognitive flexibility (DemTect, transcoding: U = 48.000, p = 0.008), and showed trends in reaction control (TAP Go/NoGo: U = 18.500, p = 0.073), paced set-shifting (TMT_B: U = 10.000, p = 0.081), and working memory (DemTect, digit span reverse: U = 68.500, p = 0.097) in favor of the EG after α-correction (Table 2 and Fig. 2).

Fig. 2

Percentage of training-induced improvements in cognitive domains for both groups. ^*: significant result; t: statistic trend.

Within-group analysis showed that the EG improved across all cognitive tests (Table 2): Significant improvements were observed for overall cognition (DemTect: z = –2.852; p = 0.003; r = 0.71), cognitive flexibility (DemTect, transcoding: z = –2.588; p = 0.008; r = 0.65), paced set-shifting (TMT_B: z = –2.380; p = 0.016; r = 0.60), and processing speed (TMT_A: z = –2.824; p = 0.002; r = 0.71). Improvements of short- and long-term memory (DemTect, immediate recall wordlist: z = –2.131; p = 0.021; r = 0.53; DemTect, delayed recall wordlist: z = –2.379; p = 0.057) showed a statistical trend after α-correction. For working memory (DemTect, digit span reverse: z = –0.905; p = 0.727), verbal fluency (DemTect, verbal fluency: z = –0.416; p = 0.754), alertness (TAP, alertness: z = –1.054; p = 0.375), and errors in reaction control (TAP Go/NoGo: z = –0.868, p = 0.418), better performance was observed post training, albeit that the differences were not statistically significant.

In the CG, no significant cognitive improvements were observed (Table 2). Rather, in reaction control, more errors were made; this difference showed a statistical trend towards a decrease (TAP Go/NoGo: z = –1.787; p = 0.094; r = 0.68). On a descriptive level, the CG also decreased in working memory, cognitive flexibility, and paced set-shifting.

3.3.2 Motor-functional performance

Both groups demonstrated motor functional improvement in the post-training assessment (Fig. 3). Between-group comparisons of the change scores showed that differences in stroke severity (NIH-SS; U = 65.500; p = 0.090; r = 0.32) and non-paretic hand function (U = 64.000; p = 0.080; r = 0.33) followed statistical trends (Table 3 and Fig. 3).

Fig. 3

Percentage of training induced improvements in motor-functional parameters in both groups. ^*: significant result; t: statistic trend.

The EG significantly improved (Table 3) in bed mobility tasks (TCT: z = -2.451; p = 0.012; r = 0.61), balance (BBS: z = –3.298; p = 0.001; r = 0.82), paretic hand function (FCF: z = –2.472; p = 0.011; r = 0.62), functional independence (FIM: z = –2.948; p = 0.002; r = 0.74), and recovery from stroke-induced impairment (NIH-SS: z = –3.376; p < 0.001; r = 0.84).

The CG also showed significant improvements (Table 3) in bed mobility (TCT: z = –2.532; p = 0.012; r = 0.70), non-paretic (FCF: z = –2.395; p = 0.013; r = 0.66) and paretic hand function (FCF: z = –2.536; p = 0.008; r = 0.70), functional independence (FIM: z = –3.064; p < 0.001; r = 0.85), and recovery from stroke-induced impairment (NIH-SS: z = –2.677; p = 0.004). The assessment of balance tasks showed a trend towards an improvement (BBS: z = –1.940; p = 0.057; r = 0.82).

3.4 Correlation analysis

Correlations were analyzed between those change scores of cognitive parameters that showed significant improvement or a trend for improvement within the groups and their motor-functional change scores.

For the EG, a correlation was observed between changes in overall cognition and changes in bed mobility (DemTect x TCT: τ _B = 0.570; p = 0.004) as well as changes in long-term memory and changes in bed mobility (DemTect, delayed recall wordlist x TCT: τ _B = 0.664; p = 0.001).

In the CG, increased errors in reaction control correlated with less improvement in balance (TAP, Go/NoGo x BBS: τ _B = 0.650; p = 0.046).

4 Discussion

We evaluated the effects of a combined cognitive and motor training on the early rehabilitation outcome after stroke compared to pure motor training of equal intensity. We also assessed the feasibility of our study protocol concerning its implementation into the routine setting of an early rehabilitation unit and the occurrence of adverse events.

In our group of stroke patients, who all showed cognitive and motor impairment in the early phase after the stroke, we observed that both the EG and the CG improved in motor functions. However, only patients receiving combined CT and MT showed significant gains in cognitive flexibility and trends for paced set-shifting, working memory, and reaction control. Within the EG, significant improvement or trends were observed across all cognitive domains assessed, which correlated with gains in bed-mobility. In contrast, the CG showed no significant improvements. Instead, a descriptive decline in cognitive flexibility, paced set-shifting, and working memory, as well as a significant trend for a decline in reaction control, which correlated with less functional progression and recovery, occurred. The observed effects had moderate to strong effect sizes and, therefore, constitute a promising target for a future RCT. Furthermore, the study protocol was successfully integrated into the daily routine of the early rehabilitation program. No adverse effects were reported.

4.1 Cognitive improvement in EG

The change scores suggest that combined CT and MT improved cognition more than the control training. Between-group comparisons revealed statistically differing cognitive gains for cognitive flexibility, reaction control, paced set-shifting, and working memory in favour of the EG. The EG improved in six of ten examined cognitive domains, while CG showed a decline in four of ten domains, even though only the difference in reaction control showed a statistical trend.

These observations confirm our first hypothesis. They also agree with the existing literature suggesting that cognition needs to be targeted directly through training and does not benefit from standard motor therapy alone (Cicerone, 2000). Our results are furthermore consistent with previous reports that multidimensional cognitive training is superior to single-domain training (Cicerone et al., 2000; Cumming et al., 2012; Rahe et al., 2015a; Petrelli et al., 2014), and that effects are more significant when a cognitive approach is combined with physical training (Carter et al., 1988; Kalra et al., 1997; Rahe et al., 2015a). Moreover, the observed decline in cognitive domains in the CG confirms previous reports, especially when therapists are unaware of the problems resulting from cognitive dysfunction (Hostenbach et al., 1998; Hostenbach & Mulder, 1999; Robertson, 1997; Skidmore et al., 2010).

4.2 Clinically relevant motor-functional improvement in EG

Training resulted in improved motor performance in both groups. The CG improved significantly more in non-paretic hand function, whereas the EG showed significantly less stroke-induced impairment (NIH-SS).

The severity of the stroke-induced impairment early after stroke is a predictor of recovery (Schlegel et al., 2003). Therefore, this improvement in the EG is of high clinical relevance. It is conceivable that no additional significant differences between groups were observed after eight days of training because such effects might need more time. In line with this assumption, descriptive results from this study confirm that overall motor-functional gains were less high in the CG than in the EG. Taken together with the observation that CG patients did not improve in cognition, these observations converge with previous evidence suggesting an impact of cognitive dysfunctions on motor relearning and rehabilitation outcome (McDonald et al., 2019; Farokhi-Sisakht et al., 2019; Rohde et al., 2019; Cumming et al., 2013; Nys et al., 2007; Hostenbach et al., 1998; Skidmore et al., 2010).

Interestingly, a significant improvement of the non-paretic hand was observed in the CG but not in the EG. In the CG, the non-paretic hand was the dominant hand in most patients (12/ 13). A significant improvement of the non-paretic hand, together with less improvement in the neurological condition and cognitive functions, is compatible with a compensatory overuse of the unaffected hand at the cost of recovery of the affected hand. Overactivity of the unaffected hand early after stroke, as seen in the CG, has previously been associated with compensation on a behavioural and neuronal level (Grefkes et al., 2010; Sunderland et al., 1989).

4.3 Cognitive gains correlate with functional performance in EG

In the EG, improvement in overall cognition and long-term memory were associated with gains in basic motor functions. This observation is relevant since basic motor functions two weeks post-stroke are a predictor of ambulation six months later and a prerequisite of higher-level motor functions (Verheyden et al., 2006; 2007).

In the CG, the decline in reaction control correlated positively with gains in balance tasks. This correlation was not expected and an explanation stays elusive. Therefore, further studies will have to investigate whether this result is replicable before any conclusion can be drawn.

Overall, the number of identified correlations between significant cognitive change scores and change scores of motor-functional performance was limited, so any conclusion needs to be drawn with caution and further research is needed to evaluate the correlation between cognitive gains and motor functional improvement.

4.4 Limitations

Some limitations have to be considered when interpreting the results of this pilot study combining CT and MT in the early phase after a stroke on eight consecutive days in a controlled approach. Limitations are the small sample size as well as the non-randomized group assignment, albeit that the group assignment followed an a-priori defined strategy.

Another limitation is that in some of the patients, the cognitive impairment resulted in missing values for paced-set shifting (missing in EG n = 8; CG n = 7) and reaction control (missing in EG n = 2; CG n = 6). However, in the early acute phase after a stroke, this constitutes a realistic setting. It would have been possible to avoid missing values by including “assessment completion” in the inclusion criteria, but that would result in adapting patients to test modalities rather than representing clinical reality in an acute setting.

As a further limitation, we did not document and report daily factors that have potentially influenced cognitive performance in CG (and EG). However, we would like to emphasize that our inclusion and exclusion criteria were broad and defined that patients with previous or current neurological or psychiatric disorders (e.g. occurrence of seizures, chronic headaches, pain syndromes or mental health issues) as well as infections such as pulmonary or urinary infections were also considered for the study as these patients are likely to miss training days or require additional medical exams due to their constitution. Nevertheless, for future trials daily records of patients’ conditions will be important when interpreting effects of (non-pharmacological) interventions.

5 Conclusion

Combined CT and MT on eight consecutive days in the early phase after a stroke improved cognition as an essential factor impacting motor relearning and rehabilitation outcome. In contrast, MT alone helped the control group to regain moderate independence, but not to recover higher-level motor functions, and did not prevent cognitive decline. The latter finding is in line with previous descriptions of cognitive decline after stroke (Sun et al., 2014; Rasquin et al., 2004; 2005; Vataja et al., 2003), indicating that cognition and motor functions should be considered as linked entities and trained in combined approaches starting immediately after a stroke. The data also suggest that neglecting the potential for rehabilitation in the early phase has a relevant, negative impact on the outcome and increases the risk of worsening cognitive dysfunctions. Further research with multicenter RCTs, allowing for larger sample sizes, and more extended follow-up periods are warranted to prove these preliminary, however promising results. Besides, including functional brain imaging in the study protocol prior and after training could help to disentangle the neural mechanisms underlying the differential effects of combined CT and MT relative to MT alone.

Conflict of interest

MG has received honoraria from Bernafon AG, Bern, Switzerland. EK has received grants from the German Ministry of Education and Research, ParkinsonFonds Deutschland gGmbH, the German Parkinson Society; honoraria from: Oticon GmbH, Hamburg, Germany; Lilly Pharma GmbH, Bad Homburg, Germany; Bernafon AG, Bern, Switzerland; Desitin GmbH, Hamburg, Germany. EK is author of the cognitive training program NEUROvitalis but receives no corresponding honoraria. LB, GRF and OAO declare no conflict of interest.

Funding

ME was funded by the Student Support Fund 17000385-4000285 of Cologne University. GRF gratefully acknowledges additional support by the Marga- and Walter Boll Foundation, Kerpen, Germany.

Footnotes

Acknowledgments

We thank all patients for volunteering for this study and all members of the multidisciplinary team of the neurology department for their valuable support. We also thank Mrs. Ann-Kristin Folkerts for her critical review of the manuscript.

References

Baller,

, Kalbe,

, Kaesberg,

, & Kessler,

(2010). NEUROvitalis. Ein Neuropsychologisches Gruppenprogramm zur Förderung der Geistigen Leistungsfähigkeit. Köln, ProLog.

Baller,

, Kalbe,

, Kaesberg,

, & Kessler,

(2017). NEUROvitalis Home. Einzelübungen für Eigentraining und Therapie. Köln, ProLog.

Barnett,

A. M.

, & Muzaffar,

(2014). Spatial cognitive rehabilitation and motor recovery after stroke. Curr Opin Neurol, 27(6), 653–658.

Berg,

, Wood-Dauphinee,

, Williams,

, & Gayton,

(1989). Measuring balance in the elderly preliminary development of an instrument. Physiother Can, 41, 304–11.

Brewer,

, Horgan,

, Hickey,

, & Williams,

(2013). Stroke rehabilitation: recent advances and future therapies. Q J Med, 106, 11–25.

Carr,

J. H.

, & Shepherd,

R. B.

(2006). The changing face of neurological rehabilitation. Rev Bras Fisioter, 10(2), 147–156.

Carter,

L. T.

, Howard,

B. E.

, & O’Neil,

W. A.

(1983). Effectiveness of cognitive skill remediation in acute stroke patients. Am J Occup Ther, 37(5), 320–326.

Carter,

, Oliveira,

, Duponte,

, & Lynch,

(1988). The relationship of cognitive skills performance to activities of daily living in stroke patients. Am J Occup Ther, 42(7), 449–455.

Chen,

, Leys,

, & Esquenazi,

(2013). The interaction between neuropsychological and motor deficits in patients after stroke. Neurology, 80(3 Suppl. 2), 27–34.

10.

Cicerone,

K. D.

, Dahlberg,

, Kalmar,

, Langenbahn,

D. M.

, Malec,

J. F.

, Bergquist,

T. F.

, et al. (2000). Evidence-based cognitive rehabilitation: Recommendations for clinical practice. Arch Phys Med Rehabil, 81(12), 1596–1615.

11.

Cicerone,

K. D.

, Dahlberg,

, Malec,

J. F.

, Langenbahn,

D. M.

, Felicetti,

, Kneipp,

, et al. (2005). Evidence-Based Cognitive Rehabilitation: Updated Review of the Literature From 1998 Through 2002. In: Arch Phys Med Rehabil, 86(8), 1681–1692.

12.

Cicerone,

K. D.

, Langenbahn,

D. M.

, Braden,

, Malec,

J. F.

, Fraas,

, Kalmar,

, et al. (2011). Evidence-based cognitive rehabilitation: updated review of the literature from 2003 through 2008. Arch Phys Med Rehabil, 92(4), 519–530.

13.

Cohen,

(1992). A power primer. Psychological Bulletin, 112(1), 155–159.

14.

Collin,

, & Wade,

(1990). Assessing motor impairment after stroke: a pilot reliability study. J Neurol Neurosurg Psychiatry, 53(7), 576–579.

15.

Criddle,

L. M.

, Bonnono,

, & Fischer,

S. K.

(2003). Standardizing stroke assessment using the national institutes of health stroke scale. J Emerg Nurs, 29(6), 541–546.

16.

Cumming,

T. B.

, Tyedin,

, Churilov,

, Morris,

M. E.

, & Bernhardt,

(2012). The effect of physical activity on cognitive function after stroke: a systematic review. Int Psychogeriatr, 24(4), 557–567.

17.

Cumming,

T. B.

, Marshall,

R. S.

, & Lazar,

R. M.

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

18.

das Nair,

, Cogger,

, Worthington,

, & Lincoln,

N. B.

(2016). Cognitive rehabilitation for memory deficits after stroke. Cochrane Database Syst Rev, 9, CD002293.

19.

Douiri,

, Rudd,

A. G.

, & Wolfe,

C. D.

(2013). Prevalence of poststroke cognitive impairment: South London Stroke Register 1995-2010. Stroke, 44, 138–45.

20.

Farokhi-Sisakht,

, Farhoudi,

, Sadigh-Eteghad,

, Mahmoudi,

, & Mohaddes,

(2019). Cognitive Rehabilitation Improves Ischemic Stroke-Induced Cognitive Impairment: Role of Growth Factors. J Stroke Cerebrovc Dis, 28(10), 104299.

21.

Feigin,

V.L.

, Forouzanfar,

M.H.

, Krishnamurthi,

, Mensah,

G.A.

, Connor,

, Bennett, et al. (2014). Global and regional burden of stroke during 1990-2010: findings from the Global Burden of Disease Study 2010. The Lancet, 383(9913), 245–254.

22.

Fischer,

(2014). Rumpfkontrolle erfassen. Trunk Control Test. Ergopraxis, 6, 32–33.

23.

Grefkes

, Nowak

D. A.

, Wang

L. E.

, Dafotakis

, Eickhoff

S. B.

, & Fink

G. R.

(2010). Modulating cortical connectivity in stroke patients by rTMS assessed with fMRI and dynamic casual modeling. Neuroimage, 50, 234–243.

24.

Hostenbach,

, Mulder,

, van Limbeck,

, Donders,

, & Schoonwaldt,

(1998). Cognitive Decline Following Stroke: A Comprehensive Study of Cognitive Decline Following Stroke. J Clin Exp Neuropsy, 20(4), 503–517.

25.

Hostenbach,

, & Mulder,

(1999). Neuropsychology and the relearning of motor skills following stroke. Int J of Rehabil Res, 22, 11–19.

26.

Ihle-Hansen,

, Thommessen,

, Wyller,

T. B.

, Engedal,

, Øksengård,

A. R.

, Stenset,

, et al. (2011). Incidence and subtypes of MCI and dementia 1 year after first-ever stroke in patients without pre-existing cognitive impairment. Dement Geriatr Cogn Disord, 32, 401–7.

27.

Kalaria,

R. N.

, Akinyemi,

, & Ihara,

(2016). Stroke injury, cognitive impairment and vascular dementia. Acta Biochim Biophys, 1862(5), 915–925.

28.

Kalbe,

, Kessler,

, Calabrese,

, Smith,

, Passmore,

A. P.

, Brand,

, et al. (2004). DemTec: a new, sensitive cognitive screening test to support the diagnosis of mild cognitive impairment in early dementia. Int J Geriatr Psychiatry, 19(2), 136–143.

29.

Kalra,

, Perez,

, Gupta,

, & Wittink,

(1997). The influence of visual neglect on stroke rehabilitation. Stroke, 28, 1386–1391.

30.

Keith,

R. A.

, Granger,

C. V.

, Hamilton,

B. B.

, & Sherwin,

F. S.

(1987). The functional independence measure: a new tool for rehabilitation. Adv Clin Rehabil, 1, 6–18.

31.

Kessler,

, Calabrese,

, & Kalbe,

(2010). DemTect-B: ein Äquivalenztest zum kognitiven Screening DemTect-A^®. Fortschr Neurol Psychiatr, 78(9), 532–535.

32.

Khan,

, Amatya,

, Galea,

M. P.

et al. (2017). Neurorehabilitation: applied neuroplasticity. J Neurol, 264, 603–615.

33.

Kleim

J. A.

(2011). Neural plasticity and neurorehabilitation: teaching the new brain old tricks. In: Journal of communication disorders, 44(5), pp. 521–528.

34.

Kwakkel,

, Kollen,

, & Linderman,

(2004). Understanding the pattern of functional recovery after stroke: facts and thoeries. Restor Neurol Neurosci, 22, 281–299.

35.

McDonald,

M. W.

, Black,

S. E.

, Copland,

D. A.

, Corbett,

, Dijkhuizen,

R. M.

, Farr,

T. D.

, et al. (2019). Cognition in stroke rehabilitation and recovery research: Consensus-based core recommendations from the second Stroke Recovery and Rehabilitation Roundtable. Int J Stroke, 12, 1747493619873600.

36.

Mehrholz,

, Carr,

, Flämig,

, Grellmann,

, Oehmichen,

, Pohl,

, Schlosser,

, & Shepherd,

R. B.

(2008). Frühphase nach Schlaganfall: Physiotherapie und medizinische Versorgung. Stuttgart, New York, Dehli: Georg Thieme.

37.

Mijajlović,

M. D.

, Pavlović,

, Brainin,

, Heiss,

W. D.

, Quinn,

T. J.

, Ihle-Hansen,

H. B.

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

38.

Morsund,

Å.H.

, Ellekjær,

, Gramstad,

, Reiestad,

M. T.

, Midgard,

, Sando

S. B.

, et al. (2019). The development of cognitive and emotional impairment after a minor stroke: A longitudinal study. Acta Neurol Scand, 140, 281–289.

39.

Nudo,

R. J.

(2013). Recovery after brain injury: Mechanisms and principles. Front Hum Neurosci, 7, 887.

40.

Nys,

G. M. S.

, van Zandvoort,

M. J.

, de Kort,

P. L.

, Jansen,

B. P.

, de Haan,

E. H.

, & Kappelle,

L. J.

(2007). Cognitive disorders in acute stroke: prevalence and clinical determinant. Cerebrovasc Dis, 23(5-6), 408–416.

41.

Nys,

G. M.

, van Zandvoort,

M. J.

, de Kort,

P. L.

, Jansen,

B. P.

, Kappelle,

L. J.

, & de Haan,

E. H.

(2005). Restrictions of the Mini-Mental State Examination in acute stroke. Arch Clin Neuropsychol, 20, 623–629.

42.

Pasi,

, Poggesi,

, Salvadori,

, & Pantoni.

(2012). Post-stroke dementia and cognitive impairment. Front Neurol Neurosci, 30, 65–69.

43.

Petrelli,

, Kaesberg,

, Barbe,

M. T.

, Timmermann,

, Fink,

G. R.

, Kessler,

, et al. (2014). Effects of cognitive training in Parkinson’s disease: A randomized controlled trial. Parkinsonism Relat Disord, 20(11), 1196–1202.

44.

Rahe,

, Petrelli,

, Kaesberg,

, Fink,

G. R.

, Kessler,

, Kalbe,

, (2015a). Effects of cognitive training with additional physical activity compared to pure cognitive training in healthy older adults. Clin Interv Aging, 10, 297–310.

45.

Rahe,

, Liesk,

, Rosen,

J. B.

, Petrelli,

, Kaesberg,

, Onur,

O. A.

, Kessler,

, Fink,

G. R.

, & Kalbe,

, (2015b). Sex differences in cognitive training effects of patients with amnestic mild cognitive impairment. Neuropsychol Dev Cogn B Aging Neuropsychol Cogn, 22(5), 620–638.

46.

Rasquin,

S. M.

, Verhey,

F. R.

, Oosterbrugge,

R. J.

, Lousberg,

, & Lodder,

(2004). Demographic and CT scan features related to cognitive impairment in the first year after stroke. In: J Neurol Neurosurg Psychiatry, 75(11), 1562–1567.

47.

Rasquin,

S. M.

, Verhey,

F. R.

, Lousberg,

, & Lodder,

(2005). Cognitive performance after first ever stroke related to progression of vascular brain damage: a 2 year follow up CT scan study. J Neurol Neurosurg Psychiatry, 76, 1075–1079.

48.

Reitan,

, & Wolfson,

(1985). The Halstead- Reitan Neuropsychological Test Battery: Therapy and clinical interpretation. Tucson, AZ: Neuropsychological Press.

49.

Rist,

P. M.

, Chalmers,

, Arima,

, Anderson,

, Macmahon,

, Woodward,

, et al. (2013). Baseline cognitive function, recurrent stroke, and risk of dementia in patients with stroke. Stroke, 44, 1790–1795.

50.

Robertson,

I. H.

, Ridgeway,

, Greenfield,

, & Parr,

(1997). Motor recovery after stroke depends on intact sustained attention: a 2-year follow-up study. Neuropsychology, 11(2), 290–295.

51.

Rohde,

, Gaynor,

, Large,

, Mellon,

, Hall,

, Brewer,

, et al. (2019). The Impact of Cognitive Impairment on Poststroke Outcomes: A 5-Year Follow-Up. J Geriatr Psychiatry Neurol, 32(5), 275–281.

52.

Schlegel,

, Kolb,

S. J.

, Luciano,

J. M.

, Tovar,

J. M.

, Cucchiara,

B. L.

, Liebeskind,

D. S.

, et al. (2003). Utility of the NIH Stroke Scale as a predictor of hospital disposition. Stroke, 34(1), 134–137.

53.

Shepherd,

R. B.

, & Carr,

(2005). Scientific Basis of Neurological Physiotherapy: Bridging the Gap between Science and Practice. Neurol Rehabil, 11(1), 1–6.

54.

Skidmore,

E. R.

, Whyte,

E. M.

, Holm,

M. B.

, Becker,

J. T.

, Butters,

M. A.

, Dew,

M. A.

, et al. (2010). Cognitive and affective predictors of rehabilitation participation after stroke. Arch Phys Med Rehabil, 91(2), 203–207.

55.

Sun,

J. H.

, Tan,

, & Yu,

J. T.

(2014). Post-stroke cognitive impairment: epidemiology, mechanisms and management. Ann Transl Med, 2(8), 80–96.

56.

Sunderland

, Tinson

, Bradley

, & Hewer

R. L.

(1989). Arm function after stroke. An evaluation of grip strength as a measure of recovery and a prognostic indicator. J Neurol Neurosurg Psychiatry, 52(11), 1267–1272.

57.

Tatemichi,

T. K.

, Desmond,

D. W.

, Stern,

, Paik,

, Sano,

, & Bagiella,

(1994). Cognitive impairment after stroke: frequency, patterns, and relationship to functional abilities. J Neurol Neurosurg Psychiatry, 57, 202–207.

58.

Theill,

, Schumacher,

, Adelsberger,

, Martin,

, & Jäncke,

(2013). Effects of simultaneously performed cognitive and physical training in older adults. In: BMC Neurosci, 14, 103.

59.

Vataja,

, Pohjasvaara,

, Mäntylä,

, Ylikoski,

, Leppävuori,

, Leskelä,

, Kalska,

, Hietanen,

, Aronen,

H.J.

, Salonen,

, Kaste,

, & Erkinjuntti,

(2003). MRI correlates of executive dysfunction in patients with ischaemic stroke. In: Eur J Neurol, 10(6), 625–631.

60.

Verheyden,

, Vereeck,

, Truijen,

, Troch,

, Herregodts,

, Lafosse,

, et al. (2006). Trunk performance after stroke and the relationship with balance, gait, and functional ability. Clinic Rehabil, 20(5), 451–458.

61.

Verheyden,

, Nieuwboer,

, De Wit,

, Feys,

, Schuback,

, Baert,

, et al. (2007). Trunk performance after stroke: an eye catching predictor of functional outcome. J Neurol Neurosurg Psychiatry, 78(7), 694–698.

62.

World Health Organization, WHO. (2008). Burden of Disease Statistics Update 2004. Geneva, Switzerland: World Health Organization, http://www.who.int/healthinfo/bod/en/index.html (November, 2016)

63.

Zimmermann,

, & Fimm,

(2012). Testbatterie zur Aufmerksamkeitsprüfung (Version Mobilität) Version 1.3.; Vera Fimm Psychologische Testsysteme.

64.

Zorowitz,

R. D.

, Gross,

, & Polinski,

D. M.

(2002). The stroke survivor. Disabil Rehabil, 24(13), 666–679.