qqq

1. ORIGINAL ARTICLE
Association between cardiac autonomic control and cognitive
performance among patients post stroke and age-matched healthy
controls—an exploratory pilot study
Noa Raphaely Beer1
& Nachum Soroker2,3
& Nathan M. Bornstein4
& Michal Katz Leurer1
Received: 28 March 2017 /Accepted: 16 August 2017 /Published online: 7 September 2017
# Springer-Verlag Italia S.r.l. 2017
Abstract Associations between autonomic nervous system
health and cognitive performance have been described in dif-
ferent populations. Autonomic disturbances are a common
phenomenon in patients post stroke. Little is known about
the relationship between post stroke disturbances of the auto-
nomic nervous system and the commonly occurring distur-
bances of cognitive functions revealed by victims of stroke.
To assess the association between heart rate variability
(HRV) and cognitive performance among patients post ische-
mic stroke and healthy age-matched controls, 13 patients post
first-ever ischemic stroke aged 40–80 years and 15 age-
matched healthy controls were evaluated. HRV was monitored
during sustained handgrip, while breathing at a rate of six
breaths per minute, while performing the serial-3 subtraction
task sitting at rest, and while cycling. Patients post stroke had
greater error rate in the serial-3 subtraction task, and lower
HRV (both at rest and during task performance) relative to
healthy controls (at rest 26 ms [10–53] vs. 43 ms [29–88]).
The HRVof stroke patients showed less sensitivity to changes
in testing conditions, and also failed to show the correlation
with cognitive performance exhibited by the healthy subjects.
Stroke patients experience autonomic nervous system dysfunc-
tion in parallel to their motor and cognitive impairments. Too
often only the latter receive appropriate treatment consider-
ation in the rehabilitation setting. The current results, and ear-
lier research, point to the importance of focusing clinical atten-
tion to the status of the autonomic nervous system, as amelio-
ration of its functioning is likely to enhance motor and cogni-
tive functioning as well.
Keywords Stroke . Autonomic nerve system . Cognition
Introduction
Heart rate variability (HRV), with alterations in sinus rhythm,
is a well-studied marker of cardiovascular autonomic function
[1]. Reduced HRV is a strong independent predictor of cardio-
vascular events and mortality [2], and is linked to a number of
risk factors for cognitive impairment, including hypertension
[3], diabetes mellitus [4], depression [5], and subclinical in-
flammation [6].
Associations between autonomic nervous system health and
cognitive performance have been described in different popu-
lations. Hansen et al. described differences in cognitive abilities
in relation to HRV in 37 young healthy male sailors—those
with high HRV performed better on executive tasks compared
to those with low HRV [7]. In people with Alzheimer’s disease,
more severe cognitive decline was found to be associated with
lower cardiac parasympathetic modulation and higher cardiac
sympathetic modulation [8]. Luft et al. described significant
differences in HRVamong 30 young high-level athletes, during
performance of executive and non-executive tasks [9].
Mukherjee et al. found in elderly adults that different levels of
mental workload had different effects on HRV—the greater the
cognitive load, the lower the HRV [10].
This work was performed in partial fulfillment of the requirements for a
Ph.D. degree of Noa Raphaely Beer, Sackler Faculty of Medicine, Tel
Aviv University, Israel.
* Michal Katz Leurer
michalkz@post.tau.ac.il
1
Physical Therapy Department, School of Health Professions, Sackler
Faculty of Medicine, Tel Aviv University, 69978 Tel Aviv, Israel
2
Neurological Rehabilitation Department, Loewenstein Hospital,
Raanana, Israel
3
Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel
4
Neurology Department, Sourasky Medical Center, Tel Aviv, Israel
Neurol Sci (2017) 38:2037–2043
DOI 10.1007/s10072-017-3097-0

2. Autonomic disturbances are a common phenomenon in
patients post stroke [11], and reduced HRV was found to be
related with adverse post-event clinical outcomes [11, 12].
Little is known about the relationship between post-stroke
disturbances of the autonomic nervous system and the com-
monly occurring disturbances of cognitive functions revealed
by victims of stroke [13]. The aim of the current pilot study
was to shed light on this relationship by an assessment of the
association between HRV and cognitive performance among
stroke patients, as compared to the association in age-matched
healthy controls.
Materials and methods
Participants
Thirteen patients (10 males, 3 females) post first-ever ische-
mic stroke, seven with right-hemisphere damage and six with
left-hemisphere damage, were recruited for the study from the
Neurology Department at Sourasky Medical Center Tel Aviv
and the Department of Neurological Rehabilitation at the
Loewenstein Rehabilitation Center, Raanana, Israel, 1–
4 weeks post the event. A convenience sample of 15 age-
matched healthy volunteers (nine males, six females) served
as the control group. Inclusion criteria for patients were as
follows: first event, unilateral hemispheric damage, age 40–
80 years at onset, with no previous neurological or cardiac
medical history, clinically stable state at the time of testing,
lack of major hearing or vision impairment, dementia, or de-
pression. Ethical approval was obtained from the Human
Research Ethics Committee of both hospitals and the Health
Ministry of Israel. All participants signed informed consent
forms prior the study.
Being a case control study, sample size calculation was
based on a one-sided assumption that a large effect size
exists in HRV parameters at rest between stroke patients
and healthy controls. Francica et al. noted a significant
and large effect size in HRV parameters (effect size > 1)
between patients post stroke and healthy controls [14].
With a type 1 error of 0.05 and power of 0.8, the minimal
sample size needed was 22 participants: 11 post stroke
and 11 as controls.
Tools
Heart rate was monitored by the Polar Advanced Heart Rate
Monitor (RS800CX) validated for this purpose [15]. An elas-
tic electrode transmitter belt (Polar sensor) was placed on the
lower chest. The electrodes detect the voltage differential on
the skin during every heart beat and send the signal wirelessly
to the Polar receiver unit, which is then transferred via Polar-
specific software (Polar® ProTrainer 5 software) to a
computer. The data is then exported as a text file to the HRV
analysis software (Kubios HRV software ver. 2.0; Biosignal
Analysis and Medical Imaging Group, Department of Physics,
University of Kuopio, Kuopio, Finland) for analysis of the
following HRV parameters: The standard deviation of normal
R–R intervals (SDNN) and the root-mean-square difference of
successive normal R–R intervals (RMSSD) were used as in-
dexes of total variance and the vagal control within the time
domain, respectively [16]. The high frequency (HF; 0.15–
0.40 Hz) was used as the index of vagal tone in the frequency
domain [15].
Manipulation of Autonomic Nervous System condition (after
Ewing [17])
Induction of increased parasympathetic tone: Participants
were asked to breathe at a slow rate (six breaths per minute).
Induction of increased sympathetic tone: Participants were
requested to sustain a handgrip at one third of maximum vol-
untary contraction pressure for 2 min using the dynamometer
scale directly in front of them.
Cognitive tests
The Montreal Cognitive Assessment Scale (MoCA) was used
to assess each patient’s basic cognitive ability. The MoCA is a
screening test including subtests that assess executive func-
tions and psychomotor speed [18] which are frequently im-
paired in patients post stroke [13]. Its score range is between 0
and 30, reflecting success tasks employed in orientation, fig-
ure drawing, processing speed, object naming, memory recall,
attention, vigilance, repetition, verbal fluency, and abstraction.
The MoCA adds one point for those whose educational level
is 12 years or less.
The serial-3 subtraction task, involving calculation, work-
ing memory, and attention systems, was used for testing the
relationship between HRV and cognitive performance.
Participants received triple digit numbers and were asked to
perform counting downwards by threes during 1 min.
Performance scoring included the total number of subtractions
in 1 min, the number of errors, and the proportion of errors out
of the total number of subtraction steps. The test was per-
formed twice—while sitting at rest (Bsingle task,^ an easier
condition) and while cycling at a comfortable pace (Bdual
task,^ a more difficult condition). Previous studies have
shown the advantage of using the serial-3 subtraction test
while performing a second task [19, 20].
In addition, patient’s activity level assessed by the
modified Rankin scale (mRS) [21] and basic activities of
daily living (ADL) was assessed by the Barthel Index (BI)
[22].
2038 Neurol Sci (2017) 38:2037–2043

3. Protocol
After being seated with the polar belt attached, subjects per-
formed the MoCA test. Then they had a rest period for 10 min,
followed by heart rate recording in the following order of
conditions: (1) during sustained handgrip using one third of
maximum voluntary contraction pressure for 2 min (sympa-
thetic tone induction); (2) breathing at a rate of six breaths per
minute (parasympathetic tone induction); (3) while
performing the timed serial-3 subtraction task sitting at rest
(single task); and (4) performing the timed serial-3 subtraction
task while cycling at a comfortable pace for 1 min (dual task).
It should be noted that this order of testing conditions is likely
to facilitate subtraction during cycling, as this condition (dual
task) comes after performance of subtraction at rest (single
task). Moreover, cycling, being a common motor activity, is
unlikely to slow down the cognitive task by recruiting atten-
tional resources.
Statistical analysis
The Kolmogorov–Smirnov test was performed for all out-
come measures. As the data distribution violated the assump-
tion of normality, non-parametric statistics were used. The
Spearman correlation coefficient was performed to assess the
correlation between the subjects’ cognitive status, as revealed
by the MoCA and serial-3 tests, and the HRV values at rest
and during the different test conditions. The Fisher r-to-z
transformation was used to assess the different correlations
between groups. The Mann–Whitney U test was used to as-
sess differences between groups and the Friedman and
Wilcoxon test for assessing differences within each group.
All statistical analyses were performed using SPSS v. 22 soft-
ware packages. A p value < 0.05 was considered statistically
significant.
Results
Based on the mRS, the patients’ functioning level was 2,
Bslight disability—able to look after one’s affairs without as-
sistance, but unable to carry out all previous activities^ [1–4].
The Barthel Index mean score was 83 [35–100]. The mean
age of stroke patients was 61.3 years (± 8.5) and that of
healthy controls 61.7 (± 9.6). The two groups were found to
differ in educational level (years of formal schooling among
stroke patients 13 [10–22] and among controls 18 [13–25];
p = 0.001). The median MoCA score of the stroke group was
26 (range 18–30) and that of healthy controls 27 (range 24–
30). The difference between the two groups was significant
(p = 0.02) (Table 1).
Stroke patients revealed a significant disadvantage relative
to healthy controls also in the timed serial-3 subtraction task,
both at rest (single task) and during cycling (dual task). The
results of the two groups (number of subtraction steps reached
in 1 min, number of errors, and percent error rate) are shown in
Table 1.
Unlike healthy controls who performed the serial-3 task
differently in the first (single) and the subsequent (dual) task
conditions (22 and 28 subtraction steps in 1 min, respectively,
p value = 0.002), the median group value among stroke pa-
tients was equal (12) in both conditions (p value = 0.98)
(Table 1).
Significant differences were noted between stroke patients
and healthy control subjects in HRV parameters. Patients post
stroke exhibited significantly lower SDNN and RMSSD
values at rest. SDNN values were significantly higher among
the healthy control subjects in all testing conditions (Fig. 1).
The RMSSD values were significantly higher in the control
group in all testing conditions except dual task (serial-3 sub-
traction during cycling, Z = 0.9, p = 0.19). (Fig. 2).
Within-group analyses revealed significant changes in
SDNN values between study conditions (Friedman test, pa-
tients post stroke χ2
4 = 11.0, p value 0.02; healthy controls
χ2
4 = 17.2, p value 0.002). Among patients post stroke, sig-
nificant differences were noted in post hoc analysis of SDNN
values between rest and the single task (serial-3 at rest) con-
dition, using the Wilcoxon signed-rank test (Z = 2.51,
p = 0.006). Among healthy controls, post hoc analysis re-
vealed significant differences between rest and each of the
consequent study conditions except dual task (serial-3 while
cycling (Z = 0.48, p = 0.31) (Fig. 1).
Table 1 Descriptive characteristic by groups
Stroke N = 13 Control N = 15 p value
Gender
Male 10 9 0.33
Female 3 6
Age (years) 60 [42–71] 63 [42–76] 0.46
Education (years) 13 [10–22] 18 [13–25] 0.001
MoCA 26 [18–30] 27 [24–30] 0.02
Serial three (number)
Single task 12 [4–24] 22 [13–32] < 0.001
Dual task 12 [4–23] 28 [9–36] < 0.001
p valuea
0.98 0.002
Mistakes (number)
Single task 1 [0–9] 0 [0–1] 0.004
Dual task 1 [0–5] 0 [0–1] 0.03
Mistakes (ratio in percentage)
Single task 10.8 [0–75] 0 [0–6] 0.006
Dual task 6.8 [0–43] 0 [0–21] 0.09
Values in table are numbers, median [min–max], p value based on χ2
,
Mann–Whitney
a
p value based on Wilcoxon test
Neurol Sci (2017) 38:2037–2043 2039

4. Post-stroke patients showed non-significant differences in
RMSSD values when comparing the different study condi-
tions (Friedman test: χ2
4 = 5.2, p value 0.14). In contrast,
significant differences were noted in the control group
(Friedman test χ2
4 = 19.2, p value < 0.001). In this group,
post hoc analysis using the Wilcoxon signed-rank test showed
significant differences in RMSSD values between rest and
paced breathing condition (Z = 2.72, p = 0.003) and between
rest and dual task (serial-3 during cycling) condition (Z = 1.93,
p = 0.02) (Fig. 2).
Computation of Spearman correlation coefficients between
MoCA scores and serial-3 task performance revealed further
differences between stroke patients and healthy controls
(Table 2). In the stroke group, MoCA scores correlated nega-
tively with the number of errors performed both in single
(serial-3 at rest) and dual (serial-3 while cycling) task condi-
tions. However, the correlation between the MoCA scores and
the total number of subtraction steps done in 1 min did not
reach significance. In the control group, the negative correla-
tion between MoCA scores and number of errors done in the
serial-3 task (both single and dual task conditions) did not
reach significance at the 0.05 level. However, in this group
the MoCA scores correlated significantly with the total num-
ber of subtraction steps done in 1 min. In both groups, perfor-
mance in the single task (serial-3 at rest) correlated with per-
formance in the dual task (serial-3 while cycling) (see
Table 2).
SDNN values did not correlate with performance level in
the cognitive tests, neither in the patient group nor in the
control group. Among healthy controls, a significant correla-
tion was noted between RMSSD values at rest and cognitive
performance as reflected in the number of steps reached in
1 min in the serial-3 subtraction task: single task (serial-3 at
rest) rs = 0.41, p < 0.05; dual task (serial-3 while cycling)
rs = .44, p < 0.05. (Fig. 3). An association was found also with
the MoCA score (Spearman correlation = 0.40, p < 0.05). In
contrast, RMSSD values did not correlate with cognitive task
performance in the stroke group (Fig. 3). Using the Fisher r-to-
z transformation to assess the difference between the two
groups in the correlation between RMSSD and cognitive per-
formance, a significant difference was noted during the single
task condition (Z = 2.07, p = 0.02).
Rest
Grip
Slow breathing
Serial-3 at rest
Serial-3 while cycling
ms
Fig. 1 Box plots of SDNN values
at the different testing conditions.
Stars represent statistically
significant differences between
line-connected measures (be-
tween groups above and within
group below the box plot)
ms
Rest
Grip
Slow breathing
Serial-3 at rest
Serial-3 while cycling
Fig. 2 Box plots of RMSSD
values at the different testing
conditions. Stars represent
statistically significant differences
between line-connected measures
(between groups above and with-
in group below the box plot)
2040 Neurol Sci (2017) 38:2037–2043

5. Discussion
The aim of the current exploratory pilot study was to assess the
association between HRV, as a measure of autonomic nervous
system functioning, and performance level in cognitive tasks
(MoCA and timed serial-3 subtraction), in patients post ischemic
stroke, as compared to healthy age-matched controls. The results
of the study point to several important differences between the
two groups. First, patients post stroke were found to have lower
MoCA as well as serial-3 subtraction test scores, with a greater
error rate in the latter task. Second, patients post stroke had lower
HRV compared to the control group, both at rest and during
cognitive task performance in different conditions. Third, among
patients post stroke, the HRV was generally less sensitive to
changes in ongoing task, and both SDNN and RMSSD param-
eters remained relatively stable along the different testing
Fig. 3 Scatter plot of RMSSD
values at rest (x-axis) vs.
cognitive task performance (y-
axis) presented separately for
each group. Black dots = single
task (serial-3 at rest); white
dots = dual task (serial-3 while
cycling). The lines represent the
linear trend (straight line = single
task; dash line = dual task)
Table 2 Spearman correlation coefficients between cognitive test achievements
DT
errors
DT
number
ST
errors
ST
number
MoCA
Control
Stroke
-.50^
.71*
*
-.42^
.70*
MoCA
-.48^
.80**
-.10
.15
Single task number
.73**
-.21
-.37
-.88**
Single task errors
-.52*
-.49*
.72**
.24
Dual task number
-.16
.64*
-0.22
-.62**
Dual task errors
SD single task (timed serial-3 at rest), DT dual task (timed serial-3 while cycling)
^p < 0.1; *p < 0.05; **p < 0.01
Neurol Sci (2017) 38:2037–2043 2041

6. conditions. In contrast, healthy controls exhibited significant al-
teration of the HRV when passing from rest to more active motor
and cognitive task performance. Finally, in stroke patients, the
baseline HRV did not correlate with performance level in the
cognitive tasks. In contrast, healthy subjects showed a significant
positive correlation between RMSSD values at rest and perfor-
mance level in the serial-3 subtraction task.
Our findings of lower achievements in cognitive test perfor-
mance in patients post stroke are in accord with previous find-
ings on this topic. Up to two thirds of stroke patients have some
cognitive impairment when tested shortly after the onset of the
disease, and more than 50% of stroke survivors have cognitive
impairments at 1 year post onset [23]. In addition, our demon-
stration of reduced HRV post stroke, in particular the
hypofunction of the parasympathetic branch of the autonomic
nervous system, has been documented before [11]. According
to several earlier studies, up to 70% the patients post stroke
show some autonomic nervous system dysfunction, a fact that
often fails to receive the appropriate consideration [24, 25].
To our knowledge, the association between post-stroke im-
pairment in autonomic nervous system function and post-stroke
impairment in cognitive functioning has not yet been investi-
gated. In previous studies with other populations—healthy in-
dividuals [10], people suffering from mild cognitive impair-
ment who show white matter changes in brain imaging [26],
and patients with Alzheimer’s disease [27]—an association was
found between cardiac autonomic parameters and cognitive
performance, especially in the domain of executive function.
In the current study, the RMSSD measured at rest was found to
correlate with performance level in the serial-3 subtraction task.
However, this association was noted only in the healthy control
subjects and not in the stroke group. The significant correlation
shown by the control subjects (higher RMSSD associated with
higher cognitive performance) may point to the importance of
keeping good cardio-vascular functioning in order to maintain
adequate cognitive functioning. This might be especially true in
the adult population (the mean age of the control group in the
current study was 63 years). The stroke subjects differed from
their age-matched controls in average level of cognitive func-
tioning as well as in HRV (both significantly lower in the stroke
group). Yet, the nature of the relationship between these two
stroke-related deviations from normality remains unclear. In
particular, a causal relationship cannot be inferred, as the cor-
relation between cardiac autonomic function (RMSSD at rest)
and cognitive function (serial-3 subtraction test performance)
shown by healthy subjects was not revealed in the stroke group.
The lack of significant correlation might reflect just the statisti-
cal outcome of analyzing a small study group with high inter-
personal variance. Alternatively, it might point to a biological
reality where as a consequence of brain damage the functioning
of the autonomic nervous system becomes more segregated,
losing to some extent the normal inter-relationship with ongo-
ing processing done in other functional systems of the brain.
This possibility is hinted by another difference between the
healthy subjects and the patients post stroke—whereas in the
former group both SDNN (Fig. 1) and RMSSD (Fig. 2) were
sensitive to changes in the experimental conditions, no such
sensitivity was shown by the stroke patients.
Autonomic nervous system function post stroke is generally
characterized by reduced activity of the para-sympathetic
branch and increased activity of the sympathetic branch of
the system [11]. Reduced para-sympathetic activation might
contribute to stroke-related cognitive malfunction. It should
be noted that in the current study, cognitive performance was
found to correlate in healthy subjects with the RMSSD param-
eter of the HRV and not with the SDNN parameter. This find-
ing suggests that short-term variations in the heart rate, which
are mediated mainly by the parasympathetic branch of the
autonomic nervous system, are more closely associated with
cognitive functioning than the SDNN, an indicator of long-
term variations in heart rate, reflecting the combined activity
of both the sympathetic and parasympathetic branches of the
system. Previous research also showed the RMSSD to be more
closely associated with cognitive functioning than the SDNN
[4]. Thus, our finding of significantly lower RMSSD values in
the stroke group compared to the control group is likely to
imply a negative contribution of the impaired autonomic reg-
ulation to the overall cognitive function following stroke. Lack
of correlation between RMSSD and cognitive performance in
the stroke group seems to reflect the added principal effects of
damages to the cortical mantle, which differ from patient to
patient in accord with lesion location and extent. This source of
variance in cognitive function does not exist in normal sub-
jects, enabling the load on cognition of autonomic nervous
system health to emerge in a more salient manner.
Beyond the small number of subjects, the current study has
some limitations that should be noted. Breathing rate, which is an
important contributor to HRV, was not controlled or measured,
although subjects were asked to breathe at a Bcomfortable^ rate.
Further, the matched controls were Bhealthy^ in as much as
elderly subjects with a negative history of stroke or cardiovascu-
lar disease are considered to be healthy. Possible effects of med-
ications taken by these subjects for other problems were not
controlled. Variance in cognitive performance of stroke patients,
introduced by differences in lesion characteristics, was also not
controlled, due to the small number of participants.
In summary, the findings of the current exploratory pilot
study may point to two different aspects of autonomic nervous
system dysfunction post stroke: (1) reduced HRV (as reflected
by lower than normal SDNN and RMSSD values) both at rest
and during activity and (2) reduced inter-connectedness be-
tween the autonomic nervous system and ongoing processing
in other functional systems of the brain (as reflected by lack of
the normal sensitivity of HRV parameters to changes in the
experimental conditions). Given the small number of stroke
participants in the current study and the large inherent
2042 Neurol Sci (2017) 38:2037–2043

7. variance stemming from differences in lesion characteristics,
the relationship between the demonstrated HRV abnormality
and the cognitive impairments of stroke survivors remains
uncertain. The role of rehabilitation training addressing spe-
cifically the impaired function of the cardiac autonomous sys-
tem, and its possible role in amelioration of concurrent cogni-
tive and motor impairments, awaits further research.
Compliance with ethical standards
Ethical approval was obtained from the Human Research Ethics
Committee of both hospitals and the Health Ministry of Israel. All partic-
ipants signed informed consent forms prior the study.
Conflict of interest All authors declare that they have no conflict of
interest.
References
1. Electrophysiology (1996) TFotESoCatNASoPa. Heart rate vari-
ability: standards of measurement, physiological interpretation,
and clinical use. Circulation 93:1043–1065
2. Ziegler D, Zentai CP, Perz S et al (2008) Prediction of mortality
using measures of cardiac autonomic dysfunction in the diabetic
and nondiabetic population: the MONICA/KORA Augsburg
Cohort Study. Diabetes Care 31(3):556–561
3. Huikuri HV, Ylitalo A, Pikkujämsä SM et al (1996) Heart rate
variability in systemic hypertension. Am J Cardiol 77(12):1073–
1077
4. Carnethon MR, Golden SH, Folsom AR, Haskell W, Liao D (2003)
Prospective investigation of autonomic nervous system function
and the development of type 2 diabetes. The Atherosclerosis Risk
In Communities Study, 1987–1998. Circulation 107(17):2190–
2195
5. Nahshoni E, Aravot D, Aizenberg D et al (2004) Heart rate vari-
ability in patients with major depression. Psychosomatics 45(2):
129–134
6. Sajadieh A, Nielsen OW, Rasmussen V, Hein HO, Abedini S,
Hansen JF (2004) Increased heart rate and reduced heart-rate vari-
ability are associated with subclinical inflammation in middle-aged
and elderly subjects with no apparent heart disease. Eur Heart J
25(5):363–370
7. Hansen AL, Johnsen BH, Thayer JF (2003) Vagal influence on
working memory and attention. Int J Psychophysiol 48:263–274
8. de Vilhena Toledo MA, Junqueira LF Jr (2010) Cardiac autonomic
modulation and cognitive status in Alzheimer’s disease. Clin Auton
Res 20(1):11–17
9. Di Bernardi LC, Takase E, Darby D (2009) Heart rate variability
and cognitive function: effects of physical effort. Biol Psychol 82:
186–191
10. Mukherjee S, Yadav R, Yung I, Zajdel DP, Oken BS (2011)
Sensitivity to mental effort and test–retest reliability of heart rate
variability measures in healthy seniors. Clin Neurophysiol 122(10):
2059–2066
11. Korpelainen JT, Sotaniemi KA, Myllylä VV (1993) Asymmetric
sweating in stroke: a prospective quantitative study of patients with
hemispheral brain infarction. NeuroImage 43:1121–1214
12. Katz-Leurer M, Shochina M (2005) Heart rate variability(HRV)
parameters correlate with motor impairments and aerobic capacity
in stroke patients. Neurorehabilitation 20:91–95
13. Zinn S, Bosworth HB, Hoenig HM, Swartzwelder HS (2007)
Executive function deficits in acute stroke. Arch Phys Med
Rehabil 88(2):173–180
14. Francica JV, Bigongiari A, Mochizuki L et al (2015) Cardiac auto-
nomic dysfunction in chronic stroke women is attenuated after sub-
maximal exercise test, as evaluated by linear and nonlinear analysis.
BMC Cardiovasc Disord 15(1):105
15. Gamelin FX, Berthoin S, Bosquet L (2006) Validity of the polar
S810 heart rate monitor to measure R-R intervals at rest. Med Sci
Sports Exerc 38(5):887–893
16. Allen JJ, Chambers AS, Towers DN (2007) The many metrics of
cardiac chronotropy: a pragmatic primer and a brief comparison of
metrics. Biol Psychol 74(2):243–262
17. Ewing D, Borsey D, Bellavere F, Clarke B (1981) Cardiac auto-
nomic neuropathy in diabetes: comparison of measures of RR in-
terval variation. Diabetologia 21(1):18–24
18. Nasreddine ZS, Phillips NA, Bédirian Vet al (2005) The Montreal
Cognitive Assessment, MoCA: a brief screening tool for mild cog-
nitive impairment. J Am Geriatr Soc 53(4):695–699
19. Srygley JM, Mirelman A, Herman T, Giladi N, Hausdorff JM
(2008) When does walking alter thinking? Age and task associated
findings. Brain Res 1253:92–99
20. Manor B, Zhou J, Jor'dan A, Zhang J, Fang J, Pascual-Leone A
(2016) Reduction of dual-task costs by noninvasive modulation of
prefrontal activity in healthy elders. J Cogn Neurosci 28(2):275–
281
21. Wolfe CD, Taub NA, Woodrow E, Burney P (1991) Assessment of
scales of disability and handicap for stroke patients. Stroke 22(10):
1242–1244
22. Wilkinson PR, Wolfe CD, Warburton FG et al (1997) Longer term
quality of life and outcome in stroke patients: is the Barthel index
alone an adequate measure of outcome? Qual Health Care 6(3):
125–130
23. Blackburn DJ, Bafadhel L, Randall M, Harkness KA (2013)
Cognitive screening in the acute stroke setting. Age Ageing
42(1):113–116
24. Bassi A, Colivicchi F, Santini M, Caltagirone C (2007) Cardiac
autonomic dysfunction and functional outcome after ischemic
stroke. Eur J Neurol 14:917–922
25. Naver HK, Blomstrand C, Wallin BG (1996) Reduced heart rate
variability after right-sided stroke. Stroke 27:247–251
26. Galluzzi S, Nicosia F, Geroldi C et al (2009) Cardiac autonomic
dysfunction is associated with white matter lesions in patients with
mild cognitive impairment. J Gerontol Ser A Biol Med Sci 64(12):
1312–1315
27. Zulli R, Nicosia F, Borroni B et al (2005) QT dispersion and heart
rate variability abnormalities in Alzheimer's disease and in mild
cognitive impairment. J Am Geriatr Soc 53(12):2135–2139
Neurol Sci (2017) 38:2037–2043 2043