Eye-Movement Training Results in Changes in qEEG and NIH Stroke Scale in Subjects Suffering from Acute Middle Cerebral Artery Ischemic Stroke: A Randomized Control Trial

Abstract

Context:
Eye-movement training (EMT) can induce altered brain activation and change the functionality of saccades with changes of the brain in general.

Objective:
To determine if EMT would result in changes in quantitative electroencephalogram (qEEG) and NIH Stroke Scale (NIHSS) in patients suffering from acute middle cerebral artery (MCA) infarction. Our hypothesis is that there would be positive changes in qEEG and NIHSS after EMT in patients suffering from acute MCA ischemic stroke.

Design:
Double-blind randomized controlled trial.

Setting and participants:
Thirty-four subjects with acute MCA ischemic stroke treated at university affiliated hospital intensive care unit.

Interventions:
Subjects were randomized into a "control" group treated only with aspirin (125 mg/day) and a "treatment" group treated with aspirin (125 mg/day) and a subject-specific EMT.

Main outcome measures:
Delta-alpha ratio, power ratio index, and the brain symmetry index calculated by qEEG and NIHSS.

Results:
There was strong statistical and substantive significant improvement in all outcome measures for the group of stroke patients undergoing EMT. Such improvement was not observed for the "control" group, and there were no adverse effects.

Conclusion:
The addition of EMT to a MCA ischemic stroke treatment paradigm has demonstrated statistically significant changes in outcome measures and is a low cost, safe, and effective complement to standard treatment.

Full text

January 2016 | Volume 7 | Article 31
ORIGINAL RESEARCH
published: 22 January 2016
doi: 10.3389/fneur.2016.00003
Frontiers in Neurology | www.frontiersin.org
Edited by:
Owen B. White,
Royal Melbourne Hospital and
The University of Melbourne, Australia
Reviewed by:
Bernard Yan,
Royal Melbourne Hospital and
The University of Melbourne, Australia
Mary Pauline Galea,
The University of Melbourne, Australia
*Correspondence:
Frederick Robert Carrick
drfrcarrick@post.harvard.edu
Specialty section:
This article was submitted to
Neuro-Ophthalmology,
a section of the journal
Frontiers in Neurology
Received: 03December2015
Accepted: 08January2016
Published: 22January2016
Citation:
CarrickFR, OggeroE, PagnaccoG,
WrightCHG, MachadoC, EstradaG,
PandoA, CossioJC and BeltránC
(2016) Eye-Movement Training
Results in Changes in qEEG and NIH
Stroke Scale in Subjects Suffering
from Acute Middle Cerebral Artery
Ischemic Stroke: A Randomized
Control Trial.
Front. Neurol. 7:3.
doi: 10.3389/fneur.2016.00003
Eye-Movement Training Results in
Changes in qEEG and NIH Stroke
Scale in Subjects Suffering from
Acute Middle Cerebral Artery
Ischemic Stroke: A Randomized
Control Trial
Frederick Robert Carrick1,2,3,4* , Elena Oggero1,5 , Guido Pagnacco1,5 ,
Cameron H. G. Wright1,5 , Calixto Machado1,3 , Genco Estrada3 , Alejandro Pando3 ,
Juan C. Cossio3 and Carlos Beltrán3
1 Neurology, Carrick Institute, Cape Canaveral, FL, USA, 2 Global Clinical Scholars Research Training Program (GCSRT),
Harvard Medical School, Boston, MA, USA, 3 Institute of Neurology and Neurosurgery, Havana, Cuba, 4 Bedfordshire Centre
for Mental Health Research, University of Cambridge, Cambridge, UK, 5 Electrical and Computer Engineering, University of
Wyoming, Laramie, WY, USA
Context: Eye-movement training (EMT) can induce altered brain activation and change
the functionality of saccades with changes of the brain in general.
Objective: To determine if EMT would result in changes in quantitative electroenceph-
alogram (qEEG) and NIH Stroke Scale (NIHSS) in patients suffering from acute middle
cerebral artery (MCA) infarction. Our hypothesis is that there would be positive changes
in qEEG and NIHSS after EMT in patients suffering from acute MCA ischemic stroke.
Design: Double-blind randomized controlled trial.
Setting and participants: Thirty-four subjects with acute MCA ischemic stroke treated
at university afliated hospital intensive care unit.
Interventions: Subjects were randomized into a “control” group treated only with
aspirin (125mg/day) and a “treatment” group treated with aspirin (125mg/day) and a
subject-specic EMT.
Main outcome measures: Delta–alpha ratio, power ratio index, and the brain symme-
try index calculated by qEEG and NIHSS.
Results: There was strong statistical and substantive signicant improvement in all
outcome measures for the group of stroke patients undergoing EMT. Such improvement
was not observed for the “control” group, and there were no adverse effects.
Conclusion: The addition of EMT to a MCA ischemic stroke treatment paradigm has
demonstrated statistically signicant changes in outcome measures and is a low cost,
safe, and effective complement to standard treatment.
Keywords: ischemic stroke, qEEG, NIHSS scores, eye moment therapy, stroke rehabilitation, saccades

January 2016 | Volume 7 | Article 32
Carrick et al.
Eye-Movement Therapy in Acute Ischemic Stroke
Frontiers in Neurology | www.frontiersin.org
INTRODUCTION
Stroke is one of the leading causes of death in the United States
and is a major cause of adult disability; although from 2001
to 2011 the relative rate of stroke death fell by 35.1% and the
actual number of stroke deaths declined by 21.2%, the number
of persons suering from a stroke is still signicant (≈795,000
each year in the United States alone) and its consequences are
serious (in 2011, stroke caused ≈1 of every 20 deaths in the
United States) (1). Its etiology is a change in blood ow to a
specic area of the brain due to ischemia or hemorrhage, and it is
usually manifested as brain dysfunction with consequent eects
such as hemiparesis, dysphasia, ataxia, diplopia, or visual eld
loss. Strokes are diagnosed by physical and neurological examina-
tion, with the help of neurological scales specically developed
to quantify the impairment caused by a stroke, in particular the
NIH Stroke Scale (NIHSS). is scale originally consisted of a
15-item examination (2), then amended to an 11-item examina-
tion (3), scored on a scale from 0 to 2, 3, or 4 depending on the
item, for a total score ranging from 0 (normal function) to 42
(severe stroke). Several studies have reported that the baseline
NIHSS (taken at hospitalization/diagnosis time) is a good predic-
tor of outcome aer a stroke (4–7). Diagnostic tools for strokes
include CT scans (with or without contrast), MRI scans (espe-
cially diusion-weighted imaging – DWI, and with magnetic
resonance angiography–MRA), Doppler ultrasound, and digital
subtraction angiography. In particular, for ischemic stroke, MRI
scans have shown a higher sensitivity and specicity than CT
scans without contrast (8). Once patients are hospitalized, elec-
troencephalograms (EEG) are used to continuously monitor their
brain function as well as to drive clinical management, since EEG
abnormalities are typical manifestation of an ischemic stroke. In
particular, quantitative electroencephalogram (qEEG) (9) has
been used for monitoring and formulating prognosis in acute and
sub-acute ischemic stroke (10). Of all the numerical parameters
that can be obtained from the qEEG, of particular interest are
the ratio of mean scalp delta to alpha power [known as the alpha
delta ratio (ADR), or its inverse the delta alpha ratio (DAR)] (11,
12), the power ratio index (PRI) of mean “slow” (delta and theta)
to mean “fast” (alpha and beta) activity (12–14), and the brain
symmetry index (BSI or mBSI) (15, 16).
Standard treatment plans for patients aected by ischemic
stroke involve brinolytic therapy (administration of recom-
binant tissue-type plasminogen activator – rt-PA), antiplatelet
agents (such as aspirin), and mechanical thrombectomy (removal
of the clot causing the blood ow obstruction). Aer the acute
phase is concluded, the most eective rehabilitation programs
involve carefully directed, well-focused, repetitive practice to
relearn skills that are lost when part of the brain is damaged.
Saccades are fast eye movements that allow humans to vol-
untarily very quickly change the direction of gaze. Extensive
studies have been conducted to characterize the dierent brain
and eye mechanisms generating such movements and how
dierent pathologies aect them (17). A number of standard
parameters have been used to characterize saccades: latency or
reaction time (the time it takes for the eyes to start moving once
a stimulus is presented), velocity (at how many deg/s the eyes
move), amplitude (how many degrees the eyes move), and dura-
tion (how much time it takes) (18). All of these eye movements
can be quantied with diagnostic equipment, such as video-
nystagmography (VNG), but they can be observed at the bedside
as well. Standardized objective examination of eye movements
is of great value in the detection and clarication of sub-clinical
lesions in the central nervous system. Even patients with multiple
sclerosis (MS) with lesions beyond the primary visual pathway
have both saccadic latency and smooth pursuit abnormalities
of oculomotor dysfunction (19). Patients suering from mild
closed-head injury also demonstrate prolonged saccadic laten-
cies, and quantitative tests of oculomotor function may provide
sensitive markers of cerebral dysfunction (20) that can assist and
direct patient assessment. For instance, a cerebral vascular lesion
in the right and/or le hemisphere produces a general slowing
in the saccadic latency and a general reduction in the accuracy
of saccades with respect to a healthy subject’s performance (21).
Abnormalities in the control of saccades have been described
in patients with cerebral pathology (22), suggesting that they
might be robust biomarkers that could be utilized in guiding and
interpreting treatment outcomes. Discrepancy in horizontal and
vertical tilt angle coecients can cause eye positions to lie on a
twisted rather than a planar surface, resulting in eye velocities
that change during a visual saccade (23). e coordination of eye
movements is dependent upon the non-linear addition of visual
saccades and the pursuit components of catch-up saccades that
can be measured to assess function and disability (24). ere
are many variables that can result in dierent clinical scenarios
for patients with similar disease states or injuries. For example,
elderly patients demonstrate an increased latency and decreased
peak velocity from age-related degenerative changes in the central
nervous system with diseases of the central nervous system oen
causing saccadic disorders (25). Dierent disease states and sites
of neurological injury may aect one component of a visual task
while not aecting another. Alzheimer’s patients show increased
latency to initiation of saccades but no dierence in their ampli-
tude and velocity when compared to healthy controls (26). We
have observed slowing of visual saccades and saccadic intrusions
of visual pursuits in patients with acute middle cerebral artery
(MCA) infarction. Abnormal saccadic intrusions consisting of
frequent sporadic horizontal square wave jerks occur in a large
percentage of patients with acute or chronic focal cerebral lesions
(27). Low-amplitude cerebral square wave jerks can be detected
clinically by fundoscopy at the bedside. Reexive visually guided
saccade triggering may be facilitated or inhibited by the cerebral
cortex. Pierrot-Deseilligny and colleagues observed pathology
of saccades made toward and away from suddenly appearing
visual targets in patients with limited unilateral cerebral infarc-
tion (28). Dierent phenomenology of eye movements have
been observed with lesions of both the right and le cerebral
hemisphers. For example, ischemic lesions of the le frontal
eye eld (FEF) have been associated with abnormal reexive
visually guided saccades (gap and overlap tasks), antisaccades,
predictive saccades, memory-guided saccades, smooth pursuit,
and optokinetic nystagmus (29). Eye-movement analysis not only

January 2016 | Volume 7 | Article 33
Carrick et al.
Eye-Movement Therapy in Acute Ischemic Stroke
Frontiers in Neurology | www.frontiersin.org
identies functional lesions but can also act as a biomarker for
treatment outcomes. Hemispatial neglect aects the ability to
explore space on the side opposite a brain lesion that is also mir-
rored in abnormal saccadic eye-movement patterns that provide
a sensitive means to assess the extent of neglect recovery (30).
Russell and colleagues provided the rst evidence for a decit
in remapping visual information across saccades underlying
right-hemisphere constructional apraxia (RHCA) (31). RHCA is
a common disorder aer right parietal stroke, oen persisting
aer initial problems such as visuospatial neglect have resolved.
Concurrent saccade programing is bilaterally impaired with
extensive right cerebral damage with an inability to produce
a corrective saccade within 100ms aer the end of a primary
saccade (32). Visual eld defects aer striate lesions are associ-
ated with changes in the frontoparietal network underlying the
cortical control of saccades, but may improve search strategies
with appropriate training of saccades (33). Nelles and colleagues
used functional magnetic resonance imaging (fMRI) to study the
eects of eye-movement training (EMT) on cortical control of
saccades (34). EMT induced altered brain activation in the striate
and extrastriate cortex as well as in oculomotor areas and a rela-
tive decrease of activation in the le FEF. e cerebellum plays a
major role in saccadic adaptation representing a well-established
model of sensory–motor plasticity (35). e cerebellum remains
intact aer MCA infarction, while the intraparietal sulcus may be
the neural substrate for remapping of the visual environment by
saccadic training (36). But saccade training may not be enough
in EMT as repetitive contralesional smooth visual pursuit train-
ing has been shown to induce superior, multimodal therapeutic
eects in mild and severe chronic stroke patients with neglect
syndrome (37).
Exploratory ndings suggest that measurements of saccades,
smooth pursuit, and vergence are useful in detecting changes
associated with mild traumatic brain injuries (38), and it is
reasonable to utilize them in other brain syndromes, including
stroke. EMT has been used with vestibular rehabilitation in the
successful treatment of Post-Traumatic Stress Disorder (PTSD)
in combat veterans aer traumatic brain injury (39–41). Dong
and colleagues evaluated the sensitivity of measuring cognitive
processing in the ocular motor system as a marker for recovery
of decit in post-stroke patients (42). ey tested ocular motor
function and compared outcomes in the NIHSS score, modied
Rankin Scale (mRS), and standard cognitive function assess-
ments. Ocular motor function was more sensitive in identifying
cognitive dysfunction and improvement compared with NIHSS or
mRS. ey concluded that ocular motor assessment demonstrates
cognitive eects of even mild stroke and may provide improved
quantiable measurements of cognitive recovery post-stroke. We
desired to see if EMT might be benecial in the treatment of acute
MCA infarction and hypothesized that it would result in positive
changes of qEEG and NIHSS.
MATERIALS AND METHODS
is study was a single-center, double-blind, randomized
controlled clinical trial performed at our Institutional Hospital
Intensive Care Unit and conducted in accordance with the
Declaration of Helsinki with equipoise. e protocol was
approved by the ethics committee of our Institution. Written
informed consent was obtained from every potential par-
ticipant prior to randomization. e eect of traditional stroke
therapy (aspirin regimen) combined with a subject-specic
EMT was investigated in subjects aected by MCA stroke, and
its outcome compared with a “control” group consisting of
subjects aected by the same pathology and receiving only the
aspirin regimen. We utilized the DAR, PRI, and BSI calculated
by qEEG, and NIHSS as outcome measures of intervention.
Participants
Subjects were recruited from patients with acute MCA ischemic
stroke admitted to our intensive care unit. Patients with a pre-
sumptive diagnosis of acute MCA ischemic stroke were screened
within 48h following stroke onset. Investigators veried eligibil-
ity and obtained written informed consent before randomization
to two groups.
Sample Size
e planned sample of 17 subjects in each treatment group was
calculated to give the study 80% power to detect a 30% reduction
in NIHSS at a 0.05 signicance level for a two-sided test. e
calculations assumed that 20% of participants would be lost to
follow-up or non-compliant or would die of other causes.
Inclusion Criteria
Non-disabling ischemic MCA stroke (mRS ≤3):
• Onset within 48h before randomization.
• No previous history of cerebral strokes and functionally inde-
pendent (mRS of 0 or 1) pre-morbidity.
• Focal neurological decit of likely atherothrombotic origin
classied as ischemic stroke by questionnaire/algorithm and
conrmed as new cerebral infarction consistent with symp-
toms by cranial computed tomography and brain magnetic
resonance imaging.
• Age >39years.
• Agreement to participate in this study.
• Written informed consent.
Exclusion Criteria
• A previous history of cerebral stroke.
• Potential sources of emboli (atrial brillation within 30days
of stroke, prosthetic cardiac valve, intracardiac thrombus or
neoplasm, or valvular vegetation).
• Other major neurological illness that would obscure evalua-
tion of recurrent stroke.
• Refractory depression, severe cognitive impairment, alcohol-
ism or other substance abuse.
• General anesthesia or hospital stay of ≥3days, any type of
invasive cardiac instrumentation, or endarterectomy, stent
placement, thrombectomy, or any other endovascular treat-
ment of carotid artery within 30days prior to admission to
intensive care unit or scheduled to be performed.

January 2016 | Volume 7 | Article 34
Carrick et al.
Eye-Movement Therapy in Acute Ischemic Stroke
Frontiers in Neurology | www.frontiersin.org
Randomization, Intervention, and
Follow-up
Fiy-seven subjects who had symptoms and signs of acute MCA
ischemic stroke within the rst 48h of clinical evolution were
admitted to our intensive care unit. Of these, 34 subjects (age
43–83years old) met the inclusion criteria and were enrolled in
this study. ey had no previous history of cerebral strokes and
were functionally independent (mRS of 0 or 1) pre-morbidity.
e subject design was reviewed with subjects and/or their fami-
lies who were oered a place in the study once informed consent
was obtained. Each subject underwent a CT scan study when
admitted to exclude hemorrhagic strokes, and a second CT study
was performed aer 72 h of stroke onset. e MCA ischemic
stroke was diagnosed according to clinical history, neurological,
and imaging exam. All participants consented to be admitted to
the study and then were randomly assigned to “treatment” and
“control” groups. e allocation of participants was programed
by the statistical coordinating center, encrypted, and entered
into a data entry program installed on a study computer at our
institution. Aer computer verication that all eligibility criteria
had been met, participants were treated according to their groups
with both groups receiving standard medical care and support
in the intensive care unit. e “control” group (11 males and
6 females, age 58 ± 10.7 years) was treated only with aspirin
(125mg/day). e “treatment” group (13 males and 4 females,
age 61.9 ± 11.67 years) received the same aspirin regimen as
well as subject-specic EMT. Figure1 illustrates the CONSORT
diagram showing the ow of participants.
Treatment Group Intervention
Ipsilateral saccades are generated by the contralateral cerebral
cortex and we desired to utilize extraocular muscle targets that
are associated with cerebellar reexogenic activation in the plane
of the anterior and posterior canals. is strategy has been used
with success in the investigation of the treatment of PTSD in
combat veterans (39, 40). We prepared video targeting of the
exercises on Apple iPADs using Apple Keynote soware with
a 2 mm red circular ball target on a blue screen background.
Subjects with le MCA ischemic stroke performed diagonal
saccades to appearing targets (with a xation duration of 2 s
each) using gap paradigm from the lower le to the upper right
corner of the tablet monitor, followed by smooth pursuit of the
target from the upper right corner to the lower le. e saccadic
EMT activates the combination of right superior rectus and le
inferior oblique muscles that have reexogenic connections to
the right cerebellum. Subjects with right MCA ischemic stroke
performed the same type of EMT along the opposite diagonal of
the display (lower right to upper le), activating the combination
of le superior rectus and right inferior oblique muscles that have
reexogenic connections to the le cerebellum. is was followed
by smooth pursuit of the target from the upper le corner to the
lower right. Each treatment session consisted of three repetitions
of the saccades/smooth pursuit sequence, and subjects received
three such treatments a day. Each repetition took ~3min with
the entire intervention session taking <15min with short breaks
between repetitions.
Intervention Both Groups
All subjects underwent EEG testing upon admission and 7days
aerward. Using a Medicid-05 (I. C. NEURONIC S.L., Zaragoza,
Spain), with a gain of 20,000, sampling frequency of 200Hz, lter
band pass of 0.3–30Hz with a “notch” lter at 60Hz. e noise
level of the EEG recording was 2μV RMS and the recordings were
performed at an environmental temperature of ~23°C. Copper
electrodes coated with silver chloride were placed on the scalp at
19 monopolar derivations of the International 10/20 System with
linked ear lobes as a reference. Electrode-skin impedance was
<10kΩ. Total time of EEG data collection persession was 330s.
Patients were recumbent, awake, and relaxed. For each subject, 24
artifact free segments of 2.56s duration were visually selected by
an expert electroencephalographer and used for the subsequent
standard qEEG analysis [power spectrum in the delta (<4Hz),
theta (4–7Hz), alpha (7–14Hz), and beta (14–30Hz) frequency
bands] using the Neuronic EEG 6.0 soware (I. C. NEURONIC
S.L., Zaragoza, Spain). A custom script in MATLAB® (e
MathWorks, Inc., Natick, MA, USA) was used to calculate the
DAR, the PRI, and the BSI that were used as outcome measures
for all subjects. e NIHSS was also administered to all subjects
upon admission and 7days aerward.
Statistical Analysis
e statistical analysis of the outcome measures was performed
using IBM® SPSS® Statistics release 20.0.0 (IBM Corporation,
Armonk, NY, USA) on the pre–post qEEG measures for both
“treatment” and “control” groups and on the pre–post NIHSS
scores. e normality of the distributions of the data was
veried using Kolmogorov–Smirnov with Lilliefors Signicance
Correction and Shapiro–Wilk tests of normality. Since these data
were found to be normally distributed, Multivariate General
Linear Model (M-GLM) analysis was performed to assess the
presence of dierences between the two groups in the pre-
treatment data, i.e., to verify if the two groups were dierent to
begin with. e existence of a dierence in pre–post changes
between the “treatment” and “control” groups was investigated
by performing a Multivariate Repeated Measures General Linear
Model (M-RM-GLM), with repeated measures being the pre and
post measures and the factor being the treatment modality. e
same M-RM-GLM was performed separately on the two groups
to verify if the two dierent treatment modalities were able to
produce statistically signicant changes in the outcome measures.
RESULTS
e descriptive statistics for the pre and post outcome measures
as well as for their paired pre–post changes for the “treatment”
and “control” groups are reported in Table1. Table2 reports the
results of the statistical analyses performed on the data: to quan-
tify the presence of statistically signicant dierences between
the two groups in the pre-treatment data, and in the pre–post
results between and within groups. eir signicance (p value)
and eect size (calculated as partial eta squared) are also reported
in the same table. A partial eta squared of 0.02 is considered a
small eect, 0.13 a medium eect and 0.26 a large eect. Figure2
depicts the pre and post DAR and BSI of “treatment” and “control”

TABLE 1 | Mean, its 95% condence interval (CI), and Standard deviation (SD) of the Pre, Post, and Pre–Post Change for the NIHSS, DAR, PRI, and BSI
measures of “treatment” (subjects receiving EMT therapy in conjunction with the standard aspirin regimen) and “control” (subjects receiving only the
standard aspirin regiment) groups.
Measure Group Pre Post Pre–post change
Mean (CI) SD Mean (CI) SD Mean (CI) SD
NIHSS Treatment 2.82 (1.38: 4.26) 3.03 1.44 (0.67: 2.21) 1.63 −0.81 (−1.34: −0.28) 1.11
Control 2.29 (1.66: 2.92) 1.33 1.86 (1.41: 2.31) 0.95 −0.43 (−0.67: −0.18) 0.51
DAR Treatment 1.77 (1.10: 2.44) 1.40 1.40 (0.81: 1.99) 1.24 −0.37 (−0.67: −0.0.07) 0.63
Control 2.48 (2.19: 2.77) 0.60 2.76 (2.46: 3.06) 0.64 0.28 (0.02: 0.54) 0.55
PRI Treatment 2.39 (1.60: 3.18) 1.66 2.17 (1.30: 3.04) 1.82 −0.22 (−0.51: 0.07) 0.61
Control 3.32 (3.09: 3.55) 0.48 3.69 (3.30: 4.08) 0.81 0.37 (−0.04: 0.78) 0.87
BSI Treatment 0.27 (0.22: 0.32) 0.11 0.33 (0.27: 0.39) 0.13 0.06 (0.03: 0.09) 0.06
Control 0.23 (0.22: 0.24) 0.02 0.23 (0.22: 0.24) 0.03 0.00 (−0.02: 0.02) 0.04
TABLE 2 | Results of the statistical analyses performed on the data,
including the question under examination, each considered parameter,
its signicance (p value) and the effect size (calculated as partial eta
squared).
Statistical question Measure Signicance
(p value)
Effect size
(partial eta
squared)
Are the two groups
signicantly different
pre-treatment?
Multivariate 0.305 0.164
NIHSS # #
DAR # #
PRI # #
BSI # #
Are the pre–post treatment
changes signicantly different
between the two groups?
Multivariate 0.004*** 0.402
NIHSS 0.162 0.066
DAR 0.003*** 0.243
PRI 0.029 0.141
BSI 0.001*** 0.279
Are the pre–post treatment
changes in the “control” group
signicant?
Multivariate 0.011* 0.699
NIHSS 0.008** 0.429
DAR 0.055 0.212
PRI 0.098 0.162
BSI 0.774 0.005
Are the pre–post treatment
changes in the “treatment”
group signicant?
Multivariate 0.008** 0.631
NIHSS 0.037* 0.243
DAR 0.026* 0.272
PRI 0.158 0.121
BSI 0.000*** 0.550
#Value not calculated because multivariate p did not reached required statistical
signicance (p<0.05).
*Statistical signicance p<0.05.
**Statistical signicance p<0.01.
***Statistical signicance p<0.005.
Bold font means signicant values.
January 2016 | Volume 7 | Article 35
Carrick et al.
Eye-Movement Therapy in Acute Ischemic Stroke
Frontiers in Neurology | www.frontiersin.org
groups. Figure3 depicts the pre and post PRI and NIHSS scores
of “treatment” and “control” groups. Figure4 illustrates using box
plots the changes pre and post in the DAR, PRI, and BSI for the
“treatment” group.
Patient Follow-up Data
We have had no patient follow-up data on this preliminary
study, but have scheduled all subjects for follow-up with repeat
diagnostics at 1year and at yearly times aer the initial long-term
follow-up. We will report our outcomes to long-term follow-up
when they are available.
Efcacy of Treatment
e M-GLM analysis on the initial measurements (“pre”) of the
outcome measurements (rst question in Tab l e 2 ) indicated that
the “treatment” and “control” groups are not dierent to begin
with, with an overall multivariate tests signicance of p=0.305
(observed power=0.341). Aer verifying the sphericity of the data
using Mauchly’s test of sphericity, the M-RM-GLM analysis on the
pre/post measures with the group as a factor (second question in
Table2, conrmed by Figures2 and 3) showed that the dierences
in the changes between the two groups are indeed statistically
signicant. Specically, the multivariate tests showed that the
changes are dierent overall with a p=0.004 and observed power
of 0.922 and the tests of between-subjects eects and parameter
estimates showed that the changes in the DAR, PRI, and BSI are
dierent between the “treatment” and “control” groups with p of
0.003, 0.029, and 0.001 and observed power of 0.875, 0.602, and
0.926 respectively, whereas the NIHSS change is not signicantly
dierent between the two groups (p=0.162). e M-RM-GLM
analysis on the pre and post measures of the “control” group (third
question in Tab l e 2 , conrmed by Figures2 and 3) showed that
there is a statistically signicant dierence in the pre and post
measurements for this group (p=0.011 with observed power of
0.879), but this dierence is produced mostly by the change in
NIHSS, which is the only measure changing signicantly with
p= 0.008 and observed power of 0.823. e same analysis on
the “treatment” group (fourth question in Tab le  2 , conrmed by
Figure4) showed that the pre/post measures are statistically sig-
nicantly dierent for this group: the multivariate tests show that
the measures are dierent overall with a p=0.008 and observed
power of 0.882 and the univariate tests and the tests of within-
subjects contrasts show that NIHSS, DAR, and BSI are dierent
with p of 0.037, 0.026, and 0.000 and an observed power of 0.568,
0.633, and 0.985 respectively, whereas the PRI is not dierent to a
statistical signicance (p=0.158, observed power of 0.286).
DISCUSSION
We did not nd any published studies that investigated outcome
measures in the treatment of acute ischemic stroke using the
NIHSS and electrical brain activity aer EMT. Our results show
that the group of stroke patients undergoing EMT, although not
initially dierent from the “control” group, had a signicant

Assessed for eligibility (n=57)
Excluded (n=23)
Not meeting inclusion criteria (n=23)
Declined to participate (n=0)
Other reasons (n=0)
Analysed (n=17)
Excluded from analysis (n=0)
Lost to follow-up (n=0)
Discontinued intervention (n=0)
Allocated to EMT and standard medical
intervention (n=17)
Received allocated intervention (n=17)
Did not receive allocated intervention (n=0 )
Lost to follow-up (n=0)
Discontinued intervention (n=0)
Allocated to standard intervention (n=17)
Received allocated intervention (n=17)
Did not receive allocated intervention (n=0)
Analysed (n=17)
Excluded from analysis (n=0)
Randomized (n=34)
FIGURE 1 | CONSORT diagram showing the ow of participants.
January 2016 | Volume 7 | Article 36
Carrick et al.
Eye-Movement Therapy in Acute Ischemic Stroke
Frontiers in Neurology | www.frontiersin.org
improvement of the electrical brain activity as measured by
the DAR and BSI qEEG indices. Such improvement was not
observed for the “control” group. Furthermore, the improve-
ments in all the qEEG indices considered, i.e., DAR, PRI, and
BSI, were signicantly larger in the patients treated with EMT
than in the controls. We did not investigate a functional rela-
tionship between qEEG ndings in this pilot study, but other
investigators have considered qEEG as a biomarker for neuro-
logical function. Song and colleagues (43) concluded that qEEG
measures of background rhythm frequency (BRF) and relative
power in the qEEG theta band are potential predictive biomark-
ers for cognitive impairment in patients with cerebral infarcts.
ese biomarkers may be valuable in the early prediction of
cognitive impairment in patients with cerebral infarcts. Our
ndings suggest that EMT might change the qEEG and have the
potential to decrease cognitive impairment in MCA ischemic
stroke patients. Song and colleagues (43) also demonstrated that
the risk hazard of developing cognitive impairment was 14 times
higher for those with low BRF than for those with high BRF
(p<0.001). We have found that EMT increases BRF and perhaps
decreases the risk hazard of developing cognitive impairment.
Schleiger and colleagues (44) also analyzed correlations between
post-stroke qEEG indices and cognition-specic functional
outcome measures. ey reported highly signicant correlations
with cognitive outcomes: frontal DAR (ρ=−0.664, p≤0.001)
and global, relative alpha power (ρ=0.67, p≤0.001). We have
demonstrated that EMT changes these qEEG indices and as a
consequence may have a functional eect specic to cognition-
specic outcomes and clinical decision-making. Other inves-
tigators have utilized electrophysiological measurements to
identify the potential therapeutic eects of various treatments in
acute stroke. For example, Liao and colleagues (45) utilized elec-
trophysiology to evaluate neural and vascular responses of the
rat cortex to peripheral sensory stimulation following ischemic
insult. ey demonstrated neural recovery and the preservation
of neurovascular function as well as an optimal time window
of treatment that might result in minimal infarct volume in
the ischemic hemisphere. Our ndings of qEEG changes aer
EMT have led us to postulate that EMT might also be associ-
ated with neural recovery and better functional outcomes. e
DAR has also been correlated with motor function recovery.
Zhang and colleagues (46) evaluated the temporal alterations of
neural activities using EEG from the acute phase to the chronic
phase, and compared EEG with the degree of post-stroke motor

January 2016 | Volume 7 | Article 37
Carrick et al.
Eye-Movement Therapy in Acute Ischemic Stroke
Frontiers in Neurology | www.frontiersin.org
FIGURE 2 | Pre and post delta–alpha ratio and brain symmetry index of “treatment” and “control” groups.
function recovery in a rat model of focal ischemic stroke. e
DAR was found to have the highest correlation coecients with
the motor function recovery. e statistically and substantively
signicant qEEG changes that we have reported aer EMT
would suggest that our therapy might be of use in the treatment
and rehabilitation of motor function. Our study was specic
to observe whether EMT would result in changes of qEEG
and NIHSS without measuring other functional neurological
changes. Other investigators have used similar technology to
explore the relationship between qEEG global indexes and their
association with functional outcome aer neurorehabilitation
in stroke patients. Leon-Carrion and colleagues (47) found that
qEEG indexes and other clinical variables were correlated with
functional recovery aer neurorehabilitation. ey suggested
that the ratio between delta and alpha may play a signicant role
in predicting and monitoring functional rehabilitation outcome.
We agree, and our ndings that EMT changes the DAR suggest
a functional application in the treatment of stroke along with
other neurorehabilitation tools. We have demonstrated statisti-
cally signicant changes in the NIHSS aer EMT. e NIHSS
oers a reliable approach to capture the true response patterns
that are associated with function, outcome, and mortality post-
stroke (48).
e addition of simple EMT to a patient’s treatment paradigm
has demonstrated statistically signicant changes in outcome
measures and is a low cost, safe, and eective complement
to standard treatment in MCA ischemic stroke. ese results
complement previous studies utilizing EMT discussed in the
introduction to this report.
Limitations
e outcome measures include only the three qEEG parameters
and the NIHSS. e NIHSS is a scale of stroke severity and does
not provide any insight as to functional changes. e study would
have beneted from the inclusion of some functional outcome
related to the rationale, e.g., change in visual tracking, cognitive
and functional testing, etc. Other investigators have found that
the outcomes we have utilized have been associated with func-
tional changes in neurological function. We expect that EMT will
also be associated with functional changes and improvement of
outcomes aer stroke treatment. We intend to address functional
outcome measurements in a new randomized controlled study

January 2016 | Volume 7 | Article 38
Carrick et al.
Eye-Movement Therapy in Acute Ischemic Stroke
Frontiers in Neurology | www.frontiersin.org
FIGURE 3 | Pre and post power ratio index and NIHSS scores of “treatment” and “control” groups.
FIGURE 4 | Box plot comparing pre and post delta–alpha ratio, power
ratio index, and brain symmetry index for the “treatment” group.
as our present investigation is considered a pilot from which to
guide and direct future investigations and did not include other
functional measurements.
AUTHOR CONTRIBUTIONS
FC: designed the study and the eye-movement strategies, wrote
the manuscript, and contributed to the statistical analysis. EO:
contributed to the study design, reviewed and edited the manu-
script, and contributed to the statistical analysis. GP: contributed
to the study design, reviewed and edited the manuscript, and
contributed to the statistical analysis. CW: reviewed and edited
the manuscript and contributed to the statistical analysis. CM:
prepared IRB submissions, patient recruitment, and review of
the manuscript. GE: coordinated subject diagnosis and treat-
ment, and reviewed the manuscript. AP: reviewed and edited the
manuscript and contributed to subject assignment. JC: reviewed
the manuscript and contributed to the data collection and com-
pilation. CB: reviewed the manuscript and contributed to subject
treatment assignments.
FUNDING
We thank the Carrick Institute and Plasticity Brain Centers for
generously funding this study.

January 2016 | Volume 7 | Article 39
Carrick et al.
Eye-Movement Therapy in Acute Ischemic Stroke
Frontiers in Neurology | www.frontiersin.org
REFERENCES
1. Mozzafarian D, Benjamin EJ, Go AS, Arnett DK, Blaha MJ, Cushman M, etal.
Heart disease and stroke statistics–2015 update: a report from the American
Heart Association. Circulation (2015) 131(4):e29–322.
2. Brott T, Adams HP Jr, Olinger CP, Marler JR, Barsan WG, Biller J, etal.
Measurements of acute cerebral infarction: a clinical examination scale. Stroke
(1989) 20:864–70. doi:10.1161/01.STR.20.7.871
3. U.S. Department of Health and Human Services. National Institute of
Neurological Disorders and Stroke (NINDS). NIH Stroke Scale Training DVD
(Version 2.0) [DVD]. Bethesda MD, (2007).
4. Adams HP Jr, Davis PH, Leira EC, Chang KC, Bendixen BH, Clarke WR,
etal. Baseline NIH Stroke Scale score strongly predicts outcome aer stroke:
a report of the Trial of Org 10172 in Acute Stroke Treatment (TOAST).
Neurology (1999) 53:126–31. doi:10.1212/WNL.53.1.126
5. Schlegel D, Kolb SJ, Luciano JM, Tovar JM, Cucchiara BL, Liebeskind DS, etal.
Utility of the NIH Stroke Scale as a predictor of hospital disposition. Stroke
(2003) 34:134–7. doi:10.1161/01.STR.0000048217.44714.02
6. Bang OY, Park HY, Yoon JH, Yeo SH, Kim JW, Lee MA, etal. Predicting the
long-term outcome aer subacute stroke within the middle cerebral artery
territor. J Clin Neurol (2005) 1:148–58. doi:10.3988/jcn.2005.1.2.148
7. Jain A, Houten DV, Sheikh L. Retrospective study on National Institutes of
Health Stroke Scale as a predictor of patient recovery aer stroke. J Cardiovasc
Nurs (2016) 31(1):69–72. doi:10.1097/JCN.0000000000000198
8. Chalela JA, Kidwell CS, Nentwich LM, Luby M, Butman JA, Demchuk AM,
etal. Magnetic resonance imaging and computed tomography in emergency
assessment of patients with suspected acute stroke: a prospective comparison.
Lancet (2007) 369:293–8. doi:10.1016/S0140-6736(07)60151-2
9. Hammond DC. e need for individualization in neurofeedback: hetero-
geneity in QEEG patterns associated with diagnoses and symptoms. Appl
Psychophysiol Biofeedback (2010) 35:31–6. doi:10.1007/s10484-009-9106-1
10. Finnigan S, van Putten MJ. EEG in ischemic stroke: quantitative EEG can
uniquely inform (sub-)acute prognoses and clinical management. Clin
Neurophysiol (2013) 124:10–9. doi:10.1016/j.clinph.2012.07.003
11. Claassen J, Hirsch LJ, Kreiter KT, Du EY, Connolly ES, Emerson RG, et al.
Quantitative continuous EEG for detecting delayed cerebral ischemia in
patients with poor-grade subarachnoid haemorrhage. Clin Neurophysiol
(2004) 115:2699–710. doi:10.1016/j.clinph.2004.06.017
12. Finnigan SP, Walsh M, Rose SE, Chalk JB. Quantitative EEG indices of sub-
acute ischemic stroke correlate with clinical outcomes. Clin Neurophysiol
(2007) 118:2525–32. doi:10.1016/j.clinph.2007.07.021
13. Nagata K, Gross CE, Kindt GW, Geier JM, Adey GR. Topographic
electroencephalographic study with power ratio index mapping in
patients with malignant brain tumors. Neurosurgery (1985) 17:613–9.
doi:10.1227/00006123-198510000-00014
14. Nagata K, Tagawa K, Hiroi S, Shishido F, Uemura K. Electroencephalographic
correlates of blood ow and oxygen metabolism provided by positron emis-
sion tomography in patients with cerebral infarction. Electroencephalogr Clin
Neurophysiol (1989) 72:16–30. doi:10.1016/0013-4694(89)90027-8
15. van Putten MJ, Peters JM, Mulder SM, de Haas JA, Bruijninckx CM, Tavy
DL. A brain symmetry index (BSI) for online EEG monitoring in carotid
endarterectomy. Clin Neurophysiol (2004) 115:1189–94. doi:10.1016/j.
clinph.2003.12.002
16. van Putten MJ. e revised brain symmetry index. Clin Neurophysiol (2007)
118:2362–7. doi:10.1016/j.clinph.2007.07.019
17. Leigh RJ, Zee DS. e Neurology of Eye Movements. 4th ed. New York, NY:
Oxford University Press (2006).
18. Baloh RW, Konrad HR, Sills AW, Honrubia V. e saccade velocity test.
Neurology (1975) 25(11):1071–6.
19. Reulen JP, Sanders EA, Hogenhuis LA. Eye movement disorders in multiple
sclerosis and optic neuritis. Brain (1983) 106(Pt 1):121–40. doi:10.1093/
brain/106.1.121
20. Heitger MH, Anderson TJ, Jones RD, Dalrymple-Alford JC, Frampton CM,
Ardagh MW. Eye movement and visuomotor arm movement decits fol-
lowing mild closed head injury. Brain (2004) 127(Pt 3):575–90. doi:10.1093/
brain/awh066
21. Pitzalis S, Di Russo F. Spatial anisotropy of saccadic latency in normal subjects
and brain-damaged patients. Cortex (2001) 37(4):475–92.
22. Williams IM, Ponsford JL, Gibson KL, Mulhall LE, Curran CA, Abel
LA. Cerebral control of saccades and neuropsychological test results
aer head injury. J Clin Neurosci (1997) 4(2):186–96. doi:10.1016/
S0967-5868(97)90072-2
23. urtell MJ, Joshi AC, Walker MF. ree-dimensional kinematics of
saccadic and pursuit eye movements in humans: relationship between
Donders’ and Listing’s laws. Vision Res (2012) 60:7–15. doi:10.1016/j.
visres.2012.02.012
24. Rahafrooz A, Fallah A, Jafari AH, Bakouie F, Zendehrouh S, Gharibzadeh
S. Saccadic and smooth pursuit eye movements: computational modeling
of a common inhibitory mechanism in brainstem. Neurosci Lett (2008)
448(1):84–9. doi:10.1016/j.neulet.2008.10.019
25. Schik G, Mohr S, Hoerberth B. Eect of aging on saccadic eye movements to
visual and auditory targets. Int Tinnitus J (2000) 6(2):154–9.
26. Bylsma FW, Rasmusson DX, Rebok GW, Keyl PM, Tune L, Brandt J. Changes
in visual xation and saccadic eye movements in Alzheimer’s disease. Int J
Psychophysiol (1995) 19(1):33–40. doi:10.1016/0167-8760(94)00060-R
27. Sharpe JA, Herishanu YO, White OB. Cerebral square wave jerks. Neurology
(1982) 32:57–62. doi:10.1212/WNL.32.1.57
28. Pierrot-Deseilligny C, Rivaud S, Gaymard B, Agid Y. Cortical control of reex-
ive visually-guided saccades. Brain (1991) 114(Pt 3):1473–85. doi:10.1093/
brain/114.3.1473
29. Rivaud S, Muri RM, Gaymard B, Vermersch AI, Pierrot-Deseilligny C. Eye
movement disorders aer frontal eye eld lesions in humans. Exp Brain Res
(1994) 102:110–20. doi:10.1007/BF00232443
30. Olk B, Harvey M, Gilchrist ID. First saccades reveal biases in recovered
neglect. Neurocase (2002) 8:306–13. doi:10.1093/neucas/8.4.313
31. Russell C, Deidda C, Malhotra P, Crinion JT, Merola S, Husain M. A decit
of spatial remapping in constructional apraxia aer right-hemisphere stroke.
Brain (2010) 133:1239–51. doi:10.1093/brain/awq052
32. Ptak R, Schnider A, Muri R. Bilateral impairment of concurrent saccade
programming in hemispatial neglect. Neuropsychologia (2010) 48:880–6.
doi:10.1016/j.neuropsychologia.2009.11.005
33. Nelles G, de Grei A, Pscherer A, Stude P, Forsting M, Hufnagel A, et al.
Saccade induced cortical activation in patients with post-stroke visual eld
defects. J Neurol (2007) 254(9):1244–52. doi:10.1007/s00415-007-0511-1
34. Nelles G, Pscherer A, de Grei A, Forsting M, Gerhard H, Esser J, etal.
Eye-movement training-induced plasticity in patients with post-stroke
hemianopia. J Neurol (2009) 256(5):726–33. doi:10.1007/s00415-009-5005-x
35. Alahyane N, Fonteille V, Urquizar C, Salemme R, Nighoghossian N, Pelisson
D, etal. Separate neural substrates in the human cerebellum for sensory-motor
adaptation of reactive and of scanning voluntary saccades. Cerebellum (2008)
7(4):595–601. doi:10.1007/s12311-008-0065-5
36. Sapir A, Hayes A, Henik A, Danziger S, Rafal R. Parietal lobe lesions disrupt
saccadic remapping of inhibitory location tagging. J Cogn Neurosci (2004)
16:503–9. doi:10.1162/089892904323057245
37. Kerkho G, Reinhart S, Ziegler W, Artinger F, Marquardt C, Keller I. Smooth
pursuit eye movement training promotes recovery from auditory and visual
neglect: a randomized controlled study. Neurorehabil Neural Repair (2013)
27(9):789–98. doi:10.1177/1545968313491012
38. Hunt AW, Mah K, Reed N, Engel L, Keightley M. Oculomotor-based vision
assessment in mild traumatic brain injury: a systematic review. J Head Trauma
Rehabil (2015). doi:10.1097/HTR.0000000000000174
39. Carrick FR, McLellan K, Brock JB, Randall C, Oggero E. Evaluation of the
eectiveness of a novel brain and vestibular rehabilitation treatment modality
in PTSD patients who have suered combat related traumatic brain injuries.
Front Public Health (2015) 3:15. doi:10.3389/fpubh.2015.00015
40. Carrick FR, Pagnacco G, McLellan K, Solis R, Shores J, Fredieu A, etal. Short-
and long-term eectiveness of a subject’s specic novel brain and vestibular
rehabilitation treatment modality in combat veterans suering from PTSD.
Front Public Health (2015) 3:151. doi:10.3389/fpubh.2015.00151
41. Carrick FR, Pagnacco G, Wright CH, Oggero E. Changes in saccadic eye
movements produced by novel brain and vestibular rehabilitation therapy.
Biomed Sci Instrum (2015) 51:9–16.
42. Dong W, Yan B, Johnson BP, Millist L, Davis S, Fielding J, et al. Ischaemic
stroke: the ocular motor system as a sensitive marker for motor and cognitive
recovery. J Neurol Neurosurg Psychiatry (2013) 84(3):337–41. doi:10.1136/
jnnp-2012-303926

January 2016 | Volume 7 | Article 310
Carrick et al.
Eye-Movement Therapy in Acute Ischemic Stroke
Frontiers in Neurology | www.frontiersin.org
43. Song Y, Zang DW, Jin YY, Wang ZJ, Ni HY, Yin JZ, etal. Background rhythm
frequency and theta power of quantitative EEG analysis: predictive biomark-
ers for cognitive impairment post-cerebral infarcts. Clin EEG Neurosci (2015)
46(2):142–6. doi:10.1177/1550059413517492
44. Schleiger E, Sheikh N, Rowland T, Wong A, Read S, Finnigan S. Frontal EEG
delta/alpha ratio and screening for post-stroke cognitive decits: the power
of four electrodes. Int J Psychophysiol (2014) 94(1):19–24. doi:10.1016/j.
ijpsycho.2014.06.012
45. Liao LD, Liu YH, Lai HY, Bandla A, Shih YY, Chen YY, et al. Rescue of
cortical neurovascular functions during the hyperacute phase of ischemia by
peripheral sensory stimulation. Neurobiol Dis (2015) 75:53–63. doi:10.1016/j.
nbd.2014.12.022
46. Zhang SJ, Ke Z, Li L, Yip SP, Tong KY. EEG patterns from acute
to chronic stroke phases in focal cerebral ischemic rats: correla-
tions with functional recovery. Physiol Meas (2013) 34:423–35.
doi:10.1088/0967-3334/34/4/423
47. Leon-Carrion J, Martin-Rodriguez JF, Damas-Lopez J, Barroso y Martin
JM, Dominguez-Morales MR. Delta-alpha ratio correlates with level of
recovery aer neurorehabilitation in patients with acquired brain injury. Clin
Neurophysiol (2009) 120:1039–45. doi:10.1016/j.clinph.2009.01.021
48. Abdul-Rahim AH, Fulton RL, Sucharew H, Kleindorfer D, K hatri P, Broderick
JP, etal. National Institutes of Health Stroke Scale item proles as predictor of
patient outcome: external validation on independent trial data. Stroke (2015)
46(2):395–400. doi:10.1161/STROKEAHA.114.006837
Conict of Interest Statement: e authors declare that the research was con-
ducted in the absence of any commercial or nancial relationships that could be
construed as a potential conict of interest.
Copyright © 2016 Carrick, Oggero, Pagnacco, Wright, Machado, Estrada, Pando,
Cossio and Beltrán. is is an open-access article distributed under the terms of the
Creative Commons Attribution License (CC BY). e use, distribution or reproduc-
tion in other forums is permitted, provided the original author(s) or licensor are
credited and that the original publication in this journal is cited, in accordance with
accepted academic practice. No use, distribution or reproduction is permitted which
does not comply with these terms.