- Email: [email protected]
Visual processing features in patients with visual spatial neglect recovering from right-hemispheric stroke
Journal Pre-proof Visual processing features in patients with visual spatial neglect recovering from right-hemispheric stroke Linlin Ye, Lei Cao, Huanxin Xie, Guixiang Shan, Jie Hu, Jubao Du, Weiqun Song
PII:
S0304-3940(19)30631-7
DOI:
https://doi.org/10.1016/j.neulet.2019.134528
Reference:
NSL 134528
To appear in:
Neuroscience Letters
Received Date:
17 October 2018
Revised Date:
9 September 2019
Accepted Date:
30 September 2019
Please cite this article as: Ye L, Cao L, Xie H, Shan G, Hu J, Du J, Song W, Visual processing features in patients with visual spatial neglect recovering from right-hemispheric stroke, Neuroscience Letters (2019), doi: https://doi.org/10.1016/j.neulet.2019.134528
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 Published by Elsevier.
Visual processing features in patients with visual spatial neglect recovering from right-hemispheric stroke Running title: ERP changes of VSN recovery after stroke Linlin Ye1, Lei Cao1, Huanxin Xie1, Guixiang Shan1, Jie Hu1, Jubao Du1, Weiqun Song1, * 1
Department of Rehabilitation, Xuanwu Hospital, Capital Medical University,
ro of
Beijing100053, China. No. 45 Changchun Road, Beijing. *Weiqun Song, Department of Rehabilitation, Xuanwu Hospital, Capital Medical University, Beijing100053, China. No. 45 Changchun Road, Beijing.
-p
Tel: +8613391516966;
re
Email: [email protected]
lP
Highlights
This was a prospective analysis approach on a cohort of 115 patients.
ERP was used to explore impairment in the temporal processing capacity in VSN.
ERP and behavioral data between patients with or without rapid recovery were
ur
na
Jo
compared.
Patients from one week to four weeks after stroke were examined.
ABSTRACT
Objective: Visual spatial neglect (VSN) is a disorder of spatial-temporal attention, often as a result of traumatic brain injury, including stroke. Accumulating evidence suggests that the recovery from VSN follows a very predictable pattern. In this study,
we aimed to determine the specific electrophysiology readout that might have predictive value for recovery from VSN in the typical early events, including the recovery rate of visual processing, within the first four weeks of recovery. Methods: This was a prospective study of 18 right ischemic stroke patients with VSN who performed a visual cue-target task within 3 days after stroke. The patients were divided into two groups according to their outcome. We compared the amplitudes;
ro of
latencies of P1, N1, and P300; and behavioral data between patients with persistent-VSN (P-VSN) and those with rapid recovery-VSN (R-VSN).
Results: The amplitudes and latencies of the P1 and N1 components were not
-p
significantly influenced by the validity of the cue-based expectancy (all p > 0.05).
re
However, a longer mean P300 latency evoked an effective cue (p < 0.001), and there was a significant difference between the P-VSN and R-VSN groups when using the
lP
left target (left hemisphere, p = 0.014; right hemisphere, p = 0.027). The recovery rate
na
found in our study (18.75% at four weeks after stroke) was lower than that of previously reported studies.
ur
Conclusions: Our findings support the use of the event-related potential as a tool for investigating rapid recovery from VSN after stroke and suggest that other factors,
Jo
such as an asymmetrical omission toward the contralateral side or impairment in the temporal processing capacity, might also be potential biomarkers of recovery. KEYWORDS: visual spatial neglect, stroke recovery, event-related potential
INTRODUCTION Visual spatial neglect (VSN) is a neurological complication that mainly occurs after traumatic insults, such as stroke, and primarily affects the right hemisphere. The hallmark of VSN spatial-temporal attention disorder is the failure to attend, respond, or orient to stimuli in the contralateral hemisphere [1]. It occurs in approximately 10– 82% of stroke victims [2,]. In the short-term, VSN can potentially prolong
ro of
hospitalization and increase the risk of falls, and the main long-term consequence is a decreased chance of community reintegration [3,4]. Although spontaneous recovery
occurs, many patients with VSN (approximately 30–40%) suffer from a variety of
-p
long-term consequences [5]. Nevertheless, accumulating evidence suggests that the
One
interesting
feature
of
re
recovery from VSN follows a predictable pattern [6, 7]. VSN
is
the
striking
laterality,
i.e.,
lP
contralateral-evoked visual processing is different from the ipsilateral-evoked one [8,
na
9]. Consistently, previous studies also have highlighted an asymmetrical omission toward the contralateral side [7, 10] and impairment in the temporal processing
ur
capacity in VSN [6, 11,12]. The electrophysiology of this asymmetry in the temporal processing may predict recovery from VSN.
Jo
Based on this information, in this study, we measured VSN through at least
two approaches, including the event-related potential (ERP), and we reasoned that the ERP might be more sensitive than the paper-and-pencil test. Furthermore, we proposed that VSN might affect visual components in a stage-specific manner. To test this hypothesis, we conducted a cohort study, i.e., studied a variety of visual
parameters in patients with and without rapid recovery, using a new variant of the classic cue-target paradigm [13]. In addition, we carefully tracked the whole process of clinical presentation development from emergency admission to hospital discharge at four weeks after stroke. Our main goal was to find the specific clinical readout that might have the potential to serve as a biomarker or have predictive value for recovery
ro of
from VSN.
MATERIALS AND METHODS Study design and participants
-p
Patients were recruited between January 2017 and January 2018 from the Stroke Unit
re
of the Department of Neurology, Emergency Room, and the Rehabilitation Department at Xuan Wu Hospital of Capital Medical University, China. This study
lP
was approved by the local ethics committee. Full written consent was obtained from
na
all patients. A total of 115 right-handed patients with stroke in the right hemisphere were recruited. Patients with stroke in the left hemisphere or with a history of a
ur
previous stroke were excluded from this study. A total of 48 patients were diagnosed as having VSN; among these patients, 21 patients were excluded from further analysis
Jo
because of age, inability to tolerate the ERP examination, structural changes of the brain, cognitive impairment, compliance issues, or artifact problems. The ERPs was recorded only once, within 3days after stroke. Complete ERP data were obtained from the remaining 27 patients with a cut-off Mini–Mental State Examination score >17. These 27 patients performed paper-and-pencil test again at four weeks after their
stroke. Nine patients with a negative performance in all tests were assigned to the rapidly-recovered group, and nine patients with positive test performance in more than two tests were assigned into the persistent-VSN group (Figure 1). Additionally, 10 age- and sex-matched healthy participants were recruited as controls from outpatient clinics. For these controls, the paper-and-pencil test was confirmed to be negative, and the absence of cerebral infarction or cerebral hemorrhage was
ro of
confirmed by computed tomography or magnetic resonance imaging.
Behavioral testing
-p
Patients were clinically examined by a neurologist and a neurologically trained
re
occupational therapist at two time points, respectively. The first examination occurred within 3 days of the stroke, and the second was on the day of discharge to
lP
neurological rehabilitation (four weeks after stroke). We also carefully checked the
na
clinical reports of all non-VSN patients to confirm the diagnosis. All patients received standard visual scanning training and performed standardized paper-and-pencil tests,
ur
including 1) line bisection test, 2) line cancellation task, 3) clock copying task, 4) gap detection test, 5) star cancellation task, and 6) text reading. Performance on all tests
Jo
was assessed as described previously [14]. In the line bisection test, a >12% rightward deviation was considered to be pathological neglect (positive performance). In the line cancellation task, an omission of ≥3 targets in the left hemifield was considered to be pathological neglect (positive performance). In the clock drawing task, omission of the left side of the clock was considered to be pathological neglect (positive
performance). In the gap detection test, an omission of ≥3 targets in the left hemifield or ≥3 gaps in the left side of the circle was considered to be pathological neglect (positive performance). In the star cancellation task, an omission of ≥ 5 targets in the left hemifield was considered to be pathological neglect (positive performance). In the text reading task, if one or more sentences on the left side, but not on the right side, were misread, this was considered to be a positive performance. Patients were
ro of
considered to have left VSN if they had positive performance on two or more of the above-mentioned clinical tests. The behavioral scores were calculated according to
-p
the previously described method (Table 1) [15].
re
Stimulus presentation procedure
Stimulus presentation was controlled by E-prime 4.5 software. The 14-inch screen
lP
was placed in parallel to the midline of the subject at a distance of 500 mm.
na
Responses were made with the laptop’s mouse. The stimuli were located in two squares with a size of 15 mm, located 60 mm to the left and right of the screen’s
ur
center, respectively. The width of the line was 0.2 cm. The cues were “>” and “<”, and the target was “*”. The ERP task consisted of 16 sessions, each consisting of 40
Jo
trials. The participants were instructed to fixate on the center of the screen for the trial. During each trial, the background was presented for 800–1000 ms, then the cue was presented for 1400–1800 ms, and the target was presented for 100 ms thereafter. The participants used the mouse to answer whether the “*” appears in the left or right square. They were asked to detect the appearance of the target by pressing a button as
soon as possible. The subjects responded by pressing their right index finger to the left mouse button when the target appeared on the left and their right middle finger to the right mouse button when the target appeared on the right. Previous exercises were carried out to ensure that the patients did not press the wrong button. The participants were asked to move the focus to the target location and then return it to the center before the next target appeared. The interval between two consecutive trials was 1200
ro of
ms. Overall, each participant completed 640 trials. If the direction of the cue was the same as the target, it was recorded as an effective cue; otherwise, it was an ineffective
cue. The effective-to-ineffective ratio was 80:20. The time course of events on each
-p
trial of the spatial attention task is shown in Figure 2. The electrical signals were
re
collected using the International electroencephalogram (EEG) system and processed
lP
using E-prime software.
na
EEG recording
The EEG was recorded using a Neuroscan system from 64 electrodes placed
ur
according to the 10-20 system montage. All scalp channels refer to reference electrode conversion of the bilateral mastoids. Both the blink and vertical eye movement were
Jo
recorded with an electrode under the left eye, with reference position Fp1VEO. The EEG from each electrode site was digitized at 250 Hz with an amplifier bandpass of 0.05–80 Hz, including a 50-Hz notch filter, and was stored for off-line analysis. The semi-automatic artifact rejection procedure was performed before signal averaging, and electrooculogram correction was performed by blink filtering. An EEG
amplitude >100 μV or <-100 μV was discarded as an artifact. The analysis time course was 1000 ms (-200 to 800 ms). The final amplitude of the different components was the highest value within the specified time window minus the baseline value at 200 ms before the stimulation. Only the correctly responded trials were analyzed. For every subject, the average ERP components were obtained at each
ro of
electrode site. Artifact rates (proportion of trials) in individual subjects ranged from
1.86% to 25.47%. The time window for the P1 component was measured as 90–160
ms, and the time window for the N1 component was measured as 160–230 ms by
-p
visually inspecting the average ERP waveforms obtained from all subjects at the P3,
re
P5, PO7, P4, P6, and PO8 electrode sites. The P1 and N1 amplitudes were determined by averaging each electrode within the time window for each individual. The time
lP
window for the P300 component was determined as 300–700 ms by visually
na
inspecting the average ERP waveforms obtained from all subjects at the F3, F4, C3, C4, P3, and P4 electrode sites. The P300 amplitude was determined by averaging each
ur
electrode within the time window for each individual.
Jo
Statistical analysis
Intragroup comparisons of behavioral scores were performed using the Wilcoxon signed-rank test. Repeated measures analysis of variance (ANOVA) was used to compare different groups. The factors of interest were different in various contexts. Separate ANOVA analyses were conducted on amplitudes and latencies for each
component (P1, N1, and P300). The Greenhouse-Geisser correction was applied to the results. The significance level was set at p < 0.05.
RESULTS General demographic data and paper-and-pencil test data of patients with persistent VSN (P-VSN) or rapid recovery VSN (R-VSN)
ro of
The general demographic data and the paper-and-pencil test data of 18 patients (9 patients with P-VSN and 9 patients with R-VSN) are summarized in Table 1. Among
the 18 patients with right-hemisphere stroke, 9 patients (7 men and 2 women) had
-p
P-VSN (an average age of 48.89 ± 15.89 years, 20–69 years) and 9 patients (9 men)
re
had R-VSN (an average age of 54.11 ± 10.90 years, 31–68 years). The average age of the normal control participants (8 men and 1 woman) was 46.60 ± 13.14 years (33–73
lP
years). It is worth mentioning that the demographic data, including the average ages
na
of the patients, of both groups were similar (F(2,25) = 0.765, p = 0.476).
ur
Behavioral scores
Jo
There was no significant difference in the behavioral scores between the two groups in the first examination, i.e., in line bisection, z = 0.943 and p = 0.336; in line cancellation, z = 0.943 and p = 0.336; in star cancellation, z = 0.707 and p = 0.699. The behavioral scores of the two groups are summarized in Table 1.
Behavioral analyses of response time (RT) and accuracy rate The behavioral analyses of the RT and accuracy rate based on stimulus presentation with E-prime 4.5 software evaluation under different contexts are summarized in Figure 3. Specifically, for each type of trial, the RT and accuracy rate were recorded and analyzed by ANOVA. For the RT, the data demonstrated that there was no difference between the P-VSN and R-VSN groups, but both groups had significantly
ro of
different RT values from the control group, regardless of the target condition. For the accuracy rate, under the left target condition, regardless of a valid or an invalid cue,
the difference between the control group and the P-VSN group was significant (valid
-p
cue, F(2,25) = 24.650, p < 0.001; invalid cue, F(2,25) = 20.479, p < 0.001). In
re
addition, there was a significant difference in the accuracy rate between the P-VSN group and the R-VSN group (valid cue, F(2,25) = 24.650, p < 0.001; invalid cue
lP
F(2,25) = 20.479, p = 0.003); but under the right target condition, only the control
na
group was different from the P-VSN group (valid cue, F(2,25) = 10.730, p < 0.001;
ur
invalid cue, F(2,25) = 4.590, p = 0.016) (Figure 3).
Electrophysiological analyses of P1, N1, and P300 components
Jo
Next, we analyzed the main EEG factors between groups (control vs. P-VSN vs. R-VSN). For each participant, the mean amplitudes were computed, and these values were subsequently statistically analyzed (Figures 4–5). The P1 amplitude was not significantly influenced by the validity of the cue-based expectancy (F (2,25) = 1.033, p = 0.371). Similarly, the P1 latency was not significantly influenced by the
above-mentioned conditions either (F(2,25) = 1.089, p = 0.352). No other main effects or interactions were significant (p > 0.05), including the spatial attention. The mean amplitude and latency of P1 are summarized in Supp. Table 2. Likewise, the N1 amplitude was not significantly influenced by the validity of the cue-based expectancy (F(2,25) = 1.051, p = 0.365), and the N1 latency was not significantly influenced either (F(2,25) = 1.065, p = 0.360). No other main effects or
ro of
interactions were significant (p > 0.05), including the spatial attention. The mean amplitude and latency of N1 are summarized in Supp. Table 3.
For P300, even though we did not see a significant effect of the amplitude
-p
(F(2,25) = 3.131, p = 0.061), we observed a main effect of spatial attention (F(2,25) =
re
37.160, p < 0.001), with a longer mean P300 latency evoked over the contralateral visual target. When the left and right targets were compared, we found that there was
lP
a significant difference between the control group and the P-VSN/R-VSN group using
na
the right target (p < 0.001); while using the left target, there was a significant difference not only between the two groups (p < 0.001) but also between the P-VSN
ur
and R-VSN groups (left hemisphere, p = 0.014; right hemisphere, p = 0.027). No other main effects or interactions were significant (all p > 0.05). The mean amplitude
Jo
and latency of P300 are summarized in Table 2.
DISCUSSION The main purpose of this study in stroke patients was to clarify the relationships between early recovery and visual processing within the first four weeks after stroke.
The results of this study suggest that under valid stimuli, the latency of P300 to the contralensional side could potentially predict recovery from VSN. In our cohort of 115 patients with right-hemispheric stroke, 41.7% (48/115) of them displayed acute VSN to the right side within the first 72 h after their stroke. These data roughly confirm the previously reported incidence of VSN ranging from 10% to 82% in right-hemispheric stroke patients. However, it is worth mentioning
ro of
that the actual incidence might be even higher in our study, since we only included
patients who understood and tolerated clinical paper-based testing. Remarkably, we
observed that only 18.75% (9/48) of the patients with VSN had totally recovered from
-p
these symptoms at four weeks after their stroke. This recovery rate is apparently lower
re
than the previously reported values of 43–77% [16-18]. However, no study has
right-hemispheric stroke.
lP
closely examined the neuropsychological recovery from VSN at four weeks after a
na
Some literature reports suggest that the initial behavioral scores may predict prognosis, but our results do not support this idea. This discrepancy is probably
ur
because of cerebral edema, a state of confusion, and changes in arousal in the acute phase. Therefore, to determine VSN reliably, we used several examinations, with
Jo
three scale scores. Although the clock test is slightly different from the others, it is still not sensitive enough to detect pathological neglect in patients with hemianopia or constructional apraxia [19]. In some studies investigating VSN, the RT and error rate (or accuracy rate in our study) are thought to be the best criteria because of the functional relevancy of
these parameters to clinical practice [20]. Some data also have indicated that the speed to complete the paper-and-pencil test is the most important predictor for recovery at 2 years after a stroke [9]. However, our study found that the RT could be used to predict an increased perceptual expectation, but it was not a good predictor of recovery in the early stage. In contrast, the accuracy rate to the contralensional side could be related to the recovery in terms of high-order cognitive processing.
ro of
Unlike the behavioral RT and accuracy rate, measurements that cannot distinguish different component processes, the stages of processing that are influenced
by attention can be directly measured by the ERP. We chose to apply the cue-target
-p
paradigm in our study. We focused on analysis of the early component P1/N1 and the
re
late component P300 to verify the processing time. However, differences were not significant for either the N1 or P1 component, which represented the activity of the
lP
striate and extrastriate areas in our study. Interestingly, Verleger et al. have noted that
na
the N1 component evoked by left visual field stimuli was reduced at the right recording site [21]. In contrast, Francesco et al. failed to record the N1a component,
ur
but the P1 component was not significantly different [8]. For this reason, we detected three components under the valid cue and
Jo
The P300 component is a positive component of the ERP that peaks at 300 ms
or more (up to 900 ms) after a stimulus [22]. P300 is known to reflect neural processing associated with task-related information, and its latency is considered to be a measure of stimulus classification speed [23], which is generally unrelated to the overt response. Whereas P300 latency is sensitive to variables involved in stimulus
evaluation [24], it is feasible to use P300 as a measurement of cognitive function; and the practical and theoretical support indeed exist, since the P300 peak latency is negatively correlated with the mental function of normal subjects, i.e., shorter latencies are associated with superior cognitive performance from neuropsychological tests of attention and immediate memory [25]. The Posner paradigm involves manipulation of executive function demands not only in attention but also in
ro of
visuospatial processing [26], which can be more demanding than a simple task. The delay of the P300 latency in our study may have resulted from the Posner task, which not only reflects the global attentional processing but also the cognitive deficits in
-p
visuospatial attention.
re
Even though it is well known that P300 is modulated by attention, only a handful of studies have so far used it to investigate the role of attention in VSN,
lP
possibly because of the difficulty in recruiting subjects who are able to participate in
na
such experiments [27]. Research regarding P300 suggests an important link between attentional dysfunction and impaired stimulus processing in VSN, i.e., in the affected
ur
hemisphere, the difference in the P300 amplitude between the identified and missed targets was correlated with the miss rate [28]. However, our study did not find a
Jo
similar difference of amplitudes. In fact, the amplitudes of patients in our study were similar to those of the healthy controls. In this regard, it is of note that P300 is also modulated by the overall arousal level, which governs the amount of attention available for task performance. To the best of our knowledge, this work is the first to evaluate the
electrophysiological processes involved in the recovery from VSN at four weeks after stroke. In the past few years, researchers have tried to use EEG, the brain-computer interface (BCI), and other methods to explore ways to predict VSN recovery. For example, in a small study, Ros et al. confirmed the feasibility of EEG as a way to predict the recovery of patients with pathological neglect. They found that the resting-state alpha amplitude variability at the right posterior parietal cortex was the
ro of
principal predictor of VSN recovery by examining the EEG of five patients [29]. Moreover, Tonin et al. detected attention task-specific brain patterns in three VSN
patients by BCI and suggested that it is possible to use BCI to predict lateralized,
-p
attention-based VSN [30]. In contrast, we suggest that ERP can be used as a
re
predictive tool for recovery from VSN.
Furthermore, an asymmetrical omission toward the contralesional side in VSN
lP
and impairment in the temporal processing capacity may represent a good marker of
na
recovery. Our findings are consistent with a report suggesting that an inability of capturing attention and a slow speed of visual processing are not only key symptoms
ur
of VSN, but they are also important predictors of functional dependency [9]. However, some limitations of this study should be mentioned. First, this study
Jo
was a single-center study with a relatively small number of patients. Second, the P300 component may be confounded by some clinical variables such as dietary and circadian factors [31]. Although we tested all of the participants during the same time frame, the ERP data may display individual differences and some participants could not be evaluated for long periods of time due to their general health status.
CONCLUSIONS We used ERP with exceptional temporal resolution to show direct evidence of temporal processing and recovery in patients at four weeks after a stroke. We concluded that the P300 latency might be an index for speed and efficiency of information processing in the brain. This study contributes to the knowledge and
ro of
understanding of the functional impairment and recovery in patients with VSN.
Funding
-p
This research was supported by the National Natural Science Foundation of China
re
(grant no. 81671048). Conflict of Interest
None.
Heilman KM, Valenstein E, Watson RT, Neglect and related disorders, Seminars
Jo
1.
ur
References
na
Acknowledgments
lP
The authors declare that they have no conflicts of interest.
in neurology. 20 (2000) 463-70. https: //doi. 10.1055/s-2000-13179 2.
Stone SP, Halligan PW, Greenwood RJ, The incidence of neglect phenomena and
related disorders in patients with an acute right or left hemisphere stroke, Age and ageing. 22 (1993) 46-52. 3.
Paolucci S, Antonucci G, Grasso MG, Pizzamiglio L, The role of unilateral
spatial neglect in rehabilitation of right brain-damaged ischemic stroke patients: a matched comparison, Archives of physical medicine and rehabilitation. 82 (2001) 743-9. 4.
Katz N, Hartman-Maeir A, Ring H, Soroker N, Functional disability and
rehabilitation outcome in right hemisphere damaged patients with and without unilateral spatial neglect, Archives of physical medicine and rehabilitation. 80 (1999) 379-84. Nijboer T, van de Port I, Schepers V, Post M, Visser-Meily A, Predicting
ro of
5.
functional outcome after stroke: the influence of neglect on basic activities in daily living,
Frontiers
in
human
neuroscience.
7
(2013)
6.
https:
//doi.
-p
10.3389/fnhum.2013.00182
182.
Husain M, Shapiro K, Martin J, Kennard C, Abnormal temporal dynamics of
re
visual attention in spatial neglect patients, Nature. 385 (1997) 154-6. https: //doi.
7. of
lP
10.1038/385154a0
Corbetta M, Shulman GL, Spatial neglect and attention networks, Annual review neuroscience.
34
(2011)
569-99.
https:
//doi.
8.
na
10.1146/annurev-neuro-061010-113731
Di Russo F, Aprile T, Spitoni G, Spinelli D, Impaired visual processing of
ur
contralesional stimuli in neglect patients: a visual-evoked potential study, Brain : a
Jo
journal of neurology. 131 (2008) 842-54. https: //doi. 10.1093/brain/awm281 9.
Viken JI, Jood K, Jern C, Blomstrand C, Samuelsson H, Ipsilesional bias and
processing speed are important predictors of functional dependency in the neglect phenomenon after a right hemisphere stroke, The Clinical neuropsychologist. 28 (2014) 974-93. https: //doi. 10.1080/13854046.2014.950335 10. van Kessel ME, van Nes IJ, Brouwer WH, Geurts AC, Fasotti L, Visuospatial
asymmetry and non-spatial attention in subacute stroke patients with and without neglect, Cortex; a journal devoted to the study of the nervous system and behavior. 46 (2010) 602-12. https: //doi. 10.1016/j.cortex.2009.06.004 11. Husain M, Rorden C, Non-spatially lateralized mechanisms in hemispatial neglect, Nature reviews Neuroscience. 4 (2003) 26-36. https: //doi. 10.1038/nrn1005 12. Samuelsson H, Hjelmquist EK, Jensen C, Ekholm S, Blomstrand C, Nonlateralized attentional deficits: an important component behind persisting
ro of
visuospatial neglect?, Journal of clinical and experimental neuropsychology. 20 (1998) 73-88. https: //doi. 10.1076/jcen.20.1.73.1481
13. Posner MI, Orienting of attention, The Quarterly journal of experimental
-p
psychology. 32 (1980) 3-25.
14. Umarova RM, Reisert M, Beier TU, Kiselev VG, Kloppel S, Kaller CP, et al.,
re
Attention-network specific alterations of structural connectivity in the undamaged
10.1002/hbm.22503 15.
lP
white matter in acute neglect, Human brain mapping. 35 (2014) 4678-92. https: //doi.
Lee BH, Kang SJ, Park JM, Son Y, Lee KH, Adair JC, et al., The Character-line
1715-24.
na
Bisection Task: a new test for hemispatial neglect. Neuropsychologia. 42 (2004)
ur
16. Farne A, Buxbaum LJ, Ferraro M, Frassinetti F, Whyte J, Veramonti T, et al.,
Jo
Patterns of spontaneous recovery of neglect and associated disorders in acute right brain-damaged patients, Journal of neurology, neurosurgery, and psychiatry. 75 (2004) 1401-10. https: //doi. 10.1136/jnnp.2002.003095 17. Cassidy TP, Lewis S, Gray CS, Recovery from visuospatial neglect in stroke patients, Journal of neurology, neurosurgery, and psychiatry. 64 (1998) 555-7. 18. Ramsey LE, Siegel JS, Baldassarre A, Metcalf NV, Zinn K, Shulman GL, et al.,
Normalization of network connectivity in hemispatial neglect recovery, Annals of neurology. 80 (2016) 127-41. https: //doi. 10.1002/ana.24690 19. Agrell BM, Dehlin OI, Dahlgren CJ, Neglect in elderly stroke patients: a comparison of five tests. Psychiatry Clin Neurosci. 51 (1997) 295-300. http://dx.doi.org/10.1111/j.1440-1819.1997.tb03201.x. 20. Polich J, Clinical application of the P300 event-related brain potential, Physical medicine and rehabilitation clinics of North America. 15 (2004) 133-61.
ro of
21. Verleger R, Heide W, Butt C, Wascher E, Kompf D, On-line brain potential correlates of right parietal patients' attentional deficit, Electroencephalography and clinical neurophysiology. 99 (1996) 444-57.
-p
22. Linden DE, The p300: where in the brain is it produced and what does it tell us?, The Neuroscientist : a review journal bringing neurobiology, neurology and
re
psychiatry. 11 (2005) 563-76. https: //doi. 10.1177/1073858405280524
lP
23. Magliero A, Bashore TR, Coles MG, Donchin E, On the dependence of P300 latency on stimulus evaluation processes, Psychophysiology. 21 (1984) 171-86. 24. Duncan-Johnson CC, Donchin E, The P300 component of the event-related brain
na
potential as an index of information processing, Biological psychology. 14 (1982) 1-52.
ur
25. Reinvang I, Cognitive event-related potentials in neuropsychological assessment,
Jo
Neuropsychology review. 9 (1999) 231-48. 26. Theeuwes J, Top-down and bottom-up control of visual selection, Acta psychologica. 135 (2010) 77-99. https: //doi. 10.1016/j.actpsy.2010.02.006 27. Deouell LY, Hamalainen H, Bentin S, Unilateral neglect after right-hemisphere damage: contributions from event-related potentials, Audiology & neuro-otology. 5 (2000) 225-34. https: //doi. 10.1159/00001388428. Saevarsson S, Kristjansson A,
Bach M, Heinrich SP, P300 in neglect, Clinical neurophysiology : official journal of the International Federation of Clinical Neurophysiology. 123 (2012) 496-506. https: //doi. 10.1016/j.clinph.2011.07.028 29. Ros T, Michela A, Bellman A, Vuadens P, Saj A, Vuilleumier P, Increased Alpha-Rhythm Dynamic Range Promotes Recovery from Visuospatial Neglect: A Neurofeedback Study. Neural Plasticity. 2017 (2017) 1-9. 30. Tonin L, Pitteri M, Leeb R, Zhang H, Menegatti E, Piccione F, Millán JDR, et al.,
ro of
Behavioral and Cortical Effects during Attention Driven Brain-Computer Interface Operations in Spatial Neglect: A Feasibility Case Study. Frontiers in Human Neuroscience, 2017 11.
-p
31. Polich J, Kok A, Cognitive and biological determinants of P300: an integrative
re
review, Biological psychology. 41 (1995) 103-46.
lP
Figure 1. Flow chart of the sequential assessment of the patients included in this
Jo
ur
na
study.
ro of -p
ur
na
lP
re
Figure 2 Schematic view of the experimental paradigm.
Jo
Figure 3. The bar graphs depict the accuracy rate (a) and the response time (RT) (b) of the control, persistent visual spatial neglect (P-VSN), and rapid recovery visual spatial neglect (R-VSN) groups in the contexts of a valid (top panels of a&b) or an invalid target (bottom panels of a&b) as well as a left-cue (left panels of a&b) or a right-cue target (right panels of a&b). *p < 0.05, **p < 0.01, ***p < 0.001.
ro of
Figure 4. Typical average waveforms of event-related potentials from electrode (P300, i.e., P3&P4, P1N1, i.e., PO7&PO8) recordings of a valid left target on the scalp. The
-p
control, rapid recovery visual spatial neglect (R-VSN), and persistent visual spatial
na
lP
re
neglect (P-VSN) groups are indicated by blue, green, and red lines, respectively.
ur
Figure 5. Typical average waveforms of event-related potentials from electrode (P300,
Jo
i.e., P3&P4, P1N1, i.e., PO7&PO8) recordings of a valid right target on the scalp. The control, rapid recovery visual spatial neglect (R-VSN), and persistent visual spatial neglect (P-VSN) groups are indicated by blue, green, and red lines, respectively.
neglect or rapid recovery visual spatial neglect
Pathoge ny
P1 P2 P3 P4 P5 P6 P7 P8 P9 R1 R2 R3 R4 R5 R6 R7 R8 R9
M M F M M M F M M M M M M M M M M M
39 66 52 32 57 20 69 54 51 61 61 53 61 31 68 49 47 56
CH CI CI CH CI CH CH CI CI CI CI CI CI CI CI CI CI CH
SC CSC C CSC CSC C C CSC CSC SC CSC SC CSC CSC SC C C C
25.27 39.71 26.27 69.93 21.38 14.55 78.16 26.29 18.75 14.22 13.79 17.62 15.16 22.65 28.25 14.49 19.14 40.18
0.33 5.00 1.67 4.33 0.11 1.29 9.00 0.50 0.50 0.50 0.33 0.33 0.11 1.67 0.50 0.71 0.71 0.21
+ + + + + + + -
na
ur
Jo
Clock drawi ng
Gap detecti on
Star cancellat ion
Senten ce readin g
+ + + + + + + + -
0.74 1.80 1.47 7.68 1.74 1.74 7.68 2.89 2.89 0.50 0.50 1.47 0.13 1.74 2.89 6.61 1.80 0.88
+ + + + + + + + + + + -
-p
Ag e
Line cancellat ion
re
Se x
Line bisectio n%
lP
Patie nt
Lesi on type
ro of
Table 1. Demographic and general clinical data of the patients with persistent visual spatial
CH, cerebral hemorrhagic; CI, cerebral ischemic; C, cortical; SC, subcortical; CSC, cortical-subcortical; +, positive; -, negative; F, female; M, male; P, persistent visual spatial neglect; R, rapid recovery visual spatial neglect.
Table 2. Amplitude and latency of the P300 component Amplitude (μV)
Latency (ms)
Left Target Left hemisph ere
Right hemisph ere
Right Target
Left Target
Right Target
Left hemisph ere
Right hemisph ere
Left hemisph ere
Right hemisph ere
Left hemisph ere
Right hemisph ere
497.15 ± 87.31
425.78 ± 87.50
6.48 4.42
± 8.41 4.55
±
4.54 4.88
±
6.46 4.39
±
521.48 ± 95.93
529.07 ± 87.82
Rapid recove ry VSN group
8.10 4.49
± 5.93 3.55
±
5.98 4.01
±
5.64 3.69
±
434.56 ± 41.64
428.56 ± 37.42
Contro l group
9.02 4.85
± 10.58 ± 4.72
9.79 5.31
±
10.17 ± 366.43 4.84 ± 23.93
-p
re lP na ur Jo
ro of
Persist ent VSN group
359.27 ± 25.07
435.22 ± 51.59
433.11 ± 87.84
386.10 ± 50.88
373.70 ± 34.57
PII:
S0304-3940(19)30631-7
DOI:
https://doi.org/10.1016/j.neulet.2019.134528
Reference:
NSL 134528
To appear in:
Neuroscience Letters
Received Date:
17 October 2018
Revised Date:
9 September 2019
Accepted Date:
30 September 2019
Please cite this article as: Ye L, Cao L, Xie H, Shan G, Hu J, Du J, Song W, Visual processing features in patients with visual spatial neglect recovering from right-hemispheric stroke, Neuroscience Letters (2019), doi: https://doi.org/10.1016/j.neulet.2019.134528
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 Published by Elsevier.
Visual processing features in patients with visual spatial neglect recovering from right-hemispheric stroke Running title: ERP changes of VSN recovery after stroke Linlin Ye1, Lei Cao1, Huanxin Xie1, Guixiang Shan1, Jie Hu1, Jubao Du1, Weiqun Song1, * 1
Department of Rehabilitation, Xuanwu Hospital, Capital Medical University,
ro of
Beijing100053, China. No. 45 Changchun Road, Beijing. *Weiqun Song, Department of Rehabilitation, Xuanwu Hospital, Capital Medical University, Beijing100053, China. No. 45 Changchun Road, Beijing.
-p
Tel: +8613391516966;
re
Email: [email protected]
lP
Highlights
This was a prospective analysis approach on a cohort of 115 patients.
ERP was used to explore impairment in the temporal processing capacity in VSN.
ERP and behavioral data between patients with or without rapid recovery were
ur
na
Jo
compared.
Patients from one week to four weeks after stroke were examined.
ABSTRACT
Objective: Visual spatial neglect (VSN) is a disorder of spatial-temporal attention, often as a result of traumatic brain injury, including stroke. Accumulating evidence suggests that the recovery from VSN follows a very predictable pattern. In this study,
we aimed to determine the specific electrophysiology readout that might have predictive value for recovery from VSN in the typical early events, including the recovery rate of visual processing, within the first four weeks of recovery. Methods: This was a prospective study of 18 right ischemic stroke patients with VSN who performed a visual cue-target task within 3 days after stroke. The patients were divided into two groups according to their outcome. We compared the amplitudes;
ro of
latencies of P1, N1, and P300; and behavioral data between patients with persistent-VSN (P-VSN) and those with rapid recovery-VSN (R-VSN).
Results: The amplitudes and latencies of the P1 and N1 components were not
-p
significantly influenced by the validity of the cue-based expectancy (all p > 0.05).
re
However, a longer mean P300 latency evoked an effective cue (p < 0.001), and there was a significant difference between the P-VSN and R-VSN groups when using the
lP
left target (left hemisphere, p = 0.014; right hemisphere, p = 0.027). The recovery rate
na
found in our study (18.75% at four weeks after stroke) was lower than that of previously reported studies.
ur
Conclusions: Our findings support the use of the event-related potential as a tool for investigating rapid recovery from VSN after stroke and suggest that other factors,
Jo
such as an asymmetrical omission toward the contralateral side or impairment in the temporal processing capacity, might also be potential biomarkers of recovery. KEYWORDS: visual spatial neglect, stroke recovery, event-related potential
INTRODUCTION Visual spatial neglect (VSN) is a neurological complication that mainly occurs after traumatic insults, such as stroke, and primarily affects the right hemisphere. The hallmark of VSN spatial-temporal attention disorder is the failure to attend, respond, or orient to stimuli in the contralateral hemisphere [1]. It occurs in approximately 10– 82% of stroke victims [2,]. In the short-term, VSN can potentially prolong
ro of
hospitalization and increase the risk of falls, and the main long-term consequence is a decreased chance of community reintegration [3,4]. Although spontaneous recovery
occurs, many patients with VSN (approximately 30–40%) suffer from a variety of
-p
long-term consequences [5]. Nevertheless, accumulating evidence suggests that the
One
interesting
feature
of
re
recovery from VSN follows a predictable pattern [6, 7]. VSN
is
the
striking
laterality,
i.e.,
lP
contralateral-evoked visual processing is different from the ipsilateral-evoked one [8,
na
9]. Consistently, previous studies also have highlighted an asymmetrical omission toward the contralateral side [7, 10] and impairment in the temporal processing
ur
capacity in VSN [6, 11,12]. The electrophysiology of this asymmetry in the temporal processing may predict recovery from VSN.
Jo
Based on this information, in this study, we measured VSN through at least
two approaches, including the event-related potential (ERP), and we reasoned that the ERP might be more sensitive than the paper-and-pencil test. Furthermore, we proposed that VSN might affect visual components in a stage-specific manner. To test this hypothesis, we conducted a cohort study, i.e., studied a variety of visual
parameters in patients with and without rapid recovery, using a new variant of the classic cue-target paradigm [13]. In addition, we carefully tracked the whole process of clinical presentation development from emergency admission to hospital discharge at four weeks after stroke. Our main goal was to find the specific clinical readout that might have the potential to serve as a biomarker or have predictive value for recovery
ro of
from VSN.
MATERIALS AND METHODS Study design and participants
-p
Patients were recruited between January 2017 and January 2018 from the Stroke Unit
re
of the Department of Neurology, Emergency Room, and the Rehabilitation Department at Xuan Wu Hospital of Capital Medical University, China. This study
lP
was approved by the local ethics committee. Full written consent was obtained from
na
all patients. A total of 115 right-handed patients with stroke in the right hemisphere were recruited. Patients with stroke in the left hemisphere or with a history of a
ur
previous stroke were excluded from this study. A total of 48 patients were diagnosed as having VSN; among these patients, 21 patients were excluded from further analysis
Jo
because of age, inability to tolerate the ERP examination, structural changes of the brain, cognitive impairment, compliance issues, or artifact problems. The ERPs was recorded only once, within 3days after stroke. Complete ERP data were obtained from the remaining 27 patients with a cut-off Mini–Mental State Examination score >17. These 27 patients performed paper-and-pencil test again at four weeks after their
stroke. Nine patients with a negative performance in all tests were assigned to the rapidly-recovered group, and nine patients with positive test performance in more than two tests were assigned into the persistent-VSN group (Figure 1). Additionally, 10 age- and sex-matched healthy participants were recruited as controls from outpatient clinics. For these controls, the paper-and-pencil test was confirmed to be negative, and the absence of cerebral infarction or cerebral hemorrhage was
ro of
confirmed by computed tomography or magnetic resonance imaging.
Behavioral testing
-p
Patients were clinically examined by a neurologist and a neurologically trained
re
occupational therapist at two time points, respectively. The first examination occurred within 3 days of the stroke, and the second was on the day of discharge to
lP
neurological rehabilitation (four weeks after stroke). We also carefully checked the
na
clinical reports of all non-VSN patients to confirm the diagnosis. All patients received standard visual scanning training and performed standardized paper-and-pencil tests,
ur
including 1) line bisection test, 2) line cancellation task, 3) clock copying task, 4) gap detection test, 5) star cancellation task, and 6) text reading. Performance on all tests
Jo
was assessed as described previously [14]. In the line bisection test, a >12% rightward deviation was considered to be pathological neglect (positive performance). In the line cancellation task, an omission of ≥3 targets in the left hemifield was considered to be pathological neglect (positive performance). In the clock drawing task, omission of the left side of the clock was considered to be pathological neglect (positive
performance). In the gap detection test, an omission of ≥3 targets in the left hemifield or ≥3 gaps in the left side of the circle was considered to be pathological neglect (positive performance). In the star cancellation task, an omission of ≥ 5 targets in the left hemifield was considered to be pathological neglect (positive performance). In the text reading task, if one or more sentences on the left side, but not on the right side, were misread, this was considered to be a positive performance. Patients were
ro of
considered to have left VSN if they had positive performance on two or more of the above-mentioned clinical tests. The behavioral scores were calculated according to
-p
the previously described method (Table 1) [15].
re
Stimulus presentation procedure
Stimulus presentation was controlled by E-prime 4.5 software. The 14-inch screen
lP
was placed in parallel to the midline of the subject at a distance of 500 mm.
na
Responses were made with the laptop’s mouse. The stimuli were located in two squares with a size of 15 mm, located 60 mm to the left and right of the screen’s
ur
center, respectively. The width of the line was 0.2 cm. The cues were “>” and “<”, and the target was “*”. The ERP task consisted of 16 sessions, each consisting of 40
Jo
trials. The participants were instructed to fixate on the center of the screen for the trial. During each trial, the background was presented for 800–1000 ms, then the cue was presented for 1400–1800 ms, and the target was presented for 100 ms thereafter. The participants used the mouse to answer whether the “*” appears in the left or right square. They were asked to detect the appearance of the target by pressing a button as
soon as possible. The subjects responded by pressing their right index finger to the left mouse button when the target appeared on the left and their right middle finger to the right mouse button when the target appeared on the right. Previous exercises were carried out to ensure that the patients did not press the wrong button. The participants were asked to move the focus to the target location and then return it to the center before the next target appeared. The interval between two consecutive trials was 1200
ro of
ms. Overall, each participant completed 640 trials. If the direction of the cue was the same as the target, it was recorded as an effective cue; otherwise, it was an ineffective
cue. The effective-to-ineffective ratio was 80:20. The time course of events on each
-p
trial of the spatial attention task is shown in Figure 2. The electrical signals were
re
collected using the International electroencephalogram (EEG) system and processed
lP
using E-prime software.
na
EEG recording
The EEG was recorded using a Neuroscan system from 64 electrodes placed
ur
according to the 10-20 system montage. All scalp channels refer to reference electrode conversion of the bilateral mastoids. Both the blink and vertical eye movement were
Jo
recorded with an electrode under the left eye, with reference position Fp1VEO. The EEG from each electrode site was digitized at 250 Hz with an amplifier bandpass of 0.05–80 Hz, including a 50-Hz notch filter, and was stored for off-line analysis. The semi-automatic artifact rejection procedure was performed before signal averaging, and electrooculogram correction was performed by blink filtering. An EEG
amplitude >100 μV or <-100 μV was discarded as an artifact. The analysis time course was 1000 ms (-200 to 800 ms). The final amplitude of the different components was the highest value within the specified time window minus the baseline value at 200 ms before the stimulation. Only the correctly responded trials were analyzed. For every subject, the average ERP components were obtained at each
ro of
electrode site. Artifact rates (proportion of trials) in individual subjects ranged from
1.86% to 25.47%. The time window for the P1 component was measured as 90–160
ms, and the time window for the N1 component was measured as 160–230 ms by
-p
visually inspecting the average ERP waveforms obtained from all subjects at the P3,
re
P5, PO7, P4, P6, and PO8 electrode sites. The P1 and N1 amplitudes were determined by averaging each electrode within the time window for each individual. The time
lP
window for the P300 component was determined as 300–700 ms by visually
na
inspecting the average ERP waveforms obtained from all subjects at the F3, F4, C3, C4, P3, and P4 electrode sites. The P300 amplitude was determined by averaging each
ur
electrode within the time window for each individual.
Jo
Statistical analysis
Intragroup comparisons of behavioral scores were performed using the Wilcoxon signed-rank test. Repeated measures analysis of variance (ANOVA) was used to compare different groups. The factors of interest were different in various contexts. Separate ANOVA analyses were conducted on amplitudes and latencies for each
component (P1, N1, and P300). The Greenhouse-Geisser correction was applied to the results. The significance level was set at p < 0.05.
RESULTS General demographic data and paper-and-pencil test data of patients with persistent VSN (P-VSN) or rapid recovery VSN (R-VSN)
ro of
The general demographic data and the paper-and-pencil test data of 18 patients (9 patients with P-VSN and 9 patients with R-VSN) are summarized in Table 1. Among
the 18 patients with right-hemisphere stroke, 9 patients (7 men and 2 women) had
-p
P-VSN (an average age of 48.89 ± 15.89 years, 20–69 years) and 9 patients (9 men)
re
had R-VSN (an average age of 54.11 ± 10.90 years, 31–68 years). The average age of the normal control participants (8 men and 1 woman) was 46.60 ± 13.14 years (33–73
lP
years). It is worth mentioning that the demographic data, including the average ages
na
of the patients, of both groups were similar (F(2,25) = 0.765, p = 0.476).
ur
Behavioral scores
Jo
There was no significant difference in the behavioral scores between the two groups in the first examination, i.e., in line bisection, z = 0.943 and p = 0.336; in line cancellation, z = 0.943 and p = 0.336; in star cancellation, z = 0.707 and p = 0.699. The behavioral scores of the two groups are summarized in Table 1.
Behavioral analyses of response time (RT) and accuracy rate The behavioral analyses of the RT and accuracy rate based on stimulus presentation with E-prime 4.5 software evaluation under different contexts are summarized in Figure 3. Specifically, for each type of trial, the RT and accuracy rate were recorded and analyzed by ANOVA. For the RT, the data demonstrated that there was no difference between the P-VSN and R-VSN groups, but both groups had significantly
ro of
different RT values from the control group, regardless of the target condition. For the accuracy rate, under the left target condition, regardless of a valid or an invalid cue,
the difference between the control group and the P-VSN group was significant (valid
-p
cue, F(2,25) = 24.650, p < 0.001; invalid cue, F(2,25) = 20.479, p < 0.001). In
re
addition, there was a significant difference in the accuracy rate between the P-VSN group and the R-VSN group (valid cue, F(2,25) = 24.650, p < 0.001; invalid cue
lP
F(2,25) = 20.479, p = 0.003); but under the right target condition, only the control
na
group was different from the P-VSN group (valid cue, F(2,25) = 10.730, p < 0.001;
ur
invalid cue, F(2,25) = 4.590, p = 0.016) (Figure 3).
Electrophysiological analyses of P1, N1, and P300 components
Jo
Next, we analyzed the main EEG factors between groups (control vs. P-VSN vs. R-VSN). For each participant, the mean amplitudes were computed, and these values were subsequently statistically analyzed (Figures 4–5). The P1 amplitude was not significantly influenced by the validity of the cue-based expectancy (F (2,25) = 1.033, p = 0.371). Similarly, the P1 latency was not significantly influenced by the
above-mentioned conditions either (F(2,25) = 1.089, p = 0.352). No other main effects or interactions were significant (p > 0.05), including the spatial attention. The mean amplitude and latency of P1 are summarized in Supp. Table 2. Likewise, the N1 amplitude was not significantly influenced by the validity of the cue-based expectancy (F(2,25) = 1.051, p = 0.365), and the N1 latency was not significantly influenced either (F(2,25) = 1.065, p = 0.360). No other main effects or
ro of
interactions were significant (p > 0.05), including the spatial attention. The mean amplitude and latency of N1 are summarized in Supp. Table 3.
For P300, even though we did not see a significant effect of the amplitude
-p
(F(2,25) = 3.131, p = 0.061), we observed a main effect of spatial attention (F(2,25) =
re
37.160, p < 0.001), with a longer mean P300 latency evoked over the contralateral visual target. When the left and right targets were compared, we found that there was
lP
a significant difference between the control group and the P-VSN/R-VSN group using
na
the right target (p < 0.001); while using the left target, there was a significant difference not only between the two groups (p < 0.001) but also between the P-VSN
ur
and R-VSN groups (left hemisphere, p = 0.014; right hemisphere, p = 0.027). No other main effects or interactions were significant (all p > 0.05). The mean amplitude
Jo
and latency of P300 are summarized in Table 2.
DISCUSSION The main purpose of this study in stroke patients was to clarify the relationships between early recovery and visual processing within the first four weeks after stroke.
The results of this study suggest that under valid stimuli, the latency of P300 to the contralensional side could potentially predict recovery from VSN. In our cohort of 115 patients with right-hemispheric stroke, 41.7% (48/115) of them displayed acute VSN to the right side within the first 72 h after their stroke. These data roughly confirm the previously reported incidence of VSN ranging from 10% to 82% in right-hemispheric stroke patients. However, it is worth mentioning
ro of
that the actual incidence might be even higher in our study, since we only included
patients who understood and tolerated clinical paper-based testing. Remarkably, we
observed that only 18.75% (9/48) of the patients with VSN had totally recovered from
-p
these symptoms at four weeks after their stroke. This recovery rate is apparently lower
re
than the previously reported values of 43–77% [16-18]. However, no study has
right-hemispheric stroke.
lP
closely examined the neuropsychological recovery from VSN at four weeks after a
na
Some literature reports suggest that the initial behavioral scores may predict prognosis, but our results do not support this idea. This discrepancy is probably
ur
because of cerebral edema, a state of confusion, and changes in arousal in the acute phase. Therefore, to determine VSN reliably, we used several examinations, with
Jo
three scale scores. Although the clock test is slightly different from the others, it is still not sensitive enough to detect pathological neglect in patients with hemianopia or constructional apraxia [19]. In some studies investigating VSN, the RT and error rate (or accuracy rate in our study) are thought to be the best criteria because of the functional relevancy of
these parameters to clinical practice [20]. Some data also have indicated that the speed to complete the paper-and-pencil test is the most important predictor for recovery at 2 years after a stroke [9]. However, our study found that the RT could be used to predict an increased perceptual expectation, but it was not a good predictor of recovery in the early stage. In contrast, the accuracy rate to the contralensional side could be related to the recovery in terms of high-order cognitive processing.
ro of
Unlike the behavioral RT and accuracy rate, measurements that cannot distinguish different component processes, the stages of processing that are influenced
by attention can be directly measured by the ERP. We chose to apply the cue-target
-p
paradigm in our study. We focused on analysis of the early component P1/N1 and the
re
late component P300 to verify the processing time. However, differences were not significant for either the N1 or P1 component, which represented the activity of the
lP
striate and extrastriate areas in our study. Interestingly, Verleger et al. have noted that
na
the N1 component evoked by left visual field stimuli was reduced at the right recording site [21]. In contrast, Francesco et al. failed to record the N1a component,
ur
but the P1 component was not significantly different [8]. For this reason, we detected three components under the valid cue and
Jo
The P300 component is a positive component of the ERP that peaks at 300 ms
or more (up to 900 ms) after a stimulus [22]. P300 is known to reflect neural processing associated with task-related information, and its latency is considered to be a measure of stimulus classification speed [23], which is generally unrelated to the overt response. Whereas P300 latency is sensitive to variables involved in stimulus
evaluation [24], it is feasible to use P300 as a measurement of cognitive function; and the practical and theoretical support indeed exist, since the P300 peak latency is negatively correlated with the mental function of normal subjects, i.e., shorter latencies are associated with superior cognitive performance from neuropsychological tests of attention and immediate memory [25]. The Posner paradigm involves manipulation of executive function demands not only in attention but also in
ro of
visuospatial processing [26], which can be more demanding than a simple task. The delay of the P300 latency in our study may have resulted from the Posner task, which not only reflects the global attentional processing but also the cognitive deficits in
-p
visuospatial attention.
re
Even though it is well known that P300 is modulated by attention, only a handful of studies have so far used it to investigate the role of attention in VSN,
lP
possibly because of the difficulty in recruiting subjects who are able to participate in
na
such experiments [27]. Research regarding P300 suggests an important link between attentional dysfunction and impaired stimulus processing in VSN, i.e., in the affected
ur
hemisphere, the difference in the P300 amplitude between the identified and missed targets was correlated with the miss rate [28]. However, our study did not find a
Jo
similar difference of amplitudes. In fact, the amplitudes of patients in our study were similar to those of the healthy controls. In this regard, it is of note that P300 is also modulated by the overall arousal level, which governs the amount of attention available for task performance. To the best of our knowledge, this work is the first to evaluate the
electrophysiological processes involved in the recovery from VSN at four weeks after stroke. In the past few years, researchers have tried to use EEG, the brain-computer interface (BCI), and other methods to explore ways to predict VSN recovery. For example, in a small study, Ros et al. confirmed the feasibility of EEG as a way to predict the recovery of patients with pathological neglect. They found that the resting-state alpha amplitude variability at the right posterior parietal cortex was the
ro of
principal predictor of VSN recovery by examining the EEG of five patients [29]. Moreover, Tonin et al. detected attention task-specific brain patterns in three VSN
patients by BCI and suggested that it is possible to use BCI to predict lateralized,
-p
attention-based VSN [30]. In contrast, we suggest that ERP can be used as a
re
predictive tool for recovery from VSN.
Furthermore, an asymmetrical omission toward the contralesional side in VSN
lP
and impairment in the temporal processing capacity may represent a good marker of
na
recovery. Our findings are consistent with a report suggesting that an inability of capturing attention and a slow speed of visual processing are not only key symptoms
ur
of VSN, but they are also important predictors of functional dependency [9]. However, some limitations of this study should be mentioned. First, this study
Jo
was a single-center study with a relatively small number of patients. Second, the P300 component may be confounded by some clinical variables such as dietary and circadian factors [31]. Although we tested all of the participants during the same time frame, the ERP data may display individual differences and some participants could not be evaluated for long periods of time due to their general health status.
CONCLUSIONS We used ERP with exceptional temporal resolution to show direct evidence of temporal processing and recovery in patients at four weeks after a stroke. We concluded that the P300 latency might be an index for speed and efficiency of information processing in the brain. This study contributes to the knowledge and
ro of
understanding of the functional impairment and recovery in patients with VSN.
Funding
-p
This research was supported by the National Natural Science Foundation of China
re
(grant no. 81671048). Conflict of Interest
None.
Heilman KM, Valenstein E, Watson RT, Neglect and related disorders, Seminars
Jo
1.
ur
References
na
Acknowledgments
lP
The authors declare that they have no conflicts of interest.
in neurology. 20 (2000) 463-70. https: //doi. 10.1055/s-2000-13179 2.
Stone SP, Halligan PW, Greenwood RJ, The incidence of neglect phenomena and
related disorders in patients with an acute right or left hemisphere stroke, Age and ageing. 22 (1993) 46-52. 3.
Paolucci S, Antonucci G, Grasso MG, Pizzamiglio L, The role of unilateral
spatial neglect in rehabilitation of right brain-damaged ischemic stroke patients: a matched comparison, Archives of physical medicine and rehabilitation. 82 (2001) 743-9. 4.
Katz N, Hartman-Maeir A, Ring H, Soroker N, Functional disability and
rehabilitation outcome in right hemisphere damaged patients with and without unilateral spatial neglect, Archives of physical medicine and rehabilitation. 80 (1999) 379-84. Nijboer T, van de Port I, Schepers V, Post M, Visser-Meily A, Predicting
ro of
5.
functional outcome after stroke: the influence of neglect on basic activities in daily living,
Frontiers
in
human
neuroscience.
7
(2013)
6.
https:
//doi.
-p
10.3389/fnhum.2013.00182
182.
Husain M, Shapiro K, Martin J, Kennard C, Abnormal temporal dynamics of
re
visual attention in spatial neglect patients, Nature. 385 (1997) 154-6. https: //doi.
7. of
lP
10.1038/385154a0
Corbetta M, Shulman GL, Spatial neglect and attention networks, Annual review neuroscience.
34
(2011)
569-99.
https:
//doi.
8.
na
10.1146/annurev-neuro-061010-113731
Di Russo F, Aprile T, Spitoni G, Spinelli D, Impaired visual processing of
ur
contralesional stimuli in neglect patients: a visual-evoked potential study, Brain : a
Jo
journal of neurology. 131 (2008) 842-54. https: //doi. 10.1093/brain/awm281 9.
Viken JI, Jood K, Jern C, Blomstrand C, Samuelsson H, Ipsilesional bias and
processing speed are important predictors of functional dependency in the neglect phenomenon after a right hemisphere stroke, The Clinical neuropsychologist. 28 (2014) 974-93. https: //doi. 10.1080/13854046.2014.950335 10. van Kessel ME, van Nes IJ, Brouwer WH, Geurts AC, Fasotti L, Visuospatial
asymmetry and non-spatial attention in subacute stroke patients with and without neglect, Cortex; a journal devoted to the study of the nervous system and behavior. 46 (2010) 602-12. https: //doi. 10.1016/j.cortex.2009.06.004 11. Husain M, Rorden C, Non-spatially lateralized mechanisms in hemispatial neglect, Nature reviews Neuroscience. 4 (2003) 26-36. https: //doi. 10.1038/nrn1005 12. Samuelsson H, Hjelmquist EK, Jensen C, Ekholm S, Blomstrand C, Nonlateralized attentional deficits: an important component behind persisting
ro of
visuospatial neglect?, Journal of clinical and experimental neuropsychology. 20 (1998) 73-88. https: //doi. 10.1076/jcen.20.1.73.1481
13. Posner MI, Orienting of attention, The Quarterly journal of experimental
-p
psychology. 32 (1980) 3-25.
14. Umarova RM, Reisert M, Beier TU, Kiselev VG, Kloppel S, Kaller CP, et al.,
re
Attention-network specific alterations of structural connectivity in the undamaged
10.1002/hbm.22503 15.
lP
white matter in acute neglect, Human brain mapping. 35 (2014) 4678-92. https: //doi.
Lee BH, Kang SJ, Park JM, Son Y, Lee KH, Adair JC, et al., The Character-line
1715-24.
na
Bisection Task: a new test for hemispatial neglect. Neuropsychologia. 42 (2004)
ur
16. Farne A, Buxbaum LJ, Ferraro M, Frassinetti F, Whyte J, Veramonti T, et al.,
Jo
Patterns of spontaneous recovery of neglect and associated disorders in acute right brain-damaged patients, Journal of neurology, neurosurgery, and psychiatry. 75 (2004) 1401-10. https: //doi. 10.1136/jnnp.2002.003095 17. Cassidy TP, Lewis S, Gray CS, Recovery from visuospatial neglect in stroke patients, Journal of neurology, neurosurgery, and psychiatry. 64 (1998) 555-7. 18. Ramsey LE, Siegel JS, Baldassarre A, Metcalf NV, Zinn K, Shulman GL, et al.,
Normalization of network connectivity in hemispatial neglect recovery, Annals of neurology. 80 (2016) 127-41. https: //doi. 10.1002/ana.24690 19. Agrell BM, Dehlin OI, Dahlgren CJ, Neglect in elderly stroke patients: a comparison of five tests. Psychiatry Clin Neurosci. 51 (1997) 295-300. http://dx.doi.org/10.1111/j.1440-1819.1997.tb03201.x. 20. Polich J, Clinical application of the P300 event-related brain potential, Physical medicine and rehabilitation clinics of North America. 15 (2004) 133-61.
ro of
21. Verleger R, Heide W, Butt C, Wascher E, Kompf D, On-line brain potential correlates of right parietal patients' attentional deficit, Electroencephalography and clinical neurophysiology. 99 (1996) 444-57.
-p
22. Linden DE, The p300: where in the brain is it produced and what does it tell us?, The Neuroscientist : a review journal bringing neurobiology, neurology and
re
psychiatry. 11 (2005) 563-76. https: //doi. 10.1177/1073858405280524
lP
23. Magliero A, Bashore TR, Coles MG, Donchin E, On the dependence of P300 latency on stimulus evaluation processes, Psychophysiology. 21 (1984) 171-86. 24. Duncan-Johnson CC, Donchin E, The P300 component of the event-related brain
na
potential as an index of information processing, Biological psychology. 14 (1982) 1-52.
ur
25. Reinvang I, Cognitive event-related potentials in neuropsychological assessment,
Jo
Neuropsychology review. 9 (1999) 231-48. 26. Theeuwes J, Top-down and bottom-up control of visual selection, Acta psychologica. 135 (2010) 77-99. https: //doi. 10.1016/j.actpsy.2010.02.006 27. Deouell LY, Hamalainen H, Bentin S, Unilateral neglect after right-hemisphere damage: contributions from event-related potentials, Audiology & neuro-otology. 5 (2000) 225-34. https: //doi. 10.1159/00001388428. Saevarsson S, Kristjansson A,
Bach M, Heinrich SP, P300 in neglect, Clinical neurophysiology : official journal of the International Federation of Clinical Neurophysiology. 123 (2012) 496-506. https: //doi. 10.1016/j.clinph.2011.07.028 29. Ros T, Michela A, Bellman A, Vuadens P, Saj A, Vuilleumier P, Increased Alpha-Rhythm Dynamic Range Promotes Recovery from Visuospatial Neglect: A Neurofeedback Study. Neural Plasticity. 2017 (2017) 1-9. 30. Tonin L, Pitteri M, Leeb R, Zhang H, Menegatti E, Piccione F, Millán JDR, et al.,
ro of
Behavioral and Cortical Effects during Attention Driven Brain-Computer Interface Operations in Spatial Neglect: A Feasibility Case Study. Frontiers in Human Neuroscience, 2017 11.
-p
31. Polich J, Kok A, Cognitive and biological determinants of P300: an integrative
re
review, Biological psychology. 41 (1995) 103-46.
lP
Figure 1. Flow chart of the sequential assessment of the patients included in this
Jo
ur
na
study.
ro of -p
ur
na
lP
re
Figure 2 Schematic view of the experimental paradigm.
Jo
Figure 3. The bar graphs depict the accuracy rate (a) and the response time (RT) (b) of the control, persistent visual spatial neglect (P-VSN), and rapid recovery visual spatial neglect (R-VSN) groups in the contexts of a valid (top panels of a&b) or an invalid target (bottom panels of a&b) as well as a left-cue (left panels of a&b) or a right-cue target (right panels of a&b). *p < 0.05, **p < 0.01, ***p < 0.001.
ro of
Figure 4. Typical average waveforms of event-related potentials from electrode (P300, i.e., P3&P4, P1N1, i.e., PO7&PO8) recordings of a valid left target on the scalp. The
-p
control, rapid recovery visual spatial neglect (R-VSN), and persistent visual spatial
na
lP
re
neglect (P-VSN) groups are indicated by blue, green, and red lines, respectively.
ur
Figure 5. Typical average waveforms of event-related potentials from electrode (P300,
Jo
i.e., P3&P4, P1N1, i.e., PO7&PO8) recordings of a valid right target on the scalp. The control, rapid recovery visual spatial neglect (R-VSN), and persistent visual spatial neglect (P-VSN) groups are indicated by blue, green, and red lines, respectively.
neglect or rapid recovery visual spatial neglect
Pathoge ny
P1 P2 P3 P4 P5 P6 P7 P8 P9 R1 R2 R3 R4 R5 R6 R7 R8 R9
M M F M M M F M M M M M M M M M M M
39 66 52 32 57 20 69 54 51 61 61 53 61 31 68 49 47 56
CH CI CI CH CI CH CH CI CI CI CI CI CI CI CI CI CI CH
SC CSC C CSC CSC C C CSC CSC SC CSC SC CSC CSC SC C C C
25.27 39.71 26.27 69.93 21.38 14.55 78.16 26.29 18.75 14.22 13.79 17.62 15.16 22.65 28.25 14.49 19.14 40.18
0.33 5.00 1.67 4.33 0.11 1.29 9.00 0.50 0.50 0.50 0.33 0.33 0.11 1.67 0.50 0.71 0.71 0.21
+ + + + + + + -
na
ur
Jo
Clock drawi ng
Gap detecti on
Star cancellat ion
Senten ce readin g
+ + + + + + + + -
0.74 1.80 1.47 7.68 1.74 1.74 7.68 2.89 2.89 0.50 0.50 1.47 0.13 1.74 2.89 6.61 1.80 0.88
+ + + + + + + + + + + -
-p
Ag e
Line cancellat ion
re
Se x
Line bisectio n%
lP
Patie nt
Lesi on type
ro of
Table 1. Demographic and general clinical data of the patients with persistent visual spatial
CH, cerebral hemorrhagic; CI, cerebral ischemic; C, cortical; SC, subcortical; CSC, cortical-subcortical; +, positive; -, negative; F, female; M, male; P, persistent visual spatial neglect; R, rapid recovery visual spatial neglect.
Table 2. Amplitude and latency of the P300 component Amplitude (μV)
Latency (ms)
Left Target Left hemisph ere
Right hemisph ere
Right Target
Left Target
Right Target
Left hemisph ere
Right hemisph ere
Left hemisph ere
Right hemisph ere
Left hemisph ere
Right hemisph ere
497.15 ± 87.31
425.78 ± 87.50
6.48 4.42
± 8.41 4.55
±
4.54 4.88
±
6.46 4.39
±
521.48 ± 95.93
529.07 ± 87.82
Rapid recove ry VSN group
8.10 4.49
± 5.93 3.55
±
5.98 4.01
±
5.64 3.69
±
434.56 ± 41.64
428.56 ± 37.42
Contro l group
9.02 4.85
± 10.58 ± 4.72
9.79 5.31
±
10.17 ± 366.43 4.84 ± 23.93
-p
re lP na ur Jo
ro of
Persist ent VSN group
359.27 ± 25.07
435.22 ± 51.59
433.11 ± 87.84
386.10 ± 50.88
373.70 ± 34.57