ARTICLE IN PRESS

Cognitive Outcomes following Thrombolysis in Acute Ischemic

Stroke: A Systematic Review
Laura Jayne Broome, BSc (Hons),* Ceri Elisabeth Battle, PhD,*
Matthew Lawrence, PhD,* Phillip Adrian Evans, MD, FRCEM,* and
Michael Stuart Dennis, MD, FRCPsych

Background: Patients treated with thrombolytic therapy within 4.5 hours after stroke
onset appear to have improved survival and functional outcomes. Poststroke cognitive impairment is associated with reduced quality of life and survival and needs
to be reviewed in consideration of the administration of thrombolysis. This review
aims to systematically evaluate literature exploring the effect of thrombolysis for
ischemic stroke on cognition. Methods: An electronic search was conducted to identify articles and gray literature applying broad Medical Subject Heading terms.
Literature was reviewed with a 2-step process against predetermined inclusion
criteria. All relevant studies were included if they investigated global or individual cognitive domains. Results: Three studies satisfied the inclusion criteria but
were diverse in outcome measures and duration, their heterogeneity limiting any
possible pooled analysis. One study examined long-term treatment effects on global
cognition and did not find a positive effect at 6 months. A positive treatment effect
was reported in the acute phase in 1 study examining domains of visuoconstructive
and perceptive abilities. One study retrospectively analyzed treatment effects on
language and found improvement in the acute phase but not in the long term.
Conclusions: The limited existing evidence on the effects of thrombolytic therapy
on long- and short-term cognition is varied in both outcome measures and diagnostic classifications, making it difficult to extrapolate results to a global stroke
population. This review should be used to inform future research in stroke treatment outcomes and highlights the immediate need for larger, more robust studies
in this area. Key Words: Strokecerebrovascular disorderthrombolytic
therapyrecombinant tissue plasminogen activatorcognitive disordercognitive
impairmentneurological impairment.
2016 National Stroke Association. Published by Elsevier Inc. All rights reserved.

From the *NISCHR Haemostasis Biomedical Research Unit, Emergency Department, Morriston Hospital, Swansea, United Kingdom;
and Psychiatry of Older People, College of Medicine, Swansea University, Swansea, United Kingdom.
Received April 18, 2016; revision received July 22, 2016; accepted
July 30, 2016.
Grant support: This study was supported by the National Institute for Social Care and Health Research (NISCHR).
Address correspondence to Laura Jayne Broome, BSc (Hons),
NISCHR Haemostasis Biomedical Research Unit, Emergency
Department, Morriston Hospital, Swansea SA6 6NL, United Kingdom.
E-mail: 493135@swansea.ac.uk.
1052-3057/$ - see front matter
2016 National Stroke Association. Published by Elsevier Inc. All
rights reserved.
http://dx.doi.org/10.1016/j.jstrokecerebrovasdis.2016.07.048

Introduction
Stroke is the second primary cause of death and the
most common life-threatening neurological disorder
worldwide.1 One third of patients suffer from dementia
in the first year following stroke, with 60% of patients
experiencing some form of cognitive decline.2 Poststroke
cognitive impairment (PSCI) can improve during followup in approximately 16%-20% of elderly patients3,4 but
is associated with increased risk of functional decline and
mortality.5 If deficits are categorized as global and severe
enough to satisfy dementia criteria, the risk of recurrent
stroke, dependency, and mortality are further increased
with resulting social and economic costs.6

Journal of Stroke and Cerebrovascular Diseases, Vol. , No. (), 2016: pp

ARTICLE IN PRESS
L.J. BROOME ET AL.

Treatment with intravenous recombinant tissue plasminogen activator (rt-PA) in the acute stage of ischemic
stroke is approved worldwide7-9 and has been seen to
improve survival and functional outcomes when administered within 4.5 hours of onset.10-12 The benefits of rtPA are thought to be due to the early recanalization of
occluded arteries, reducing the amount of tissue at risk
of infarction.13 The reduction of lesion volume resulting
from increased cerebral reperfusion is considered to be
one of the best predictors of a good outcome following
ischemic stroke.14 A recent systematic review examining
the safety and efficacy of thrombolytic therapy for acute
ischemic stroke strengthens previous evidence, and indicates that thrombolysis reduces the risk of mortality
and dependency despite the risk of symptomatic intracranial hemorrhage.15 Theoretically, thrombolytic drugs
may therefore also improve cognitive outcomes and reduce
neurological deficits by restoring blood flow in occluded
vessels.16 The impact of thrombolysis on PSCI warrants
detailed review to improve knowledge around the treatment and management of such patients. The main objective
of the current review is to evaluate the effect of thrombolytic therapy after ischemic stroke on cognitive outcomes.

Table 1. Systematic search strategy

Search
stage
1
2
3
4
5
6
7
8
9
10
11
12
13
14

Strategies
Stroke/ or cerebrovascular disorders/ or brain
ischemia/
Thrombolytic therapy/ or fibrinolysis/ or
plasminogen/ or fibrinolysin/ or plasmin.mp.
(Blood clot lysis or alteplase or urokinase).mp.
Streptokinase/ or fibrinolytic agents/ or tissue
plasminogen activator/ or rtpa.mp. or rt-pa.mp.
Plasminogen activators/ or recombinant tissue
activator.mp.
2 or 3 or 4 or 5
1 and 6
Limit 7 to (English language and humans)
Cognition disorders/ or cognitive impairment.mp.
or neurological impairment.mp.
Dementia/ or delirium/ or confusion/ or acute
confusion.mp. or chronic confusion.mp.
Quality of life/ or qol.mp. or quality of life.mp.
Neuropsychological tests/ or neuropsychological
behaviour.mp. or neurocognitive tests.mp.
9 or 10 or 11 or 12
8 and 13

Methods
An extensive electronic search was conducted on the
Cochrane Library, PubMed and MedLine, Embase, and
PsychInfo databases. The search terms were Medical Subject
Heading terms covering stroke, thrombolytic treatment,
and cognitive outcomes (Table 1), and literature was limited
to stroke trials published in English on human subjects
between 1990 and July 2015.
An electronic search was also conducted for relevant
peer-reviewed journals (Stroke, Journal of Neurology, Neurology, Age and Ageing, The Lancet/The Lancet Neurology,
and Journal of Stroke & Cerebrovascular Diseases), reference lists of appropriate papers, and international
conference proceedings on stroke and thrombolysis. Conference abstracts from the International Stroke Conference,
Journal of Neurology, and Age and Ageing were searched,
and gray literature was examined by searching OpenGrey,
which provides access to European gray literature between
1980 and 2005.
An initial search was conducted by means of a 2-step
selection process in which 2 reviewers (L.B. and C.B.) independently reviewed the titles and abstracts to exclude
obviously irrelevant articles, reducing the risk of selection bias. In a second review, reviewers (L.B. and C.B.)
considered full papers against the inclusion and exclusion criteria shown in Table 2. Data were extracted from
full papers to determine if they met the inclusion criteria. Uncertainty or a difference of opinion was resolved
through discussion between the 2 reviewers (L.B. and C.B.).
If no agreement could be reached, discussions were referred to a third person (M.D.). A pilot data extraction

form was designed using guidelines in the Preferred Reporting Items for Systematic Reviews and Meta-Analyses
(PRISMA) statement and piloted on a sample of 10 articles to ensure the inclusion criteria could be applied
generally and consistently. Studies were selected based
on the participants, interventions, comparisons, outcomes, and study design (PICOS) formulation and
hierarchical models of analysis to examine group effects
and multiple observations of the same outcome (Table 3).
The primary outcome measure of the present review
was cognition. Cognitive impairment could be measured by either a validated instrument of global cognitive
function (such as the Mini-Mental State Examination, Montreal Cognitive Assessment, or Abbreviated Mental Test
Score) or instruments examining individual cognitive
domains such as executive function, memory, language,
and visuospatial and constructional abilities.
Assessment of risk bias was independently assessed
by the reviewers using the Cochrane Collaboration tool
for assessing risk of bias.17 Any differences of opinion
between the 2 reviewers were resolved by way of
discussion.
Due to methodological heterogeneity between the studies,
we were unable to compare studies and perform a
meta-analysis.

Results
Figure 1 indicates the process of study selection. The
search yielded a paucity of randomized controlled trials

ARTICLE IN PRESS
COGNITIVE OUTCOMES FOLLOWING THROMBOLYSIS

Table 2. Study inclusion and exclusion criteria

Inclusion

Exclusion

First-stage review
Studies that recruited adults 18 years and over
with confirmed ischemic stroke
Studies that examined outcomes of all types of
thrombolytic drugs associated with ischemic
stroke given either intravenously or intraarterially, to include rt-PA, u-PA, and SK
All observational studies
Second-stage review
Studies investigating cognitive outcomes by
a. A validated instrument of global cognition
b. Instruments examining individual cognitive domains (executive
function, memory, language, and visuospatial and
constructional abilities)
Studies examining cognitive-related QoL

Studies examining hemorrhagic stroke and animal studies

Studies examining other forms of treatment, including
antiplatelet therapy
If no comparison group was included, such as case studies
or review articles
Retrospective studies that did not carry out cognitive tests
using validated instruments
Studies that aimed to assess individual domains using
NIHSS only
Studies investigating
a. Functional outcomes only
b. Mortality only
c. Health-related QoL only

Abbreviations: NIHSS, National Institutes of Health Stroke Scale; QoL, quality of life; rt-PA, recombinant tissue plasminogen activator;
SK, streptokinase; u-PA, urokinase.

(RCTs) and a diverse range of study types and outcome

measures. Tabulated data showing the heterogeneity of
included studies are presented in terms of study characteristics and outcome measures in Table 4. Five published
articles were identified as meeting the inclusion criteria18-22;
3 papers separately reported subsets and summaries of
results from the same group of patients19-21 and are
therefore reported as 1 study for the purpose of this
review.
Studies recruited first-ever symptomatic ischemic stroke
patients diagnosed by computed tomography or magnetic resonance imaging. Treatment was administered
depending on the protocol and standard clinical practice at participating hospitals, with standard administration
of rt-PA given within 3 hours of stroke onset. Left-

hemisphere stroke patients were excluded from 1 study

assessing the effect of thrombolysis on visuoconstructive
and visuoperceptive abilities known to be affected by righthemisphere (RH) infarcts.19-21 One study recruited only
left middle cerebral artery infarct patients with aphasia.22
Patients over the age of 80 were excluded from 1 study19-21
and patients aged 85 were excluded from another study18
due to aging effects on cognition. All studies excluded
patients with previous stroke, other brain injury, and psychiatric disorders. Stroke severity was assessed by the
National Institutes of Health Stroke Scale (NIHSS) on admission and at follow-up. In assessing aphasia, those
patients with an NIHSS score higher than 24 or with any
other neurological condition that could cause speech disorders were also excluded.22 Patients were split into 2

Table 3. PICOS formulation of study characteristics

Formulation
Participant
Interventions
Comparisons

Outcomes

Study design

Characteristics
Number recruited, age, sex, stroke location, treatment received in addition to intervention, difference in
baseline NIHSS scores between intervention and control group, subgroups measured
Type of intervention (including means of administration and dosage) and time administered
Comparisons will be made between the intervention and control group to assess change from baseline and
stated time points. Statistical methods and their appropriateness will be considered where appropriate as
well as the need for reanalysis
To include a primary outcome of cognition (global functioning, executive functioning, memory, language,
and visuospatial and visuoconstructive abilities) and a secondary outcome of cognitive-related QoL;
assessment of tools used, including whether tools are validated and time points assessed
Study design (RCTs, prospective, retrospective)

Abbreviations: NIHSS, National Institutes of Health Stroke Scale; QoL, quality of life; RCT, randomized controlled trial.

ARTICLE IN PRESS
L.J. BROOME ET AL.

4
Citations identified through database search n = 3478
Citations identified from article citations n = 23
Total citations n = 3501

Excluded after screening titles and abstracts n = 3362

Ineligible disease state n = 1096
Ineligible intervention n = 497
Inappropriate design n = 1383
No clinically relevant outcomes n = 386

Figure 1. PRISMA flow diagram representing the

process of study selection.

Full papers retrieved n = 139

Excluded following detailed evaluation n = 134

No clinically relevant outcomes n = 111
Inappropriate design n = 23

Total Papers included n = 5

Total studies included n = 3 (3 multiple publications from 1 study)

groups (treated and nontreated) based on the clinical treatment decision made by the recruiting hospital.
Risk of bias was assessed across the studies; none of
the included studies were RCTs and were therefore at
risk of sequence and allocation bias. None of the studies
reported measures of blinding when analyzing outcome
measures and were therefore at risk of performance bias.
No other bias was identified. Limitations to the studies
are addressed in the discussion.

Global Cognition
Nys et al18 was the first to examine treatment effects
of rt-PA on global cognition in ischemic stroke patients.
The study by Nys et al was primarily a prospective cohort
analytic study but additionally included a casecontrol
element. Seven cognitive domains were assessed between
6 and 12 months from onset using the tools outlined in
Table 4. Outcomes were categorized and presented as crude
and adjusted odds ratio (OR) with 95% confidence intervals (CIs) with logistical regression. In measuring
cognitive outcomes, patients were defined as cognitively
intact versus cognitively impaired (in at least 1 domain),
and severity of impairment was assessed by the
unweighted average of the 7 domain scores. Functional
status was assessed by a measure of activities of daily
living (ADLs). The modified Barthel Index was used to
assess basic ADL, which assesses basic functions such as

hygiene and dressing with a value of 19 or lower being

indicative of impaired function. Instrumental ADL assesses more complex activities such as social activities,
shopping, and household management, and was assessed by the Frenchay Activity Index with an overall
score of 15 or lower being defined as impaired. A control
group was matched for age, education, and gender
and was used as a reference sample for neurological
testing.
No significant differences were observed between treated
and nontreated patients in any cognitive domain at 6
months (OR = 1.0, 95% CI: .2-4.3). Severity of impairment was not influenced by treatment of rt-PA, with a
reported adjusted difference of 1.0 (95% CI: .4-.6). The
study did observe a treatment effect in basic (OR = 13.5,
95% CI: 1.4-129.4) and instrumental ADL (OR = 7.1, 95%
CI: 1.2-42.2) at 6 months in line with previous research.15
To further explore any confounding effects on stroke severity, 25 nontreated patients with the highest NIHSS scores
where compared with 25 treated patients with the lowest
NIHSS score. Again, no significant differences were found
in treatment effect on cognition between the groups
(OR = .9) with a bigger effect seen in basic (OR = 47.1)
and instrumental ADL (OR = 12.7).
Nys et al reports that thrombolysis may have shortterm improvements in focal cognition but is not sufficient
to improve cognition at 6 months. Furthermore, rt-PA treatment may lead to improved independence following

Study ID
Global cognition
Nys et al18

N/age

Study design

T+, 25; M, 59.9

T, 67; M, 61.7
C, 75; matched

Prospective cohort
and casecontrol
study

Methods

Visual perception
and construction
(multiple
publications)
Laihosalo et al19
Kettunen et al20
(RW bias)
Kettunen et al21

Language
Jacquin et al22

T+, 28; Md, 63.5

T, 28; Md, 64
T+, 34; M, 60.5
T, 43; M, 62
C, 62; M, 56.2
T+, 34; Md, 60.5
T, 43; Md, 62

Matched case
control study
prospective cohort
and casecontrol
study

T+, 37; M, 68
T, 38; M, 66

Casecontrol study

Prospective cohort
study

Language (NIHSS, Boston Diagnostic Aphasia Short Form)

rt-PA treatment influences basic (OR = 13.5, 95% CI:

1.4-129.4) and instrumental functional outcomes
(OR = 7.1, 95% CI: 1.2-42.2), but not cognitive
outcomes at 6 months (OR = 1.0, 95% CI: .2-4.3).

T patients had higher odds of visual neglect 4 days

after stroke onset (OR = 4.366, 95% CI: .99419.175; P = .05). A positive effect was observed in
visuoconstructive functioning (P = .002) with a
mean of 4 days post stroke, with no significant
differences between the groups in immediate
(P = .180) or delayed (P = .0508) visual memory
and reasoning (P = .454).
A difference in attentional processing was observed
between healthy controls and treated patients in all
cancellation tasks 4 days after onset. Additionally,
defective rightward orientation bias was more
evident in T patients with a significant difference
between the groups in the line cancellation task
(P = .05). No significant difference was observed
in the star cancellation test (P = .08) or letter
cancellation.
rt-PA was found to be significantly associated with a
favorable outcome of aphasia at 1 week (P = .028)
but not at 3 months (P = .872). Trend analysis
reports less severe aphasia in T+ patients at 3
months (P = .011). Recovery of previous activities
was associated with conduction and atypical and
mild aphasia at 6 months and 1 year.

Abbreviations: ADL, activity of daily living; BADS, Behavioural Assessment of the Dysexecutive Syndrome; BI, Barthel Index; BITC, Behavioural Inattention Test Conventional; C, control; CI, confidence interval; M,
mean; MD, median; MMSE, Mini-Mental State Examination; NIHSS, National Institutes of Health Stroke Scale; OR, odds ratio; rt-PA, recombinant tissue plasminogen activator; T, nontreated patients; T+, treated patients; WAIS-II, Wechsler Adult Intelligence Scale; WMS, Wechsler Memory Scale.

ARTICLE IN PRESS

Stroke status (NIHSS)

Functional status (BI)
Abstract reasoning (WAIS-II)
Verbal memory (Rey Auditory Verbal Learning, WAIS-II)
Executive functioning (Brixton Spatial Anticipation Test, Visual
Elevator, Stroop Color and Word Test, Zoo Test, BADS)
Visuoperception/construction (Judgment of Line Orientation, Test
of Facial Recognition, WAIS-III (block design))
Visual memory (WMS-R (visual reproduction), Location Learning
Task, Corsi Block Span)
Language (Token Test, Boston Naming Test, Chapman reading task)
Unilateral neglect (BITC (star cancellation))
Stroke status, hemiparesis, hemianopia (NIHSS)
Visuoconstructive ability (WAIS-R (block design))
Visual search and reasoning (WAIS-R (picture completion))
Visual memory (WMS-R (visual reproduction))
Visual neglect (BITC Subtests)
Attentional processes (BITC (line, letter, and star cancellation))
Global cognition (MMSE)
Basic ADL (BI)

Results

COGNITIVE OUTCOMES FOLLOWING THROMBOLYSIS

Table 4. Tabulated study characteristics and outcome measures of included studies

ARTICLE IN PRESS
L.J. BROOME ET AL.

discharge in more complex activities, that is, household

management and food shopping.

Visual Perception and Construction

This was the first prospective cohort study with a case
control element to assess whether visuoperceptive and
visuoconstructive functions differ between treated and
nontreated patients in the acute stage of RH stroke.19-21
Neuropsychological examinations were carried out with
a mean of 4 days after onset to assess visuoconstructive
abilities, visual search and reasoning, visual neglect (VN),
visual memory, and pathological attention processes
(Table 4). NIHSS was used on admission and with a mean
of 4 days after onset. Treated and nontreated patient groups
were matched for stroke severity, age, and education.
The groups did not significantly differ in stroke severity in baseline NIHSS score; however, nontreated patients
had more severe strokes 4 days after onset (P = .017,19
P = .00920,21). Neuropsychological test results (Table 4) were
converted into z scores centered on the scores from a
matched healthy control group and dichotomized.19
Visuoperceptual abilities in each domain (visuoconstructive
abilities, visual search and reasoning, and visual reproduction) were considered impaired in stroke patients with
a score of 1 SD or lower below the mean performance
of the control group. Additionally, patients scoring below
129 or lower on at least two of the Behavioural Inattention Test Conventional (BITC) subtests with a mean of
4 days after onset were viewed as having VN.21 Visual
attention deficits (rightward orientation bias) were assessed by analyzing the starting points in 3 BITC
cancellation tasks, comparing results between treated and
nontreated patients and healthy controls.20
Treated patients performed significantly better in
visuoconstructive functioning (P = .002), with no significant difference in the presence of neglect between the
groups (P = .453).19 No significant differences were found
in immediate (P = .180) or delayed visual memory (P = .508)
and visual reasoning (P = .454), although fewer treated
patients experienced deficits in these examinations.19
Subgroup analysis examining the presence of VN after
thrombolysis found that although VN was present in a
higher proportion of nontreated patients, the difference
was not statistically significant (P = .168).21 After adjusting for confounding factors, logistic regression found that
thrombolysis independently predicts the absence of neglect,
with nontreated patients having higher odds of deficit
(OR = 4.366, 95% CI: .994-19.175; P = .05).21
In examining attentional processes between the groups,
starting points in 3 cancellation points were compared
using cross-tabulations.20 When comparing healthy controls to treated patients, a statistical difference was found
between the groups, with treated patients illustrating a
stronger rightward bias tendency in all cancellation tasks
(line, P = .015; letter, P = .005; star cancellation tasks,

P = .000). This difference was significant regardless of the

presence of VN. Results between stroke patient groups
found that nontreated patients had a tendency toward
rightward orientation bias in the cancellation tasks, with
a significant difference in the line cancelation task (P = .05)
and a reported trend toward significance for the star cancellation test (P = .08). No significant difference was found
in the letter cancellation task. Results are suggested to
be indicative of mild or residual inattention in the acute
phase of stroke.20

Language
One retrospective casecontrol study analyzed the effect
of rt-PA on aphasia outcome and compared the recovery of previous activities at 6 and 12 months on left middle
cerebral artery infarct patients. 22 Language was assessed on admission using the NIHSS and tested 1 week
and 3 months after onset using the short form of the Boston
Diagnostic Aphasia Examination. No significant differences were found in NIHSS scores between the groups
(P = .050). Two baseline characteristics differed in the acute
phase; treated patients were significantly more likely to
have a mutism (OR = .12, 95% CI: .04-.40; P = .002) compared to fluent or nonfluent aphasia (OR = .22, 95% CI:
.06-.77; P = .018), and to be admitted to a rehabilitation
center (OR = 7.18, 95% CI: 2.01-25.66; P = .002).
Aphasia was significantly less severe in treated patients 1 week after onset (P = .028) but not at 3 months
(P = .872). Trend analysis using likelihood ratios was carried
out to examine differences in aphasia severity at 3 months,
which found aphasia in treated patients appeared to be
less severe (P = .011). Multivariate analysis found thrombolysis to be associated with a favorable outcome on the
language assessments at both 1 week (P = .002) and 3
months (P = .001). Bivariate analysis found conduction and
mild atypical aphasia to be more frequent in treated patients during the first week after stroke (P = .033), with
no significant difference between the groups at 3 months
(P = .665). 22 Improved recovery at both 6 months
(OR = 12.96, 95% CI: 1.56-107.48; P = .002) and 1 year
(OR = 13.84, 95% CI: 1.67-114.86; P = .015) was associated with conduction and atypical aphasia. Overall, the
present study reports that aphasia in patients treated with
thrombolytic therapy was less severe 1 week after stroke
onset, with trend analysis suggesting improvements at
3 months.22

Discussion
The main aim of the present systematic review was to
summarize the effect of thrombolysis after acute ischemic stroke on cognition. To our knowledge, this paper
is the first to systematically review studies that have assessed the treatment effect of thrombolysis on cognition
using validated instruments, and highlights how research in this area continues to focus on functional

ARTICLE IN PRESS
COGNITIVE OUTCOMES FOLLOWING THROMBOLYSIS

outcomes, mortality, and symptomatic intracranial hemorrhage. The 3 studies selected were of variable
methodological quality and comprised a mix of prospective cohort and casecontrol studies. Because of the lack
of quality RCTs, it was decided to include all 3 identified studies regardless of their quality and to discuss their
limitations.
Overall results from included studies were conflicting
and showed high levels of heterogeneity. Only 1 study
examined long-term cognitive outcomes of both RH and
left-hemisphere stroke patients and found that thrombolysis did not have a positive treatment effect on cognition
at 6 months.18 Nys et al18 suggests that reducing lesion
size with thrombolysis may not be sufficient to improve
long-term cognitive functioning, with other factors such
as lesion location or volume hypoperfusion playing a more
important role in recovery. Alternatively, it is also suggested that the neurotoxic properties of rt-PA may cause
long-term dysfunction in some patients regardless of any
improvement in focal cognition in the acute phase.18 It
is important to note that Nys et al did not assess shortterm cognitive functioning and only assessed stroke severity
on admission. Additionally, the heterogeneity found
between treated and nontreated patients in the baseline
NIHSS score may have caused a bias in treatment effects.
The study aimed to address this confounding statistically by carrying out a risk adjustment on patients with the
highest and lowest NIHSS scores returning the same results,
suggesting that stroke severity cannot entirely explain the
lack of treatment effect.
Short-term effects of thrombolysis on cognition were
examined through assessment of individual domains of
cognition with a mean of 4 days after stroke.19-21 Unlike
Nys et al, a positive treatment effect was observed in shortterm cognition in terms of VN,21 attentional processing,20
and visuoconstructive abilities.19 These results indicate that
recovery in residual pathological attention processes may
be incomplete in the acute stage of stroke in nontreated
patients. Nys et als18 suggestion that thrombolysis does
not have a positive effect on long-term cognition is supported by Laihosalo et al,19 who proposed that differences
in long-term cognition are not sustained between the
groups due to the gradual recovery of nontreated patients. This conclusion must be considered cautiously as
Laihosalos study did not investigate long-term recovery processes or global cognition. Ischemic stroke may
lead to domain-specific impairments such as aphasia and
neglect, but also more global impairments, which need
to be explored to support this statement. Additionally,
investigating visuoperception in RH stroke patients only19-21
makes it difficult to extrapolate results to a global stroke
population.
A retrospective study analyzing aphasia outcome after
thrombolysis concluded that the severity of aphasia was
significantly milder in treated patients 1 week after onset
but not at 3 months, although a trend analysis at 3 months

was significant, suggesting that the probability of aphasia

may be higher in nontreated patients.22 The present study
evaluated speech using standardized clinical tests, but is
a retrospective casecontrol study and therefore vulnerable to a number of potential biases, including selection
bias, differential misclassification bias, and inadequate
power.
The present review failed to identify any RCTs. Identified articles based their inclusion of treated patients on
the clinical practice of the recruiting hospital and showed
a number of limitations. First, there was considerable heterogeneity between the studies. No standardized definition
for cognitive impairment or means of assessment post
stroke exists and therefore studies use a range of tests to
assess the same outcome. Second, frequency of dysfunction may differ between the studies depending on the
methods used to assess the treatment effect. Global cognitive impairment was assessed by the unweighted average
of 7 domain scores, 18 whereas impairment for
visuoperception was dichotomized from a z score.19 A previous review23 that aimed to outline the methodical
limitations of cognitive assessment following stoke found
that studies defining cognitive impairment as a z score
identified higher levels of dysfunction24 compared to studies
requiring patients to be impaired on 4 or more domains.25
The varying methodologies used to test cognition between
the studies, as well as the differing classifications of impairment, may have confounding effects on the results.
All studies examined stroke severity using the NIHSS score,
which may have caused a possible lateralization bias in
the study population as the NIHSS places less emphasis
on task-dependent tasks associated with RH infarcts. This
finding can lead to lower NIHSS scores despite the substantial infarct volume,26 underestimating stroke severity
in RH stroke patients and therefore biasing the results.
Studies on global cognition18 and visuoperception19-21
had small sample sizes, particularly in treated groups,
which may influence the power of the study and their
subsequence outcomes. A trend toward significance was
reported in two of the studies,19-22 which may have been
due to them being underpowered. Future studies need
to be carried out on a larger sample to explore this trend.
Furthermore, the only study to have reported the exclusion of patients with pre-existing cognitive impairment
was Nys et al,18 who excluded patients who scored 3.6
or higher on the short Informant Questionnaire on Cognitive Decline in the Elderly (IQCODE). It is therefore
not possible to ascertain whether impairment was present
prior to the stroke in those studies. Finally, studies only
included patients with first-ever symptomatic stroke; findings can therefore not be extrapolated to the stroke
population as a whole.
In summary, cognitive decline, especially in an elderly
population, is associated with reduced quality of life
and survival.27 PSCI has been recently identified as a
priority research area among stroke patients, with life

ARTICLE IN PRESS
L.J. BROOME ET AL.

after stroke and a patient-centered approach to research being recognized as priority areas in the
management of treatment.28 This review highlights the
paucity of studies examining treatment effects of thrombolysis on cognition at a time of importance in current
research priorities. Additionally, this review may be used
as a tool to inform researchers and practitioners on existing evidence exploring the impact of thrombolytic
treatment and cognition. Moreover, the review should be
used to direct future research exploring the psychological impact of stroke and its associated treatment measures.
An emphasis on larger prospective cohort studies and
RCTs is urgently needed to explore whether thrombolysis leads to a favorable cognitive outcome both in the
short term and long term.

References
1. The top ten causes of death. 2014. World Health
Organization web site. Available at: http://www.who.int/
mediacentre/factsheets/fs310/en/index.html. Accessed
July 12, 2014.
2. Pendlebury ST, Rothwell PM. Prevalence, incidence, and
factors associated with pre-stroke and post-stroke
dementia: a systematic review and meta-analysis. Lancet
Neurol 2009;8:1006-1018.
3. Patel M, Coshall C, Rudd AG, et al. Natural history of
cognitive impairment after stroke and factors associated
with recovery. Clin Rehabil 2003;17:158-166.
4. Ballad C, Rowan E, Stephens S, et al. Prospective follow
up study between 3 and 15 months after stroke:
improvements and decline in cognitive function among
dementia-free stroke survivors >75 years of age. Stroke
2003;34:2440-2444.
5. Oksala NK, Jokinen H, Melkas S, et al. Cognitive
impairment predicts post stroke death in long term follow
up. J Neurol Neurosurg Psychiatry 2009;80:1230-1235.
6. Lee M, Saver JL, Hong KS, et al. Cognitive impairment
and risk of future stroke: a systematic review and
meta-analysis. CMAJ 2014;186:E536-E546.
7. European Stroke Organisation (ESO) Executive Committee,
ESO Writing Committee. Guidelines for management of
ischaemic stroke and transient ischaemic attack 2008.
Cerebrovasc Dis 2008;25:457-507.
8. Minematsu K, Toyoda K, Hirano T, et al. Guidelines for
the intravenous application of recombinant tissue-type
plasminogen activator (alteplase), the second edition,
October 2012: a guideline from the Japan Stroke Society.
J Stroke Cerebrovasc Dis 2013;22:571-600.
9. Jauch EC, Saver JL, Adams HP Jr, et al. Guidelines for
the early management of patients with acute ischaemic
stroke: a guideline for healthcare professionals from the
American Heart Association/American Stroke Association.
Stroke 2013;44:870-947.
10. The National Institute of Neurological Disorders and
Stroke rt-PA Stroke Study Group. Tissue plasminogen
activator for acute ischaemic stroke. N Engl J Med
1995;333:1581-1588.
11. Hacke W, Donnan G, Fieschi C, et al. Association of
outcome with early with early stroke treatment: pooled
analysis of ATLANTIS, ECASS and NINDS rt-PA stroke
trials. Lancet 2004;363:768-774.

12. Lees KR, Bluhmki E, von Kummer R, et al. Time to

treatment with intravenous alteplase and outcome in
stroke: an updated pooled analysis of ECASS, ATLANTIS,
NINDS and EPITHET trials. Lancet 2010;375:1695-1703.
13. Neumann-Haefelin T, du Mesnil de Rochemont R, Fiebach
JB, et al. Effect of incomplete (spontaneous and
postthrombolytic) recanalization after middle cerebral
artery occlusion: a magnetic resonance imaging study.
Stroke 2004;35:109-114.
14. Barber PA, Davis SM, Infeld B, et al. Spontaneous
reperfusion after ischemic stroke is associated with
improved outcome. Stroke 1998;29:2522-2528.
15. Wardlaw JM, Murray V, Berge E, et al. Thrombolysis for
acute ischaemic stroke. Cochrane Database Syst Rev
2014;(7):CD000213.
16. Roman GC, Hakim AM. rtPA in acute stroke patients
with history of cognitive impairment: do not hesitate to
treat. Neurology 2014;82:2044-2045.
17. Higgins J, Green S, eds. Cochrane handbook for systematic
reviews of interventions version 5.1.0 (updated March
2011). The Cochrane Collaboration, 2011 Available at:
www.cochrane-handbook.org.
18. Nys GMS, van Zandvoort MJE, Algra A, et al. Cognitive
and functional outcome after intravenous recombinant
tissue plasminogen activator treatment in patients with
a first symptomatic brain infarct. J Neurol 2006;253:237241.
19. Laihosalo M, Kettunen JE, Koivisto AM, et al.
Thrombolytic therapy and visuoperceptual functions in
right hemisphere infarct patients. J Neurol 2011;258:10211025.
20. Kettunen JE, Laihosalo M, Ollikainen J, et al. Rightward
bias in right hemisphere infarct patients with or without
thrombolytic treatment and in healthy controls. Neurocase
2012;18:359-365.
21. Kettunen JE, Nurmi M, Koivisto AM, et al. The presence
of visual neglect after thrombolytic treatment in patients
with right hemisphere stroke. ScientificWorldJournal
2012;2012:434120.
22. Jacquin A, Virat-Brassaud ME, Rouaud O, et al. Vascular
aphasia outcome after intravenous recombinant tissue
plasminogen activator thrombolysis after stroke. Eur
Neurol 2014;71:288-295.
23. Gottesman RF, Hillis AE. Predictors and assessment of
cognitive dysfunction resulting from ischaemic stroke.
Lancet Neurol 2010;9:895-905.
24. Nys GMS, van Zandvoort MJE, De Kort PLM, et al.
Domain-specific cognitive recovery after first-ever stroke:
a follow-up study of 111 cases. J Int Neuropsychol Soc
2005;11:795-806.
25. Tatemichi TK, Desmond DW, Stern Y, et al. Cognitive
impairment after stroke: frequency, patterns, and
relationship to functional abilities. J Neurol Neurosurg
Psychiatry 1994;57:202-207.
26. Fink JN, Frampton CM, Lyden P, et al. Does hemispheric lateralization influence functional and
cardiovascular outcomes after stroke? Stroke 2008;39:33353340.
27. Neale R, Brayne C, Johnson AL. Cognition and survival:
an exploration in a large multicentre study of the
population aged 65 years and over. Int J Epidemiol
2009;30:1383-1388.
28. Pollock A, George BS, Fenton M, et al. Top 10 research
priorities relating to life after strokeconsensus from
stroke survivors, caregivers and health professionals. Int
J Stroke 2014;9:313-320.