International Journal of Stroke, 17(3)
https://doi.org/10.1177/17474930211034331
International Journal of Stroke
2022, Vol. 17(3) 251 –259
© 2021 World Stroke Organization
Article reuse guidelines:
sagepub.com/journals-permissions
DOI: 10.1177/17474930211034331
journals.sagepub.com/home/wso
Review
Frailty and cerebrovascular disease:
Concepts and clinical implications for
stroke medicine
Nicholas R Evans1 , Oliver M Todd2, Jatinder S Minhas3 ,
Patricia Fearon4, George W Harston5 , Jonathan Mant6,
Gillian Mead7 , Jonathan Hewitt8, Terence J Quinn9 and
Elizabeth AWarburton10
Abstract
Frailty is a distinctive health state in which the ability of older people to cope with acute stressors is compromised by an
increased vulnerability brought by age-associated declines in physiological reserve and function across multiple organ
systems. Although closely associated with age, multimorbidity, and disability, frailty is a discrete syndrome that is
associated with poorer outcomes across a range of medical conditions. However, its role in cerebrovascular disease
and stroke has received limited attention. The estimated rise in the prevalence of frailty associated with changing
demographics over the coming decades makes it an important issue for stroke practitioners, cerebrovascular research,
clinical service provision, and stroke survivors alike. This review will consider the concept and models of frailty, how
frailty is common in cerebrovascular disease, the impact of frailty on stroke risk factors, acute treatments, and rehabili-
tation, and considerations for future applications in both cerebrovascular clinical and research settings.
Keywords
Frailty, stroke, inflammageing, rehabilitation
Received: 6 April 2021; accepted: 2 July 2021
Introduction
Frailty—the state of vulnerability characterized by the
cumulative multisystem decline of physiological reserves
to maintain homeostasis following a stressor event1—is
associated with increased morbidity and mortality
across a range of medical conditions,2 though only
recently has attention been paid to its role in cerebro-
vascular disease. Stroke represents an archetypal stres-
sor event, and frailty may affect stroke risk factors,
disease trajectory, and outcomes (Figure 1).
Frailty is a distinct clinical syndrome discrete
from—but closely related to—age, multimorbidity,
and disability (Figure 2). Although these conditions
frequently co-exist, an individual may be frail in the
absence of significant co-morbidity and disability, and
without being elderly. This distinction is important, as
it may be possible to attenuate or reverse frailty trajec-
tories in order to reduce its burden on health
outcomes.3
The prevalence of frailty rises markedly with age.4
However, as people are living longer, and living for an
extended proportion of that time with greater disability
and comorbidity, there is a wide variation in the health
of older people. Chronological age is insufficient to cap-
ture this variation in the ageing process. Despite advo-
cacy of the ‘‘Compression of Morbidity’’
paradigm—where postponement of chronic disease
1Department of Medicine, University of Cambridge, Cambridge, UK
2Academic Unit for Ageing and Stroke Research, University of Leeds,
Leeds, UK
3NIHR Leicester Biomedical Research Centre, University of Leicester,
Leicester, UK
4Department of Stroke Medicine, Royal Victoria Hospital, Belfast, UK
5Acute Stroke Programme, Oxford University Hospitals NHS Foundation
Trust, Oxford, UK
6Department of Public Health & Primary Care, University of Cambridge,
Cambridge, UK
7Centre for Clinical Brain Sciences, University of Edinburgh, Edinburgh,
UK
8Division of Population Medicine, Cardiff University, Cardiff, UK
9Institute of Cardiovascular and Medical Sciences, University of Glasgow,
Glasgow, UK
10Department of Clinical Neurosciences, University of Cambridge,
Cambridge, UK
Corresponding author:
Nicholas Evans, Department of Medicine, University of Cambridge,
Cambridge CB2 0QQ, UK.
Email: ne214@cam.ac.uk
International Journal of Stroke, 0 0)
252 International Journal of Stroke 17(3)
International Journal of Stroke, 17(3)
outweighs any increase in life-expectancy, thereby redu-
cing time in later life with chronic disability5—some
western countries have experienced worsening health
across multiple age ranges.6 Shifting demographic
trends with rising numbers of older, multimorbid, frai-
ler individuals necessitate a move away from consider-
ation of single organ disease-specific processes to a
more nuanced frailty-based consideration of how the
multisystem decline in physiological reserves and con-
sequent vulnerability modifies the natural history of
stroke.
This review will consider the models of frailty and
how it is evaluated, prior to considering the effect of
frailty along the natural history of stroke (including
effects on cardiovascular risk factors preceding stroke,
its role during acute stroke presentation and treatment,
and impact after stroke on rehabilitation and secondary
prevention). Finally, we will consider future directions
and applications for frailty in both clinical care and
research.
Concepts of frailty
Two predominant approaches to evaluating frailty have
developed based around measuring deficits versus
assessing a frailty phenotype.
Cumulative deficit model
This operationalized model of frailty considers that
‘‘the more things individuals have wrong with them,
the higher the likelihood that they will be frail.’’7 This
model is predicated upon recognition that physiological
changes (‘‘deficits’’) may not necessarily achieve disease
status, yet their accumulation is associated with higher
levels of frailty and adverse outcomes. The Cumulative
Deficit Model quantifies frailty through a frailty index
consisting of a number of equally weighted deficits
across different domains (including cognition, function,
mobility, and continence), where the number of deficits
present in the individual is divided by the total number
of possible scoring deficits to give a ratio between zero
to one which reflects the spectrum of frailty (Table 1).
Frailty, as defined by this deficit accumulation, is asso-
ciated with increased mortality and rates of
institutionalization.8
The frailty phenotype model
In contrast to the Cumulative Deficit Model, the Fried
Phenotype Model recognizes five main phenotypical
characteristics of frailty:
1. Weight loss
2. Self-reported exhaustion
Figure 2. Schema illustrating the relationships between
frailty, disability, and multimorbidity.
Figure 1. Differing trajectories in disability following stroke events in non-frail (a) and frail (b) individuals.
t 0(0)
International Journal of Stroke 0 0
Evans et al. 253
International Journal of Stroke, 17(3)
3. Low levels of activity
4. Slow gait speed
5. Weak grip strength.
When comparing those with no criteria (non-frail),
one or two criteria (intermediate frailty), and three or
more criteria (frail) in an unselected population, there is
a clear increase in mortality with increasing frailty, as
well as associations with falls, worsening mobility,
functional disability, and hospitalization.9
Measuring frailty
Recognition of the importance of frailty has resulted in
policymakers advocating frailty screening in unsched-
uled admissions. The measures used may reflect:
1. Different models of frailty: frailty indices (Cumulative
Deficit Model), or measures including grip strength
and walking speed (Frailty Phenotype Model).
2. Different clinical contexts:
(i) Secondary care: The Hospital Frailty Risk Score
(HFRS) considers 109 routinely collected ICD-10
diagnoses to produce a score associated with length
of hospitalization and in-patient mortality.10
(ii) Community: The electronic frailty index (eFI)
using 36 deficits in primary health datasets meas-
ures frailty at a population level, and demonstrates
associations with hospitalization, nursing home
admission, and all-cause one-year mortality.11
3. Different data settings:
(i) Bedside assessment using the Clinical Frailty Scale
(CFS), which correlates strongly with the frailty
index, evaluates how an individual aged over 65
years was, two weeks prior to admission (and
importantly not as they appear at time of
admission).12
(ii) Routinely collected health data, e.g. eFI, HFRS.
(iii) Research study data, e.g. grip strength, gait speed.
Although premorbid modified Rankin Scale (mRS)
is frequently used to determine eligibility for partici-
pation in stroke clinical trials, it is important to rec-
ognize that pre-stroke mRS (a measurement of
disability) is not a substitute for frailty assessment.
Pre-stroke mRS demonstrates reasonable agreement
with a frailty index, though only one-third of individ-
uals had evidence of frailty yet over half were classed
as dependent on pre-stroke mRS, and there was a
cohort with frailty but low disability on the pre-
stroke mRS.13 Other studies have reported moderate
agreement between pre-stroke mRS and a frailty
index, but only slight agreement between pre-stroke
mRS and phenotypical frailty measurements.14
However, other studies have reported no statistically
significant correlation between CFS and mRS.15
Future work needs to consider the best method for
evaluating frailty in the stroke setting.
Such considerations highlight a challenge for the
operationalized use of frailty measurements: there
remains debate over whether frailty should be con-
sidered according to individual domains (physical, cog-
nitive, brain appearances) or the total burden of frailty
for the individual. The relative weighting of these dif-
ferent domains vary within different frailty scales, and
consequently may make direct comparisons between
studies challenging. This review will consider the total
burden of frailty on the individual, but will explore the
associations described with different frailty domains.
Arguably, the abundance of neuroimaging in Stroke
may facilitate the operationalized radiological evalu-
ation of ‘‘brain frailty,’’ but for the clinician seeing
the patient it is often the totality of frailty that is
Table 1. Exemplar of a frailty index used in individuals presenting with stroke.14
Frailty index
Depression
Anxiety
Polypharmacy
Previous cerebrovascular disease
Atrial fibrillation
Diabetes
Hypertension
Previous myocardial infarction
Heart failure
Vascular disease
Hyperlipidaemia
Haemoglobin (low)
Care-home resident
Carers
Hearing aid
Sensory impairment (e.g. blind/deaf)
Continence bladder
Continence bowel
Falls
Fracture
Chronic Obstructive Pulmonary Disease
Cancer
Liver disease
Peptic ulcer
Arthritis
Impaired external ADL
Impaired ADL
Mobility aid
Assistance walking
Calcium
Albumin (low)
High glucose
Renal failure
Note: This approach considers equally weighted deficits across different domains (including function, mobility, continence, co-morbidities, and bio-
chemical values). To be robust, a frailty index should have approximately 30–40 potential deficits spanning multiple domains.
International Journal of Stroke, 0 0
l.
outweighs any increase in life-expectancy, thereby redu-
cing time in later life with chronic disability5—some
western countries have experienced worsening health
across multiple age ranges.6 Shifting demographic
trends with rising numbers of older, multimorbid, frai-
ler individuals necessitate a move away from consider-
ation of single organ disease-specific processes to a
more nuanced frailty-based consideration of how the
multisystem decline in physiological reserves and con-
sequent vulnerability modifies the natural history of
stroke.
This review will consider the models of frailty and
how it is evaluated, prior to considering the effect of
frailty along the natural history of stroke (including
effects on cardiovascular risk factors preceding stroke,
its role during acute stroke presentation and treatment,
and impact after stroke on rehabilitation and secondary
prevention). Finally, we will consider future directions
and applications for frailty in both clinical care and
research.
Concepts of frailty
Two predominant approaches to evaluating frailty have
developed based around measuring deficits versus
assessing a frailty phenotype.
Cumulative deficit model
This operationalized model of frailty considers that
‘‘the more things individuals have wrong with them,
the higher the likelihood that they will be frail.’’7 This
model is predicated upon recognition that physiological
changes (‘‘deficits’’) may not necessarily achieve disease
status, yet their accumulation is associated with higher
levels of frailty and adverse outcomes. The Cumulative
Deficit Model quantifies frailty through a frailty index
consisting of a number of equally weighted deficits
across different domains (including cognition, function,
mobility, and continence), where the number of deficits
present in the individual is divided by the total number
of possible scoring deficits to give a ratio between zero
to one which reflects the spectrum of frailty (Table 1).
Frailty, as defined by this deficit accumulation, is asso-
ciated with increased mortality and rates of
institutionalization.8
The frailty phenotype model
In contrast to the Cumulative Deficit Model, the Fried
Phenotype Model recognizes five main phenotypical
characteristics of frailty:
1. Weight loss
2. Self-reported exhaustion
Figure 2. Schema illustrating the relationships between
frailty, disability, and multimorbidity.
Figure 1. Differing trajectories in disability following stroke events in non-frail (a) and frail (b) individuals.
International Journal of Stroke, 0(0)
2 International Journal of Stroke 0(0)
254 International Journal of Stroke 17(3)
International Journal of Stroke, 17(3)
Impact of frailty on stroke presentation
and outcomes
Stroke presentation, hyperacute therapies, and
mortality
Pre-stroke frailty is independently associated with
stroke severity in the acute setting, as measured by
the National Institute of Health Stroke severity scale
(NIHSS).27 Mediation analysis in a single center
study suggested pre-stroke frailty status is not asso-
ciated with poorer outcomes directly, but rather the
effect is mediated by this association between frailty
and stroke severity.28 However, other studies report
the association between premorbid frailty and early
outcomes remains significant after adjustment for
stroke severity. In a retrospective single-center
study, the CFS was associated with increased 30-
day mortality after ischemic stroke after adjustment
for age, vascular risk factors, and NIHSS.15
Ultimately, large prospective studies are required to
elucidate this pathway in terms of the relative contri-
butions from frailty promoting bigger strokes,
impaired resilience to withstand the stroke, or a com-
bination of the two.
As well as severity of presentation, frailty may dem-
onstrate a treatment-modifying effect in hyperacute
reperfusion therapies, and consequently poorer recov-
ery. In a proof-of-principle study, pre-stroke frailty was
independently associated with an attenuated improve-
ment in NIHSS following thrombolysis, with each one-
point increase in CFS associated with a reduction of
one point in the NIHSS improvement.15 Following
mechanical thrombectomy, frailty is present in around
a third of individuals and is associated with poorer
neurological status and increased mortality after 90
days.29,30
Both pre-stroke pre-frailty and frailty are inde-
pendently associated with shorter survival time after
stroke in individuals aged under 80 years of age, but
not in individuals older than this.31 When consider-
ing the components of the Fried phenotype, slow
walking speed and low grip strength were consist-
ently and independently associated with reduced sur-
vival time.31
Related syndromes and surrogate markers for
frailty may predict outcomes following stroke.
Sarcopenia—the loss of skeletal muscle and function
that is a major component of frailty—is independently
associated with more severe strokes at presentation
and poorer outcomes after three months.32 Similarly,
after controlling for age and stroke type, the only
other independent predictor of death after any
stroke was poor performance on a timed walk—a sur-
rogate marker of frailty—measured prior to the inci-
dent stroke.33
Stroke recovery
Frailty may influence other non-physical aspects of
stroke recovery. The premorbid frailty index demon-
strates a borderline significant association with the
development of post-stroke delirium after adjustment
for age, sex, and medication count.14 Pre-stroke frailty
is independently associated with poorer post-stroke
cognition after adjustment for age, delirium, pre-
stroke cognitive impairment, and stroke severity.34
Pre-stroke frailty phenotypes of slow walking speed
and low grip strength are also independently associated
with post-stroke cognitive decline and reduced ability
to perform activities of daily living.31 Such associations
have potential repercussions for reduced effectiveness
of rehabilitation for individuals with frailty-associated
post-stroke cognitive impairment, whilst also represent-
ing an important avenue of research to consider
whether frailty exerts a treatment-modifying effect on
post-stroke cognitive rehabilitation.
Frailty is associated with a marked reduction in self-
reported quality of life after stroke, where frail individ-
uals reported poorer quality of life compared to the
non-frail group, driven by significant reductions in
mobility and self-care categories, after adjustments for
age, sex, and NIHSS score.35 This study considered the
frailty phenotype using self-reported exhaustion, low
physical activity, and weight loss from the pre-stroke
setting, combined with post-stroke measures of walking
speed and grip strength.
Frailty may modulate the response to psychosocial
intervention following stroke, with non-frail individuals
demonstrating significant improvements in activities of
daily living in response to such interventions, whilst no
significant improvement (and a trend towards worsen-
ing outcomes) was observed in the frail cohort. Similar
treatment-modifying effects of frailty upon psycho-
social intervention for physical performance and mor-
tality were also observed.36
Discharge destination
In 7258 individuals receiving stroke care in the United
States through Medicare, 46.9% of pre-morbidly frail
individuals were discharged to a nursing institution,
compared to 28% of pre-frail and 18.5% of non-frail
individuals. Furthermore, non-frail individuals were
71% more likely than frail (and 16% more likely than
pre-frail) to be discharged to in-patient rehabilitation
after adjustment for demographics, stroke severity, and
co-morbidities.25
Hemorrhagic stroke
In a retrospective single center observational study,
frailty was not associated with mortality following
International Journal of Stroke, 0(0)
Evans et al. 5
important. The relative strengths and weaknesses of a
total versus sub-type evaluation of frailty, and whether
these vary according to the aspect of stroke care repre-
sent important avenues of future research into the bio-
logical mechanisms underlying the impact of frailty in
stroke etiology and outcomes.
Frailty and vascular risk factors
Frailty is associated with increasing 10-year
Framingham Risk Scores, with Scores particularly
pronounced in the presence of weight loss, weakness,
and slowness components of the frailty phenotype.16
Frailty frequently co-exists with conventional
cardiovascular risk factors, where it demonstrates dis-
ease-modifying and treatment-modifying effects. In
hypertension, achieving a systolic blood pressure
below 140mmHg was associated with a 14% reduction
in all-cause mortality in non-frail individuals, yet no
difference in all-cause mortality was seen in frailer indi-
viduals.17 In individuals with diabetes, frailty is asso-
ciated with increased mortality, hospital admission,
disability, and cognitive impairment.18 Atherosclerotic
burden is also associated with frailty,19 potentially
through sub-clinical effects on end-organs contributing
to decreased function and physiological reserve. As dis-
cussed in subsequent sections, such end-organ effects on
the brain may contribute to findings of ‘‘brain frailty’’
and negatively impact cognitive reserve.
Frailty is common in individuals with atrial fibrilla-
tion, with approximately two-thirds of individuals
being pre-frail or frail, and independently associated
with higher rates of hospitalization, all-cause mortality,
bleeding, and stroke.20 Frailty is associated with lower
odds of being prescribed anticoagulation at the time of
hospitalization, but higher odds of being prescribed
anticoagulation in community settings.21 Additionally,
frailty is a major factor influencing discontinuation of
therapy for those already taking anticoagulants.22 Such
findings illustrate the perpetual dilemma for prescribing
anticoagulation in co-existent atrial fibrillation, frailty,
and risk of falls, particularly given the rising rate-
adjusted fall death rate as more individuals are surviv-
ing with stroke disability.23
Frailty and the risk of stroke
Frailty in stroke is common. A recent meta-analysis of
18 studies with 48,009 participants reported the preva-
lence of pre-frailty and frailty in individuals with stroke
as 49% and 22%, respectively.24 Frail individuals with
stroke are typically older and more likely to be female.25
Although much of the focus on associations between
frailty and stroke have considered the impact of frailty
on stroke, it is important to recognize the impact of
stroke on frailty. The neurological deficits following a
stroke are likely to exacerbate the phenotypic charac-
teristics of frailty, and prior stroke has been found to be
an important factor in the transition from robust to
frail, as well as a worsening of a frailty trajectory.26
Whether this bi-directional relationship becomes a
self-propagating cycle (Figure 3), and whether it may
represent a target for intervention, requires further
research.
Figure 3. Factors influencing propagation of frailty and stroke risk.
i l 0(0)
4 International Journal of Stroke 0 0
Evans et al. 255
International Journal of Stroke, 17(3)
Impact of frailty on stroke presentation
and outcomes
Stroke presentation, hyperacute therapies, and
mortality
Pre-stroke frailty is independently associated with
stroke severity in the acute setting, as measured by
the National Institute of Health Stroke severity scale
(NIHSS).27 Mediation analysis in a single center
study suggested pre-stroke frailty status is not asso-
ciated with poorer outcomes directly, but rather the
effect is mediated by this association between frailty
and stroke severity.28 However, other studies report
the association between premorbid frailty and early
outcomes remains significant after adjustment for
stroke severity. In a retrospective single-center
study, the CFS was associated with increased 30-
day mortality after ischemic stroke after adjustment
for age, vascular risk factors, and NIHSS.15
Ultimately, large prospective studies are required to
elucidate this pathway in terms of the relative contri-
butions from frailty promoting bigger strokes,
impaired resilience to withstand the stroke, or a com-
bination of the two.
As well as severity of presentation, frailty may dem-
onstrate a treatment-modifying effect in hyperacute
reperfusion therapies, and consequently poorer recov-
ery. In a proof-of-principle study, pre-stroke frailty was
independently associated with an attenuated improve-
ment in NIHSS following thrombolysis, with each one-
point increase in CFS associated with a reduction of
one point in the NIHSS improvement.15 Following
mechanical thrombectomy, frailty is present in around
a third of individuals and is associated with poorer
neurological status and increased mortality after 90
days.29,30
Both pre-stroke pre-frailty and frailty are inde-
pendently associated with shorter survival time after
stroke in individuals aged under 80 years of age, but
not in individuals older than this.31 When consider-
ing the components of the Fried phenotype, slow
walking speed and low grip strength were consist-
ently and independently associated with reduced sur-
vival time.31
Related syndromes and surrogate markers for
frailty may predict outcomes following stroke.
Sarcopenia—the loss of skeletal muscle and function
that is a major component of frailty—is independently
associated with more severe strokes at presentation
and poorer outcomes after three months.32 Similarly,
after controlling for age and stroke type, the only
other independent predictor of death after any
stroke was poor performance on a timed walk—a sur-
rogate marker of frailty—measured prior to the inci-
dent stroke.33
Stroke recovery
Frailty may influence other non-physical aspects of
stroke recovery. The premorbid frailty index demon-
strates a borderline significant association with the
development of post-stroke delirium after adjustment
for age, sex, and medication count.14 Pre-stroke frailty
is independently associated with poorer post-stroke
cognition after adjustment for age, delirium, pre-
stroke cognitive impairment, and stroke severity.34
Pre-stroke frailty phenotypes of slow walking speed
and low grip strength are also independently associated
with post-stroke cognitive decline and reduced ability
to perform activities of daily living.31 Such associations
have potential repercussions for reduced effectiveness
of rehabilitation for individuals with frailty-associated
post-stroke cognitive impairment, whilst also represent-
ing an important avenue of research to consider
whether frailty exerts a treatment-modifying effect on
post-stroke cognitive rehabilitation.
Frailty is associated with a marked reduction in self-
reported quality of life after stroke, where frail individ-
uals reported poorer quality of life compared to the
non-frail group, driven by significant reductions in
mobility and self-care categories, after adjustments for
age, sex, and NIHSS score.35 This study considered the
frailty phenotype using self-reported exhaustion, low
physical activity, and weight loss from the pre-stroke
setting, combined with post-stroke measures of walking
speed and grip strength.
Frailty may modulate the response to psychosocial
intervention following stroke, with non-frail individuals
demonstrating significant improvements in activities of
daily living in response to such interventions, whilst no
significant improvement (and a trend towards worsen-
ing outcomes) was observed in the frail cohort. Similar
treatment-modifying effects of frailty upon psycho-
social intervention for physical performance and mor-
tality were also observed.36
Discharge destination
In 7258 individuals receiving stroke care in the United
States through Medicare, 46.9% of pre-morbidly frail
individuals were discharged to a nursing institution,
compared to 28% of pre-frail and 18.5% of non-frail
individuals. Furthermore, non-frail individuals were
71% more likely than frail (and 16% more likely than
pre-frail) to be discharged to in-patient rehabilitation
after adjustment for demographics, stroke severity, and
co-morbidities.25
Hemorrhagic stroke
In a retrospective single center observational study,
frailty was not associated with mortality following
International Journal of Stroke, 0(0)
Evans et al. 5
256 International Journal of Stroke 17(3)
International Journal of Stroke, 17(3)
spontaneous intracerebral hemorrhage (ICH), nor was
frailty associated with post-stroke mRS after adjust-
ment for the Intracerebral Hemorrhage Score.37
Higher frailty scores were associated with lower rates
of surgical intervention, and in those with more exten-
sive ICH the frailty scores were higher in those who
died following withdrawal of care versus those who
died despite active management. In those undergoing
surgery for spontaneous ICH, frailty was independently
associated with higher mortality and poorer longterm
neurological recovery 6–8 months after ICH.38 In con-
trast, age was independently associated with poorer
neurological recovery but not mortality.
Frailty and secondary prevention:
Overall the effect of frailty on secondary prevention
after stroke has received little attention. However,
there has been some consideration of its role in carotid
revascularization. In a study of 1,426,343 individuals
undergoing carotid revascularization, 59,158 (4.2%)
were identified as frail. Compared to non-frail individ-
uals, frailty was independently associated with
increased post-procedure mortality, stroke, myocardial
infarction, and longer length of hospital stay.39 Other
studies have reported higher rates of frailty (up to
27.3%), but also supported the independent association
of frailty with procedural complications, mortality, and
30-day readmission.40 Subgroup analysis suggested that
frailty may not be associated with complications and
mortality in individuals undergoing carotid stenting,
but was unable to determine this definitively.40
Frailty and cerebrovascular
pathophysiology
The challenge for frailty research within cerebrovascu-
lar disease is to move beyond the reporting of associ-
ations to understanding the biological mechanisms
through which frailty affects outcomes. Crucially, cen-
tral and peripheral vascular hemodynamic changes
occur in response to ageing, and frailty is associated
with impaired cerebral autoregulation.41
Distinguishing pathological disease states from
‘‘healthy’’ ageing is paramount for the development
of effective interventions. For example, age is negatively
correlated with penumbral volume (but not core
volume) in individuals undergoing CT perfusion in
the hyperacute stroke setting.42 However, this work
considered only chronological age, not frailty, and it
would be advantageous for future work to evaluate
the role of frailty in this relationship.
Cross-sectional neuroimaging studies have suggested
links between systemic frailty and chronic brain patho-
physiology. Frailty is associated with cortical atrophy
(predominantly in men),43 deep white matter hyperin-
tensities,19 severe periventricular white matter hyperin-
tensities, and cortical superficial siderosis.44
The presence of white matter hyperintensities
(WMH)—but not baseline infarcts or cerebral micro-
bleeds—was associated with frailty progression inde-
pendently of other small vessel disease markers in a
longitudinal population-based study.45 In a further lon-
gitudinal study, although WMH volume at baseline
was associated with a higher likelihood of progression
in frailty phenotype severity, no association was found
between the progression of WMH volume over the
study period and frailty progression, though both the
sample size and WMH volume increase over the study
period were small.46
Features of ‘‘brain frailty’’ (leukoaraiosis, atrophy,
and old vascular lesions/infarcts) were associated with
poorer functional and cognitive outcomes at 90 days in
individuals following ischemic stroke.47 Atrophy and
leukoaraiosis are also associated with increased 90-
day mortality after thrombolysis treatment.48 Such
imaging criteria may indicate a more ‘‘vulnerable’’
brain with poorer neurological and cognitive reserve,
accounting for poorer outcomes, but further work is
required to establish the interaction between the frail
individual and frail brain, as well as elucidating any
underlying biological mechanisms. In addition to the
associations of a frail brain with poorer clinical out-
come, any attenuation of treatment effect size has also
yet to be clearly established.
Future directions for frailty in stroke
medicine
Future work needs to consider the best methods to
evaluate frailty in individuals with stroke. The different
approaches, relative weightings of different domains of
frailty, and scoring systems for evaluating frailty pose
challenges for comparing studies and how they may be
employed in clinical practice. A single assessment of
gait speed and grip strength in the immediate post-
stroke setting for a phenotype model may not reflect
early neurological recovery or associated complications
that may be seen after a stroke, and consequently may
over-estimate frailty, and may not be practical in some
settings. In clinical practice the quality of pre-morbid
data to calculate a frailty index or phenotype is likely to
vary, and outside of population research datasets it is
unlikely that individuals will have premorbid walking
speed and grip strength measured routinely. Frailty
indices and the CFS may be more pragmatic and argu-
ably easier to score retrospectively in a general
population.
Stroke recovery is multifactorial, consisting of not
only physical but also cognitive and psychological
I J l f Strok , 0(0)
6 International Journal of Stroke 0 0
Evans et al. 257
International Journal of Stroke, 17(3)
recovery. Expansion of the Fried phenotype model to
include cognitive frailty, where physical and cognitive
deficits frequently co-exist, has been argued to be a
better predictor of long-term dependency and death
than either domain alone, and may have important
implications for identifying those at high risk of post-
stroke delirium and mood deficits.49,50 Furthermore, an
important question relevant for stroke rehabilitation
will be whether modifying physical frailty is able to
modify cognition, and vice-versa.
Frailty may represent an important target for inter-
vention, either to prevent further deterioration in frailty
or possibly to reverse the frailty trajectory in order to
reduce its impact on post-stroke outcomes.
Multifaceted intervention programmes—including
physical, cognitive, and nutritional interventions—have
been proposed,3 through which interventions hold the
most promise in a stroke setting remains unclear.
Frequently, research studies have excluded older
people with frailty. Furthermore, there is often a
sense of fatalism that results in frailer individuals not
being offered the usual evidence-based treatments due
to a belief that they will not respond or have a higher
risk of adverse events. Consequently, there is a need for
robust evidence for prognostication, treatment, and
management applicable to frail individuals with
stroke who are more representative of the general
population seen in clinical practice, and hence necessi-
tate measuring frailty in clinical trials. Incorporation of
frailty measures into electronic record systems and
national stroke databases may also be advantageous
in establishing such trends at a population level.
Conclusion
Frailty is emerging as an important clinical risk factor
for stroke, and is independently associated with a range
of poor post-stroke outcomes. Shifting demographics,
and the consequent rise in frailty, means that the
burden of frailty and its effect on cerebrovascular dis-
ease is likely to increase. How best to assess frailty in
stroke, attenuate its effects, and incorporate assessment
of frailty into treatment decisions, are pressing concerns
for both clinical care and research.
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with
respect to the research, authorship, and/or publication of this
article.
Funding
The author(s) disclosed receipt of the following financial sup-
port for the research, authorship and/or publication of this
article: NRE and JSM are supported by National Institute of
Health Research Clinical Lectureships. OMT is supported by
a Research Training Fellowship from The Dunhill Medical
Trust [RTF107/0117]. The collaboration of authors is sup-
ported by a Joint British Association of Stroke Physicians
and NIHR CRN: Stroke Writing Group Award.
ORCID iDs
Nicholas R Evans https://orcid.org/0000-0002-7640-4701
Jatinder S Minhas https://orcid.org/0000-0002-0576-9105
George W Harston https://orcid.org/0000-0003-4916-5757
Gillian Mead https://orcid.org/0000-0001-7494-2023
Terence J Quinn https://orcid.org/0000-0003-1401-0181
References:
1. Clegg A, Young J, Iliffe S, Rikkert MO and Rockwood
K. Frailty in elderly people. Lancet 2013; 381: 752–762.
2. Wallis SJ, Wall J, Biram RW and Romero-Ortuno R.
Association of the clinical frailty scale with hospital out-
comes. QJM 2015; 108: 943–949.
3. Walston J, Buta B and Xue QL. Frailty screening and
interventions: considerations for clinical practice. Clin
Geriatr Med 2018; 34: 25–38.
4. Gale CR, Cooper C and Sayer AA. Prevalence of frailty
and disability: findings from the English Longitudinal
Study of Ageing. Age Ageing 2015; 44: 162–165.
5. Fries JF. Frailty, heart disease, and stroke: the compres-
sion of morbidity paradigm. Am J Prev Med 2005; 29:
6. Welsh CE, Matthews FE and Jagger C. Trends in life
expectancy and healthy life years at birth and age 65 in
the UK, 2008–2016, and other countries of the EU28: an
observational cross-sectional study. Lancet Reg Health
Eur 2021; 2: 100023.
7. Rockwood K and Mitnitski A. Frailty in relation to the
accumulation of deficits. J Gerontol A Biol Sci Med Sci
2007; 62: 722–727.
8. Rockwood K, Mitnitski A, Song X, Steen B and Skoog I.
Long-term risks of death and institutionalization of eld-
erly people in relation to deficit accumulation at age 70. J
Am Geriatr Soc 2006; 54: 975–979.
9. Fried LP, Tangen CM, Walston J, et al. Frailty in older
adults: evidence for a phenotype. J Gerontol A Biol Sci
Med Sci 2001; 56: M146–M156.
10. Street A, Maynou L, Gilbert T, Stone T, Mason S and
Conroy S. The use of linked routine data to optimise
calculation of the Hospital Frailty Risk Score on the
basis of previous hospital admissions: a retrospective
observational cohort study. Lancet Healthy Longev
2021; 2: e154–e62.
11. Clegg A, Bates C, Young J, et al. Development and val-
idation of an electronic frailty index using routine pri-
mary care electronic health record data. Age Ageing
2016; 45: 353–360.
12. Rockwood K, Song X, MacKnight C, et al. A global
clinical measure of fitness and frailty in elderly people.
CMAJ 2005; 173: 489–495.
13. Fearon P, McArthur KS, Garrity K, et al. Prestroke
modified rankin stroke scale has moderate interobserver
reliability and validity in an acute stroke setting. Stroke
2012; 43: 3184–3188.
International Journal of Stroke, 0 0
l.
spontaneous intracerebral hemorrhage (ICH), nor was
frailty associated with post-stroke mRS after adjust-
ment for the Intracerebral Hemorrhage Score.37
Higher frailty scores were associated with lower rates
of surgical intervention, and in those with more exten-
sive ICH the frailty scores were higher in those who
died following withdrawal of care versus those who
died despite active management. In those undergoing
surgery for spontaneous ICH, frailty was independently
associated with higher mortality and poorer longterm
neurological recovery 6–8 months after ICH.38 In con-
trast, age was independently associated with poorer
neurological recovery but not mortality.
Frailty and secondary prevention:
Overall the effect of frailty on secondary prevention
after stroke has received little attention. However,
there has been some consideration of its role in carotid
revascularization. In a study of 1,426,343 individuals
undergoing carotid revascularization, 59,158 (4.2%)
were identified as frail. Compared to non-frail individ-
uals, frailty was independently associated with
increased post-procedure mortality, stroke, myocardial
infarction, and longer length of hospital stay.39 Other
studies have reported higher rates of frailty (up to
27.3%), but also supported the independent association
of frailty with procedural complications, mortality, and
30-day readmission.40 Subgroup analysis suggested that
frailty may not be associated with complications and
mortality in individuals undergoing carotid stenting,
but was unable to determine this definitively.40
Frailty and cerebrovascular
pathophysiology
The challenge for frailty research within cerebrovascu-
lar disease is to move beyond the reporting of associ-
ations to understanding the biological mechanisms
through which frailty affects outcomes. Crucially, cen-
tral and peripheral vascular hemodynamic changes
occur in response to ageing, and frailty is associated
with impaired cerebral autoregulation.41
Distinguishing pathological disease states from
‘‘healthy’’ ageing is paramount for the development
of effective interventions. For example, age is negatively
correlated with penumbral volume (but not core
volume) in individuals undergoing CT perfusion in
the hyperacute stroke setting.42 However, this work
considered only chronological age, not frailty, and it
would be advantageous for future work to evaluate
the role of frailty in this relationship.
Cross-sectional neuroimaging studies have suggested
links between systemic frailty and chronic brain patho-
physiology. Frailty is associated with cortical atrophy
(predominantly in men),43 deep white matter hyperin-
tensities,19 severe periventricular white matter hyperin-
tensities, and cortical superficial siderosis.44
The presence of white matter hyperintensities
(WMH)—but not baseline infarcts or cerebral micro-
bleeds—was associated with frailty progression inde-
pendently of other small vessel disease markers in a
longitudinal population-based study.45 In a further lon-
gitudinal study, although WMH volume at baseline
was associated with a higher likelihood of progression
in frailty phenotype severity, no association was found
between the progression of WMH volume over the
study period and frailty progression, though both the
sample size and WMH volume increase over the study
period were small.46
Features of ‘‘brain frailty’’ (leukoaraiosis, atrophy,
and old vascular lesions/infarcts) were associated with
poorer functional and cognitive outcomes at 90 days in
individuals following ischemic stroke.47 Atrophy and
leukoaraiosis are also associated with increased 90-
day mortality after thrombolysis treatment.48 Such
imaging criteria may indicate a more ‘‘vulnerable’’
brain with poorer neurological and cognitive reserve,
accounting for poorer outcomes, but further work is
required to establish the interaction between the frail
individual and frail brain, as well as elucidating any
underlying biological mechanisms. In addition to the
associations of a frail brain with poorer clinical out-
come, any attenuation of treatment effect size has also
yet to be clearly established.
Future directions for frailty in stroke
medicine
Future work needs to consider the best methods to
evaluate frailty in individuals with stroke. The different
approaches, relative weightings of different domains of
frailty, and scoring systems for evaluating frailty pose
challenges for comparing studies and how they may be
employed in clinical practice. A single assessment of
gait speed and grip strength in the immediate post-
stroke setting for a phenotype model may not reflect
early neurological recovery or associated complications
that may be seen after a stroke, and consequently may
over-estimate frailty, and may not be practical in some
settings. In clinical practice the quality of pre-morbid
data to calculate a frailty index or phenotype is likely to
vary, and outside of population research datasets it is
unlikely that individuals will have premorbid walking
speed and grip strength measured routinely. Frailty
indices and the CFS may be more pragmatic and argu-
ably easier to score retrospectively in a general
population.
Stroke recovery is multifactorial, consisting of not
only physical but also cognitive and psychological
International Journal of Stroke, 0(0)
6 International Journal of Stroke 0(0)
164–168.
258 International Journal of Stroke 17(3)
International Journal of Stroke, 17(3)
14. Taylor-Rowan M, Cuthbertson G, Keir R, et al. The
prevalence of frailty among acute stroke patients, and
evaluation of method of assessment. Clin Rehabil 2019;
33: 1688–1696.
15. Evans NR, Wall J, To B, Wallis SJ, Romero-Ortuno R
and Warburton EA. Clinical frailty independently pre-
dicts early mortality after ischaemic stroke. Age Ageing
2020; 49: 588–591.
16. Fernandes J, Gomes CDS, Guerra RO, et al. Frailty syn-
drome and risk of cardiovascular disease: analysis from
the International Mobility in Aging Study. Arch Gerontol
Geriatr 2021; 92: 104279.
17. Todd OM, Wilkinson C, Hale M, et al. Is the association
between blood pressure and mortality in older adults dif-
ferent with frailty? A systematic review and meta-analy-
sis. Age Ageing 2019; 48: 627–635.
18. Hanlon P, Faure´ I, Corcoran N, et al. Frailty measure-
ment, prevalence, incidence, and clinical implications in
people with diabetes: a systematic review and study-level
meta-analysis. Lancet Healthy Longev 2020; 1: e106–e116.
19. Newman AB, Gottdiener JS, McBurnie MA, et al.
Associations of subclinical cardiovascular disease with
frailty. J Gerontol A Biol Sci Med Sci 2001; 56:
M158–M166.
20. Gugganig R, Aeschbacher S, Leong DP, et al. Frailty to
predict unplanned hospitalization, stroke, bleeding, and
death in atrial fibrillation. Eur Heart J Qual Care Clin
Outcomes 2021; 7: 42–51.
21. Wilkinson C, Todd O, Clegg A, et al. Management of
atrial fibrillation for older people with frailty: a system-
atic review and meta-analysis. Age Ageing 2018; 48 :
196–203.
22. Buck J, Fromings Hill J, Martin A, et al. Reasons for
discontinuing oral anticoagulation therapy for atrial
fibrillation: a systematic review. Age Ageing 2021; 50 :
1108–1117.
23. Hartholt KA, Lee R, Burns ER, et al. Mortality from
falls among US adults aged 75 years or older, 2000–
2016. JAMA 2019; 321: 2131–2133.
24. Palmer K, Vetrano DL, Padua L, et al. Frailty syndromes
in persons with cerebrovascular disease: a systematic
review and meta-analysis. Frontiers Neurol 2019; 10:
1255.
25. Seamon BA and Simpson KN. The effect of frailty on
discharge location for medicare beneficiaries after acute
stroke. Arch Phys Med Rehabil 2019; 100: 1317–1323.
26. Lee Y, Kim J, Han ES, Ryu M, Cho Y and Chae S.
Frailty and body mass index as predictors of 3-year mor-
tality in older adults living in the community. Gerontology
2014; 60: 475–482.
27. Kanai M, Noguchi M, Kubo H, et al. Pre-stroke frailty
and stroke severity in elderly patients with acute stroke. J
Stroke Cerebrovasc Dis 2020; 29: 105346.
28. Noguchi M, Kubo H, Kanai M, Nozoe M and Shimada
S. Relationship between pre-stroke frailty status and
short-term functional outcome in older patients with
acute stroke – a mediation analysis. Arch Gerontol
Geriatr 2021; 94: 104370.
29. Pinho J, Ku¨ppers C, Nikoubashman O, et al. Frailty is an
outcome predictor in patients with acute ischemic stroke
receiving endovascular treatment. Age Ageing 2021
[Online ahead of print].
30. Schnieder M, Ba¨hr M, Kirsch M, et al. Analysis of frailty
in geriatric patients as a prognostic factor in endovascular
treated patients with large vessel occlusion strokes. J Clin
Med 2021; 10: 2171.
31. Winovich DT, Longstreth Jr WT, Arnold AM, et al.
Factors associated with ischemic stroke survival and
recovery in older adults. Stroke 2017; 48: 1818–1826.
32. Nozoe M, Kanai M, Kubo H, Yamamoto M, Shimada S
and Mase K. Prestroke sarcopenia and functional out-
comes in elderly patients who have had an acute stroke:
a prospective cohort study. Nutrition 2019; 66: 44–47.
33. Longstreth Jr WT, Bernick C, Fitzpatrick A, et al.
Frequency and predictors of stroke death in 5,888 par-
ticipants in the Cardiovascular Health Study. Neurology
2001; 56: 368–375.
34. Taylor-Rowan M, Keir R, Cuthbertson G, et al. Pre-
stroke frailty is independently associated with post-
stroke cognition: a cross-sectional study. J Int
Neuropsychol Soc 2019; 25: 501–506.
35. Wæhler IS, Saltvedt I, Lydersen S, et al. Association
between in-hospital frailty and health-related quality of
life after stroke: the Nor-COAST study. BMC Neurol
2021; 21: 100.
36. Ertel KA, Glymour MM, Glass TA and Berkman LF.
Frailty modifies effectiveness of psychosocial intervention
in recovery from stroke. Clin Rehabil 2007; 21: 511–522.
37. Kim MG, Gandhi C, Azizkhanian I, et al. Frailty and
spontaneous intracerebral hemorrhage: Does the modi-
fied frailty index predict mortality? Clin Neurol
Neurosurg 2020; 194: 105816.
38. Imaoka Y, Kawano T, Hashiguchi A, et al. Modified
frailty index predicts postoperative outcomes of spontan-
eous intracerebral hemorrhage. Clin Neurol Neurosurg
2018; 175: 137–143.
39. Mandelbaum AD, Hadaya J, Ulloa JG, et al. Impact of
frailty on clinical outcomes after carotid artery revascu-
larization. Ann Vasc Surg 2021 [online ahead of print].
40. Pandit V, Lee A, Zeeshan M, et al. Effect of frailty syn-
drome on the outcomes of patients with carotid stenosis.
J Vasc Surg 2020; 71: 1595–1600.
41. Maguire F, Romero-Ortuno R, O’Connor JD, Reilly RB,
Knight SP and Kenny RA. One-dimensional statistical
parametric mapping identifies impaired orthostatic cere-
brovascular and cardiovascular response in frailty index.
J Gerontol A Biol Sci Med Sci 2021; 76: 885-892.
42. Agarwal S, Scoffings DJ, Jones PS, et al. Interaction of
age with the ischaemic penumbra, leptomeningeal collat-
eral circulation and haemodynamic variables in acute
stroke: a pilot study. J Neurol Neurosurg Psychiatr
2013; 84: 271–276.
43. Gallucci M, Piovesan C and Di Battista ME.
Associations between the frailty index and brain atrophy:
The Treviso Dementia (TREDEM) Registry.
J Alzheimers Dis 2018; 62: 1623–1634.
44. Jordan N, Gvalda M, Cody R, Galante O, Haywood C
and Yates P. Frailty, MRI, and FDG-PET measures in
an Australian memory clinic cohort. Front Med
(Lausanne) 2020; 7: 578243.
i l 0(0)
8 International Journal of Stroke 0 0
Evans et al. 259
International Journal of Stroke, 17(3)
45. Siejka TP, Srikanth VK, Hubbard RE, et al. White
matter hyperintensities and the progression of frailty –
the tasmanian study of cognition and gait. J Gerontol A
Biol Sci Med Sci 2020; 75: 1545–1550.
46. Maltais M, de Souto Barreto P, Moon SY, Rolland Y
and Vellas B. Prospective association of white matter
hyperintensity volume and frailty in older adults. Exp
Gerontol 2019; 118: 51–54.
47. Appleton JP, Woodhouse LJ, Adami A, et al. Imaging
markers of small vessel disease and brain frailty, and out-
comes in acute stroke. Neurology 2020; 94: e439–e52.
48. Delcourt C, Wang X, Zhou Z, et al. Brain imaging
abnormalities and outcome after acute ischaemic
stroke: the ENCHANTED trial. J Neurol Neurosurg
Psychiatry 2020; 91: 1290–1296.
49. Kelaiditi E, Cesari M, Canevelli M, et al. Cognitive
frailty: rational and definition from an (I.A.N.A./
I.A.G.G.) international consensus group. J Nutr Health
Aging 2013; 17: 726–734.
50. Aliberti MJR, Cenzer IS, Smith AK, Lee SJ, Yaffe K and
Covinsky KE. Assessing risk for adverse outcomes in
older adults: the need to include both physical frailty
and cognition. J Am Geriatr Soc 2019; 67: 477–483.
International Journal of Stroke, 0 0
l.