Post-stroke fatigue following minor infarcts: a prospective study
N Radman2, F Staub1, T. Aboulafia-Brakha2, A Berney4, J Bogousslavky3, J-M
1 Department of Neurology, Lausanne University Hospital, 1011 Lausanne,
2 Department of Neurology, University of Fribourg, Switzerland
3 Center for Brain and Nervous System Disorders (Neurocenter), Genolier
4 Psychiatry Liaison Service, Lausanne University Hospital, Switzerland
Word count: Title: 58 characters; abstract: 293 words; main text: 4678 words;
bibliography: 49 references.
Illustrations: 2 Tables, 1 Apendix.
Classification of the manuscript: Original article
Corresponding author: Julien Bogousslavsky
Center for Brain and Nervous System Disorders (Neurocenter), Genolier Swiss
Medical Network, and Department of Neurology and Neurorehabilitation, Clinique
Valmont, Glion/Montreux, Switzerland. email@example.com
Keywords: fatigue, stroke, neuropsychology
Objective: Subjective fatigue affects 30-72% of patients following a stroke; it may persist for
a long time and may have an adverse influence on the patient’s rehabilitation and, ultimately,
survival, even in the absence of significant sensory motor impairment. The aim of this study is
to explore the potential relationship between fatigue and post-stroke mood, cognitive
dysfunction, disability and infarct site and to determine the predictive factors in the
development of post-stroke fatigue (PSF) after minor infarct.
Method: Ninety-nine functionally active patients over 70 years old, victims of a first, non-
disabling stroke (NIHSS≤ 6 in acute phase and < 3 after 6 months) were assessed during the
acute phase and then at six (T1) and twelve months (T2) after their stroke. Scores in the
fatigue assessment inventory (FAI) were described and correlated to age, gender, neurological
and functional impairment, lesion site, mood scores, neuropsychological data, laboratory data
and quality of life at T1 and T2 using a multivariate logistic regression analysis in order to
determine which variables recorded at T1 best predicted fatigue at T2.
Result: As many as 30.5% of the patients at T1 and 34.7% at T2 (11.6% of new cases between
T1 and T2) reported fatigue. The presence of severe fatigue and elevated depression scores at
T1 were predictors of severe fatigue at T2. At both six and twelve months, there was a
significant association between fatigue and a reduction in professional activity. Attentional-
executive impairment and anxiety levels remained associated with PSF throughout this time
period, underlining the critical role of these variables in the genesis of PSF. There was no
significant association between the lesion site and PSF.
Conclusion: This study suggests that attentional and executive impairment, as well as anxiety,
may play a critical role in the development of PSF. Moreover, the presence of early severe
fatigue and high depression scores could be predictive factors of long-lasting PSF.
The feeling of fatigue (subjective fatigue) is a complex phenomenon with motor-perceptive,
emotional and cognitive components which may be difficult to understand. It can be part of a
regular physiological reaction (e.g., as a result of lack of sleep) or a consequence of various
medical conditions, which include cancer and autoimmune, endocrinological, infectious,
rheumatological, neurological and psychiatric diseases, as well as sleep disorders (Guessous
et al., 2006). Normal fatigue is usually acute and can be alleviated by rest, while its
pathological presentation is characterized by a longer duration with “a feeling of early
exhaustion, weariness and aversion to effort” (de Groot et al., 2003, Staub and
Objective measuring of fatigue usually reveals a decline in performance during the repetition
of physical or mental tasks, while subjective measuring is characterized by self-reports of
decreased/loss of abilities associated with a heightened sensation of physical or mental strain,
even without conspicuous effort (Staub et al., 2002). Objective and subjective dimensions
are generally poorly correlated(Claros-Salinas D et al., 2010).
Pathological fatigue is frequent in stroke patients (accounting for 30-72% of this population)
and is usually referred to as post-stroke fatigue (PSF) (Carlsson et al., 2003, Glader et al.,
2002, Ingles et al., 1999, Schepers et al., 2006, Staub et al., 2004, Van der Werf et
al., 2001). It may persist for a long time after the acute event and may have detrimental
influences on rehabilitation and survival, as well as on the patient’s family, social and
professional life (Glader et al., 2002, Röding et al., 2003, Schepers et al., 2006). The
correlates and mechanisms of PSF remain relatively underexplored compared with post-stroke
cognitive and affective changes, including depression. However, some recent studies suggest
that a syndrome of « primary » post-stroke fatigue may develop independently of post-stroke
depression, significant cognitive or neurological sequelae, and of other clinical dysfunctions
potentially associated with fatigue (McGeough E et al., 2009, Snaphaan L et al., 2011,
Tseng BY et al., 2010). Such dissociation between depression and fatigue has already been
highlighted in patients with Parkinson’s disease (Friedman and Friedman, 1993), and
multiple sclerosis (Van der Werf et al., 1998, Vercoulen et al., 1996) and also seems to
be valid in stroke patients (Ingles et al., 1999, Ingles et al., 1999, Van der Werf et al.,
2001). Data resulting from pathophysiological studies (Bruno et al., 1998, Chaudhuri and
Behan, 2000, Colombo et al., 2000) in post-viral syndromes, MS and basal ganglia
diseases, suggest that “primary” PSF might be linked to specific damage to the reticular
formation and other cortex-activating systems involved in the functioning of the neural
The present study aims to investigate the subjective dimension of PSF in the absence of
significant sensory motor impairment, by characterizing and identifying its persistence and by
exploring potential relationships between fatigue and post-stroke mood, cognitive
dysfunction, disability and infarct site. To achieve these aims, we evaluated the patients
(victims of a non-severe first stroke) longitudinally, in order to obtain predictive neurological,
affective and demographic factors for the development of fatigue. We also evaluated fatigue
in the subgroup of non-depressed patients in order to better define the determinant of fatigue
in the absence of depression. Finally, we also evaluated separately the professionally active
patients, who were most likely to feel disabled by the presence of fatigue. (Carlsson et al.,
2003, Staub and Bogousslavsky, 2001).
Materials and Methods
Subjects were prospectively recruited from patients examined consecutively in our
department during the acute stage of a first ischemic or hemorrhagic stroke (< 7 days of
onset), and who were included in the Lausanne Stroke Registry (Bogousslavsky et al.,
1988). The inclusion criteria included: i) a “non-disabling” (minor) stroke, defined by the
total score of the National Institute of Health Stroke Scale (NIHSS) ≤ 6 at discharge from
hospital and ≤ 3 after 6 months; ii) age ≤ 70 years old; iii) regular involvement in working
activities (job, housework etc.) before the stroke.
The exclusion criteria included prior impairment in intermediate activities of daily living,
recurrent stroke, transient ischemic attack, psychiatric history, known sleep disorders and
coexisting disease interfering with fatigue assessment.
Acute phase assessment (< 7 days of stroke onset):
We recorded the patients’ characteristics and stroke features according to the standard
protocol of tests from the Lausanne Stroke Registry, including epidemiological data,
neurological status and brain imaging (Bogousslavsky et al., 1988). Subjects were
categorized into two groups, namely ≤45 years and >45years, and distributed in a statistically
balanced manner. We also registered the professional status of each patient at the time of their
stroke (professionally active, retired, no lucrative activity). The patients were informed about
the study and, if they accepted to participate in it, in accordance with the rules of the local
ethical committee, they were given a follow-up appointment after six months.
Brain MRI or CT imaging was performed at the acute post-stroke stage for each patient.
Images were analyzed in a “blind” manner (i.e., without disclosing the patients’ names) by
two of the authors (FS and JMA). By means of anatomical templates, we differentiated the
following anatomical regions: frontal, temporal, occipital, parietal, mesencephalic, basal
ganglia, thalamic, white matter and cerebellar. In some cases, lesions simultaneously affected
two different areas (for example, parietal and temporal); in these cases both areas were
recorded. Thus the final total number of lesions is greater than the total number of patients.
According to their lesion site, patients were categorized into five groups: “cortical”,
“subcortical”, “cortical-subcortical”, “cerebellar” and “brainstem”.
Chronic phase assessment: fatigue, neurological neuropsychological and emotional
evaluation, 6 and 12 months follow-up (respectively T1 and T2):
The same protocol was administered at T1 and at T2. The patients were once again informed
about the aims of the study before the T1 evaluation.
Instruments and scoring:
Fatigue was assessed using the Fatigue Assessment Instrument (FAI) (Schwartz et al.,
1993), a multidimensional 29-item auto-evaluation scale that allows the quantification of
fatigue through a fatigue severity score (see Appendix 1). The FAI was developed in order to
assess fatigue across a range of medical conditions and has been validated in different
diagnoses. It has four sub-scores: severity, specificity, consequences of fatigue and
responsiveness to rest/sleep. The FAI severity score was selected as our outcome measure. It
has indeed been shown to be reliable in distinguishing patients from controls; 81.3% of all the
medical patients (Lyme disease, chronic fatigue syndrome, systemic lupus erythematosus,
multiple sclerosis, affective disorders) scoring 4 or above and 89.2% of healthy subjects
scoring less than 4 (Schwartz et al., 1993). In a previous study, severity scores were > 4 in
nearly one third of non-disabling stroke patients (Gramigna et al., 2007). Thus, we retained
a cut-off of 4 in selecting a patient with a severe fatigue syndrome.
A clinician-neurologist (JMA) carried out the standard general neurological assessment,
including NIHSS and any history of sleep apnea. Patients with sleep disorders verified by
formal sleep assessment were excluded.
Disabilities were evaluated using the modified Rankin scale (Rankin, 1957). A detailed
cognitive examination was conducted by a neuropsychologist (FS) . The following tests were
used for different functions: Attention: phasic alert and divided attention (North et al.,
1994) and « D2 »(Brickenkamp, 1981). Language: object naming from line drawing
(French version of the Boston naming test)(Assal, 1985) and written comprehension (Boston
Diagnostic Aphasia Examination). Executive functions: a modified version of the Stroop test
(Assal, 1985); category and letter fluency tasks and a non-verbal directed fluency task (« 5
points »). Short term verbal and non-verbal memory: digit span and the Corsi blocks test
(Assal, 1985). Long-term memory: Rey auditory verbal memory task (Assal, 1985).
Each test was rated in a dichotomic way (normal vs. abnormal, cut-off performance set at
below 2 SD) according to a standardized neuropsychological battery (Assal, 1985). For each
domain (language, memory, attention, executive functions), the patient’s score was
considered abnormal if one of the scores was below 2 SD. Thus,, a global cognitive score
was developed internally, based on the percentage of tests (> 10%) showing abnormal
A psychiatrist (AB) performed the psychiatric evaluation. The Hamilton Depression Rating
Scale (HDRS) and the Hamilton Anxiety Rating Scale (HARS)(Hamilton, 1960) criteria
were fulfilled, and the diagnosis of a major depression established according to DSM-IV
criteria. HDRS and HARS scores were treated as continuous variables for most statistical
analyses. In order to complete the analysis in the subgroup of non-depressed patients, the
diagnosis of depression was applied to patients who met the DSM-IV criteria for major
depression or/and in patients whose total HDRS score was ≥10, a value which has been
frequently used as a cut-off in stroke populations.
A measure of the quality of life was also obtained (Stroke-Specific Quality of life
Scale)(Williams et al., 1999). Data on plasma levels of cortisol, ACTH, T4 (free) and TSH
were systematically collected. Laboratory work-up was considered as pathological when one
anomaly was registered, ending up as a dichotomous variable.
Finally, any modification of professional activity following the stroke was recorded. Thus, for
each patient, we took into consideration normal, reduced or interrupted lucrative activities at
T1 and T2, as compared with the premorbid state. This variable did not concern retired
The socio-demographical characteristics, cognitive status and FAI scores were presented for
the six and 12 months evaluations. The fatigue severity was presented in terms of percentage
of patients with fatigue severity score>4. A general cognitive status was defined based on the
percentage of tests showing abnormal performances.
The potential relationships between fatigue on one hand, and neurological,
neuropsychological, functional, emotional and physiological variables on the other, were
investigated at T1 and T2. Statistical analyses were carried out on the whole group, and
separately on the two subgroups, “non-depressed patients” and “professionally active
patients” (at the onset of the stroke).
While the FAI severity score was used as our outcome measure, the independent variables
consisted of age, gender, changes in activity rate (for patients still in active professional life at
onset of stroke), neurological and functional impairment, level of depression, degree of
anxiety, lesion side and site, neuropsychological data, laboratory data, quality of life and
abulia. First, a univariate analysis was performed for each variable (one-way analysis of
variance for categorical variables and linear regression model for continuous variables).
Variables with a p-value<0.05 were kept for the final stepwise multiple regression analysis, as
well as the variables related to our hypothesis. Level of significance was set at 0.05.
We also used multivariate logistic regression analysis to determine which variables recorded
at T1 best predicted T2 fatigue (later fatigue).
Analyses were conducted by a statistician using the Stata program(StataCorp LP, 1996-
Over a continuous period of two years, 125 patients fulfilled the inclusion criteria upon
discharge from hospital. One hundred and nine (87.2%) patients (37 women (34%) and 72
men (66%)) were evaluated at a six months follow-up (T1). Among the missing subjects, two
had moved and could not be contacted, ten refused the assessment and four were excluded
because of a NIHSS score outside the limits of inclusion criteria at the time of evaluation (at
six months). At the 12 months follow-up (T2), 99 (79.2%) subjects (33 women (33%) and 66
men (67%)) completed the examination. Of the ten dropout patients between T1 and T2, three
had had a stroke recurrence, one had died, three were excluded because we diagnosed a sleep
apnea syndrome and three refused to continue the study.
Table 1 shows socio-demographical characteristics, cognitive status and scores in the FAI.
Among the patients who reported fatigue at six months following a stroke, 77.3% still
complained of fatigue at the 12 months follow-up (30% at T1 and 23.2% at T2). The total
percentage of patients reporting fatigue at the second follow-up was 34 %. A significant
number (11.6%) were new cases who only started to complain of fatigue later on (Fig 1),
which was mostly attributable to social changes, such as resuming working. Fatigue was rated
as the worst symptom by 23% of the patients (n=25) at T1 and by 25% (n=25) at T2, and it
was considered as a new phenomenon (different both qualitatively and quantitatively from
fatigue experienced before the stroke), by 49.5% (n=54) and 44.4% (n=44) patients at T1 and
T2 respectively. When taking into account the patients involved in a professional activity at
the time of their stroke (“professionally active” subgroup, n=56), we observed that 64.3%
(n=36) had reduced or given up lucrative activity at the six months follow-up, the percentage
decreasing slightly to 55.4% (n=31) at 12 months post-stroke.
Concerning the cognitive functions, nearly one third of the patients showed some impairment
in global cognitive scores, i.e., produced more than 10% of abnormal or marginal
performances when tested (Table 1). Concerning stroke locations at the six months follow-
up, we assessed 41 patients with lesions in the right hemisphere, 48 in the left, three with
bilateral lesions and 17 with lesions in the sub-tentorial area. At the 12 months follow-up, 38
patients had right and 39 had left hemispheric damage, two had bilateral lesions and 20 had
With respect to lesion sites, at T1 and T2, we found respectively 29 and 27 with subcortical
damage, 28 and 24 cortical lesions, 22 and 20 with cortico-subcortical damage, 25 and 24
brainstem lesions and five cerebellar strokes.
At T1, the highest correlation coefficients were found for NIHSS, Rankin, HDRS and HARS
total scores, sustained attention, phasic alert, global cognitive scores, age and modification of
professional activity (reduction or suppression). At T2 the strongest association concerned
HDRS, HARS, language, long-term memory, executive functions, sustained attention, phasic
alert, divided attention, global cognitive scores and modification of professional activity.
There was no correlation between infarct site and fatigue severity scores, except for a
tendency towards left parietal lesions, which was not statistically significant (p = 0.086). At
T1, the multivariate analysis based on variables with a p-value<0.05 (Table 2) revealed that
severe fatigue (FAI>4) was independently associated with anxiety levels, as shown by the
HARS score. When considering non-depressed patients, significant variables were younger
age (≤45) and anxiety levels, while in the “professionally active” group, severe fatigue was
associated with anxiety levels and modification (reduction or suppression) of professional
At T2 (one year), severe fatigue was independently associated with higher depression scores
(HDRS) and with sustained attention dysfunction. In non-depressed subjects, the level of
depression scores and impairment in executive functions constituted the relevant associated
variables. In the “professionally active” group, the significant factors were higher depression
scores (HDRS) and sustained attention deficit.
We did not find any statistically significant correlation between PSF and plasma levels of
cortisol, ACTH, T4 (free) and TSH. Moreover, no significant association between PSF and
lesion side and/or site was found.
In this study, our purpose was to characterize PSF longitudinally in the first year following
the first minor strokes (NHSS < 6). Our data suggest that: 1) fatigue is a frequent, stable and
persisting symptom following a non-disabling stroke, affecting 30% of patients at six months
and 34% at one year after a minor infarct; 2) there is no clear association between fatigue and
lesion site; and 3) cognitive (attentional or executive) impairment and levels of anxiety remain
independently associated at T2, underlining the critical role of these two variables in the
persistence of PSF. Other factors, such as the high psychophysiological cost of a delayed
resumption of professional activity (>6months), seem also to contribute to the lack of
improvement in fatigue.
We found a slightly lower incidence (or frequency) of PSF when compared to other studies
that describe this phenomenon in up to 72% of patients assessed (Carlsson et al., 2003,
Glader et al., 2002, Gramigna et al., 2007, Ingles et al., 1999, Naess et al., 2011,
Schepers et al., 2006, Van der Werf et al., 2001). As fatigue has also been correlated
with motor impairment, this might in part be explained by the fact that we included only
minor strokes and “non-disabled” patients. All patients had a NIHSS < 3 at 6 and 12 months.
and the motor component as well as the exertion fatigue were less likely to participate to the
phenomenon. As for post-stroke depression, this variability in reported prevalence may also
be related to methodological factors, including age selection of patients, inclusion criteria,
time of evaluation and investigation tools.
A number of longitudinal studies have been carried out on fatigue (Carlsson et al., 2003,
Christensen et al., 2008, Glader et al., 2002, Paradiso et al., 1997, Schepers et al.,
2006, Snaphaan L et al., 2011, van Eijsden et al., 2012). We found a remarkable
stability of fatigue across time (mean FAI severity score of 3.2±1.8 at the first evaluation (T1)
and 3.3 ±1.7 at the second (T2); 30% of patients at T1 and 34% at T2 had FAI>4). Compared
to the six months follow-up, one-year self-rated fatigue remains unchanged despite significant
improvement in neurological and functional outcomes, as well as depression. Therefore, our
data show that PSF is a stable symptom in the first year after a stroke. A stabilizing of fatigue
frequency after the subacute period has been suggested by other longitudinal studies, as
reported by 59% of patients at ten days, and then by 44% , 38%, and 40% of stroke patients
respectively three months, one year, and two years following hospitalization for stroke
(Christensen et al., 2008). Another longitudinal study (Schepers et al., 2006), showed an
increased percentage of self-rated fatigue between admission and the six months and one year
follow-ups; fatigue was indeed reported by 51.5%, 64.1% and 69.5% of patients respectively.
The Fatigue Severity Scale scores (a first and very similar version of the FAI severity score)
increased from 4.1 to 4.5 and then 4.7 across time. Our results are consistent with these data.
In cross-sectional studies, the time elapsed since stroke was found to have no influence on the
presence and severity of fatigue (Ingles et al., 1999, Van der Werf et al., 2001).
The percentage of patients rating fatigue as their worst symptom was 23% (n=25) at six
months and 25% (n=25) at one year. In the ‘active’ subgroup, a reduction (or suppression) of
the professional activity rate was associated with severe fatigue at the six months evaluation.
These data are in agreement with clinical observations and studies in which many patients
showing good neurological recovery complained of disabling fatigue; this was, in some cases,
their only sequelae (Carlsson et al., 2003, Staub and Bogousslavsky, 2001), and could
be associated with a poor functional outcome (Christensen et al., 2008). One aspect of this
explanation may be that patients with severe neuropsychological and neurological impairment
saw fatigue as only a ‘minor’ symptom compared to their other symptoms. Furthermore,
patients with major disabilities are generally partially or completely dependent and are
therefore not confronted by the demands of an active social and professional life. Thus,
patients with a good neurological outcome may be the group most likely to feel disabled by
the presence of fatigue, which may prevent them from resuming social, familial, and
professional activities, even when neurological or cognitive sequelae cannot be found
(Carlsson et al., 2003, Staub and Bogousslavsky, 2001). It is striking to note that only a
small proportion (20-40%) of young stroke patients return to full-time work after disease
onset (Wozniak et al., 1999), even in the absence of neurological and cognitive deficits
which could be detected by a classical clinical examination. Young stroke patients often
report nonspecific symptoms, such as fatigue, irritability, anxiety, headache, emotional
instability, forgetfulness, and concentration difficulties (Carlsson et al., 2003, Leegaard,
1983, Röding et al., 2003). In the specific case of small subcortical strokes, which do not
usually lead to major cognitive dysfunction, the patients’ quality of life has been found to be
less favorable than expected, and patients commonly complain of fatigue and of ‘being
different from before the stroke’, despite a good functional recovery (Van Zandvoort et al.,
Mood factors are closely related to fatigue (general or severe) at any time following a stroke.
Paradoxically, the level of anxiety and depression remain closely associated with fatigue
when considering the non-depressed group.
The overlap between fatigue and depression is undeniable, and the presence of fatigue
constitutes one of the main diagnostic criteria for depression in most standardized scales for
assessment of depression. Nevertheless, fatigue can also occur in the absence of depression, a
dissociation already highlighted in patients with Parkinson’s disease (Friedman and
Friedman, 1993) or multiple sclerosis(Van der Werf et al., 1998, Vercoulen et al., 1996)
and which is equally valid in stroke patients (Ingles et al., 1999, Van der Werf et al.,
2001). For example, only 38% of patients with severe fatigue after a stroke were found to be
depressed (Van der Werf et al., 2001) and 57% of non-depressed patients reported frequent
fatigue (Crosby et al., 2011).
Until recently, fatigue after a stroke was misleadingly regarded by many clinicians as a mere
component of post-stroke depression. While it has been shown that over three-quarters of
patients examined eight months after a stroke may complain of fatigue whilst being assessed
for depression, it has become clear that post-stroke depression and fatigue are commonly
dissociated from each other (Claros-Salinas D et al., 2010, Staub et al., 2000, Staub
and Bogousslavsky, 2001): affective distress and vegetative and fatigue manifestations
may evolve independently, although it is this combination which leads to the highest score on
the Hamilton depression scale. In another study (Paradiso et al., 1997) it was found that
fatigue and sleep disturbances were common after a stroke in patients who also suffered from
depressive moods. In contrast, fatigue was also reported by many patients who did not suffer
from depressive moods (Crosby et al., 2011). However, some studies reported a “dose-
response” relationship between depression and fatigue; the depressed patients had higher
fatigue severity scores (Glader et al., 2002, Schepers et al., 2006)
PSF seems to be developed by cognitive dysfunction (Annoni et al., 2008, Choi-Kwon and
Kim, 2011). Cognitive factors, and especially impairment in attentional and executive
functions, may play a critical role in the more chronic stages. Physical and cognitive
impairment have been found to have a cumulative effect on PSF: mean FSS scores were
higher in patients who had both physical and cognitive impairment in comparison to patients
who had neither physical nor cognitive impairment and patients who had either physical or
cognitive impairment (p\0.001 )(Passier et al., 2011). In contrast, Naess (Naess et al.,
2005) and colleagues found no association between cognitive dysfunction (based on MMSE
scores) and PSF, but this result could also be explained by the fact that MMSE is of limited
sensitivity and specificity in stroke patients (Godefroy et al., 2011). Our results showed
specifically that, 12 months after a stroke, severe fatigue was associated with sustained
attention dysfunction, and in non-depressed patients impairment in executive functions was
correlated to severe fatigue. Such an interaction between executive dysfunction and fatigue
has been already highlighted in chronic brain-injured patients (Dimoska-Di Marco et al.,
2011), but has received little attention for stroke patients. Our interpretation is that, as this is
the first study in which PSF has been analyzed using such a large neuropsychological
evaluation, we were in a position to analyze more subtle cognitive changes. However, such a
correlation does not necessarily mean that lasting PSF is due to executive dysfunction. But
such a correlation, between fatigue and executive dysfunction after a stroke, is supported by
observations that cognitive function is mainly affected by fatigue in patients with lower
physical disabilities (van Eijsden et al., 2012).
On the other hand, we found no statistically significant associations (in multivariate analyses)
between PSF and lesion site and side, functional and neurological impairment and what could
be called localized cognitive dysfunction. The only trend that we found in univariate analyses
was that of high fatigue severity scores in patients with left parietal lesions, but these were not
significant. Whatever the mechanism, the link between fatigue and brain lesions appears to be
very complex. Like PSD, PSF is a complex, heterogeneous, and mainly subjective
phenomenon, which cannot be easily explained by lesion localization. In fact, published
studies (Carlsson et al., 2003, Glader et al., 2002, Ingles et al., 1999, Leegaard, 1983,
Röding et al., 2003, Schuitemaker et al., 2004, Van der Werf et al., 2001, Van
Zandvoort et al., 1998), together with some unpublished data, have shown that it is difficult
to find definite predictors for PSF. Lesion location may contribute only to a small extent to the
risk of developing PSF, which could be due to the multiplicity of fatigue expressions
(Gramigna et al., 2007). In fact, such clinical cohort analyses suggest that there are two
different types of fatigue in stroke: a “task-specific” fatigue linked to cognitive or physical
sequels (such as residual aphasia following left parietal lesions, when mental fatigue appears
after speaking for a certain time) and a “primary” PSF linked to subtle attention difficulties
after brain stem or subcortical strokes which impair the cortical activating system (Staub
and Bogousslavsky, 2001). It is indeed tempting to sketch a pathophysiological model of
PSF in which damage to specific subcortical structures would lead to fatigue through a
decrease in cortical activation. However, whilst two studies have suggested that infratentorial
infarction (Snaphaan L et al., 2011) and infarction in basal ganglia may play a role in
development of PSF (Tang et al., 2010), the majority of the findings in literature on strokes
are in contrast with the findings of studies on other neurological pathologies. In a study on
multiple sclerosis patients (Colombo et al., 2000), is it suggested that fatigue might be the
result of functional deafferentation of the cortex due to cortical-subcortical damage, the main
structures involved being the periventricular areas, internal capsule, and trigon. In Parkinson’s
disease, both lesions to basal ganglia and strata dopaminergic input impairment, which are
known to diminish cortical activation and reduce voluntary attention, have been associated
with fatigue (Owen et al., 1993). Other techniques, such as Voxel Based Lesion Symptom
Mapping, could help to solve this problem in future studies.
In conclusion, post-stroke fatigue is an important problem which may have adverse effects on
daily life and the rehabilitation process. There is still controversy over the correlation between
lesion sites and PSF, but high depression scores and cognitive impairment (especially in the
executive function and sustained attention domains) play important roles in the development
Our research project was supported by the Swiss National Science Foundation (32-138497-
and 3200-061342). Thanks to Sandrine Gramigna and Andrea Brioschi for their help with the
patients and to Ann Travis for editing.
Annoni J, Staub F, Bogousslavsky J, Brioschi A. Frequency, characterisation and
therapies of fatigue after stroke. Neurol Sci. 2008;29:244-6.
Assal G. Batterie des examens neuropsychologiques du CHUV. Lausanne1985.
Bogousslavsky J, Van Melle G, Regli F. The Lausanne Stroke Registry. Analysis of
1,000 consecutive patients with first stroke. Stroke. 1988;19:1083-92.
Brickenkamp R. Test D2: Aufmerksamkeits-Belastungs-Test (Handanweisung, 7th
ed.) Göttingen: Verlag fur Psychologie. 1981.
Bruno R, Creange S, Frick N. Parallels between post-polio fatigue and chronic fatigue
syndrome : a common pathophysiology ? . Am J Med. 1998;105(suppl.1):66-73.
Carlsson G, Möller A, Blomstrand C. Consequences of mild stroke in persons < 75
years-A 1-year follow-up. Cerebrovascular Diseases. 2003;16:383-8.
Chaudhuri A, Behan P. Fatigue and basal ganglia. Journal of the Neurological
Choi-Kwon S, Kim J. Post-Stroke fatigue: an emerging, critical issue in stroke
medicine. International Journal of Stroke. 2011;6:328-36.
Christensen D, Johnsen S, Watt T, Harder I, Kirkevold M, Andersen G. Dimensions
of post-stroke fatigue: a two-year follow-up study. Cerebrovasc dis. 2008;26(2):134-
Claros-Salinas D, Bratzke D, Schroter H. Subjective and objective assessment of
fatigue. Journal of the Neurological Sciences. 2010;298:165-6.
Colombo B, Martinelli Boneschi F, Rossi P, Rovaris M, Maderna L, Filippi M, et al.
MRI and motor evoked potential findings in nondisabled multiple sclerosis patients
with and without symptoms of fatigue. J Neurol. 2000;247(7):506-9.
Crosby G, Munshi S, Karat A, Worthington E, Lincoln N. Fatigue after stroke:
frequency and effect on daily life. Disability and Rehabilitation. 2011;1-5, Early
de Groot MH, Phillips S, Eskes G. Fatigue associated with stroke and other
neurologic conditions: implications for stroke rehabilitation. Archives of Physical
Medicine and Rehabilitation. 2003;84(11):1714-20
Dimoska-Di Marco A, McDonald S, Kelly M, Tate R, Johnstone S. A meta-analysis of
response inhibition and Stroop interference control deficits in adults with traumatic
brain injury (TBI). J Clin Exp Neuropsychol. 2011;33(4):471-85.
Friedman J, Friedman H. Fatigue in Parkinson’s disease. Neurology. 1993;43:2016-
Glader EL, Stegmayr B, Asplund K. Poststroke fatigue : a 2-year follow-up study of
stroke patients in Sweden. Stroke. 2002;33:1327-33.
Godefroy O, Fickl A, Roussel M, Auribault C, Bugnicourt J, Lamy C, et al. Is the
Montreal Cognitive Assessment superior to the Mini-Mental State Examination to
detect poststroke cognitive impairment? A study with neuropsychological evaluation.
Gramigna S, Schluep M, Staub F, Bruggimann L, Simioni S, Bogousslavsky J, et al.
Dimensions multiples de la fatigue d‘origine neurologique: differences entre l‘accident
vasculaire cerebral et la sclerose en plaques. Revue Neurologique (Paris).
Guessous I, Favrat B, Cornuz J, Verdon F. Fatigue: review and systematic approach
to potential causes. Rev Med Suisse. 2006;29(2):2725-31.
Hamilton M. A rating scale for depression. Journal of Neurology, Neurosurgery &
Ingles J, Eskes G, Phillips S. Fatigue after stroke. Arch Phys Med Rehabil.
Ingles J, Eskes G, Phillips S. Fatigue after stroke. Arch Phys Med Rehabil.
Leegaard O. Diffuse cerebral symptoms in convalescents from cerebral infarction and
myocardial infarction. Acta Neurol Scan. 1983;67:348-55.
McGeough E, Pollock A, Smith L N, Smith L C, Dennis M, Sharpe M, et al.
Interventions for post-stroke fatigue. Cochrane Database of Systematic Reviews.
Naess H, Lunde L, Brogger J, Waje-Andreassen U. Fatigue among stroke patients on
long-term follow-up. The Bergen Stroke Study. Journal of the Neurological Sciences.
2011;[Epub ahead of print].
Naess H, Nyland H, Thomassen L, Aarseth J, Myhr K. Fatigue at long-term follow-up
in young adults with cerebral infarction. Cerebrovasc dis. 2005;20:245-50.
North P, Leclercq M, Crémel N, Tassi P, Jeromin D. Tests d’évaluation de l’attention
(TEA), adaptation française. Psytest. 1994.
Owen A, Roberts A, Hodges J, Summers B, Polkey C, Robbins T. Contrasting
mechanisms of impaired attentionnal set-shifting in patients with frontal lobe damage
or Parkinson’s disease. Brain. 1993;116:1159-75.
Paradiso S, Ohkubo T, Robinson R. Vegetative and Psychological Symptoms
Associated with Depressed Mood Over the First Two Years After Stroke. The
International Journal of Psychiarty in Medicine. 1997;27:137-57.
Passier PECA, Post MWM, van Zandvoort MJE, Rinkel GJE, Lindeman E, Visser-
Meily JMA. Predicting fatigue 1 year after aneurysmal subarachnoid Hemorrhage. J
Rankin J. Cerebral vascular accidents in patients over the age of 60. Scottish Medical
Röding J, Lindström B, Malm J, Öhman A. Frustrated and invisible - younger stroke
patients’ experiences of the rehabilitation process. Disability and Rehabilitation.
Schepers V, Visser-Meily A, Ketelaar M, Lindeman E. Poststroke fatigue: course and
its relation to personal and stroke-related factors. Arch Phys Med Rehabil.
Schuitemaker G, Dinant G, van der Pol G, Appels A. Vital exhaustion as a risk
indicator for first stroke. Psychosomatics. 2004;45:114-8.
Schwartz J, Jandorf L, Krupp L. The measurement of fatigue : a new instrument.
Journal of Psychosomatic Research. 1993;37(7):753-62.
Snaphaan L, Van der Werf S, De Leeuw F. E. Time course and risk factors of post-
stroke fatigue: aprospective cohort study. European Journal of Neurology.
StataCorp LP TU. Stata 8, Data Analysis and Statstical Support. 1996-2008.
Staub F, Annoni J, Bogousslavsky J. Post-stroke fatigue : a major problem in “non-
disabling” stroke. Cerebrovasc Dis. 2002;13(suppl 3):96.
Staub F, Annoni J, Bogousslavsky J. Fatigue after stroke : a pilot study. Cerebrovasc
Dis. 2000;10(suppl 2):62.
Staub F, Annoni J, Gramigna S, Bogousslavsky J. Time dynamics of post-stroke
fatigue. Cerebrovasc dis. 2004;17:49.
Staub F, Bogousslavsky J. Fatigue after stroke : a major but neglected issue.
Cerebrovasc dis. 2001;12:75-81.
Tang W, Chen Y, Mok V, Chu W, Ungvari G, Ahuja A, et al. Acute basal ganglia
infarcts in poststroke fatigue: an MRI study. Journal of Neurology. 2010;257:178-82.
Tseng BY, Billinger SA, Gajewski BJ, Kluding PM. Exertion fatigue and chronic
fatigue are two distinct constructs in people post-stroke. Stroke. 2010;41(12):2908-
Van der Werf S, Jongen P, Lycklama a Nijeholt G, Barkhof F, Hommes O,
Bleijenberg G. Fatigue in multiple sclerosis : interrelations between fatigue
complaints, cerebral MRI abnormalities and neurological disability. J Neurol Sci.
Van der Werf SP, van den Broek HLP, Anten HWM, Bleijenberg G. Experience of
severe fatigue long after stroke and its relation to depressive symptoms and disease
characteristics. Eur Neurol. 2001;45:28-33.
van Eijsden HM, van de Port IGL, Visser-Meily JMA, Kwakkel G. Poststroke Fatigue:
Who Is at Risk for an Increase in Fatigue? Stroke Research and Treatment.
Van Zandvoort M, Kappelle L, Algra A, De Haan E. Decreased capacity for mental
effort after single supratentorial lacunar infarct may affect performance in everyday
life. J Neurol Neurosurg Psychiatry. 1998;65:697-702.
Vercoulen J, Hommes O, Swanink C, Jongen P, Fennis J, Galama J, et al. The
measurement of fatigue in patients with multiple sclerosis : a multidimensional
comparison with patients with chronic fatigue syndrome and healthy subjects. Arch
Williams L, Weinberger M, Harris L, Clark D, Biller J. Development of a Stroke-
Specific Quality of Life Scale. stroke. 1999;30:1362-9.
Wozniak MA, Kittner SJ, Price TR, Hebel JR, Sloan MA, Gardner JF. Stroke Location
In Not Associated With Return to work After First Ischemic Stroke. Stroke.
Table 1. Socio-demographical characteristics, cognitive status and FAI scores Download full-text
6 months 12 months
72 M 66M
Normal cognitive score a
Mean FAI score
FAI score >4
a Patients showing normal performances in more than 90% of the neuropsychological tests
b Among patients involved in professional lucrative activity