(Stroke. 2000;31:637.)
© 2000 American Heart Association, Inc.
Original Contributions |
Functional and Neuroanatomic Correlations in Poststroke Depression
The Sunnybrook Stroke Study
From the Department of Psychiatry (A.S., N.H.); Medicine (Neurology), Research Program in Aging (S.E.B., F.S.L., P.L.E., J.L.); and Research Design and Biostatistics (J.P.S.), Sunnybrook & Women’s College Health Sciences Centre and University of Toronto, Toronto, Canada.
Correspondence to Dr Sandra E. Black, Head, Division of Neurology, Sunnybrook & Women’s College Health Sciences Centre, 2075 Bayview Ave, Room A421, Toronto, Ontario M4N 3M5, Canada. E-mail black{at}sten.sunnybrook.utoronto.ca
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Background and Purpose—The purpose of our study was to determine the functional and neuroanatomic correlates of poststroke depressive symptoms.
Methods—Patients with consecutive admissions to a regional stroke center for new-onset unilateral hemispheric stroke who met World Health Organization and National Institute of Neurological and Communicative Disorders and Stroke criteria were eligible for inclusion in a longitudinal study. Acutely, patients underwent CT scanning, and at 3 months and 1 year after stroke, depressive symptoms were assessed by using both the Montgomery-Asberg Depression Rating Scale and the Zung Self-Rating Depression Scale. The Functional Independence Measure (FIM) served as an indication of functional outcome and was obtained at 1 month, 3 months, and 1 year after stroke, along with other demographic information. The Talairach and Tournoux stereotactic atlas was used for the primary determination of CT lesion localization. Lesion proximity to the anterior frontal pole was also measured.
Results—Eighty-one patients participated in the longitudinal study. Stepwise linear regression analyses generated a highly significant model (F3,76=9.8, R2=28%, P<0.0005), with lower 1-month total FIM scores, living at home, and damage to the inferior frontal region predicting higher depression scores at 3 months. Similarly, lower 3-month total FIM scores correlated with higher 3-month depression scores, and lower 1-year total FIM scores correlated with higher 1-year depression scores.
Conclusions—Functional measures correlated with poststroke depression across time and, together with neuroanatomic measures, predicted depressive symptoms longitudinally. Although inferior frontal lesion location, irrespective of side, appeared to play a role as a risk factor in this study, the degree of functional dependence after stroke imparted the greatest risk.
Key Words: depression • stroke assessment • stroke outcome • tomography, x-ray computed
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Excessive mortality and significant cognitive, physical, and psychosocial morbidity may occur after stroke. Mood disturbance can be an important sequela of stroke.1 Estimates of poststroke depression range from 18% to 78%,2 3 with the greatest risk occurring in the first 2 years, especially at 3 to 6 months after stroke.4 5 6
Sparked by the early laboratory rat studies of Robinson and Coyle,7 8 attempts have been made by many researchers to determine the importance of lesion characteristics in depression after stroke in humans. Robinson, Starkstein, and colleagues3 4 5 9 10 11 12 have repeatedly demonstrated a specific relationship between left anterior or basal ganglia stroke lesion location and depression in the early stage of stroke, a finding that they attribute to asymmetry in the hemispheric distribution of catecholamines. Specifically, they speculated that disrupted cortical noradrenergic pathways originating in the locus ceruleus and projecting to the frontal cortex and then to the posterior cortical regions explain the association between anterior lesion location and the severity of depression.13 In a positron emission tomographic study,14 they suggested that failure to upregulate serotonin receptors after left hemisphere damage (LHD) contributed to the development of depression.
Although the role of lesion laterality and frontal proximity in poststroke depression has been widely accepted, attempts to replicate have yielded conflicting results, and numerous methodological issues remain unresolved. The original 1983 study (Robinson et al9 ) had 48 patients consisting largely of middle-aged African American men from lower socioeconomic circumstances in a large urban center. Of the LHD patients, 30% had aphasia, and 25% had no visible CT lesions. LHD patients had higher depression scores than did patients with right hemisphere damage (RHD) in the acute phase of stroke, and left anterior lesions were associated with greater frequency and severity of depression. Given the particular characteristics of this sample, generalizability may not hold. Some studies3 10 11 15 have replicated the findings of Robinson et al, but other studies2 16 17 18 19 20 have not, and comprehensive reviews21 22 23 24 25 attributed the divergent results to methodological differences, including cohort selection from different settings,26 previous history of psychiatric or neurological illness,24 definition and classification of stroke,21 phenomenology and nosology of depressive symptoms,23 and the validity of the rating methodology used.25 More recently, Shimoda and Robinson27 have suggested that time elapsed after onset may be the key factor, with the left frontal correlation applying primarily to depression in acute stroke. Very few studies have provided detailed lesion localization15 28 ; most simply classify lesions as anterior or posterior or measure proximity to the frontal pole.
Cognizant of the methodological pitfalls inherent to this area of research, the authors designed the present study to follow prospectively a large relatively unselected sample of carefully diagnosed acute-stroke patients and to determine the frequency, severity, and persistence of depressive symptoms and their clinical and neuroanatomic correlates. A clearer understanding of the relationship between neurological damage and depression after stroke would provide insight into the neurobiology of mood disorders and assist clinicians in the early identification of patients at highest risk for mood disturbance and those most likely to benefit from treatment interventions. This, in turn, might lead to shortened hospital stays, contained healthcare costs, improved quality of life, and reduced mortality.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Site and Patients
The Sunnybrook Stroke Study29 was conducted at a university-affiliated health science center serving as a regional stroke center to a predominantly middle-class residential northern suburb of Toronto, Ontario, Canada, with a catchment area population of
250 000 residents. Consecutive admissions from August
1990 and May 1993 were eligible for inclusion in the present study.
Diagnosis of stroke was made clinically by an experienced stroke
neurologist using World Health Organization Monitoring Trends and
Determinants in Cardiovascular Disease (MONICA)
Project30 and National Institute of Neurological and
Communicative Disorders and Stroke (NINCDS)31 criteria. A
multidisciplinary stroke care team including neurology, psychiatry,
occupational therapy, physiotherapy, speech therapy, and social work
was involved with each patient according to a standardized stroke care
protocol, and rehabilitation was provided as appropriate, as part of
the universal healthcare service. All patients underwent CT scanning,
and those with subarachnoid hemorrhage and
vertebrobasilar strokes were excluded as described
previously.29 Patients with aphasia were not excluded and
were always assessed unless they suffered from global aphasia or severe
comprehension deficits as assessed by the Western Aphasia
Battery.32 Written and informed consent was obtained from
all patients or their substitute consent givers. An important and unique feature of the present localization study was that a careful accounting of all exclusions was provided to give perspective on generalizability. By our predetermined exclusion criteria, the following 248 patients had to be excluded from the 449 originally entered into the study: 114 who died (3 with bilateral hemisphere damage [BHD], 56 with LHD, and 55 with RHD), 10 who experienced a subsequent stroke while in the study, 8 with BHD, 72 with previous history of stroke (41 with LHD and 31 with RHD), and 44 with negative CT scans (18 with LHD and 26 with RHD). Of the 201 patients remaining, 112 (65 with LHD and 47 with RHD) were unable to complete the depression questionnaires for the following reasons: 27 were too ill medically or neurologically to cooperate with assessment (15 with LHD and 12 with RHD), 12 were too cognitively impaired (7 with LHD and 5 with RHD), 5 experienced a subsequent stroke before 3 months (4 with LHD and 1 with RHD), 4 were unable to speak English (1 with LHD and 3 with RHD), 24 had moved away or were unable to be located (13 with LHD and 11 with RHD), 23 refused to participate (10 with LHD and 13 with RHD), and 17 had comprehension deficits too severe to meaningfully understand the questions (15 with LHD and 2 with RHD). The number of patients excluded was counterbalanced between RHD and LHD, except for aphasia, yielding 89 patients for analysis. Finally, 8 patients (2 with LHD and 6 with RHD) were on antidepressants at 1 month and were analyzed separately because of the possible effect on depression scores of treated or partially treated states. This yielded a total of 81 patients (29 with LHD and 52 with RHD) in the localization study. This attrition underlines the challenge of conducting prospective studies in acute-stroke populations.
Measures
Shortly after admission, a standardized questionnaire was used
to collect data on demographics, concurrent illness, past medical and
psychiatric history, and medications for all acute-stroke patients.
Those patients who consented to participate in the longitudinal study
were followed up at 1, 3, and 12 months after stroke and were
administered an assessment battery comprising demographics, depressive
symptomatology, and functional assessment.
Demographics
Demographics included age at stroke onset, sex, concomitant
illness, and history of depression. Living arrangement was recorded
dichotomously as residing at home or not, and partner status was coded
as married or not. These variables were selected as possible
important covariates to be entered in all regression analyses.
Antidepressant medication use was also recorded at each
patient visit.
Depressive Symptomatology
Depressive symptomatology was evaluated at 3 and 12 months after
stroke by a research nurse blind to lesion location and trained by a
psychiatrist specializing in management of poststroke depression. Three
months was chosen for initial depressive symptom assessment because it
was close enough to the stroke to be temporally connected with the
event but allowed sufficient time for early transient mood changes
related to the acute insult to have settled. The objective
observer-rated Montgomery-Asberg Depression Rating Scale
(MADRS)33 is a 20-item scale measuring symptom severity,
with a score 
7 indicating depressive symptomatology.34 A
subjective measure, the Zung Self-Rated Depression Scale
(SDS)35 36 was obtained, with a score 50 indicating
depressive symptomatology. Both scales measure severity and frequency
dimensions and use complementary reporting techniques, ie, observer
rating and self-report. A combined factor score was derived from the
MADRS and SDS by using principal components factor
analysis.37 Only 1 factor emerged from the factor
analysis, which accounted for 78% of the variance and had an
eigenvalue of 1.6. A cutoff score for the combined factor score was
calculated by maximizing sensitivity and specificity to both the MADRS
and SDS. A score above the cutoff was categorized as displaying
depressive symptomatology. Sensitivity and specificity of the combined
score to the MADRS were 73% and 86% and to the SDS were 90% and
88%, respectively. Last, any patient suspected by the stroke care team
to be clinically depressed was assessed and treated as warranted by a
consulting psychiatrist.
Functional Assessment
Functional assessment used the 18-item Functional Independence
Measure (FIM),38 which scores levels of dependence from 18
(total assistance in all areas) to 126 (complete independence in all
areas). In a previously published study,29 the FIM was
shown to be highly correlated with both the standardized Hemispheric
Stroke Scale39 and the Oxford Handicap
Scale.40 A total FIM score generated through summing
subscale scores assessing self-care, sphincter control, mobility,
locomotion, communication, and social cognition served as the overall
measure of neurological status and functioning. FIM data were collected
at 1 month, 3 months, and 1 year after stroke.
CT Analysis
CT scans were used for lesion localization and volume. CT
scanning was generally carried out within 48 hours of hospital
admission on a GE 9800 CT scanner (General Electric Medical Systems).
Because CT scans may be negative if performed early after stroke,
patients with an initially negative scan usually underwent repeat
scanning within the first week after stroke. When lesions were evident
on >1 CT scan, the film that maximally represented the
lesion was used. According to a standardized stroke protocol, axial
scans were performed parallel to the orbital-meatal line, with a 1-cm
slice thickness, and printed on x-ray film. The films were digitized
with a Konica KFDR-S laser film scanner (Konica Corp) and saved as a
bitmap file. The resulting image resolution was 8 bits per pixel
with 175 µm of the CT image being represented by
each pixel.
Neuroanatomic localization was determined with observers blind to the clinical data. For the primary determination of localization, lesions on CT scans were matched to the best fitting template from a set of 24 transverse slices selected from the Talairach and Tournoux atlas,41 a commonly used stereotactic anatomic reference system. To obtain an estimation of the degree of damage, a quantification method devised by Leibovitch et al42 was used to index the vertical extent of a lesion according to the ratio of the number of slices in which a particular region was damaged over the total number of slices in which that same region appeared in the atlas. Quantification by this method was derived for the following 15 brain regions of interest: superior, middle and inferior frontal, medial frontal/anterior cingulate, sensorimotor strip, parietal, superior, and middle/inferior temporal, medial and lateral occipital, basal ganglia, thalamus, and anterior, central, and posterior white matter. Because there was a specific a priori interest in the relation of the frontal lobe and poststroke depression, 4 frontal subregions were examined. To improve skewness and kurtosis, square root transformation was applied to the CT data.
A further method sometimes used to analyze the effects of neurological damage on clinical outcome is the overlap of lesions of a small sample of patients grouped according to common criteria. Some stroke patients were diagnosed to be clinically depressed early in their course and were on antidepressants at 1 month of poststroke follow-up. Treatment of their depression frequently brought their depressive symptom scores into the normal range, precluding them from lesion analyses in the larger sample. Because their lesion localization was relevant to the lesion correlations, their lesions were traced onto 1 template series to evaluate the distribution of damage. The overlap method was then used to find areas of common damage.
To allow comparison with previous literature on the localization of
damage in poststroke depression, a replication of the methodology
reported by Robinson and colleagues43 44 45 was undertaken.
Lesions traced on the digitized CT scans were used to generate
numerical ratios (AP ratio), which were determined by measuring the
distance of the most anterior border of the lesion from the frontal
pole and dividing this by the overall anterior-posterior (AP) length of
the cerebral hemisphere in the same slice (Figure 1
).28 46 For those scans in
which the lesion was visible on multiple slices, an average based on
the AP ratio of each slice was calculated and used as the measure of
proximity of the lesion to the frontal pole. AP ratio intrarater and
interrater reliability was 0.98 and 0.85, respectively, with use of
intraclass correlation. In keeping with previous
studies,43 44 45 lesions were classified into 4 groups in
relation to the AP diameter, namely, anterior, posterior, extended, and
intermediate as demonstrated in Figure 1, panels A to D,
respectively. AP measures were calculated online on a 17-in computer
monitor with the use of Adobe PhotoShop 4.0 (Adobe Systems Inc).
|
Lesion volume was measured on lesion tracings with the use of a digitizing tablet (Sigma Scan, version 3.0, Jandel Scientific). The lesion area was traced on each 1-cm-thick slice, and the areas were summed to derive the volume.
Data Analysis
Demographic and functional covariates followed by the CT data
were entered into multiple linear regression analyses for the
purposes of predicting and correlating with the combined depression
scores at 3 months and 1 year. Similar regression equations were
repeated with the AP ratio data. Five models were tested, 3 of which
were predictive in nature and 2 of which were correlative. For the 3
predictive models, we used 1-month data to predict 3-month depression
scores (model 1), 1-month data to predict 1-year depression scores
(model 2), and 3-month data to predict 1-year depression scores (model
3). For the correlative models, we correlated 3-month data with 3-month
depression scores (model 4) and 1-year data with 1-year depression
scores (model 5). Thus, we used a Bonferroni correction of 
=0.05/5
to correct for multiple models. To make some adjustment for allowing
multiple variables the opportunity to enter the model, we also set
the entry criterion with an acceptable level of error at 1%. We opted
for the multiple linear rather than logistic regression modeling
because dichotomization of the depression variables was arbitrary,
and we believed that using the depression score as a continuous
variable would be more sensitive.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Patients participating in the longitudinal study have been described previously,29 and their characteristics are briefly summarized in Table 1
. At
3 months, 36% of the sample (47% of RHD patients and 17% of LHD
patients) were rated as having depressive symptoms on the basis of
their combined depression scores. Comparisons between patients with and
without marked depressive symptoms were conducted by use of
univariate statistics (Table 1). There were no
significant differences between the groups apart from those with
depressive symptoms scoring more poorly on the FIM (86 versus 107,
P<0.0005) and demonstrating greater frequency of damage
(34% versus 8%, P<0.03) and vertical extent of damage
(2.0 versus 0.4, P<0.02) to the inferior
frontal region. Although 83% of those with depressive symptoms had RHD
compared with 52% in the group without depressive symptoms
(P<0.001), this right hemisphere predominance did not
emerge in the multivariate analyses.
|
Stepwise least squares linear regression was conducted with the
combined depression score as the dependent measure. Age, sex, living
arrangements, marital status, history of depression, and total FIM
score representing the demographic and functional
covariates, followed by CT lesion variables (including CT lesion
volume and side of stroke), were all entered as potential predictors of
depression score (Table 2). A
highly significant model was produced, with 1-month covariate data
predicting depression scores at 3 months
(F3,76=9.8, P<0.0005), accounting for
28% of the variance. The significant variables contributing to
this model were lower total FIM score at 1 month, living at home, and
damage to the inferior frontal region. Regression performed
with only the 15 CT regions revealed that damage to the
inferior frontal region still emerged as the only
significant CT variable (P<0.001, standardized
ß=0.36) predicting depression scores, accounting for 13% of the
model variance. Predictive models for 1-year depression scores were not
significant.
|
Regression analyses were repeated with AP ratio data instead of
the 15 CT lesion variables, and the results were similar to those
reported in the preceding paragraph. Lower 1-month FIM score
(R2=15%, standardized ß=-0.62),
living at home (R2=11%, standardized
ß=-0.45), and lower AP ratios
(R2=9%, standardized ß=-0.31),
representing more frontal lesion location, significantly
predicted higher 3-month depression scores (P<0.05),
accounting for 35% of the model variance. Regression performed with
only the AP ratio also revealed a significant model predicting
depression scores at 3 months (P<0.001, standardized
ß=-0.38), accounting for 15% of the model variance. When the
qualitative AP lesion categories were analyzed by
2 test of independence, no significant
differences emerged in relation to depressive symptom category (ie,
depressed versus not depressed).
Correlative models (ie, correlating 3-month covariate data with 3-month depression scores and 1-year covariate data with 1-year depression scores) were also significant. FIM scores were strongly correlated with depression scores at 3 months (r=0.38, P<0.0005) and at 1 year (r=0.36, P<0.002). No CT variables entered the correlation analyses.
CT scan overlaps were carried out to compare those patients on
antidepressant medication with those patients with or without
depressive symptoms that were not on antidepressants (Figure 2). As illustrated, patients with
depressive symptoms, including those on antidepressants, had the most
significant overlap in the central region of the centrum
semiovale.
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Summary of Findings
The major finding of the present study is that functional status in activities of daily living is the major correlate of depressive symptoms at each assessment and that this status at 1 month is the strongest predictor of 3-month depressive symptoms. Inferior frontal damage or proximity to the frontal pole as measured by the AP diameter (along with living arrangements) strengthens the prediction. The finding that functional status was an important predictor is consonant with previous studies.5 16 29 47 48 49 That patients living at home by 1 month were more likely to have depressive symptoms at 3 months may at first seem counterintuitive47 because their functional status was higher (mean FIM of 118±8 for patients living at home versus mean FIM of 91±21 for patients not living at home). A plausible psychosocial explanation, however, might be that discharged patients were facing the impact of the stroke on their previous daily routine at this time, whereas those still in acute care or in rehabilitation had continuing multidisciplinary team support and were preoccupied with their physical recovery.
Methodological Issues
The present study distinguishes itself from previously
published work in this area by the prospective follow-up of a large
stroke sample coupled with detailed CT localization methodology. The
authors chose to replicate frequently cited CT analysis
methods43 44 45 to ensure the comparability of the
present study with others but also carefully classified regional
brain damage on the basis of a widely used standard reference
system.41 Several relevant risk factors for poststroke
depression, including age, sex, past history of depression, living
arrangement, marital status, and degree of functional independence,
were entered into each multivariate analysis
along with the CT lesion parameters to elucidate the
relative contribution of each potential risk factor. Few previous
studies have had a sufficient sample size to allow detailed lesion
information to be included in the multivariate
analysis. Furthermore, few studies have accounted as
meticulously for all exclusions. The 4:1 screening-to-entry ratio of
the survivors is not atypical for a lesion localization study and
demonstrates the difficulties inherent in sampling a consecutive stroke
population, whose average age of 70 years was more
representative of other stroke registries than the
patients in the original seminal studies of Robinson and
colleagues.3 4 5 6 7 8 9 10 11 12 Forty percent of the survivors were
disqualified by lesion requirements (eg, a positive scan with a single
lesion), and depression scores could not be obtained in a third of the
remainder. Exclusions were counterbalanced between the risk factors,
including the 7% who refused participation, except for more LHD
participants with aphasia too severe for assessment. This resulted in a
smaller number of LHD patients in the sample. Therefore, it is possible
that depressive symptoms were underestimated, but these essentially
nonverbal patients would not be assessable by
Diagnostic and Statistical Manual of Mental
Disorders, edition 3 (DSM-III) criteria either. If they were
assessed clinically to be depressed by the multidisciplinary team, they
were seen by a geriatric psychiatrist who prescribed antidepressants on
the basis of clinical judgment. For the purpose of the present
study, such individuals would have been classified as depressed if they
were on antidepressants.
Localization Within and Between Hemispheres
The landmark study by Robinson et al,9 which
generated great interest in lesion patterns and poststroke depression,
reported data on a unique sample of mostly younger black males of low
socioeconomic status. Stroke registries and community studies of stroke
patients50 51 reveal sample characteristics more closely
aligned with the present study population. Despite this difference
in sampling, our findings are consistent with those of Morris
et al,15 who reported a specific relationship between
lesion proximity to the frontal pole and severity of poststroke
depression. Unlike these investigators, our research did not
demonstrate hemispheric lateralization of depressive symptoms. There
are several possible explanations for this discrepancy, including
differences in sample selection already discussed as well as depression
assessment and statistical analysis. One important reason may
be the time of assessment, which was subacute in the present
study, with follow-up at 1 year. The present study measured
depressive symptomatology with 2 widely accepted, valid, and reliable
depression scales. Other researchers18 20 demonstrating
left-right differences in poststroke depression relied on modified
diagnostic criteria and clinical interviews. The lack of
structured diagnostic interviews for depression could be
considered a potential limitation of the present study. In an
earlier published report on this population, however, the easily
administered MADRS and SDS both demonstrated acceptable external and
concurrent validity, as well as internal
consistency.29 Furthermore, these scales
provide continuous data, which permit multiple regression
analysis. Right stroke predominance has been reported in some
other poststroke depression studies,2 19 and
univariate statistical analysis of our data
revealed a right hemisphere predominance in patients with depressive
symptoms. However, multivariate analyses, which
included other potential factors in poststroke depression, failed to
maintain any effect of lesion laterality. Noteworthy is the relatively
lower incidence of depressive symptoms in the LHD patients. In our
sample, depressive symptoms emerged in almost half of the RHD patients
compared with 17% of the LHD patients. Interestingly, a recent review
of the growing literature on poststroke depression reveals that the
majority of CT lesion localization studies to date do not demonstrate
hemispheric specificity for depressive symptoms.24
The frontal lobe, however, has repeatedly emerged as playing an important role in mood regulation. Numerous structural and functional neuroimaging studies of affective disorders have been conducted on normal control subjects and those with primary and secondary depression, in addition to those studies in poststroke depression that have found an anterior lesion correlation.2 3 9 28 43 46 47 52 53 Positron emission tomographic imaging data suggest that inferior and orbitofrontal activation54 and anterior cingulate regions55 are related to behavioral and subjective changes in the affective experience of self-induced dysphoria. Working with patients experiencing primary depression, Sackeim et al56 revealed depressed patients to have a global decrease in cortical blood flow and a variety of regional cerebral blood flow changes, including decreased flow to inferior frontal regions. Greater left anterior cerebral blood flow was also noted in depressed patients at rest than in normal control subjects at rest by Uytdenhoef et al.57 A careful study of depressed patients suffering from unipolar depression, bipolar depression, and obsessive-compulsive disorder with secondary depression by Baxter et al58 demonstrated a common reduction of left anterolateral prefrontal cortex glucose metabolism in all 3 depression types. Studies of depressed patients with brain injury other than stroke have also pointed to frontal lobe dysfunction. Mayberg et al59 demonstrated an inverse relationship between the magnitude of hypometabolism in the inferior and orbital regions of the frontal lobe and the severity of depression in patients with Parkinson’s disease. Likewise, in early Huntington’s disease, hypometabolism of the orbital frontal-inferior prefrontal cortex distinguished depressed from nondepressed patients and normal control subjects.60
Emotional Valence
The neurological basis of emotion has intrigued clinicians and
researchers alike. The occurrence of poststroke depression provides the
opportunity for examination of potential neuroanatomic correlates of
mood disturbance. In an early study of brain-damaged patients,
Ross and Rush61 used clinicopathological conditions to put
forward an initial model for the neurology of depression in which both
hemispheres participate, each modulating certain signs and symptoms.
Around the same time, Tucker62 reviewed the literature on
hemispheric lateralization of human emotion, and Alexander et
al63 proposed that limbic and prefrontal striatal pathways
mediate human emotion. Hypotheses implicating the role of nonmotor
basal ganglia–thalamocortical–frontocortical circuits in the
modulation of certain affective states may help explain how lesions
lying along tracts traversing the basal ganglia or frontal cortex would
result in a dysfunction of the circuit and, consequently, mood
dysregulation.
The present study attempts to uncover the functional and neuroanatomic correlations of poststroke depression. Although inferior frontal lesion location appears to play a role as a risk factor, the degree of functional dependence after stroke imparts the greatest risk in the present study. This finding highlights the need for clinicians to carefully screen for depressive symptoms in all stroke patients, regardless of lesion location.
|
Acknowledgments |
|---|
This study was supported by grants to Dr Black from the Heart and Stroke Foundation of Canada. We wish to express thanks to K. Barbour, who assisted with the lesion localization database, and Dr M. Yaffe for access to the CT film scanner.
Received July 26, 1999; revision received November 23, 1999; accepted December 8, 1999.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
1. Folstein MF, Maiberger R, McHugh PR. Mood disorder as a specific complication of stroke. J Neurol Neurosurg Psychiatry. 1977;40:1018–1020.
2. Sharpe M, Hawton K, House A, Molyneux A, Sandercock P, Bamford J, Warlow C. Mood disorders in long-term survivors of stroke: associations with brain lesion and location and volume. Psychol Med. 1990;20:815–828.[Medline] [Order article via Infotrieve]
3.
Robinson RG, Kubos KL, Starr LB, Rao K, Price TR. Mood
disorders in stroke patients: importance of location of lesion.
Brain. 1984;107:81–93.
4.
Robinson RG, Starr LB, Price TR. A two year
longitudinal study of mood disorders following stroke: prevalence and
duration at six months follow-up. Br J Psychiatry. 1984;144:256–262.
5.
Robinson RG, Bolduc PL, Price TR. Two-year
longitudinal study of poststroke mood disorders: diagnosis and outcome
at one and two years. Stroke. 1987;18:837–843.
6.
Parikh RM, Robinson RG, Lipsey JR, Starkstein SE,
Fedoroff JP, Price TR. The impact of poststroke depression on recovery
in activities of daily living over a 2-year follow-up. Arch
Neurol. 1990;47:785–789.
7.
Robinson RG. Differential behavioral and biochemical
effects of right and left hemispheric cerebral infarction in the rat.
Science. 1979;205:707–710.
8. Robinson RG, Coyle JT. The differential effect of right versus left hemispheric cerebral infarction on catecholamines and behavior in the rat. Brain Res. 1980;188:63–78.[Medline] [Order article via Infotrieve]
9. Robinson RG, Kubos KL, Starr LB, Rao K, Price TR. Mood changes in stroke patients: relationship to lesion location. Compr Psychiatry. 1983;24:555–566.[Medline] [Order article via Infotrieve]
10.
Starkstein SE, Robinson RG, Price TR. Comparison of
cortical and subcortical lesions in the production of
poststroke mood disorders. Brain. 1987;110:1045–1059.
11.
Starkstein SE, Robinson RG, Berthier ML, Parikh RM,
Price TR. Differential mood changes following basal ganglia vs thalamic
lesions. Arch Neurol. 1988;45:725–730.
12.
Starkstein SE, Cohen BS, Fedoroff P, Parikh RM, Price
TR, Robinson RG. Relationship between anxiety disorders and depressive
disorders in patients with cerebrovascular injury. Arch Gen
Psychiatry. 1990;47:246–251.
13. Robinson RG. The Clinical Neuropsychiatry of Stroke: Cognitive, Behavioral, and Emotional Disorders Following Vascular Brain Injury. Cambridge, UK: Cambridge University Press; 1998.
14.
Mayberg HS, Robinson RG, Wong DF, Parikh RM, Bolduc PL,
Starkstein SE, Price TR, Dannals RF, Links JM, Wilson AA, Hayden TR,
Wagner HN Jr. PET imaging of cortical S2
serotonin receptors after stroke: lateralized changes and
relationship to depression. Am J Psychiatry. 1988;145:937–943.
15.
Morris PLP, Robinson RG, Raphael B, Hopwood MJ. Lesion
location and poststroke depression. J Neuropsychiatry Clin
Neurosci. 1996;8:399–403.
16. Schwartz JA, Speed NM, Brunberg JA, Brewer TL, Brown M, Greden JF. Depression in stroke rehabilitation. Biol Psychiatry. 1993;33:694–699.[Medline] [Order article via Infotrieve]
17.
Stern RA, Bachman DL. Depressive symptoms following
stroke. Am J Psychiatry. 1991;148:351–356.
18.
Morris PLP, Robinson RG, de Carvalho ML, Albert P,
Wells JC, Samuels JF, Eden-Fetzer D, Price TR. Lesion characteristics
and depressed mood in the stroke data bank study. J
Neuropsychiatry Clin Neurosci. 1996;8:153–159.
19.
Sinyor D, Jacques P, Kaloupek DG, Becker R,
Goldenberg M, Coopersmith H. Poststroke depression and lesion location.
Brain. 1986;109:537–546.
20. Dam H, Pedersen HE, Ahlgren P. Depression among patients with stroke. Acta Psychiatr Scand. 1989;80:118–124.[Medline] [Order article via Infotrieve]
21. Primeau F. Post-stroke depression: a critical review of the literature. Can J Psychiatry. 1988;33:757–765.[Medline] [Order article via Infotrieve]
22. Starkstein SE, Robinson RG. Depression following cerebrovascular lesions. Semin Neurol. 1990;10:247–253.[Medline] [Order article via Infotrieve]
23. Ramasubbu R, Kennedy SH. Factors complicating the diagnosis of depression in cerebrovascular disease, 1: phenomenological and nosological issues. Can J Psychiatry. 1994;39:596–600.[Medline] [Order article via Infotrieve]
24. Singh A, Herrmann N, Black SE. The importance of lesion location in poststroke depression: a critical review. Can J Psychiatry. 1998;43:921–927.[Medline] [Order article via Infotrieve]
25.
House A. Depression associated with stroke.
J Neuropsychiatry Clin Neurosci. 1996;8:453–457.
26. Agrell B, Dehlin O. Depression in stroke patients with left and right hemisphere lesions: a study in geriatric rehabilitation in-patients. Aging (Milano). 1994;6:49–56.[Medline] [Order article via Infotrieve]
27. Shimoda K, Robinson RG. The relationship between poststroke depression and lesion location in long-term follow-up. Biol Psychiatry. 1999;45:187–192.[Medline] [Order article via Infotrieve]
28.
Herrmann M, Bartels C, Schumacher M, Wallesch CW.
Poststroke depression: is there a pathoanatomic correlate for
depression in the postacute stage of stroke? Stroke. 1995;26:850–856.
29.
Herrmann N, Black SE, Lawrence J, Szekely C, Szalai JP.
The Sunnybrook Stroke Study: a prospective study of depressive symptoms
and functional outcome. Stroke. 1998;29:618–624.
30. WHO MONICA Project Principal Investigators. The World Health Organization MONICA Project (monitoring trends and determinants in cardiovascular disease): a major international collaboration. J Clin Epidemiol.. 1988;41:105–114.[Medline] [Order article via Infotrieve]
31.
Foulkes MA, Wolf PA, Price TR. The Stroke Data Bank:
design, methods and baseline characteristics. Stroke. 1988;19:547–554.
32. Kertesz A. Western Aphasia Battery. New York, NY: Grune & Stratton; 1982.
33.
Montgomery SA, Asberg M. A new depression scale
designed to be sensitive to change. Br J Psychiatry. 1979;134:382–389.
34.
Snaith RP, Harrop FM, Newby DA, Teale C. Grade scores
of the Montgomery-Asberg depression and the clinical anxiety scales.
Br J Psychiatry. 1986;148:599–601.
35. Zung WK. A self-rating depression scale. Arch Gen Psychiatry. 1965;12:67–70.
36. Zung WW, Magruder-Habib K, Velez R, Alling W. The comorbidity of anxiety and depression in general medical patients: a longitudinal study. J Clin Psychiatry. 1990;51(suppl):77–80. Discussion.
37. Kim J, Mueller CW. Quantitative applications in the social sciences. In: Factor Analysis: Statistical Methods and Practical Issues. Beverly Hills, Calif: Sage Publications; 1978. Series 14.
38. Keith RA, Granger CV, Hamilton BB, Sherwin FS. The functional independence measure: a new tool for rehabilitation. Adv Clin Rehabil. 1987;1:6–18.[Medline] [Order article via Infotrieve]
39.
Adams RJ, Meador KJ, Sethi KD, Grotta JC, Thomson DS.
Graded neurological scale for use in acute hemisphere stroke treatment
protocols. Stroke. 1987;18:665–669.
40. Bamford JM, Sandercock PAG, Warlow CP, Slattery J. Interobserver agreement for the assessment of handicap in stroke patients. Stroke. 1989;20:828. Letter.[Medline] [Order article via Infotrieve]
41. Talairach J, Tournoux P. Co-Planar Stereotaxic Atlas of the Human Brain. New York, NY: Georg Thieme; 1988.
42.
Leibovitch FS, Black SE, Caldwell CB, Ebert PL, Ehrlich
LE, Szalai JP. Brain-behavior correlations in hemispatial neglect using
CT and SPECT: the Sunnybrook Stroke Study. Neurology. 1998;50:901–908.
43. Robinson RG, Szetela B. Mood change following left hemispheric brain injury. Ann Neurol. 1981;9:447–453.[Medline] [Order article via Infotrieve]
44. Robinson RG, Benson DF. Depression in aphasic patients: frequency, severity, and clinical-pathological correlations. Brain Lang. 1981;14:282–291.[Medline] [Order article via Infotrieve]
45.
Lipsey JR, Robinson RG, Pearlson GD, Rao K, Price TR.
Mood change following bilateral hemisphere brain injury. Br
J Psychiatry. 1983;143:266–273.
46.
House A, Dennis M, Warlow C, Hawton K, Molyneux A. Mood
disorders after stroke and their relation to lesion location.
Brain. 1990;113:1113–1129.
47.
Astrom M, Adolfsson R, Asplund K. Major depression in
stroke patients: a 3-year longitudinal study. Stroke. 1993;24:976–982.
48.
Ramasubbu R, Robinson RG, Flint AJ, Kosier T, Price TR.
Functional impairment associated with acute poststroke depression: the
stroke data bank study. J Neuropsychiatry Clin
Neurosci. 1998;10:26–33.
49.
Pohjasvaara T, Leppavuori A, Siira I, Vataja R, Kaste
M, Erkinjuntti T. Frequency and clinical determinants of poststroke
depression. Stroke. 1998;29:2311–2317.
50.
Burvill PW, Johnson GA, Jamrozik KD, Anderson CS,
Stewart-Wynne EG, Chakera TM. Prevalence of depression after stroke:
the Perth Community Stroke Study. Br J Psychiatry. 1995;166:320–327.
51.
Bamford J, Sandercock P, Dennis M, Warlow C, Jones L,
McPherson K, Vessey M, Fowler G, Molyneux A, Hughes T. A prospective
study of acute cerebrovascular disease in the community: the
Oxfordshire Community Stroke Project 1981–86, 1: methodology,
demography and incident cases of first-ever stroke. J Neurol
Neurosurg Psychiatry. 1988;51:1373–1380.
52.
Eastwood MR, Rifat SL, Nobbs H, Ruderman J. Mood
disorder following cerebrovascular accident. Br J
Psychiatry. 1989;154:195–200.
53.
Robinson RG, Price TR. Post-stroke depressive
disorders: a follow-up study of 103 patients. Stroke. 1982;13:635–641.
54.
Pardo JV, Pardo PJ, Raichle ME. Neural correlates of
self-induced dysphoria. Am J Psychiatry. 1993;150:713–719.
55.
Mayberg HS, Liotti M, Brannan SK, McGinnis S, Mahurin
RK, Jerabek PA, Silva JA, Tekell JL, Martin CC, Lancaster JL, Fox PT.
Reciprocal limbic-cortical function and negative mood: converging PET
findings in depression and normal sadness. Am J
Psychiatry. 1999;156:675–682.
56.
Sackeim HA, Prohovnik I, Moeller JR, Brown RP, Apter S,
Prudic J, Devanand DP, Mukherjee S. Regional cerebral blood flow in
mood disorders, I: comparison of major depressives and normal controls
at rest. Arch Gen Psychiatry. 1990;47:60–70.
57.
Uytdenhoef P, Portelange P, Jacquy J, Charles G,
Linkowski P, Mendlewicz J. Regional cerebral blood flow and lateralized
hemispheric dysfunction in depression. Br J Psychiatry. 1983;143:128–132.
58.
Baxter LRJ, Schwartz JM, Phelps ME, Mazziotta JC, Guze
BH, Selin CE, Gerner RH, Sumida RM. Reduction of prefrontal cortex
glucose metabolism common to three types of depression.
Arch Gen Psychiatry. 1989;46:243–250.
59. Mayberg HS, Starkstein SE, Sadzot B, Preziosi T, Andrezejewski PL, Dannals RF, Wagner HNJ, Robinson RG. Selective hypometabolism in the inferior frontal lobe in depressed patients with Parkinson’s disease. Ann Neurol. 1990;28:57–64.[Medline] [Order article via Infotrieve]
60.
Mayberg HS, Starkstein SE, Peyser CE, Brandt J, Dannals
RF, Folstein SE. Paralimbic frontal lobe hypometabolism in
depression associated with Huntington’s disease. Neurology. 1992;42:1791–1797.
61.
Ross ED, Rush AJ. Diagnosis and neuroanatomical
correlates of depression in brain-damaged patients: implications for a
neurology of depression. Arch Gen Psychiatry. 1981;38:1344–1354.
62. Tucker DM. Lateral brain function, emotion, and conceptualization. Psychol Bull. 1981;89:19–46.[Medline] [Order article via Infotrieve]
63. Alexander GE, DeLong MR, Strick PL. Parallel organization of functionally segregated circuits linking basal ganglia and cortex. Annu Rev Neurosci. 1986;9:357–381.[Medline] [Order article via Infotrieve]
This article has been cited by other articles:
 |
|
 |
|
  L. De Wit, M. Molas, E. Dejaeger, W. De Weerdt, H. Feys, W. Jenni, N. Lincoln, K. Putman, W. Schupp, and E. Lesaffre The Use of a Biplot in Studying Outcomes After Stroke Neurorehabil Neural Repair, October 1, 2009; 23(8): 825 - 830. [Abstract] [PDF]
|
|
|
|
 |
|
 I. G.L. van de Port, G. Kwakkel, I. van Wijk, and E. Lindeman Susceptibility to Deterioration of Mobility Long-Term After Stroke: A Prospective Cohort Study Stroke, January 1, 2006; 37(1): 167 - 171. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. L. Hackett and C. S. Anderson Predictors of Depression after Stroke: A Systematic Review of Observational Studies Stroke, October 1, 2005; 36(10): 2296 - 2301. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Carota, A. Berney, S. Aybek, G. Iaria, F. Staub, F. Ghika-Schmid, L. Annable, P. Guex, and J. Bogousslavsky A prospective study of predictors of poststroke depression Neurology, February 8, 2005; 64(3): 428 - 433. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Vataja, A. Leppavuori, T. Pohjasvaara, R. Mantyla, H. J. Aronen, O. Salonen, M. Kaste, and T. Erkinjuntti Poststroke Depression and Lesion Location Revisited J Neuropsychiatry Clin Neurosci, May 1, 2004; 16(2): 156 - 162. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. K. Bhogal, R. Teasell, N. Foley, and M. Speechley Lesion Location and Poststroke Depression: Systematic Review of the Methodological Limitations in the Literature Stroke, March 1, 2004; 35(3): 794 - 802. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Bogousslavsky William Feinberg Lecture 2002: Emotions, Mood, and Behavior After Stroke Stroke, April 1, 2003; 34(4): 1046 - 1050. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Hanna-Pladdy, K. M. Heilman, and A. L. Foundas Ecological implications of ideomotor apraxia: Evidence from physical activities of daily living Neurology, February 11, 2003; 60(3): 487 - 490. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Berg, H. Palomaki, M. Lehtihalmes, J. Lonnqvist, and M. Kaste Poststroke Depression: An 18-Month Follow-Up Stroke, January 1, 2003; 34(1): 138 - 143. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Vataja, T. Pohjasvaara, A. Leppavuori, R. Mantyla, H. J. Aronen, O. Salonen, M. Kaste, and T. Erkinjuntti Magnetic Resonance Imaging Correlates of Depression After Ischemic Stroke Arch Gen Psychiatry, October 1, 2001; 58(10): 925 - 931. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||