837
Two-Year Longitudinal Study of Poststroke
Mood Disorders: Diagnosis and Outcome at One
and Two Years
Robert G. Robinson, MD, Paula L. Bolduc, MS, and Thomas R. Price, MD
Downloaded from http://stroke.ahajournals.org/ by guest on July 28, 2017
As part of a prospective study of mood disorders in stroke patients, interviews were obtained from 37
patients at 1 year and 48 patients at 2 years follow-up. In-hospital evaluations for these 65 follow-up
patients found that 9 patients (14%) had symptom clusters of major depression, 12 patients (18%) had
symptom clusters of dysthymic or minor depression, and 44 patients (68%) did not meet the DSM i n
diagnostic criteria for depression. Although overall prevalence of depression did not change significantly over time, the prognosis for individual patients, depending on diagnostic group, was different.
All of the follow-up patients with major depression in-hospital were improved by 2 years, with a
significant reduction in their mean depression scores and improvement in their activities of daily
living, whereas only 30% of follow-up patients with dysthymic depression improved by this time.
There was no significant improvement in their mean depression scores or mean activities of daily living
score. Of the patients followed up who were not depressed in-hospital, 34% had developed major or
minor depression by 2 years, and their mean depression scores were significantly increased. These
data suggest that the prevalence of depression among the follow-up patients remains high (between 30
and 40%) for the first 2 years after stroke, but that untreated poststroke major depression has a
natural course of about 1-2 years, with associated improvement in activity of dairy living scores,
whereas the prognosis for poststroke dysthymic depression is frequently unfavorable and often
persists for > 2 years. (Stroke 1987; 18:837-843)
D
uring the past few years, we have reported our
I findings from a group of 103 patients who
were examined during their acute hospitalization for stroke and followed for 2 years. '~3 At 6 months
follow-up, 10 of 13 patients with symptoms of major
depression in-hospital continued to have these symptoms, whereas 5 of 9 patients with minor (dysthymic)
depression in-hospital continued to have these symptoms; the remaining 4 patients with dysthymic depression had developed symptoms of major depression.2
Of the 28 patients who were not depressed at the time
of the initial in-hospital evaluation, 9 had developed
either major or minor depressive symptoms by 6
months follow-up.2 Thus, based on our initial followup, 19 of 22 patients who were depressed in-hospital
snowed no significant improvement 6 months later,
and another 32% of the follow-up patients developed
depression after hospital discharge.
Although the Diagnostic and Statistical Manual HI
(DSM ED)4 does not recognize more than 1 category of
depression in patients with brain injury (i.e., organic
affective disorder), our previous investigations have
From the Departments of Psychiatry (R.G.R., P.L.B.) and Neuroscience (R.G.R), Johns Hopkins University School of Medicine,
and the Departments of Psychiatry (R.G.R.) and Neurology
(T.R.P.), University of Maryland School of Medicine, Baltimore,
Md.
Supported in part by National Institute of Mental Health Research Scientist Development Award (to R.G.R.) MH00163,
NS15178, MH40355, NS15O8O, NS16332, and NS9-2302.
Address for reprints: Robert G. Robinson, MD, Department of
Psychiatry and Behavioral Science, Johns Hopkins University
School of Medicine, Meyer 4-119, Baltimore, MD 21205.
Received December 8, 1986; accepted May 21, 1987.
demonstrated that the diagnostic differentiation into
major and dysthymic depression appears to be important. Major depression was strongly associated with
left anterior lesion location, whereas dysthymic depression occurred primarily with posterior brain injury. 5 The severity of depression, as measured by either the Zung or Hamilton Depression Rating Scales,
was significantly different for patients with major
compared with dysthymic depression.2 A positive dexamethasone suppression test (i.e., failure to suppress
serum cortisol below 5 /ug/ml following 1 mg of dexamethasone) was significantly associated with major
but not dysthymic depression.6 Cognitive impairment
associated with depression was also found only in
stroke patients with major depression, whereas dysthymic depression did not appear to significantly affect
cognitive impairment.7 Thus, there are several lines of
evidence that suggest that the differentiation of major
and dysthymic depression may be an important element in the evaluation of mood disorder associated
with brain injury. In the present study, we examined
whether in-hospital diagnosis of major depression,
dysthymic depression, or no depression was significantly related to diagnostic status or physical or intellectual impairment at 1 and 2 years follow-up.
Subjects and Methods
Study Population
The sample population consisted of patients who
were admitted to the University of Maryland Hospital
for acute treatment of thromboembolic, atherosclerotic, or hemorrhagic stroke and gave informed consent
838
Stroke
Vol 18, No 5, September-October 1987
Downloaded from http://stroke.ahajournals.org/ by guest on July 28, 2017
for participation in this study. They were also included
in the National Institute of Neurological and Communicative Diseases and Stroke (NINCDS) Pilot Stroke
Data Bank.8 A total of 154 consecutive admissions
were evaluated; of these, 43 patients with severe comprehension deficits were excluded, as were 7 patients
who were discharged before an interview could be
done and 1 patient who refused examination. Thus,
103 patients were included in the prospective longitudinal study. During the 24-months follow-up, 8 patients died and 9 patients were lost completely, not
having received any follow-up, while 86 patients were
reexamined at either 3, 6, 12, or 24 months after
stroke. The present study involves 65 patients, 37 seen
at 12 months and 48 seen at 24 months follow-up (20
patients were seen at both times).
exact correspondence between PSE items and DSM i n
symptom criteria, we have shown in previous publications that there is a close correspondence between
DSM HI, Research Diagnostic Criteria, and PSE diagnoses.1 The conversion criteria for making DSM i n
diagnoses from PSE criteria is available upon request.
Evaluations
In-hospital and follow-up neurologic examinations
were done by the attending neurologist using the standardized examination and rating criteria established by
the NINCDS Pilot Stroke Data Bank.8 Neurologic diagnoses were made using the Stroke Data Bank criteria, which are based on both computed tomography
(CT) scan and other laboratory tests as well as on
clinical examination.
Psychiatric examination was done in the hospital by
a psychiatrist or a social worker trained in the administration of these scales in the late morning so that any
possible effect of diurnal mood variation on the interview outcome would be standardized. The psychiatric
examination included 3 quantitative measures of mental state, the Present State Examination (PSE), 9 the
Hamilton Depression Scale,10 and the Zung Depression Scale." The reliability and validity of these measures in this brain-injured population has been demonstrated in previous publications. 11213
In addition to these quantitative measures of depression, overall cognitive impairment was assessed using
the Mini-Mental State Examination (MMSE).14 Scores
range from 0 to 30, with scores < 23 indicating significant impairment. Functional physical impairment
(i.e., activities of daily living, ADL) was quantified
using the Johns Hopkins Functioning Inventory
(JHFI).13 Scores range from 0 to 27, with higher scores
indicating greater impairment. The quality of social
functioning was determined using the Social Functioning Exam (SFE).16 This is a 28-item semistructured
interview assessment of social functioning, with total
scores reported as a fraction of the maximum possible
score. We have demonstrated the reliability and validity of these instruments in this brain-injured population
in previous publications. 1121316
At the follow-up evaluation, a detailed longitudinal
medical and psychiatric history was taken. Any record
of treatment for depression or intervening medical illness was carefully noted.
Results
Study Population and Neurologic Findings
Demographic characteristics for the 37 patients seen
at 12 months and the 48 patients seen at 24 months are
shown in Table 1. To determine whether the follow-up
groups were in some way "aberrant samples" from our
original group of 103, the no-follow-up group demographics were compared with the 2 follow-up groups
(Table 1). There were no significant intergroup differences in any of the demographic data except age. The
no-follow-up group was slightly younger (p < 0.05)
than the follow-up groups.
Neurologic findings for patients evaluated at either
12 or 24 months as well as in-hospital findings in the
no-follow-up group are shown in Table 2. There were
no significant differences in neurologic findings between these groups. Almost all of the patients had a
hemiparesis or monoparesis with or without sensory
deficit.
Ten of the 65 patients seen at either 12 or 24 months
had suffered subsequent strokes by the follow-up inter-
Diagnosis
Using the symptoms elicited from the PSE, DSM HI
diagnoses were generated. Although there is not an
Statistical Analysis
Means and SDs were calculated for all quantitative
measurement scales. Groups were compared using
analysis of variance (ANOVA); a group was compared
over time using repeated-measures ANOVA. If the
ANOVA was significant, cells were compared using
post-hoc planned orthogonal t tests. Nonparametric
data were analyzed using %2 statistics with Yates'
modification for expected cell sizes of < 5.
Table 1. In-Hospltal Demographic Characteristics of 12- and
24-Months Follow-Up and No-Follow-Up Groups
12-Months 24-Months
follow-up follow-up No-follow-up
(#i = 38)
(" = 37)
(#i = 48)
63±11
Age (mean±SD)
76%
Race (% black)
57%
Sex (% male)
Marital status
32%
% married
35%
% widowed
33%
% other
Socioeconomic status Hollingshead
24%
Class I-m
76%
Class IV-V
History of CVA
19%
History of depression
0%
63+12
60%
63%
56±13*
63%
58%
42%
23%
35%
45%
21%
34%
30%
70%
27%
4%
29%
71%
21%
5%
CVA, cardiovascular accident.
*Significantly different from follow-up groups (p<0.05); all
other demographic variables not different by x2-
Robinson et al
Poststroke Diagnosis and Outcome at 1 and 2 Years
Table 2. Neurologic Findings and Diagnoses in 12- and 24Months Follow-Up and No-Follow-Up Groups
Stroke type
Thromboembolic,
atherosclerotic,
or unknown
cause
12 Months
follow-up
(i = 37)
24 Months
follow-up
(n = 48)
No-follow-up
(« = 38)
35
42
6
31
7
Downloaded from http://stroke.ahajournals.org/ by guest on July 28, 2017
2
Hemorrhage
Lesion location
22
Left hemisphere
17
14
Right hemisphere
13
Bilateral or
12
7
brains tem
Neurologic deficits
Monoparesis or
33
hemiparesis
29
Monosensory or
hemi sensory
15
deficit
13
Visual field
5
2
deficit
Aphasia
5
Broca's
3
1
1
Wemicke
1
3
Other
Computed tomography findings
19
12
Lesion visualized
Mean lesion vol4.7 + 6.1
5.6±4.8
ume ±SD
Mean distance of
lesion from
frontal pole
(% of A-P distance ± SD)
41.5±21.8 40.0+17.2
19
12
7
29
11
3
0
1
3
9
8.3+10
34.1 + 14.6
A-P, anterior-posterior distance.
views. Of these 10 patients, 6 were not depressed inhospital or at follow-up. Two patients had symptoms
of depression subsequent to a recurrent stroke; 1 had
major depression, the other had dysthymic depression.
One patient, who had an in-hospital diagnosis of major
depression, had a 24-month follow-up diagnosis of
dysthymic depression. One patient with dysthymic depression in-hospital had dysthymic depression subsequent to recurrent stroke.
In-Hospital Evaluation
In-hospital evaluations of the 37 patients seen at 12
months found 5 (14%) with symptom clusters of major
depression and 7 (19%) with symptom clusters of
dysthymic depression. Of the 48 patients seen at 24
months, in-hospital evaluations found 6 (13%) with
symptom clusters of major depression and 10 (21%)
with symptom clusters of dysthymic depression. Inhospital evaluations of the 38 no-follow-up patients
revealed 11 (29%) with major depression and 9 (24%)
with dysthymic depression. The frequency of depres-
839
sion in the no-follow-up group was not significantly
different from either follow-up group (x2 = 3.62,
df= 2, not significant [NS] for 12-months follow-up;
f = 4.34, df= 2, NS for 24-months follow-up).
Mean in-hospital depression scores on the Hamilton
(F 243 = 68.6, p < 0 . 0 0 1 ) , Zung (F2>4S = 33.1, p<
0.001), and PSE (F2,« = 69.2, p < 0.001) for both the
12- and 24-months groups were significantly higher
for patients with major compared with dysthymic depression or nondepressed patients (Table 3). Similarly,
mean depression scores for patients with dysthymic
depression were significantly higher than the nondepressed group (Table 3). The different depression
groups, however, did not differ significantly in either
MMSE (F2>43 = 1.6, NS) or SFE scores (F2i4J = 1.7,
NS). Patients with major depression, however, had
higher JHFI scores (F2>43 = 3.5, p < 0.05) than the nondepressed patients but were not significantly different
from patients with dysthymic depression.
Patients with major, dysthymic, or no depression
seen only in-hospital, at 12 months, or at 24 months
did not differ in their mean in-hospital depression
scores. In addition, in-hospital scores of physical
(JHFI) and intellectual (MMSE) impairment or social
functioning (SFE) did not differ significantly among
the 12-months, 24-months, or no-follow-up groups.
Thus, the no-follow-up group did not differ significantly from the follow-up groups in frequency or severity of depression or severity of impairment.
Follow-up Evaluation at 12 Months
Of the 37 patients interviewed at 12 months, 5
(14%) had major depressive symptoms, 7 (19%) had
dysthymic symptoms, and 25 (67%) were not depressed. Although these were the same overall percents found in-hospital (i.e., 14, 19, and 67%, respectively), the composition of the diagnostic groups had
changed (Figure 1). Of the 5 patients with major depression in-hospital, only 1 continued to have major
depression at 12 months, 2 had dysthymic depression
symptoms, and 2 were not depressed (Figure 1). Of the
7 patients with dysthymic depression in-hospital, 2 had
developed major depressive symptoms, 2 continued to
have dysthymic symptoms, and 3 had no depression.
Among the 25 patients who were not depressed inhospital, 2 had developed major depression, 3 had
developed dysthymic depression, and 20 patients remained not depressed.
Only 1 depressed patient (in-hospital diagnosis major depression, patient taking medication at 12-months
interview) had received treatment with antidepressant
medication. This patient was not depressed at 12
months. A hypothesis of unequal frequency of diagnostic change (i.e., 4 of 5 major depressions and 3 of 7
dysthymic depressions improved) based on in-hospital
depression category was not substantiated (X2Y«CJ =
1.5, df = 1, NS). Similarly, there was no significant
change (using repeated-measures ANOVA) in mean
scores over time or the diagnosis X time interaction
for the Zung (F 234 = 0.57, NS), Hamilton (F 234 =
0.54, NS), or PSE (F2t34 = 0.27, NS) scores be-
Stroke
840
Vol 18, No 5, September-October 1987
Table 3. Scores Obtained In-hospital and at 12- and 24-Months Follow-Up
No-follow-up
In-hospital diagnosis
Major depression
Zung
Hamilton
PSE
MMSE
JHFI
SFE
Dysthymic depression
Zung
Hamilton
PSE
MMSE
JHFI
Downloaded from http://stroke.ahajournals.org/ by guest on July 28, 2017
SFE
Nondepressed
Zung
Hamilton
PSE
MMSE
JHFI
SFE
In-hospital
(i=H)
60+14
14±7
25 ±10
16±5
8+7
0.25 + 0.07
12-Months follow-up
In-hospital
12 months
(n = 5)
d = 5)
51 ± 15
53 ±13
15 + 3
15±3
23 ±4
22±5
23±6
22±7
9±7
(n = 9)
45 ±9
0.10±0.06
d = 7)
44±10
9±6
9+3
13±5
2O±8
10±8
0.28±0.21
(n=18)
34±6
14±7
23 ±8
6±3
0.20 + 0.13*
(i = 7)
45±11
12±8 (n = 6)
13±11
21±7
24-Months follow-up
In-hospital
24 months
d = 6)
d = 6)
42 ±8*
60±ll
18±5
8±7*
23±4
9±7*
16 + 5
23 ±7
13 + 9
0.26 ±0.11
(*=10)
46±10
5 + 6*
0.20±0.09
(n =10)
13±6
21 ±6
50±16t
13±9t
18±12t
22±8
8±4
6±6
4±4
9±7
7±7
0.24±0.1
0.23±0.16
d = 32)
35±6
0.31 ±0.13
d = 25)
40±ll
4+3
4±3
4±3
5±5
4+3
6±7
0.21 ±0.14
d = 32)
44+14*
8 ±7*
9±8*
22±5
23±7
(i = 25)
4±4
5±5
25±6
2±3*
0.22±0.14
0.19±0.11
0.21+0.14
35±6
4 + 4(n = 31)
6+4
22 + 6
6±6
0.16±0.11
24 + 6
3±4*
0.17±0.10
Depression scores in-hospital and at 12-months follow-up are significantly different between each diagnostic category. F values for
statistical analyses are given in text. PSE, Present State Examination; MMSE, Mini-Mental State Examination; JHFI, Johns Hopkins
Functional Inventory; SFE, Social Functioning Exam.
•Significantly different from in-hospital scores.
tSignificantly different from either major depression or nondepressed at 24-months follow-up.
tween the in-hospital and 12-months evaluation for
patients with in-hospital diagnoses of major, dysthymic, or no depression.
Between the in-hospital and 12-months evaluations,
there was significant improvement in mean JHFI
scores over time but only in the nondepressed group
(F, M = 6.21, p<0.05). The major and dysthymic depression groups had not significantly changed from inhospital scores. There were no significant diagnosis X
time interactions (F2>34 = 0.36, NS). Thus, only nondepressed patients showed a significant improvement in
their ADL scores during the first year after stroke.
Scores on the MMSE were not significantly different between the in-hospital and 12-months evaluations
for any of the depression groups (F, M = 0.27, NS),
and there was no significant diagnosis x time interaction (F2,34 = 0.68, NS). In contrast, however, SFE
scores were significantly worse at 12 months only
for patients with major depression in-hospital
(FlM = 7.26, p < 0.01) while patients with dysthymic
or no depression did not deteriorate in social functioning. There were no significant diagnosis x time interactions in SFE scores (F2i34 = 1.10, NS).
Follow-up Evaluation at 24 Months
Of the 48 patients seen at 24 months, 10 (21%) had
major depression, 10 (21%) had dysthymic depression
symptoms, and 28 (58%) were not depressed. Although these percents were not significantly different
from those found in-hospital (i.e., 14, 19, and 67%,
respectively), the composition of the depression
groups had changed (Figure 1). All patients with major
depression at 24 months were different from patients
with major depression in-hospital, and 7 of 10 patients
with dysthymic depression were different from the patients with dysthymic depression in-hospital.
Of the 6 patients with major depression in-hospital,
4 had no depression and 2 had symptoms of dysthymic
depression (Figure 1). In contrast, however, among
the 10 patients with dysthymic depression in-hospital,
at 24 months 4 had developed symptoms of major
depression, 3 continued to have symptoms of dysthymic depression, and only 3 had recovered. Of the 32
patients who were not depressed at the initial evaluation, 6 had developed symptoms of major and 5 symptoms of dysthymic depression, whereas 21 remained
nondepressed.
In this study, only 2 patients had a previous (i.e.,
prior to stroke) personal history of depression defined
as seeing a health professional for evaluation or treatment of depression. These 2 patients were included in
the 48 patients receiving 24-months follow-up. One
patient had dysthymic depression in-hospital and remained depressed at 24 months, and the other had
Robinson et al
Poststroke Diagnosis and Outcome at 1 and 2 Years
in~hospltal
n.65
Downloaded from http://stroke.ahajournals.org/ by guest on July 28, 2017
m»)of depression
12 months
n-37
dysthymic depression
24 months
n.48
no depression
FIGURE 1. Diagnostic status in the hospital, 12 and 24 months
after stroke, for individual patients indicated by a numbered
bar; subsequent follow-up diagnosis is horizontally displayed.
major depression in-hospital but was not depressed by
24 months. Neither patient received a 12-months evaluation. Thus, a history of depression did not have a
significant impact on the results of this study.
Remarkably, only 1 patient with dysthymic depression (this patient was being treated at the 6-months
interview but still had dysthymic symptoms at 24
months) and 1 patient who was not depressed in-hospital but developed major depression by the 24-months
follow-up (this patient was receiving treatment at 24
months) had received antidepressant medication.
In contrast to our findings at 12 months, at 24
months the hypothesis of unequal frequency of improvement (i.e., improvement was defined as in-hospital major depression going to dysthymic or no depression at follow-up, while in-hospital dysthymic
depression was improved if no depression was found at
follow-up) for in-hospital major vs. dysthymic depres-
841
sion was statistically substantiated (X2YUM = 4 . 3 5 ,
df- \,p <0.025). Thus, patients with major depression in-hospital were more likely to improve by 24
months than patients with dysthymic depression inhospital.
The improvement in diagnostic status of patients
with major depression between in-hospital and 24months follow-up was also reflected by significant
improvement in depression scores between in-hospital and 24-months follow-up using repeated-measures ANOVA of the diagnosis x time interaction
for the Zung (F 2 4 5 =10.1, p<0.001), Hamilton
(F M 5 =12.6, p<6.001), and PSE (F 245 =12.6,
p<0.001). Post-hoc comparisons revealed that depression scores between in-hospital and 24-months
follow-up significantly improved for patients with inhospital major depression but were not improved in
patients with dysthymic depression (e.g., Hamilton inhospital vs. 24 months for major depression t = 3.4,
df = 5, p <0.05; Hamilton in-hospital vs. 24 months
for dysthymic depression t - \.%,df=9, NS). On the
other hand, by 24 months patients who were not depressed in the hospital had significantly worse scores
on the Zung (r = 3.7, df= 31, p<0.05), Hamilton
(r = 3.1,4f=31,p<0.05),andPSE(r= 1.9,rf/=31,
p<0.05; one-tailed). At the 24-months follow-up, patients with dysthymic depression in-hospital had significantly higher scores on the Zung (F24j = 4.6,
p<0.01), Hamilton (F245 = 5.9, p<0.01), and PSE
(F243 =7.0, /?<0.01) than patients with in-hospital
diagnoses of either major or no depression, who were
not significantly different from one another.
Patients showed significant improvements in their
JHFI scores over 2 years (F M 3 =17.4, p<0.05),
whereas post-hoc comparisons indicated that this occurred only in patients with major or no depression inhospital and that patients with dysthymic depression
in-hospital were not significantly improved. The diagnosis x time interaction was not significant (F2 43 =
2.7, p<0.08). MMSE scores (F M5 = 2.3, NS)'and
SFE (FM3 = 0.82, NS) scores were not different between the in-hospital and 24-months evaluations for
any diagnostic category.
Discussion
This study demonstrated that the prevalence of depression remains high but relatively stable for the first
2 years after stroke. The composition of the depressed
groups, however, changed over the 2 years. All of the
patients with major depression improved by 2 years,
and their depression scores were significantly lower
compared with their in-hospital evaluation. In contrast, however, only 30% of the patients with dysthymic depression improved in their diagnostic classification, and there was no significant improvement in their
mean depression scores. As a group, patients who
were not depressed in-hospital deteriorated in their
mood state by 2 years, with more than a third of them
developing clinically diagnosed depression and a significant increase in their mean depression scores.
The major uncertainties in this study are whether the
842
Downloaded from http://stroke.ahajournals.org/ by guest on July 28, 2017
follow-up population is representative of the initial
group and whether generalizations can be made to an
overall stroke population based on the findings in this
predominantly lower socioeconomic class population.
Although these questions cannot be completely answered, we have shown that the follow-up patients did
not differ in their diagnostic categories, severity of
impairment, or demographic features from patients
who were not followed up. Second, although the findings in these lower socioeconomic class patients may
not always apply to the general stroke population,
studies by other investigators have reported similar
findings concerning the relation between severity of
depression and lesion location,17 the association between failure to suppress serum cortisol following
dexamethasone administration and the existence of depression,18 the lack of statistical association between
severity of impairment (cognitive or functional physical) and severity of depression,17"19 and treatment response.20 Thus, although further studies will be needed
to establish the general applicability and reproducibility of these findings, there appears to be some indication that this stroke population may provide some insights for the general stroke population.
The most important finding from this study was the
poor prognosis of patients with an in-hospital diagnosis of dysthymic depression. This is in contrast to the
improvement in depression scores and ADL scores
(JHFI) seen in patients with major depression over the
2 years follow-up. This finding is certainly consistent
with our previous reports demonstrating the importance of differentiating major from dysthymic poststroke depression.5"7 It is also consistent with the findings of Jacoby et al21 and Post,22 who found that
depression in the elderly is frequently associated with a
poor prognosis, particularly when there was associated
ventricular enlargement.
Although this kind of differential prognosis for major and dysthymic depression might be expected for
patients without brain injury, to find this same clinical
differentiation within the brain-injured population surprised us. Both major and dysthymic depression recognized in-hospital generally met the DSM HI diagnostic
criteria of 2 weeks' duration of symptoms for major
depression and 2 years' duration of symptoms for
dysthymic disorder. It is possible, however, that some
patients with dysthymic depression in-hospital improved and then relapsed by the 2-years follow-up.
Although we do not know why major depressive
disorder remits between 1 and 2 years after stroke, this
spontaneous remission is consistent with the untreated
course of major depression in patients without brain
injury.23 It is also consistent with our previous suggestion of neurochemical or neurophysiological contributions to the development and remission of these depressive disorders. 3i24M These major depressions may
remit based on neural regeneration or biochemical
compensatory mechanisms such as changes in receptor
regulation or the rate of transmitter or enzyme
synthesis.
The reason dysthymic depression disorders do not
Stroke
Vol 18, No 5, September-October 1987
significantly improve over 2 years is uncertain. In a
previous study, we examined patients with dysthymic
depression in-hospital and compared them with those
who developed dysthymic depression several months
after discharge.26 Although we hypothesized that these
delayed-onset dysthymic depressions might be "reactive" to the severity of physical or intellectual impairment, we found no significant relation of depression to
severity of impairment. Thus, dysthymic depressions
do not appear to be psychological reactions to the impairment. These depressions, however, might be related to premorbid personality vulnerabilities, and this
"depressive" personality trait may lead to prolonged
"characterological depressions." This is only one possibility, and others might be posed.
The other striking finding in this study was the association between improvement in ADL (JHFI) and the
existence of depression. At the 12-months follow-up,
only nondepressed patients significantly improved in
JHFI scores. We have suggested in a previous publication26 that, although physical impairment does not
cause depression, once depression occurs, there is an
interaction that impacts significantly on the course of
the recovery. It has been suggested by many clinicians27 that depression adversely affects stroke rehabilitation. These data provide some empirical support
for this idea and also emphasize the need for ongoing
clinical evaluation and treatment of depression in
stroke patients. Although the demonstration of the efficacy of treatment of poststroke depressions by ourselves28 and others20 had not been published when this
study was conducted (i.e., 1981), the lack of treatment
of these disorders was striking.
Acknowledgments
The authors wish to thank Lynn Book Starr, MSW,
who did some of the patient interviews and Sergio
Starkstein, MD, for his assistance with the statistics.
References
1. 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
2. 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;144:256262
3. Robinson RG, Starr LB, Lipsey JR, Rao K, Price TR: A two
year longitudinal study of post-stroke mood disorders: In-hospital prognostic factors associated with six month outcome. /
NervMentDis 1985;173:221-226
4. American Psychiatric Association: Diagnostic and Statistical
Manual of Mental Disorders, ed 3. Washington, DC, 1980
5. Robinson RG, Kubos KL, Starr LB, Rao K, Price TR: Mood
disorders in stroke patients: Importance of location of lesion.
Brain 1984;187:81-93
6. Lipsey JR, Robinson RG, Pearlson GD, Rao K, Price TR: The
dexamethansone suppression test and mood following stroke.
Am J Psychiatry 1985;142:318-323
7. Robinson RG, Bolla-Wilson K, Kaplan E, Lipsey JR, Price
TR: Evidence for intellectual impairment related to depression
in stroke patients. Br J Psychiatry 1986;148:54I-547
8. Kunitz SC: The pilot stroke data bank: Definition, design and
data. Stroke 1984;15:740-746
9. Wing JK, Cooper JE, Sartorius N: Measurement and classifica-
Robinson et al Poststroke Diagnosis and Outcome at 1 and 2 Years
10.
11.
12.
13.
14.
15.
16.
17.
Downloaded from http://stroke.ahajournals.org/ by guest on July 28, 2017
18.
19.
tion of psychiatric symptoms, in: Instruction Manual for the
PSE. London, Cambridge University Press, 1974
Hamilton MA: A rating scale for depression. J Neurol Neurosurg Psychiatry 1960;23:56-62
Zung WWK: A self-rating depression scale. Arch Gen Psychiatry 1965;12:63-70
Robinson RG, Szetela B: Mood change following left hemisphere brain injury. Am Neurol 1981;9:447—453
Robinson RG, Starr LB, Kubos KL, Price TR: A two year
longitudinal study of post-stroke mood disorders: Findings during the initial evaluation. Stroke 1983;14:736-744
Folstein MF, Folstein SE, McHugh PR: Mini-mental state: A
practical method for grading the cognitive state of patients for
the clinician. J Psychiat Res 1975; 12:189-198
Robinson RG, Benson DF: Depression in aphasic patients:
Frequency, severity, and clinical pathology correlations. Brain
Lang 1981;14:282-291
Robinson RG, Bolduc PL, Kubos KL, Starr LB, Price TR:
Social functioning assessment in stroke patients. Arch Phys
MedRehabil 1985;66:496-500
Sinyor D, Jacques P, Kaloupek DG, Becker R, Goldenberg M,
Coopersmith H: Post-stroke depression and lesion location: An
attempted replication. Brain 1986; 109:537-546
Finkelstein S, Bernowit LI, Baldessarrini RJ, Arana GW, Levine D, Wood E, Near D, Moya K, Stroll AL: Mood, vegetative disturbance and dexamethasone suppression test after
stroke. Ann Neurol 1982; 12:463-468
Folstein MF, Maiberger R, McHugh PR: Mood disorder as
specific complication of stroke. J Neurol Neurosurg Psychiatry 1977;40:1018-1020
843
20. Redding MJ, Orto LA, Winter SW, Fortuna IM, Di Ponte P,
McDowell FH: Antidepressant therapy after stroke: A doubleblind trial. Arch Neurol 1986;43:763-765
21. Jacoby RJ, Levy R, Bird JM: Computed tomography and the
outcome of affective disorder: A follow-up study of elderly
patients. Br J Psychiatry 1981;139:288-292
22. Post F: The significance of affective symptoms in old age
(Maudsley monograph no. 10). London, Oxford University
Press, 1962
23. Rennie TAC: Prognosis in manic-depressive psychoses. Am J
Psychiatry 1942;98:801-814
24. Robinson RG: Differential behavioral and biochemical effects
of right and left hemispheric cerebral infarction in the rat.
Science 1979;205:707-710
25. Robinson RG, Lipsey JR, Bolla-Wilson K, Bolduc PL, Pearlson GD, Rao K, Price TR: Mood disorders in left-handed
stroke patients. Am J Psychiatry 1985;142:1424-1429
26. Robinson RG, Lipsey JR, Rao K, Price TR: A two year longitudinal study of post-stroke mood disorders: Comparison of
acute onset with delayed onset depression. Am J Psychiatry
1986;143:1238-1244
27. Adams GF, Hurvitz LJ: Mental barriers to recovery from
strokes. Lancet 1963; 1:533-537
28. Lipsey JR, Robinson RG, Pearlson GD, Rao K, Price TR:
Nortriptyline treatment of post-stroke depression: a doubleblind treatment trial. Lancet 1984;l:297-300
KEY WORDS • major depression • stroke • prognosis
minor (dysthymic) depression • longitudinal study
Two-year longitudinal study of poststroke mood disorders: diagnosis and outcome at one
and two years.
R G Robinson, P L Bolduc and T R Price
Stroke. 1987;18:837-843
doi: 10.1161/01.STR.18.5.837
Downloaded from http://stroke.ahajournals.org/ by guest on July 28, 2017
Stroke is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 1987 American Heart Association, Inc. All rights reserved.
Print ISSN: 0039-2499. Online ISSN: 1524-4628
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://stroke.ahajournals.org/content/18/5/837
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in
Stroke can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office.
Once the online version of the published article for which permission is being requested is located, click Request
Permissions in the middle column of the Web page under Services. Further information about this process is
available in the Permissions and Rights Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Stroke is online at:
http://stroke.ahajournals.org//subscriptions/