Log in using OpenID

Neuropsychological profile linked to low dopamine

Downloaded from on December 29, 2014 - Published by
Journal of Neurology, Neurosurgery, and Psychiatry 1990;53:915-917
Neuropsychological profile linked to low
dopamine: in Alzheimer's disease, major
depression, and Parkinson's disease
Nicola Wolfe, Douglas I Katz, Martin L Albert, Abraham Almozlino, Raymon Durso,
Marcia C Smith, Ladislav Volicer
Boston University
School of Medicine
and Boston Veterans
Medical Center,
Boston, Massachusetts
Department of
A distinct pattern of neuropsychological
deficits was associated with low homovanillic acid (HVA) in the cerebrospinal
fluid of 21 patients with: Alzheimer's disease (9), Parkinson's disease (8) and
major depressive disorders (4). Regardless of clinical diagnosis, patients with
low HVA were slower on a *test of
efficiency of processing timed information, and showed greater benefit from
semantic structure on a verbal fluency
task than patients with high HVA. However, low HVA subjects were not significantly impaired on confrontation
naming (Boston Naming Test). Across
three diagnostic groups, patients with
lower HVA also tended to have more
extrapyramidal motor signs and were
significantly more depressed. These
results demonstrate a significant relationship between specific neurobehavioural deficits and dopaminergic
activity which cuts across traditional
diagnostic categories.
N Wolfe
M L Albert
A Almozlino
R Durso
Avenue, Boston,
Massachusetts 02130, United
Based on a description of the role of dopamine
in dementia by Bachman and Albert,' we
suspected that patients with dopamine
deficiency might exhibit a characteristic
neurobehavioural profile that cuts across the
traditional diagnostic categories of Alzheimer's disease, Parkinson's disease, and
major depressive disorders. Low dopamine
has been associated with extrapyramidal
motor signs, cognitive impairment and
depression within each of these disorders. We
suspected that low dopamine could be a common neurochemical pathology which accounts
for similarities observed across these diagnostic groups. In this pilot study, we tested the
hypothesis that regardless of diagnosis,
subjects with low cerebrospinal fluid (CSF)
homovanillic acid (HVA) would exhibit a syndrome consisting of the following: Parkinsonian motor signs, slowed cognition (bradyphrenia), and depression.
Received 29 November 1988
and in final revised form 26
January 1990.
Accepted 5 February 1990
Patients and methods
We evaluated twenty one male inpatients with
Department of
Psychiatry and
L Volicer
Braintree Hospital,
D I Katz
Department of
Clinical Psychology,
University of Florida,
Gainesville, Florida
M C Smith
Dr Wolfe,
c/o Dr M L Albert,
Department of Neurology,
Boston VA Medical Center,
150 South Huntington
diagnoses of dementia of the Alzheimer type
(9) (ADRDA-NINCDS criteria),2 Parkinson's
disease (8) (clinical consensus by two
on (a) the presence of at least
three of the four cardinal features: resting
temor, rigidity, bradykinesia, postural instability, and (b) absence of other known
aetiologies of extrapyramidal symptoms), or
major depressive disorder (4) (DSMIII).3
Exclusion criteria for all patients were:
evidence of structural lesion on CT scan,
Hachinski Ischaemia Score greater than 4, or
history of seizure. Neuropsychological testing
included the following range of tests: the
Mini-Mental State Test (MMS), a standard
index of overall cognitive ability, and the
Varied Interval Summation Task (VIST), a
simplified version of the Paced Auditory
Serial Addition Task4 designed to measure
efficiency of processing speeded information.
In this test, subjects hear a number (between
3 and 9) followed by a random series of the
numbers 1 or 2. The task is to add the series
and report the sum after the tenth number.
The interstimulus interval ranges from 5s on
the first trial to 0 7s on the last. Thus successive trials increase the rate of presentation of
stimuli, but the verbal response is not timed.
The test score is the total number of successful trials. A pilot study of this instrument
(Katz, unpublished data) showed it to be
sensitive to efficiency of processing speeded
information and to attentional deficits.
The Controlled Oral Word Association
Test (FAS) of verbal fluency5 was used to
assess benefit from semantic structure in word
list generation. Benefit from structure was
calculated as: number of correct responses in
Semantic Categories (animals, fruits and
vegetables, first names); number of correct
responses in Phonemic Categories (starting
with first letter F, A, and S successively)/
(Total correct responses). The Boston Naming Test (BNT) (30 item version) was
included as an index of language function
predicted to be uncorrelated with dopamine
deficiency. Extrapyramidal motor signs were
assessed with the Parkinson's Disease Motor
Performance Scale (PDS), based on the scale
by Duvoisin6; and affect was measured with
the Hamilton Depression Rating Scale
Downloaded from on December 29, 2014 - Published by
Wolfe, Katz, Albert, Almozlino, Durso, Smith, Volicer
Figure Cognitive profiles
ofpatients with low versus
high CSF HVA.
Results in patients with
lower HVA (n = 9),
compared to high HVA
(n = 12). Statistically
significant differences:
*p < 0-025, **p < 0 01
*Benefit from semantic
structure was computed as:
(number correct using
semantic categories minus
number correct using
phonemic categories)/
(total correct responses).
Data in Table.
with "high HVA" (12). Raw scores were
residualised to correct for baseline differences
in group means.9 (Group effect = grand meangroup mean). The "low HVA" subset of
patients, defined as those with HVA level
below the mean of residualised scores included
four with Alzheimer's disease, three with Parkinson's disease and two with major depressive
disorder. The "low HVA" patients, regardless
of clinical diagnosis, exhibited a distinctive
pattern of performance on the neuropsychological tests selected to assess bradyphrenia,
ability to benefit from structured tasks, and
anomia (fig, table).
On the efficiency of processing speeded
information (VIST), patients with low HVA
performed significantly worse compared with
high HVA patients (t(16) -231, p < 0 025)
(one-tailed t-tests were used to test this directional hypothesis). On the Verbal Fluency
Task, subjects with low HVA displayed significantly greater benefit from the structured
condition compared with the unstructured
condition (t(16) 2-86, p < 0-01, one-tailed).
Patients in the low HVA group had slightly
lower overall Mini Mental State Scores (t(16)
= -219, p < 0-025, one-tailed). However,
cognitive impairment in patients with low CSF
HVA was not present in all domains. For
example, scores on confrontation naming (Boston Naming Test) were not significantly
impaired in patients with lower HVA compared with the high HVA group. Subjects in
the low HVA group also had a significantly
greater number of Parkinsonian motor features
(t(16) 2-31, p < 0-025, one-tailed) and more
depression (t(16) 2-08, p < 0-05, one-tailed).
Lumbar puncture was performed using standard clinical procedure with the patient in the
lateral recumbent position at 9 am after the
patient had fasted and remained in bed from
midnight of the previous day. CSF was quick
frozen at the bedside and stored at 70°C until
the day of assay. HVA in the 7-9th cc was
measured by high pressure liquid chromatography based on the method described by
Matson et al' using absolute coulometric
detection with high selectivity and resolution.
HVA levels obtained were comparable with
other studies.8
Most patients were not taking drugs, but
14% (3/21) had received major tranquilisers
(minimum washout period 10 days) and 19%
(4/21) had previously received levodopa
therapy (Sinemet for Parkinson's disease
minimum washout period two days). The
length of the washout period was not significantly correlated with CSF HVA in these
patients, however, the number ofthese patients
is too small to rule out the possibility of residual
drug effects. Age and education were not
significantly different across the diagnostic
The mean (SD) cerebrospinal fluid HVA levels
were: 40 9 (17-1) ng/ml for Alzheimer's disease, 34 6 (16) ng/ml for Parkinson's disease,
and 29-9 (17 2) ng/ml for major depression. To
define the pattern of cognitive deficits
associated with dopamine deficiency, we compared patients with "low HVA" (9) to those
In this study, patients with low CSF HVA,
regardless of clinical diagnostic category,
exhibited a characteristic profile of cognitive
deficits, depressive symptoms, and extrapyramidal motor signs. Other investigators
have described aspects of this neurobehavioural profile in subgroups of patients in
each clinical category.8 0 This analysis draws
together these aspects.
Research on cognitive deficits in Parkinson's
disease provides the best comparison to the
neuropsychological pattern described in the
current study. Bachman and Albert' emphasised the pattern of slowed thinking, impaired
Table Cognitive profiles of patients with low versus high CSF HVA
Mini Mental State (MMS)
Varied Interval Summation Task (VIST)*
Verbal Fluency (FAS)** (Benefit from Structure)
Boston Naming Test (BNT) (30 item version)
Hamilton Depression Rating (HDRS)
Parkinson Disease Motor Performance Scale (PDS)
High HVA
(n= 12)
(n= 9)
23-7, 5 45 (790, 18-2)
7-8, 3-16 (55-7, 22-5)
0-169, 0-234 (16-9, 23-4)
22-4, 5-7 (747, 19 0)
8-39, 49 (11-3, 66)
4 76, 35 (14-87, 10 9)
18-9, 4-2f (63-0, 14-0)
4 89, 2-42tt (349, 17 3)
0-494, 0-285ff1 (49-4, 28-5)
18 6, 6-07 (62-0. 20 2)
14-1, 7-6t (19-1, 10-3)
9-82, 65 (307, 20-2)
Values are given as: mean, standard deviation (scores corrected for diagnostic group effects) and in brackets (percent of total
possible score).
*VIST score is the number of correct trials (presented with successively decreasing interstimulus interval) (Maximum possible
score equals 14).
**FAS Benefit from semantic structure was computed as (number correct using semantic categories minus number correct using
phonemic categories)/(total correct responses). (Maximum possible score = 1-0).
tStudent t test for significance of difference between low and high CSF HVA: t p (one-tailed) < 0-05, ttP (one-tailed) < 0-025,
tttp (one-tailed)
Downloaded from on December 29, 2014 - Published by
Neuropsychological profile linked to low dopamine: in Alzheimer's disease, major depression, and Parkinson's disease
manipulation of acquired knowledge, and cribed a potential neurobehavioural "sigimpaired memory, without aphasia. Studies nature" associated with relatively low
measuring cognitive speed in Parkinson's dis- dopamine, but only the HVA level in CSF has
ease and multiple sclerosis, and Huntington's been examined to date. It is likely that other
disease often use the common term brady- neurotransmitter systems contribute to the
phrenia, but methods to measure cognitive neurobehavioural profile we have identified.
speed vary considerably. For example, cognitive speed has been measured using: speed of
We thank Edith Kaplan for advice on the design of the
scanning elements in short term memory,'1 12 neuropsychological
evaluations, Errol Baker for statistical
time to prepare a manual movement,'1 advice and Guila Glosser
for comments on an earlier draft of this
was supported by NIH training fellowship
difference in time to perform two digit symbol paper. This study
the Veterans Administration Medical Research
tasks which differed in cognitive complexity,'0 (NS07239-04),
Service, and the Seidel Fund for the Study of Dementia.
attention and vigilance using a continuous
performance task,8 reaction time on tasks
DL, Albert ML. The dopaminergic syndromes of
differing in cognitive complexity'3 and the 1 Bachman
dementia. In: Pilleri G, Tagliavini F, eds. Cerebral aging
Paced Auditory Serial Addition Test.'4 In this
and degenerative dementias. Brain Pathology, Vol 1. Waldau Bern, Switzerland: University of Berne, 1984:91-119.
pilot study, we used a simpler serial addition 2 McKhann
G, Drachman D, Folstein M, Katzman R, Price
task (VI ST) to disentangle motor speed and the
D, Stadlan EM. Clinical diagnosis of Alzheimer's disease.
Report of the NINCDS-ADRDA work group under the
efficiency of processing speeded information.
auspices of Department of Health and Human Services
Because the motor response from the patient
Task Force on Alzheimer's disease. Neurology
(Cleveland) 1984;34:939-44.
was not timed, the score represents an 3 American
Psychiatric Association. Diagnostic and statistical
individual's success at processing auditory
manualfor mental disorders, Third Ed, Washington, DC:
stimuli at increasingly rapid intervals. 4 Gronwall DMA, Sampson H. The psychological effects of
However, this efficiency of processing speeded
concussion. Auckland, NZ: Auckland University Press,
information cannot be equated with cognitive 5 Benton
AL, Hamsher KS. Multilingual aphasia examination.
speed. Performance on this task is also depenIowa City: University of Iowa, 1976 (revised 1978).
RC. The evaluation of extrapyramidal disease. In:
dent on motivation, auditory acuity, short term
Ajuriaguerra J, ed. Monoamines, noyaux gris centraux et
memory, distractibility, difficulty with calculasyndrome de Parkinson. Paris: Masson, 1970:313-25.
7 Matson WR, Gamache PG, Beal MF, Bird ED. EC Array
tion, and other factors.
and data. Life Sci 1987;41:905-8.
Many investigators have reported decreased 8 Mayeux concepts
R, Stern Y, Sano M, Cote L, Williams JBW.
correlates of bradyphrenia in
verbal fluency in Parkinson's disease'5 16 (using
Parkinson's disease. Neurology 1987;37:1130-4.
the FAS test),5 but have not specifically
9 Rosenthal R, Rosnow RL. Essentials of behavioral research.
Methods and data analysis. New York: McGraw-Hill,
examined the benefit provided by semantic
categories compared with phonemic categories. 10 Rogers D, Lees AJ, Smith E, Trimble M, Stern GM.
in Parkinson's disease and psychomotor
In this study, we observe a need for structure
retardation in depressive illness. An experimental study.
from semantic categories on a verbal fluency
Brain 1987;l1O:761-76.
RD, Posner MI, Walker JA, Friedrich FJ. Cognition
test. Comparable observations in the literature 11 Rafal
and the basal ganglia: separating mental and motor
on Parkinson's disease include the description
components of performance in Parkinson's disease. Psychol Med 1984;16:257-65.
by Lees and Smith of "mental inflexibility",'5 12 Wilson
RS, Kaszniak AW, Klawans HL, Garron DC. High
and observations by Taylor et al 7 of impairspeed memory scanning in parkinsonism. Cortex 1980;
ment in the ability to generate efficient 13 Dubois B, Pillon B, Legault F, Agid Y, Lhermitte F.
strategies on spontaneously self-directed tasks.
Slowing of Cognitive Processing in Progressive Supranuclear Palsy. A Comparison with Parkinson's Disease. Arch
Most reviews of dementia in Parkinson's
Neurol 1988;45:1194-9.
disease agree that naming is relatively spared'6 14 Litvan I, Grafman J, Vendrell P, Martinez JM. Slowed
Information Processing in Multiple Sclerosis. Arch
and not affected by levodopa treatment.'8 Our
Neurol 1988;45:281-5.
findings that naming is relatively spared sup- 15 Lees AJ, Smith E. Cognitive deficits in the early stages of
Parkinson's disease. Brain 1983;106:257-70.
ports the hypothesis that only a specific set of 16 El-Awar
M, Becker JT, Hammond KM, Nebes RD, Boller
cognitive functions is impaired and that we are
F. Learning deficit in Parkinson's disease. Comparison
disease and normal aging. Arch Neurol
not simply reporting a global cognitive decline.
1987;44: 180-4.
The results of our study, as well as those 17 Taylor AE, Saint-Cyr JA, Lang AE. Frontal lobe dysfunction in Parkinson's disease. Brain 1986;109:845-83.
from other laboratories,'° '9 seem to support the 18 Delis
D, Direnfeld L, Alexander MP, Kaplan E. Cognitive
earlier theory on the association between
fluctuations associated with on-off phenomenon in Parkinson's disease. Neurology 1982;32:1049-52.
dopamine and cognitive impairment.' Alth- 19 Kaye
JA, Kay AD, May C, et al. Cerebrospinal fluid
ough there may be a role for mesolimbocortical
monoamine markers are reduced in dementia of the
Alzheimer's type with extrapyramidal features. Neurology
dopamine neurons to account for this
1987;37 (suppl 1):332.
behaviour, other neurotransmitters have also 20 Pillon B, Dubois B, Cusimano G, Bonnet AM, Lheymitte F,
Y. Does cognitive impairment in Parkinson's disease
been implicated, for example, in cognitive
result from non-dopaminergic lesions? J Neurol Neurosurg
impairment in Parkinsonism.20 We have desPsychiatry 1989;52:201-6.
Downloaded from on December 29, 2014 - Published by
Neuropsychological profile linked to low
dopamine: in Alzheimer's disease, major
depression, and Parkinson's disease.
N Wolfe, D I Katz, M L Albert, A Almozlino, R Durso, M C Smith and L
J Neurol Neurosurg Psychiatry 1990 53: 915-917
doi: 10.1136/jnnp.53.10.915
Updated information and services can be found at:
These include:
Email alerting
Receive free email alerts when new articles cite this article. Sign up in the
box at the top right corner of the online article.
To request permissions go to:
To order reprints go to:
To subscribe to BMJ go to:
Report inappropriate content