|
Neuropsychological evaluation of deep-brain-stimulation in patients with parkinson´s disease or essential tremor Dipl. Psych. I. Fresenborg, Dr. med. D. Gruia, Dr. med. U. Sander |
Abstract The implantation of a deep brain electrode for functional deactivation
of the nucleus ventralis intermedius
(VIM) results in a suppression of tremor which is one of the main symptoms of parkinson´s desease (PD) or
essential tremor (ET).
This elegant method is not
completely risc-free. In addition to complications as
bleedings and hemiparesis, cognitive impairments have
been described.
Since february
1996 implantations of chronic deep brain electrode have been performed in 15
cases in the neurosurgical clinic of the Evangelical hospital in
A careful neuropsychological
evaluation was undertaken:
An extensive neuropsychological diagnostic examination was performed pre and post surgery to examine the differential effects of the stimulation on the cognitive capacity respectively to detect undesirable side effects. Special emphasis was placed on mnestic functions and attention.
In this paper we would like
to emphasize on the effect of stimulation on the patients cognitive capacity.
Index terms Parkinson´s disease (PD), essential tremor (ET), stimulation of VIM, cognitive impairments.
II.
Results of neuropsychological research in PARKINSON´S DISEASE patients
Today
many studies have established the view that non-demented patients with
idiopathic PD even of an early or middle stage of the disease develop various
neuropsychological deficits across a range of cognitive functions. We would
like to describe in short the main findings:
·
Some of these impairments closely resemble those commonly attributed to
frontal lobe dysfunction, but are not as severe as in patients with frontal
lobe lesions. Set-shifting and set-formation (Downes
[2]) often tested with the Wisconsin Card Sorting Test (WCST [3]) is impaired.
Set means the state, which enables a person to react in a specific way if there
are reaction-alternatives. You can also say it is the ability to attend to a
specific stimulus dimension. The patients show more non- and perseveratoric errors and select not as much categories as
controls. The performance of the patients is disturbed only when there are no
external information for the shift, when they have to generate effective
strategies for the shift internally. Their performance in matching-to-sample-tasks
is preserved.
Some results show
that PD patients have problems in planning simple and complex movements
(movement-sequencing, Sullivan [4]). Their premotor
reaction times are prolonged in simple, but not in choice-reaction paradigms
which means that they can not use cues effectively in order to decrease their
reaction times.
PD patients show
problems in temporal ordering and time estimation, due to their increased
temporal discrimination threshold. The ability to develop a successful approach
to problem solving is reduced in treated PD (Morris [5]). Performance on the
tasks varies across patients, however: Some patients in early stages of PD show
no impairment (Canavan [6]), but others do.
·
Some of the visual spatial deficits found in patients suffering from PD
are due to specific impairments in motor-programming and performance, but
cognitive dysfunction have been found in the domain of visuospatial
perception even when no intellectual deficit is present and the tests include
no motor component (e.g. Boller [7]).
·
PD patients show visuomotor disabilities.
Research of eye
movements are correlated with the research of motor programming. External
triggering of saccadic eye movements plays the major role. There is e.g. an
increased dependence of saccadic eye movements on external cues, due to an
impaired disinhibition process in PD patients.
·
In memory capacity, PD patients show impaired verbal and non-verbal
short-term recall (Sullivan and Sagar [8]), deficits
in working memory, recognition and associative learning (Sahakian
[9]), but relative preserved long-term recognition (Sullivan and Sagar [10]). Long-term recall is mildly impaired and the
capacity to date past events is disproportionately disrupted (Sagar [11]). Procedural learning is mildly impaired in
non-demented PD-patients (Heindel [12]) and they show
slowed scanning on visual recognition tasks (
·
Language processing (e.g. object naming and vocabulary) and
comprehension are, in general, preserved; defective performance on language
tests may be due to impaired self-generation (Matison
[14]) and motor dysfunction. E. g. researchers found decreased language
production and semantic fluency, when the criterion for recall was alternating.
This is taken as a hint for a deficit of inhibitoric
attention processes and a deficit of maintaining internal performance-relevant
representation. PD patients often show speech problems (dysartric
speech) and in writing they show micrographia (also
called: amyostatic syndrome).
·
There are findings of a reduced rate of information processing in PD
patients. E.g. Zimmermann (1992) found a disproportional increase of only
simple reaction times in untreated PD patients in early stages with no
cognitive impairments. Patients with slight cognitive impairments were
additionally impaired in automatic and controlled processing and motor
programming. Slowing on motor response tasks may reflect both bradykinesia and a central defect of motor programming, but
the concept of bradyphrenia itself is controversial.
·
Results of the research of attention (e.g. selective attention in the
visual modality) are inhomogeneous: E.g. some researchers found a deficit of
maintenance of attention and overall prolonged reaction time (Sharpe [16]), but
others did not.
·
General intelligence as measured by the WAIS-R is relatively intact,
particularly in terms of verbal intelligence.
The
relationship between motor and cognitive disorders of PD is controversial:
E.g.
positive associations have been shown between severity of bradykinesia
and visuospatial deficit, in contrast to tremor
severity, which paradoxically correlated with good spatial memory (Mortimer
[17]). In a large longitudinal study presence of cognitive deficits was
associated with hypokinesia and rigidity, rather than
tremor (Portin [18]). Other researchers could not
confirm these results (Mortimer [19]).
In
comparison with patients suffering from other chronical
medical disorders a higher proportion of patients with PD shows evidence of
depression. Starkstein [20] reports a rate of 21%
with major and 20% with minor depression in his prospective study. Some
researchers found, that motor disability correlates strongly with the severity
of depression but weakly with cognitive impairment, which was not associated
with depression (Cooper [21]). Because of these findings depression is often
interpreted as a reaction to the physical disability the patient experience.
But other results show, that the severity of depression does not bear a strong
relationship to the extend of physical disability, but does relate to cognitive
dysfunction (Stern [22]). This suggests that depression is an intrinsic part of
the disease process. Anyway depression is a common phenomenon in PD and should
be taken into consideration when evaluating test performance.
The
results of research of the effect of long term medication on cognitive capacity
have been conflicting too: Neuropsychological deficits are not likely to
improve with medication, which means that cognitive performance seems to be
independent of an intact dopaminergic system.
III.
Treatment of the symptom: Tremor
The
implantation of a deep brain electrode for functional deactivation of the
nucleus ventralis intermedius
(VIM) results in a suppression of tremor. The permanent stimulation leads to a
deactivation of the neurons due to depolarization (Pollak
et al. [23]). This elegant method is not completely risk-free: In addition to
complications such as bleedings and hemiparesis,
cognitive impairments have been described: Especially in cases of implantation
in the left cerebral hemisphere a reduction of linguistic abilities (in first
case: loudness and articulation) has been reported (Benabid
et al. [24]).
The
thalamic stimulation of the VIM has been applied on 17 patients in
IV.
Method
A. Remarks to the surgical
procedure
Detailed
description of the surgical treatment we performed can be found elsewhere (e.g.
Benabid [24]).
Our
experience with fixing the stereotatic frame in a
short anesthesia (Dormicum) is very good. The
patients do report a significant lower stress-level after the surgical
treatment and there are no difficulties to get the people awaken and to
reproduce the tremor, when the electrode has to be positioned.
B. Procedure
In
general the patients have to stay nearly 24 days in clinic for treatment. The
electrode implantation is done in an average of one week and the generator is implantated nearly 4 days later (see figure 1.).
Because
of the findings, that PD patients even in early stages of disease show
cognitive impairments as described before we perform an extensive
neuropsychological examination before
treatment
to be able to assess the differential effect of stimulation on cognitive
capacity. This is also done to exclude patients with dementia. Emphasis was
placed on global cognitive functions as well as mnestic
and attention functions. We used the Hamburg-Wechsler Adult Intelligence
Test-Revised (Tewes [25]), the Testbatterie
zur Aufmerksamkeitsprüfung
(Zimmermann et al. [26]) and the
After
surgery neuropsychological tests are additionally done to detect even slight
side effects such as visual disorders or speech dysfunction. We found, that the
examination situation is very useful for adjusting the parameters of
stimulation.
In
general the voltage amplitude is set between 1 and 5 volts, pulse width between
60 and 200 microseconds and pulse frequency between 130 and 180 HZ.
Quantification
of the tremor was achived by rating with the Clinical
Tremor Rating Scale (Fahn et. al. [28]) before and
after treatment and by videotaping.
Depression
was quantified with the Beck-Depression Inventory (BDI, Hautzinger
et al. [29/30]).
We
used parameter-free tests (Mann-Whitney-U-Test) for statistical analysis
because of our small patient groups. An Anova was
also used to be able to take the effect of depression into consideration.
C. Description of the
patients
Nine
of our patients suffered from PD, six patients from essential tremor, one
patient suffered from Multiple Sclerosis and one patient from tremor after pontomesencephal bleeding. We will report the results for
the both main patient groups here only. There was only a significant difference
between disease duration of patients with PD and essential tremor. Further
information about the patients can be seen in figure 2.
The
severity of PD (see figure 3.) is rated according to the Scale of Hoehn and Yahr (Hoehn & Yahr [31]).
V.
Results
We
had to replace the electrode in two cases because of the appearance of visual
side effects but had no permanent complication.
·
The effect of stimulation on tremor
is seen in figure 4. It was rated according to the Clinical Tremor Rating
Scale which includes the rating of performance in activities of daily living (Fahn et al. [29]). The effect of electrode implantation on
tremor severity was significant for both groups (significance level for PD: p =
0.008, significance level for ET: p = 0.043).
·
Depression (figure 5.): We found with
using the Beck-Depression Inventory (BDI) no clinical relevant depression in PD
or ET groups and no significant difference between the severity of depression
between PD and ET patients. There was also no significant effect of surgical
treatment on the depression score of the patients.
·
General Intelligence (figure 6): The level of
performance of ET patients was slightly better than the performance of PD patients.
But the level of difference between PD and ET patients in the performance part
of the WAIS-R failed significance (significance level: p = 0.066). If taken
into consideration the depression score as a covariant there was a significant
difference between the patient groups (kind of disease: p = 0.019) and a
significant influence of the covariant (depression: p = 0.028) which was mainly
due to the subtest „picture ordering“ (effect of the covariant depression: p =
0.067, kind of disease: p = 0.085) and „object assembling“ (effect of the
covariant depression: p = 0.028, kind of disease: p = 0.051).
·
Mnestic functions: The performance in short term recall tested with the subtest „digit
span“ of the WAIS-R was significantly impaired in both groups after treatment
(PD: p = 0.047, ET: p = 0.003, compare to figure 7). But taking depression into
consideration there was no significant influence of surgical treatment. Verbal
learning (figure 8), recall and recognition was tested with the California
Verbal Learning Test [see for a detailed description of the test: Hildebrand et
al [27]: There was no significant difference between PD and ET patients before
treatment, but we found a significant covariance of depression for errors in
free recall (effect of depression on free recall: p = 0.001).The ability to
recall was significantly reduced for PD patients after implantation (figure 9,
free recall in PD: p = 0.037, cued recall in PD: p = 0.05, prolonged free
recall in PD: p = 0.020), but failed significance for ET patients (free recall
in ET: p = 0.063, cued recall in ET: p = 0.066). Prolonged recall was not
changed significantly for ET patients (free recall in ET: p = 0.066, cued
recall in ET: p = 0.066).There was a significant difference between the
patients groups and the side of implantation for errors in the interference
trial (kind of disease: p = 0.023, side of implantation: p = 0.005). The
performance in recognition was significantly dependent on the covariant
depression (effect of the covariant depression: p = 0.024) for both groups. The
error-level in recognition was significantly increased in ET patients (p =
0.042, see figure 10) after treatment, but failed significance for PD patients
(p = 0.058).
|
Figure 1. Clinical stay for treatment
Figure 2. Profile of study subjects
Figure 3. Rating of tremor severity Severity of Parkinson´s
disease Hoen and Yahr motor disability rating
|
·
Attention was tested with the „Testbatterie zur Aufmerksamkeitsprüfung“ (TAP, Zimmermann and Fimm, 1993): Simple reaction task (Alertness): In this
test the patients have to answer simple visual stimuli with or without cueing
as fast as they can. Before treatment the patients with ET profit more from
cueing than PD patients in this task (profit score: p = 0.012). The profit from
cueing was significantly increased for ET-patients after treatment (p = 0.042)
as you can see in figure 11. The implantation of the deep brain electrode
reduced the speed of PD patients in the cueing condition significantly (p =
0.05). The side of implantation had a significant effect on the profit from
cueing (profit score: side of implantation: p = 0.005, kind of disease p =
0.087, combined main effect: p = 0.010, figure 12). Working memory: The test require active holding and manipulation of verbal
information at the same time. The patients were impaired in all tasks of
complex attention measures (figure 13). The covariance of depression score and
missing in working memory was significant after surgical treatment (p = 0.032).
The missing increased significantly after treatment for PD patients (p = 0.042)
but not for ET patients (p = 0.066). Divided attention: In this test the patients have to
react to visual and acustic stimuli at the same time.
There was a significant effect of the side of implantation on the reaction time
in the divided attention task (p = 0.005). The main effect of kind of disease
failed significance (p = 0.080). This was also true for the errors in a task of
mental flexibility (side of implantation: p = 0.043, kind of disease: p
= 0.055).
VI.
Discussion
Every
patient was very satisfied after surgery. The patients who received a short
global anesthesia during fixation of the stereotactic frame reported not as
much stress and pain as those who received only local anesthesia, so it is an
integral part of the surgical treatment in our clinic now when implanting a
deep brain electrode.
To
sum up we found a significant effect of the deep brain stimulation on tremor
severity, which improved the performance of the patients in acitvities
of daily living significantly.
We
fond no clinical relevant depression in our both patient groups, but only
moderate depression symptoms, which may be the reaction to the fact of being
chronically ill. The two groups, PD and ET patients, did not differ in their
depression score. That arguments against the view, that depression is an
intrinsic part of Parkinson´s disease.
The
patients with ET performed slightly better than the PD patients in the measures
of global intelligence tested with the WAIS-R. But only if taken depression
into consideration as a covariant, there was a significant difference between
PD and ET patients in the performance part of the test.
In
memory capacity we found a significant reduction of performance in verbal
short-term recall, but this was mainly due to the covariant depression and not
due to the surgical treatment.
The
patient didn´t differ in respect to verbal learning,
recall and verbal recognition before treatment. The PD patients were impaired
after treatment but not under the condition of prolonged cued recall. We found
no significant effect of the side of implantation. The only significant
influence of the side of implantation was found for the errors under the
condition of interference. Depression covariated
significantly with the performance in verbal recognition. ET patients produced
more objects under the recognition condition, but also significant more false
positives after treatment.
ET
patients did profit more from cueing than PD patients in a simple reaction task
before treatment. Their profit from cueing increased significantly after
treatment. The treatment reduced the speed of PD patients in the cuing
condition significantly.
Implantation
of a deep brain electrode in the right cerebral hemisphere resulted in a
reduction of profit from cueing.
The missings in a task of working memory were significantly
influenced of the depression score after treatment. PD patients missed
significantly more critical stimuli after treatment.
The
side of implantation had an effect on the reaction task in a task of divided
attention (reduced speed after implantation in the reight
cerebral hemisphere, increased reaction times after implantation in the left
cerebral hemisphere) and an effect on the error rate in a task of mental
flexibility (fewer errors after implantation in the right cerebral hemisphere,
increased reaction times after implantation in the left cerebral hemisphere).
During
their stay in the clinic we questioned our patients about the subjective
outcome of the treatment. Although slight impairments of verbal memory and
parameters of attention could be found with neuropsychological tests only two
of our patients recognized any impairments themselfes.
Seemingly they don´t recognize their cognitive impariments due to the well structured situation they
experienced during their stay in our clinic. We do believe that we have to
examine our patients a third time after staying home for a longer period.
One
problem of the treatment of tremor with the VIM stimulation is that the
knowledge of the afferent and efferent connections of the VIM and the
biochemical basis of PD does not provide an explanation for the role of the VIM
nucleus in the arrest of tremor during. thalamic stimulation (see for a
detailed discussion: Benabid [24]).
Our
groups are very small and inhomoginious so it is not
easy to make any conclusions. We need more cases to confirm our results.
[1]
Poeck, K
(1994): Neurologie. Springer,
[2] Downes JJ, Roberts AC, Sahakian BJ, Evenden JL, Morris RG, Robbins TW (1989): Impaired extrapyramidal shift performance in medicated and unmedicated Parkinson´s disease: evidence for a specific attentional dysfunction. Neuropsychologia, 27, 1329-1343.
[3] Hea