close

Вход

Забыли?

вход по аккаунту

?

Changes in cerebral activity pattern due to subthalamic nucleus or internal pallidum stimulation in Parkinson's disease.

код для вставкиСкачать
ORIGINAL ARTICLES
Changes in Cerebral Activity Pattern Due to
Subthalamic Nucleus or Internal Pallidurn
Stimulation in Parkinson’s Disease
Patricia Limousin, MD,*t John Greene, MD,* Pierre Pollak, M D , t John Rothwell, PhD,*
Alim-Louis Benabid, MD, PhD,t and Richard Frackowiak, MD, DSc*
High-frequency electrical stimulation of the internal pallidum (GPi) or the subthalamic nucleus (STN) improves clinical
symptoms of Parkinson’s disease. In 12 parkinsonian patients, 6 with STN and 6 with GPi stimulators, we used H,I5O
positron emission tomography to evaluate whether changes in movement performance were accompanied by change in
regional cerebral blood flow (rCBF). Patients were scanned both at rest and while performing a free-choice joystick
movement, under conditions of effective and ineffective electrostimulation.During effective STN stimulation, movementrelated increases in rCBF were significantly higher in supplementary motor area, cingulate cortex, and dorsolateral
prefrontal cortex (DLPFC) than during ineffective stimulation. No significant change was observed in any of these areas
during GPi stimulation. The difference between the effect of STN and GPi stimulation on movement-related activity was
mainly localized to DLPFC. These results confirm the dominant role of nonprimary motor areas in the control of
movement in parkinsonian patients and demonstrate the importance of STN input in the control of these areas.
Limousin P, Greene J, Poll& P, Rorhwell J, Benabid A-L, Frackowiak R. Changes in cerebral activity partern due
to subthalamic nucleus or internal pallidum stimulation in Parkinson’s disease. Ann Neurol 1997;42:283-29 1
Present concepts of the functional organization of the
basal ganglia are based on anatomically demonstrated
multiple parallel loops of signal flow from the cortex
through the basal ganglia and back to the cortex [I, 21.
In Parkinson’s disease (PD), the loss of neurons in substantia nigra pars compacta leads to reduced dopaminergic innervation of the putamen and caudate nucleus,
which in turn results in overactivity of cells in the internal globus pallidus (GPi) and substantia nigra reticulata (SNr) [ 3 ] .These nuclei form the main outputs
of the basal ganglia. Via their projections to the thalamus, they are assumed to cause disruption of activity in
the final cortical target zones. Positron emission tomographic (PET) studies have confirmed that changes in
cortical activity do occur in P D [4, 51. During freechoice joystick movements, there is less activation of
supplementary motor areas (SMAs), anterior cingulate
cortex, and dorsolateral prefrontal cortex (DLPFC),
compared with that seen in age-matched controls [4].
Injection of apomorphine, a dopaminergic agonist
drug, decreases akinesia and increases cortical activity
bl.
The main driving force for the increased discharge of
cells in GPi and SNr is thought to be overactivity of
From the ‘Wellcome Department of Cognitive Neurology and
MRC Human Movement and Balance Unit, Institute of Neurology,
Queen Square, London, UK; and ?Department of Clinical and Biological Neurosciences, Joseph Fourier University, Grenoble,
France.
the subthalamic nucleus (STN), which has excitatory
projections to both structures. As a result, both GPi
and STN have been identified as potential targets for
neurosurgical procedures aimed at reducing basal ganglia outputs [6-111. Pallidotomy has now been performed successfully by many groups [6, 7,9, 111. Two
PET activation studies have confirmed that pallidotomy is accompanied by improvement of cortical
activity during movement [ 12-14]. Theoretically, STN
surgery has an advantage over pallidal surgery in that it
can influence SNr as well as GPi activity and therefore
could be more effective. A recent procedure has been
to implant electrodes in STN or GPi for chronic electrical stimulation [8, lo]. Continuous high-frequency
stimulation is believed to inactivate the target structure
and to have an effect similar to a lesion.
The advantage of the technique is that the effect is
reversible and easily graded by changing electrical parameters. The effect is dependent on the frequency;
frequencies lower than 30 Hz are ineffective and those
greater than 50 Hz are effective [lo]. The aim of our
study was to compare the effect of STN and GPi stimulation on the pattern of cortical activation during selfselected movements. We wanted ( I ) to examine the
Received Nov 14, 1996, and in revised form Feb 3, 1997. Accepted
for publication Feb 25, 1997.
Address correspondence to Dr Frackowiak, Wellcome Department
of Cognitive Neurology, 12 Queen Square, London, W C 1N3BG,
UK.
Copyright 0 1997 by the American Neurological Association
283
modulation of patterns of activation associated with
the movement paradigm, by effective, as opposed to
ineffective, STN or G P i electrostimulation, (2) to compare modulation induced by STN a n d by G P i electrostimulation, a n d (3) to compare th'e results of G P i
stimulation in our study to the effects of pallidotomy
described in the literature.
the same electrical stimulation parameters used in the PET
study.
The patients gave informed consent and the project was
approved by the Joint Ethics Committee of the National
Hospital for Neurology and Neurosurgery. Permission to administer radioisotopes was given by the Administration of
Radioactive Substances Advisory Committee of the Department of Health (UK).
Subjects and Methods
Subjects
Twelve parkinsonian patients were studied with PET. Before
surgery they all suffered from a predominantly akinetic-rigid
form of PD with severe motor fluctuations (Tables 1 and 2).
Six patients had a chronic electrode placed in the STN and 6
in the GPi. Electrodes (Model 3387; Medtronic, Minneapolis, MN) were implanted under stereotaxic guidance [lo]
and connected to a telemetrically controllable pulse generator
(Itrell 11, Medtronic) placed in the subclavicular area. The
accuracy of electrode placement was checked by imaging
techniques and per-operative electrophysiology [lo]. The target, STN or GPi, was visualized by preoperative stereotaxic
magnetic resonance imaging (MRI), then drawn on the peroperative radiographs. The superimposition of the tip of the
electrode and the target position was checked at the end of
surgery and the precise location was confirmed, after surgery,
by MRI [ 101. Chronic electrical stimulation with frequencies
greater than 50 Hz [lo] decreased parkinsonian symptoms
(see Table 2). Electrical stimulation parameters were adjusted
to produce maximal clinical benefit while avoiding side effects due to overstimulation. The final values for these 12
patients are presented in Table 1. PET scans were collected
at least 8 hours after withdrawal of levodopa therapy and 18
hours after withdrawal of any dopaminrrgic agonist drugs.
Unified Parkinson's Disease Rating Scale (UPDRS) [ 151
scores were evaluated off drug with and without effective
stimulation. The time taken to perform a self-paced 20" flexion of the elbow was measured on the same arm and with
Activation P a r a d i p
The study was performed using a 2 X 2 factorial design allowing comparison of the effects of type of stimulation on
task-associated activities. The following four conditions were
studied: (1) rest (M-) plus ineffective stimulation (2 Hz)
(S-), (2) rest (M-) plus effective stimulation (130 Hz)
(S+), (3) tone-paced free-choice four-direction joystick
movement (M+) plus ineffective stimulation (S-), and (4)
tone-paced free-choice four-direction joystick movement
( M f ) plus effective stimulation (S+). A further (third) level
was introduced (2 X 2 X 2 factorial design) by performing
the experiment with stimulation at two anatomical sites
(STN and GPi). Each PET study included 12 scans. Each
condition was replicated three times. The order of the conditions was fully counterbalanced to eliminate time and order effects. Patients were stimulated continuously during
each scan, at either an effective or an ineffective frequency.
The most akinetic hand was selected to perform joystick
movements, and unilateral stimulation of the contralateral
STN or GPi was used. Subjects were asked to move the joystick in one of the four possible directions (right, left, forward, and backward), selected in a random order, paced with
an auditory stimulus every 3 ? 0.3 seconds. The same tone
was played during the rest conditions, and the subjects were
instructed not to think about making movements.
The x and y coordinates of the joystick position were recorded with a sampling rate of 100 Hz, allowing measurements of a combined reaction time-movement time function
Table 1. Characteristics of the Patients and Electrical Variables of Stimukation
Surgery
STN
Subject
1
2
3
4
5
6
M e a n t SD
GPi
1
2
3
4
5
6
Mean t SD
Age (yr)
PD Duration
(yr)
Sex
Handedness
51
60
49
56
50
52
53 + - 4
16
15
10
23
15
14
16 i 4
M
F
M
M
M
F
2 Fl4 M
R
R
R
R
R
R
6R
R
R
R
R
48
50
46
54
56
48
50k 4
6
15
14
M
M
M
F
M
R
L
R
23
9
14 2 6
=
284
Vol 42
No 3
R
R
R
1-
L
R
R
3 R/3 L
R
5 R/1 L
1 Fl5 M
September 1997
L
L
4 R/2 L
L
M
subthalaniic nucleus; R = right; L
PI) = Parkinson's disease; STN
Annals of Neurology
19
=
left; GPi
Side of
Stimulation
=
Pulse Width
Voltage (V)
(w4
2.5
3.0
3.6
3.1
2.8
1.8
2.8 2 0.6
60
60
60
60
60
60
60 t 0
3.6
3.6
3.6
3.6
3.6
3.6
3.6 ? 0
90
90
90
60
90
90
85 i 12
internal globus pallidus
Tuble 2. Clinical Scores, Hoehn & Yahr Before and After Surgery, Motor Part o f the UPDRS ajer Surgery With
and Without Effective Stimulation, and Movement Time
Surgery
STN
Subject
1
2
3
4
5
6
GPi
1
2
3
4
5
6
H&Y Before
Off DruglOn Drug
H&Y After
Off DruglOn Drug
513
513
512
412
412
512.5
512.3"
2.512.5
412.5
2.512
212
2.512.5
2.512.5
2.512.5"
412.5
413
412.5
513
413
412
412.75"
212
313
2.512.5
313
313
2.512.5
2.7512.75"
UPDRS 111 After
Off StimIOn Stim
68113
37124
6413 1
30110
67/26
48/26
53 ? 16/22' t
2219
32/18
57134
44139
52133
50122
43 2 13/26' 2 l l b
MT After
2-Hz StimIl30-Hz Stim
408.31348.5
322.71224.3
423.21199.5
4 18.1l301.1
437.71298.3
368.21339.3
396.4 +- 60.7/285.2d t 43.0b
132.51134.9
373.31279.4
408.31400.3
842.51553.5
319.71274.9
321.31283.2
399.6 ? 141.61321.0 ? 237.0b
"Median.
'Mean ? SD.
' p < 0.005, ' p < 0.05, within subject-within group effect of stimulation condition.
= Hoehn & Yahn; UPDRS I11 = Unified Parkinson's Disease Rating Scale, part 111; M T = movement time; STN = subthalamic
nucleus; GPi = internal globus pallidus.
H&Y
and analysis of the randomness of movement directions. Estimation of the randomness of movement direction, during
both effective and ineffective stimulation, was calculated
from the frequency distribution of movement in each direction, yielding a x2 value that was compared with the Evans
RNG index (the higher the value of the index, the less random is the series [16]).The reaction-movement time function was calculated from the time of the metronome tone to
the time at which two-thirds of the movement was completed.
Data Acquisition
Subjects were studied using a CTI-Siemens ECAT EXACT
HRt dedicated PET scanner (Knoxville, T N ) operated in
high-sensitivity three-dimensional (3D) mode [ 171. A venous
cannula to administer tracer was inserted in an antecubital
fossa vein in the arm opposite the side of joystick movement.
For each scan, 330 MBq H,150 in 3 ml of normal saline
was loaded into intravenous tubing and flushed into the subject over 20 seconds at a rate of 10 mllmin by an automatic
pump. After a delay of approximately 35 seconds, a rise in
head counts occurred, peaking 30 to 40 seconds later. The
interval between successive H,I5O administrations was 8
minutes. The data were acquired in one 90-second frame,
beginning with the rising phase of the head curve, and preceded by a 30-second background frame.
Correction for attenuation was made with a transmission
scan performed using external sources of positron-emitting
"GeIGsGa at the beginning of each study. Images were reconstructed by 3 D filtered back projection (Hanning filter;
cut-off frequency, 0.5 cycleslpixel), giving a transaxial resolution of 6.5 mm full width at half maximum, and displayed
in a 128 X 128 pixel format with 63 planes creating approximately 2-mm cubic voxels.
Image and Statistical Analysis
Image analysis was performed on a SUN SPARC 20 (Sun
Microsystems Europe Inc, Surrey, UK), using an interactive
image display software package (ANALYZE, Biodynamic Research Unit, Mayo Clinic, Rochester, MN) [18] and statistical parametric mapping (SPM 96, Wellcome Department
of Cognitive Neurology, London, UK) [ 19-2 1 1. Calculation
and image matrix manipulations were performed in PRO
MATLAB (The Mathworks, Natick, MA).
The data were realigned parallel to the intercommissural
line and normalized to the dimensions of the atlas of Talairach and Tournoux [22].The images were smoothed using
an isotropic 16-mm kernel to account for variation in gyral
anatomy and individual variability in structure-function relationships, and to improve the signal-to-noise ratio. For the
purposes of statistical analysis, images of patients performing
the task with the left hand were flipped along the x axis, such
that all patients were analyzed as if joystick movements were
performed with the right hand. This procedure makes the
reasonable assumption that movements elicit similar although largely lateralized patterns of activation.
The following comparisons were made: (1) the main effect
of joystick movement versus rest during ineffective stimulation (M+S-) - (M-S-), in STN and GPi groups; (2) the
main effect of effective stimulation versus ineffective stimulation in each group combining M- and M + conditions,
[(M+S+) + (M-S+)] - [(M+S-)
(M-S-)]; (3) the
interaction between motor task and stimulation in each
group, [(M+S+) - (M-S+)] - [(M+S-) - (M-S-)];
this interaction identifies areas where motor activation is
modulated as a result of effective electrostimulation. These
three comparisons were evaluated independently in STN and
GPi groups. And (4) a comparison of patterns of modulated
cortical activity due to GPi and STN electrostimulation; this
+
Limousin et al: Cerebral Activity and Akinesia
285
represents a factorial design with three levels in which the
third level is provided by the site of electrostimulation (see
legend to Fig 3).
On the basis of previous data [4, 5, 1;!-14] our preexisting hypothesis was that areas showing significant modulation
of motor task activations by electrostimulation would themselves be components of basal ganglia-thalamo-cortical motor loops, including striatum, pallidum, thalamus, and motor
cortical areas. We thus interrogated these areas by using an
uncorrected spatially restricted p value of p < 0.01. For
other comparisons, in the absence o f a preexisting hypothesis
about areas involved, we determined significance by using a
corrected p value of p < 0.05.
Comparison of the effects of frequency of stimulation,
motor scores, and Evans RNG index coefficients [ 161 were
tested by using multivariate analysis of variance.
of contralateral sensorimotor cortex (pre- and postcentral gyrus) and ipsilateral cerebellum, in both the STN
and the GPi groups. Posterior SMA and cingulate cortex were also activated, whereas anterior SMA and
DLPFC were not.
Muin Efect of Stimulation
Effective compared with ineffective STN stimulation
significantly deactivated sensorimotor and lateral premotor cortex ipsilateral to stimulation (Table 3 , Fig 1).
No significant activation was caused by effective stimulation of either STN or GPi. Effective STN and GPi
stimulation increased local STN and GPi rCBF to
some extent, compared with ineffective stimulation,
but the differences did not reach statistical significance.
Results
Clinical Improvement
STN and GPi stimulation both improved akinetic parkinsonian symptoms in the off-drug st-ate (see Table 2).
The motor score of the UPDRS (part 111) decreased
significantly, by 58% with effective STN stimulation
and by 40% with effective GPi stimulation, compared
with no stimulation at all. The UPDRS motor score
without stimulation was not significantly different between the STN-stimulated and GPi-stimulated groups.
Movement time in the elbow flexion task decreased by
28% with effective STN stimulation and by 20% with
effective GPi stimulation (see Table 2). The effects of
stimulation at each of the two sites were not significantly different.
The randomness of joystick movements during PET
scanning was similar between groups and between frequencies of stimulation
0.78, STN S - ; 0.71,
STN S + ; 0.75, GPi S-; 1.20, GPi S + ; Evans RNG
index: 0.61 ? 0.06, STN S - ; 0.60 k: 0.07, STN S + ;
0.67 ? 0.12, GPi S-; 0.61 2 0.07, GPi S+). Reaction-movement time in the joystick task did not
change significantly during either effective STN or GPi
stimulation (reaction-movement time: 617.44 &
368.26 msec, STN S - ; 497.97 ? 283.94 msec, STN
S + ; 646.58 2 234.54 msec, GPi S-; 545.81 t141.00 msec, GPi S+).
Interaction Between Stimulation and Movement
This term identifies modulations of movement-induced
patterns of activation by effective stimulation. During
effective stimulation of the STN, joystick movement
caused a greater activation in DLPFC (Brodmann area
[BA] 10 and BA 46), SMA (BA 6 ) , and cingulate (BA
24 and BA 32) than when ineffective stimulation was
used (movement by type of stimulation interaction,
[(M+S+) - (M-S+) versus (M+S-) - (M-S-)]
(Table 4 , Fig 2). In GPi-stimulated patients, we failed
to reveal any significant modulation of the activation
pattern induced by movement by effective stimulation.
Comparison of Interactions Across GPi and
STN Groups
In the final comparison, we asked whether STN or
(x2:
GPi stimulation had significantly different effects on
the amount of cortical activation produced by joystick
movement. The DLPFC was the only area with a significantly greater effect of STN compared with GPi
stimulation (Table 5, Fig 3). The differences were not
significant in SMA and cingulate cortex.
The data can be further addressed by studying the
nature of the interactions by means of plots. We studied the adjusted rCBF data for each of the four conditions at the most significant voxel in each of the three
areas identified by STN stimulation. During ineffective
STN stimulation, DLPFC rCBF decreased when patients made joystick movements, in comparison with
rest (Fig 4A); during effective STN stimulation
Main Efect of joystick Movement
Joystick movement execution compared with rest, in
the ineffective stimulation condition., led to activation
Table 3 Deactivations Due to Mdtn Effert of 130-Hz Stimulation in STN Pattents
-
__
Area
Coordinates
Left precentral gyrus (BA 4 & BA 6)
-8
-380
-20
Corrected p < 0.05,
110
STN = subrhalarnic
nucleus;
-20
-10
-20
prior hypothesis; significanr at cluster level.
BA
=
Brodmann area.
286 Annals of Neurology Vol 42
No 3
September 1997
62
38
64
z Score
Corrected p Value
3.89
3.82
0.004
3.75
Fig 1. Relative deactivations due to main effect of efective
subthalamic nucleus stimulation ([M- S-] + [ M f S- J versus
s+J + [M+ s+]). M- = rest,, Mi- = movement;
Si- = effective stimulation; S- = ineffective stimulation.
DLPFC rCBF increased. The difference between these
effects represents a significant interaction for the STN
group. During ineffective GPi stimulation, DLPFC
rCBF changed little when patients made joystick
movements, in comparison with rest (see Fig 4A), and
there was no change in this pattern during effective
stimulation. The lack of difference between these ef-
Fig 3. Areas where movement-by-stimulation interaction in
subtbalamic nucleus (STN) patients is signiJicantlygreater
than in internal globus pallidus (GPi) patients ([M+S+/ [M-Si-] versus [M+S-1 - [M-S-1 STN) versus
([M+S+J - [M-S+l versus [M+S-] - [M-S-] GPi).
M - = rest; M+ = movement; S+ = effective stimulation;
S- = ineffective stimulation.
fects indicates the reason for the absence of a significant interaction in the GPi group.
A similar examination of S M A (Fig 4B) and cingulate (Fig 4 C ) results showed similar patterns, ie, effective stimulation producing a slight but insignificant increase in movement-related rCBF compared with that
seen with ineffective stimulation.
Discussion
In our patients, the pattern of brain activation produced by joystick movement during ineffective STN or
GPi stimulation was similar to that described previously in nonimplanted P D patients 14,51; unlike the
situation in normal subjects, movement failed to activate DLPFC, anterior SMA, basal ganglia, and thalamus. The new finding was that the pattern of activation during movement could be normalized by deep
brain stimulation, and that this was more effective with
stimulation in STN than in GPi, in particular for the
DLPFC. This reinforces the suggestion that parkinsonian bradykinesia may be directly related to underactivation of S M A , DLPFC, and cingulate [4, 51, because
effective stimulation produced significant improvement
in movement performance.
Fig 2. Movement-by-stimulation interaction in subthalamic
nucleus patients ([Mi- S+/ - [M- S+ J versus [Mi- S- J (M-S-1). M - = rest; M t = movement; S+ = effective
stimulation; S- = ineffective stimulation.
The Effect of STN Stimulation
STN stimulation produced a more significant change
in movement-induced activation than GPi stimulation.
This observation is consistent with a slightly greater
Limousin et al: Cerebral Activity and Akinesia
287
Table 4. Movement-by-Stimulation Interaction in STN Patient5
Area
Coordinates
z Score
Uncorrected p Value
Left prefrontal (BA 10 & BA 46)
-38
-20
-28
2
-4
-14
-2
-2
3.11
2.79
2.72
2.82
2.59
2.76
2.58
2.55
0.001
0.003
0.003
0.002
0.005
0.003
0.005
0.005
Supplementary motor area (BA 6 )
Cingulate gyrus (BA 24)
Uncorrected p
STN =
52
6
64 -4
58 -2
4 70
8 62
2 34
16 44
8 40
< 0.05, prior hypothesis.
subthalamic nucleus; BA
=
Brodmann area.
Table 5. Areas Where Movement-by-Stimur‘ation Interaction in STN Patients Is Sign;ficantly Greater than in GPi Patients
Area
Coordinates
_______
Middle frontal gyms (BA 10 & BA 46)
-38
-36
-32
52
44
52
z
8
12
0
Score
3.01
2.92
2.68
Uncorrected p Value
0.001
0.002
0.004
Uncorrected p < 0.0 1, prior hypothesis.
STN
=
suhthalamic nucleus; CI’i = internal globus pallidus; BA = Brodmann area
clinical effect with STN stimulation. Activation produced by joystick movement was significantly augmented in SMA, cingulate, and DLPFC, with no
change in motor cortex during effective versus ineffective stimulation. These effects are similar to those observed after acute administration of apomorphine to
PD patients [ 5 ] . In this respect, it appears that STN
Stimulation produces its clinical effects by bringing the
pattern of brain activity during movement back toward
that described in control subjects. Chronic highfrequency stimulation is thought to produce an effect
analogous to that seen after structural lesions [8, 10,
231. If so, then the present results are likely to be due
to interruption of STN outputs to GPi and SNr. In
the current model of basal ganglia function [ 1-31, this
would decrease excitatory drive from STN to GPi and
SNr and hence reduce the inhibitory output from these
structures to thalamus. In monkeys, inputs to the thalamus from GPi and from SNr are segregated [24],with
each thalamic territory sending fibers separately to motor cortex, premotor cortex, SMA, and DLPFC [ I , 2,
24-26]. Reduced inhibition from basal ganglia output
nuclei may therefore explain the observed increases in
movement-related cortical activity.
Comparison of GPi Stimulation with
STN Stimulation
A comparison of the modulatory effects on movementrelated cerebral activity during effective STN and GPi
stimulation showed that there was only a modest (nonsignificant) difference in the degree of activation of
SMA and cingulate cortex. In DLPFC, on the other
288
Annals of Neurology Vol 42
No 3
September 1997
hand, effective STN stimulation resulted in markedly
greater activation during movement whereas effective
GPi stimulation had no significant effect. Differences
in task performance cannot explain this effect because
the randomness and speed of the movement sequences
was the same in both the STN- and GPi-stimulated
patients. One possibility is that because of the smaller
size of the structure, stimulation in STN could influence a larger proportion of the output neurons than
direct stimulation of GPi. In addition, STN stimulation may have an advantage by influencing activity in
other pathways to cortex, for example, through the
STN-SNr-ventrolateral nucleus-cortical pathway, the
STN-SNr-cen tromedian nucleus-cortical pathway and
the STN-parafascicularis nuclear connections. In particular, it should be noted that DLPFC is a prominent
target of the thalamic projection area of the SNt,
which would therefore not be influenced by stimulation of GPi [24, 271. GPi is a large structure, and variation in the precise location of electrodes within GPi
across subjects could be a factor responsible for some of
the variability in clinical and cerebral activation results.
In our study, we compared effective and ineffective
stimulation, to exclude modifications related to electrical stimulation itself without clinical change. However,
ineffective stimulation could have changed cerebral activity in comparison with no stimulation and therefore
may conceivably have attenuated differences between
conditions.
Finally, we should consider whether the experimental design and the number of patients are sufficient to
detect activation. O u r data clearly indicate that impor-
.
76t
74)-
z
.
'
1
4
4
1
L
4
.
5:
t
:
68
66
:
i
M-S- M-S+ M+S- ..J+S+ M-S- M-S+ I+S- M+S+
M-S- M-S+ M+S- M+S+ M-S- M-S+ M+S- M+S+
STN patients
.
.
Ijl 1
4
4
4
a
a
-
!
STN patients
GPi patients
GPi patients
B
A
66t
'
1
t
a
1
M-S- M-S+ M+S- M+S+ M-S- M-S+ M+S- M+S+
STN patients
C
GPi patients
tant activations are detectable. Although there is good
control over false positives, it is possible that there are
false negatives, and this might be one explanation for
the lack of demonstration of the simple effect of GPi
stimulation. However, an introduction of a third level
to the factorial design clearly indicates that differential
activation in DLPFC is detectable, thus suggesting that
the power of the experiment is adequate. The power of
the inference about differential activation between
stimulation groups is the same as the power to detect
activation within groups because we tested for this specific interaction effect with a contrast (using a fixed
effect model).
It is also possible that stimulation caused global
changes in blood flow that may have differed between
subjects. These effects, which do not themselves reflect
neuronal activation necessarily, are biologically unlikely
and, in any event, are controlled out by normalization.
Fig 4. (A) Plots of regional cerebral blood flaw (rCBF) f i r the
voxel in prefiontal cortex, showing the maximal z score in the
interaction, mean and value for each scan, for subthalamic
nucleus (STN) and internal globus pallidus (GPi) patients.
(B) Plots of rCBF in supplementary motor area cortex. (C)
Plots of rCBF in cingulate cortex. (M- = rest; M+ =
movement; S+ = effective stimulation; S- = ineffective
stimulation.)
Comparison of GPi Stimulation and Pallidotomy
Two previous studies have used activation paradigms
during rCBF scans before and after pallidotomy [ 12141. The initial report, a study from 1 patient, showed
that surgery resulted in increased activity of premotor,
SMA, and DLPFC during self-selected joystick movements [12]. A later study in 6 patients used a visually
cued grasping task and again revealed increased SMA
and premotor activation after surgery [13, 141. However, these results of this study are difficult to interpret
because pallidotomy failed to have any significant clinical effect [13, 141. In our study, GPi stimulation produced a significant improvement in off-drug motor
function, but only slight (non-significant) increases in
movement-related activity of SMA and cingulate cortex. Clearly, we must await further results from pallidotomy patients before detailed comparisons can be
made.
Lirnousin et al: Cerebral Activity and Akinesia
289
Role of DLPFC, SMA, und Cinplate in Movement
What is the role of DLPFC, SMA, and cingulate in the
joystick motor task we used? DLPFC is known to be
involved in working memory [28], and is active, for
example, during the production of randomized motor
sequences [4, 29, 301. SMA and DLPFC are both activated during internally generated tasks [29, 301. Anterior SMA and cingulate activity have been associated
with the selection and preparation of movements [31,
321 and with performance of internally generated tasks
[29, 301. The dorsal bank of anterior cingulate and
posterior SMA are coactivated during the execution of
movements [32]. Randomly generated joystick movements may involve all these processes. Working memory is needed as people have to remember previous directions selected. In addition, internal decisions are
made about the direction of the next movement, which
is then prepared and executed.
Efect of STN and GPi Stimulation on
Brain Activity
Conclusions
There is a suspicion that effective STN stimulation improves parkinsonian akinesia better than GPi stimulation, although formal data comparisons do not provide
proof for this clinical impression. In P D patients, the
pattern of brain activity associated with randomly selected movements is abnormal. That pattern is modified by effective STN stimulation in a highly specific
Annals of Neurology
Vol 42
No 3
R. F. and J. G. are supported by the Wellcome Trust. J. R. is supported by the MRC. P. L. is supported by the EEC.
We thank the subjects for participating in the study and Amanda
Brennan, Amanda Carrol, Graham Lewington, and Jon Galliers for
technical help.
References
Paradoxically, effective STN stimulation decreased motor cortical activity, while improving the quality of
movement. The most likely explanation for this is phenomenological. Clinically, effective STN stimulation
decreased rigidity, tremor, and off-dystonia; and therefore, the resolution of these symptoms may have attenuated motor cortical activity. N o increase in motor
cortical activity has been reported in the literature for
PD versus normal subjects, but it may be that the
changes are relatively small and difficult to detect.
Other literature suggests that such effects may occur.
For example, research in monkeys has shown changes
in cortical electrical activity after treatment with MPTP
(I-methyl-4-phenyl- 1,2,3,6-tetrahydropyridine) [33],
and magnetic stimulation has shown differences in cortical excitability in PD in comparison with normal subjects [34].
Neither STN nor GPi stimulation changed basal
ganglia activity directly. Local activation of STN occurred during effective STN stimulation and in GPi
during effective GPi stimulation. Whether this change
was a nonspecific local effect or due to feedback is unclear. Two studies of thalamic stimulation in tremor
did not show similar local activation [35, 361. However, tremor may itself have changed thalamic activity
and masked any directly induced effects.
290
manner, increasing the movement-related activity of
DLPFC, anterior cingulate, and SMA. STN thus appears to be a structure of major importance for the
control of neural activity associated with movement in
nonprimary motor cortical areas. Comparison of the
effect of STN and GPi modulation indicates a specific
action of STN on DLPFC, which may be related to
the clinical differences found with STN and GPi stimulation. The differential effect may reflect that STN
projects to the cortex via two routes (GPi and SNr)
whereas GPi projects via one route.
September 1997
1. Alexander GE, Delong MR, Strick PL. Parallel organization of
functionally segregated circuits linking basal ganglia and cortex.
Annu Rev Neurosci 1986;9:357-381
2. Alexander GE, Crutcher MD. Functional architecture of basal
ganglia circuits. Neural substrates of parallel processing. Trends
Neurosci 1990;13:266-271
3. DeLong MR. Primate models of movement disorders of basal
ganglia origin. Trends Neurosci 1990;13:281-285
4. Playford ED, Jenkins IH, Passingham RE, et al. Impaired mesial frontal and putamen activation in Parkinson’s disease: a
positron emission tomography study. Ann Neurol 1992;32:
151-161
5. Jenkins IH, Fernandez W, Playford ED, et al. Impaired activation of the supplementary motor area in Parkinson’s disease is
reversed when akinesia is treated with apomorphine. Ann Neurol 1992;32:743-757
6. Svennilson E, Torvik A, Lowe R, Leksell L. Treatment of parkinsonism by stereotactic thermolesions in the pallidal region.
Acta Psychiatr Scand 1960;35:358-377
7. Laitinen LV, Bergenmheim AT, Hariz MI. Leksell’s posteroventral pallidotomy in che treatment of Parkinson’s disease.
J Neurosurg 1992;76:53-61
8. Siegfried J , Lippitz B. Bilateral chronic electrostimulation of
ventroposterolateral pallidum: a new therapeucic approach for
alleviating all parkinsonian symptoms. Neurosurgery 1994;35:
1126-1130
9. Lozano A, Lang AE, Galvez-Jimenez N , et al. Effect of GPi
pallidotomy on motor function in Parkinson’s disease. Lancet
1995;346: 1383-1387
10. Limousin P, Pollak P, Benazzouz A, et al. Effect on parkinsonian signs and symptoms of bilateral subthalamic nucleus stimulation. Lancet 1995;348:91-95
11. Baron MS, Vitek JL, Bakey RA, et al. Treatment of advanced
Parkinson’s disease by posterior GPi pallidotomy: 1-year results
of a pilot study. Ann Neurol 1996;40:355-366
12. Ceballos-Baumann AO, Obeso JA, Vitek JL, ec al. Restoration
of thalamocortical activity after posteroventral pallidotomy in
Parkinson’s disease. Lancet 1994;344:814 (Abstract)
13. Grafton ST, Sutton J, Couldwell W , et al. Network analysis of
motor system connectivity in Parkinson’s disease: modulation
of thalamocortical interactions after pallidotomy. H u m Brain
Map 1934;2:45-55
14. Grafton ST, Waters C, Sutton J. Pallidotomy increases activity
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
of moror association cortex in Parkinson’s disease: a positron
emission tonlographic study. Ann Neurol 1995;37:776-783
Fahn S, Elton R. Unified Parkinson’s Disease Rating Scale. In
Fahn S, Marsden CD, Calne D, Goldstein M, eds. Recent developments in Parkinson’s disease, vol 11. Florham Park, NJ:
MacMillan Healthcare Information, 1987:153-1 63
Evans FJ. Monitoring arrention deployment by random number generation: an index to measure subjective randomness.
Bull Psychonom Sac 1978;12:35-38
Grooroonk S, Newiger H , Byars L, et al. Performance of the
ECAT EXACT H R + PET scanner in functional brain imaging. Neuroimage 1996;3:S24 (Abstracr)
Robb RA, Hanson DP. A software system for interactive and
quantitative visualization of multidimensional biomedical images. Australas Phys Eng Sci Med 1991;14:9-30
Friston KJ, Frith CD, Liddle PF, Frackowiak RSJ. Comparing
functional (PET) images: the assessment of significant change. J
Cereb Blood Flow Metab 1991;11:690-699
Frisron KJ, Holmes AP, Worsley KJ, et al. Statistical parametric
maps in functional imaging: a general linear approach. Hum
Brain Map 1995;2:189-210
Friston KJ, Ashburner J, Frith CD, et al. Spatial registration
and normalisation of images. Hum Brain Map 1995;2:165-189
Talairach J , Tournoux P. Co-planar srereotactic atlas of the human brain. Stuttgart: Thieme, 1988
Benabid AL, Pollak P, Gervason C, et al. Long-term suppression of tremor by c h r h i c stimularion of the ventral intermediate thalamic nucleus. Lancer 1991;337:401-406
Ilinsky IA, Jouandet ML, Goldnian-Rakic PS. Organizarion of
the nigrothalamocorrical system in rhe rhesus monkey. J Comp
Neurol 1985;236:315-330
Goldman-Rakic PS, Selemon LD. New frontiers in basal ganglia research. Trends Neurosci 1990; 13:241-243
Hoover JE, Strick PL. Multiple output channels in rhe basal
ganglia. Science 1993;259:819- 821
27. Barbas H , Haswell Henion T H , Dermon CR. Diverse rhalamic
projections ro the prefrontal cortex in the rhesus monkey.
J Comp Neurol 1991;313:65-94
28. Petrides M, Alivisatos B, Meyer E, Evans AC. Functional activation of the human frontal cortex during the performance of
verbal working memory task. Proc Natl Acad Sci USA 1993;
90:878-882
29. Deiber M-P, Passingham RE, Colebatch JG, er al. Cortical areas and the selection of movement: a study with posirron emission tomography. Exp Brain Res 1991;84:393-402
30. Frith CD, Friston KJ, Liddle PF, Frackowiak RSJ. Willed acrion and rhe prefrontal cortex in man: a srudy wirh PET. Proc
R Soc Lond [Biol] 1991;244:241-246
31. Passingham RE. The fronral lobes and voluntary actions. Oxford psychological series N 21. Oxford: Oxford University
Press, 1993
32. Stephan KM, Fink GR, Passingham RE, et al. Functional anatomy of the mental representation of upper extremity movements in healthy subjecrs. J Neurophysiol 1995;73:373-386
33. Douder DJ, Gross C, Arluison M, Bioulac B. Modifications of
precentral cortex discharges and EMG activity in monkeys with
MPTP-induced lesions of DA nigral neurons. Exp Brain Res
1990;80:177-188
34. Ridding MC, Inzelberg R, Rothwell JC. Changes in excitability
of motor cortical circuitry in patients with Parkinson’s disease.
Ann Neurol 1995;37:181-188
35. Parker F, Tzourio N , Blond S, et al. Evidence for a common
network of brain strucrures involved in parkinsonian tremor
and voluntary repetitive movements. Brain Res 1992;584:
11-17
36. Deiber M-P, Pollak P, Passingham R, et al. Thalamic stimulation and suppression of parkinsonian tremor: evidence of a cerebellar deactivation using positron emission tomography. Brain
1993;116:267-279
Limousin et al: Cerebral Activity a n d Akinesia
291
Документ
Категория
Без категории
Просмотров
1
Размер файла
889 Кб
Теги
patterns, change, due, pallidus, activity, disease, parkinson, subthalamic, interna, nucleus, stimulating, cerebral
1/--страниц
Пожаловаться на содержимое документа