Brain (1999), 122, 315–327
The effects of posteroventral pallidotomy on the
preparation and execution of voluntary hand and
arm movements in Parkinson’s disease
P. Limousin,1 R. G. Brown,1 M. Jahanshahi,1 P. Asselman,1 N. P. Quinn,2 D. Thomas,3 J. A. Obeso4
and J. C. Rothwell1
1MRC
Human Movement and Balance Unit, 2Department
of Clinical Neurology and 3Department of Neurosurgery,
Institute of Neurology, London, UK and 4Functional
Neurology and Neurosurgery Center, Clinica Quiron, San
Sebastian, Spain
Correspondence to: P. Limousin, MRC Human Movement
and Balance Unit, Institute of Neurology, University
College London, 23 Queen Square, London WC1 N3BG,
UK
Summary
We studied the effect of posteroventral pallidotomy on
movement preparation and execution in 27 parkinsonian
patients using various motor tasks. Patients were
evaluated after overnight withdrawal of medication before
and 3 months after unilateral pallidotomy. Surgery had
no effect on initiation time in unwarned simple and choice
reaction time tasks, whereas movement time measured
during the same tasks was improved for the contralesional
hand. Movement times also improved for isometric and
isotonic ballistic movements. In contrast, repetitive, distal
and fine movements measured in finger-tapping and
pegboard tasks were not improved after pallidotomy.
Preparatory processes were investigated using both
behavioural and electrophysiological measures. A precued
choice reaction time task suggested an enhancement of
motor preparation for the contralesional hand. Similarly,
movement-related cortical potentials showed an increase
in the slope of the late component (NS2) when the patients
performed joystick movements with the contralesional
hand. However, no significant change was found for the
early component (NS1) or when the patient moved the
ipsilesional hand. The amplitude of the long-latency
stretch reflex of the contralesional hand decreased after
surgery. In summary, the data suggest that pallidotomy
improved mainly the later stages of movement preparation
and the execution of proximal movements with the
contralesional limb. These results provide detailed
quantitative data on the impact of posteroventral
pallidotomy on previously described measures of upper
limb akinesia in Parkinson’s disease.
Keywords: Parkinson’s disease; pallidotomy; motor function
Abbreviations: CRT 5 choice/complex reaction time; IT 5 initiation time; MRCP 5 motor-related cortical potential; MT 5
movement time; pcCRT 5 precued complex reaction time; SMA 5 supplementary motor area; SRT 5 simple reaction time
Introduction
Limitations of drug therapy in the long-term medical
management of Parkinson’s disease have led to a renewal of
interest in functional neurosurgery for severely disabled
parkinsonian patients. At the present time, posteroventral
pallidotomy is one of the most widely used techniques, and
there are already many reports of its clinical effects on the
major symptoms of Parkinson’s disease (Svennilson et al.,
1960; Laitinen et al., 1992; Dogali et al., 1995; Lozano et al.,
1995; Baron et al., 1996; Lang et al., 1997; Samuel et al.,
1998; Scott et al., 1998). Virtually all confirm the dramatic
reduction in drug-induced dyskinesias. However, many of
the other symptoms respond less well. Off-period akinesia
and rigidity show a more modest improvement of ~30%,
© Oxford University Press 1999
whilst gait, balance and other axial signs are largely
unchanged. The effect can be bilateral even after unilateral
lesions, although the improvement is generally greater on
the side contralateral to the lesion.
The aim of the previous studies has been to quantify the
effectiveness of pallidotomy as a medical treatment, and as
such, in virtually all of them investigators have assessed
function with clinical rating scales. However, clinical scales
often hide much of the complexity of the underlying motor
dysfunction. The aim of the present study was to provide
some insight into this complexity by measuring the effect of
pallidotomy using various of the physiological and
psychomotor techniques that have been used previously to
316
P. Limousin et al.
Table 1 Clinical characteristics of the patients for each task
Task
Task
Task
Task
Task
Task
1: SRT and CRT
2: precued CRT
3: peg and tap
4: flex and squeeze
5: MRCP
6: stretch reflex
n
Side of
pallidotomy
Sex
16
17
21
18
12
11
6 R/10 L
8 R/ 9 L
9 R/12 L
4 R/14 L
5 R/ 7 L
1 R/10 L
8
7
7
6
4
3
Age (years)
F/ 8
F/10
F/14
F/12
F/ 8
F/ 8
M
M
M
M
M
M
53.13
53.47
54.45
56.61
56.33
61.00
6
6
6
6
6
6
2.22
2.20
2.11
2.03
2.43
1.71
Duration of Parkinson’s
disease (years)
14.81 6 1.47
14.12 6 1.41
14.90 6 1.44
14.67 6 1.43
13.92 6 1.89
14.91 6 2.04
Mean 6 SEM. n 5 number of patients.
study akinesia in patients with Parkinson’s disease. We have
focused on akinesia/bradykinesia for two reasons. First,
akinesia is the symptom of Parkinson’s disease most closely
related to striatonigral dopamine depletion (Vingerhoets et al.,
1997). Secondly, the rationale for posteroventral pallidotomy
comes directly from the physiological model of basal ganglion
function that deals mainly with symptoms of hyper- and
hypomobility (Alexander et al., 1986, 1990; Alexander and
Crutcher, 1990; DeLong, 1990; Wichmann and DeLong,
1996). This model predicts that in Parkinson’s disease, lesions
of the overactive pallidal output to primary, premotor and
supplementary motor areas (SMAs) should lead to an
improvement in the processes of movement preparation,
initiation and execution.
The tests we have used can be divided broadly into two
main categories: (i) tests of preparation and initiation of
movement, and (ii) tests of movement execution. The former
included a series of reaction time tests and recordings of the
premovement EEG activity. For the latter we measured
movements that were aimed or non-aimed, simple or complex,
proximal or distal. Finally, to test the possible contribution
of excessive stretch reflexes, we measured the size of the
EMG response to phasic stretch in wrist flexor muscles.
Patients
Twenty-seven patients with idiopathic Parkinson’s disease
(19 men and 8 women) were evaluated before and 3 months
after unilateral pallidotomy. Ten patients were operated on
the right globus pallidus internus and 17 on the left, as
indicated by clinical need. Their mean age (6 SEM) at the
time of surgery was 55.38 6 1.70) years and the mean
duration of the disease was 14.31 6 1.27 years. Before
surgery, all suffered from a severe form of Parkinson’s
disease with motor fluctuations. The range of the Hoehn and
Yahr score before surgery was 2.5–5 in the off-drug state
and 2–4 in the on-drug state. The target was located using
CT scanning and microrecordings. The surgical technique is
detailed elsewhere (Samuel et al., 1998).
Owing to patient disability and practical constraints of the
tests, not all tests could be performed on every patient. The
clinical characteristics of the patients who took part in each
experiment are given in Table 1. All patients gave informed
consent prior to participating in the study and the combined
Ethical Committee of The Institute of Neurology and The
National Hospital for Neurology and Neurosurgery approved
the procedures.
Method
Patients were evaluated in the morning on successive days,
after overnight withdrawal of medication. Both upper limbs
were studied and results concerning the limb ipsilateral and
that contralateral to surgery were averaged separately.
Unwarned simple reaction time and choice
reaction time (n J 16)
Patients performed an unwarned visual simple reaction time
(SRT) task and choice reaction time (CRT) task as described
previously (Brown et al., 1993a). The response apparatus
had six buttons, each 2.5 cm in diameter. The two central
buttons served as the ‘home’ keys for the left and right hands
and two pairs of response keys were situated 20 cm above
and below the central keys. In the SRT task, one home key
and one of the upper response keys were exposed. In the
CRT task, all six keys were exposed. Stimuli were presented
on a computer screen. A central cross acted as a fixation
point. The patient initiated each SRT trial by placing the
index finger of either the right or the left hand on the home
key. After a random and variable delay of 2–6 s, the
imperative stimulus appeared. This was a 1-cm solid square,
which appeared at the fixation point. The patient had to
respond as quickly as possible by moving the finger from
the home key to the appropriate response key. The time from
the appearance of the target to the lifting of the finger from
the home key was the response initiation time (IT). The time
from this to the pressing of the response key was the
movement time (MT). Patients received auditory feedback
that a response had been made. The patients then returned,
in their own time, to the home key, thus initiating the next
trial. The time until the next imperative stimulus varied
randomly between 2 and 5 s. Patients received 40 SRT trials
with each hand, the order counterbalanced across patients.
Median IT and MT were calculated for each hand.
In the CRT task, patients placed the index fingers of their
right and left hands on the two home keys. The imperative
Effect of pallidotomy on motor function
317
response keys. Median IT was calculated for the two hands for
each of the five S1–S2 intervals.
Pegboard and finger-tapping (n J 21)
Fig. 1 Simple and complex reaction time tasks before and 3
months after surgery. SRT 5 simple reaction time (or initiation
time); SMT 5 simple movement time; CRT 5 complex reaction
time (or initiation time); CMT 5 complex movement time (see
text for statistics).
stimulus was a square appearing above or below and to the
right or left of the fixation point in pseudorandom order.
Patients had to respond by moving their index finger to the
upper or lower response keys with the appropriate hand.
Median IT and MT for each hand were measured as before.
Patients received 80 trials, half with the left hand and half
with the right. Equal numbers of responses were made to the
upper and lower keys.
Precued choice initiation time (n J 17)
To determine the precued choice initiation time (pcCRT), the
task was similar to the unwarned CRT task with the exception
that the patients were provided with a precue that informed
them in advance about where the imperative stimulus was
going to appear (Jahanshahi et al., 1992a). The precue was the
outline of a square in the stimulus location. After a variable
delay (S1–S2 interval) the outline was filled, at which point
the patient had to respond as before. The duration of the S1–
S2 interval was 200, 800, 1600 or 3200 ms. A control condition
was recorded with the precue at the same time as the ‘go’ signal.
Eighty trials were given: 16 for each interval in pseudorandom
order, with an equal number of responses to each of the four
These two tasks were administered as described previously
(Brown et al., 1993b; Brown and Jahanshahi, 1998). The
pegboard task involved peg placement using the Purdue
pegboard (Purdue Research Foundation, 1948). Patients had
to pick up metal pegs (3 3 25 mm) one by one from a well
in front of them, and place the pegs in a vertical row of
holes drilled into a board. Patients performed the task three
times, once with the right hand, once with the left hand and
once with both hands simultaneously. On each occasion, the
patients had to place as many pegs as possible in a 30-s
period. Unimanual and bimanual peg scores were calculated
for each hand for each 30-s period.
In the tapping task, patients were required to tap a response
button repetitively using the index finger for 30 s. The button
had full travel of ~4 mm and activated a standard 150 g
microswitch. The tapping task was performed three times,
once with the left hand, once with the right and once
bimanually. Unimanual and bimanual tapping scores were
calculated for each hand for each 30-s period.
In addition to performing the tapping and pegboard tasks
individually, the patients also performed a combined task. In
this, they tapped with one hand while placing pegs with the
other. Both hand combinations were assessed. Patients were
instructed to do both tasks as well as they could, and not to
concentrate on one to the exclusion of the other. As before,
scores for the two hands were calculated for each of the
combined tasks. One patient was unable to perform this
combined task preoperatively.
Flex and squeeze (n J 18)
Patients were comfortably seated on a chair, the forearm was
flexed through 90°, resting on a manipulandum, and the hand
held a vertical grip containing a strain gauge. Patients
performed two types of movement, a 20° flexion of the elbow
and an isometric 30 N hand squeeze (Benecke et al., 1986,
1987a, b). The position of the manipulandum and the strength
developed were displayed by means of two vertical bars on
an oscilloscope screen, placed 1.5 m in front of the patient,
allowing visual control of the amplitude of the movement.
Patients were instructed to execute a self-paced movement,
as fast as possible, and were informed that the end-point
accuracy was less important. Different tasks using these two
movements were studied: (i) isolated flexion of the elbow
(simple flex); (ii) isolated hand squeeze (simple squeeze);
(iii) simultaneous flexion of the elbow and hand squeeze
(complex flex and complex squeeze).
After three to five training runs, patients performed 10–
15 movements in each condition. The two hands were studied
successively. Movement data were recorded and analysed
off-line. Movement times were determined by visual
318
P. Limousin et al.
Fig. 2 Initiation time according to the interval between the cue and the go signal in the precued choice
reaction time task before and 3 months after surgery (see text for statistics).
inspection from the point where the position trace deflected
reliably from the baseline to the end-point of the task, when
the speed of the movement reached a zero level.
Movement-related cortical potential (n J 12)
To determine movement-related cortical potential (MRCP),
patients were seated in a comfortable chair and required to
perform self-initiated joystick movements at the rate of ~1
per 5–10 s, in a randomly chosen direction, either left, right,
forwards or backwards. They held the joystick between the
thumb and the index finger. Sixty to 80 movements without
eye movement or other artefacts were recorded for each hand.
Twelve EEG channels were recorded, using silver–silver
chloride electrodes fixed with collodion, in a monopolar
configuration with a linked earlobe reference. Electrodes
were placed in three sagittal rows, according to the 10–20
international system and three additional electrodes in FC
(F3, FC3, C3, P3 in the left side, Fz, FCz, Cz, Pz on the
midline and F4, FC4, C4, P4 on the right side). EMG activity
of the first dorsal interosseous and the abductor pollicis brevis
was recorded with surface electrodes. EEG and EMG signals
were amplified and filtered using Digitimer D 150 amplifiers
(Digitimer Ltd, Welwyn Garden City, Herts, UK). A 5-s time
constant and 100 Hz high-frequency cut-off were used for
EEG signals and a 3-ms time constant and 3 kHz highfrequency cut-off for EMG signals. EEG activity was recorded
on a PC computer via an analogue/digital converter
Effect of pallidotomy on motor function
319
Fig. 3 Finger-tapping and pegboard tasks before and 3 months after surgery. uni 5 task executed with
one hand; bi 5 task executed with both hands; PEG 1 tap 5 number of pegs picked up when the
other hand is tapping; TAP 1 peg 5 number of taps when the other hand is picking up pegs (see text
for statistics).
(CED1401 Plus, Cambridge Electronic Design, Cambridge,
UK) using SigAvg software (Cambridge Electronic Design)
to collect data sweeps of 4 s (3 s before and 1 s after joystick
movement) at a sampling rate of 250 Hz.
EEG activity was back-averaged off-line after manual
realignment of the traces to movement onset. A grand average
was calculated for each hand. As some patients were operated
in the left and others in the right globus pallidus internus,
channels were arranged in order to obtain an average for the
channels of the hemisphere ipsilateral to surgery and an
average for the channels of the hemisphere contralateral to
surgery. The early (NS1) and the late (NS2) components of
the averaged MRCP were identified by visual inspection.
The onset of the early component was set at the onset of the
rise of the slope from baseline and the end at the point of
change of the slope, which corresponded to the onset of the
late component. The end of the late component corresponded
to the time of onset of EMG activity. The slopes of each
component were measured.
Forearm flexors stretch reflex (n J 11)
Stretch reflex of the forearm flexors was measured in both
arms. Patients were seated on a chair with the forearm secure
in position, and the hand was placed in a wrist manipulandum,
which could induce extensions of the wrist by means of a
torque motor attached to the underside of the manipulandum.
The patient was instructed to keep the hand in a rest position
between the stretches, with a slight extension of the wrist,
helped by the visual display of the hand position on an
oscilloscope screen. The torque motor applied a constant
strength of 8 N and the patient had to develop a basal steady
contraction of the forearm flexors to keep the hand in the
rest position. Additionally, every 3–5 s the motor induced a
sudden extension of the wrist with three levels of force (13,
26, 39 N) in random order, each lasting 200 ms. Rectified
EMG activity of the forearm flexors was recorded with
surface electrodes by back-averaging to the wrist position,
with a 3-ms time constant and a 3-kHz high-frequency cutoff. Data were recorded for 300-ms sweeps 50 ms before
and 250 ms after the movement using the same equipment
as described for MRCP recordings. Sixteen responses were
averaged in each condition. The latency and duration of
the responses were measured using visual inspection. The
amplitude of the response was defined as the ratio of the
average amplitude of the response to the average background
activity. We log-transformed the ratio before statistical
analysis.
Statistics
Analysis of variance for repeated measures was carried out
using the GLM (general linear model) procedure of SPSS
(SPSS Inc., 1996). Time (pre-/postsurgery) and hand
(ipsilesional/contralesional) were within-subject factors in all
analyses. Other within-subject factors were used as detailed
below for individual measures. Statistics for non-significant
320
P. Limousin et al.
averaged MT across tasks was considered. Analysis revealed
a significant hand 3 time interaction [F(1,15) 5 6.6, P ,
0.05]. Average MT for the contralesional hand improved
from 581 6 55 to 426 6 35 ms [F(1,15) 5 7.1, P , 0.05],
while the ipsilesional hand improved non-significantly from
497 6 47 to 433 6 33 ms.
A final analysis was performed to compare the relative
improvements in IT and MT, averaged across the two tasks.
The average improvement in MT was significantly greater
than the improvement in IT [F(1,15) 5 12.1, P , 0.01].
However, there was no differential effect of hand relative to
surgery [time 3 IT/MT 3 hand, F(1,15) , 1].
Precued choice initiation time
Fig. 4 Flex and squeeze task before and 3 months after surgery.
SF 5 simple flex; CF 5 complex flex, associated with a squeeze;
SS 5 simple squeeze; CS 5 complex squeeze, associated with a
flex (see text for statistics).
effects (P . 0.10) will not be given. Marginal effects (P ,
0.10 and P . 0.05) are given for information.
Results
Unwarned SRT and CRT task
Hand, time and task (SRT/CRT) were within-subject factors.
Initiation time (IT) and movement time (MT) were analysed
separately.
For IT, patients were reliably slower in the CRT task than
in the SRT task, as expected (Fig. 1) [F(1,15) 5 146.9, P ,
0.001]. However, the effect of neither hand nor time was
significant, although there was a trend for patients to show
a greater preoperative–postoperative difference for the
contralesional hand [F(11,15) 5 4.3, P , 0.06]. Average IT
across tasks for the ipsilesional hand decreased from 543 6 28
ms preoperatively to 526 6 29 ms postoperatively (Fig. 1),
while the contralesional hand showed a larger decrease, from
559 6 29 to 511 6 28 ms. None of the other interaction
terms were significant.
A similar analysis was carried out on the MT data. None
of the interactions involving task were significant, and so
As with the above analyses, hand and time were two of the
within-subject factors in the ANOVA (analysis of variance).
The third factor was S1–S2 interval. Only the IT data are
reported here (Fig. 2). Because the main purpose of the
analysis was to look for differential effects of time and hand
at the various S1–S2 intervals, the interactions will be
considered first. Analysis revealed a significant three-way
(task 3 hand 3 interval) interaction [F(4,13) 5 3.6, P ,
0.05]. With the exception of the expected effect of S1–S2
interval on IT [F(4,13) 5 16.9, P , 0.001], none of the
other main effects or interactions were significant.
To help interpret the three-way interaction term, the two
hands were considered separately. The ipsilesional hand
showed no time 3 interval interaction, with a similar decrease
in IT with increasing S1–S2 interval on the two occasions.
However, a significant interaction was found for the
contralesional hand [F(4,13) 5 4.3, P , 0.05]. Further post
hoc t tests comparing pre- and postsurgery data revealed that
the patients were significantly faster after surgery, with an
800 ms precue interval [t(1,16) 5 3.9, P , 0.01], but no
difference was found with shorter or longer S1–S2 intervals.
Finger-tapping and pegboard tasks
For the simple tasks (finger-tapping and the pegboard task
performed alone), task (unimanual/bimanual) was used as a
within-subject factor. Patients tapped more in the unimanual
than in the bimanual condition (average taps 6 SEM across
hand and time: 89.3 6 5.7 for the unimanual task and
77.3 6 5.9 for the bimanual) [task, F(1,20) 5 34.3, P , 0.001]
(Fig. 3). None of the other main effects were significant. There
was no difference between the hands or across time. None of
the two-way or three-way interaction effects were significant.
Similar results were found for the pegboard task. Patients
performed better with the unimanual task than with the
bimanual [task, F(1,20) 5 72.8, P , 0.001] (average across
hands and time, 6.9 6 0.8 for the unimanual task and
4.5 6 0.6 for the bimanual task) (Fig. 3). The main effect
of hand just reached significance [F(1,20) 5 4.2, P 5
0.05]. The contralesional hand was more impaired than the
ipsilesional hand before the operation and remained so
Effect of pallidotomy on motor function
321
Table 2 Slope of the early and the late part of the MRCP (µV/s) (mean 6 SEM) before (t0) and 3 months after surgery (t3)
Electrode
location*
Hand contralateral to surgery
Early t0
FCc
Cc
Fz
FCz
Cz
Pz
FCi
Ci
2.32
3.71
3.69
4.56
5.54
4.06
2.78
4.26
6
6
6
6
6
6
6
6
0.88
0.77
0.97
1.00
1.00
1.17
0.89
1.02
Early t3
3.88
3.21
3.62
4.58
4.61
1.57
2.35
2.88
6
6
6
6
6
6
6
6
1.12
0.97
1.10
0.96
1.17
1.39
0.84
0.78
Hand ipsilateral hand to surgery
Late t0
4.76
5.99
3.77
5.89
7.89
2.94
2.99
4.52
6
6
6
6
6
6
6
6
1.29
1.67
1.09
1.61
1.91
1.59
1.16
1.47
Late t3
Early t0
Early t3
8.14 6 1.31
8.80 6 1.01
7.57 6 1.60
9.80 6 1.76
13.00 6 1.99
7.44 6 1.58
7.06 6 1.05
9.35 6 1.28
1.46 6 0.65
2.76 6 0.90
1.68 6 0.54
3.03 6 0.75
4.20 6 1.28
2.40 6 0.84
2.86 6 0.97
3.52 6 1.33
2.23
1.42
1.39
3.03
3.03
0.57
1.84
2.93
6
6
6
6
6
6
6
6
0.78
0.93
0.50
0.83
0.91
0.52
0.88
0.97
Late t0
Late t3
5.68 6 1.22 2.85 6 1.80
5.77 6 1.19 5.01 6 1.78
5.81 6 1.53 3.52 6 1.72
8.10 6 1.10 6.36 6 1.60
11.55 6 1.63 11.54 6 2.50
4.61 6 1.45 4.66 6 1.63
6.90 6 1.13 4.72 6 1.66
9.20 6 1.58 8.57 6 2.01
*c 5 hemisphere contralateral to the hand moving; i 5 hemisphere ipsilateral to the hand moving.
was no significant effect of the simultaneous tapping
performance on the number of pegs placed (Fig. 3). In
contrast, tapping performance deteriorated substantially when
performed with the pegboard task compared with tapping
performed alone [F(1,19) 5 313.9, P , 0.001]. However,
none of the main effects and interactions involving time or
hand were significant and there was no differential effect of
time on the combined relative to unimanual task.
Flex and squeeze task
Fig. 5 Movement-related cortical potentials before (grey curve)
and after pallidotomy (black curve). Mean data from all patients
performing a joystick movement with the contralesional hand (A)
or the ipsilesional hand (B) (see text for statistics).
afterwards (average across task and time, 6.0 6 0.7 for the
ipsilesional hand and 5.5 6 0.7 for the contralesional hand).
The effect of time was not significant. None of the two-way
or three-way interaction effects were significant.
Performance in the combined conditions (tapping with
pegboard, or pegboard with tapping) was compared with
performance on each task when performed alone and
unimanually as described elsewhere (Brown et al., 1993b;
Brown and Jahanshahi, 1998). For the pegboard task, there
Analyses were performed as above, with hand, time, task
(flexion/squeeze) and complexity (simple movement/complex
movement) as within-subject factors. As previously reported,
movement time to perform the complex task was longer than
movement time to perform the simple task (Benecke et al.,
1986, 1987a, b) [complexity, F(1,17) 5 11.8, P , 0.005].
Contralesional arm and ipsilesional arm were differently
influenced by surgery, as shown by a significant time 3 hand
interaction [F(1,17) 5 6.8, P , 0.05]. In addition there was
a significant task 3 time 3 complexity interaction [F(1,17) 5
8.4, P , 0.05].
To help interpret these interactions, the two sides were
then considered separately. For the ipsilesional arm, there
was no significant change in MT after surgery (Fig. 4). For
the contralesional arm, MT was significantly faster after
surgery [time, F(1,17) 5 11.0, P , 0.005] (Fig. 4). In
addition, for the contralesional arm the improvement in flex
MT was greater in the complex than in the simple task
[time 3 complexity interaction, F(1,17) 5 9.1, P , 0.01].
This time 3 complexity interaction was not significant for
the squeeze task.
Movement-related cortical potential
The MRCPs recorded when patients performed joystick
movements are presented in Fig. 5 and the slopes of the
central and frontocentral electrodes in Table 2. The slopes
of the early and the late components were analysed separately.
Because of the sample size it was necessary to consider sets
of electrodes in separate analyses.
322
P. Limousin et al.
The first analysis considered the four midline channels.
Hand, time and site (Fz/FCz/Cz/Pz) were the within-subject
factors. The analysis of the late slope showed a significant
hand 3 time interaction [F(1,11) 5 8.9, P , 0.05]. To
help interpret this interaction we considered the two hands
independently. For movements made by the contralesional
hand, the slope of the late component increased significantly
after surgery [time, F(1,11) 5 5.8, P , 0.05], whereas the
ipsilesional hand showed no significant change (Table 2).
For the slope of the early component, none of the interactions
were significant (Table 2).
The second analysis considered the three central channels.
Hand, time and side (C ipsilesional side/Cz /C contralesional
side) were the within-subject factors. For the late slope, the
only significant interaction was hand 3 time [F(1,11) 5 5.99,
P , 0.05]. To help interpret this interaction we considered
the two hands independently. The slope of the late component
increased significantly after surgery when movements were
made by the contralesional hand [time, F(1,11) 5 6.1, P ,
0.05]. There was no significant difference in the effect of
surgery according to the side. The ipsilesional hand showed
no significant change.
Forearm flexor stretch reflex
The latency, duration and amplitude ratio (see Method) were
analysed separately for the early and the late components of
the stretch reflex. Level (of torque), side and time were
within-subject factors. The amplitude increased with the level
of torque applied both for the early [F(2,8) 5 16.41, P ,
0.005] and the late [F(2,8) 5 131.41, P , 0.001] components.
The early component ratio did not change with time.
Considering the late component, the ratio decreased in both
arms after surgery (Fig. 6). The time 3 level interaction was
significant only for the contralesional arm when analysed
separately [F(2,8) 5 7.62, P , 0.05]. Further post hoc t tests
revealed that the effect of time was significant for the highest
level of torque [t(1,9) 5 2.30, P , 0.05]. The duration of
the late component was decreased after surgery [F(1,6) 5
8.13, P , 0.05], independently of the hand, from 57.05 6 2.92
ms before surgery to 53.28 6 4.33 ms after surgery (average
across torques and hands). The duration of the early
component did not change significantly (18.60 6 1.36 ms
before surgery and 20.8 6 1.63 after surgery). The latency
of the late component was increased from 54.67 6 2.00 ms
before to 56.82 6 1.80 ms after surgery and the increase
was close to statistical significance [F(1,6) 5 5.35, P 5
0.06]. The latency of the early component did not change
significantly (27.33 6 1.04 ms before surgery and
28.01 6 2.21 ms 3 months after surgery).
Discussion
In the discussion, we will consider the effect of pallidotomy
on the processes involved in both the preparation and
execution of voluntary upper limb movement. The results
will be discussed in relation to patterns of impairment
previously described in patients with Parkinson’s disease and
the known effects of dopaminergic therapy. Finally, we
will consider the implications for current pathophysiological
models of akinesia in Parkinson’s disease.
Movement preparation
Three measures in the present results give some insight into
the effects of pallidotomy on movement preparation: simple
versus choice reaction time, precued reaction times and
premovement EEG activity.
Simple versus choice reaction time
Simple reaction times are faster than choice reaction times
in part because subjects have the chance to prepare in advance
the movement they will have to make at the time of the ‘go’
signal. Several authors have reported a selective impairment
in SRT relative to CRT in patients with Parkinson’s disease
in comparison with healthy subjects, and concluded that
patients do not prepare as well as normal subjects for
forthcoming movements (Evarts et al., 1981; Bloxham et al.,
1984; Sheridan et al., 1987; Pullman et al., 1988; Goodrich
et al., 1989). However, patients may also have difficulty in
keeping a prepared motor response in store, and this may be
an additional factor which could affect simple more than
choice reaction time. The present study did not include a
control group, but we were able to compare our data with
data from previous studies using the same procedures. The
patients of the present study were more severely affected
(Hoehn and Yahr range, 2.5–5; mean, 3.6) than the patients
of the previous study (Hoehn and Yahr range, 1–3; mean,
2.1) (Brown et al., 1993a). Preoperatively, the patients’ SRT
was slower than data obtained in our previous study in
parkinsonian patients (Brown et al., 1993a) (previous data
mean IT, 390 ms), but their preoperative CRTs were similar
to previous data (previous data mean IT, 670 ms). This
suggests that the more severe disease of the patients in the
present study was reflected in a differential slowing of SRT,
the condition in which advance preparation is possible.
However, while pallidotomy led to a modest, non-significant
improvement in IT there was no differential effect on the
two tasks.
Precued reaction times
Whilst simple and choice reaction times reflect the ability of
patients to prepare and maintain preparation for a forthcoming
movement, the precued task investigates the time needed to
use advance information to speed up response initiation. A
cue about the next movement is given at different times in
advance of a reaction signal. Healthy subjects can use the
cue to speed CRT to the level of SRT if information is
presented 800 ms before the reaction signal (Jahanshahi
et al., 1992a). In contrast, our patients needed 3200 ms to
Effect of pallidotomy on motor function
323
Fig. 6 Amplitude ratio (average amplitude of the response/average EMG background) of the early and
late components of the stretch reflex for different sizes of torque pulse (L1, L2, L3) before and 3
months after surgery (see text for statistics).
achieve the full benefit of the advance information when
they were tested before the operation, confirming the result of
our previous study (Jahanshahi et al., 1992a). Postoperatively,
marked benefit was seen at 800 ms, but only for responses
with the contralesional hand. The patients had become more
like normal subjects in terms of the speed with which they
could incorporate advance information into their movement
plan. However, they remained slow in the time needed to
translate this plan into movement.
The fact that this effect was restricted to the contralesional
hand is of interest. The use of cues to prepare movements
in advance is considered to depend on conscious, attentiondemanding processes (Goodrich et al., 1989). If pallidotomy
had improved these, we might have expected both sides to
be affected. A more likely explanation is that the cue was
used to improve later, purely motor stages of movement
preparation, which are more lateralized. Requin (1992)
divided motor preparation into three stages. First, a decision
is made on the goal of the movement (e.g. reach for a glass).
This is held to be non-motoric and symbolic. Next, there is
the motor programme, which is also non-motoric and specifies
relatively abstract response parameters. Finally, there is the
formulation of the motor commands necessary to perform the
specific action. The present results suggest that pallidotomy
decreases the time needed to incorporate cue information
into this last stage.
Premovement potentials
The average EEG activity preceding self-paced voluntary
movements is usually divided into two phases, the NS1 (or
BP) and the NS2 (or NS9, MP) components. Subdural
recordings indicate that the early part of the potential is
generated in the bilateral supplementary and primary motor
cortices, whilst later activity is more lateralized and dominant
in the primary motor cortex (Ikeda et al., 1992). Previously,
some reports have shown differences between parkinsonian
patients and controls mainly in the early portion of the MRCP
(Dick et al., 1989; Feve et al., 1992; Jahanshahi et al., 1995;
Touge et al., 1995). This was also true of the present sample
in their preoperative condition compared with data in normal
subjects obtained in previous studies (mean amplitude at Cz:
normal subjects, 4.9 µV in Touge et al., 1995; patients in
the current study, 3.3 µV for movements of the contralesional
hand). However, perhaps because of their advanced clinical
state, we also found that the NS2 component was smaller
than normal (normal subjects, 8.1 µV in Touge et al., 1995;
patients in the current study, 4.3 µV).
Pallidotomy resulted in a selective increase in the amplitude
of the later, NS2 component of the potential. In the context
of the reaction time results, this is consistent with the idea
that the lesion produces its effect by enhancing the later
stages of movement preparation rather than any earlier
preparatory processes.
Movement execution
Movement execution was measured in both simple and
repetitive tasks. Within the former group we employed three
measures: (i) the time to move the hand from one button to
another in a jump reaction time task, the buttons providing
fixed start and end points for the movement; (ii) the time to
make a self-terminated elbow flexion (isotonic movement)
or hand squeeze (isometric movement); and (iii) the time
324
P. Limousin et al.
to perform simultaneous elbow flexion and hand squeeze.
Preoperatively, all of these movements were slower than
normal, with values within the same range as previous studies
on Parkinson’s disease from this laboratory (Benecke et al.,
1986, 1987a, b). Pallidotomy decreased movement time, but
the effect was much greater in the button-pressing tasks than
for the self-terminated movements. The greater effect in the
button-pressing task might be related to the fact that it has
an end-point, whereas the flex and squeeze is a self-terminated
task. According to a previous study, parkinsonian patients
have more difficulty in executing self-terminated movements
(Flowers, 1976).
Although simple, single-joint ballistic movements are slow
in Parkinson’s disease, they offer a poor reflection of the
patient’s clinical akinesia (Berardelli et al., 1986). Rather, it
is with complex movements that the deficits are most clearly
seen (Berardelli et al., 1986). Benecke et al. (1986, 1987a,
b) clearly demonstrated that making a flexion movement
either simultaneously with or following a squeeze led to an
increased deficit compared with flexion performed alone. The
same pattern was seen in the present patients preoperatively.
Postoperatively, however, this deficit was improved, at least
for the contralesional hand. In other words, pallidotomy
improved complex movements more than simple movements.
The repetitive tasks that we used were pegboard and fingertapping. Both involved distal movements, the pegboard
requiring substantial sensorimotor co-ordination. Preoperatively the movements were slow, but, surprisingly, there
was little change in either task after pallidotomy. Additionally,
there was no differential impact when the same movement
was performed bimanually or when two different movements
were performed simultaneously. We can only speculate on
the possible reasons for this lack of effect. The movements
involved fractionated distal finger movements that were not
tested in the SRT, CRT and flex and squeeze tasks. This
suggests that pallidal lesions have a larger effect on proximal
than distal muscles. Alternatively, perhaps the highly
fractionated nature of the finger movements was the crucial
factor that made them resistant to change with pallidotomy.
Stretch reflexes
The long-latency component of the EMG response to muscle
stretch is known to be enhanced in patients with Parkinson’s
disease (Rothwell et al., 1983; Cody et al., 1986). In contrast,
the short-latency response is the same size as in healthy
controls. The same was true of the present patients in their
preoperative state. Pallidotomy selectively decreased the
amplitude of long-latency responses in the contralesional
arm, which may contribute to the reduction of rigidity that
has been reported in many clinical studies. The mechanism
of the effect is less clear. The long-latency component of the
stretch reflex in wrist flexor muscles may result from activity
in two separate nervous pathways, one involving a
transcortical relay of fast-conducting (group I) muscle
afferents, the other reflecting activity in slower-conducting
(group II) muscle afferents in a shorter spinal pathway.
Pallidotomy could affect the transcortical pathway through
its effect on pallidothalamocortical projections. The reported
change in excitability of motor cortex inhibitory circuits
detected in magnetic stimulation studies after pallidotomy
would be consistent with this idea (Strafella et al., 1997;
Young et al., 1997). Alternatively, pallidotomy might reduce
the sensitivity of the secondary muscle spindle endings
through an action on the fusimotor system. As yet, this
notion is unexplored, but could involve pallidal outputs to
brainstem nuclei.
Comparison of the effects of pallidotomy and
levodopa
In some respects, the effects of pallidotomy and levodopa
are remarkably similar. Both improve movement time and
lead to an additional improvement in complex versus simple
arm movements (simultaneous flex and squeeze compared
with each alone) (Benecke et al., 1987b). Both have little
effect on initiation times in the SRT and CRT tasks used in
this study (Jahanshahi et al., 1992b).
However, there are some aspects of akinesia that improve
following administration of levodopa but which show little
change following pallidotomy, and vice versa. Thus,
pallidotomy increased the speed with which patients could
incorporate cue information to speed up reaction time whereas
levodopa has been previously shown to have no effect
(Jahanshahi et al., 1992b). In addition, pallidotomy improved
the late part of the MRCP just prior to movement onset,
whereas levodopa appears to enhance only the early
component (Dick et al., 1987; Feve et al., 1992). Simple
repetitive movements, such as finger-tapping, and tasks
involving fine distal co-ordination, such as the pegboard
task, have provided very sensitive measures of levodopa
effectiveness in treating Parkinson’s disease (Meier and
Martin, 1970; Hughes et al., 1994). In contrast, pallidotomy
led to no appreciable change. The greater effect of levodopa
in some tasks could be explained by the more diffuse action
of levodopa. Pallidotomy was unilateral and was positioned
to have maximum effect on the motor component of the
cortex–basal ganglia–thalamocortical loops. Levodopa acts
diffusely on the striatum and can modify the activity of both
direct and indirect pathways. It also affects the substantia
nigra pars reticulata, the other major output structure of the
basal ganglia. There are even dopaminergic receptors in the
globus pallidus internus, the subthalamic nucleus and some
cortical areas which could play a role in mediating some of
the effects of levodopa. Indeed, clinically also, levodopa
and pallidotomy can have opposite effects. For example,
pallidotomy improves levodopa-induced dyskinesias. It is
more difficult to explain why precued CRT and the late
MRCP were more affected by pallidotomy. Perhaps the
effect of levodopa on other structures could prevent these
improvements.
Effect of pallidotomy on motor function
Implications and conclusions
The current model of basal ganglion anatomy envisages
several parallel striatopallidocortical loops, among them an
associative or complex loop and a motor loop (Alexander
et al., 1986, 1990; Alexander and Crutcher, 1990; DeLong,
1990; Wichmann and DeLong, 1996). Some aspects of this
model have been reconsidered, but it is still useful to discuss
the mechanisms of action of basal ganglion surgery (Levy
et al., 1997; Parent and Cicchetti, 1998). The motor loop
includes the putamen, the posteroventrolateral globus pallidus
internus, the ventrolateral thalamus and, at cortical level, the
primary motor cortex, the premotor cortex and the SMA
proper. The associative loop includes the caudate nucleus,
part of the putamen, the anterodorsal globus pallidus internus,
the substantia nigra reticulata, the ventroanterior and some
mediodorsal thalamus and frontal association cortical areas,
including the dorsolateral prefrontal cortex and the preSMA. In our study, the target for pallidotomy was the
ventroposterolateral part of the globus pallidus internus,
which is part of the motor loop. This procedure is more
effective against parkinsonian motor symptoms than anterior
pallidotomies (Svennilson et al., 1960). The increase in
globus pallidus internus activity found by microelectrode
recordings during pallidotomy is mainly localized to the
posteroventral part (Hutchinson et al., 1994). Consequently,
the primary motor cortex, premotor cortex and SMA proper
should be the main areas modified by pallidotomy, while the
dorsolateral prefrontal cortex and pre-SMA should be less
influenced.
The role of these various cortical areas can be summarized
as follows. Early preparatory processes, decision and selection
of movement may involve the pre-SMA and dorsolateral
prefrontal cortex (Brinkman and Porter, 1979; Matsuzawa
et al., 1992; Jahanshahi et al., 1995). Following this, the
motor cortex needs to be set in the state required for
movement initiation, processes which involve mainly the
SMA proper, the premotor cortex and the motor cortex
according to electrophysiological and metabolic studies in
animal models and humans (Brinkman and Porter, 1979;
Roland et al., 1980; Tanji and Kurata, 1985; Mushiake et al.,
1991; MacKinnon et al., 1996). Finally, movement execution
may be the domain of the motor cortex, and of the SMA
also (Mushiake et al., 1991).
Our results are consistent with most aspects of this
organization. Posteroventral pallidotomy improved the last
stages of movement preparation and execution, presumably
by acting on the motor loop to improve function in the
primary motor cortex, premotor cortex and SMA proper. Early
preparatory processes, characteristic of projection targets of
the frontal, associative basal ganglion loops were unaffected.
Studies on regional CBF and glucose metabolism have also
noted a consistent change in SMA activation after pallidotomy
(Ceballos-Baumann et al., 1994; Grafton et al., 1994, 1995;
Eidelberg et al., 1996; Samuel et al., 1997). However, changes
are also seen in regions including the dorsolateral prefrontal
325
cortex (Ceballos-Baumann et al., 1994; Samuel et al., 1997).
The significance of this prefrontal change is unclear, as there
are no reports of any consistent changes following surgery
in cognitive functions known to involve this region. If
anything, there is a suggestion that some abilities may
deteriorate, such as aspects of working memory (Troster
et al., 1997; Jahanshahi et al., 1997).
It is acknowledged that the present study is largely
descriptive. We deliberately chose tasks for which we already
had empirical data, allowing us to compare the results of
surgery with known patterns of impairment in medicated and
unmedicated patients. The way is now open for further
experimental studies to test specific hypotheses about the
role of the basal ganglia–thalamocortical circuitry in akinesia
in Parkinson’s disease and voluntary movement in general.
Acknowledgements
We thank the patients for participating in the study. P.L. was
supported by an EEC grant and M.J. by the Wellcome
Trust. J.A.O. received a grant from the National Parkinson
Foundation.
References
Alexander GE, Crutcher MD. Functional architecture of basal
ganglia circuits. Neural substrates of parallel processing [see
comments]. [Review]. Trends Neurosci 1990; 13: 266–71. Comment
in: Trends Neurosci 1991; 14: 55–9.
Alexander GE, Delong MR, Strick PL. Parallel organization of
functionally segregated circuits linking basal ganglia and cortex.
[Review]. Annu Rev Neurosci 1986; 9: 357–81.
Alexander GE, Crutcher MD, DeLong MR. Basal ganglia–
thalamocortical circuits: parallel substrates for motor, oculomotor,
‘prefrontal’, and ‘limbic’ functions. [Review]. Prog Brain Res 1990;
85:119–46.
Baron MS, Vitek JL, Bakay RA, Green J, Kaneoke Y, Hashimoto T,
et al. Treatment of advanced Parkinson’s disease by posterior GPi
pallidotomy: 1-year results of a pilot study [see comments]. Ann
Neurol 1996; 40: 355–66. Comment in: Ann Neurol 1996; 40: 341–
3. Comment in: Ann Neurol 1997; 41: 834–6, Comment in: Ann
Neurol 1997; 42: 129–30.
Benecke R, Rothwell JC, Dick JP, Day BL, Marsden CD.
Performance of simultaneous movements in patients with
Parkinson’s disease. Brain 1986; 109: 739–57.
Benecke R, Rothwell JC, Dick JP, Day BL, Marsden CD. Disturbance
of sequential movements in patients with Parkinson’s disease. Brain
1987a; 110: 361–79.
Benecke R, Rothwell JC, Dick JP, Day BL, Marsden CD. Simple
and complex movements off and on treatment in patients with
Parkinson’s disease. J Neurol Neurosurg Psychiatry 1987b; 50:
296–303.
Berardelli A, Accornero N, Argenta M, Meco G, Manfredi M. Fast
complex arm movements in Parkinson’s disease. J Neurol Neurosurg
Psychiatry 1986; 49: 1146–9.
326
P. Limousin et al.
Bloxham CA, Mindel TA, Frith CD. Initiation and execution of
predictable and unpredictable movements in Parkinson’s disease.
Brain 1984; 107: 371–84.
Brinkman C, Porter R. Supplementary motor area in the monkey:
activity of neurons during performance of a learned motor task.
J Neurophysiol 1979; 42: 681–709.
Brown RG, Jahanshahi M. An unusual enhancement of motor
performance during bimanual movement in Parkinson’s disease.
J Neurol Neurosurg Psychiatry 1998; 64: 813–6.
Brown RG, Jahanshahi M, Marsden CD. Response choice in
Parkinson’s disease. The effects of uncertainty and stimulus–
response compatibility. Brain 1993a; 116: 869–85.
Brown RG, Jahanshahi M, Marsden CD. The execution of bimanual
movements in patients with Parkinson’s, Huntington’s and cerebellar
disease [see comments]. J Neurol Neurosurg Psychiatry 1993b; 56:
295–7. Comment in: J Neurol Neurosurg Psychiatry 1994; 57: 126.
Ceballos-Baumann AO, Obeso JA, Vitek JL, DeLong MR, Bakay R,
Linazasoro G, et al. Restoration of thalamocortical activity after
posteroventral pallidotomy in Parkinson’s disease [letter]. Lancet
1994; 344: 814.
Cody FW, Macdermott N, Matthews PB, Richardson HC.
Observations on the genesis of the stretch reflex in Parkinson’s
disease. Brain 1986; 109: 229–49.
DeLong MR. Primate models of movement disorders of basal
ganglia origin. [Review]. Trends Neurosci 1990; 13: 281–5.
Dick JP, Cantello R, Buruma O, Gioux M, Benecke R, Day BL,
et al. The Bereitschaftspotential, L-DOPA and Parkinson’s disease.
Electroencephalogr Clin Neurophysiol 1987; 66: 263–74.
Dick JP, Rothwell JC, Day BL, Cantello R, Buruma O, Gioux M,
et al. The Bereitschaftspotential is abnormal in Parkinson’s disease.
Brain 1989; 112: 233–44.
Dogali M, Fazzini E, Kolodny E, Eidelberg D, Sterio D, Devinsky
O, et al. Stereotactic ventral pallidotomy for Parkinson’s disease.
Neurology 1995; 45: 753–61.
Eidelberg D, Moeller JR, Ishikawa T, Dhawan V, Spetsieris P,
Silbersweig D, et al. Regional metabolic correlates of surgical
outcome following unilateral pallidotomy for Parkinson’s disease.
Ann Neurol 1996; 39: 450–9.
Evarts EV, Teräväinen H, Calne DB. Reaction time in Parkinson’s
disease. Brain 1981; 104: 167–86.
Feve AP, Bathien N, Rondot P. Chronic administration of L-dopa
affects the movement-related cortical potentials of patients with
Parkinson’s disease. Clin Neuropharmacol 1992; 15: 100–8.
Flowers KA. Visual ‘closed-loop’ and ‘open-loop’ characteristics
of voluntary movement in patients with parkinsonism and intention
tremor. Brain 1976; 99: 269–310.
modulation of thalamocortical interactions after pallidotomy. Hum
Brain Mapp 1994; 2: 45–55.
Grafton ST, Waters C, Sutton J, Lew MF, Couldwell W. Pallidotomy
increases activity of motor association cortex in Parkinson’s disease:
a positron emission tomographic study. Ann Neurol 1995; 37:
776–83.
Hughes AJ, Frankel JP, Kempster PA, Stern GM, Lees AJ. Motor
response to levodopa in patients with parkinsonian motor
fluctuations: a follow-up study over three years. J Neurol Neurosurg
Psychiatry 1994; 57: 430–4.
Hutchinson WD, Lozano AM, Davis KD, Saint-Cyr JA, Lang AE,
Dostrovsky JO. Differential neuronal activity in segments of globus
pallidus in Parkinson’s disease patients. Neuroreport 1994; 5:
1533–7.
Ikeda A, Lüders HO, Burgess RC, Shibasaki H. Movement-related
potentials recorded from supplementary motor area and primary
motor area. Brain 1992; 115: 1017–43.
Jahanshahi M, Brown RG, Marsden CD. Simple and choice reaction
time and the use of advance information for motor preparation in
Parkinson’s disease. Brain 1992a; 115: 539–64.
Jahanshahi M, Brown RG, Marsden CD. The effect of withdrawal
of dopaminergic medication on simple and choice reaction time and
the use of advance information in Parkinson’s disease. J Neurol
Neurosurg Psychiatry 1992b; 55: 1168–76.
Jahanshahi M, Jenkins HJ, Brown RG, Marsden CD, Passingham
RE, Brooks DJ. Self-initiated versus externally triggered movements.
I. An investigation using measurement of regional cerebral blood
flow with PET and movement-related potentials in normal and
Parkinson’s disease subjects [see comments]. Brain 1995; 118: 913–
33. Comment in: Brain 1996; 119: 1045–8.
Jahanshahi M, Brown RG, Limousin P, Rothwell JC, Quinn N.
Posteroventral pallidotomy (PVP) in Parkinson’s disease (PD): II.
Effects on executive function and working memory [abstract]. Mov
Disord 1997; 12 Suppl 1: 130.
Laitinen LV, Bergenheim AT, Hariz MI. Leksell’s posteroventral
pallidotomy in the treatment of Parkinson’s disease [see comments].
J Neurosurg 1992; 76: 53–61. Comment in: J Neurosurg 1992; 77:
487–8.
Lang AE, Lozano AM, Montgomery E, Duff J, Tasker R,
Hutchinson W. Posteroventral medial pallidotomy in advanced
Parkinson’s disease [see comments]. N Engl J Med 1997; 337: 15–
42. Comment in: N Engl J Med 1998; 338: 262–3.
Levy R, Hazrati LN, Herrero MT, Vila M, Hassani OK, Mouroux
M, et al. Re-evaluation of the functional anatomy of the basal
ganglia in normal and parkinsonian states. [Review]. Neuroscience
1997; 76: 335–43.
Goodrich S, Henderson L, Kennard C. On the existence of an
attention-demanding process peculiar to simple reaction time:
converging evidence from Parkinson’s disease. Cogn Neuropsychol
1989; 6: 309–31.
Lozano A, Lang AE, Galvez-Jimenez, Miyasaki J, Duff J,
Hutchinson WD, et al. Effect of GPi pallidotomy on motor function
in Parkinson’s disease [see comments] [published erratum appears
in Lancet 1996; 348: 1108]. Lancet 1995; 346: 1383–7. Comment
in: Lancet 1996; 347: 1490.
Grafton ST, Sutton J, Couldwell W, Lew M, Waters C. Network
analysis of motor system connectivity in Parkinson’s disease:
MacKinnon CD, Kapur S, Hussey D, Verrier MC, Houle S, Tatton
WG. Contributions of the mesial frontal cortex to the premovement
Effect of pallidotomy on motor function
potentials associated with intermittent hand movements in humans.
Hum Brain Mapp 1996; 4: 1–22.
Matsuzaka H, Aizawa H, Tanji J. A motor area rostral to the
supplementary motor area (presupplementary motor area) in the
monkey: neuronal activity during a learned motor task. J
Neurophysiol 1992; 68: 653–62.
Meier MJ, Martin WE. Measurement of behavioural changes in
patients on L-dopa. Lancet 1970; 1: 352–3.
Mushiake H, Inase M, Tanji J. Neuronal activity in the primate
premotor, supplementary, and precentral motor cortex during visually
guided and internally determined sequential movements. J
Neurophysiol 1991; 66: 705–18.
Parent A, Cicchetti F. The current model of basal ganglia
organization under scrutiny [editorial]. [Review]. Mov Disord 1998;
13: 199–202.
Pullman SL, Watts RL, Juncos JL, Chase TN, Sanes JN.
Dopaminergic effects on simple and choice reaction time
performance in Parkinson’s disease. Neurology 1988; 38: 249–54.
Purdue Research Foundation. Purdue Pegboard Test: examination
manual. Chicago: Science Research Associates; 1948.
Requin J. From action representation to movement control. In:
Stelmach GE, Requin J editors. Tutorials in motor behavior II.
Advances in psychology 87. Amsterdam: North-Holland; 1992.
p. 159–79.
Roland PE, Larsen B, Lassen NA, Skinhoj E. Supplementary motor
area and other cortical areas in organization of voluntary movements
in man. J Neurophysiol 1980; 43: 118–36.
Rothwell JC, Obeso JA, Traub MM, Marsden CD. The behaviour
of the long-latency stretch reflex in patients with Parkinson’s disease.
J Neurol Neurosurg Psychiatry 1983; 46: 35–44.
Samuel M, Ceballos-Baumann AO, Turjanski N, Boecker H,
Gorospe A, Linazasoro G, et al. Pallidotomy in Parkinson’s disease
increases supplementary motor area and prefrontal activation during
performance of volitional movements an H2(15)O PET study. Brain
1997; 120: 1301–13.
Samuel M, Caputo E, Brooks DJ, Schrag A, Scaravilli T, Branston
NM, et al. A study of medial pallidotomy for Parkinson’s disease:
Clinical outcome, MRI location and complications. Brain 1998;
121: 59–75.
327
Scott R, Gregory R, Hines N, Carroll C, Hyman N, Papanasstasiou V,
et al. Neuropsychological, neurological and functional outcome
following pallidotomy for Parkinson’s disease. A consecutive series
of eight simultaneous bilateral and twelve unilateral procedures.
Brain 1998; 121: 659–75.
Sheridan MR, Flowers KA, Hurrell J. Programming and execution
of movement in Parkinson’s disease. Brain 1987; 110: 1247–71.
SPSS for Windows (7.5.1). Chicago: SPSS Inc; 1996.
Strafella A, Ashby P, Lozano A, Lang AE. Pallidotomy increases
cortical inhibition in Parkinson’s disease. Can J Neurol Sci 1997;
24: 133–6.
Svennilson E, Torvik A, Lowe R, Leksell L. Treatment of
parkinsonism by stereotactic thermolesions in the pallidal region.
Acta Psychiatr Scand 1960; 35: 358–77.
Tanji J, Kurata K. Contrasting neuronal activity in supplementary
and precentral motor cortex of monkeys. I. Responses to instructions
determining motor responses to forthcoming signals of different
modalities. J Neurophysiol 1985; 53: 129–41.
Touge T, Werhahn KJ, Rothwell JC, Marsden CD. Movementrelated cortical potentials preceding repetitive and random-choice
hand movements in Parkinson’s disease. Ann Neurol 1995; 37:
791–9.
Troster AI, Fields JA, Wilkinson SB, Pahwa R, Miyawaki E, Lyons
KE, et al. Unilateral pallidal stimulation for Parkinson’s disease:
neurobehavioral functioning before and 3 months after electrode
implantation. Neurology 1997; 49: 1078–83.
Vingerhoets FJ, Schulzer M, Calne DB, Snow BJ. Which clinical
sign of Parkinson’s disease best reflects the nigrostriatal lesion?
Ann Neurol 1997; 41: 58–64.
Wichmann T and DeLong MR. Functional and pathophysiological
models of the basal ganglia. [Review]. Curr Opin Neurobiol 1996;
6: 751–8.
Young MS, Triggs WJ, Bowers D, Greer M, Friedman WA.
Stereotactic pallidotomy lengthens the transcranial magnetic cortical
stimulation silent period in Parkinson’s disease. Neurology 1997;
49: 1278–83.
Received June 19, 1998. Revised September 9, 1998.
Accepted October 5, 1998