DOI: 10.

1093/brain/awh237

Brain (2004), 127, 21312143

Effects of subthalamic nucleus stimulation and medication on resting and postural tremor in Parkinsons disease
Molly M. Sturman,1 David E. Vaillancourt,1 Leo Verhagen Metman,4,5 Roy A. E. Bakay4,5 and Daniel M. Corcos14
Departments of 1Movement Sciences, 2Bioengineering and Physical Therapy, University of Illinois at Chicago and Departments of 4Neurological Sciences and 5Neurosurgery, Rush Presbyterian-St Lukes Medical Center, Chicago, IL, USA
3

Correspondence to: Molly M. Sturman, Department of Movement Sciences (M/C 994), College of Applied Health Sciences, University of Illinois at Chicago, 808 S. Wood Street, 690 CME, Chicago, IL 60612, USA E-mail: msturm3@uic.edu

Summary
Deep brain stimulation (DBS) of the subthalamic nucleus (STN) and antiparkinsonian medication have proved to be effective treatments for tremor in Parkinsons disease. To date it is not known how and to what extent STN DBS alone and in combination with antiparkinsonian medication alters the pathophysiology of resting and postural tremor in idiopathic Parkinsons disease. The purpose of this study was to examine the effects of STN DBS and antiparkinsonian medication on the neurophysiological characteristics of resting and postural hand tremor in Parkinsons disease. Resting and postural hand tremor were recorded using accelerometry and surface electromyography (EMG) from 10 Parkinsons disease patients and 10 matched control subjects. The Parkinsons disease subjects were examined under four treatment conditions: (i) off treatment; (ii) STN DBS; (iii) medication; and (iv) medication plus STN DBS. The amplitude, EMG frequency, regularity, and 18 Hz tremorEMG coherence were analysed. Both STN DBS and medication reduced the amplitude, regularity and tremorEMG coherence, and increased the EMG frequency of resting and postural tremor in Parkinsons disease. STN DBS was more effective than medication in reducing the amplitude and increasing the frequency of resting and postural tremor to healthy physiological levels. These ndings provide strong evidence that effective STN DBS normalizes the amplitude and frequency of tremor. The ndings suggest that neural activity in the STN is an important modulator of the neural network(s) responsible for both resting and postural tremor genesis in Parkinsons disease.

Keywords: tremor; Parkinsons disease; deep brain stimulation; subthalamic nucleus Abbreviations: ANOVA = analysis of variance; ApEn = approximate entropy; DBS = deep-brain stimulation; RMS = root mean square; STN = subthalamic nucleus; UPDRS = Unied Parkinsons Disease Rating Scale; VIM = Ventral intermediate Received March 31, 2004. Revised May 6, 2004. Accepted May 13, 2004. Advanced Access publication July 7, 2004

Introduction
Tremor is one of three cardinal signs of idiopathic Parkinsons disease, and individuals with the disease often exhibit both resting and postural tremor (Deuschl et al., 1998). Tremor in either form can be a debilitating neurological symptom because its presence severely disrupts activities of daily living, especially self-care tasks (Wasielewski et al., 1998). Deep brain stimulation (DBS) of the subthalamic nucleus (STN) (Deep Brain Stimulation for Parkinsons Disease Study Group, 2001; Benabid, 2003; Krack et al., 2003) and antiparkinsonian medication (Cotzias et al., 1969; Yahr et al., 1969) are effective treatments for Parkinsons disease. The extent to which STN Brain Vol. 127 No. 9
#

DBS and medication change the neurophysiology of bradykinesia (Vaillancourt et al., 2004), gait (Faist et al., 2001; Bastian et al., 2003) and postural instability (Maurer et al., 2003) in Parkinsons disease to healthy levels have been established. While STN DBS and medication reduce tremor rating scales (Olanow et al., 1991; Friedman et al., 1997; Krack et al., 1998; Rodriguez et al., 1998), it is not currently known how or to what extent STN DBS alters the pathophysiology of resting and postural tremor in Parkinsons disease. Previous research has used multiple dependent measures to study the neurophysiological mechanisms of tremor in

Guarantors of Brain 2004; all rights reserved

2132

M. M. Sturman et al.
and control groups. All clinical and motor control evaluations for all of the patients were performed an average of 6.7 months after surgery and after optimization of DBS parameters had occurred. A priori inclusion criteria were established for the study, and the rst 10 Parkinsons disease patients who received STN DBS and met the criteria were chosen for the study. Patients were examined by a movement disorders neurologist and included in the study if they: (i) had idiopathic Parkinsons disease as outlined by the Parkinsons Disease Society Brain Bank diagnostic criteria (Hughes et al., 1992a, b); (ii) were deemed to have resting and/or postural tremor which was consistent with the Movement Disorders Consensus Statement on Tremor Type I or Type II classication (Deuschl et al., 1998); (iii) had a resting and/or postural tremor which both responded positively to levodopa therapy; and (iv) had a positive clinical improvement [greater than a 15% reduction in scores from the entire motor section of the Unied Parkinsons Disease Rating Scale (UPDRS)] from the off-treatment condition compared with the on STN DBS condition. All subjects gave informed consent to all experimental procedures, which were approved by the local institutional review board at The University of Illinois at Chicago, USA. In all cases a quadripolar stimulation electrode was placed in the STN opposite to the most severely affected body side. Surgery was performed as part of clinical care according to standard procedures (Deep Brain Stimulation for Parkinsons Disease Study Group, 2001; Starr et al., 1998). Briey, the anatomical target was based on standard stereotactic coordinates while the patient was in the MRI scanner with a Leksell stereotactic head frame in place (Cohn et al., 1998). MRI images were subsequently reformatted and coordinates rened using a Stealth system. Intra-operatively, microelectrode penetrations were made through a burr hole to verify the neurophysiological target (Starr et al., 1998, 2000; Vitek et al., 1998). Responses to sensory stimuli (passive movements of upper and lower limbs) were sought to identify the sensorimotor portion of the STN, where the DBS lead was subsequently placed. Postoperatively, MRI demonstrated appropriate placement of the DBS lead. The same movement disorders neurologist administered the entire motor section of the UPDRS to each patient before (pre) and after (post) surgery, and these scores are reported bilaterally. Following surgery, stimulation parameters were optimized in order to reduce presurgery scores on the motor section of the UPDRS. Optimization took place during several visits over the ensuing months. At the time of the experiments, in each case the pulse width and frequency were 60 ms and 185 Hz, respectively. The stimulation intensity varied from 1.8 to 3.1 V (mean = 2.26 V, SD = 0.42), and stimulation was delivered in a monopolar fashion (Table 1). The average presurgical motor UPDRS score was 46.05 (SD = 12.90) off medication. Average postsurgical UPDRS scores were 37.85 (11.88) off treatment, 14.2 (10.42) on medication, 24.17 (11.51) on STN DBS, and 10.75 (8.98) on medication plus STN DBS. Presurgery UPDRS scores were signicantly different from postsurgery off-treatment scores using a Wilcoxon matched pairs test (Z = 2.29, P < 0.05).

Parkinsons disease. These measures include frequency (Homberg et al., 1987), amplitude (Stiles and Pozos, 1976), regularity (Vaillancourt and Newell, 2000) and tremor electromyogram (EMG) coherence (Brown et al., 1997; Vaillancourt and Newell, 2000). In contrast to the 812 Hz frequency of healthy physiological tremor (Halliday and Redfearn, 1956; Lakie et al., 1986), the frequency of resting and postural tremor in Parkinsons disease is reduced to 36 and 412 Hz, respectively (Findley et al., 1981; Elble and Koller, 1990; Deuschl et al., 1998). There is a negative relation between the frequency and amplitude of tremor oscillations, such that if the frequency decreases the amplitude increases (Stiles and Pozos, 1976). The degree of rhythmicity in tremor, assessed using approximate entropy (ApEn) (Pincus and Goldberger, 1994), is greater and more regular in patients with Parkinsons disease compared with healthy control subjects (Vaillancourt and Newell, 2000). In addition, the amount of squared correlation between the tremor and EMG signals (tremorEMG coherence) in patients with Parkinsons disease is increased in the low-frequency bands (Vaillancourt and Newell, 2000). The purpose of this study was to examine the effects of therapeutic doses of medication and STN DBS on the neurophysiological characteristics of resting and postural hand tremor in patients with idiopathic Parkinsons disease. The study had three main objectives. First, we examined whether STN DBS and medication reduced resting and postural tremor amplitude, regularity and tremorEMG coherence, and whether the treatments increased resting and postural tremor EMG frequency. Secondly, the study directly compared the effects of STN DBS with those of medication on the neurophysiology of resting and postural tremor. The third objective was to determine the extent to which STN DBS alone, medication alone, or the combination of STN DBS and medication altered the neural control of resting and postural tremor to healthy physiological levels. Ten individuals diagnosed with Parkinsons disease who received STN DBS participated in the study. The patients hand tremor data were compared with data from neurologically healthy controlsubjects who were matched according to age and gender. Patients were tested in four treatment conditions: (i) off treatment; (ii) STN DBS; (iii) medication; and (iv) medication plus STN DBS. Each patient and control subject was examined during resting and postural hand tremor conditions. The ndings from this study establish neurophysiological correlates underlying the clinical efcacy of medication and STN DBS in the treatment of resting and postural hand tremor in Parkinsons disease. Furthermore, the rst evidence is presented on whether STN DBS and/or medication restore the neurophysiology of tremor to healthy physiological levels.

Experimental setup
The experiments for the patients were performed on two consecutive days in each of four treatment conditions: (i) off treatment; (ii) STN DBS; (iii) medication; and (iv) medication plus STN DBS. Examination of Figs 16 does not show trends in the data, suggesting that test order was not a limitation in the study. On day 1, testing of condition 1 took place between 9 and 11 a.m. and condition 2 occurred between 1 and 3 p.m. On day 2, the same testing schedule as day 1 was set for

Methods Subjects
Ten patients with Parkinsons disease, aged 52.9 years (SD = 8.03), and 10 age- and gender-matched control subjects, aged 52.7 years (8.26), participated in the study. A paired t test conrmed that there were no statistically signicant differences in age between the patient

Effects of STN DBS/medication on tremor in PD Table 1 Prole of each subject

Subject Age (years) 43 49 53 63 65 46 47 47 Gender Post-surgical UPDRS off treatment Item 20 3 0 3 3 3 3 1 3 3 1 2 3 3 2 1 3 Deep brain stimulation parameters Item 21 2.8 2.6 2.0 2.2 1.8 2.1 2.0 2.0 Monopolar: 0 C Monopolar: 1 C Monopolar: 1 C Monopolar: 0 C Monopolar: 2 C Monopolar: 1 C Monopolar: 1 C Monopolar: 1 C Anti-parkinsonian medications

2133

1 2 3 4 5 6 7 8

F F F F F M M M

53

2.0

Monopolar: 1 C

10

63

3.1

Monopolar: 0 C

Carbidopa/levodopa 25/100 0.5 tab Carbidopa/levodopa 25/100 0.5 tab Pergolide 2 mg 1 tab Amantadine 100 mg 1 tab Carbidopa/levodopa 25/100 1 tab Carbidopa/levodopa CR 50/200 1 tab Carbidopa/levodopa 25/100 1 tab Amantadine 100 mg 1 tab Carbidopa/levodopa CR 25/100 1 tab Pramipexole 1.5 mg 1 tab Carbidopa/levodopa 25/100 3 tabs Pramipexole 1 mg 3 tabs Carbidopa/levodopa 25/100 1.5 tab Carbidopa/levodopa CR 50/200 1 tab Carbidopa/levodopa 25/100 1 tab Pergolide 1 mg 1 tab Amantadine 100 mg 1 tab Carbidopa/levodopa 25/100 2 tabs Ropinirole 5 mg 2 tabs Amantadine 100 mg 1 tab Entacapone 200 mg 2 tabs Carbidopa/levodopa CR 50/200 1.5 tab Carbidopa/levodopa 25/100 1 tab Entacapone 200 mg 1 tab Pergolide 1 mg 1 tab

conditions 3 and 4, respectively. The control subjects were tested on one day. To ensure that each Parkinsons disease patient was in the operationally dened off state, patients were always tested in the off condition following 12 h without the specic treatment (Langston et al., 1992). For instance, the night prior to day 1 both treatments were withheld, and the night prior to day 2 only STN DBS was withheld. Each patient stayed in a hotel close to the laboratory and was examined using a Medtronic Console Programmer (Model 7432 Minneapolis, Minnisota, USA) on each night prior to days 1 and 2 to make certain that the stimulator was off. To ensure that each patient was on STN DBS, the motor section of the UPDRS was administered 90 min after activation of the stimulator. All patients had reduced UPDRS values 90 min following reactivation of the stimulator. The patients normal medication dose schedule was restarted on day 1 after tremor testing was completed. For all patients, resumption of optimal levodopa therapy began at 3 p.m. on day 1 and continued through day 2 of testing. On day 2, all patients took medication between 6 and 7 a.m. and then again at 9 a.m. Forty-ve minutes after the 9 a.m. dose, the motor section of the UPDRS was administered to ensure that medication efcacy was optimized. All patients responded to levodopa therapy, so tremor testing proceeded 50 min after administration of the 9 a.m. levodopa dose. For the afternoon session on day 2, patients took medication at 12:30 p.m., and the same protocol for the morning session was followed. For all patients in this study, the afternoon tremor testing commenced 50 min later. At each dose, patients took the medications listed in Table 1.

Tremor testing
Subjects were positioned in a 42 cm high straight-back chair with an arm that was 43 cm in length and 58 cm from the ground. The arm of

the chair supported the subjects forearm, with the ulnar styloid process aligned with the end of the chair. The elbow joint was exed to approximately 90 , the shoulder joint was slightly abducted, and the forearm was pronated. A small table was positioned 30 cm in front of the subject. The height of the table was adjusted to be level with the tip of the subjects hand when the wrist and ngers were extended parallel with the oor. The table served as the visual target to enable the subject to maintain the wrist in a neutral position. The width of the table edge was 0.5 inches. A calibrated Coulbourn type V 9441 (Allen town Pennsylvania, USA) miniature solid-state piezoresistive accelerometer was taped to the hand (2 cm proximal to the middle of the second metacarpophalangeal joint). A Coulbourn Lab Linc V System V72-25A resistive bridge strain gauge transducer with an excitation voltage of 65 V amplied the acceleration signal. The accelerometers resolution was 0.01 m/s2, and a 12-bit analoguedigital converter sampled the signal at 1000 samples per second. To remove the DC effects of gravity, all accelerometry data were collected in AC mode with a 1 Hz analogue high-pass lter (Elble and Koller, 1990). Surface EMG was used to measure the neuromuscular activity in the extensor digitorum and the exor digitorum supercialis. Electrode placement was determined by muscle palpation during active nger extension and exion with the wrist in a neutral, extended position. The EMG signal was amplied (gain = 1000) and bandpass-ltered between 20 and 450 Hz (Delsys, Boston, MA, USA). The study examined two types of tremor. In the resting tremor condition, subjects were instructed to relax their forearm and hand muscles and allow the wrist to dangle unsupported over the edge of the supportive surface for 30 s (Marsden et al., 1969; Burne et al., 1984;

2134

M. M. Sturman et al.

Henderson et al., 1994; Raethjen et al., 2000). In the postural tremor condition, subjects were asked to maintain their wrist and hand in a neutral, extended position while keeping it level with the target for 30 s (Elble and Randall, 1976; Homberg et al., 1987; Henderson et al., 1994; Raethjen et al., 2000). Subjects performed three trials at each respective condition and were explicitly instructed throughout each trial not to suppress resting or postural tremor.

Coherence between acceleration and extensor EMG activity

The coherence between the acceleration and EMG signals estimates the degree of motor unit entrainment affecting the acceleration signal (Elble and Randall, 1976). The cross-spectrum was calculated using Welchs averaged periodogram method (bin width = 0.7813 Hz). Coherence between the acceleration and extensor EMG signals (tremorEMG coherence) was calculated, and a 95% condence interval was determined based on the number of disjoint sections from the sample record (Rosenberg et al., 1989; Farmer et al., 1993). This method revealed a broad band of signicant coherence. Next, we determined the peak coherence value in the coherence spectrum in the 18, 915, 1530 and 3550 Hz bins.

Data analysis
Before performing all data analyses, the acceleration and EMG data were conditioned by the following methods. The EMG data were digitally rectied, and the acceleration and EMG data were downsampled to 200 Hz (Vaillancourt and Newell, 2000). The acceleration and EMG data were digitally ltered using a fourth-order Butterworth lter with a low-pass cutoff frequency of 60 Hz for the EMG signals and 30 Hz for the acceleration signals. All data analyses were performed on the entire 30 s data segmentthe same for each subject, trial, and condition. The dependent measures were averaged across the three trials for each condition. All data processing and subsequent time and frequency analyses were performed using software written in Matlab (MathWorks, Natick, MA, USA).

Frequency
The dominant frequency in the EMG spectrum recorded during resting and postural hand tremor was examined using autospectral analysis of the extensor EMG signal between 3 and 15 Hz.

Statistical analysis
The dependent variables described in the preceding sections were evaluated by analysis of variance (ANOVA). First the effects of medication and STN DBS on tremor were evaluated using a three-way ANOVA: medication (off versus on) STN DBS (off versus on) tremor type (resting versus postural). Secondly, a direct comparison between medication and STN DBS was performed using a one-way ANOVA: treatment (on medication versus on STN DBS). Thirdly, a comparison of each treatment alone and the combination of both treatments with healthy tremor was performed using a two-way ANOVA: group (Parkinsons disease versus control) by tremor type (resting versus postural). The clinical postsurgical UPDRS hand tremor data were evaluated using a non-parametric Wilcoxon matched pairs test. Postsurgical, off-treatment, unilateral UPDRS hand tremor scores (Table 1) were compared with scores on medication, STN DBS, and medication plus STN DBS. UPDRS hand tremor scores were considered unilaterally because patients received unilateral STN DBS, and the dependent measures quantied hand tremor unilaterally. All statistical analyses were interpreted as signicant when there was no more than a 5% chance of making a type I error. For important statistical comparisons, we wanted to protect against making a type II error. Therefore, additional power analyses were conducted for non-signicant ndings when comparing between medication and STN DBS, and comparing between each treatment and control subjects. The results from this power analysis are reported after the F values in two ways. Both the percentage of power in the current study with the current sample size and the sample size required to achieve 90% power are reported. All statistical analyses were completed using Statistica (StatSoft, Tulsa, OK, USA).

Amplitude and regularity of tremor

The amplitude and regularity of hand tremor were quantied in the same way as in our study of essential tremor (Vaillancourt et al., 2003). The amplitude of tremor was calculated using a method in which spectral analysis is used to extract the root mean square (RMS) tremor displacement from the dominant frequency of tremor (Stiles, 1976). The RMS displacement is obtained as if it had resulted from a singlefrequency, periodic oscillation. This assumption is reasonable when examining hand tremor because the 510 Hz peak dominates the spectrum (Elble and Koller, 1990). First, the spectrum of the tremor acceleration is calculated. Secondly, the sum of the central three spectral values of the principal band from the spectrum is taken as the variance of the tremor oscillations at the dominant frequency ( f ). Thirdly, the RMS acceleration at this frequency ( f ) was obtained as the square root of this sum and then converted to units of displacement by dividing by the square of v (v = 2f ). The RMS displacement value is reported in centimetres. The degree of rhythmicity or regularity of hand tremor was calculated using approximate entropy (ApEn), a measure of the time-dependent structure of the signal (Pincus, 1991). ApEn returns a value in the approximate range of 02, and it reects the predictability of future values in a time series based on previous values. For example, a sine wave has accurate short- and long-term predictability, and this corresponds to an ApEn value near 0. If varying amplitudes of white Gaussian noise are added to a sine wave, then the ApEn value will increase. This increases the uncertainty of making future time series predictions when random elements are added. For a completely random signal (viz., white Gaussian noise), each future value in the time series is independent and unpredictable from previous values, and the ApEn value tends towards 2. Because ApEn is inuenced by ltering characteristics, data length, and other factors (Pincus and Goldberger, 1994), the same algorithm and parameter settings (m = 2; r = 0.2 SD of the signal) were used to be consistent with previous work (Pincus and Goldberger, 1994; Vaillancourt et al., 2003).

Results Clinical effects

All 10 patients demonstrated positive clinical benets from either medication or STN DBS. Postsurgical, unilateral UPDRS resting hand tremor scores (item 20) averaged across subjects were 2.2 (SD = 1.32) off treatment, 1.4 (1.26) on medication, 0.80 (0.92) on STN DBS, and 0 (0)

Effects of STN DBS/medication on tremor in PD

2135

Fig. 1 Resting tremor acceleration, extensor EMG and exor EMG of a patient with Parkinsons disease off treatment, on medication and on STN DBS, and a control subject. (A) The Parkinsons disease subject off treatment. (B) The same subject on medication. (C) The same subject on STN DBS. (D) The control subject. The time series shows a 2 s recording during the same time period from the 30 s trial. The units for the y axis of the extensor EMG are in millivolts (mV) and the units for the acceleration y axis are in m/s2.

Fig. 2 Acceleration, extensor EMG and exor EMG during postural tremor of a patient with Parkinsons disease off treatment, on medication, on STN DBS, and a control subject. The recording status in AD is the same as in Fig. 1AD.

Characterization of resting and postural tremor in individual subjects

Figure 1 depicts acceleration, extensor EMG and exor EMG signals from a patient with Parkinsons disease (Case 6, Table 1) and a control subject during the resting tremor condition. Off treatment (Fig. 1A), the Parkinsons disease subject had a high-amplitude tremor (RMS = 11.05 cm) with a high degree of regularity, which corresponded to a low ApEn value equal to 0.50. The EMG revealed the pattern characteristic of Parkinsons disease resting tremoralternating bursts of exor and extensor EMG activity. In Fig. 1B, medication decreased the amplitude of tremor to 1.33 cm and increased ApEn to 0.52. On STN DBS (Fig. 1C), the amplitude of the tremor also decreased to 0.03 cm and ApEn increased to 0.60. Resting physiological tremor data collected from a representative control subject (Fig. 1D) had a minute amplitude of 0.008 cm and ApEn equal to 0.66. In contrast to the bursting

on medication plus STN DBS. Compared with off treatment, medication (Z = 2.2, P < 0.05), STN DBS (Z = 2.37, P < 0.01) and medication plus STN DBS (Z = 2.52, P < 0.01) reduced unilateral hand tremor UPDRS scores for item 20. For postural tremor (item 21), the average postsurgical unilateral hand tremor UPDRS scores were 1.90 (1.10) off treatment, 1.1 (1.10) on medication, 1.1 (0.99) on STN DBS, and 0.2 (0.42) on medication plus STN DBS. Compared with off treatment, medication (Z = 2.37, P < 0.01), STN DBS (Z = 2.03, P < 0.05) and the combination of medication plus STN DBS (Z = 2.52, P < 0.01) reduced postsurgical unilateral postural hand tremor UPDRS scores. Furthermore, patient reports concurred with UPDRS scores that both treatments reduced the amount of tremor experienced by the patients.

2136

M. M. Sturman et al.

Fig. 3 Resting and postural tremor amplitude of the Parkinsons disease patients (n = 10) off medication (closed hexagons), patients on medication (open triangles) and control subjects (n = 10) (closed squares). Each value was averaged across subjects (mean 6 SE) under each treatment condition during resting and postural tremor. (A) RMS displacement amplitude (cm) measured from the acceleration signal and converted to displacement units (Stiles, 1976) for resting tremor. (B) RMS displacement for postural tremor. Refer to text for signicant differences.

Fig. 4 Resting and postural tremor regularity of the Parkinsons disease patients (n = 10) off medication (closed hexagons), patients on medication (open triangles) and control subjects (n = 10) (closed squares). Each value was averaged across subjects (mean 6 SE) under each treatment condition during resting and postural tremor. (A) Regularity of the acceleration signal quantied with ApEn for resting tremor. (B) Regularity (ApEn) for postural tremor. Refer to text for signicant differences.

EMG pattern in Fig. 1A, the EMGs depicted in Fig. 1BD show a small, evenly dispersed amount of EMG activity consistent with a resting limb state. Figure 2 shows time series data from the postural tremor condition. Off treatment (Fig. 2A), the Parkinsons disease patient (Case 3, Table 1) had a high-amplitude tremor of

18.95 cm with ApEn equal to 0.46. Similar to resting tremor, postural tremor for the Parkinsons disease patient off treatment presented with alternating bursts of exor and extensor EMG activity. Medication (Fig. 2B) reduced the amplitude of postural tremor to 1.32 cm and increased ApEn to 0.58. STN DBS (Fig. 2C) reduced the amplitude of

Effects of STN DBS/medication on tremor in PD

2137

Fig. 5 Resting and postural tremor EMG frequency for Parkinsons disease patients (n = 10) and control subjects (n = 10), off medication (closed hexagons), on medication (open triangles) and control subjects (closed squares). (A) EMG spectrum for resting tremor from a Parkinsons disease patient off treatment. (B) EMG spectrum for postural tremor from the Parkinsons disease patient shown in A off treatment. (C) EMG spectrum for resting tremor from the Parkinsons disease patient shown in A on STN DBS. (D) EMG spectrum for postural tremor from the Parkinsons disease patient shown in A on STN DBS. (E) Resting tremor EMG frequency averaged across subjects (mean 6 SE). (F) Postural tremor EMG frequency averaged across subjects (mean 6 SE). Refer to text for signicant differences.

postural tremor to 0.18 cm and increased ApEn to 0.63. Both treatments reduced the prominent alternating bursts of EMG activity that occurred off treatment. In Fig. 2D the representative control subjects postural tremor had an

amplitude of 0.03 cm. The control subjects physiological postural tremor had an ApEn value of 0.76, which was higher than the value for the Parkinsons disease patient on each treatment independently or in combination.

2138

M. M. Sturman et al.

Fig. 6 Resting and postural 1-8 Hz tremor-EMG coherence between the acceleration and extensor EMG of Parkinsons disease patients (n = 10) and control subjects (n = 10), patients off medication (closed hexagons), patients on medication (open triangles) and control subjects (closed squares). (A) Resting tremorEMG coherence from a Parkinsons disease patient off treatment (dotted line) and on STN DBS (solid line). (B) Postural tremorEMG coherence from the same Parkinsons disease patient shown in A off treatment and on STN DBS. (C) Resting tremorEMG coherence averaged across subjects (mean 6 SE). (D) Postural tremorEMG coherence averaged across subjects (mean 6 SE). The thick dotted line in A and B is the 95% condence line for signicant coherence. Refer to text for signicant differences.

Amplitude of tremor Effects of medication and STN DBS

Figure 3A and B depict across-subject average RMS displacements for resting and postural tremor from Parkinsons disease patients and control subjects. The within-subject ANOVA revealed that postural tremor displacement was greater than resting tremor displacement for the Parkinsons disease subjects [F(1,9) = 5.76, P < 0.05]. There was no interaction between medication and/or STN DBS and tremor type. However, there was a signicant interaction between medication and STN DBS [F(1,9) = 9.59, P < 0.01]. Because of the interaction, one-way ANOVAs were conducted on the medication and STN DBS data separately. Medication signicantly

reduced displacement when patients were off STN DBS [F(1,9) = 10.80, P < 0.01]. Medication did not affect displacement when patients were on STN DBS [F(1,9) = 1.42, P = 0.26, 10%, n = 168]. STN DBS reduced displacement when patients were either off medication [F(1,9) = 13.02, P < 0.01] or on medication [F(1,9) = 7.15, P < 0.05]. In order to determine directly which of the two treatments was more efcacious, tremor amplitude of Parkinsons disease patients on medication was compared with amplitude on STN DBS. STN DBS reduced tremor amplitude below that of medication [F(1,9) = 6.71, P < 0.05]. These results show that each treatment reduces tremor, but that STN DBS is more effective than medication for tremor amplitude reduction.

Effects of STN DBS/medication on tremor in PD

2139

Comparison of each treatment with controls

Figure 3A and B also illustrate that resting and postural tremor amplitude for control subjects was similar to Parkinsons disease patients when on STN DBS [F(1,18) = 1.51, P = 0.23, 9%, n = 272] alone, and under the combination of medication plus STN DBS [F(1,18) = 3.37, P = 0.08, 11%, n = 108]. However, patients in the medication-alone condition had greater tremor amplitude compared with control subjects [F(1,18) = 7.15, P < 0.01]. The results from the three between-group ANOVAs showed non-signicant interactions between group and tremor type, and non-signicant main effects for tremor type. These ndings demonstrate that only STN DBS normalized the amplitude of tremor.

The statistical analysis resulted in a main effect for medication [F(1,9) = 10.48, P < 0.01] and STN DBS [F(1,9) = 19.17, P < 0.01] on resting and postural tremor EMG frequency (Fig. 5E and F). There was also a main effect of tremor type [F(1,9) = 14.05, P < 0.01]. Figure 5E and F present group data, which show that postural tremor EMG frequency is greater than resting tremor EMG frequency. There were no signicant interactions between treatment condition and tremor type. When directly comparing the two treatments, Fig. 5E and F and the statistical analysis show that Parkinsons disease patients on STN DBS had greater EMG frequencies compared with the on-medication condition [F(1,9) = 9.28, P < 0.01]. Thus, both medication and STN DBS increased tremor frequency, yet STN DBS had a greater effect on frequency than medication.

Approximate entropy of tremor Effects of medication and STN DBS

Figure 4A and B and statistical analysis revealed that medication [F(1,9) = 15.45, P < 0.01] and STN DBS [F(1,9) = 15.04, P < 0.01] increased ApEn (reduced regularity) during resting and postural tremor conditions. There were no differences in ApEn between resting and postural tremor, and no signicant interactions with the treatment conditions. Furthermore, there was no difference between the medication alone and STN DBS alone conditions [F(1,9) = 0.00, P = 0.99, 9%, n = 250]. These ndings show that medication and STN DBS equally reduce the regularity of tremor.

Comparison of each treatment with healthy controls

Figure 5E and F present mean EMG frequencies from healthy subjects for resting tremor and postural tremor. Patients on medication alone had lower EMG frequencies than healthy subjects [F(1,18) = 23.65, P < 0.01]. However, EMG frequencies of Parkinsons disease patients on STN DBS alone [F(1,18) = 2.72, P = 0.12, 12%, n = 102] and the combination of medication plus STN DBS [F(1,18) = 2.81, P = 0.11, 12%, n = 105] were statistically indistinguishable from those of control subjects. For each of the three between-group comparisons there was a main effect of tremor type (P values < 0.01), such that postural tremor EMG frequency was greater than resting tremor EMG frequency. In summary, STN DBS was the only treatment that normalized tremor EMG frequency.

Comparison of each treatment with controls

When comparing Parkinsons disease patients under each treatment condition with control subjects, Parkinsons disease patients always had lower ApEn values (Fig. 4A and B) [medication alone, F(1,18) = 9.29, P < 0.01; STN DBS alone, F(1,18) = 11.53, P < 0.01; medication STN DBS, F(1,18) = 6.52, P < 0.01]. For each of the three between group comparisons, there was a main effect of tremor type (P values < 0.05), with ApEn values greater for postural tremor than for resting tremor. These analyses show that the tremor of Parkinsons disease patients is still more regular than that of control subjects regardless of treatment interventions.

TremorEMG coherence Effects of medication and STN DBS

Figure 6A depicts resting tremor from a Parkinsons disease patient (Case 4, Table 1) showing a reduction in 18 Hz resting tremorEMG coherence from 0.70 to 0.32 as a result of STN DBS. Similarly, in the same patient during postural tremor, STN DBS (Fig. 6B) decreased tremorEMG coherence in the 18 Hz bin from 0.75 to 0.42. We also examined resting and postural tremorEMG coherence between 9 and 15 Hz, 15 and 30 Hz, and 35 and 50 Hz using exactly the same methods as described for the 18 Hz bin (see Methods section: Coherence between acceleration and extension EMG activity). However, we did not nd statistically signicant differences between treatment conditions in these higher frequency bins. Figure 6C and D depict across subject average resting and postural tremorEMG coherence for Parkinsons disease patients and control subjects. From the within-subject ANOVA, there was no effect of tremor type on tremor EMG coherence, and no interactions between the treatments and tremor type. Conversely, there was a signicant interaction between medication and STN DBS [F(1,9) = 5.12, P < 0.05]. Due to this interaction, one-way ANOVAs were conducted on the medication and STN DBS data separately. STN DBS reduced tremorEMG coherence both when patients were

Frequency of tremor Effects of medication and STN DBS

Figure 5A and C depict the EMG spectrum during resting tremor from a representative Parkinsons disease patient (Case 8, Table 1) off treatment and on STN DBS. Figure 5B and D show the EMG spectrum during postural tremor from the same Parkinsons disease patient off treatment and on STN DBS. In the off-treatment condition (Fig. 5A and B), the peak frequency in the EMG spectrums is below 10 Hz, and there is high peak power in the spectrum in the low peak frequency range. In the on STN DBS condition, the peak frequency in the EMG spectrum is at a greater frequency ($10 Hz), and the amount of power in the spectrum is reduced for both resting (Fig. 5C) and postural (Fig. 5D) tremor.

2140

M. M. Sturman et al. differences are important because they suggest that STN DBS operates with different magnitudes of clinical efcacy based on the task and/or the specic motor decit. Our results are discussed in terms of the clinical signicance and insight they provide for the central oscillator(s) of resting and postural tremor in Parkinsons disease.

off medication [F(1,9) = 23.42, P < 0.01] and when they were on medication [F(1,9) = 6.42, P < 0.05]. Medication also signicantly reduced tremorEMG coherence when patients were off STN DBS [F(1,9) = 10.48, P < 0.01]. However, when patients were on STN DBS the addition of medication did not affect tremorEMG coherence [F(1,9) = 0.87, P = 0.37, 9%, n = 150]. Finally, when medication and STN DBS were compared directly, tremorEMG coherence for the STN DBS condition was lower than for the medication condition, yet the analysis did not reach statistical signicance [F(1,9) = 3.61, P = 0.09, 29%, n = 41]. In summary, medication and STN DBS reduced tremorEMG coherence compared with off treatment, but only STN DBS provided additional reductions in coherence when on medication.

Tremor amplitude
The nding that medication (Koller, 1986; Koller et al., 1989; Henderson et al., 1994) and STN DBS (Krack et al., 1998; Rodriguez et al., 1998) reduced the amplitude of resting and postural hand tremor in patients with Parkinsons disease is consistent with previous literature. However, the results from this study are the rst to determine the extent to which STN DBS normalizes resting and postural tremor to healthy levels. Specically, we found that the resting and postural tremor amplitude of Parkinsons disease patients on STN DBS alone were both similar to the tremor amplitude of control subjects. In contrast, medication alone did not reduce tremor amplitude to the level of control subjects. While it is possible that suprathreshold doses of medication could have further reduced tremor, such doses would also cause dyskinesias, thereby compromising the clinical benets of medical therapy. Previous reports demonstrate that DBS of the ventral intermediate (VIM) thalamic nucleus reduces postural tremor amplitude in patients with Parkinsons disease (OSuilleabhain et al., 2003), but the extent to which VIM DBS restores tremor to control levels remains unclear. Nevertheless, clinical reports indicate that STN DBS also reduces rigidity, bradykinesia, and postural instability in patients with Parkinsons disease (DeepBrain Stimulation for Parkinsons Disease Study Group, 2001; Benabid, 2003), supporting the current view that the STN rather than the VIM thalamic nucleus may be the preferred neurosurgical target for the treatment of parkinsonian patients exhibiting the classic triad of abnormal motor signs.

Comparison of each treatment with healthy controls

Mean resting (Fig. 6C) and postural (Fig. 6D) tremorEMG coherence for control subjects was lower than tremorEMG coherence for patients in each treatment condition. The statistical analysis conrmed the observation from Fig. 6C and D. Medication [F(1,18) = 15.10, P < 0.01], STN DBS [F(1,18) = 5.80, P < 0.05] and the combination of medication plus STN DBS [F(1,18) = 7.85, P < 0.01] did not reduce tremorEMG coherence in Parkinsons disease patients to healthy physiological levels. The results from the three between-group ANOVAs also showed non-signicant interactions between group and tremor type, and non-signicant main effects for tremor type. These analyses demonstrate that the treatment interventions did not normalize tremorEMG coherence.

Discussion
The purpose of this study was to examine the effects of therapeutic doses of medication and STN DBS on the neurophysiological characteristics of resting and postural hand tremor in patients with Parkinsons disease. Consistent with previous work on essential tremor (Vaillancourt et al., 2003), the tremor amplitude, regularity, tremorEMG coherence and EMG frequency were quantied. There were three important ndings from this study. First, medication and STN DBS reduced resting and postural tremor amplitude, regularity and tremor EMG coherence while increasing tremor frequency. Secondly, STN DBS was more effective than medication in reducing the amplitude and increasing the frequency of resting and postural tremor. Thirdly, STN DBS caused resting and postural tremor amplitude and frequency to approximate healthy physiological tremor, but medication did not. This is similar to the normalizing effect of STN DBS on gait stride length and gait velocity during treadmill walking in Parkinsons disease patients (Faist et al., 2001). In contrast, STN DBS did not normalize tremor regularity and 18 Hz tremorEMG coherence, which is comparable to the effects of STN DBS on self-paced, overground walking (Bastian et al., 2003), postural control (Maurer et al., 2003) and bradykinesia (Vaillancourt et al., 2004). These

Tremor regularity, frequency and coherence

Increased regularity from physiological output is considered a biomarker for ageing and disease in the cardiovascular, endocrine and nervous systems (Lipsitz and Goldberger, 1992; Vaillancourt and Newell, 2002). Previous research has shown that Parkinsons disease patients have more regular tremor compared with control subjects during resting and postural tremor (Vaillancourt and Newell, 2000). Our ndings of reduced tremor regularity following STN DBS and medication are consistent with the effects that DBS of the VIM thalamic nucleus has on the regularity of essential tremor (Vaillancourt et al., 2003). The decrease in regularity demonstrates that STN DBS and medication actually change the time-dependent structure of tremor in patients with Parkinsons disease rather than merely suppressing the amplitude of the pathological oscillations. It is important to note that neither treatment alone or in combination decreased resting and postural tremor regularity to control levels.

Effects of STN DBS/medication on tremor in PD Medication and STN DBS also reduced the pathological 18 Hz tremorEMG coherence in Parkinsons disease patients. Due to the established relation between tremor EMG coherence and motor unit synchronization (Halliday et al., 1999), the reduction in tremorEMG coherence suggests that either treatment reduced motor unit synchronization. Primate electrophysiology has shown that STN lesions can desynchronize oscillatory neural activity in the STN while simultaneously reducing the amount of limb tremor (Wichmann et al., 1994). Similarly, dopamine and dopamine receptor agonists have been shown to reduce the 38 Hz oscillatory activity (Levy et al., 2001), and the coherence between these oscillatory neurons in the STN (Brown et al., 2001). This suggests that when previously synchronized basal ganglia oscillations become desynchronized, motor units are allowed to re more independently rather than in synchrony. Since STN DBS and medication reduced tremorEMG coherence, we postulate that motor units red more independently, which also reduced tremor regularity and tremor amplitude. An important nding from this study was that STN DBS and medication altered the EMG frequency of resting and postural tremor. In addition, STN DBS operated with sufcient efcacy such that resting and postural tremor EMG frequency in Parkinsons disease patients was similar to that in control subjects. While previous studies have found that medication did not affect the frequency of tremor in Parkinsons disease (Koller et al., 1989; Henderson et al., 1994), we found that medication did alter the frequency of tremor. These earlier studies determined the frequency of tremor by analysing the acceleration signal, which includes contributions from both the mechanical and the central oscillator(s). In contrast, the present study examined the EMG spectrum to determine the frequency of tremor, thereby focusing on the effect of treatment on the pathological central oscillator(s). Differences in disease severity and medication dosage between the Parkinsons disease patients in our study and the Parkinsons disease patients in previous studies may also explain the null effect medication had on tremor frequency in the preceding literature. The mechanisms responsible for the increases in resting and postural tremor EMG frequencies are unclear. One possible mechanism is that the treatments directly changed the frequency of the central oscillator(s). While this explanation is plausible, we prefer the hypothesis that STN DBS and medication suppressed the pathological central oscillator(s), which would allow input from other oscillators (mechanical-reex) to become more dominant in the system. The fact that the EMG frequency of tremor following STN DBS was similar to that in control subjects strengthens the argument that STN DBS unmasked oscillations which are characteristic of physiological tremor. Further study is necessary to determine if STN DBS resets or suppresses the central oscillator(s) frequency, thus allowing mechanical-reex factors to play a larger role in the genesis of tremor.

2141

Physiology of tremor reduction

Circuit models of the basal ganglia suggest that the STN is involved in the genesis of parkinsonian tremor (Bergman et al., 1994; Rodriguez et al., 1998; Liu et al., 2002). In primates the administration of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) causes dopamine depletion, which increases the neuronal ring rate in the STN and the internal segment of the globus pallidus (Bergman et al., 1990; DeLong, 1990). When neuronal activity was examined in the primate STN following administration of MPTP, 4 Hz oscillatory activity was prominent and was correlated with limb tremor (Bergman et al., 1994). Single-unit recordings in Parkinsons disease patients also implicate the STN as a site for oscillatory neuronal activity that may be tremor-related (Krack et al., 1998; Rodriguez et al., 1998; Levy et al., 2000; Liu et al., 2002). STN DBS could be acting via conduction block, inhibition (Benazzouz et al., 1995, 2000), or excitation (Hashimoto et al., 2003; Hershey et al., 2003) to alter the output of STN neurons. Although previous research in primates does not distinguish between resting and postural tremor, behaviourally these two types of oscillations have distinct characteristics. For instance, in Parkinsons disease the frequency of postural tremor has been shown to be greater than the frequency of resting tremor (Lance et al., 1963; Findley et al., 1981; Elble and Koller, 1990). Furthermore, according to the Consensus Statement on Tremor (Deuschl et al., 1998), type II parkinsonian tremor is characterized by the occurrence of both resting and postural tremor, where the frequency of postural tremor is higher than resting tremor. Although a re-emergent resting tremor can occur during postural tasks (Deuschl et al., 1998; Jankovic et al., 1999), the patients in this study did not demonstrate this phenomenon. Evidence for this from our study is that the frequency of postural tremor was higher than resting tremor, further supporting the behavioural distinction between these two types of oscillations. A single- or dual-circuit model could account for the difference in frequency between resting and postural tremor. The output from a single circuit could be differentially modulated depending on a resting or an active limb state. An alternative explanation is the existence of separate circuits oscillating at different frequencies. Nevertheless, the important nding from this study was that medication and STN DBS reduced both resting and postural tremor in Parkinsons disease. However, unlike medication, STN DBS enabled resting and postural tremor amplitude and frequency to approximate healthy physiological levels.

Acknowledgements
We wish to thank Medtronic for donating the Medtronic Console Programmer (Model 7432) for use in this study. We also thank Jean Arzbaecher, Sue Leurgans and the staff at the Section for Movement Disorders in the Department of Neurological Sciences at Rush Presbyterian-St Lukes Medical Center. This research was supported in part by grants from the National Institutes of Health (R01-AR-33189, R01NS-28127, R01-NS-40902 and F32-NS-044727).

2142

M. M. Sturman et al.
Henderson JM, Yiannikas C, Morris JG, Einstein R, Jackson D, Byth K. Postural tremor of Parkinsons disease. Clin Neuropharmacol 1994; 17: 27785. Hershey T, Revilla FJ, Wernle AR, McGee-Minnich L, Antenor JV, Videen TO, et al. Cortical and subcortical blood ow effects of subthalamic nucleus stimulation in PD. Neurology 2003; 61: 81621. Homberg V, Hefter H, Reiners K, Freund HJ. Differential effects of changes in mechanical limb properties on physiological and pathological tremor. J Neurol Neurosurg Psychiatry 1987; 50: 56879. Hughes AJ, Ben-Shlomo Y, Daniel SE, Lees AJ. What features improve the accuracy of clinical diagnosis in Parkinsons disease: a clinicopathologic study. Neurology 1992a; 42: 11426. Hughes AJ, Daniel SE, Kilford L, Lees AJ. Accuracy of clinical diagnosis of idiopathic Parkinsons disease: a clinico-pathological study of 100 cases. J Neurol Neurosurg Psychiatry 1992b; 55: 1814. Jankovic J, Schwartz KS, Ondo W. Re-emergent tremor of Parkinsons disease. J Neurol Neurosurg Psychiatry 1999; 67: 64650. Koller WC. Pharmacologic treatment of parkinsonian tremor. Arch Neurol 1986; 43: 1267. Koller WC, Vetere-Overeld B, Barter R. Tremors in early Parkinsons disease. Clin Neuropharmacol 1989; 12: 2937. Krack P, Benazzouz A, Pollak P, Limousin P, Piallat B, Hoffmann D, et al. Treatment of tremor in Parkinsons disease by subthalamic nucleus stimulation. Mov Disord 1998; 13: 90714. Krack P, Batir A, Van Blercom N, Chabardes S, Fraix V, Ardouin C, et al. Five-year follow-up of bilateral stimulation of the subthalamic nucleus in advanced Parkinsons disease. N Engl J Med 2003; 349: 192534. Lakie M, Walsh EG, Wright GW. Passive mechanical properties of the wrist and physiological tremor. J Neurol Neurosurg Psychiatry 1986; 49: 66976. Lance JW, Schwab RS, Peterson EA. Action tremor and the cogwheel phenomenon in Parkinsons disease. Brain 1963; 86: 95110. Langston JW, Widner H, Goetz CG, Brooks D, Fahn S, Freeman T, et al. Core assessment program for intracerebral transplantations (CAPIT). Mov Disord 1992; 7: 213. Levy R, Hutchison WD, Lozano AM, Dostrovsky JO. High-frequency synchronization of neuronal activity in the subthalamic nucleus of parkinsonian patients with limb tremor. J Neurosci 2000; 20: 776675. Levy R, Dostrovsky JO, Lang AE, Sime E, Hutchison WD, Lozano AM. Effects of apomorphine on subthalamic nucleus and globus pallidus internus neurons in patients with Parkinsons disease. J Neurophysiol 2001; 86: 24960. Lipsitz LA, Goldberger AL. Loss of complexity and aging. Potential applications of fractals and chaos theory to senescence. JAMA 1992; 267: 18069. Liu X, Ford-Dunn HL, Hayward GN, Nandi D, Miall RC, Aziz TZ, et al. The oscillatory activity in the parkinsonian subthalamic nucleus investigated using the macro-electrodes for deep brain stimulation. Clin Neurophysiol 2002; 113: 166772. Marsden CD, Meadows JC, Lange GW, Watson RS. The role of the ballistocardiac impulse in the genesis of physiological tremor. Brain 1969; 92: 64762. Maurer C, Mergner T, Xie J, Faist M, Pollak P, Lucking CH. Effect of chronic bilateral subthalamic nucleus (STN) stimulation on postural control in Parkinsons disease. Brain 2003; 126: 114663. Olanow CW, Nakano K, Nausieda P, Tetrud JA, Manyam B, Last B, et al. An open multicenter trial of Sinemet CR in levodopa-naive Parkinsons disease patients. Clin Neuropharmacol 1991; 14: 23540. OSuilleabhain PE, Frawley W, Giller C, Dewey RB Jr. Tremor response to polarity, voltage, pulsewidth and frequency of thalamic stimulation. Neurology 2003; 60: 78690. Pincus SM. Approximate entropy as a measure of system complexity. Proc Natl Acad Sci USA 1991; 88: 2297301. Pincus SM, Goldberger AL. Physiological time-series analysis: what does regularity quantify? Am J Physiol 1994; 266: H164356. Raethjen J, Pawlas F, Lindemann M, Wenzelburger R, Deuschl G. Determinants of physiologic tremor in a large normal population. Clin Neurophysiol 2000; 111: 182537. Rodriguez MC, Guridi OJ, Alvarez L, Mewes K, Macias R, Vitek J, et al. The subthalamic nucleus and tremor in Parkinsons disease. Mov Disord 1998; 13 Suppl 3: 1118.

References Bastian AJ, Kelly VE, Revilla FJ, Perlmutter JS, Mink JW. Different effects of unilateral versus bilateral subthalamic nucleus stimulation on walking and reaching in Parkinsons disease. Mov Disord 2003; 18: 10007. Benabid AL. Deep brain stimulation for Parkinsons disease. Curr Opin Neurobiol 2003; 13: 696706. Benazzouz A, Piallat B, Pollak P, Benabid A. Responses of substantia nigra pars reticulata and globus pallidus complex to high frequency stimulation of the subthalamic nucleus in rats: electrophysiological data. Neurosci Lett 1995; 189: 7780. Benazzouz A, Gao DM, Ni Z, Piallat B, Bouali-Benazzouz R, Benabid AL. Effect of high frequency stimulation of the subthalamic nucleus on the neuronal activities of the substantia nigra pars reticulata and ventrolateral nucleus of the thalamus in the rat. Neuroscience 2000; 99: 28995. Bergman H, Wichmann T, DeLong MR. Reversal of experimental parkinsonism by lesions of the subthalamic nucleus. Science 1990; 249: 14368. Bergman H, Wichmann T, Karmon B, DeLong MR. The primate subthalamic nucleus. II. Neuronal activity in the MPTP model of parkinsonism. J Neurophysiol 1994; 72: 50720. Brown P, Corcos DM, Rothwell JC. Does parkinsonian action tremor contribute to muscle weakness in Parkinsons disease? Brain 1997; 120: 4018. Brown P, Oliviero A, Mazzone P, Insola A, Tonali P, Di Lazzaro V. Dopamine dependency of oscillations between subthalamic nucleus and pallidum in Parkinsons disease. J Neurosci 2001; 21: 10338. Burne JA, Lippold OC, Pryor M. Proprioceptors and normal tremor. J Physiol 1984; 348: 55972. Cohn MC, Hudgins PA, Sheppard SK, Starr PA, Bakay RA. Pre- and postoperative MR evaluation of stereotactic pallidotomy. AJNR Am J Neuroradiol 1998; 19: 107580. Cotzias GC, Papavasiliou PS, Gellene R. Modication of parkinsonism chronic treatment with L-dopa. N Engl J Med 1969; 280: 33745. Deep-Brain Stimulation for Parkinsons Disease Study Group. Deepbrain stimulation of the subthalamic nucleus or the pars interna of the globus pallidus in Parkinsons disease. N Engl J Med 2001; 345: 95663. DeLong MR. Primate models of movement disorders of basal ganglia origin. Trends Neurosci 1990; 13: 2815. Deuschl G, Bain P, Brin M. Consensus statement of the Movement Disorder Society on Tremor. Ad Hoc Scientic Committee. Mov Disord 1998; 13 Suppl 3: 223. Elble RJ, Koller WC. Tremor. Baltimore (MD): Johns Hopkins University Press; 1990. Elble RJ, Randall JE. Motor-unit activity responsible for 8- to 12-Hz component of human physiological nger tremor. J Neurophysiol 1976; 39: 37083. Faist M, Xie J, Kurz D, Berger W, Maurer C, Pollak P, et al. Effect of bilateral subthalamic nucleus stimulation on gait in Parkinsons disease. Brain 2001; 124: 1590600. Farmer SF, Bremner FD, Halliday DM, Rosenberg JR, Stephens JA. The frequency content of common synaptic inputs to motoneurones studied during voluntary isometric contraction in man. J Physiol 1993; 470: 12755. Findley LJ, Gresty MA, Halmagyi GM. Tremor, the cogwheel phenomenon and clonus in Parkinsons disease. J Neurol Neurosurg Psychiatry 1981; 44: 53446. Friedman JH, Koller WC, Lannon MC, Busenbark K, Swanson-Hyland E, Smith D. Benztropine versus clozapine for the treatment of tremor in Parkinsons disease. Neurology 1997; 48: 107781. Halliday AM, Redfearn JW. An analysis of the frequencies of nger tremor in healthy subjects. J Physiol 1956; 134: 60011. Halliday DM, Conway BA, Farmer SF, Rosenberg JR. Load-independent contributions from motor-unit synchronization to human physiological tremor. J Neurophysiol 1999; 82: 66475. Hashimoto T, Elder CM, Okun MS, Patrick SK, Vitek JL. Stimulation of the subthalamic nucleus changes the ring pattern of pallidal neurons. J Neurosci 2003; 23: 191623.

Effects of STN DBS/medication on tremor in PD

Rosenberg JR, Amjad AM, Breeze P, Brillinger DR, Halliday DM. The Fourier approach to the identication of functional coupling between neuronal spike trains. Prog Biophys Mol Biol 1989; 53: 131. Starr PA, Vitek JL, Bakay RA. Deep brain stimulation for movement disorders. Neurosurg Clin N Am 1998; 9: 381402. Starr PA, Subramanian T, Bakay RA, Wichmann T. Electrophysiological localization of the substantia nigra in the parkinsonian nonhuman primate. J Neurosurg 2000; 93: 70410. Stiles RN. Frequency and displacement amplitude relations for normal hand tremor. J Appl Physiol 1976; 40: 4454. Stiles RN, Pozos RS. A mechanical-reex oscillator hypothesis for parkinsonian hand tremor. J Appl Physiol 1976; 40: 9908. Vaillancourt DE, Newell KM. The dynamics of resting and postural tremor in Parkinsons disease. Clin Neurophysiol 2000; 111: 204656. Vaillancourt DE, Newell KM. Changing complexity in human behavior and physiology through aging and disease. Neurobiol Aging 2002; 23: 111.

2143

Vaillancourt DE, Sturman MM, Verhagen Metman L, Bakay RA, Corcos DM. Deep brain stimulation of the VIM thalamic nucleus modies several features of essential tremor. Neurology 2003; 61: 91925. Vaillancourt DE, Prodoehl J, Verhagen Metman L, Bakay RA, Corcos DM. Effects of deep brain stimulation and medication on bradykinesia and muscle activation in Parkinsons disease. Brain 2004; 127: 491504. Vitek JL, Bakay RA, Hashimoto T, Kaneoke Y, Mewes K, Zhang JY, et al. Microelectrode-guided pallidotomy: technical approach and its application in medically intractable Parkinsons disease. J Neurosurg 1998; 88: 102743. Wasielewski PG, Burns JM, Koller WC. Pharmacologic treatment of tremor. Mov Disord 1998; 13 Suppl 3: 90100. Wichmann T, Bergman H, DeLong MR. The primate subthalamic nucleus. III. Changes in motor behavior and neuronal activity in the internal pallidum induced by subthalamic inactivation in the MPTP model of parkinsonism. J Neurophysiol 1994; 72: 52130. Yahr MD, Duvoisin RC, Schear MJ, Barrett RE, Hoehn MM. Treatment of parkinsonism with levodopa. Arch Neurol 1969; 21: 34354.