The impact of ventrolateral thalamotomy on high-frequency components of tremor

Clinical Neurophysiology 116 (2005) 1391–1399 www.elsevier.com/locate/clinph

The impact of ventrolateral thalamotomy on high-frequency components of tremor Christian Duvala,*, Antonio P. Strafellab, Abbas F. Sadikotb b

a Faculty of Applied Health Science, Brock University, 500 Glenridge Avenue, St Catharines, Ont., Canada H2S 3A1 Department of Neurology and Neurosurgery, Montreal Neurological Institute, McGill University, Montreal, Que., Canada

Accepted 22 January 2005 Available online 28 March 2005

Abstract Objective: The present study assessed the impact of ventrolateral (VL) thalamotomy on the high-frequency components of tremor in patients with Parkinson’s disease (PD). Methods: Tremor was recorded prior to, and 7 days post-surgery using a laser displacement sensor. In addition, tremor was recorded in 10 age-matched patients with PD showing low amplitude tremor (named PD controls) and in 10 age-matched control subjects. Tremor recordings in patients were performed after a 12 h withdrawal from anti-Parkinsonian drugs. Tremor characteristics such as amplitude, median power frequency (MPF) and power dispersion (a measure of concentration of power in the frequency domain) were assessed for all groups (i.e. controls, PD controls, pre-surgery and post-surgery). Results: All tremor characteristics were similar between controls and PD controls. Tremor amplitude was significantly reduced post-surgery, to become statistically similar to that of controls and PD controls. However, MPF and power dispersion remained lower post-surgery, indicating that although there was normalization of tremor amplitude, tremor showed systematically slower oscillations after the surgical procedure. In order to eliminate amplitude as a possible confounding factor, epochs of post-surgical tremor (5 s in duration) were paired for equal amplitude with 5 s tremor epochs from matched controls. Results show once again that MPF and power dispersion were lower postsurgery compared to controls. In addition, when amplitude of power was compared within specific frequency bands (0–3.5, 3.5–7.5, 7.5– 12.5, 12.5–16.5, 16.5–30 and 30–45), power regained normal values at frequencies below 7.5 Hz. Power within higher frequency bands was systematically lower, indicating that the surgical procedure had an impact on high-frequency components of tremor. Conclusions: Results from the present study showed that VL thalamotomy reduced tremor amplitude by selectively targeting centrally driven components of PD tremor. The high-frequency component of physiological tremor failed to emerge after amplitude normalization. Significance: The thalamus should then be considered as an important component of the generation and/or propagation of high-frequency components of physiological tremor. q 2005 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved. Keywords: Tremor; Power spectrum; High frequency; Laser displacement sensor

1. Introduction Physiological tremor is defined as an oscillatory, roughly sinusoidal involuntary movement of a body part (Elble and Koller, 1990). It is now widely accepted that tremor oscillations are primarily the consequence of the interaction between mechanical characteristics of the limb in which * Corresponding author. Tel.: C1 905 688 5550x4496; fax: C1 905 688 8364. E-mail address: [email protected] (C. Duval).

tremor is measured (Marsden et al., 1969; Robson, 1959; Stiles and Randall, 1967; Van Buskirk et al., 1966; Yap and Boshes, 1967) and background motor-unit activity (Hagbarth and Young, 1979; Vaillancourt and Newell, 2000; Young and Hagbarth, 1980). Tremor oscillations associated with these ‘mechanical-reflex’ components are widely distributed throughout the power spectrum, up to 30 Hz for the finger (Stiles and Randall, 1967). Also to be considered are the centrally generated components of tremor, such as the one located within the 8–12 Hz band. These oscillations are believed to originate from a central

1388-2457/$30.00 q 2005 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.clinph.2005.01.012

1392

C. Duval et al. / Clinical Neurophysiology 116 (2005) 1391–1399

oscillator located within the olivocerebellar (Lamarre et al., 1975; Llinas, 1984; Poirier et al., 1966) and/or cerebellothalamo-cortical pathways (Duval et al., 2000; Lamarre, 1995). Centrally driven components might also exist at higher frequencies; around 25 Hz (Conway et al., 1995; Halliday et al., 1999; McAuley et al., 1997; Vaillancourt and Newell, 2000). However, others imply that these highfrequency components are produced by the muscle–spinal reflex loop, while the low frequency component of about 10 Hz originates from the central nervous system or from supraspinal reflex loop (Sakamoto et al., 1998). Whether these high-frequency components are indeed under direct central influence, and/or share similar neural pathways with the 8–12 Hz component is unknown. Parkinson’s disease (PD) tremor is usually large in amplitude and presents a well-defined frequency peak at around 4–5 Hz. Pathologically induced changes in firing patterns of neurons located within the basal ganglia (Hutchison et al., 1997; Pare et al., 1990), in combination with those in the thalamus (Hua et al., 1998; Lenz et al., 1988) and/or cortex (Volkmann et al., 1996), may be responsible for the creation of a central oscillator or oscillators, ultimately resulting in relatively high amplitude PD tremor. Here, centrally generated oscillations are large enough to overwhelm other mechanical-reflex components, and probably replace or overcome otherwise existing centrally driven components of physiological tremor. The impact of ventrolateral (VL) thalamotomy on tremor oscillations was recently examined by comparing the power spectrum of post-surgical tremor with those of agematched control subjects, using epochs of tremor paired for equal amplitude (Duval et al., 2000). Results showed that the surgical procedure selectively eliminated the narrowly distributed Parkinsonian power peak, suggesting that the VL lesion targeted oscillations related to the centrally driven components of PD tremor. Furthermore, thalamotomy seemed to prevent the resurgence of the 8–12 Hz component

of PT after amplitude normalization, suggesting that this component shared similar neural pathways with PD tremor at the level of the VL thalamus. If the VL thalamus is implicated in the generation and/or propagation of the highfrequency components of tremor, such as the ones believed to exit above 20 Hz, the surgical procedure may also have an impact on the power related to these frequencies. Accordingly, the goal of the present study was to assess the impact of VL thalamotomy on high-frequency components of tremor. Tremor from 10 patients who underwent VL thalamotomy was retrospectively examined using a proven methodology to determine the impact of the surgical procedure on high-frequency components of tremor (Duval et al., 2000). This time, the mean lesion location was confirmed using a probabilistic map methodology (Atkinson et al., 2002). Post-surgical tremor characteristics such as amplitude, median power frequency (MPF) and power dispersion were compared with those seen in age-matched control subjects. In addition, the amount of power found in post-surgical tremor, within specific frequency bands, was compared with those from matched controls using epochs of tremor paired for equal amplitude. Tremor from patients with PD showing low amplitude tremor was also examined; they served as a control group within the patient population.

2. Methods 2.1. Patients Postural index finger tremor recordings from 10 patients with PD who underwent unilateral VL thalamotomy for tremor were analyzed (age range: 43–76, mean age: 63.7G 10.3SD). Individual age, gender, handedness and UPDRS tremor scores are shown in Table 1. Tremor was recorded using a laser displacement sensor (LTS 90/45; LMI Technology, The Netherlands). The measuring range of

Table 1 Description of patients Patient #

1 2 3 4 5 6 7 8 9 10 Mean SD Range

Age

63 70 66 70 43 70 60 68 70 72 65.2 8.6 43–72

Sex

M M M M M M F M M F

Laterality

R R R R R R R R R R

Lesion side

L L L R R L L L L L

UPDRS, unified Parkinson’s disease rating scale; SD, standard deviation.

Tremor (UPDRS) rest condition

Tremor (UPDRS) posture condition

Pre

Post

Pre

Post

4 1.5 4 2 3 3 2 3 3 3 2.85 0.82 1.5–4

1 0 0.5 0 0 0 0 0 0 0 0.15 0.34 0–1

4 3 4 3 3 3.5 2 3 3 4 3.25 0.63 2–4

0 0 0 0 0 0 0 0 0 0 0 0 0

C. Duval et al. / Clinical Neurophysiology 116 (2005) 1391–1399

this device was 45 mm, with a frequency response up to 20 kHz and a resolution better than 0.002 mm. This laser system is equally sensitive to low- and high-frequency components, hence allowing accurate quantification of tremor range in displacement and accurate velocity or acceleration measurements following signal differentiation. Tremor recording occurred 2 days prior to, and 7 days postsurgery. Surgical patients refrained from taking their antiParkinsonian drugs at least 12 h prior to testing pre-surgery, and 12 h prior to post-surgical testing, as their tremor recording coincided with clinical evaluation. In addition, tremor from control subjects matched for age, gender and handedness were examined (age range: 43–75, mean age: 63.8G10.22SD). Finally, in order to ensure that the surgical procedure was responsible for eventual changes of power spectral characteristics, tremor recordings obtained from 10 patients diagnosed with PD that showed low amplitude tremor were also analyzed (age range: 49–77, mean age: 60.0G8.8SD). They also refrained from taking their antiParkinsonian medication 12 h prior to testing. In the present paper, this group is labeled as ‘PD controls’. All procedures were performed after the study was explained, and informed written consent was obtained from all participants. 2.2. Surgical procedure and lesion mapping In the present paper, the terminology for thalamic nuclei of Hirai and Jones (1989) is used. The surgical procedure, target localization techniques, and lesion mapping have been previously described in detail elsewhere (Atkinson et al., 2002; Bertrand, 1966; Bertrand et al., 1969; St-Jean et al., 1998). The target was localized using stereotactic MR imaging. A deformable volumetric atlas of the thalamus and basal ganglia (St-Jean et al., 1998) was integrated with the patient’s stereotactic MR scan. An automated non-linear warping algorithm was used to produce the transformation of a model atlas-integrated MR volume to each patient’s MR scan to further confirm target localization within the VL thalamus. Stereotactic ventriculography was also used as an intraoperative aid. Also, the localization of the target was established in the awake patient using a curved retractable monopolar stimulation electrode to localize the neighboring sensory thalamus and motor fibers within the internal capsule (Atkinson et al., 2002; Bertrand, 1966; Bertrand et al., 1969; St-Jean et al., 1998). Tailored lesions were performed in the VL nucleus using a modified loop-shaped leukotome with a millimetric scale (Obrador, 1957; St-Jean et al., 1998). Lesion volume and location were confirmed post-operatively using MR images obtained within 24 h of surgery in the 10 patients. First, individual lesions were manually segmented on a pixel-by-pixel basis, using the native MR volume of individual patients, and hence producing individual lesion volumes. Average lesion location was then determined by taking each patient’s MR images and warping into a common MRI reference space which also contained an integrated computerized version of

1393

the atlas of Schalterbrand and Wahren, allowing for tridimensional (3D) visualization of thalamic target and stimulation probe (Atkinson et al., 2002; St-Jean et al., 1998). The transformation resulted in a probabilistic map that was used to assess the average lesion location. 2.3. Tremor recording For each recording session, the forearm of subjects was allowed to rest comfortably on a foam-padded support with the elbow slightly flexed. A palm support was adjusted so as to allow isolation of the index finger. Recording of postural index finger tremor was recorded contra-lateral to the lesion side. During recording, patients were instructed to maintain their index finger in a roughly horizontal position. When tremor was present, patients simply adjusted their finger position so that the midpoint of upward and downward movements was approximately located at the center of the laser detection range. A small piece of white cardboard paper was placed on the fingernail in order to provide an accentuated reflective surface for the laser beam. Patients were asked to keep their eyes closed during recording. A minimum of three 60-s recordings were done for each condition, with a sample rate of 2 kHz. Amplitudes of power for frequencies below 1 Hz and above 45 Hz were set to zero (high- and low-passed with FFT-inverse FFT method; tapered from 0.5 to 1 and from 44.5 to 45 with a ramp of width 0.5 Hz). Voltage signal from the laser was then transformed into millimeters. Each tremor recording of a particular subject was then separated into 5 s epochs. Next, 3 tremor characteristics were computed on each 5 s epoch and subsequently averaged together: (a) tremor amplitude (root mean square times two), provided a measure of tremor range. (b) MPF provided the frequency at which half of the power was located on either side. (c) Power dispersion, which represents the width of a frequency band containing 68% of the power centered at the MPF (Beuter and Edwards, 1999), provided a measure of concentration of tremor power. Typical Parkinsonian tremor usually possesses a sharp peak in the power spectrum, which is revealed by a low power dispersion value. Tremor analysis was performed in two distinct steps. First the aforementioned analyses were done on all tremor traces recorded in each group. Then, the same analysis was performed on 5 s tremor epochs that were paired for equal amplitude with 5 s epochs of tremor from an age-matched subject. The goal was to compare tremor characteristics where amplitude was no longer a confounding factor. This approach provides a much better assessment of the consequence of the surgical procedure on each tremor components (Duval et al., 2000). To do so, an automated algorithm found 5 s epochs of equal amplitude in matched controls, for each 5 s epochs of post-surgical tremor. One hundred and sixteen such 5 s epochs could be paired for equal amplitude. From the power spectrum of these

C. Duval et al. / Clinical Neurophysiology 116 (2005) 1391–1399

tremor epochs, the amplitude of power was calculated within predetermined frequency bands: [1] 0–3.5 Hz, probably related to slow irregular, aperiodic oscillations; [2] 3.6–7.5 Hz, in which major PD pathological oscillations are usually found; [3] 7.6–12.5 Hz, in which the centrally generated component of physiological tremor is believed to be located; [4] 12.6–16.5 where higher frequency components of physiological tremor may be found; [5] 16.5–30 Hz, where the resonance frequency and possibly some centrally driven components may be found, and [6] 30–45 Hz where other high-frequency centrally components may be found. Amplitude (log(mm))

100 B 10 1 0.1 0.01

C

30 D

E

20 10 0 30

F

G

20 10 Post-surgery (matched)

Controls (matched)

Post-surgery

0 Pre-surgery

Fig. 1 shows a mid-sagittal view of the probabilistic map of lesions from the 10 surgical patients. In addition, Fig. 1 shows the results for amplitude, MPF and power distribution from all groups. Comparisons of amplitude, MPF and power dispersion between paired 5 s epochs are also shown. The map confirms that lesions were centered upon the ventral lateral (VLp) nucleus of the thalamus, the optimal target for tremor alleviation. The lesions encroached upon the most rostral shell region of the ventral posterior lateral (VPL) nucleus, but spared the cutaneous sensory thalamus. The lesion also partially extended into the anterior part of the VL (VLa), the target of pallido-thalamic pathways. ANOVA results indicate that there was a group effect for amplitude (FZ15.84, P!0.0000), MPF (FZ35.74, PZ0.0000) and power dispersion (FZ37.74, PZ0.0000). More specifically, post hoc analysis showed that tremor amplitude was similar between controls and PD controls (PO0.05), pre-surgery tremor was higher when compared to other groups (P!0.05) but was significantly reduced postsurgery (P!0.05). Although the amplitude of post-surgery tremor did not regain controls’ values, there were no statistical differences between post-surgery tremor amplitude and controls or PD controls (PO0.05). There were no significant differences between controls and PD controls for MPF (PO0.05). Power dispersion increased slightly after the surgical procedure, but not significantly (PO0.05); it remained much lower than controls and PD controls (P!0.05). Once again, there were no significant

0

Controls

3. Results

PUL VPLa

MPF (Hz)

In the present study, each condition (i.e. controls, PD controls, pre-surgery and post-surgery) was considered as an independent group. Accordingly, a one-way ANOVA was used to determine statistical significance, and when necessary, post hoc analysis was performed using Tukey’s honestly significant differences (HDS) method. When needed, T tests were used to compare results between paired 5 s epochs described above.

IC 1

Power dispersion (Hz)

2.4. Statistical methods

VLp

A

PD controls

1394

Fig. 1. (A) Mid-sagittal view of the probabilistic map, obtained from 10 patients, superposed over a common brain. The lesion is centered in the posterior segment of the ventral lateral (VLp) nucleus and extends into the anterior VL (VLa). (B) Mean amplitude from each group; (C) mean amplitude of the 116 5 s epochs of post-surgical tremor and paired 5 s epochs from controls; (D) mean median power frequency from each group; (E) mean median power frequency of paired 5 s epochs; (F) mean power dispersion from each group; (G) mean power dispersion of paired 5 s epochs. Note that the median power frequency and power dispersion did not return to values seen in controls (D and F), even when amplitudes were identical in the case of paired 5 s epochs (E and G). IC, internal capsule; PUL, pulvinar. Statistical differences are indicated in the text.

difference between controls and PD controls for power dispersion (PO0.05). As for matched 5 s epochs, the amplitude, MPF and power dispersion of 5 s epochs of post-surgical tremor paired for amplitude with matched controls. Amplitude was identical, a consequence of the matching procedure (TZ0.01, PZ0.99). Post-surgical tremor’s MPF remained lower (TZ16.55, PZ0.000), as did the power dispersion (TZ9.00, PZ0.000), confirming that even when amplitude was not a confounding factor, post-surgical tremor did not regain normal values in the frequency domain.

C. Duval et al. / Clinical Neurophysiology 116 (2005) 1391–1399

1395

0.08

0-3.5

0.06 0.04 0.02 0.00 0.08

3.5-7.5

0.06 0.04 0.02

12.5-16.5

Power

7.5-12.5

0.00 0.05 0.04 0.03 0.02 0.01 0.00 0.05 0.04 0.03 0.02 0.01 0.00

16.5-30

0.020 0.015 0.010 0.005 0.000 0.020

30-45

0.015 0.010 0.005 0.000

0.1

0.2

0.3

0.4

0.5

0.6

0.7

Amplitude (mm) Fig. 2. From top to bottom: in the ordinate axis of each plot, the sum of power found in post-surgical tremor (white dots) and matched controls for each frequency bands. Frequency bands are indicated on the far left. Abscises represents the tremor amplitude placed in increasing order. Solid lines represent the second order regression. Note that within the 0–3.5, 3.5–7.5 and 12.5–16.5 Hz bands, the power increased concomitantly in both groups as higher 5 s epochs amplitudes were compared. For the 7.5–12.5 Hz band, the power increased mostly for controls. As for the 16.5–30 and 30–45 Hz bands, there was also clearly more power for the controls.

Fig. 2 shows the amplitude of tremor power within specific frequency bands (from the tremor velocity power spectrum) for each of the paired 5 s epochs. The power seemed to increase proportionally with tremor amplitude in the 0–3.5 and 3.5–7.5 Hz bands. However, this proportional increase was not as apparent in the 7.5–12.5 Hz band where the power of matched controls was consistently higher than post-surgical patients. In the 12.6–16.5 Hz band, the power seemed to increase concomitantly for both groups, although once again there was more power for matched controls. This is even more apparent for the 16.5–30 Hz band. As for the 30–45 Hz, matched controls showed higher power, without proportional increase for either group, suggesting that

the level of power was independent of tremor amplitude. Except for the 0–3.5 and 3.5–7.5 Hz bands, power related to high frequencies was consistently higher for matched controls. Fig. 3 shows the tremor recording of a 43 year old man during intra-cerebral stimulation at the optimum tremor relief site, which is the border of the VLp and VPLa. Prior to stimulation, tremor was of relatively large in amplitude and presented a well-defined peak in the power spectrum that is typical of Parkinsonian tremor. When stimulation was applied, tremor was markedly reduced. The tremor power became clearly concentrated within lower frequencies, without resurgence of high-frequency components. This is

1396

C. Duval et al. / Clinical Neurophysiology 116 (2005) 1391–1399

OFF

4

A

OFF

ON

0

B

Amplitude (mm)

–4 0

5

10

15

20

25

30

35

40

45

50

55

0.4 0.2 0.0 –0.2 –0.4 22

24

26

28

30

36

38

40

0.4 0.2 0.0 –0.2 –0.4

C

32

34 VPLa

E Power (mm/s)

D

VLp

Stimulator Stimulator probe tip

Time (s) 700 600 500 400 300 200 100 0

F 1.2 0.9 0.6 0.3

0

15

30

0.0 45 0

15

30

45

Frequencies (Hz)

Fig. 3. (A) Intraoperative tremor recording in a 42 year old man. (B) Selected epoch of tremor during which electrical stimulation was applied within the VL thalamus (1 V, 180 Hz, 2 ms pulse duration), more specifically within the VLp and near the VPLa. Note the smoother tremor compared to a sample of tremor with identical amplitude from a matched control (in C). (D) Mid-sagittal 3D representation of the actual VLp (green) and VPLa medial (pink) and lateral (blue) thalamic target of that patient. The stimulating electrode is represented in yellow with the tip extending out to the border of the VLp and VPLa, the optimal site for tremor alleviation. (E) The power spectrum of the patient’s tremor before stimulation was applied. (F) In black, the power spectrum of the patient’s tremor while stimulation is applied, in dark gray, the power distribution of post-surgical tremor from that patient. Finally, the dashed line represents the power spectrum from the matched control. Arrows indicates the peaks seen in that control. It is evident that stimulation reduced tremor amplitude, without resurgence of higher frequency components, such as seen post-surgery.

evident when the power spectrum of tremor under stimulation was compared with the power spectrum of tremor (paired for amplitude) from one healthy, matched control subject.

4. Discussion Results from the present study show that tremor amplitude was normalized after VL thalamotomy, from an average of 10 mm pre-surgery to less than 0.5 mm after the surgical procedure, hence attesting to the efficacy of the surgical procedure. However, several findings suggest that tremor was not normalized in the frequency domain. First, MPF and power dispersion remained lower post-surgery. After this drastic tremor amplitude reduction, one would expect to find a significant increase of MPF and power dispersion to levels seen in controls and PD controls. The fact that values of these tremor characteristics remained lower than controls and PD controls implies that the surgical procedure was responsible for the low MPF and power dispersion seen post-surgery. Supporting this is the fact

that when the MPF and power dispersion of 5 s tremor epochs paired for amplitude were compared, these tremor characteristics were once again lower in post-surgical tremor. This particular finding indicates that even when amplitude was removed as a possible confounding factor, post-surgery tremor characteristics in the frequency domain did not regain normal values. Second, when the power of the aforementioned paired 5 s tremor epochs was compared within specific frequency bands, results showed clearly that the power was similar and increased proportionally with tremor amplitude within the 0–3.5 and 3.5–7.5 Hz bands. This suggests that irregular, aperiodic oscillations and/or mechanical-reflex components in these frequency bands re-emerged after amplitude normalization. The power within the 7.5– 12.6 Hz band and above were systematically lower in post-surgical tremor, suggesting that the surgery had an impact on oscillations related to these frequency components. Finally, electrical stimulation within the VL thalamus in a patient seemed to reproduce these results. Taken together, these results show clearly that post-surgical tremor oscillations were systematically slower those seen in controls and PD controls.

C. Duval et al. / Clinical Neurophysiology 116 (2005) 1391–1399

4.1. The thalamus and PD tremor A consensus now seems to exist about the fact that the origin of PD tremor resides in supraspinal structures, and is primarily the consequence of the loss of dopaminergic neurons in the substantia nigra pars compacta (Kish et al., 1988). Peripheral pathways may be involved in the modulation of amplitude and/or phase of PD tremor. While change in firing patterns of output neurons from the basal ganglia is believed to be mostly responsible for the generation of PD tremor (Pare et al., 1990), changes of firing patterns of neurons within the thalamus probably strengthen tremor oscillations through neural synchronization (Bertrand and Jasper, 1965; Bertrand et al., 1969), and allow for transmission of oscillatory neural patterns to cortical areas projecting to the affected limb (Albe-Fessard et al., 1966). Therefore, the thalamus seems to be directly involved in the consolidation and/or propagation of oscillatory neural impulses to the cortex. The consistent success of VL thalamotomy for various kinds of pathological tremors confirms the predominant role of the thalamus in tremor (Ohye, 2000; Ohye et al., 1982). Cortical involvement in PD tremor has also been suspected for a long time. Alberts (1972) showed that stimulation of the rolandic fissure generated hand movement resembling PD tremor. Duffau et al. (1996) and Parker et al. (1992) supported this claim by showing that brain structures involved in tremor were similar to the ones used for fast repetitive movements. Volkmann et al. (1996) demonstrated that the motor cortex showed neural activity at the tremor frequency in patients with PD. Accordingly, these observations suggest that changes of firing patterns within the basal ganglia and thalamus would affect motor cortex output to produce PD tremor. Since the VLp nucleus of the thalamus shows neural activity that is highly synchronized with PD tremor (Bertrand and Jasper, 1965; Bertrand et al., 1969), it is reasonable to assume that thalamotomy reduced PD tremor by targeting an important central component of neural networks involved in the generation and/or propagation of these tremor oscillations. Recently, Sturman et al. (2004) have shown that deep brain stimulation (DBS) in the subthalamic nucleus (STN) allows for the resurgence of high-frequency components of physiological tremor. Because such resurgence is not seen after VL thalamotomy, it is reasonable to suggest that DBS in the STN reduces PD tremor amplitude while allowing the centrally driven components of physiological tremor to flow through the thalamo-cortical pathways. According to this hypothesis, a patient who is treated with DBS in the STN, but also has had a previous lesion within the VL thalamus for tremor, would not present with these high-frequency components of physiological tremor. This remains to be investigated. 4.2. The thalamus and physiological tremor As mentioned in the introduction, tremor is made of several types of oscillations. Some are mechanically

1397

induced (Marsden et al., 1969; Robson, 1959; Stiles and Randall, 1967; Van Buskirk et al., 1966; Yap and Boshes, 1967), others under the influence of motor-unit activity, i.e. reflex activity and/or central oscillators (Conway et al., 1995; Duval et al., 2000; Hagbarth and Young, 1979; Halliday et al., 1999; Lamarre, 1995; Lamarre et al., 1975; Llinas, 1984; McAuley et al., 1997; Poirier et al., 1966; Vaillancourt and Newell, 2000; Young and Hagbarth, 1980). It is reasonable to assume that VL thalamotomy would not modify the irregular, aperiodic oscillations or mechanical-reflex components because they are not under direct central influence. In the present study, where the amplitude of tremor was normalized, power related to irregular, aperiodic oscillations or mechanical-reflex components re-emerged between 0 and 7.5 Hz. Based on previous observations (Duval et al., 2000), it was also assumed that any tremor oscillations in which the thalamocortical pathway may be implicated would be affected by the surgical procedure. Because there was a drastic reduction of tremor power within the 8–12, 16–30 and 30–45 Hz bands in the present study, the present findings seemed to suggest that oscillations within these frequency bands are to some extent under central influence. In physiological tremor, McAuley et al. (1997) have identified distinct peaks in tremor at 10, 20 and 40 Hz, and suggested based on coherence analysis between tremor acceleration and EMG activity that these oscillations might be of central origin. They supported this claim by showing that mechanical loading and partial anesthesia did not affect these frequency peaks. Vaillancourt and Newell (2000) agreed with these results by showing that frequencies remained stable within the 8–12, 20–25 and 40 Hz components during increased finger loading, consistent with central involvement. The present study clearly showed that VL thalamotomy affected all these frequency components, suggesting that the thalamus may be involved in the generation and/or propagation of these tremor components. Lamarre (1995) proposed that in primate species, where cerebello-thalamo-cortical pathways assume predominance, cerebello-thalamic pathways could be critical to the central component of physiological tremor. The thalamus, especially the VLp, is believed to be an integral part of a cerebello-thalamo-cortical pathway in which neural synchronization in frequencies between 8 and 27 Hz could be found (Marsden et al., 2000). Other evidence suggests that there is coherence between motor cortical activity and electromyography activity of muscles during sustained muscle contraction around 20–25 Hz (Conway et al., 1995; Gross et al., 2000; McAuley et al., 1997), and that pyramidal tract neurons would transmit these neural rhythms to the spinal motor neurons (Baker et al., 1997; Gross et al., 2000). The rhythmic interaction between cortex and muscle, namely the cortico-muscular coherence, seem to be higher during sustained, static phases of a motor task (see Salenius and Hari 2003 for review), a condition similar to that of the task performed in the present study. A partial resurgence of this

1398

C. Duval et al. / Clinical Neurophysiology 116 (2005) 1391–1399

cortico-muscular coherence has been observed in patients with PD under Levodopa treatment (Salenius et al., 2002) and subthalamic stimulation (Marsden et al., 2001). In these cases, treatments targeting brain structures upstream to the thalamus, i.e. in the basal ganglia, may allow for the resurgence of the central component of physiological tremor. This is not surprising since normal tremor amplitude and power distribution characteristics is possible in patients with PD, as demonstrated in the present study (PD control group). Here, where the VL thalamus was targeted, central components of physiological tremor failed to emerge. This suggests that the surgical procedure may have acted as a ‘partial low-pass filter’, diminishing considerably the central output of these frequency components while preserving power related to irregular, aperiodic oscillations and mechanical-reflex components. Indeed, VL thalamotomy is known to have an impact on cortical activity by reducing cortical activation ipsilateral to the lesion (Boecker et al., 1997; Obrist et al., 1975). Accordingly, it is reasonable to suggest that the thalamo-cortical pathways are involved in cortico-muscular coherence and that centrally driven tremor oscillations may be one component of this neural network. Of course, this hypothesis should be tested further by examining cortex–muscle interaction after thalamotomy to determine if the cortico-muscular coherence is affected by the surgical procedure. Also, long-term impact of the surgical procedure on these frequency characteristics should be studied since they may reappear following reorganization of neural pathways. Such reorganization has been observed after thalamic stroke (Ohye et al., 1985). This is especially relevant if cortico-muscular coherence is found to have a significant role in motor the act. To date, the physiological importance of cortico-muscular coherence remains to be determined. In conclusion, results from the present study showed that VL thalamotomy reduced tremor amplitude by selectively targeting centrally driven components of PD tremor. Highfrequency component of physiological tremor failed to emerge after amplitude normalization. The thalamus should then be considered as an important component of the generation and/or propagation of high-frequency components of physiological tremor. Acknowledgements This research project was funded by an Operating Grant from the Canadian Institute for Health Research (AFS, APS), Canada Innovation Fund (CD), Ontario Innovation Trust (CD) and Parkinson Society Canada (CD).

References Albe-Fessard D, Guiot G, Lamarre Y, Arfel G. Activation of thalamocortical projections related to tremogenic processes. In: Purpura DP, Yahr MD, editors. The Thalamus. New York: Columbia University Press; 1966. p. 237–49.

Alberts WW. A simple view of Parkinsonian tremor. Electrical stimulation of cortex adjacent to the rolandic fissure in awake man. Brain Res 1972; 44:357–69. Atkinson JD, Collins L, Bertrand G, Peters TM, Pike BG, Sadikot AF. Optimal location of thalamotomy lesions for tremor associated with Parkinson’s disease: a probabilistic analysis based on postoperative magnetic resonance imaging and integrated digital atlas. J Neurosurg 2002;96:854–66. Baker SN, Olivier E, Lemon RN. Coherent oscillations in monkey motor cortex and hand muscle EMG show task-dependent modulation. J Physiol 1997;501(Pt 1):225–41. Bertrand G. Stimulation during stereotactic operations for dyskinesias. J Neurosurg 1966;24:419–28. Bertrand G, Jasper HH. Microelectrode recording of unit activity in the human thalamus. Confin Neurol 1965;26:205–8. Bertrand G, Jasper A, Wong A, Matthews G. Microelectrode recording during stereotactic surgery. Clin Neurosurg 1969;16:328–55. Beuter A, Edwards R. Using frequency domain characteristics to discriminate physiologic and Parkinsonian tremors. J Clin Neurophysiol 1999;6(5):484–94. Boecker H, Wills AJ, Ceballos-Baumann A, Samuel M, Thomas DG, Marsden CD, Brooks DJ. Stereotactic thalamotomy in tremor-dominant Parkinson’s disease: an H2(15)O PET motor activation study. Ann Neurol 1997;41:108–11. Conway BA, Halliday DM, Farmer SF, Shahani U, Maas P, Weir AI, Rosenberg JR. Synchronization between motor cortex and spinal motoneuronal pool during the performance of a maintained motor task in man. J Physiol 1995;489(Pt 3):917–24. Duffau H, Tzourio N, Caparros-Lefebvre D, Parker F, Mazoyer B. Tremor and voluntary repetitive movement in Parkinson’s disease: comparison before and after L-dopa with positron emission tomography. Exp Brain Res 1996;107(3):453–62. Duval C, Panisset M, Bertrand G, Sadikot AF. Evidence that ventrolateral thalamotomy may eliminate the supraspinal component of both pathological and physiological tremors. Exp Brain Res 2000;132(2): 216–22. Elble RJ, Koller DC. Tremor. Baltimore, MD: John Hopkins University Press; 1990 p. 10–89. Gross J, Tass PA, Salenius S, Hari R, Freund HJ, Schnitzler A. Corticomuscular synchronization during isometric muscle contraction in humans as revealed by magnetoencephalography. J Physiol 2000; 527(Pt 3):623–31. Hagbarth KE, Young RR. Participation of the stretch reflex in human physiological tremor. Brain 1979;102(3):509–26. Halliday DM, Conway BA, Farmer SF, Rosenberg JR. Load-independent contributions from motor-unit synchronization to human physiological tremor. J Neurophysiol 1999;82(2):664–75. Hirai T, Jones EG. A new parcellation of the human thalamus on the basis of histochemical staining. Brain Res Rev 1989;14:1–34. Hua SE, Lenz FA, Zirh TA, Reich SG, Dougherty PM. Thalamic neuronal activity correlated with essential tremor. J Neurol Neurosurg Psychiatry 1998;64:273–6. Hutchison D, Lozano AM, Tasker RR, Lang E, Dostrovsky JO. Identification and characterization of neurons with tremor-frequency activity in human globus pallidus. Exp Brain Res 1997;113:557–63. Kish SJ, Shannak K, Hornykiewich O. Uneven pattern of dopamine loss in the striatum of patients with idiopathic Parkinson’s disease. Pathophysiologic and clinical implications. N Engl J Med 1988; 318(14):876–80. Lamarre Y. Central mechanisms of experimental tremor and their clinical relevance. In: Findley LJ, Koller WC, editors. Handbook of Tremor Disorders. New York: Marcel Dekker; 1995. p. 103–18. Lamarre Y, Joffroy AJ, Dumont M, De Montigny C, Grou F, Lund JP. Central mechanism of tremor in some feline and primate models. Can J Neurol Sci 1975;2:227–33.

C. Duval et al. / Clinical Neurophysiology 116 (2005) 1391–1399 Lenz FA, Tasker RR, Kwan HC, Schnider S, Kwong R, Murayama Y, Dostrovsky JO, Murphy J. Single unit analysis of the human ventral thalamic nuclear group: correlation of thalamic ‘tremor cells’ with the 3–6 Hz component of Parkinsonian tremor. J Neurosci 1988;8:754–64. Llinas R. Rebound excitation as the physiological basis for tremor: a biophysical study of the oscillating properties of mammalian central neurons. In: Findley LJ, Calpildeo R, editors. Movement Disorders: Tremor. London: Macmillan Press; 1984. p. 165–82. Marsden CD, Meadows JC, Lange GW, Watson RS. The role of the ballistocardiac impulse in the genesis of physiological tremor. Brain 1969;92:647–62. Marsden JF, Ashby P, Limousin-Dowsey P, Rothwell JC, Brown P. Coherence between cerebellar thalamus, cortex and muscle in man: cerebellar thalamus interactions. Brain 2000;123(Pt 7):1459–70. Marsden J, Limousin-Dowsey P, Fraix V, Pollak P, Odin P, Brown P. Intermuscular coherence in Parkinson’s disease: effects of subthalamic nucleus stimulation. Neuroreport 2001;12(6):1113–7. McAuley JH, Rothwell JC, Marsden CD. Frequency peaks of tremor, muscle vibration and electromyographic activity at 10, 20 and 40 Hz during human finger muscle contraction may reflect rhythmicities of central neural firing. Exp Brain Res 1997;114(3):525–41. Obrador S. A simplified neurosurgical technique for approaching and damaging the region of the globus pallidus in Parkinson’s disease. J Neurol Neurosurg Psychiatry 1957;20:47–9. Obrist WD, Thompson Jr. HK, Wang HS, Wilkinson WE. Regional cerebral blood flow estimated by 133-xenon inhalation. Stroke 1975; 6(3):245–56. Ohye C. Use of selective thalamotomy for various kinds of movement disorder, based on basic studies. Stereotact Funct Neurosurg 2000;75: 54–65. Ohye C, Hirai T, Miyazaki M, Shibazaki T, Nakajima H. VIM thalamotomy for the treatment of various kinds of tremor. Appl Neurophysiol 1982; 45:275–80. Ohye C, Shibasaki T, Hirai T, Kawashima Y, Hirato M, Matsumura M. Plastic change of thalamic organization in patients with tremor after stroke. Appl Neurophysiol 1985;48:288–92. Pare D, Curro’Dossi R, Steriade M. Neural basis of the Parkinsonian resting tremor: a hypothesis and its implications for treatment. Neuroscience 1990;35:217–26.

1399

Parker F, Tzourio N, Blond S, Petit H, Mazoyer B. Evidence for a common network of brain structures involved in Parkinsonian tremor and voluntary repetitive movement. Brain Res 1992;584(1–2):11–17. Poirier LJ, Sourkes TL, Bouvier G, Boucher R, Carabin S. Striatal amines, experimental tremor and the effect of harmaline in the monkey. Brain 1966;89:37–55. Robson JG. The effect of loading on the frequency of muscle tremor. J Physiol (London) 1959;149:29P–330. Sakamoto K, Miao T, Arihara M. Analysis of interaction of spinal and supraspinal reflex pathways involved in physiological tremor. Electromyogr Clin Neurophysiol 1998;8(2):103–13. Salenius S, Hari R. Synchronous cortical oscillatory activity during motor action. Curr Opin Neurobiol 2003;13:678–84. Salenius S, Avikainen S, Kaakkola S, Hari R, Brown P. Defective cortical drive to muscle in Parkinson’s disease and its improvement with levodopa. Brain 2002;125(Pt 3):491–500. St-Jean P, Sadikot AF, Collins L, Clonda D, Kasrai R, Evans AC, Peters TM. Automated atlas integration and interactive three-dimensional visualization tools for planning and guidance in functional neurosurgery. IEEE Trans Med Imaging 1998;17(5):672–80. Stiles RN, Randall JE. Mechanical factors in human tremor frequency. J Appl Physiol 1967;23:324–30. Sturman MM, Vaillancourt DE, Metman LV, Bakay RA, Corcos DM. Effects of subthalamic nucleus stimulation and medication on resting and postural tremor in Parkinson’s disease. Brain 2004;127(Pt 9): 2131–43. Vaillancourt DE, Newell KM. Amplitude changes in the 8–12, 20–25 and 40 Hz oscillations in finger tremor. Clin Neurophysiol 2000;111: 1792–801. Van Buskirk C, Wolbarsht ML, Stecher Jr K. The non-nervous causes of normal physiologic tremor. Neurology 1966;16:217. Volkmann J, Joliot M, Mogilner A, Ioannides AA, Lado F, Fazzini E, Ribary U, Llinas R. Central motor loop oscillations in Parkinsonian resting tremor revealed by magnetoencephalography. Neurology 1996; 46:1359–70. Yap CB, Boshes B. The frequency and pattern of normal tremor. Electroencephalogr Clin Neurophysiol 1967;22(3):197–203. Young RR, Hagbarth KE. Physiological tremor enhanced by maneuvers affecting the segmental stretch reflex. J Neurol Neurosurg Psychiatry 1980;43(3):248–56.