Skip Navigation


Brain Advance Access originally published online on August 22, 2003
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
126/12/2597    most recent
awg267v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (61)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Silberstein, P.
Right arrow Articles by Brown, P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Silberstein, P.
Right arrow Articles by Brown, P.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Brain, Vol. 126, No. 12, 2597-2608, December 2003
© 2003 Guarantors of Brain
doi: 10.1093/brain/awg267

Patterning of globus pallidus local field potentials differs between Parkinson’s disease and dystonia

Paul Silberstein1, Andrea A. Kühn1,3, Andreas Kupsch2, Thomas Trottenberg2, Joachim K. Krauss4, Johannes C. Wöhrle5, Paolo Mazzone6, Angelo Insola7, Vincenzo Di Lazzaro8, Antonio Oliviero8, Tipu Aziz2 and Peter Brown1

1 Sobell Department of Motor Neuroscience and Movement Disorders, Institute of Neurology, London and 2 Department of Neurological Surgery, Radcliffe Infirmary, Oxford, UK, 3 Department of Neurology, Charité Campus Virchow, Humboldt University Berlin, Berlin, 4 Department of Neurosurgery and 5 Department of Neurology, University Hospital, Klinikum Mannheim, Mannheim, Germany, 6 Operative Unit of Functional and Stereotactic Neurosurgery and 7 Operative Unit of Neurophysiology, CTO ‘A. Alesini’ Hospital and 8 Institute of Neurology, Università Cattolica, Rome, Italy

Correspondence to: Dr P. Brown, Sobell Department of Motor Neuroscience and Movement Disorders, Institute of Neurology, Queen Square, London WCIN 3BG, UK. E-mail: p.brown{at}ion.ac.uk

Received March 20, 2003. Revised June 2, 2003. Accepted June 15, 2003.


   Summary
Top
 Summary
Introduction
 Methods
 Results
 Discussion
 References
 
Here we test the hypothesis that there are distinct temporal patterns of synchronized neuronal activity in the pallidum that characterize untreated and treated parkinsonism and dystonia. To this end we recorded local field potentials (LFPs) from the caudal and rostral contact pairs of macroelectrodes implanted into the pallidum of patients for the treatment of Parkinson’s disease (12 cases recorded on and off medication, 17 macroelectrodes) and dystonia (10 cases, 19 macroelectrodes). Percentage LFP power in the 11–30 Hz band was decreased and that in the 4–10 Hz band increased across both contact pairs in treated Parkinson’s disease compared with untreated Parkinson’s disease. Dystonic patients had even less 11–30 Hz power and greater 4–10 Hz power compared with untreated or treated Parkinson’s disease patients. The change in the 4–10 Hz band in patients with dystonia was particularly manifest in the more rostral contact pair, presumed to be within or bridging the globus pallidus externus. We conclude that untreated and treated Parkinson’s disease and dystonia are characterized by different spatiotemporal patterns of activity in the human pallidum.

Keywords: local field potentials; dystonia; Parkinson’s disease

Abbreviations: A–D = analogue-to-digital conversion; GPe = globus pallidus externus; GPi = globus pallidus internus; HFS = high-frequency stimulation; LFP = local field potential; MPTP = 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine; STN = subthalamic nucleus


   Introduction
Top
 Summary
 Introduction
Methods
 Results
 Discussion
 References
 
High-frequency stimulation (HFS) of the pallidum has been shown to be of therapeutic benefit in the treatment of Parkinson’s disease (Siegfried and Lippitz, 1994Go; Bejjani et al., 1997Go, 1998; Davis et al., 1997Go; Gross et al., 1997Go; Pahwa et al., 1997Go; Siegfried and Wellis, 1997Go; Yelnik et al., 2000Go) and dystonia (Kumar et al., 1999Go; Coubes et al., 2000Go). In Parkinson’s disease, stimulation of the pallidum improves motor performance by 54% (SD 33%) on evaluation of the UPDRS (Unified Parkinson’s Disease Rating Scale) Part III (Volkmann et al., 2001Go) and may completely abolish levodopa-induced dyskinesias (Krack et al., 1998Goa; Yelnik et al., 2000Go). Generalized dystonia can be improved by up to 90% by pallidal stimulation (Kumar et al., 1999Go; Coubes et al., 2000Go).

The therapeutic success of pallidal functional neurosurgery for movement disorders has revealed two related paradoxes. First, how does lesioning or HFS of one of the two major outflow stations of the basal ganglia, the globus pallidus internus (GPi), improve parkinsonism without any clear deleterious effects upon motor function (Marsden and Obeso, 1994Go; Bar-Gad and Bergman, 2001Go)? Secondly, how do the same interventions improve such disparate states as parkinsonism (Baron et al., 1996Go; Tronnier et al., 1997Go; Yelnik et al., 2000Go), dyskinesia (Baron et al., 1996Go; Krack et al., 1998Gob; Yelnik et al., 2000Go) and dystonia (Tronnier and Fogel, 2000Go; Coubes et al., 2000Go; Vercueil et al., 2001Go)?

It has been suggested that the first of these paradoxes might be explained if the parkinsonian state was characterized by abnormal patterning of pallidal outflow, so that no activity is better than bad or noisy activity, although the precise nature of such abnormal patterning was not specified (Marsden and Obeso, 1994Go). In line with this, microelectrode studies from the GPi in normal monkeys and monkeys treated with 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) as well as Parkinson’s disease patients on and off levodopa therapy have shown alterations not only in neuronal firing rates but also in the temporal patterning and synchronization of neuronal activity (Raz et al., 2000Go; Levy et al., 2001Go). Local field potential (LFP) recordings in the GPi, which probably represent the product of synchronous activity in a population of neurons (Brown et al., 2001Go), also demonstrate an altered pattern of frequency-dependent coupling between the GPi, the subthalamic nucleus (STN) and the motor cortex in patients with Parkinson’s disease, dependent on the dopaminergic state (Cassidy et al., 2002Go; Williams et al., 2002Go). Thus the bulk of evidence points to a difference in patterning in the off and on states. The knock-on effects of such patterning could be either obliterated by pallidotomy or suppressed by high-frequency pallidal stimulation (Anderson et al., 2003Go).

The paradoxical improvement in differing disease states with pallidal lesions and stimulation might also be explained if there was more than one type of abnormal patterning of pallidal outflow. Here, the evidence is more limited, but in dystonia patterns of neuronal discharge have been reported that differ from those in normal primates (Lenz et al., 1999Go; Vitek et al., 1999Go;), although the extent to which they may differ from those in parkinsonian primates and Parkinson’s disease patients is unclear.

Here we test the hypothesis that there are distinct temporal patterns of synchronized neuronal activity in the pallidum that characterize parkinsonism and dystonia. LFPs were recorded from the different contacts of macroelectrodes inserted into the pallidum of patients for the treatment of Parkinson’s disease and dystonia. Patients were studied while alert, during the interval between implantation and subsequent connection of the macroelectrode to a subcutaneous stimulating device. By recording the LFP from macroelectrode contacts rather than single-unit activity from microelectrodes we were able to concentrate on changes in the synchronous activity across many neurons, as unsynchronized neuronal activity is cancelled out and lost in the LFP. We focused our attention on LFP activity in the 4–10, 11–30 and 65–85 Hz bands, as past work suggests that these activities may differ in functional significance (Brown et al., 2001Go; Williams et al., 2002Go). The results confirm that untreated Parkinson’s disease, treated Parkinson’s disease and dystonia can be distinguished by the pattern of synchronization of local activity evident in the LFPs recorded from the human globus pallidus.


   Methods
Top
 Summary
 Introduction
 Methods
Results
 Discussion
 References
 
Patients and surgery
All patients participated with informed consent and with the agreement of the joint ethical committee of the National Hospital for Neurology and Neurosurgery and the Institute of neurology, London; the Ethics Committee of The Clarité-Universitätsmedizin, Berlin, Campus Virchow-Klinikum and the Ethics Committee of the A. Alesini Hospital, Rome. Clinical details are summarized in Table 1.

Indications for surgery were medication-refractory severe dystonia and treatment-resistant Parkinson’s disease with or without levodopa-induced dyskinesias. Dyskinesias were severe preoperatively in nine out of the 12 Parkinson’s disease patients (13 out of a total of 17 macroelectrodes in the Parkinson’s disease patients). Patients were recorded 1–6 days postoperatively, at a time when levodopa-induced dyskinesias tended to be less evident because of temporary microlesional effects. There was no difference in the delay between surgery and recording between Parkinson’s disease (mean 3.4 days) and dystonic (mean 2.9 days) patients, and Parkinson’s disease patients were recorded on and off medication on the same day. Off-period dystonias were not seen in any of the Parkinson’s disease patients during recording. The operative procedure and beneficial clinical effects of stimulation have been described previously (Siegfried and Lippitz, 1994Go; Volkmann et al., 1998Go, 2001). Macroelectrodes were inserted bilaterally, with the exception of six Parkinson’s disease patients in whom implantation was unilateral. The macroelectrode used was model Medtronic 3387 (Medtronic Neurological Division, Minneapolis, MN, USA) with four platinum–iridium cylindrical surfaces (1.27 mm diameter and 1.5 mm length) and centre-to-centre separation of 3 mm. Contact 0 was the most caudal and contact 3 was the most rostral. Macroelectrodes were connected to a battery-operated programmable pulse- generator (Itrel II or Kinetra 7428; Medtronic) after an interval of 2–7 days. Eight of the nine dyskinetic patients and all 10 dystonic patients improved following pallidal stimulation.

Electrode placement
The intended coordinates at the tip of contact 0 were 19– 22 mm from the midline of the patient, 2–3 mm in front of the midcommissural point and 6 mm below the anterior commissure–posterior commissure line, i.e. within the GPi. Under these circumstances, with a trajectory that is angled by 70–80° in the sagittal plane and by 75–85° in the coronal plane, the most rostral contact pair is likely to lie in the globus pallidus externus (GPe) (Yelnik et al., 2000Go; Peppe et al., 2001Go). Intraoperatively, the correct placement of the electrodes was verified by teleradiography using ventriculography, Guiot’s landmarks/volumetric fusion and a long-distance biorthogonal X-ray system. In addition, macrostimulation of the pallidum and surrounding structures, including the internal capsule and the optic tract, was performed intraoperatively. All patients received preoperative high-resolution MRI. Postoperative imaging consisted of CT (n = 2, Cases 12 and 13) or MRI (n =13, Cases 2, 4, 6, 9, 10, 14, 16–22). Where postoperative CT was performed, the macroelectrode position was superimposed on preoperative MRI using image fusion systems. No postoperative imaging was performed in seven cases (Cases 1, 3, 5, 7, 8, 11 and 15). Postoperative imaging, where available, was consistent with the placement of macroelectrode contacts 0 and/or 1 and 2 and/or 3 in the GPi and GPe/GPe–putamen border, respectively (Fig. 1). Note that postoperative images have been previously published in Cases 1 (Brown et al., 2001Go), 4 (Brown et al., 2002Go) and 13 (Brown et al., 2002Go).



View larger version (111K):
[in this window]
[in a new window]
  		Fig. 1 Axial MRI sections of 3 mm thickness (0 mm between sections) presented rostrocaudally (A–D) in Case 14, with cervical dystonia. Contacts 01 are in the region of GPi (C, D) whereas contacts 23 are in the region of GPe (A, B) bilaterally. The macroelectrode tip was visualized in the section below D (not shown).

 
Recordings
Subjects were seated or supine and recorded at rest. In the parkinsonian patients, recordings were divided into those performed in the practically defined off state, after overnight withdrawal of antiparkinsonian medication, and those performed over a period 45–135 min after ingestion of levodopa, during the clinical effect of this drug. These states will be termed ‘on’ and ‘off’ (drug). The mean test dose of levodopa given in combination with a decarboxylase inhibitor was 200 mg. LFPs were recorded bipolarly from the adjacent four contacts of each macroelectrode (0–1, 1–2, 2–3). They were filtered at 1–300 Hz and amplified (x100 000 to x500 000). Signals were sampled at 1 kHz and monitored on-line. Amplification, filtering and recording were performed using a custom-made 9 V battery-operated portable high-impedance amplifier (its front end input stage was the INA128 instrumentation amplifier; Texas Instruments, Dallas, TX, USA) and recorded through a 12-bit analogue-to-digital conversion (A–D) card (PCM-DAS16S; ComputerBoards, Middleboro, MA, USA) with a maximum throughput rate of 330 kHz, onto a portable computer using a custom-written program in Berlin (except in Cases 21 and 22, in whom A–D conversion was performed with a CED 1401; Cambridge Electronic Design, Cambridge, UK), Mannheim and Oxford. Amplification, filtering and recording were performed using a Nicolet Viking IIe and thereafter through an A–D card onto a portable computer in Rome and, using a Schwartzer 34 amplifier system (Schwartzer Medical Diagnostic Equipment, Munich, Germany) and Brainlab software (OSG, Rumst, Belgium) in Amsterdam. With one exception, recordings were made by either P.B. and/or A.A.K.

Analysis
Only LFPs recorded from contact pairs 01 and 23 were analysed, as contacts 0 and/or 1 and 2 and/or 3 were likely to be placed in GPi and GPe, respectively (see above) and did not share a contact. LFPs were digitally filtered at 50 Hz. A fast Fourier transform was performed on non-overlapping sections of equal length using commercial software (Spike 4; Cambridge Electronic Design). Results were averaged across sections and the autospectra determined over 4–100 Hz. The mean number of segments averaged was 800 per subject (range 110–2200). Spectral estimates were derived from long records, so that changes in power could have arisen from a persistent change in activity or periods of altered activity. Inspection of the signals suggested that the latter was only important in the 4–10 Hz band, where bursts of activity were frequently seen to last 1–4 s in dystonic patients. The segment length used was 1024 points, giving a frequency bin size approximating 1 Hz. Thus segment lengths were 1 s in duration, so that significant departures from local stationarity within these blocks was unlikely, even in the case of the 4–10 Hz bursts. Power was expressed as the percentage of the total power in the 4–45 and 55–100 Hz range in each subject and summed across bins to give the percentage power in the 4–10, 11–30 and 65–85 Hz bands. These bands were chosen on the basis of past work that suggests that these activities may differ in functional significance (Brown et al., 2001Go; Williams et al., 2002Go). The 0–4 and 45–55 Hz ranges were omitted so as to avoid inclusion of movement artefact and mains noise, respectively. Relative rather than absolute power was measured to allow comparison across disease groups, as absolute power would be more likely to be dependent on proximity to the LFP source than relative power and to vary with minor changes in recording technique. At this stage it remains unclear whether absolute or proportional changes in neuronal synchronization across different frequency bands are likely to be the more relevant to abnormal motor activity. Statistical analysis was performed using a repeated measures general linear model, as described in the Results section (SPSS for Windows version 11; SSPS, Chicago, IL, USA). Where results were non-spherical, Greenhouse–Geisser correction was used and the corrected degrees of freedom are shown in the text. Post hoc paired and unpaired two-tailed t tests were applied where appropriate. As multiple post hoc tests were performed, significance levels are presented with Bonferroni correction. Discrete peaks in LFP power spectra were defined as more than two contiguous bins with mean power greater than twice that of the adjacent two contiguous bins on each side.


   Results
Top
 Summary
 Introduction
 Methods
 Results
Discussion
 References
 
Representative autospectra, expressed as a percentage of the total power up to 100 Hz, are shown for a patient with Parkinson’s disease (treated and untreated) and a patient with dystonia in Fig. 2. The three candidate frequency bands (4–10, 11–30 and 65–85 Hz) are shown by the boxed areas. A discrete peak was found consistently only in the 11–30 Hz band. Only five Parkinson’s disease patients showed a discrete peak in the 65–85 Hz band, and this was present only after drug treatment. A further two patients with dystonia had a discrete peak in the 65–85 Hz band.



View larger version (44K):
[in this window]
[in a new window]
  		Fig. 2 Raw data and LFP autospectra in Case 11 with Parkinson’s disease off (A, B, C) and on (D, E, F) levodopa and in Case 13 with dystonia (G, H, I). Raw data (A, D, G) from electrode pairs 01 and 23 are labelled. LFP autospectra from contact pair 01 are shown in B, E and H and from contact pair 23 in C, F and I. Autospectra are expressed as a percentage of total power up to 100 Hz. The three candidate frequency bands, 4–10, 11–30 and 65–85 Hz, are shown by the boxed areas.

 
Comparison of LFPs in treated and untreated Parkinson’s disease
The percentages of total power picked up at 17 macroelectrodes in 12 parkinsonian patients were entered into the general linear model with frequency band (three levels: 4–10, 11–30 and 65–85 Hz), treatment state (two levels: on and off levodopa) and contact pair (two levels: contacts 01 and 23) as main effects. A significant main effect for frequency [F(1,19) = 47.841, P < 0.001] and significant interactions between drug state and frequency [F(1,19) = 9.510, P = 0.005] and between contact and frequency [F(1,22) = 6.663, P = 0.011] were found. Post hoc analysis revealed an increase in the relative power in the 4–10 Hz band (P < 0.05) and a decrease in the 11–30 Hz band in the treated state (P < 0.05). The interaction between contact and frequency across treatment states was due to an increase in power in the 4–10 Hz band at the most caudal contact, presumed to be in the GPi (P < 0.01). The results are summarized in Fig. 3.



View larger version (14K):
[in this window]
[in a new window]
  		Fig. 3 Mean percentage total LFP power in (A) the 4–10 Hz and (B) 11–30 Hz bands for contact pairs 01 and 23 in untreated and treated Parkinson’s disease and dystonia. Percentage power in the 4–10 Hz band is greater at contact pair 01 than 23 in treated and untreated Parkinson’s disease. Percentage power in the 4–10 and 11–30 Hz bands was increased and decreased, respectively, when Parkinson’s disease patients were treated. These power changes were even more marked in dystonic patients, particularly at contact 23. See Results section for statistical analysis. No significant effects were seen in the 65–85 Hz band and these data are not illustrated. Error bars denote the standard errors of the mean.

 
A subgroup analysis between LFPs recorded in untreated Parkinson’s disease patients with and without tremor was also performed. The percentage of total power picked up at seven macroelectrodes in five tremulous patients was compared with that in 10 macroelectrodes in seven non-tremulous patients. The percentage of total power was entered into the general linear model with frequency band (three levels: 4–10, 11–30 and 65–85 Hz), presence of tremor (two levels: tremor and no tremor) and contact pair (two levels: contacts 01 and 23) as main effects. A significant main effect for frequency [F(1,17) = 46.514, P < 0.001] was found. No main effect for tremor or significant interactions were found.

Comparison of LFPs in untreated Parkinson’s disease and dystonia
The percentages of total LFP power detected at 19 macroelectrodes in the dystonic patients were compared with those in the untreated parkinsonian patients. The percentage of total power was entered into the general linear model with frequency band (three levels: 4–10, 11–30 and 65–85 Hz), disease (two levels; untreated Parkinson’s disease and dystonia) and contact pair (two levels: contacts 01and 23) as main effects. A significant main effect for frequency [F(1,41) = 90.169, P < 0.001] and a significant interaction between frequency and disease state [F(1,41) = 23.847, P < 0.001] were found. Post hoc analysis showed this to be due to marked differences in power in the 4–10 Hz (P < 0.001) and 11–30 Hz bands (P < 0.0001). The results are summarized in Fig. 3.

Comparison of LFPs in treated Parkinson’s disease and dystonia
The percentage of total LFP power picked up in the dystonic patients was also compared with that in the treated parkinsonian patients. The percentage of total power was entered into the general linear model with frequency band (three levels: 4–10, 11–30 and 65–85 Hz), disease (two levels: treated Parkinson’s disease and dystonia) and contact pair (two levels: contacts 01 and 23) as main effects. There was a significant main effect for frequency [F(1,43) = 82.310, P < 0.001], a significant two-way interaction between frequency and disease [F(1,43) = 9.234, P = 0.002] and a significant three-way interaction between patient disease, frequency and contact [F(2,56) = 4.063, P = 0.030]. Post hoc tests indicated that the two-way interaction between frequency and disease state was due to an increase in the 4– 10 Hz (P < 0.05) and a decrease in the 11–30 Hz (P < 0.01) percentage power in the dystonic patients. The three-way interaction between disease, frequency and contact was due to an increase in the 4–10 Hz activity at the most rostral electrode (P < 0.05). Finally, dystonic patients on medication and those without medication at the time of recording (Table 1) were compared using the general linear model. There was a main effect for frequency [F(1,20) = 63.459, P < 0.001], but no significant interactions involving treatment or contact, suggesting that drug treatment did not affect our results.


View this table:
[in this window]
[in a new window]
  		Table 1 Summary of cases
 

   Discussion
Top
 Summary
 Introduction
 Methods
 Results
 Discussion
References
 
We have demonstrated that untreated Parkinson’s disease, treated Parkinson’s disease and dystonia can be distinguished by the pattern of the LFP recorded from the human globus pallidus. That fluctuations in the cortical LFP are linked to the timing of synchronous neuronal discharges has been known for some time (Creutzfeldt et al., 1966Go; Frost, 1968Go) but it has only recently become evident that the same may be true of LFP changes in the basal ganglia. Thus there is good concordance between oscillations evident in LFPs and those in single units (Goto and O’Donnell, 2001Go; Levy et al., 2002Goa) and LFP fluctuations are tightly linked to postsynaptic effects in terms of coupling between LFPs in distant sites (Brown et al., 2001Go; Williams et al., 2002Go). Our results therefore suggest that untreated Parkinson’s disease, treated Parkinson’s disease and dystonia may be distinguished by different patterns of synchronization of local activity within the globus pallidus.

Figure 4 summarizes the core results. However, before examining our findings in detail, we should bear in mind some of the potential limitations of our study.



View larger version (15K):
[in this window]
[in a new window]
  		Fig. 4 Intended orientation and localization of the macroelectrode shown superimposed on the traced Schaltenbrand and Wahren atlas at the posterior commissure (PC) +20 mm (after Yelnik et al., 2000) with a summary of the LFP findings. Note that the 4–10 Hz activity is greater at the lower contact pair in Parkinson’s disease, but that this difference is lost in dystonia due to the greater increase in 4–10 Hz activity at the rostral contact pair.

 
Limitations of human recordings
One of the major limitations with pathophysiological investigations of patients receiving HFS is that the placement of macroelectrodes within their targets is presumptive and not backed up by histological confirmation, as in animal studies. Postoperative imaging cannot provide absolute confirmation of electrode placement and, even so, was available in only 70% of cases. We limited the effects of variance in electrode positioning by recording many patients. To achieve this we collaborated across several surgical centres, but this itself introduced the possibility of systematic bias, albeit non-intended, in the achieved target between centres.

On the other hand, the intended surgical coordinates were the same across patients, postoperative imaging (where available) was consistent with positioning of the distal contacts of the macroelectrode in GPi, and all but one of the dyskinetic parkinsonian patients and all of the dystonic patients were clinically improved following deep-brain stimulation, consistent with reasonable target accuracy. In addition, two findings point to some overall consistency of electrode placement between patients. First, a significant two-way interaction between contact and frequency was seen across treatment groups in Parkinson’s disease, indicating relatively consistent localization of macroelectrode contacts between patients with Parkinson’s disease. Secondly, a significant three-way interaction involving contact was observed between treated Parkinson’s disease and dystonic patients, pointing to relatively consistent localization of macroelectrode contacts within these patient groups. Excessive variance in the position of each contact between patients would have obviated any interactions involving contact as a main effect. Nevertheless, the possibility that differences between Parkinson’s disease and dystonic patients, which were treated largely at different centres, arose from minor systematic bias in achieved targeting should be borne in mind.

Another limitation in clinical studies is the inability to entirely control for muscle activity. All our patients were recorded at rest, but the three clinical groups are likely to have had different degrees of involuntary activity: rigidity and rest tremor in untreated parkinsonian patients, dyskinesias in treated parkinsonian patients and spontaneous dystonic activity in subjects with dystonia. The question therefore arises whether differences in LFPs between the patient groups were directly or indirectly responsible for the clinical differences or were the product of the different patterns of afferent activity accompanying the differing types of involuntary muscle contraction. In practice, however, levodopa-induced dyskinesias were diminished or absent at the time of recording due to microlesional effects and there was no difference between tremulous and non-tremulous Parkinson’s disease patients. Nevertheless, in the light of the limitations discussed above we must be cautious in our interpretation of differences in LFPs between clinical groups, seeking, wherever possible, corroboration from animal studies. In this regard it is particularly notable that our finding that LFP power is elevated in the 4–10 Hz band and decreased in the beta band in dystonic patients compared with either treated or untreated Parkinson’s disease patients is closely paralleled by the LFP changes in the globus pallidus of the dtsz mutant hamster. This mutant has attacks of predominantly sustained dystonic posturing lasting several hours (Gernert et al., 1998Go). In this animal model of paroxysmal dystonia, spectral changes in the theta and beta bands persist between attacks, suggesting that the changes are not a consequence of the attacks and accompanying afferent activity.

It should also be noted that changes in the power at 4–10 Hz could have potentially included movement-related artefact. This will have been limited by the fact that the spectral content of dystonia and dyskinesias is generally <4 Hz (Manson et al., 2000Go) and by our use of a differential amplifier. In practice, it also seems unlikely that changes in the 4–10 Hz band were substantially affected by movement, as differences tended to be asymmetrically distributed across contacts.

The effect of levodopa upon LFPs in Parkinson’s disease was marked, raising the question whether drug treatments in the dystonic patients may have accounted for the differences between these and Parkinson’s disease subjects. However, no differences were found between treated and untreated dystonic patients.

Finally, it should be remembered that recordings were performed several days after implantation, when regional oedema may still have been significant, as suggested by the microlesional effect noted in some patients. Nevertheless, the delay between operation and recording did not differ systematically between patient groups or across treatment conditions in the same Parkinson’s disease patients, so that it cannot account for the spectral differences between or within groups.

Pallidal LFP changes and bradykinesia
The finding of an increase in synchronization at 11–30 Hz when Parkinson’s disease patients were withdrawn from levodopa is consistent with previous studies. Nini et al. (1995Go) demonstrated a pronounced tendency towards synchronization at frequencies <30 Hz in the primate GPi following treatment with MPTP, which was not seen in healthy monkeys (Nini et al., 1995Go). Microelectrode studies in patients with Parkinson’s disease have also demonstrated synchronization of single units in the STN (Levy et al., 2000Go, 2002b) and GPi (Levy et al., 2001Go) at 11–30 Hz, particularly in tremulous patients. Some of this synchronization of pallidal activity may be related to the greater influence of striatal tonically active neurons in the parkinsonian state (Raz et al., 2001Go). In addition, coherence (coupling) has been reported between LFPs in the STN and GPi at 11–30 Hz and between these structures and the cerebral cortex in parkinsonian patients (Brown et al., 2001Go; Williams et al., 2002Go). The LFP power and coupling in this frequency band are attenuated by the ingestion of levodopa (in keeping with the present results) and during movement (Cassidy et al., 2002Go; Williams et al., 2002Go), leading to the hypothesis that activity at 11–30 Hz may be essentially antikinetic in nature (Brown, 2003Go). If this is correct, pallidal functional neurosurgery could improve parkinsonism without any clear deleterious effects upon motor function by reducing an abnormal and antikinetic patterning of pallidal outflow in the 11–30 Hz band (Brown, 2003Go).

It is noteworthy that no difference emerged in pallidal LFPs in the high gamma band (65–85 Hz) between treated and untreated Parkinson’s disease. This is in contrast to LFP spectra from the human subthalamic nucleus, which commonly demonstrate a discrete peak in the high gamma band following treatment with levodopa (Cassidy et al., 2002Go). The result also contrasts with studies of the coherence between LFPs in STN and GPi that show coupled activity in the high gamma band following treatment (Brown et al., 2001Go). However, as coherence is effectively normalized by autospectral power, strong coupling may be evident despite very little pallidal power at these frequencies. Neither is it necessarily the case that an increase in the coupling between spatially segregated neuronal populations in a particular band should entail an increase in local synchronization at the same frequency. Within the motor areas of the human cerebral cortex, for example, local power and inter-regional coherence/coupling may relate to different aspects of cortical function that might, to some extent, operate independently of each other (Gerloff et al., 1998Go).

It is also of interest that no difference was found between tremulous and non-tremulous patients with respect to proportional LFP power. A significant postoperative reduction in tremor due to microlesional effects in our tremulous patients may potentially account for this, but, in addition, it seems likely that tremor-synchronized pallidal oscillatory activity makes only a small contribution to the LFP. Hutchison et al. (1997Go) specifically searched for hand-related internal pallidal cells with tremor-synchronized oscillatory activity in three patients undergoing microelectrode exploration prior to pallidotomy, but found such activity in only 12% of 228 neurons sampled. Raz and colleagues (Raz et al., 2000Go) reported a 40% increase in oscillatory pallidal single-cell activity after two vervet monkeys were treated with MPTP, with frequencies distributed bimodally around 7 and 13 Hz. They found that oscillatory activity and tremor were phasic and dynamic phenomena and that in only 11% of records was there significant dependence between the timing of tremor and neuronal oscillations. Hurtado et al. (1999Go), in their case report of recordings from a single parkinsonian pallidum, similarly found that single cells may, over time, oscillate in both a phase-locked and an independent manner from limb tremor at a single joint. Thus, whilst cells with tremor-synchronized activity are clearly present in the parkinsonian pallidum, they may make a relatively small contribution to LFP rhythmicity. Our averaging over long segments, including non-tremor periods, is further likely to have diminished the impact of tremor-related oscillation on the LFP.

Pallidal LFP changes and hyperkinesia
Excessive movement (hyperkinesia) clearly dominated the clinical picture in our dystonic patients, but was also present in our Parkinson’s disease patients when treated with levodopa (although dyskinesias were often temporally suppressed shortly after surgery). Thus, it is interesting that there was a gradient of LFP change, whereby treated Parkinson’s disease patients had more 4–10 Hz and less 11–30 Hz activity than untreated Parkinson’s disease patients and dystonic patients had more 4–10 Hz activity and less 11–30 Hz activity than treated parkinsonian patients. These changes were even more marked when compared with untreated Parkinson’s disease patients, consistent with the conclusion that the difference in oscillatory activity between dystonia and Parkinson’s disease cannot be ascribed to treatment with levodopa. As noted above, it has been suggested previously that synchronization of neuronal activity at 11–30 Hz may have an essentially antikinetic effect in the basal ganglia (Brown, 2003Go). Consequently, the reduction in beta activity amongst our treated Parkinson’s disease and dystonic patients may have contributed to hyperkinesia.

What is the possible basis for the 4–10 Hz LFP activity at the neuronal level? Microelectrode studies demonstrate a tendency for neuronal discharge in the GPi, and its input STN, to occur in three modes: irregular, bursting (i.e. an inconstant period of time between grouped discharges) and oscillatory (Levy et al., 2001Go). The bursting and oscillatory modes, in particular, are likely to have a major spectral content in the 4–10 Hz band. The oscillatory mode is accompanied by a strong tendency to synchronization between neurons at the frequency of parkinsonian rest and action tremor, and is more prominent in the untreated parkinsonian state (reviewed in Brown, 2003Go). Presumably, therefore, the overall increase in power at 4–10 Hz with levodopa treatment in Parkinson’s disease represents an exacerbation of the bursting mode. The degree of synchronization between neurons during non-oscillatory bursting is unclear, but the available evidence is in favour of this mode actually increasing when parkinsonian patients are treated with the dopamine agonist apomorphine (Merello et al., 1999Go; Levy et al., 2001Go) or when MPTP-treated green monkeys undergo STN lesioning (and become dyskinetic but less tremulous) with ibotenic acid (Wichmann et al., 1994Go). Indeed, bursting may be particularly prominent during dyskinesias in primates and Parkinson’s disease patients (Yoshida, 1991Go; Levy et al., 2001Go) and during hemiballismus (Suarez et al., 1997Go).

What is the evidence that non-oscillatory bursting is also a feature of dystonia? In addition to changes in firing rate, microelectrode recordings from the globus pallidus in dystonic patients reveal marked abnormalities in the temporal patterning of neuronal discharge, neurons firing in irregularly grouped discharges with intermittent pauses (Lenz et al., 1999Go; Vitek et al., 1999Go; Vitek, 2002Go). Furthermore, analysis of interspike intervals in the endopeduncular nucleus (the homologue of GPi) of dystonic dtsz hamsters shows an increased proportion of neurons with a burst-like firing pattern compared with normal animals (Gernert et al., 2002Go). Evidence also exists from animal models of dystonia of burst discharge in other nodes of corticobasal ganglionic loops involved in motor control. Increased bursting has been noted in the monkey motor thalamus after induction of dystonia by intrathalamic injection of the GABA antagonist bicuculline (Macia et al., 2002Go), and dystonic movements corresponding to putaminal burst discharge have been seen after intraputaminal injection of bicuculline in the cat (Yamada et al., 1995Go).

Together, these observations suggest that abnormal neuronal bursting may be an important pathophysiological substrate for the development of hyperkinesias, including levodopa-induced dyskinesia and dystonia, and that this may be related to the increase in 4–10 Hz activity in our parkinsonian patients after treatment with levodopa and the even greater relative 4–10 Hz activity in dystonic compared with parkinsonian patients.

If this is correct, pallidal functional neurosurgery could improve dyskinesias and dystonia (as opposed to parkinsonism) by reducing an abnormal pattern of pallidal outflow in the 4–10 Hz band. Consonant with this is the report that myoclonic dystonia is associated with a preponderance of LFP power at 4–8 Hz in the pallidum (Liu et al., 2002Go). Note that the presence of increased 4–10 Hz activity in patients in whom surgery has led to a temporary amelioration of dyskinesias through microlesional effects does not obviate any association between these oscillations and the propensity to dyskinesias. On the other hand, it does suggest that these changes are not simply the consequence of hyperkinesia, perhaps through afferent activity. Similar observations have been made of the dystonic dtsz mutant hamster, in which increases in LFP activity in the theta band occur during and between paroxysmal hyperkinesias (Gernert et al., 1998Go).

Spatiotemporal differences between Parkinson’s disease and dystonia
Topographic differences in spectral activity were also noted between macroelectrode contacts. In Parkinson’s disease patients there was a gradient of 4–10 Hz activity such that this was maximal at the most caudal contact, whether or not patients were treated. Interestingly, other studies have also suggested that there may be functional topographic differences within the globus pallidus in Parkinson’s disease (Bejjani et al., 1997Go; Krack et al., 1998Gob; Yelnik et al., 2000Go). In contrast, when dystonic patients were compared with Parkinson’s disease patients there was a general increase in 4–10 Hz activity that was most marked in the rostral contact pair. As discussed in the Methods section, the likely localization of one or both rostral contacts was the GPe, and the localization of abnormalities to the GPe in dystonia is not unprecedented. In humans, microelectrode studies have shown changes in the frequency and pattern of neuronal firing between the GPe and GPi. Mean firing rate is reduced in the GPe but not the GPi in dystonia compared with Parkinson’s disease on medication (Lenz et al., 1998Go). Consistent with this finding, the mean firing rate in the GPe was significantly lower than in the GPi in two out of the three cases of generalized dystonia reported by Vitek et al. (1999Go).

Conclusion
In conjunction with other lines of evidence, our studies of pallidal LFPs suggest that untreated Parkinson’s disease is associated with an increased tendency to synchronization over the 11–30 Hz range and a decreased tendency to synchronization at 4–10 Hz. The major change in dystonia is a decrease in relative power in the 11–30 Hz band and an increase in relative power in the 4–10 Hz range compared with treated or untreated Parkinson’s disease. The latter is focused at macroelectrode contacts that are more likely to be in the GPe than the GPi. The observations potentially offer an explanation for two of the enduring paradoxes of pallidal functional neurosurgery, namely, how this intervention improves such disparate states as parkinsonism, levodopa induced dyskinesias and dystonia, and how such improvement is achieved without any clearly deleterious effects upon motor function. It seems likely that each pathological state is associated with its own abnormal spatiotemporal pattern of pallidal activity, so that the success of functional neurosurgery may lie in the suppression of disruptive noisy activity across a number of frequency bands.


   Acknowledgements
 
We wish to thank David Williams, Gerard van Bruggen, Andries Bosch and Hans Speelman for their help in recording Case 15 and Klaus Niemann for his advice regarding imaging. A.A.K. was supported by a fellowship to the German Academic Exchange Service (DAAD).


   References
Top
 Summary
 Introduction
 Methods
 Results
 Discussion
 References
 
Anderson ME, Postupna N, Ruffo M. Effects of high-frequency stimulation in the internal globus pallidus on the activity of thalamic neurons in the awake monkey. J Neurophysiol 2003; 89: 1150–60.[Abstract/Free Full Text]

Bar-Gad I, Bergman H. Stepping out of the box: information processing in the neural networks of the basal ganglia. Curr Opin Neurobiol 2001; 11: 689–95.[CrossRef][ISI][Medline]

Baron MS, Vitek JL, Bakay RA, Green J, Kaneoke Y, Hashimoto T, et al. Treatment of advanced Parkinson’s disease by posterior GPi pallidotomy: 1-year results of a pilot study. Ann Neurol 1996; 40: 355–66.[CrossRef][ISI][Medline]

Bejjani B, Damier P, Arnulf I, Bonnet AM, Vidailhet M, Dormont D, et al. Pallidal stimulation for Parkinson’s disease. Two targets? Neurology 1997; 49: 1564–9.[Abstract/Free Full Text]

Bejjani BP, Damier P, Arnulf I, Papadopoulos S, Bonnet AM, Vidailhet M, et al. Deep brain stimulation in Parkinson’s disease: opposite effects of stimulation in the pallidum. Mov Disord 1998; 13: 969–70.[CrossRef][ISI][Medline]

Brown P. Oscillatory nature of human basal ganglia activity: relationship to the pathophysiology of Parkinson’s disease. Mov Disord 2003; 18: 357–63.[CrossRef][ISI][Medline]

Brown P, Oliviero A, Mazzone P, Insola A, Tonali P, Di Lazzaro, V. Dopamine dependency of oscillations between subthalamic nucleus and pallidum in Parkinson’s disease. J Neurosci 2001; 21: 1033–8.[Abstract/Free Full Text]

Brown P, Williams D, Aziz T, Mazzone P, Oliviero A, Insola A, et al. Pallidal activity recorded in patients with implanted electrodes predictively correlates with eventual performance in a timing task. Neurosci Lett 2002; 330: 188–92.[CrossRef][ISI][Medline]

Cassidy M, Mazzone P, Oliviero A, Insola A, Tonali P, Di Lazzaro V, et al. Movement-related changes in synchronization in the human basal ganglia. Brain 2002; 125: 1235–46.[Abstract/Free Full Text]

Coubes P, Roubertie A, Vayssiere N, Hemm S, Echenne B. Treatment of DYT1-generalised dystonia by stimulation of the internal globus pallidus. Lancet 2000; 355: 2220–1.[CrossRef][ISI][Medline]

Creutzfeldt OD, Watanabe S, Lux HD. Relations between EEG phenomena and potentials of single cortical cells. I. Evoked responses after thalamic and epicortical stimulation. Electroencephalogr Clin Neurophysiol 1966; 20: 1–18.[CrossRef][ISI][Medline]

Davis KD, Taub E, Houle S, Lang AE, Dostrovsky JO, Tasker RR, et al. Globus pallidus stimulation activates the cortical motor system during alleviation of parkinsonian symptoms. Nat Med 1997; 3: 671–4.[CrossRef][ISI][Medline]

Frost JD Jr. EEG–intracellular potential relationships in isolated cerebral cortex. Electroencephalogr Clin Neurophysiol 1968; 24: 434–43.[CrossRef][ISI][Medline]

Gerloff C, Uenishi N, Hallett M. Cortical activation during fast repetitive finger movements in humans: dipole sources of steady-state movement-related cortical potentials. J Clin Neurophysiol 1998; 15: 502–13.[CrossRef][ISI][Medline]

Gernert M, Richter A, Rundfeldt C, Loscher W. Quantitative EEG analysis of depth electrode recordings from several brain regions of mutant hamsters with paroxysmal dystonia discloses frequency changes in the basal ganglia. Mov Disord 1998; 13: 509–21.[CrossRef][ISI][Medline]

Gernert M, Bennay M, Fedrowitz M, Rehders JH, Richter A. Altered discharge pattern of basal ganglia output neurons in an animal model of idiopathic dystonia. J Neurosci 2002; 22: 7244–53.[Abstract/Free Full Text]

Goto Y, O’Donnell P. Network synchrony in the nucleus accumbens in vivo. J Neurosci 2001; 21: 4498–504.[Abstract/Free Full Text]

Gross C, Rougier A, Guehl D, Boraud T, Julien J, Bioulac B. High-frequency stimulation of the globus pallidus internalis in Parkinson’s disease: a study of seven cases. J Neurosurg 1997; 87: 491–8.[ISI][Medline]

Hurtado JM, Gray CM, Tamas LB, Sigvardt KA. Dynamics of tremor-related oscillations in the human globus pallidus: a single case study. Proc Natl Acad Sci USA 1999; 96: 1674–9.[Abstract/Free Full Text]

Hutchison WD, Lozano AM, Tasker RR, Lang AE, Dostrovsky JO. Identification and characterization of neurons with tremor-frequency activity in human globus pallidus. Exp Brain Res 1997; 113: 557–63.[CrossRef][ISI][Medline]

Krack P, Pollak P, Limousin P, Hoffmann D, Xie J, Benazzouz A, et al. Subthalamic nucleus or internal pallidal stimulation in young onset Parkinson’s disease. Brain 1998a; 121: 451–7.[Abstract/Free Full Text]

Krack P, Pollak P, Limousin P, Hoffmann D, Benazzouz A, Le Bas JF, et al. Opposite motor effects of pallidal stimulation in Parkinson’s disease. Ann Neurol 1998b; 43: 180–92.[CrossRef][ISI][Medline]

Kumar R, Dagher A, Hutchison WD, Lang AE, Lozano AM. Globus pallidus deep brain stimulation for generalized dystonia: clinical and PET investigation. Neurology 1999; 53: 871–4.[Abstract/Free Full Text]

Lenz FA, Suarez JI, Metman LV, Reich SG, Karp BI, Hallett M, et al. Pallidal activity during dystonia: somatosensory reorganisation and changes with severity. J Neurol Neurosurg Psychiatry 1998; 65: 767–70.[Abstract/Free Full Text]

Lenz FA, Jaeger CJ, Seike MS, Lin YC, Reich SG, DeLong MR, et al. Thalamic single neuron activity in patients with dystonia: dystonia-related activity and somatic sensory reorganization. J Neurophysiol 1999; 82: 2372–92.[Abstract/Free Full Text]

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: 7766–75.[Abstract/Free Full Text]

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 Parkinson’s disease. J Neurophysiol 2001; 86: 249–60.[Abstract/Free Full Text]

Levy R, Ashby P, Hutchison WD, Lang AE, Lozano AM, Dostrovsky JO. Dependence of subthalamic nucleus oscillations on movement and dopamine in Parkinson’s disease. Brain 2002a; 125: 1196–209.[Abstract/Free Full Text]

Levy R, Hutchison WD, Lozano AM, Dostrovsky JO. Synchronized neuronal discharge in the basal ganglia of parkinsonian patients is limited to oscillatory activity. J Neurosci 2002b; 22: 2855–61.[Abstract/Free Full Text]

Liu X, Griffin IC, Parkin SG, Miall RC, Rowe JG, Gregory RP, et al. Involvement of the medial pallidum in focal myoclonic dystonia: a clinical and neurophysiological case study. Mov Disord 2002; 17: 346–53.[CrossRef][ISI][Medline]

Macia F, Escola L, Guehl D, Michelet T, Bioulac B, Burbaud P. Neuronal activity in the monkey motor thalamus during bicuculline-induced dystonia. Eur J Neurosci 2002; 15: 1353–62.[CrossRef][ISI][Medline]

Manson AJ, Brown P, O’Sullivan JD, Asselman P, Buckwell D, Lees AJ. An ambulatory dyskinesia monitor. J Neurol Neurosurg Psychiatry 2000; 68: 196–201.[Abstract/Free Full Text]

Marsden CD, Obeso JA. The functions of the basal ganglia and the paradox of stereotaxic surgery in Parkinson’s disease. Brain 1994; 117: 877–97.[Abstract/Free Full Text]

Merello M, Balej J, Delfino M, Cammarota A, Betti O, Leiguarda R. Apomorphine induces changes in GPi spontaneous outflow in patients with Parkinson’s disease. Mov Disord 1999; 14: 45–9.[CrossRef][ISI][Medline]

Nini A, Feingold A, Slovin H, Bergman H. Neurons in the globus pallidus do not show correlated activity in the normal monkey, but phase-locked oscillations appear in the MPTP model of parkinsonism. J Neurophysiol 1995; 74: 1800–5.[Abstract/Free Full Text]

Pahwa R, Wilkinson S, Smith D, Lyons K, Miyawaki E, Koller WC. High-frequency stimulation of the globus pallidus for the treatment of Parkinson’s disease. Neurology 1997; 49: 249–53.[Abstract/Free Full Text]

Peppe A, Pierantozzi M, Altibrandi MG, Giacomini P, Stefani A, Bassi A, et al. Bilateral GPi DBS is useful to reduce abnormal involuntary movements in advanced Parkinson’s disease patients, but its action is related to modality and site of stimulation. Eur J Neurol 2001; 8: 579–86.[CrossRef][ISI][Medline]

Raz A, Vaadia E, Bergman H. Firing patterns and correlations of spontaneous discharge of pallidal neurons in the normal and the tremulous 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine vervet model of parkinsonism. J Neurosci 2000; 20: 8559–71.[Abstract/Free Full Text]

Raz A, Frechter-Mazar V, Feingold A, Abeles M, Vaadia E, Bergman H. Activity of pallidal and striatal tonically active neurons is correlated in mptp-treated monkeys but not in normal monkeys. J Neurosci 2001; 21: RC128.

Siegfried J, Lippitz B. Bilateral chronic electrostimulation of ventroposterolateral pallidum: a new therapeutic approach for alleviating all parkinsonian symptoms. Neurosurgery 1994; 35: 1126–9.[ISI][Medline]

Siegfried J, Wellis G. Chronic electrostimulation of ventroposterolateral pallidum: follow-up. Acta Neurochir Suppl (Wien) 1997; 68: 11–3.[Medline]

Suarez JI, Metman LV, Reich SG, Dougherty PM, Hallett M, Lenz FA. Pallidotomy for hemiballismus: efficacy and characteristics of neuronal activity. Ann Neurol 1997; 42: 807–11.[CrossRef][ISI][Medline]

Tronnier VM, Fogel W. Pallidal stimulation for generalized dystonia. Report of three cases. J Neurosurg 2000; 92: 453–6.[ISI][Medline]

Tronnier VM, Fogel W, Kronenbuerger M, Steinvorth S. Pallidal stimulation: an alternative to pallidotomy? J Neurosurg 1997; 87: 700–5.[ISI][Medline]

Vercueil L, Pollak P, Fraix V, Caputo E, Moro E, Benazzouz A, et al. Deep brain stimulation in the treatment of severe dystonia. J Neurol 2001; 248: 695–700.[CrossRef][ISI][Medline]

Vitek JL. Pathophysiology of dystonia: a neuronal model. Mov Disord 2002; 17 Suppl 3: S49–S62.

Vitek JL, Chockkan V, Zhang JY, Kaneoke Y, Evatt M, DeLong MR, et al. Neuronal activity in the basal ganglia in patients with generalized dystonia and hemiballismus. Ann Neurol 1999; 46: 22–35.[CrossRef][ISI][Medline]

Volkmann J, Sturm V, Weiss P, Kappler J, Voges J, Koulousakis A, et al. Bilateral high-frequency stimulation of the internal globus pallidus in advanced Parkinson’s disease. Ann Neurol 1998; 44: 953–61.[CrossRef][ISI][Medline]

Volkmann J, Allert N, Voges J, Weiss PH, Freund HJ, Sturm V. Safety and efficacy of pallidal or subthalamic nucleus stimulation in advanced PD. Neurology 2001; 56: 548–51.[Abstract/Free Full Text]

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: 521–30.[Abstract/Free Full Text]

Williams D, Tijssen M, Van Bruggen G, Bosch A, Insola A, Di Lazzaro V, et al. Dopamine-dependent changes in the functional connectivity between basal ganglia and cerebral cortex in humans. Brain 2002; 125: 1558–69.[Abstract/Free Full Text]

Yamada H, Fujimoto K, Yoshida M. Neuronal mechanism underlying dystonia induced by bicuculline injection into the putamen of the cat. Brain Res 1995; 677: 333–6.[CrossRef][ISI][Medline]

Yelnik J, Damier P, Bejjani BP, Francois C, Gervais D, Dormont D, et al. Functional mapping of the human globus pallidus: contrasting effect of stimulation in the internal and external pallidum in Parkinson’s disease. Neuroscience 2000; 101: 77–87.[CrossRef][ISI][Medline]

Yoshida M. The neuronal mechanism underlying parkinsonism and dyskinesia: differential roles of the putamen and caudate nucleus. Neurosci Res 1991; 12: 31–40.[CrossRef][ISI][Medline]


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 BrainHome page
X. Liu, S. Wang, J. Yianni, D. Nandi, P. G. Bain, R. Gregory, J. F. Stein, and T. Z. Aziz
The sensory and motor representation of synchronized oscillations in the globus pallidus in patients with primary dystonia
Brain, June 1, 2008; 131(6): 1562 - 1573.
[Abstract] [Full Text] [PDF]

Home page
 J. Neurosci.Home page
E. Lalo, S. Thobois, A. Sharott, G. Polo, P. Mertens, A. Pogosyan, and P. Brown
Patterns of Bidirectional Communication between Cortex and Basal Ganglia during Movement in Patients with Parkinson Disease
J. Neurosci., March 19, 2008; 28(12): 3008 - 3016.
[Abstract] [Full Text] [PDF]

Home page
 BrainHome page
A. Sharott, P. Grosse, A. A. Kuhn, F. Salih, A. K. Engel, A. Kupsch, G.-H. Schneider, J. K. Krauss, and P. Brown
Is the synchronization between pallidal and muscle activity in primary dystonia due to peripheral afferance or a motor drive?
Brain, February 1, 2008; 131(2): 473 - 484.
[Abstract] [Full Text] [PDF]