Abstract

We studied the time course and nature of interactions between the subthalamic nucleus (STN) and the motor cortex in 8 Parkinson disease (PD) patients with chronically implanted STN deep-brain stimulation (DBS) electrodes. We first identified the cortical evoked potentials following STN stimulation. The most consistent potential was positive wave with peak latency of 22.2 ± 1.2 ms from stimulation of clinically effective contacts. We then stimulated the motor cortex with transcranial magnetic stimulation (TMS) at 2–15 ms and at the latency of the evoked potential (∼ 23 ms) following STN DBS. TMS induced currents in 3 directions: lateral–medial (LM) direction activated corticospinal axons directly, posterior–anterior (PA), and anterior–posterior (AP) directions activated corticospinal neurons transynaptically. Motor-evoked potentials (MEP) elicited by AP and PA TMS were facilitated at short (2–4 ms) and medium latencies (21–24 ms). However, MEPs elicited by LM TMS were not modified by STN DBS. Short-latency antidromic stimulation of the corticosubthalamic projections and medium latency transmission likely through the basal ganglia–thalamocortical circuit led to cortical evoked potentials and increased motor cortex excitability at specific intervals following STN stimulation at clinically effective contacts. Cortical activation may be related to the clinical effects of STN DBS in PD.

Introduction

High-frequency stimulation of the subthalamic nucleus (STN) improves motor symptoms in Parkinson's disease (PD) (Krack et al. 2003; Rodriguez-Oroz et al. 2005; Deuschl et al. 2006). However, its mechanism of action is not fully understood. The STN modulates the activities of basal ganglia output structures. STN activities in PD (Ceballos-Baumann et al. 1999; Magnin et al. 2000; Levy et al. 2002) are characterized by augmented synchrony of neuronal firing, loss of specificity of the receptive fields, and increased firing rates with bursting activities (Hutchison et al. 1998; Magarinos-Ascone et al. 2000; Hamani et al. 2004). This pathological drive from STN is hypothesized to disrupt the activities of the substantia nigra pars reticulata (SNr), globus pallidus pars interna (GPi), globus pallidus pars externa (GPe), pedunculopontine nucleus, thalamus, and various cortical areas (Benazzouz et al. 2000; Dostrovsky and Lozano 2002; Vitek 2002; Hamani et al. 2005).

Currently several hypotheses exist to explain the mechanism of action of deep-brain stimulation (DBS) including depolarization blockade (Beurrier and Lozano 2001), synaptic inhibition (Dostrovsky and Lozano 2002), synaptic depression (Wang and Zucker 1998; Zucker and Regehr 2002), and stimulation-induced disruption of the pathological network (Montgomery and Baker 2000; Brown and Eusebio 2008). Recent animal studies suggested that inhibition of the STN is not sufficient to produce motor benefit and STN DBS may work by activation of the motor cortex (Dejean et al. 2009; Gradinaru et al. 2009). STN DBS may act by different mechanisms on the various parkinsonian signs and multiple mechanisms may operate in parallel (Temperli et al. 2003).

The interactions between STN and motor cortex have been studied in animal models (Hazrati et al. 1995; Nambu et al. 1997; Inase et al. 1999; Dejean et al. 2009; Gradinaru et al. 2009) and in patients with STN DBS (Cunic et al. 2002; Strafella et al. 2004; Gaynor et al. 2008; Eusebio et al. 2009). Transcranial magnetic stimulation (TMS) of the motor cortex changed the firing rate of STN neurons (Strafella et al. 2004) and the oscillatory activities of the STN (Gaynor et al. 2008). Several authors reported evoked potentials recorded from scalp electrodes following low frequency STN stimulation at short (2–8 ms), medium (18–25 ms), and long latencies (more than 50 ms) (Ashby et al. 2001; Baker et al. 2002; MacKinnon et al. 2005). A previous study from our group demonstrated that single-pulse STN stimulation facilitated the ipsilateral motor cortex tested with TMS with induced current in the posterior–anterior (PA) current direction at short latencies of 3–4 ms (Hanajima et al. 2004). However, the study was performed shortly after surgery. It was not known if the contacts used were capable of producing clinically effective stimulation and the results could have been influenced by microlesion effects. No study has evaluated the interaction between STN and ipsilateral motor cortex at medium latencies. In the present study, we tested the effects of STN stimulation on cortical excitability at interstimulus intervals (ISIs) corresponding to the short and medium latency–evoked potentials using TMS at different current directions. The different current directions were used to determine whether the changes occurred at the cortical or subcortical levels. TMS with induced currents in the anterior–posterior (AP) and PA directions activate pyramidal neurons transsynaptically through interneurons and are sensitive to changes in cortical excitability, whereas induced current in the lateral–medial (LM) direction directly activates corticospinal axons and is not sensitive to changes in cortical excitability (Kaneko et al. 1996; Sakai et al. 1997; Di Lazzaro et al. 2001). Because animal studies showed that STN DBS activates the motor cortex (Dejean et al. 2009; Gradinaru et al. 2009), we hypothesized that motor cortex will be facilitated at the times of the short and medium latency–evoked potentials from STN DBS.

Materials and Methods

Subjects

We studied 8 PD patients (Table 1) who had bilateral STN DBS for more than 3 months. Patients were recruited from the Movement Disorders Clinic at the Toronto Western Hospital. All patients gave written informed consent, and the protocol was approved by the University Health Network Research Ethics Board.

Table 1

Clinical profile of patients studied

Subjects Age (years) Disease duration (years) Time after surgery (years) H&Y DBS parameters UPDRS* Medication—levodopa equivalent (mg/day) 
P1 75 16 11 Right: 2.3 V, 60 μs, 130 Hz, 2− C+ 47/41 800 
    Left: 2.8 V, 120 μs, 130 Hz, 1+ 2−   
P2 58 14 Right: 3 V, 120 μs, 80 Hz, 2− 1− C+ 39/30 1500 
    Left: 2.5 V, 120 μs, 80 Hz, 2− 1+   
P3 67 15 Right: 3.15 V, 60 μs, 185 Hz, 2− 3− C+ 32/20 1000 
    Left: 3.6 V, 90 μs, 185 Hz, 3− C+   
P4 57 18 2.5 Right: 2.5 V, 60 μs, 130 Hz, 3− C+ 21/20 262 
    Left: 3.5 V, 60 μs, 185 Hz, 3− 2− C+   
P5 59 11 Right: 3 V, 60 μs, 185 Hz, 2− C+ 15/8 600 
    Left: 2.6 V, 60 μs, 160 Hz, 1− 2− C+   
P6 62 13 Right: 3 V, 60 μs, 185 Hz, 1− 2− C+ 49/37 1000 
    Left: 2.6 V, 60 μs, 160 Hz, 1− 2− C+   
P7 52 13 2.5 Right: 3.3 V, 60 μs, 185 Hz, 1− C+ 27/16 1000 
    Left: 4 V, 60 μs, 185 Hz, 3- C+   
P8 61 12 Right: 3.5 V, 90 μs, 180 Hz, 1− 2− C+ 24/10 900 
    Left: 3.1 V, 60 μs, 180 Hz, 1− 2− C+   
Mean ± SD 61 ± 6.9 14 ± 2.2 5.1 ± 3.3 2.6 ± 0.7    
Subjects Age (years) Disease duration (years) Time after surgery (years) H&Y DBS parameters UPDRS* Medication—levodopa equivalent (mg/day) 
P1 75 16 11 Right: 2.3 V, 60 μs, 130 Hz, 2− C+ 47/41 800 
    Left: 2.8 V, 120 μs, 130 Hz, 1+ 2−   
P2 58 14 Right: 3 V, 120 μs, 80 Hz, 2− 1− C+ 39/30 1500 
    Left: 2.5 V, 120 μs, 80 Hz, 2− 1+   
P3 67 15 Right: 3.15 V, 60 μs, 185 Hz, 2− 3− C+ 32/20 1000 
    Left: 3.6 V, 90 μs, 185 Hz, 3− C+   
P4 57 18 2.5 Right: 2.5 V, 60 μs, 130 Hz, 3− C+ 21/20 262 
    Left: 3.5 V, 60 μs, 185 Hz, 3− 2− C+   
P5 59 11 Right: 3 V, 60 μs, 185 Hz, 2− C+ 15/8 600 
    Left: 2.6 V, 60 μs, 160 Hz, 1− 2− C+   
P6 62 13 Right: 3 V, 60 μs, 185 Hz, 1− 2− C+ 49/37 1000 
    Left: 2.6 V, 60 μs, 160 Hz, 1− 2− C+   
P7 52 13 2.5 Right: 3.3 V, 60 μs, 185 Hz, 1− C+ 27/16 1000 
    Left: 4 V, 60 μs, 185 Hz, 3- C+   
P8 61 12 Right: 3.5 V, 90 μs, 180 Hz, 1− 2− C+ 24/10 900 
    Left: 3.1 V, 60 μs, 180 Hz, 1− 2− C+   
Mean ± SD 61 ± 6.9 14 ± 2.2 5.1 ± 3.3 2.6 ± 0.7    

Note: All patients are men. H&Y—Hoehn and Yahr staging, UPDRS*—UPDRS motor score off/on stimulation. We calculated Levodopa equivalent dose by the formula; 100 mg of standard levodopa = 133 mg of controlled-release levodopa or 75 mg of levodopa plus entacapone or 1 mg of pramipexole or 5 mg of ropinirole or 10 mg of bromocriptine.

Magnetic Resonance Imaging (MRI) Localization of STN DBS Electrode Contacts

The methods to assess the location of DBS electrode contacts have been described in detail (Hamani et al. 2008). Briefly, preoperative T2 and postoperative axial 3D inversion recovery images were fused. We then registered the anterior and posterior commissures and established the location of each electrode contact relative to the midcommissural point and the internal anatomy of the subthalamic region. We considered the region extending from 2 mm ventral to 2 mm dorsal to the MR border of the STN as the dorsal STN/zona incerta (ZI) region (dSTN-ZI) (Saint-Cyr et al. 2002).

Experimental Design

The study consisted of 2 experiments performed on separate days. Experiment 1 identified the latencies of cortical evoked potentials from STN stimulation. In Experiment 2, the effects of STN DBS on motor cortex (M1) excitability were examined. Patients continued their usual scheduled dopaminergic medications during the study.

Experiment 1: Recording of Evoked Potentials from STN DBS

DBS Settings

The DBS frequency was changed to 10 Hz with same pulse width as that used in clinical setting (60–90 μs). In our preliminary studies, we recorded evoked potential using 3, 5, and 10 Hz of STN stimulation from 2 patients. Because the potentials recorded were identical, we chose 10 Hz stimulation to reduce the duration of study and to increase the number of epochs for averaging. The stimulation voltage was set at the highest level without adverse effect. All bipolar stimulation montages using adjacent contact pairs were tested (e.g., −0 + 1, −1 + 2, −2 + 3, −1 + 0, −2 + 1, and −3 + 2). Bipolar stimulations were used to reduce the stimulus artifact. Both sides were studied separately, and the contralateral stimulator was switched off.

Electroencephalographic (EEG) Recording

Potentials evoked at the scalp from STN DBS were recorded with 5-mm diameter gold-plated cup electrodes at 7 positions (Cz, C3, C4, F1, F2, Fp1, and Fp2) of the international 10–20 system. The reference electrodes were placed on the linked ears and the ground electrode was placed over Fz. Three minutes of EEG was recorded during stimulation with each pair of bipolar contacts.

Data Acquisition and Processing

EEG signals were amplified (gain 10k; band pass filter 5–500 Hz), sampled at 5 kHz using Synamps (Compumedic Neuroscan, Texas, United States) and Scan 4 software (NeuroScan Inc., El Paso, TX), and were stored in a computer for offline analysis. The continuous EEG data were transformed to epochs using the DBS artifacts as triggers. One thousand eight-hundred epochs were averaged. The latencies of evoked potentials from DBS were measured from the averaged signal.

Experiment 2: Effects of STN DBS on Motor Cortex Excitability

Electromyographic Recording

Electromyography (EMG) was recorded from the first dorsal interosseous (FDI) muscle contralateral to the stimulated motor cortex using disposable 9-mm diameter surface electrodes with a belly-tendon montage. EMG was amplified at 1K (Intronix Technologies Corporation Model 2024F, Bolton, Ontario, Canada), filtered (band pass 20 Hz–2.5 kHz), digitized at 5 kHz (Micro 1401, Cambridge Electronics Design, Cambridge, United Kingdom) and stored in a laboratory computer for offline analysis. Subjects were asked to relax throughout the experiment with EMG monitored on a computer screen and via loudspeakers.

DBS Protocol

The DBS frequencies were set to 3 or 30 Hz, at the same pulse width used in the clinical setting (60–90 μs). Stimulation frequencies of 3 and 30 Hz were used to explore the effects of different stimulation frequencies. We could not use frequencies higher than 30 Hz because the time between pulses would be too short for the medium-latency potentials. Monopolar stimulation was used for the study with the implanted programmable pulse generator case as the anode and the DBS electrode contact as the cathode (3− case+, 2− case+, 1− case+, and 0− case+), using the same contacts used in the clinical setting. The side showing greater clinical benefit but without significant rest tremor was chosen for the study. The other stimulator was switched off. The voltage for STN stimulation was set at 0.5 V below the threshold for eliciting motor-evoked potentials (MEP) in contralateral FDI muscle by direct current spread to the corticospinal tract. This threshold was estimated by averaging 500 epochs of STN DBS triggered EMG responses from resting contralateral FDI muscle.

Transcranial Magnetic Stimulation

A figure of 8–shaped coil (external diameter was 9.5 cm) connected to a Magstim 200 stimulator (Magstim, Whitland, Dyfed, United Kingdom) was used to deliver single-pulse TMS over the M1, ipsilateral to STN stimulated. Three different coil orientations were studied. To induce PA current in the brain, the handle of the coil pointed backward and laterally at about 45 deg from the midsagittal line, which induced current perpendicular to the central sulcus. In this orientation, pyramidal neurons were activated transsynaptically and produced early indirect (I1) waves (Kaneko et al. 1996; Sakai et al. 1997; Di Lazzaro et al. 2001). The coil was rotated by 180 deg to induce AP current in the brain, which recruited later I3 waves. LM current was induced by holding the handle of the coil laterally and perpendicular to the direction used for PA current. The LM direction preferentially activates corticospinal neurons directly and induces the direct (D) wave (Sakai et al. 1997; Hanajima et al. 1998, 2002). The optimal position for activating the contralateral FDI muscle in each current direction was established by moving the coil in 0.5-cm steps around the presumed hand motor area and was marked on the scalp. TMS intensity to produce MEP of 1 mV (MEP1mv) in the relaxed contralateral FDI muscle was determined separately for all 3 coil orientations.

Experimental Design

STN DBS were used as conditioning stimuli and were paired with ipsilateral TMS over M1. STN stimulation artifact recorded with surface electrodes placed over the chest was used to trigger TMS at the following ISIs: 2, 3, 4, 5, 6, 7, 8, 10, and 15 ms; evoked potential (EP) latency, EP +2 ms, EP –2 ms. EP referred to the peak of the middle latency EP (∼22 ms) found in Experiment 1 using the same electrode contact (cathode was considered the active contact) in the same patient. These ISIs were selected based on the results of Experiment 1. The trigger pulses for TMS were delivered from a Micro1401 interface (Cambridge Electronics Design) controlled by Signal Software (v3.07). TMS was delivered every 6 s.

In Experiment 2.1, TMS over ipsilateral M1 was delivered in the AP direction. STN stimuli were delivered at 3 Hz. Experiment 2.2 was the same as Experiment 2.1 except that STN stimuli were delivered at 30 Hz instead of 3 Hz to test the effects of different background stimulation rates. Experiments 2.3 and 2.4 tested the effects of different current directions of TMS. Experiment 2.3 was the same as Experiment 2.1 except that TMS was delivered in the PA direction, whereas Experiment 2.4 used TMS in the LM direction.

In each experimental run, the 12 ISIs between STN stimulation and TMS over M1 were delivered in random order, and each was repeated 10 times. TMS over M1 was adjusted to produce MEPs of ∼1-mV amplitude without DBS in the corresponding coil direction. Control responses were obtained by single-pulse TMS over M1 (20 trials) at the same intensity for each coil direction with DBS switched off.

Data Analysis

For each ISI, MEP amplitudes were expressed as a ratio to the mean amplitude of the control MEP without DBS. Repeated-measures analysis of variance (rmANOVA) was performed for Experiments 2.1, 2.3, and 2.4 with the TMS current direction and ISI as the within-subject factors, and for Experiments 2.1 and 2.2 with the STN DBS frequency and ISI as the within-subject factors. Separate rmANOVA was also conducted for each experiment followed by post hoc analysis with Fischer's protected least-square difference to determine which ISIs showed a significant change from baseline. Significance level was set at P < 0.05. Statview version 5.01 software was used for statistical analysis. Unless otherwise stated, values are reported as mean ± standard deviation (SD).

Results

The clinical profile and STN DBS stimulation parameters of the 8 PD patients are summarized in Table 1.

Experiment 1

MRI Localization of STN DBS Contacts

The locations of the contacts based on postoperative MRI are shown in Table 2. Postoperative MRI was not available for patients 1 and 8. In the 6 patients with MRI results, clinically effective contacts were in dSTN-ZI in 8 of 12 (66%) electrodes, STN proper in 3 (25%) electrodes and thalamus (P7) in 1 (8%) electrode. Patient 7 had tremor predominant PD.

Table 2

Localization of STN electrode contacts based on postoperative MRI

Patients Right side
 
Left side
 
P2 SNr STNa dSTN/ZIa Thal SNr STNa dSTN/ZIa Thal 
P3 SNr SNr/STN STNa ZIa SNr SNr/STN dSTN ZIa 
P4 SNr SNr STN STNa SNr SNr SNr/STNa STNa 
P5 SNr STN dSTN/ZIa Thal SNr STNa ZIa Thal 
P6 SNr STNa dSTN/ZIa Thal SNr STNa dSTN/ZIa Thal 
P7 SNr STNa ZI Thal SNr STN ZI/Thal Thala 
Patients Right side
 
Left side
 
P2 SNr STNa dSTN/ZIa Thal SNr STNa dSTN/ZIa Thal 
P3 SNr SNr/STN STNa ZIa SNr SNr/STN dSTN ZIa 
P4 SNr SNr STN STNa SNr SNr SNr/STNa STNa 
P5 SNr STN dSTN/ZIa Thal SNr STNa ZIa Thal 
P6 SNr STNa dSTN/ZIa Thal SNr STNa dSTN/ZIa Thal 
P7 SNr STNa ZI Thal SNr STN ZI/Thal Thala 

Note: ZI—zona incerta, dSTN—dorsal subthalamic nucleus, SNr—substantia nigra reticulata, Thal—thalamus, STN—subthalamic nucleus, and 0,1,2, and 3—electrode contacts.

a

Clinically effective electrode contacts, MRI for P1 and P8 was not available.

Evoked Potentials from STN Stimulation

Positive and negative potentials were observed time locked to STN stimuli at various latencies. The most consistent finding was a large amplitude, medium latency, and positive potential obtained in 7 of 8 patients studied. The medium latency–evoked potentials elicited in all patients are summarized in Table 3. These potentials had amplitudes of up to 15 μV (11 ± 3 μV) and were most prominent over the ipsilateral central leads (C3 or C4) in all patients and were absent or less prominent in Fp1 and Fp2. These medium-latency potentials were elicited following stimulation of bipolar contacts in the ZI-thalamic area (ZI-Thal) and dSTN-ZI. Stimulation of ZI-Thal evoked medium-latency potentials with onset latencies of 12.8 ± 2.9 ms and peak latencies of 18.9 ± 2.8 ms (Fig. 1A) in 8 of 12 sides studied in 6 patients who had postoperative MRI. Stimulation of dSTN-ZI evoked medium-latency potentials with latencies of 14.2 ± 0.6 ms and peak latencies of 22.7 ± 1.7 ms (Fig. 1B) in 10 of 12 sides studied in 6 patients. However, stimulation of contacts in the STN proper and SNr did not produce these potentials (Fig. 1C,D). The potentials from stimulation of bipolar contacts in ZI-Thal had significantly shorter onset and peak latencies than potentials from stimulation of contacts in dSTN-ZI (P = 0.0007). The medium latency–evoked potentials were elicited from clinically effective contacts in 11 of 16 sides studied in 8 patients. In patient 7, on the left side, the clinically effective contact was in thalamus. Medium-latency potentials were not elicited in patient 4, in whom no contacts were located in the dSTN, ZI, or thalamus (Table 2). This patient had suboptimal clinical improvement with high frequency STN DBS (Table 1).

Table 3

Latencies of medium latency–evoked potential from STN stimulation

Patient Right
 
Left (ms)
 
dSTN–ZI
 
ZI–Thal
 
dSTN–ZI
 
ZI–Thal
 
Onset (ms) Peak (ms) Onset (ms) Peak (ms) Onset (ms) Peak (ms) Onset (ms) Peak (ms) 
P2 14.2a 22.3a 11.2 19.5 13.8a 23.2a 12.9 21.9 
P3 14.6a 20.5a NR NR 14.1a 20.8a NRa NRa 
P4 NR NR NR NR NR NR NR NR 
P5 14.24a 23.37a 11.46 21.09 13.71a 23.45a 11.81 20.47 
P6 13.38a 22.56a 11.55 18.35 14.21a 22.61a 11.65 20.35 
P7 14.43a 21.54a 11.78 19.47 13.76 24.52 12.54a 17.35a 
EC 1+ 2− 2+ 3− 5+ 6− (1+ 2−) 6+ 7−(2+ 3−) 
P1 14.38a 23.56a 12.35 20.45 NRa NRa NR NR 
P8 NRa NRa NR NR 15.67a 24.23a NR NR 
Patient Right
 
Left (ms)
 
dSTN–ZI
 
ZI–Thal
 
dSTN–ZI
 
ZI–Thal
 
Onset (ms) Peak (ms) Onset (ms) Peak (ms) Onset (ms) Peak (ms) Onset (ms) Peak (ms) 
P2 14.2a 22.3a 11.2 19.5 13.8a 23.2a 12.9 21.9 
P3 14.6a 20.5a NR NR 14.1a 20.8a NRa NRa 
P4 NR NR NR NR NR NR NR NR 
P5 14.24a 23.37a 11.46 21.09 13.71a 23.45a 11.81 20.47 
P6 13.38a 22.56a 11.55 18.35 14.21a 22.61a 11.65 20.35 
P7 14.43a 21.54a 11.78 19.47 13.76 24.52 12.54a 17.35a 
EC 1+ 2− 2+ 3− 5+ 6− (1+ 2−) 6+ 7−(2+ 3−) 
P1 14.38a 23.56a 12.35 20.45 NRa NRa NR NR 
P8 NRa NRa NR NR 15.67a 24.23a NR NR 

Note: P4 had no electrode contact points in the ZI or thalamus. ZI—zona incerta, dSTN—dorsal STN, EC—electrode contact, NR—no response, Thal—thalamus. P1 and P8 did not have MRI information of electrode contacts.

a

Clinically effective contact points.

Figure 1.

Examples of evoked potentials elicited with bipolar stimulation in different locations in the subthalamic area. (A) Medium latency–evoked potential with onset latency of 11.6 ms and peak latency of 20.3 ms following stimulation of electrode contacts in ZI–thalamus in patient 6. The stimulation parameters were left side, 2–3+, 8 V, 60 μs, 10 Hz. Maximum amplitude was observed in C3 and Cz. Short latency negative-evoked potential was also observed with a peak latency of 3.2 ms in bilateral frontal and central leads. (B) Medium latency–evoked potential with onset latency of 13.7 ms and peak latency of 24.5 ms following stimulation of electrode contacts in dSTN-ZI in the patient 7. The stimulation parameters were right side, 1–2+, 7.5 V, 60 μs, 10 Hz. Maximum amplitude was seen in C4 and Cz. Short latency negative-evoked potential was also seen with a peak latency of 4 ms in bilateral frontal and central leads. (C) Short latency negative-evoked potential with peak latency of 6.4 ms following stimulation of electrode contacts in SNr–STN in patient 4 and was observed in bilateral frontal and central leads, more prominent on left side. The stimulation parameters were left side, 1–2+, 8 V, 60 μs, 10 Hz. (D) Short latency negative-evoked potentials with peak latency of 3.5 and 11.4 ms following stimulation of electrode contacts in SNr-STN in patient 5 and was observed in bilateral frontal and central leads, more prominent on the right side. The stimulation parameters were right side, 0–1+, 8 V, 60 μs, 10Hz.

Figure 1.

Examples of evoked potentials elicited with bipolar stimulation in different locations in the subthalamic area. (A) Medium latency–evoked potential with onset latency of 11.6 ms and peak latency of 20.3 ms following stimulation of electrode contacts in ZI–thalamus in patient 6. The stimulation parameters were left side, 2–3+, 8 V, 60 μs, 10 Hz. Maximum amplitude was observed in C3 and Cz. Short latency negative-evoked potential was also observed with a peak latency of 3.2 ms in bilateral frontal and central leads. (B) Medium latency–evoked potential with onset latency of 13.7 ms and peak latency of 24.5 ms following stimulation of electrode contacts in dSTN-ZI in the patient 7. The stimulation parameters were right side, 1–2+, 7.5 V, 60 μs, 10 Hz. Maximum amplitude was seen in C4 and Cz. Short latency negative-evoked potential was also seen with a peak latency of 4 ms in bilateral frontal and central leads. (C) Short latency negative-evoked potential with peak latency of 6.4 ms following stimulation of electrode contacts in SNr–STN in patient 4 and was observed in bilateral frontal and central leads, more prominent on left side. The stimulation parameters were left side, 1–2+, 8 V, 60 μs, 10 Hz. (D) Short latency negative-evoked potentials with peak latency of 3.5 and 11.4 ms following stimulation of electrode contacts in SNr-STN in patient 5 and was observed in bilateral frontal and central leads, more prominent on the right side. The stimulation parameters were right side, 0–1+, 8 V, 60 μs, 10Hz.

Negative potentials at ∼3 ms were seen in 5 of 16 sides tested. The onset latency was 3.7 ± 0.6 ms (range 3.1–4.7 ms; Fig. 1D). These potentials had a maximum amplitude over the ipsilateral frontal and central leads. They were seen following stimulation of contact points in ZI, STN, and SNr. In 6 of 16 sides tested, a negative potential ∼7 ms was seen with an onset latency of 7.4 ± 0.9 ms (range 6.4–8.9 ms; Fig. 1C). These potentials were most prominent in the ipsilateral frontal leads and were elicited following stimulation of the ZI, STN, and SNr. Both ∼3 and ∼7-ms negative potentials were elicited from clinically effective as well as ineffective contacts in dorsal and ventral STN area. The short latency–evoked potentials in all patients are summarized in Table 4.

Table 4

Short latency–evoked potentials from stimulation of dorsal and ventral contacts

Patient Right (ms)
 
Left (ms)
 
 SNr-STN dSTN-ZI SNr-STN dSTN-ZI 
P2 NR NRa NR NRa 
P3 NR NRa NR NRa 
P4 6.5a NR 4.7 & 8.9a NR 
P5 3.5 & 11.4 NRa NR 3.7a 
P6 NR NRa NR NRa 
P7 3.1 & 7.5 3.8 & 6.9a Nil 3.5 &7.3a 
EC 1+ 2− 2+ 3− 1+ 2− 2+ 3− 
P1 6.34 7.43a 7.2 NRa 
P8 NR NRa NR NRa 
Patient Right (ms)
 
Left (ms)
 
 SNr-STN dSTN-ZI SNr-STN dSTN-ZI 
P2 NR NRa NR NRa 
P3 NR NRa NR NRa 
P4 6.5a NR 4.7 & 8.9a NR 
P5 3.5 & 11.4 NRa NR 3.7a 
P6 NR NRa NR NRa 
P7 3.1 & 7.5 3.8 & 6.9a Nil 3.5 &7.3a 
EC 1+ 2− 2+ 3− 1+ 2− 2+ 3− 
P1 6.34 7.43a 7.2 NRa 
P8 NR NRa NR NRa 

Note: The onset latencies of evoked potentials are shown. ZI—zona incerta, dSTN—dorsal subthalamic nucleus, EC—electrode contact, and NR—no response, SNr—substantia nigra reticulata.

a

Clinically effective contact points.

Experiment 2

Recording of MEP from STN Stimulation

Thresholds for eliciting MEP in the contralateral FDI muscle by monopolar stimulation of the clinically effective STN contacts were 4.57 ± 0.97 V (range 3–6 V) on the side selected for TMS study. The MEP latencies following STN stimulation were 19.7 ± 0.99 ms (range 18–21 ms). Figure 2A and B shows the averaged EMG response recorded from the left FDI muscle with right STN stimulation at 3 and 7 V in patient 1. The thresholds and mean MEP latencies following direct STN stimulation in all patients are given in Table 5.

Table 5

Stimulation parameters for Exps 1 and 2 and the MEP latencies in contralateral FDI muscle in Exp 2

Patient STN stimulation (V) (Exp 1)
 
STN threshold (V) STN stimulation setting (Exp 2) TMS intensity (% of output)
 
Mean MEP latencies (ms)
 
AP PA LM STN AP PA LM 
P1 6.5 Right: 3.5 V, 60 μs, 3/30 Hz, 2− C+ 45 41 38 20 26 23.5 22 
P2 6.5 Right: 4.5 V, 120 μs, 3/30 Hz, 2− 1− C+ 59 47 39 20.5 27 24 22 
P3 6.5 Right: 3.5 V, 60 μs, 3/30 Hz, 2− 3− C+ 49 42 37 20 26 24.5 22 
P4 Right: 2.5 V, 60 μs, 3/30 Hz, 3− C+ 53 38 45 19 24 21.4 19.5 
P5 7.5 Left: 5.5 V, 60 μs, 3/30 Hz, 1− 2− C+ 55 44 44 21 25.5 23 24 
P6 6.5 Left: 4.5 V, 60 μs, 3/30 Hz, 1− 2− C+ 64 50 41 18 27 24 23 
P8 6.5 Right: 4.5 V, 90 μs, 3/30 Hz, 1− 2− C+ 50 40 32 19.5 26 23 22 
Patient STN stimulation (V) (Exp 1)
 
STN threshold (V) STN stimulation setting (Exp 2) TMS intensity (% of output)
 
Mean MEP latencies (ms)
 
AP PA LM STN AP PA LM 
P1 6.5 Right: 3.5 V, 60 μs, 3/30 Hz, 2− C+ 45 41 38 20 26 23.5 22 
P2 6.5 Right: 4.5 V, 120 μs, 3/30 Hz, 2− 1− C+ 59 47 39 20.5 27 24 22 
P3 6.5 Right: 3.5 V, 60 μs, 3/30 Hz, 2− 3− C+ 49 42 37 20 26 24.5 22 
P4 Right: 2.5 V, 60 μs, 3/30 Hz, 3− C+ 53 38 45 19 24 21.4 19.5 
P5 7.5 Left: 5.5 V, 60 μs, 3/30 Hz, 1− 2− C+ 55 44 44 21 25.5 23 24 
P6 6.5 Left: 4.5 V, 60 μs, 3/30 Hz, 1− 2− C+ 64 50 41 18 27 24 23 
P8 6.5 Right: 4.5 V, 90 μs, 3/30 Hz, 1− 2− C+ 50 40 32 19.5 26 23 22 

Note: Exp 1—Experiment 1; Exp 2—Experiment 2; R—right side, L—left side; STN threshold—lowest stimulation voltage that evoked MEP from the contralateral FDI muscle; TMS intensity—intensity of TMS required to evoke 1 mV MEP in the contralateral FDI muscle; AP—anterior–posterior direction of TMS; PA—posterior–anterior direction of TMS; LM—lateral–medial direction of TMS; and Data were not available for P7 as TMS study was not performed due to worsening of disease.

Figure 2.

Examples of MEPs from the contralateral FDI muscle following STN stimulation and TMS over the motor cortex in Patient 1. Each trace represents the average of 20 trials. (A) STN stimulation at 3 V produced no MEP. (B) STN stimulation at 7 V produced MEP with a latency of 20 ms. (C) LM direction TMS over M1 produced MEP with latency of 22 ms. (D) PA direction TMS produced MEP with latency of 23.5 ms. (E) AP direction TMS produced MEP with a latency of 26.5 ms.

Figure 2.

Examples of MEPs from the contralateral FDI muscle following STN stimulation and TMS over the motor cortex in Patient 1. Each trace represents the average of 20 trials. (A) STN stimulation at 3 V produced no MEP. (B) STN stimulation at 7 V produced MEP with a latency of 20 ms. (C) LM direction TMS over M1 produced MEP with latency of 22 ms. (D) PA direction TMS produced MEP with latency of 23.5 ms. (E) AP direction TMS produced MEP with a latency of 26.5 ms.

Single Pulse TMS over M1

The intensity for generating 1 mV MEP in the contralateral FDI muscle by TMS over M1 was 53.5 ± 6.4% of stimulator output for AP current direction, 43.1 ± 4.1% for PA direction, and 39.4 ± 4.4% for LM direction. The MEP latencies were 25.9 ± 1.0 ms for AP, 23.2 ± 1 ms for PA, and 21.9 ± 1.4 ms for LM direction. Figure 2 shows the MEP latencies following TMS with LM, PA, and AP current directions in Patient 1. TMS intensities used and the MEP latencies following M1 TMS in all patients are given in Table 5.

STN DBS Followed by Ipsilateral M1 TMS

The results for Experiments 2.1–2.4 are shown in Figure 3A–D. Repeated measures of ANOVA on ratio of MEP amplitudes for Experiment 2.1 (AP TMS paired with 3-Hz STN stimulation) and Experiment 2.2 (AP TMS paired with 30-Hz STN stimulation) showed significant main effects of ISI (F = 2.76; P = 0.004) but not for frequency. There were no significant interactions between frequency and ISI.

Repeated-measures of ANOVA on MEP amplitudes ratios for Experiments 2.1 (PA TMS), 2.3 (AP TMS), and 2.4 (LM TMS) showed significant main effects of ISI (F = 3.69; P = 0.001) but not for current direction (F = 2.74; P = 0.14). However, the interaction between current direction and ISI was significant (F = 1.73; P = 0.04). Post hoc analysis revealed significant differences between control TMS response and STN-M1 paired-pulse response at 2, 3, 4, 5, and 15 ms and at the peak of the medium latency–evoked potential (20–25 ms).

To further explore the significant interaction between current direction and ISI, we performed separate repeated measures of ANOVA for each experiment. There were significant main effects of ISI in Experiments 2.1 (F = 1.98; P = 0.04), 2.2 (F = 2.36; P = 0.0129), and 2.3 (F = 1.94; P = 0.049). However, for Experiment 2.4, the effect of ISI was not significant (F = 0.71; P = 0.72). Therefore, the significant current direction and ISI interaction can be explained by a significant effect of ISI for Experiments 2.1 and 2.3 but not for Experiment 2.4. Post hoc analysis for Experiments 1 and 2 showed significant facilitation of the STN-M1 paired-pulse response compared with the control TMS response at 2, 3, 4, and 5 ms and at medium latency–evoked potential (20–25 ms) (Fig. 3A,B). Experiment 2.3 showed significant facilitation of the STN-M1 paired-pulse response compared with control response at 2, 3, and 4 ms and at the medium latency–evoked potential (20–25 ms) (Fig. 3C).

Figure 3.

Shows time course of cortical facilitation following STN stimulation. The figures show the MEP amplitude elicited by TMS at different times after STN stimulation expressed as ratios to MEP amplitude produced by TMS with STN stimulator turned off. Error bars represent standard error. (A) AP direction TMS and STN stimulation at 3 Hz. Significant MEP facilitation was noted at 2, 3, 4, and 5 ms and medium latency–evoked potential latency. (B) AP direction TMS and STN stimulation at 30 Hz. Significant MEP facilitation was noted at 2, 3, 4, and 5 ms and medium latency–evoked potential latency. (C) PA direction TMS and STN stimulation at 3 Hz. Significant MEP facilitation was noted at 2, 3, and 4 ms and medium latency–evoked potential latency. (D) LM direction TMS and STN stimulation at 3 Hz. No significant MEP facilitation was found.

Figure 3.

Shows time course of cortical facilitation following STN stimulation. The figures show the MEP amplitude elicited by TMS at different times after STN stimulation expressed as ratios to MEP amplitude produced by TMS with STN stimulator turned off. Error bars represent standard error. (A) AP direction TMS and STN stimulation at 3 Hz. Significant MEP facilitation was noted at 2, 3, 4, and 5 ms and medium latency–evoked potential latency. (B) AP direction TMS and STN stimulation at 30 Hz. Significant MEP facilitation was noted at 2, 3, 4, and 5 ms and medium latency–evoked potential latency. (C) PA direction TMS and STN stimulation at 3 Hz. Significant MEP facilitation was noted at 2, 3, and 4 ms and medium latency–evoked potential latency. (D) LM direction TMS and STN stimulation at 3 Hz. No significant MEP facilitation was found.

Discussion

STN stimulation at 10 Hz produced evoked potentials recorded from the scalp with onset latencies at about 3, 7, and 13 ms (peak latencies of about 22 ms). Following STN stimulation, increased MEP amplitude was observed after 2–5 and 15–25 ms with TMS in AP and PA current directions but not with TMS in the LM direction over the ipsilateral motor cortex.

Short Latency–Evoked Potentials from STN Stimulation

We obtained short latency–evoked potential with onset latencies around 3 and 7 ms following STN stimulation from about one-third of the DBS contacts studied. The evoked potentials were most prominent in the ipsilateral frontal and central areas, consistent with the known anatomical projection of the basal ganglia. Previous studies have reported similar short latency–evoked potentials at 3, 5, and 8 ms in about two-third of the sides tested. Based on the short chronaxie and refractory period, these potentials likely arose from stimulation of myelinated axons (Ashby et al. 2001; Baker et al. 2002). One study found that the contacts that evoked short-latency potentials were often located in the ventral STN but could be located in the dorsal STN as well (Ashby et al. 2001). This is consistent with our findings.

Medium Latency–Evoked Potentials from STN Stimulation

We found that the most consistent response was a medium latency–evoked potential with an average onset latency of 14 ms and peak latency of 22 ms. It occurred only with stimulation of dorsal STN contact points near the ZI and was most prominent over the ipsilateral central region. This is similar to the findings of previous studies (MacKinnon et al. 2005; Eusebio et al. 2009), although postoperative MRI contact localization was not reported. In our study, the contacts that produced the medium latency–evoked potential were also the contacts that produced clinical benefit with high-frequency stimulation. However, MacKinnon et al. (2005) reported that the contacts used for clinical benefit were often ventral to the contacts that produced the largest medium latency–evoked potentials. The topography of potentials evoked by stimulation through clinically used contacts and contacts with largest medium latency–evoked potential were often different (MacKinnon et al. 2005). The different findings may be because we used electrodes with longer distance between contacts (Medtronic 3387) than those used by MacKinnon et al. (2005) (Medtronic 3389). Moreover, the selection of the clinically used contacts not only depends on clinically efficacy but also incorporates the need to avoid side effects. The monopolar stimulation montage used for clinical benefit also produced a larger stimulation volume than the bipolar montage used for the evoked potential studies. Therefore, it is likely that the circuits involved in medium latency–evoked potentials are also activated by the parameters used for chronic DBS. This is consistent with findings of Eusebio et al. (2009) that stimulation of the contacts used for clinical benefits also consistently produced this potential. Our results are similar to recent reports showing that 95% of clinically effective contact points for STN DBS in PD are located near the dorsal border of the STN (Godinho et al. 2006) and structures adjacent to the STN such as the ZI and Forel's field H2 could mediate some of the clinical benefit from STN DBS (Hamel et al. 2003; Benazzouz et al. 2004).

Medium latency–evoked responses have been reported following GPi and ventrolateral thalamic stimulation. GPi stimulation produced an evoked potential with a peak latency of 26.6 ± 1.6 ms (onset latencies were 10.9 ± 0.77 ms; Tisch et al. 2008), whereas ventrolateral thalamic stimulation produced a peak latency of 8–12 ms (onset latencies 4–6 ms; Nishimoto and Matsumoto 1970). The evoked potentials were similar from all 3 sites and were distributed mainly over the ipsilateral hemisphere, maximum centrally. The finding that similar evoked potentials are elicited following stimulation of dorsal STN, GPi, and ventrolateral thalamus suggest that the basal ganglia thalamocortical circuit are involved, but further work is needed to examine the pathways involved. Interestingly, the medium latency–evoked potentials are enhanced with STN DBS at 20 Hz compared with lower or higher frequencies, possibly related to a resonance frequency of the subthalamo–cortical circuit (Eusebio et al. 2009).

MEP Facilitation Following STN DBS

MEP facilitation with AP and PA TMS was observed at 2–5 ms following STN stimulation. These findings suggest STN stimulation increases cortical excitability at short latencies. This is similar to a previous study in which MEP facilitation was observed at 3 ms with AP TMS (Hanajima et al. 2004). However, the clinically effective contacts were not known in that study, whereas we used the clinically effective contacts. In addition, cortical excitability was examined for only 10 ms after DBS. TMS studies that examined cortical excitability at medium latency–evoked potentials have not been reported and we found cortical facilitation following STN stimulation at 15–25 ms. Our finding suggests the interaction between STN and motor cortex at medium latencies (15–25 ms) is facilitatory.

We studied patients in the on-mediation state because our study was lengthy and DBS was switched to ineffective stimulation rates or switched off during the study. Many of our patients with advanced PD did not tolerate off medication and off stimulation for long periods. We also avoided increased tremor and difficulties in relaxation of muscles in the off-medication state that could have affected our study. However, the clinical effects of STN DBS is greater in the off-medication than in the on-medication state (Deuschl et al. 2006) Therefore, we may have potentially underestimated the effects on STN DBS on cortical excitability

Several previous studies found that high-frequency STN DBS increased motor cortical inhibition but had no effect on motor threshold or MEP amplitude (Cunic et al. 2002; Dauper et al. 2002; Pierantozzi et al. 2002). This is likely because, in these studies, TMS was not time locked to STN DBS and changes in motor cortex excitability only occur at specific time intervals after STN DBS.

It is not possible to study directly cortical effects time locked to clinically effective higher frequency DBS because of the stimulus artifacts and the short time between pulses (∼6 ms for 180 Hz). However, we examined STN DBS at 3 Hz (1 pulse every 333 ms) and 30 Hz (1 pulse every 33 ms) and found virtually identical cortical facilitatory effects (Fig. 3A,B). Thus, the facilitatory effects we observed are likely due to STN DBS pulse immediately preceding the TMS rather than the cumulative effects of preceding pulses. High-frequency stimulation likely influences these circuits, but it is not known whether they are activated in the same manner and they may even be inhibited. Previous studies demonstrated that low-frequency STN stimulation (<20 Hz) may exacerbate PD symptoms, especially bradykinesia (Eusebio et al. 2008). However, the effects are subtle and only studies in “off” medication state have been reported. It is possible that at low frequencies, cortical activation may augment pathological oscillations in the basal ganglia–thalamo–cortical network.

Site of MEP Facilitation Following STN Stimulation

The neuronal populations activated by TMS depend on the direction of induced currents in brain. AP TMS leads to the late I3 wave, PA TMS leads to the I1 wave, and LM TMS preferentially evokes the earliest D wave (Kaneko et al. 1996; Sakai et al. 1997; Di Lazzaro et al. 2001). As a result, the MEP latencies following AP TMS are ∼4.5 ms and PA TMS ∼1.5 ms longer than LM TMS, and this is consistent with our findings (Table 5). Therefore, AP TMS and PA TMS activate corticospinal neurons located in layer V of the motor cortex transsynaptically through interneurons and are sensitive to changes in cortical excitability. In contrast, LM TMS directly activates corticospinal axons and is not sensitive to changes in cortical excitability. We observed MEP facilitation following STN stimulation with both AP and PA TMS but not with LM TMS (Fig. 3). These findings suggest that MEP facilitation occurs at the cortical level. Following STN stimulation, significant facilitation was observed only at 2–5 and 15–25 ms with TMS at AP and PA current directions, suggesting that facilitatory interactions between STN and interneurons in the motor cortex responsible for generating the I1 and I3 waves occur at these specific intervals that coincided with the peaks of the evoked potentials. Some previous studies reported that the MEP onset latencies after TMS over the M1 were shorter after DBS surgery in the resting state. They postulated that TMS induced currents in scalp leads underneath the TMS coil leading to induced currents at the DBS electrodes that activated the corticospinal tract (Kuhn et al. 2002; Hidding et al. 2006). However, in our study, the MEP latencies following direct stimulation of STN were significantly shorter than those produced by TMS (Fig. 2), suggesting that in our patients TMS did not cause activation of corticospinal tract by inducing current in the DBS leads. Average MEP latency following STN stimulation was 19.7 ms in our study, which is similar to the results obtained in previous studies (Ashby et al. 1999; Costa et al. 2007). The different MEP latencies with different current directions of TMS also make it unlikely that MEP from TMS can be attributed to induction of currents at the STN electrodes. Because we used subthreshold intensities for STN stimulation, it is unlikely that cortical activation at short latencies was due to the direct effects of the stimuli on corticospinal fibers. However, the possibility that fibers were partially depolarized and became facilitated with a subsequent stimulus cannot be completely excluded. Previous studies found TMS to be safe in patients with implanted STN DBS electrodes (Kumar et al. 1999; Cunic et al. 2002; Dauper et al. 2002; Pierantozzi et al. 2002), and the study was conducted in accordance with the guidelines published by the Safety of TMS consenus group (Rossi et al. 2009).

Possible Pathways for Evoked Potentials and Cortical Facilitation

Because we used voltage subthreshold for eliciting corticospinal effects to stimulate the STN, it is unlikely that our results are due to antidromic activation of the corticospinal tract to stimulate the motor cortex or orthodromic stimulation to activate the contralateral FDI muscle. A previous study (Hanajima et al. 2004) showed the early cortical facilitation at ∼3 ms with STN stimulation below the active motor threshold for corticospinal tract activation. Because inhibition rather than facilitation of EMG activity was observed at ∼22 ms after stimulation of corticospinal tract near the STN (Ashby et al. 1999; Costa et al. 2007), subthreshold corticospinal activation cannot account for the evoked potential or MEP facilitation at medium latencies.

Evoked potentials and cortical facilitation provide information regarding the STN-cortical interaction and the mechanism of action of STN DBS. The neurophysiological effects of STN stimulation could be due to activation of afferent fibers projecting to STN, activation of efferent fibers projecting from STN, inhibition of neurons within STN, or activation of structures adjacent to STN (McIntyre, Mori, et al. 2004; McIntyre, Savasta, et al. 2004; McIntyre, Savasta, Walter, et al. 2004; Valls-Sole et al. 2008). We will consider these possibilities separately.

Antidromic Activation of Cortico-Subthalamic Fibers

Direct projections from the motor cortex, the supplementary motor area, and the premotor cortex to the STN have been reported in monkeys and rats (Kunzle 1978; Monakow et al. 1978; Hazrati et al. 1995; Nambu et al. 1996; Inase et al. 1999). Electrical stimuli are more likely to activate axons than cell bodies, and the effects obtained with low-intensity stimulation are usually due to activation of axons in long tracts in the vicinity of the electrode (Nowak and Bullier 1998; Ashby et al. 1999). The short latency–evoked potentials and cortical facilitation (∼3 ms) observed in our study are most likely due to antidromic stimulation of cortical axons projecting to STN. This is supported by short-latency (2–4 ms) facilitation in STN observed following cortical stimulation in primates (Fujimoto and Kita 1993; Nambu et al. 2000). In addition, intracellular recording in rats revealed antidromic spiking of neurons in deep cortical layers following STN DBS (Maurice, Deniau, Glowinski, and Thierry, 1998; Maurice, Deniau, Menetrey, et al. 1998; Li et al. 2007; Dejean et al. 2009).

Activation of Subthalamo–Cortical Pathway

A subthalmo-cortical pathway has been described in the rat (Degos et al. 2008). Orthodromic activation of this pathway is possible, but we consider it less likely as it projects mainly to the orofacial part of the motor cortex (Degos et al. 2008).

Activation of Pallidosubthalamic Fibers and Subthalamopallidal Fibers

Activation of GABAergic GPe fibers projecting to the STN can inhibit the STN (Hashimoto 2000). Similarly, activation of glutaminergic STN projection fibers to GPe can stimulate GPe, which can in turn inhibit STN (Filali et al. 2004). Inhibition of the STN leads to decreased activation of GABAergic GPi efferents, which results in increased thalamocortical activity (Alexander et al. 1990; DeLong 1990). The latencies for these pathways are likely more than 6 ms and may explain the 7 ms evoked potential observed in some patients in our study. Activation of STN fibers projecting to GPi can increase GABAergic output to the thalamus thereby decreasing thalamocortical facilitatory projection (Alexander et al. 1990; DeLong 1990). This pathway can lead to decreased cortical facilitation, but we did not find cortical inhibition following STN stimulation. However, rebound bursting of thalamic neurons is a potential mechanism (see below).

Activation of Neurons within STN

Because we used low-frequency rather than high-frequency stimulation, inhibition of STN neurons observed with high-frequency (100–300 Hz) stimulation cannot explain our results (Filali et al. 2004). Activation of STN neurons would have similar effects to activating STN fibers projecting to GPi.

Activation of Structures Adjacent to STN

STN is surrounded by dense bundles of myelinated fibers (Yelnik and Percheron 1979). Forel's field H2 and ZI lies along the dorsal border of STN and separates it from ventral thalamus (Schaltenbrand et al. 1971; Parent and Hazrati 1995). Forel's field H2 contains pallidothalamic GABAergic fibers that inhibit ventral anterior thalamus (Parent et al. 2000). The ZI has rich GABAergic projections to ventrolateral thalamic nucleus (Bartho et al. 2002). Stimulation of dorsal STN contacts can activate adjacent pallidothalamic fibers and ZI (Godinho et al. 2006). Excessive GABAergic inhibition may cause hyperpolarization of thalamocortical neurons followed by rebound burst firing (Jahnsen and Llinas 1984a, 1984b). Because the medium latency–evoked potential seen with peak latency around 22 ms (mean onset 15 ms) was observed with stimulation of contacts located in the dorsal STN and ZI area (Table 3), this potential could be explained by activation of this pathway. Studies using GPi and ventral lateral thalamic stimulation have elicited similar medium latency–evoked potentials (Nishimoto and Matsumoto 1970; Tisch et al. 2008). Moreover, a TMS study showed that thalamic DBS increases motor cortex excitability (Molnar et al. 2005). These findings support our hypothesis that either pallidothalamic fibers or ZI stimulation may be responsible for medium latency–evoked potentials.

Possible Relationship between MEP Facilitation and Clinical Effects of STN DBS

The sensory-motor region of the STN is located in the dorsal STN, which receives afferent input from the supplementary motor area, premotor cortex, motor cortex, and motor portions of pallidum and centromedian nucleus of thalamus (Monakow et al. 1978; Nambu et al. 1996; Hamani et al. 2004).

Several recent animal studies provided direct evidence that STN DBS activates the cortex, and this may be related to the therapeutic effect of STN DBS. High-frequency STN DBS in rats produced short-latency (∼2 ms) antidromic spiking of deep-layer cortical neurons, which dampened the oscillation of local field potentials in cortex. The amplitude of antidromic activation correlated with suppression of slow waves and beta band activity during STN DBS (Li et al. 2007). Moreover, in awake animals, the amplitude of this short-latency cortical antidromic activation from STN DBS correlated with the degree of improvement in akinesia produced by dopaminergic blockade (Dejean et al. 2009). An elegant study used optogenetic and solid-state optics to show that stimulation or inhibition of STN neurons failed to improve parkinsonian rats but stimulation of afferent fibers of the STN produced marked improvement in motor signs (Gradinaru et al. 2009). Afferent fibers stimulated were the cortico-STN fibers rather than the GABAergic projections to the STN. Importantly, the stimulation of STN afferents antidromically activated layer V neurons in the M1 and selective high-frequency stimulation of layer V in the M1 also ameliorated parkinsonian symptoms (Gradinaru et al. 2009). Because MEPs generated by TMS are due to activation of corticospinal neurons in layer V of the M1, our findings are consistent STN DBS activating layer V neurons in the M1 at specific intervals. Therefore, the short latency–evoked potential and excitatory STN–motor cortical interaction at ∼3 ms demonstrated in our study could be related to antidromic activation of cortical circuits and desynchronization of abnormal noisy signals in PD. Previous studies showed that this potential can follow stimulation rates up to 100 Hz (Ashby et al. 2001; Baker et al. 2002) and may therefore be activated with high-frequency DBS used in the clinical setting.

We found medium latency STN–motor cortical facilitation following stimulation of clinically effective dorsal STN area contacts. This could be also due to activation of ZI and/or pallidothalamic fibers that lies close to the dorsal border of STN (Godinho et al. 2006). One study suggested that high-frequency stimulation of the ZI produced greater improvement in contralateral motor scores in PD patients than stimulation of the STN itself (Plaha et al. 2006). It has been postulated that in PD, ZI receives abnormal synchronized oscillations from basal ganglia output nuclei and motor cortex (Merello et al. 2006). The ZI is in a position to transmit these abnormal oscillations via its efferent connections to the thalamic nuclei (centromedian/parafascicular and ventral intermedius), the brainstem locomotor center and back to the basal ganglia output nuclei (GPi and SNr) and the cortex (Foffani et al. 2006). High-frequency stimulation of caudal ZI could potentially override these abnormal oscillations and lead to more profound effects in controlling PD symptoms than stimulation of the STN, whose efferents are predominantly confined to the basal ganglia output nuclei (Plaha et al. 2006).

Conclusions

STN DBS led to medium latency–evoked potentials observed mainly following stimulation of clinically effective contacts located near dSTN border. Short latency–evoked potentials were elicited from stimulation of contacts in both dorsal and ventral STN areas. STN stimulation using contacts that produced clinical benefit increased the excitability of the motor cortex at specific short and medium latencies. This is likely due to short-latency antidromic stimulation of corticosubthalamic projections and the medium-latency facilitatory basal ganglia–thalamo–cortical interactions following dSTN stimulation. These findings support animal studies (Gradinaru et al. 2009) suggesting that cortical activation could be one of the mechanisms mediating the clinical effects of STN DBS in PD.

Funding

Canadian Institutes of Health Research grant number MOP 15128 to Dr Chen; Canadian Institutes of Health Research—Industry Partnered (Medtronic) Investigator Award and the Catherine Manson Chair in Movement Disorders to Dr Chen; Canada Research Chair in Neurosciences to Dr Lozano.

Conflict of Interest: Dr Hamani acted as a consultant for St Jude Medical. Dr Moro received honorarium from Medtronic Inc for lecturing and consulting services. Dr Lozano acted as a consultant for Medtronic Inc. Dr Chen received research grant and acted as a consultant for Medtronic Inc.

References

Alexander
GE
Crutcher
MD
DeLong
MR
Basal ganglia-thalamocortical circuits: parallel substrates for motor, oculomotor, “prefrontal” and “limbic” functions
Prog Brain Res
 , 
1990
, vol. 
85
 (pg. 
119
-
146
)
Ashby
P
Kim
YJ
Kumar
R
Lang
AE
Lozano
AM
Neurophysiological effects of stimulation through electrodes in the human subthalamic nucleus
Brain
 , 
1999
, vol. 
122
 (pg. 
1919
-
1931
)
Ashby
P
Paradiso
G
Saint-Cyr
JA
Chen
R
Lang
AE
Lozano
AM
Potentials recorded at the scalp by stimulation near the human subthalamic nucleus
Clin Neurophysiol
 , 
2001
, vol. 
112
 (pg. 
431
-
437
)
Baker
KB
Montgomery
EB
Jr.
Rezai
AR
Burgess
R
Luders
HO
Subthalamic nucleus deep brain stimulus evoked potentials: physiological and therapeutic implications
Mov Disord
 , 
2002
, vol. 
17
 (pg. 
969
-
983
)
Bartho
P
Freund
TF
Acsady
L
Selective GABAergic innervation of thalamic nuclei from zona incerta
Eur J Neurosci
 , 
2002
, vol. 
16
 (pg. 
999
-
1014
)
Benazzouz
A
Piallat
B
Ni
ZG
Koudsie
A
Pollak
P
Benabid
AL
Implication of the subthalamic nucleus in the pathophysiology and pathogenesis of Parkinson's disease
Cell Transplant
 , 
2000
, vol. 
9
 (pg. 
215
-
221
)
Benazzouz
A
Tai
CH
Meissner
W
Bioulac
B
Bezard
E
Gross
C
High-frequency stimulation of both zona incerta and subthalamic nucleus induces a similar normalization of basal ganglia metabolic activity in experimental parkinsonism
FASEB J
 , 
2004
, vol. 
18
 (pg. 
528
-
530
)
Beurrier
C
Bioulac
B
Audin
J
Hammond
C
High-frequency stimulation produces a transient blockade of voltage-gated currents in subthalamic neurons
J Neurophysiol
 , 
2001
, vol. 
85
 (pg. 
1351
-
1356
)
Brown
P
Eusebio
A
Paradoxes of functional neurosurgery: clues from basal ganglia recordings
Mov Disord
 , 
2008
, vol. 
23
 (pg. 
12
-
20
)
Ceballos-Baumann
AO
Boecker
H
Bartenstein
P
von
F
I
Riescher
H
Conrad
B
Moringlane
JR
Alesch
F
A positron emission tomographic study of subthalamic nucleus stimulation in Parkinson disease: enhanced movement-related activity of motor-association cortex and decreased motor cortex resting activity
Arch Neurol
 , 
1999
, vol. 
56
 (pg. 
997
-
1003
)
Costa
J
Valls-Sole
J
Valldeoriola
F
Rumia
J
Tolosa
E
Motor responses of muscles supplied by cranial nerves to subthalamic nucleus deep brain stimuli
Brain
 , 
2007
, vol. 
130
 (pg. 
245
-
255
)
Cunic
D
Roshan
L
Khan
FI
Lozano
AM
Lang
AE
Chen
R
Effects of subthalamic nucleus stimulation on motor cortex excitability in Parkinson's disease
Neurology
 , 
2002
, vol. 
58
 (pg. 
1665
-
1672
)
Dauper
J
Peschel
T
Schrader
C
Kohlmetz
C
Joppich
G
Nager
W
Dengler
R
Rollnik
JD
Effects of subthalamic nucleus (STN) stimulation on motor cortex excitability
Neurology
 , 
2002
, vol. 
59
 (pg. 
700
-
706
)
Degos
B
Deniau
JM
Le Cam
J
Mailly
P
Maurice
N
Evidence for a direct subthalamo-cortical loop circuit in the rat
Eur J Neurosci
 , 
2008
, vol. 
27
 (pg. 
2599
-
2610
)
Dejean
C
Hyland
B
Arbuthnott
G
Cortical effects of subthalamic stimulation correlate with behavioral recovery from dopamine antagonist induced akinesia
Cereb Cortex
 , 
2009
, vol. 
19
 (pg. 
1055
-
1063
)
DeLong
MR
Primate models of movement disorders of basal ganglia origin
Trends Neurosci
 , 
1990
, vol. 
13
 (pg. 
281
-
285
)
Deuschl
G
Schade-Brittinger
C
Krack
P
Volkmann
J
Schafer
H
Botzel
K
Daniels
C
Deutschlander
A
Dillmann
U
Eisner
W
, et al.  . 
A randomized trial of deep-brain stimulation for Parkinson's disease
N Engl J Med
 , 
2006
, vol. 
355
 (pg. 
896
-
908
)
Di Lazzaro
V
Oliviero
A
Saturno
E
Pilato
F
Insola
A
Mazzone
P
Profice
P
Tonali
P
Rothwell
JC
, et al.  . 
The effect on corticospinal volleys of reversing the direction of current induced in the motor cortex by transcranial magnetic stimulation
Exp Brain Res
 , 
2001
, vol. 
138
 (pg. 
268
-
273
)
Dostrovsky
JO
Lozano
AM
Mechanisms of deep brain stimulation
Mov Disord
 , 
2002
, vol. 
17
 
Suppl. 3
(pg. 
S63
-
S68
)
Eusebio
A
Chen
CC
Lu
CS
Lee
ST
Tsai
CH
Limousin
P
Hariz
M
Brown
P
Effects of low-frequency stimulation of the subthalamic nucleus on movement in Parkinson's disease
Exp Neurol
 , 
2008
, vol. 
209
 (pg. 
125
-
130
)
Eusebio
A
Pogosyan
A
Wang
S
Averbeck
B
Gaynor
LD
Limousin
P
Cantiniaux
S
Witjas
T
Limousin
P
Azulay
JP
Brown
P
Resonance in subthalamo-cortical circuits in Parkinson's disease
Brain 132(Pt 8):
 , 
2009
(pg. 
2139
-
50
)
Filali
M
Hutchison
WD
Palter
VN
Lozano
AM
Dostrovsky
JO
Stimulation-induced inhibition of neuronal firing in human subthalamic nucleus
Exp Brain Res
 , 
2004
, vol. 
156
 (pg. 
274
-
281
)
Foffani
G
Ardolino
G
Egidi
M
Caputo
E
Bossi
B
Priori
A
Subthalamic oscillatory activities at beta or higher frequency do not change after high-frequency DBS in Parkinson's disease
Brain Res Bull
 , 
2006
, vol. 
69
 (pg. 
123
-
130
)
Fujimoto
K
Kita
H
Response characteristics of subthalamic neurons to the stimulation of the sensorimotor cortex in the rat
Brain Res
 , 
1993
, vol. 
609
 (pg. 
185
-
192
)
Gaynor
LM
Kuhn
AA
Dileone
M
Litvak
V
Eusebio
A
Pogosyan
A
Androulidakis
AG
Tisch
S
Limousin
P
Insola
A
, et al.  . 
Suppression of beta oscillations in the subthalamic nucleus following cortical stimulation in humans
Eur J Neurosci
 , 
2008
, vol. 
28
 (pg. 
1686
-
1695
)
Godinho
F
Thobois
S
Magnin
M
Guenot
M
Polo
G
Benatru
I
Xie
J
Salvetti
A
Garcia-Larrea
L
Broussolle
E
, et al.  . 
Subthalamic nucleus stimulation in Parkinson's disease: anatomical and electrophysiological localization of active contacts
J Neurol
 , 
2006
, vol. 
253
 (pg. 
1347
-
1355
)
Gradinaru
V
Mogri
M
Thompson
KR
Henderson
JM
Deisseroth
K
Optical deconstruction of parkinsonian neural circuitry
Science
 , 
2009
, vol. 
324
 (pg. 
354
-
359
)
Hamani
C
Richter
E
Schwalb
JM
Lozano
AM
Bilateral subthalamic nucleus stimulation for Parkinson's disease: a systematic review of the clinical literature
Neurosurgery
 , 
2005
, vol. 
56
 (pg. 
1313
-
1321
)
Hamani
C
Richter
E
Schwalb
JM
Lozano
AM
Bilateral subthalamic nucleus stimulation for Parkinson's disease: a systematic review of the clinical literature
Neurosurgery
 , 
2008
, vol. 
62
 
Suppl. 2
(pg. 
863
-
874
)
Hamani
C
Saint-Cyr
JA
Fraser
J
Kaplitt
M
Lozano
AM
The subthalamic nucleus in the context of movement disorders
Brain
 , 
2004
, vol. 
127
 (pg. 
4
-
20
)
Hamel
W
Fietzek
U
Morsnowski
A
Schrader
B
Weinert
D
Muller
D
Deuschl
G
Mehdorn
HM
Subthalamic nucleus stimulation in Parkinson's disease: correlation of active electrode contacts with intraoperative microrecordings
Stereotact Funct Neurosurg
 , 
2003
, vol. 
80
 (pg. 
37
-
42
)
Hanajima
R
Ashby
P
Lozano
AM
Lang
AE
Chen
R
Single pulse stimulation of the human subthalamic nucleus facilitates the motor cortex at short intervals
J Neurophysiol
 , 
2004
, vol. 
92
 (pg. 
1937
-
1943
)
Hanajima
R
Ugawa
Y
Terao
Y
Enomoto
H
Shiio
Y
Mochizuki
H
Furubayashi
T
Uesugi
H
Iwata
NK
Kanazawa
I
Mechanisms of intracortical I-wave facilitation elicited with paired-pulse magnetic stimulation in humans
J Physiol
 , 
2002
, vol. 
538
 (pg. 
253
-
261
)
Hanajima
R
Ugawa
Y
Terao
Y
Sakai
K
Furubayashi
T
Machii
K
Kanazawa
I
Paired-pulse magnetic stimulation of the human motor cortex: differences among I waves
J Physiol
 , 
1998
, vol. 
509
 (pg. 
607
-
618
)
Hashimoto
T
[Pathophysiology of abnormal movements in Parkinson's disease]
Nippon Rinsho
 , 
2000
, vol. 
58
 (pg. 
1994
-
1999
)
Hazrati
LN
Pinault
D
Parent
A
The thalamic reticular nucleus does not send commissural projection to the contralateral parafascicular nucleus in the rat
Brain Res
 , 
1995
, vol. 
679
 (pg. 
123
-
134
)
Hidding
U
Baumer
T
Siebner
HR
Demiralay
C
Buhmann
C
Weyh
T
Moll
C
Hame
W
Munchau
A
MEP latency shift after implantation of deep brain stimulation systems in the subthalamic nucleus in patients with advanced Parkinson's disease
Mov Disord
 , 
2006
, vol. 
21
 (pg. 
1471
-
1476
)
Hutchison
WD
Allan
RJ
Opitz
H
Levy
R
Dostrovsky
JO
Lang
AE
Lozano
AM
Neurophysiological identification of the subthalamic nucleus in surgery for Parkinson's disease
Ann Neurol
 , 
1998
, vol. 
44
 (pg. 
622
-
628
)
Inase
M
Tokuno
H
Nambu
A
Akazawa
T
Takada
M
Corticostriatal and corticosubthalamic input zones from the presupplementary motor area in the macaque monkey: comparison with the input zones from the supplementary motor area
Brain Res
 , 
1999
, vol. 
833
 (pg. 
191
-
201
)
Jahnsen
H
Llinas
R
Electrophysiological properties of guinea-pig thalamic neurones: an in vitro study
J Physiol
 , 
1984
, vol. 
349
 (pg. 
205
-
226
)
Jahnsen
H
Llinas
R
Ionic basis for the electro-responsiveness and oscillatory properties of guinea-pig thalamic neurones in vitro
J Physiol
 , 
1984
, vol. 
349
 (pg. 
227
-
247
)
Kaneko
K
Kawai
S
Fuchigami
Y
Morita
H
Ofuji
A
The effect of current direction induced by transcranial magnetic stimulation on the corticospinal excitability in human brain
Electroencephalogr Clin Neurophysiol
 , 
1996
, vol. 
101
 (pg. 
478
-
482
)
Krack
P
Batir
A
Van Blercom
N
Chabardes
S
Fraix
V
Ardouin
C
Koudsie
A
Limousin
PD
Benazzouz
A
LeBas
JF
, et al.  . 
Five-year follow-up of bilateral stimulation of the subthalamic nucleus in advanced Parkinson's disease
N Engl J Med
 , 
2003
, vol. 
349
 (pg. 
1925
-
1934
)
Kuhn
AA
Trottenberg
T
Kupsch
A
Meyer
BU
Pseudo-bilateral hand motor responses evoked by transcranial magnetic stimulation in patients with deep brain stimulators
Clin Neurophysiol
 , 
2002
, vol. 
113
 (pg. 
341
-
345
)
Kumar
R
Chen
R
Ashby
P
Safety of transcranial magnetic stimulation in patients with implanted deep brain stimulators
Mov Disord
 , 
1999
, vol. 
14
 (pg. 
157
-
158
)
Kunzle
H
An autoradiographic analysis of the efferent connections from premotor and adjacent prefrontal regions (areas 6 and 9) in macaca fascicularis
Brain Behav Evol
 , 
1978
, vol. 
15
 (pg. 
185
-
234
)
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
 , 
2002
, vol. 
125
 (pg. 
1196
-
1209
)
Li
S
Arbuthnott
GW
Jutras
MJ
Goldberg
JA
Jaeger
D
Resonant antidromic cortical circuit activation as a consequence of high-frequency subthalamic deep-brain stimulation
J Neurophysiol
 , 
2007
, vol. 
98
 (pg. 
3525
-
3537
)
MacKinnon
CD
Webb
RM
Silberstein
P
Tisch
S
Asselman
P
Limousin
P
Rothwell
JC
Stimulation through electrodes implanted near the subthalamic nucleus activates projections to motor areas of cerebral cortex in patients with Parkinson's disease
Eur J Neurosci
 , 
2005
, vol. 
21
 (pg. 
1394
-
1402
)
Magarinos-Ascone
C
Riva-Meana
C
Figueiras-Mendez
R
[Neuronal activity in the subthalamic nucleus in Parkinson disease]
Rev Neurol
 , 
2000
, vol. 
31
 (pg. 
66
-
71
)
Magnin
M
Morel
A
Jeanmonod
D
Single-unit analysis of the pallidum, thalamus and subthalamic nucleus in parkinsonian patients
Neuroscience
 , 
2000
, vol. 
96
 (pg. 
549
-
564
)
Maurice
N
Deniau
JM
Glowinski
J
Thierry
AM
Relationships between the prefrontal cortex and the basal ganglia in the rat: physiology of the corticosubthalamic circuits
J Neurosci
 , 
1998
, vol. 
18
 (pg. 
9539
-
9546
)
Maurice
N
Deniau
JM
Menetrey
A
Glowinski
J
Thierry
AM
Prefrontal cortex-basal ganglia circuits in the rat: involvement of ventral pallidum and subthalamic nucleus
Synapse
 , 
1998
, vol. 
29
 (pg. 
363
-
370
)
McIntyre
CC
Mori
S
Sherman
DL
Thakor
NV
Vitek
JL
Electric field and stimulating influence generated by deep brain stimulation of the subthalamic nucleus
Clin Neurophysiol
 , 
2004
, vol. 
115
 (pg. 
589
-
595
)
McIntyre
CC
Savasta
M
Kerkerian-Le Goff
L
Vitek
JL
Uncovering the mechanism(s) of action of deep brain stimulation: activation, inhibition, or both
Clin Neurophysiol
 , 
2004
, vol. 
115
 (pg. 
1239
-
1248
)
McIntyre
CC
Savasta
M
Walter
BL
Vitek
JL
How does deep brain stimulation work? Present understanding and future questions
J Clin Neurophysiol
 , 
2004
, vol. 
21
 (pg. 
40
-
50
)
Merello
M
Tenca
E
Cerquetti
D
Neuronal activity of the zona incerta in Parkinson's disease patients
Mov Disord
 , 
2006
, vol. 
21
 (pg. 
937
-
943
)
Molnar
GF
Sailer
A
Gunraj
CA
Cunic
DI
Lang
AE
Lozano
AM
Moro
E
Chen
R
Changes in cortical excitability with thalamic deep brain stimulation
Neurology
 , 
2005
, vol. 
64
 (pg. 
1913
-
1919
)
Monakow
KH
Akert
K
Kunzle
H
Projections of the precentral motor cortex and other cortical areas of the frontal lobe to the subthalamic nucleus in the monkey
Exp Brain Res
 , 
1978
, vol. 
33
 (pg. 
395
-
403
)
Montgomery
EB
Baker
KB
Mechanisms of deep brain stimulation and future technical developments
Neurol Res
 , 
2000
, vol. 
22
 (pg. 
259
-
266
)
Nambu
A
Takada
M
Inase
M
Tokuno
H
Dual somatotopical representations in the primate subthalamic nucleus: evidence for ordered but reversed body-map transformations from the primary motor cortex and the supplementary motor area
J Neurosci
 , 
1996
, vol. 
16
 (pg. 
2671
-
2683
)
Nambu
A
Tokuno
H
Hamada
I
Kita
H
Imanishi
M
Akazawa
T
Ikeuchi
Y
Hasegawa
N
Excitatory cortical inputs to pallidal neurons via the subthalamic nucleus in the monkey
J Neurophysiol
 , 
2000
, vol. 
84
 (pg. 
289
-
300
)
Nambu
A
Tokuno
H
Inase
M
Takada
M
Corticosubthalamic input zones from forelimb representations of the dorsal and ventral divisions of the premotor cortex in the macaque monkey: comparison with the input zones from the primary motor cortex and the supplementary motor area
Neurosci Lett
 , 
1997
, vol. 
239
 (pg. 
13
-
16
)
Nishimoto
A
Matsumoto
K
Bilateral, three negative, cortical evoked potentials after unilateral stimulation of the thalamic VL nucleus in man
Confin Neurol
 , 
1970
, vol. 
32
 (pg. 
349
-
366
)
Nowak
LG
Bullier
J
Axons, but not cell bodies, are activated by electrical stimulation in cortical gray matter. II. Evidence from selective inactivation of cell bodies and axon initial segments
Exp Brain Res
 , 
1998
, vol. 
118
 (pg. 
489
-
500
)
Parent
A
Hazrati
LN
Functional anatomy of the basal ganglia. II. The place of subthalamic nucleus and external pallidum in basal ganglia circuitry
Brain Res Brain Res Rev
 , 
1995
, vol. 
20
 (pg. 
128
-
154
)
Parent
A
Sato
F
Wu
Y
Gauthier
J
Levesque
M
Parent
M
Organization of the basal ganglia: the importance of axonal collateralization
Trends Neurosci
 , 
2000
, vol. 
23
 (pg. 
S20
-
S27
)
Pierantozzi
M
Palmieri
MG
Mazzone
P
Marciani
MG
Rossini
PM
Stefani
A
Giacomini
P
Peppe
A
Stanzione
P
Deep brain stimulation of both subthalamic nucleus and internal globus pallidus restores intracortical inhibition in Parkinson's disease paralleling apomorphine effects: a paired magnetic stimulation study
Clin Neurophysiol
 , 
2002
, vol. 
113
 (pg. 
108
-
113
)
Plaha
P
Ben Shlomo
Y
Patel
NK
Gill
SS
Stimulation of the caudal zona incerta is superior to stimulation of the subthalamic nucleus in improving contralateral parkinsonism
Brain
 , 
2006
, vol. 
129
 (pg. 
1732
-
1747
)
Rodriguez-Oroz
MC
Obeso
JA
Lang
AE
Houeto
JL
Pollak
P
Rehncrona
S
Kulisevsky
J
Albanese
A
Volkmann
J
Hariz
MI
, et al.  . 
Bilateral deep brain stimulation in Parkinson's disease: a multicentre study with 4 years follow-up
Brain
 , 
2005
, vol. 
128
 (pg. 
2240
-
2249
)
Rossi
S
Hallett
M
Rossini
PM
Pascual-Leone
A
of TMS Consensus Group 2009. Safety ethical considerations, and application guidelines for the use of transcranial magnetic stimulation in clinical practice and research
Clin Neurophysiol. 120(12):2008–2039
 
Saint-Cyr
JA
Hoque
T
Pereira
LC
Dostrovsky
JO
Hutchison
WD
Mikulis
DJ
Abosch
A
Sime
E
Lang
AE
Lozano
AM
Localization of clinically effective stimulating electrodes in the human subthalamic nucleus on magnetic resonance imaging
J Neurosurg
 , 
2002
, vol. 
97
 (pg. 
1152
-
1166
)
Sakai
K
Ugawa
Y
Terao
Y
Hanajima
R
Furubayashi
T
Kanazawa
I
Preferential activation of different I waves by transcranial magnetic stimulation with a figure-of-eight-shaped coil
Exp Brain Res
 , 
1997
, vol. 
113
 (pg. 
24
-
32
)
Schaltenbrand
G
Spuler
H
Wahren
W
Rumler
B
Electroanatomy of the thalamic ventro-oral nucleus based on stereotaxic stimulation in man
Z Neurol
 , 
1971
, vol. 
199
 (pg. 
259
-
276
)
Strafella
AP
Vanderwerf
Y
Sadikot
AF
Transcranial magnetic stimulation of the human motor cortex influences the neuronal activity of subthalamic nucleus
Eur J Neurosci
 , 
2004
, vol. 
20
 (pg. 
2245
-
2249
)
Temperli
P
Ghika
J
Villemure
JG
Burkhard
PR
Bogousslavsky
J
Vingerhoets
FJ
How do parkinsonian signs return after discontinuation of subthalamic DBS?
Neurology
 , 
2003
, vol. 
60
 (pg. 
78
-
81
)
Tisch
S
Rothwell
JC
Zrinzo
L
Bhatia
KP
Hariz
M
Limousin
P
Cortical evoked potentials from pallidal stimulation in patients with primary generalized dystonia
Mov Disord
 , 
2008
, vol. 
23
 (pg. 
265
-
273
)
Valls-Sole
J
Compta
Y
Costa
J
Valldeoriola
F
Rumia
J
Human central nervous system circuits examined through the electrodes implanted for deep brain stimulation
Clin Neurophysiol
 , 
2008
, vol. 
119
 (pg. 
1219
-
1231
)
Vitek
JL
Mechanisms of deep brain stimulation: excitation or inhibition
Mov Disord
 , 
2002
, vol. 
17
 
Suppl. 3
(pg. 
S69
-
S72
)
Wang
C
Zucker
RS
Regulation of synaptic vesicle recycling by calcium and serotonin
Neuron
 , 
1998
, vol. 
21
 (pg. 
155
-
167
)
Yelnik
J
Percheron
G
Subthalamic neurons in primates: a quantitative and comparative analysis
Neuroscience
 , 
1979
, vol. 
4
 (pg. 
1717
-
1743
)
Zucker
RS
Regehr
WG
Short-term synaptic plasticity
Annu Rev Physiol
 , 
2002
, vol. 
64
 (pg. 
355
-
405
)