Subthalamic Neuronal Firing in Obsessive-Compulsive Disorder and Parkinson Disease
Although electrophysiologic dysfunction of the subthalamic nucleus is putative, deep brain stimulation of this structure has recently been reported to improve obsessions and compulsions. In Parkinson disease, sensorimotor subthalamic neurons display high-frequency burst firing, which is considered as an electrophysiologic signature of motor loop dysfunction. We addressed whether such neuronal dysfunction of the subthalamic nucleus also exists in the nonmotor loops involved in patients with obsessive-compulsive disorder. We compared the neuronal activity of the subthalamic nucleus recorded in 9 patients with obsessive-compulsive disorder with that of 11 patients with Parkinson disease measured during intraoperative exploration for deep brain stimulation. The mean subthalamic neuron discharge rate was statistically lower in patients with obsessive-compulsive disorder than in patients with Parkinson disease (20.5 ± 11.0 Hz, n = 100 and 30.8 ± 15.6 Hz, n = 93, respectively, p < 0.001). The relative proportion of burst neurons did not differ significantly between the 2 diseases (75% vs 73%). Interestingly, burst neurons were predominantly left-sided in obsessive-compulsive disorder. The recording of burst neurons within the nonmotor subthalamic nucleus in patients with obsessive-compulsive disorder is a novel finding that suggests the existence of deregulation of the nonmotor basal ganglia loop, possibly left-sided. Potentially, burst activity might interfere with normal processes occurring within nonmotor loops.
Subthalamic Neuronal Firing in Obsessive-
Compulsive Disorder and
Brigitte Piallat, PhD,
Mircea Polosan, MD,
rie Fraix, MD, PhD,
Laurent Goetz, MSc,
Olivier David, PhD,
Albert Fenoy, MD,
Napoleon Torres, MD, PhD,
Jean-Louis Quesada, MSc,
Eric Seigneuret, MD,
Pierre Pollak, MD,
Paul Krack, MD, PhD,
Thierry Bougerol, MD,
Alim L. Benabid, MD, PhD,
Objective: Although electrophysiologic dysfunction of the subthalamic nucleus is putative, deep brain stimulation of
this structure has recently been reported to improve obsessions and compulsions. In Parkinson disease, sensorimotor
subthalamic neurons display high-frequency burst firing, which is considered as an electrophysiologic signature of
motor loop dysfunction. We addressed whether such neuronal dysfunction of the subthalamic nucleus also exists in
the nonmotor loops involved in patients with obsessive-compulsive disorder.
Methods: We compared the neuronal activity of the subthalamic nucleus recorded in 9 patients with obsessive-
compulsive disorder with that of 11 patients with Parkinson disease measured during intraoperative exploration for
deep brain stimulation.
Results: The mean subthalamic neuron discharge rate was statistically lower in patients with obsessive-compulsive
disorder than in patients with Parkinson disease (20.5 6 11.0 Hz, n ¼ 100 and 30.8 6 15.6 Hz, n ¼ 93, respectively,
p < 0.001). The relative proportion of burst neurons did not differ significantly between the 2 diseases (75% vs 73%).
Interestingly, burst neurons were predominantly left-sided in obsessive-compulsive disorder.
Interpretation: The recording of burst neurons within the nonmotor subthalamic nucleus in patients with obsessive-
compulsive disorder is a novel finding that suggests the existence of deregulation of the nonmotor basal ganglia
loop, possibly left-sided. Potentially, burst activity might interfere with normal processes occurring within nonmotor
ANN NEUROL 2010;00:000–000
bsessive-compulsive disorder (OCD) is a relatively
common psychiatric disease, affecting 2 to 3% of
It is characterized by recurrent
unwanted ideas, images, or impulses (obsessions), and re-
petitive stereotyped behaviors or mental acts (compul-
sions), often intended to neutralize the anxiety induced
by the obsessions. Despite pharmacologic treatment and
cognitive behavioral therapy, approximately 20% of
patients with OCD remain severely disabled.
lines of evidence support the concept that OCD pathoge-
nesis may result from dysfunction in the frontobasal gan-
glia-thalamocortical loops, because OCD symptoms have
been documented in various diseases involving striatal
and neuroimaging of OCD has shown
significant structural and functional abnormalities within
the orbitofrontal cortex and the striatum, which partly
respond to pharmacologic and cognitive behavioral ther-
The subthalamic nucleus (STN), a key structure
View this article online at wileyonlinelibrary.com. DOI: 10.1002/ana.22222
Received Mar 29, 2010, and in revised form Aug 4, 2010. Accepted for publication Aug 6, 2010.
Address correspondence to Pr Chabarde
s, Grenoble Institut des Neurosciences, INSERM-U836, Equipe 11, Universite
Joseph Fourier, Site Sante, La
Tronche, BP170, 38042 Grenoble ce
dex 9, France. E-mail: SChabardes@chu-grenoble.fr
Grenoble University, France;
INSERM-U836, Grenoble Institute of Neurosciences, France;
Department of Psychiatry,
Department of Neurology,
Department of Neuroradiology,
Department of Neurosurgery,
Center of Clinical Investigation and INSERM-003, University Hospital of Grenoble, France;
Atomic Energy Commission of Grenoble, France.
Additional Supporting Information may be found in the online version of this article.
2010 American Neurological Association 1
of the frontobasal ganglia-thalamocortical motor loop,
also seems crucial for the nonmotor (noM) parallel
loops that are involved in the emotional and cognitive
components of behavior.
Furthermore, acute STN
deep brain stimulation (DBS) can induce mood changes
in patients with Parkinson disease (PD),
the role of STN in emotional behavior. In monkeys,
anatomic studies have shown that the cortical areas
involved in motor, cognitive, and emotional functions
are processed in distinct territories within the STN
large dorsolateral area receives sensorimotor (SM) input
and a noM area, encompassing ventromedial and ante-
romedial parts, receives associative and limbic system
Lesional surgery or DBS of the cingulum,
the limbic sys-
and the nucleus accumbens
have been considered
as therapeutic options in patients with OCD. Recently, the
DBS in OCD study group (ST OC) found that STN-DBS
markedly improved OCD symptoms.
However , the sub-
strate of the involvement of the STN in OCD remains
unknown. In patients with PD and in monkeys rendered
parkinsonian, clinical efficacy of STN-DBS suggests the
key role of SM-STN in the pathophysiology of motor
Furthermore, the depletion of striatal dopa-
mine is related to changes in the firing rate (FR) and pat-
tern of STN neurons that become hyperactive, exhibiting
and potentially contributing to the
appearance of motor symptoms. If cognitive or emotional
loops are involved in OCD, an electrophysiologic signature
should be found in the noM-STN as well. To address this
hypothesis, we compared the electrophysiologic properties
of noM-STN neurons in patients with OCD to those of
SM-STN neurons in patients with PD.
Patients and Methods
Ten patients with OCD were initially recruited for this study,
but 1 was excluded because microrecordings were not amenable
to analysis. Ultimately, 7 women and 2 men with OCD, aged
34 to 52 years (mean, 39.5 6 5.5 years), and 3 women and 8
men with PD, aged 28 to 68 years (mean, 54.3 6 15.3 years)
were included in the study. We retrospectively analyzed data
gathered during routine microrecording performed for the im-
plantation of DBS leads into the STN, using the same proce-
dures conducted at the University Hospital of Grenoble
the last 17 years. Patients with OCD were all referred to our
center after unsuccessful medical and cognitive behavioral
TABLE 1: Clinical Characteristics of Patients with Obsessive-Compulsive Disorder at Time of Surgery
yr YBOCS GAF CGI
Medication, Total Daily Dose, mg
SNRI and SRI FGA and SGA Other
O1 M 39 18 18þ19 26 6 Fluvoxamine 200 Pimozide 1 Clonidine 0.3
O2 F 43 32 14þ18 32 6 Sertraline 100 Risperidone 4 Alprazolam 0.75
O3 F 42 25 15þ15 36 5 Fluoxetine 60,
O4 F 34 24 17þ17 39 5 Escitalopram 30 Clobazam 5;
O5 F 35 15 14þ15 36 6 Paroxetine 60
O6 F 37 5 14þ18 33 6 Venlafaxine 150 Bromazepam 9,
O7 F 52 25 20þ20 33 6 Citalopram 20,
Cyamemazine 50 Clobazam 20,
O8 F 38 11 18þ18 32 6 Sertraline 100 Clonazepam 2
O9 M 36 17 17þ15 40 5 Paroxetine 40
YBOCS scores range from 0 to 40, with higher scores indicating worse function. The 2 YBOCS subscores (obsession þ compul-
sion) range from 0 to 20. GAF scores range from 1 to 90, with higher scores indicating the normal global functional status. CGI
scores range from 1 to 7, with higher scores indicating the severity of the disease.
YBOCS ¼ Yale-Brown Obsessive Compulsive Scale; GAF ¼ Global Assessment of Functioning; CGI ¼ Clinical Global Impres-
sion; SNRI ¼ serotonin and norepinephrine reuptake inhibitors; SRI ¼ serotonin reuptake inhibitors; FGA ¼ first-generation
antipsychotics; SGA, second-generation antipsychotics.
ANNALS of Neurology
2 Volume 000, No. 000
therapy. All patients gave informed consent. We followed strict
ethical guidelines given by the Comite
and we applied established inclusion criteria for
DBS in OCD.
Tables 1 and 2 summarize the clinical characteristics of
patients. For patients with OCD, all right-handed, the primary
diagnosis was based on the Diagnostic and Statistical Manual of
Mental Disorders-IV criteria. The Yale-Brown Obsessive Com-
score was >25, the Global Assessment Function-
ing was <45, and the Clinical Global Impression (severity) was
>4. All 9 patients with OCD were resistant to adequate drug
trials (>12 weeks at the maximum tolerated dose) including at
least 3 serotonin reuptake inhibitors, clomipramine and lith-
ium, clonazepam, and buspirone and pimozide, and at least 1
year of well-conducted cognitive behavioral therapy. Patients
were referred for bilateral STN-DBS electrode implantation.
Three patients were also included in the recent French prospec-
tive study on bilateral STN-DBS for OCD (STOC study).
With the exception of 1 patient (O1) who suffered from
comorbid Tourette syndrome and another patient (O2) with a
very mild akinesia as a side effect of neuroleptic treatment
(Unified Parkinson’s Disease Rating Scale [UPDRS] part III ¼
3/108), patients with OCD did not suffer from PD or abnor-
mal movements (UPDRS part III ¼ 0/108).
For patients with PD, the mean disease duration was
>10 years. The mean motor UPDRS off-state was 45.4 6
15.3/108. All PD patients suffered from motor complications
of dopaminergic treatment (see Table 2; mean L-dopa equiva-
lent daily dose of 1,302.0 6 658.4mg/d; mean UPDRS items
32 þ 33 was 3.2 6 1.6/8). They did not suffer from severe
cognitive impairment, defined objectively by scores >130/144
on the Mattis dementia rating scale score. Medical treatment
was withheld for at least 12 hours before surgery.
Surgery and Targeting
Targeting was based on stereotactic T1 and T2 magnetic reso-
nance imaging scans fused with televentriculography (Fig 1;
In patients with PD, the SM-STN was targeted: 6/12 of the
anterior-posterior commissural (AC-PC) length posterior to the
AC, 12mm lateral to the midline, and 3mm below the AC-PC line.
In patients with OCD, the presumed noM-STN was targeted 2mm
anterior and 1mm medial to the previous target. It was then
adjusted during surgical implantation by using multiunit recordings
with 5 microelectrodes. In patients with PD, the SM-STN was
identified as the region in which neurons responded to passive or
active movements and where microstimulation induced improve-
ment of rigidity or tremor or both. Based on that definition, the
most anterior microelectrode was called the noM microelectrode,
because it explored the presumed noM-STN. In patients with
OCD, the noM-STN was identified by the lack of neuronal
responses to passive or active movements and by the lack of SM
side effects such as contralateral paresthesias or motor contractions
induced by microstimulations. In patients with OCD, the most
posterior microelectrode was called the SM microelectrode,becauseit
explored the presumed SM-STN (see F ig 1D).
Microrecording Sessions and Offline Analysis
Under local anesthesia, 5 microelectrodes (2mm apart) were
lowered into the brain using a motorized microdrive (Alpha-
Omega, Nazareth, Israel) on a stereotactic robotic arm (Neuro-
mate, Schaerer-Mayfield, Bron, France). Single-unit events were
discriminated offline using template-matching spike-sorting
software (Spike 2, CED, Cambridge, United Kingdom).
average neuronal FR was first calculated. Second, neurons were
classified as bursts or irregular as follows. Burst activity showed
a wide or bimodal interval interspike distribution and a
TABLE 2: Clinical Characteristics of Patients with Parkinson Disease at Time of Surgery
Sex Age, yr Duration of
(item 32 1 33)
Daily Dosage, mg
P1 M 57 18 69 1 3,000
P2 F 66 12 52 4 384
P3 M 54 7 27 5 1,060
P4 M 52 8 40 0 1,390
P5 M 68 14 37 4 860
P6 M 28 3 28 4 1,200
P7 F 61 13 43 4 1,495
P8 M 48 9 60 4 1,350
P9 M 56 9 26 2 1,084
P10 M 60 14 64 5 900
P11 F 48 13 54 3 1,600
UPDRS ¼ Unified Parkinson’s Disease Rating Scale.
Piallat et al: STN Firing in OCD and PD
Month, 2010 3
significant single peak on the autocorrelation function
2A, B). Irregular activity was characterized by a wide interval
interspike distribution and a flat autocorrelogram (see Fig
The Supporting Information provides details.
Radiographic Assessment of Microelectrode
Positions and 3-Dimensional Reconstruction
of Recording Sites
Microelectrodes were localized at the beginning and end of the
trajectory using a flat detector-based x-ray system (BioScan,
Geneva, Switzerland) that allowed the 3-dimensional (3-D)
reconstruction of the trajectory of each microelectrode, in rela-
tion to the ventricular landmarks (AC, PC, height of the thala-
mus). The x (anteroposterior), y (lateral), and z (depth) coordi-
nates of each recording site were calculated based on the
distance covered with the microdrive and the corresponding 3-
D trajectory. They were then normalized in the bicommissural
system and plotted into the Morel stereotactic atlas.
Statistical analyses were performed using Prism (GraphPad Soft-
ware, La Jolla, CA). Quantitative data were summarized as
mean 6 standard deviation or median and the 25th and 75th
percentiles. The Mann-Whitney U test was used to compare
continuous data. Chi-square analysis was used to test for the
association between the proportion of burst neurons and the
diseases, the hemispheres, and the SM/noM-STN in OCD and
PD. Therefore, p values were Bonferroni corrected for multiple
testing of 5 proportions of burst neurons. The level of signifi-
cance for all statistical tests was defined as p < 0.05 (2-tailed).
FRs and Firing Patterns
We recorded 193 cells (OCD, n ¼ 100; PD, n ¼ 93).
The mean recording duration was 72.4 seconds, with a
mean of 1585 events. The mean FR in OCD was signifi-
cantly lower than that in PD (20.5 6 11.0Hz vs 30.8 6
15.6Hz, respectively, p < 0.001; Fig 3A). Burst and
irregular STN neurons were clearly identified (see Fig 2)
in proportions that did not differ significantly between
OCD and PD (p ¼ 0.76; see Fig 3B). Most neurons in
OCD and PD exhibited burst firing (75% and 73%,
respectively; see Fig 3B). In OCD, the proportion of
burst neurons recorded within the noM-STN reached
83% and was significantly higher than the proportion of
burst neurons recorded by the SM microelectrode (50%,
p ¼ 0.02 after Bonferroni correction). Conversely, in
FIGURE 1: (A) Schematic representation of the sensorimotor
(SM)-subthalamic nucleus (STN) target used for Parkinson
disease (PD) surgery (blue spot) and the nonmotor (noM)-STN
target used for obsessive-compulsive disorder (OCD) (red
spot). Targets are plotted onto a coronal T2 magnetic
resonance imaging (MRI) scan superimposed with an
anteroposterior x-ray ventriculogram. The same targets are
represented on a parasagittal T1 MRI scan (B) and a lateral
x-ray ventriculogram (C). (D) Schema of the position of the 5
microelectrodes used for neuronal recording in the noM-
STN (red part) and the SM-STN (blue part) reported on the
corresponding horizontal section of the Morel stereotactic
atlas. Note that the noM microelectrode explored the noM-
STN in patients with PD and that the SM microelectrode
explored the SM-STN in patients with OCD. The target used
for OCD is 1mm medial and 2mm anterior to that used for
PD. AC 5 anterior commissure; PC 5 posterior commissure;
RN 5 red nucleus.
ANNALS of Neurology
4 Volume 000, No. 000
PD, the proportion of burst neurons recorded from the
SM-STN reached 81% and was higher than the propor-
tion of burst neurons recorded with the noM microelec-
trode (55%, p ¼ 0.05 after Bonferroni correction). The
burst characteristics differed significantly between OCD
and PD (Table 3). Burst duration was significantly lower
and the number of spikes per burst was higher in PD as
compared to OCD, and intraburst frequen cy was signifi-
cantly higher in PD than in OCD.
Lateralization of Neuronal Activity
Because of targeting strategy, recorded neurons in
patients with PD were mostly located in the dorsolateral
part, whereas those recorded in patients with OCD were
more ventral, medial, and anterior. In patients with PD,
burst neurons were distributed symmetrically and homo-
genously (70% in the right and 75% in the left side,
p ¼ 0.64; see Figs 3C and 4). In patients with OCD,
burst neurons were located in the ventromedial part of
the STN in an asymmetrical distribution, with 55/66
(83%) burst neurons on the left side compared to 20/34
(59%) on the right side (p ¼ 0.03 after Bonferroni cor-
rection). The mean FR of irregular and burst neurons
did not differ significantly between left and right sides
(all OCD, 22.2 6 11.4Hz right vs 19.6 6 10.7Hz left,
p ¼ 0.25; all PD, 31.4 6 15.9Hz right vs 29.9 6
15.3Hz left, p ¼ 0.53; burst OCD, 22.4 6 12.1Hz right
vs 19.3 6 10.8Hz left, p ¼ 0.35; burst PD, 29.2 61
3.6Hz right vs 30.8 6 4.9Hz left, p ¼ 0.76).
From intraoperative recordings performed during DBS
implantation, we report here the properties of STN neu-
rons in patients with severe OCD compared to patients
with PD. Three main characteristics distinguished OCD
from PD neurons: (1) a lower spontaneous FR, (2) a
burst pattern with a lower intraburst frequency, and (3) a
left-sided burst distribution. Although these neurons
were not recorded in the same STN subdivisions, these
electrophysiologic findings suggest noM-STN dysfunc-
tion in OCD.
FR and Firing Pattern
In line with previous studies,
we observed a
high incidence of burst activity in PD. We measured a
mean FR of 30Hz, which is lower than the ranges of 33
to 65Hz already reported, possibly because of our more
restrictive methodology for spike detection (see Support-
ing Information). It remained, however, above values
reported in normal monkeys
(19Hz) and in patients
with essential tremor
(19Hz), which are considered as
nonparkinsonian controls. The way by which we proc-
essed PD data is thus validated by the knowledge that
dopamine depletion significantly increases neuronal FR
along with the development of abnormal burst patterns
FIGURE 2: Example of electrophysiologic characteristics of
subthalamic nucleus neurons from patients with obsessive
compulsive disorder (OCD) and Parkinson’s Disease (PD).
The sorted spikes are displayed above the spike train with
burst detection displayed only for burst neurons. The raw
data, interval interspike (ISI) histograms, and
autocorrelograms (AC) are displayed beneath the spike
train. The firing patterns were determined on the basis of
ISI and AC, calculated for 1 second (1-millisecond bin
width). The AC confidence intervals (5% and 95%) were
calculated with the null assumption that the number of
spikes in any bin in the histogram should fit an independent
Poisson process. Burst neurons were characterized by a
wide or bimodal ISI distribution and by a significant single
peak on the AC in patients (A) with PD and (B) with OCD
(C) Irregular firing (obtained from patient with OCD) was
characterized by a wide ISI distribution and a flat AC.
Piallat et al: STN Firing in OCD and PD
Month, 2010 5
in the STN.
Interestingly, in OCD, the mean FR of
STN neurons (20Hz) was significantly lower than that in
PD patients (by 55%) and similar to assumed normal
values reported in the above-referenced movement disor-
ders studies (19Hz).
However, in the absence of con-
trol data and animal models of OCD, it is difficult to
firmly conclude whether the mean FR in patients with
OCD is normal. Nonetheless, an increased mean FR is
unlikely, because no dopaminergic depletion occurs in
OCD. Conversely, the possibility that the lower mean
FR in OCD is due to a regional effect is also not sup-
ported by our data, which did not show any anatomic
differences in mean FR along the 5 microelectrodes that
sampled large portions of SM-STN in OCD and of
noM-STN in PD. This is in accordance with a recent
study in which no correlation between the recording site
and the FR was found in patients with PD using com-
plete mapping of the STN, from 0 to 7mm below AC-
PC and from 9 to 15mm lateral relative to the midline.
In addition, no regional difference in FR was reported in
a study largely sampling the STN in parkinsonian mon-
keys (disease induced with 1-methyl-2-phenyl-1,2,3,6-tet-
rahydropyridine) and normal monkeys.
we found burst neurons in most anterior STN (noM-
STN) in OCD and posterior STN (SM-STN) in PD.
The ratio of burst neurons was equivalent in OCD and
PD and significantly larger (about 35%) than ratios
reported in patients with essential tremor. Both findings
suggest that the anatomically increased distribution of
burst neurons may follow the STN functional organiza-
tion and, therefore, that those neurons may be good
markers of specific deregulation of basal ganglia loops
associated with specific symptoms. At least in PD, this
functional specialization is supported by STN recordings
in patients with PD, which showed that no neuron
located in the ventral segment of the STN exhibited SM
responses, in contrast to the SM-STN located in the dor-
of the nucleus.
In addition, several ana-
tomic and physiologic studies from the monkey suggest
this subdivision with a dorsolateral part of the STN is
involved in the integration of SM inputs and an antero-
ventral part not related to body movement.
Moreover, in OCD, bursts did not exhibit the same
characteristics, possibly because of a lack of dopamine
depletion. According to neurocomputational theories, for
example, synfire chains,
bursts are efficient solutions
TABLE 3: Burst Characteristics in OCD and PD
OCD 0.307 6 0.177 12.3 6 4.3 50.9 6 27.8
PD 0.243 6 0.131 14.5 6 5.8 83.4 6 131.2
p 0.01 0.007 <0.0001
Values are means 6 standard deviation.
OCD ¼ obsessive-compulsive disorder; PD ¼ Parkinson
FIGURE 3: (A) Distribution of the mean frequencies of all
subthalamic nucleus (STN) neurons recorded from patients
with obsessive-compulsive disorder (OCD) and Parkinson
disease (PD). The box plot summarizes the distribution, with
the limits of the box representing the 25th and 75th
quartiles and the central line showing the median sample
value. The whiskers extend to the 95% confidence intervals.
The mean neuronal firing rate recorded in OCD was
significantly lower than that in PD (Mann Whitney U test;
**p < 0.001). (B) The relative proportion of the 2 discharge
patterns among neurons sampled in patients with OCD and
PD. No significant difference was found (chi-square test, p
5 0.76). (C) The relative proportion of the 2 discharge
patterns among neurons recorded within the right and left
STN from OCD and PD. The percentage of burst neurons is
significantly higher in the left side than the right side in
OCD (chi-square test, **p 5 0.03 after Bonferroni
correction), whereas no significant differences were found
in PD (chi-square test, p 5 0.64).
ANNALS of Neurology
6 Volume 000, No. 000
for information encoding. The specific bursts found in
OCD may thus reflect specific pathologic information
processing within the noM-STN basal ganglia loops.
Lateralization of Burst Activity in OCD
Burst neurons were asymmetrically distributed mainly in
the left anterior region of STN in OCD, all right-
handed, whereas the majority of burst neurons in PD
were symmetrically distributed in the dorsal part of the
motor STN (see Fig 4). Conflicting results have emerged
from imaging studies with regard to the distribution and
lateralization of functional abnormalities in OCD. Bilat-
eral dysfunction of the orbitofrontal cortex, the anterior
cingulate, and the head of the caudate nucleus have been
observed using metabolic and functional imaging.
Other authors have reported a right dysfunction of the
and have pro-
posed unilateral modulation of the right accumbens nu-
cleus as a therapeutic option.
However, left-sided dys-
function has also been underlined in the orbito frontal
and a volumetric study found decreased gray
matter in the left lateral orbitofrontal, left inferior fron-
tal, and left dorsolateral prefrontal cortex in OCD.
single lesion in the left ventral caudate can also induce
which supports the hypot hesis of an
involvement of the left caudate in OCD.
electrophysiologic data recently obtained in the caudate
of 3 patients with OCD failed to demonstrate any
This asymmetry may be due mainly to the hetero-
geneity of OCD symptoms that involve several
FIGURE 4: Above, distribution of recorded neuronal
activities with respect to the Morel stereotactic atlas of the
human thalamus and basal ganglia. Filled circles illustrate
burst cells, open circles show irregular cells. Below, 3-
dimensional distribution densities of burst neurons
constructed with normalized coordinates, from all patients,
pooled together and considered as a cloud of points. An
isolated neuron corresponds to a small sphere, and clusters
of neurons are represented as larger blobs. They are
overlaid on a T1 magnetic resonance imaging slice of a
canonical brain in the MNI (Montreal Neurological Institute)
space, at average coordinates of neurons (frontal: z 523.5
mm; horizontal: y 5 12 mm). Parkinson disease and
obsessive-compulsive disorder neuronal activities are
illustrated in blue and red, respectively. Right hemisphere
values are expressed negatively. The midanterior border of
the posterior commissure represents the origins of the
coordinate system. (A) Frontal sections centered on
subthalamic nucleus (STN). Recording sites were plotted
from their normalized coordinates on 3 anteroposterior
overlay sections (gray 5 A 10, dark gray 5 A 12, black 5 A
14). Note that some spots are located outside the boundary
of the STN due to the lack of normalization of the laterality.
(B) Horizontal sections centered on STN. Recording sites
were plotted from their normalized coordinates on 3
overlay dorsoventral sections (black 5 v 4.5, dark gray 5 v
3.6, gray 5 v 2.7). SN 5 substantia nigra; ZI 5 zona incerta;
RN 5 red nucleus; ic 5 internal capsule.
Piallat et al: STN Firing in OCD and PD
Month, 2010 7
structures, with relative lateralization.
In addition to
pathophysiologic consideration, the asymmetry of dys-
function might have therapeutic consequences.
Medication and Gender as Possible
All patients with OCD were receiving serotoninergic
treatment, and 4 were receiving low-dose neuroleptics
(see Table 1). Despite the washout of 12 hours, chronic
pharmacologic treatment of OCD may thus have influ-
enced recorded electrophysiologic activity. Indeed, neuro-
leptic agents have an antidopaminergic effect, which,
according to the rate model of basal ganglia, should indi-
rectly induce an increase of the FR of STN neurons. In
our study, however, because neuroleptic doses were very
low and did not induce any motor side effects, neurolep-
tic agents may have interfered weakly with the activity of
The same conclusion might be drawn for serotoni-
nergic treatment. It has been shown that acute adminis-
tration of serotonin reuptake inhibitors (SSRIs) activates
and that chronic SSRI administration
induces robust inhibition of dopaminergic neurons
(which should induce an excitation of STN neurons).
However, this expected increase of FR from serotoniner-
gic medications was not found in our study. The fact
that burst patterns were lateralized on the left side is
another argument against excessive influence from neuro-
leptic and serotoninergic medications on neuronal
The number of women in the OCD group is sig-
nificantly higher and constitutes a possible confounding
factor. To our knowledge, a single study reported some
gender-related differences in local field potentials
recorded in the STN of patients with PD treated by
L-dopa, showing that the power of the high gamma
band was significantly higher in women.
However, it is
difficult to extrapolate these results to single-unit activ-
ities recorded in nonparkinsonian patients.
OCD, STN, and noM Loops of the Basal Ganglia
The dorsolateral part of the STN plays a key role in the
motor loop of the basal ganglia, whereas the ventrome-
dial part is involved in cognitive and emotional loops.
In PD, an increase in FR and burst activity in the SM-
STN leads to excessive inhibition of thalamocortical and
brainstem nuclei, resulting in hypoactivation of the pre-
motor and motor cortex. This prevents linkages between
distinct motor programs and could result in akinesia. In
OCD, a disruption in information processing within the
noM loops that connect the prefrontal cortex, the orbito-
frontal cortex, the cingulum, the basal ganglia, and the
thalamus seems to play a central role in compulsive and
Based on our study, we hypothesize
that burst activity reflects a dysfunction in the associative
cortex circuitry mentioned above, preventing the release
of ‘‘chunks of behaviors,’’ which, under normal situations,
create ‘‘coordinated sequential motor actions and streams
of thought and motivation,’’ as already suggested by
Graybiel and Rauch.
As a result, this could block the
shifting from 1 concept to another in a patient with
OCD, thus creating a ‘‘freezing of thoughts.’’
Despite unknown mechanisms of action, it is now well
established that high-frequency stimulation (HFS)-DBS
of the SM-STN improves motor symptoms in PD. It is
usually accepted that HFS modifies the abnormal STN
discharge pattern and thus restores the neuronal stream
from the basal ganglia output structure to the thalamus.
Here we demonstrate that the noM-STN also has its
own specific burst pattern. We suggest that, first, simi-
larly to the role of the SM-STN in PD, the noM-STN
plays a key role in mediating obsessive-compulsive behav-
ior in OCD. This hypothesis fits well with the significant
proportion of burst neurons found in the noM STN in
OCD. Second, high-frequency DBS-STN would alter
this abnor mal burst pattern, which might explain the
clinical improvements recently reported in OCD.
To conclude, our electrophysiologic data suggest
that neuronal processing within the noM-STN is dys-
functional in OCD. These results could explain the
potential effect of electrical HFS of the noM-STN to
treat severe OCD.
We thank Dr A. Kistner, for help acquiring the clinical
Potential Conflicts of Interest
N.T. has received inscription fees and hotel and travel
expenses for the Neurosurgical Congress from Medtronic.
P.K. has had travel/accommodations expenses reimbursed
by Medtronic. A.L.B. is a board member of University
1. Stein DJ. Obsessive-compulsive disorde r. Lancet 2002;360:
2. Thobois S, Jouanneau E, Bouvard M, Sindou M. Obsessive-com-
pulsive disorder after unilateral caudate nucleus bleeding. Acta
Neurochir (Wien) 2004;146:1027–1031; discussion 1031.
ANNALS of Neurology
8 Volume 000, No. 000
3. Cheyette SR, Cummings JL. Encephalitis lethargica: lessons for
contemporary neuropsychiatry. J Neuropsychiatry Clin Neurosci
4. Saxena S, Rauch SL. Function al neuroimaging and the neuroanat-
omy of obsessive-compulsive disorder. Psychiatr Clin North Am
5. Saxena S, Brody AL, Schwartz JM, Baxter LR. Neuroimaging and
frontal-subcortical circuitry in obsessive-compulsive disorder. Br J
Psychiatry Suppl 1998;(35):26–37.
6. Baxter LR Jr, Schwartz JM, Bergman KS, et al. Caudate glucose
metabolic rate changes with both drug and behavior therapy for
obsessive-compulsive disorder. Arch Gen Psychiatry 1992;49:
7. Alexander GE, DeLong MR, Strick PL. Parallel organization of func-
tionally segregated circuits linking basal ganglia and cortex. Annu
Rev Neurosci 1986;9:357–381.
8. Cummings JL. Frontal-subcortical circuits and human behavior.
Arch Neurol 1993;50:873–880.
9. Krack P, Kumar R, Ardouin C, et al. Mi rthful laughter induced by
subthalamic nucleus stimulation. Mov Disord 2001;16:867–875.
10. 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;20:128–154.
11. Dougherty DD, Baer L, Cosgrove GR, et al. Prospective long-term fol-
low-up of 44 patients who received cingulotomy for treatment-refractory
obsessive-compulsive disorder. Am J Psychiatry 2002;159:269–275.
12. Kihlstrom L, Hindmarsh T, Lax I, et al. Radiosurgical lesions in the
normal human brain 17 years after gamma knife capsulotomy.
Neurosurgery 1997;41:396–401; discussion 401-402.
13. Nuttin B, Cosyns P, Demeulemeester H, et al. Electrical stimula-
tion in anterior limbs of internal capsules in patients with obsessi-
ve-compulsive disorder. Lancet 1999;354:1526.
14. Greenberg BD, Price LH, Rauch SL, et al. Neurosurgery for intrac-
table obsessive-compulsive disorder and depression: critical
issues. Neurosurg Clin North Am 2003;14:199–212.
15. Montoya A, Weiss AP, Price BH, et al. Magnetic resonance imag-
ing-guided stereotactic limbic leukotomy for treatment of intracta-
ble psychiatric disease. Neurosurgery 2002;50:1043–1049;
16. Sturm V, Lenartz D, Koulousakis A, et al. The nucleus accumbens:
a target for deep brain stimulation in obsessive-compulsive- and
anxiety-disorders. J Chem Neuroanat 2003;26:293–299.
17. Mallet L, Polosan M, Jaafari N, et al. Subthalamic nucleus stimula-
tion in severe obsessive-compulsive disorder. N Engl J Med 2008;
18. Benabid AL, Chabardes S, Mitrofanis J, Pollak P. Deep brain stim-
ulation of the subthalamic nucleus for the treatment of Parkinson’s
disease. Lancet Neurol 2009;8:67–81.
19. Hutchison WD, Allan RJ, Opitz H, et al. Neurophysiological identi-
fication of the subthalamic nucleus in surgery for Parkinson’s dis-
ease. Ann Neurol 1998;44:622–628.
20. Magarinos-Ascone CM, Figueiras-Mendez R, Riva-Meana C, Cor-
doba-Fernandez A. Subthalamic neuron activity related to tremor
and movement in Parkinson’s disease. Eur J Neurosci 2000;12:
21. Magnin M, Morel A, Jeanmonod D. Single-unit analysis of the pal-
lidum, thalamus and subthalamic nucleus in parkinsonian patients.
22. Steigerwald F, Potter M, Herzog J, et al. Neuronal activity of the
human subthalamic nucleus in the parkinsonian and nonparkinso-
nian state. J Neurophysiol 2008;100:2515–2524.
23. Bergman H, Wichmann T, Karmon B, DeLong MR. The primate
subthalamic nucleus. II. Neuronal activity in the MPTP model of
parkinsonism. J Neurophysiol 1994;72:507–520.
24. National Consultative Ethics Committee on Health and Life Scien-
ces Opinion on functional neurosurgery for severe psychiatric dis-
order Number 71 2002-04-25.
25. Goodman WK, Price LH, Rasmussen SA, et al. The Yale-Brown
Obsessive Compulsive Scale. I. Development, use, and reliability.
Arch Gen Psychiatry 1989;46:1006–1011.
26. Benabid AL, Krack PP, Benazzouz A, et al. Deep brain stimulation
of the subthalamic nucleus for Parkinson’s disease: methodologic
aspects and clinical criteria. Neurology 2000;55:S40–S44.
27. Piallat B, Chabardes S, Torres N, et al. Gait is associated with an
increase in tonic firing of the sub-cuneiform nucleus neurons. Neu-
28. Murer MG, Riquelme LA, Tseng KY, Pazo JH. Substantia nigra
pars reticulata single unit activity in normal and 60HDA-lesioned
rats: effects of intrastriatal apomorphine and subthalamic lesions.
29. Abeles M. Quantification, smoothing, and confidence limits for
single-units’ histograms. J Neurosci Methods 1982;5:317–325.
30. Morel A. Stereotactic atlas of the human thalamus and basal gan-
glia. New York, NY: Informa Healthcare USA, 2007.
31. Levy R, Hutchison WD, Lozano AM, Dostrovsky JO. High-fre-
quency synchronization of neuronal activity in the subthalamic nu-
cleus of parkinsonian patients with limb tremor. J Neurosci 2000;
32. Rodriguez-Oroz MC, Rodriguez M, Guridi J, et al. The subthalamic
nucleus in Parkinson’s disease: somatotopic organization and
physiological characteristics. Brain 2001;124:1777–1790.
33. Wichmann T, Bergman H, DeLong MR. The primate subthalamic
nucleus. I. Functional properties in intact animals. J Neurophysiol
34. Alexander GE, Crutcher MD, DeLong MR. Basal ganglia-thalamo-
cortical circuits: parallel substrates for motor, oculomotor, ‘‘pre-
frontal’’ and ‘‘limbic’’ functions. Prog Brain Res 1990;85:119–146.
35. Parent A, Smith Y. Organization of efferent projections of the sub-
thalamic nucleus in the squirrel monkey as revealed by retrograde
labeling methods. Brain Res 1987;436:296–310.
36. 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;33:
37. DeLong MR, Crutcher MD, Georgopoulos AP. Primate globus pal-
lidus and subthalamic nucleus: functional organization . J Neuro-
38. Abeles M. Neuroscience. Time is precious. Science 2004;304:
39. Baxter LR Jr, Schwartz JM, Mazziotta JC, et al. Cerebral glucose
metabolic rates in nondepressed patients with obsessive-compul-
sive disorder. Am J Psychiatry 1988;145:1560–1563.
40. Molina V, Montz R, Martin-Loeches M, et al. Drug therapy and
cerebral perfusion in obsessive-compulsive disorder. J Nucl Med
41. Nordahl TE, Benkelfat C, Semple WE, et al. Cerebral glucose met-
abolic rates in obsessive compulsive disorder. Neuropsychophar-
42. Simpson S, Baldwin B. Neuropsychiatry and SPECT of an acute
obsessive-compulsive syndrome patient. Br J Psychiatry 1995;166:
43. Lacerda AL, Dalgalarrondo P, Caetano D, et al. Elevated thalamic
and prefrontal regional cerebral blood flow in obsessive-compul-
sive disorder: a SPECT study. Psychiatry Res 2003;123:125–134.
44. Swedo SE, Schapiro MB, Grady CL, et al. Cerebral glucose me-
tabolism in childhood-onset obsessive-compulsive disorder. Arch
Gen Psychiatry 1989;46:518–523.
Piallat et al: STN Firing in OCD and PD
Month, 2010 9
45. van den Heuvel OA, Remijnse PL, Mataix-Cols D, et al. The major
symptom dimensions of obsessive-compulsive disorder are medi-
ated by partially distinct neural systems. Brain 2009;132:853–868.
46. Flor-Henry P. The obsessive-compulsive syndrome: reflection of
fronto-caudate dysregulation of the left hemisphere? [in French].
47. Guehl D, Benazzouz A, Aouizerate B, et al. Neuronal correlates of
obsessions in the caudate nucleus. Biol Psychiatry 2008;63:
48. Zhang QJ, Liu X, Liu J, et al. Subthalamic neurons show increased
firing to 5-HT2C receptor activation in 6-hydroxydopamine-
lesioned rats. Brain Res 2009;1256:180–189.
49. Dremencov E, El Mansari M, Blier P. Effects of sustained sero-
tonin reuptake inhibition on the firing of dopamine neurons in the
rat ventral tegmental area. J Psychiatry Neurosci 2009;34:
50. Marceglia S, Mrakic-Sposta S, Foffani G, et al. Gender-related dif-
ferences in the human subthalamic area: a local field potential
study. Eur J Neurosci 2006;24:3213–3222.
51. Aouizerate B. Pathophysiological bases of obsessive-compulsive disor-
der and therapeutic implications [in French]. Rev Prat 2007;57:59–63.
52. Graybiel AM, Rauch SL. Toward a neurobiology of obsessive-com-
pulsive disorder. Neuron 2000;28:343–347.
53. Fontaine D, Mattei V, Borg M, et al. Effect of subthalamic nucleus
stimulation on obsessive-compulsive disorder in a patient
with Parkinson disease. Case report. J Neurosurg 2004;100:
ANNALS of Neurology
10 Volume 000, No. 000