Current Status of Deep Brain Stimulation for Obsessive-Compulsive Disorder: A Clinical Review of Different Targets
- First Online:
- Cite this article as:
- de Koning, P.P., Figee, M., van den Munckhof, P. et al. Curr Psychiatry Rep (2011) 13: 274. doi:10.1007/s11920-011-0200-8
- 1.3k Views
Obsessive-compulsive disorder (OCD) is a chronic psychiatric disorder that affects 2% of the general population. Despite optimal cognitive-behavioral and pharmacologic therapy, approximately 10% of patients remain treatment resistant. Currently, deep brain stimulation (DBS) is being investigated as an experimental therapy for treatment-refractory OCD. This review focuses on the efficacy and adverse events of all published DBS targets for OCD: anterior limb of the internal capsule, ventral striatum/ventral capsule, nucleus accumbens, nucleus subthalamicus, and inferior thalamic peduncle. Small studies with various designs indicate an overall average Yale-Brown Obsessive Compulsive Scale score decrease ranging from 6.8 to 31 points. The average overall responder rate is ±50%. The frequency of adverse events seems to be limited. Larger prospective studies including neuroimaging are needed to estimate adequately the true potential of DBS in treatment of OCD and to elucidate its underlying mechanism of action and optimal brain target. We conclude that DBS may be a promising and safe therapy for treatment-resistant OCD.
KeywordsDeep brain stimulationObsessive-compulsive disorderNeurosurgeryNucleus accumbensInternal capsuleVentral capsuleVentral striatumSubthalamic nucleusInferior thalamic peduncleNeuroimagingPETMRISPECTParkinsonAdverse eventsCognitive functionCTSCResponderY-BOCS
Obsessive-compulsive disorder (OCD) is a heterogeneous, chronic, and disabling anxiety disorder. According to the DSM-IV definition, the essential features of OCD are recurrent obsessions and/or compulsions that are severe and time consuming (more than 1 h/day); cause marked distress; or significantly interfere with an individual’s normal routine, occupational functioning, usual social activities, or relationships. At some point during the course of the disorder, the individual has recognized that the obsessions or compulsions are excessive or unreasonable. OCD has an estimated lifetime prevalence of 1% to 3% and afflicts men and women equally [1, 2]. If left untreated, OCD can destroy a person’s capacity to function at work, socially, and even at home. Specific treatments for OCD have been developed, such as cognitive-behavioral therapy and pharmacotherapy with serotonin reuptake inhibitors. It is estimated that these treatments provide an average of 40% to 60% symptom reduction in half of patients. However, even when the best available treatments are applied, approximately 10% of patients remain severely affected and suffer from treatment-refractory OCD .
Data for this review were identified by searches on Medline using different combinations of the following MeSH and free text terms: deep brain stimulation, obsessive-compulsive disorder, neurosurgery, nucleus accumbens (NAc), internal capsule, ventral capsule, ventral striatum, subthalamic nucleus, inferior thalamic peduncle, neuroimaging, PET (positron emission tomography), MRI (magnetic resonance imaging), SPECT (single photon emission computed tomography), Parkinson, responder, Y-BOCS (Yale-Brown Obsessive Compulsive Scale), and adverse events. By using various combinations of these search terms, all Medline-listed studies as of January 2011 on DBS in OCD were identified, and the reference lists of the relevant publications were screened. We included human studies that assessed the efficacy of DBS in OCD and accepted case studies if they included some estimate of efficacy. Due to differences in outcome measures, we only included studies that had used the Y-BOCS for the efficacy analysis. The primary outcome measure was the mean improvement in the Y-BOCS score after follow-up. The secondary outcome measure was the responder rate, defined as the proportion of patients with a minimum of 35% decrease on the Y-BOCS.
Efficacy of Deep Brain Stimulation for Obsessive-Compulsive Disorder
Open case studies of deep brain stimulation in the treatment of obsessive-compulsive disorder
Follow-up change in Y-BOCS score (all patients), points
Responders (≥35% Y-BOCS ↓), n (%)a
Mallet et al.  (2002)
185 + 130 Hz
60 + 90 μs
3.1 + 3.2 V
Anderson and Ahmed  (2003)
Sturm et al.  (2003)c
No Y-BOCS scores
Fontaine et al.  (2004)
Right = 3.5 V; left = 1.3 V
Aouizerate et al.  (2004)
Greenberg et al.  (2006)
Jiménez-Ponce et al. [24•] (2009)
Franzini et al.  (2010)
5.0 + 5.5 V
Controlled studies with blinded on–off phase of deep brain stimulation in treatment of obsessive-compulsive disorder
Follow-up change in Y-BOCS score (all patients), points
Follow-up (responders), n (%)a
Blinded on/off phase
Y-BOCS score, on vs off
Nuttin et al.  (1999)
1 patient (duration unknown)
No Y-BOCS scores
Nuttin et al.  (2002)
MDT 3487Ac and 3887d
4 patients (3 month on, 5–10 week off)
19.8 vs 32.3
Abelson et al.  (2005)
4 blinded on–off periods (duration unknown)
26.5 vs 29.3
Mallet et al. [22••] (2008)
3 month on, 3 month off
19.8 vs 28.7
2.0 ± 0.9 V
Goodman et al. [13•] (2010)
Staggered-onset (stimulation 30 or 60 d after surgery)
Huff et al. [16•] (2010)f
3 month on, 3 month off
27.9 vs 31.1
Denys et al. [19••] (2010)
2 week on, 2 week off
8.3-point difference between on and off
Anterior Limb of Internal Capsule
The anterior limb of internal capsule (ALIC) is part of the internal capsule in front of the genu, between the head of the caudate nucleus and the lenticular nucleus. It contains fibers connecting the prefrontal cortex and the subcortical nuclei, including the dorsomedial thalamus. The choice of the ALIC as a brain target for DBS was based on the experience with the anterior capsulotomy for refractory OCD. This neurosurgical procedure had shown positive response in approximately 50% of participants . The first experimental DBS for OCD was performed in Stockholm in 1998 at the Karolinska Hospital. Two patients received bilateral stimulation of the ALIC with use of external batteries. The results were never published, but there were no significant effects on OCD symptoms after 3 to 4 weeks of stimulation in these patients (Andreewitch and Meyerson, personal communication).
One year later, Nuttin et al.  published the first article on bilateral ALIC DBS in four patients. The authors reported “some beneficial effects” in three of four patients. Symptom severity scores were not based on a validated questionnaire such as the Y-BOCS but rather on clinical observation only.
Another study by the same group in 2002 described six patients with DBS in the ALIC for a period of 21 months . Four patients participated in the crossover design phase. Three of four patients showed a 35% or greater reduction in symptoms on the Y-BOCS after follow-up. An average of 40% symptom decrease was observed in the double-blind, controlled, on–off phase of the study. In 2003, a single case study involving ALIC DBS by Anderson and Ahmed  reported a 79% reduction in symptoms after 3 months of open stimulation. There was an overall symptom decrease of 23 points on the Y-BOCS at the 10-month follow-up. Abelson et al.  reported two responders of four patients in 2005. Patients in this study received stimulation in a randomized on–off sequence of four 3-week blocks, followed by an open stimulation phase. Only one patient had a decrease of 35% or more (13-point decrease on the Y-BOCS) in the double-blind phase. This patient was the only one with monopolar stimulation settings. He had a 73% (22 points) improvement during the open-phase follow-up. Another patient failed to show improvement until the open phase, with a final reduction of 44% (16 points) compared with baseline.
Ventral Capsule/Ventral Striatum
In subsequent years, adjacent structures of the internal capsule were targeted for DBS. One of these adjacent regions is the ventral striatal (VS) area. The VS contains the ventral caudate nucleus and NAc and is thought to be associated with reward and motivation. Combined with the ventral capsule (VC), it is referred to as the VC/VS region. This brain target was chosen based on positive results following gamma knife capsulotomy at the ventral region of the ALIC for refractory OCD . In 2006, Greenberg et al.  published the results of 10 patients who underwent bilateral stimulation of the VC/VS. Eight were observed for 3 years. Four of eight patients were considered responders (≥35% symptom reduction). They reported an overall average decrease of 12.3 points on the Y-BOCS during follow-up. In 2010, combined long-term results from 26 patients with VC/VS implantation were published by the same American–Belgian group [12••]. There was an overall responder rate of 62% after a mean of 31.4 months of follow-up. The overall average Y-BOCS decrease for all participants was 13.1 points. Of importance was that results generally improved for patients implanted more recently, which resulted in lower pulse width and voltage settings. Refinement of the implantation site to a more posterior location toward the junction of the anterior capsule, anterior commissure (AC), and bed nucleus of the stria terminalis (BST) seemed to be the main factor accounting for this gain. A recent pilot study by Goodman et al. [13•], using a blinded, staggered-onset design of six OCD patients with VC/VS DBS, showed four of six responders after 12 months’ follow-up. Stimulation had started under blinded conditions 30 or 60 days after surgery. The overall symptom decrease was 46% (15.7 points).
The NAc is part of the ventral striatum. It is located where the head of the caudate and the anterior portion of the putamen meet, just beneath the ALIC, and is involved in functions ranging from reward processing to motivation and addiction. The NAc is considered a promising target for DBS because there is evidence of dysfunction of the reward system in OCD. In a study by Figee et al. [14•] using a monetary incentive delay task and functional MRI, OCD patients showed attenuated reward anticipation activity in the NAc compared with healthy controls. In 2003, Sturm et al.  published the first DBS results of unilateral, right-sided NAc implantation in four OCD patients. In this open study, after 24 to 30 months, three of four patients were considered responders, although no Y-BOCS scores were reported in the article. In 2010, the same group published a double-blind study on unilateral, right-sided NAc DBS in 10 OCD patients [16•]. The symptom improvement observed in the double-blind part of the study was, however, limited to an average of 10%. The mean Y-BOCS score went from 27.9 in the active stimulation to 31.1 during sham stimulation. At 1-year follow-up, only one patient showed 35% or greater symptom improvement. Five patients were considered partial responders (≥25% symptom improvement). The average Y-BOCS symptom decrease for all participants at 1-year follow-up was 21% (6.8 points). An earlier open stimulation case study by Aouizerate et al.  on NAc/ventral caudate DBS for OCD and depression had reported a delayed 52% decrease in symptoms at 15-month follow-up. In 2010, Franzini et al.  reported on NAc DBS for OCD in two patients. They reported an average symptom improvement of 38% (12 points). Denys et al. [19••] published a study on 16 patients with NAc DBS for OCD in 2010. This study consisted of an open 8-month treatment phase, followed by a double-blind, crossover phase with randomly assigned 2-week periods of active or sham stimulation. It ended with an open 12-month maintenance phase. This resulted in an average 46% symptom decrease after 8 months. Nine of 16 patients were responders during follow-up. These nine individuals had a mean Y-BOCS score decrease of 72% (23.7 points). The average symptom decrease at 21 months’ follow-up for all 16 individuals was 48% (17.5 points). In the double-blind, sham-controlled phase (n = 14), the mean Y-BOCS difference between active and sham stimulation was 25% (8.3 points).
The subthalamic nucleus (STN) is located ventral to the thalamus, dorsal to the substantia nigra, and medial to the corticospinal tract, and is part of the basal ganglia. Studies of DBS in Parkinson’s disease (PD) have by serendipity highlighted the presumable role of the STN in behavioral alteration and reducing OCD symptoms. In 2002 and 2004, two case reports were published targeting the STN. The STN is an effective target for PD and was chosen as an implantation site for three patients with PD who also had OCD. Mallet et al.  reported two of two responders with an average Y-BOCS decrease of 82% (20 points) after 6 months of stimulation. Fontaine et al.  reported a 97% (31 points) Y-BOCS decrease after 12 months in a single case study. In 2008, Mallet et al. [22••] reported on the efficacy of bilateral STN stimulation in 16 OCD patients. Twelve of 16 patients were categorized as responders, although responders were defined by a mean decrease of 25% or greater in Y-BOCS score in this study. The overall average Y-BOCS decrease from the active versus sham phase was 31% (8.9 points).
Inferior Thalamic Peduncle
The inferior thalamic peduncle (ITP) is part of the orbitofrontal-thalamic system and links the thalamus to the orbitofrontal cortex (OFC). Because these structures and system are central in the pathophysiology of OCD , it was hypothesized that electrical stimulation of this white matter bundle could reduce OCD symptoms. Jiménez-Ponce et al. [24•] conducted the only study on DBS for OCD in the ITP. They reported five of five responders on the Y-BOCS after 12 months’ follow-up. The average decrease on the Y-BOCS was 49% (17.2 points).
DBS is an invasive procedure. It is associated with different types of adverse events: procedure-related complications, device-related problems (technical problems), and undesired effects caused by stimulation or cessation of stimulation. There is a clear difference in knowledge of adverse events related to DBS in movement disorders as compared with OCD because of the extensive literature on patients implanted for movement disorders (>80,000) and the relatively small number of DBS procedures for OCD (±100). Although different implantation sites are used, many operating procedures are similar. Therefore, the risk of procedure-related complications and device-related problems could be derived from the comprehensive experience of DBS for the treatment of movement disorders.
Several groups have reported on surgical complications in large series of patients who underwent DBS in the treatment of movement disorders. Infections, misplaced leads, and seizures can occur in up to 30% of patients . A potential serious risk of the introduction of the electrodes is intracerebral hemorrhage. In patients with movement disorders, the recent literature has reported a 0% to 5% risk of intracerebral hemorrhage [26, 27], with sulcal and transventricular electrode trajectories, elevated blood pressure, and older age as major risk factors . In OCD patients, intracerebral hemorrhage was only reported in 1 of 10 patients by Greenberg et al. , and in 1 patient in the study by Mallet et al. [22••]. Superficial wound infection was reported in 1 of 10 OCD patients by Greenberg et al. , and in 2 of 16 OCD patients by Mallet et al. [22••]. In the latter study, the implanted electrodes had to be removed in one patient.
Lead breakage and failure of the neurostimulator, reported to occur in as many as 8% of cases in 2002 , have rarely been noted in more recent literature. Greenberg et al.  reported a break in the electrode and subcutaneous extension cable requiring a replacement in one OCD patient. Nuttin et al.  reported that OCD patients disturbingly felt the material within their body to the extent that one patient (of four) wanted it to be removed.
Undesired Stimulation Effects
Undesired stimulation effects vary widely (physical to mental) and can be divided into acute effects and effects of chronic stimulation. The latter can be divided into effects on mood and effects on cognition. These effects are usually reversible by cessation or adjustment of stimulation parameters. Okun et al.  reported acute olfactory, gustatory, and motor sensations that were strongly associated with the most ventral electrode positions, as well as physiologic responses. All effects reversed after DBS cessation or parameter changes.
Acute mood changes during the first few days of stimulation of the ALIC and NAc have been reported, specifically transient sadness, anxiety , and euphoria, sometimes to the extent of hypomanic and manic symptoms . Transient mania or hypomania after DBS implantation has been reported in several targets, including the globus pallidus, STN, and the ALIC–NAc region . In 2010, a case report by Tsai  reported a hypomanic episode following bilateral ALIC stimulation. The hypomanic symptoms diminished after lowering the voltage from 4.0 to 2.0 V, but the OCD symptoms persisted during follow-up. All hypomanic and manic episodes associated with DBS stimulation dissolved after the field density was readjusted by changing the voltage and/or the active contact. Transient hypomania is the side effect most commonly observed immediately after stimulation. Transient hypomanic episodes seem to occur more often in the VC/VS–NAc region. Occurrence is estimated to be as high as 50% to 67% in ALIC–NAc DBS, as contrasted with 4% to 8% in STN DBS patients .
Chronic mood improvement is an unintended but favorable side effect of DBS because most treatment-refractory OCD patients suffer from comorbid major depression. Denys et al. [19••], Abelson et al. , and Greenberg et al.  reported improvement of depression after accumbens, ALIC, and VS/VC stimulation, respectively. Antidepressive effects thus seem to be related to DBS of the ventral striatum in particular [13•, 17, 18, 22••], because no mood improvement was observed following STN stimulation [22••]. Stimulation cessation can result in severe worsening of mood [7, 9], although this can be reversed by reactivation of the stimulation.
Effects on Cognition
Apart from transient diminished concentration and verbal perseverations [12••], DBS has not been associated with evident cognitive decline and/or cognitive function improvement. However, the literature on this topic is sparse. Analysis of neuropsychological testing in the study by Aouizerate et al.  (1 patient) showed no deterioration in memory or attentional or executive function tests. Abelson et al.  administrated a large battery of tests at baseline and at 6-month follow-up in all four patients. They reported no obvious patterns of cognitive changes following stimulation. Goodman et al. [13•] performed extended neuropsychological tests in six patients. Overall, results indicated that the clinical effectiveness in this population was achieved without significant neuropsychological dysfunctions. All patients in the study by Greenberg et al.  completed neuropsychological assessment before implantation and after a mean of 10 months of chronic DBS. Analysis found no pattern of pervasive decline or improvement in any one patient. Denys et al. [19••] reported mild forgetfulness in 5 of 16 patients and word-finding problems in 3 of 16 patients following NAc DBS. An extensive neuropsychological test battery was carried out, but given the extensiveness of the data, the outcomes will be published in a separate article.
Positive Side Effects
Thus far, no proven response predictors for DBS in OCD have been identified. One American group observed that the onset of laughter during the neurosurgical procedure and optimization of DBS settings predicted improvement in OCD symptoms [36, 37]. More intense laughter was associated with a greater reduction in Y-BOCS scores 2 years after implantation. They hypothesized that laughter is an epiphenomenon of stimulation of a putative “sweet spot,” a region that ameliorates OCD symptoms when chronically stimulated.
Mechanism of Action
Circuits connecting the OFC, medial prefrontal cortex, basal ganglia, and thalamus are central to OCD pathophysiology and treatment response . Although the mechanism of DBS is still unknown, a widely accepted hypothesis is that OCD is associated with hyperactivity of the cortical-striatal-pallidal-thalamic-cortical network . It is plausible that DBS inhibits or functionally overrides this pathological network hyperactivity [39•]. Stimulation frequency seems to be a key factor in determining clinical efficacy. Low frequencies are considered to activate neurons, whereas high-frequency stimulation results in neuronal inhibition . Combined imaging and DBS studies that may confirm the inhibitory characteristics of DBS are sparse. Le Jeune et al. [41•] found hyperactivity in the OFC of patients with severe and refractory OCD compared with healthy controls. The same study compared the metabolic effects of STN DBS (n = 10) in the on-versus-off stimulation conditions. During on stimulation, a significant decrease in glucose metabolism was observed in the anterior cingulated gyrus. In addition, therapeutic effects correlated with a decrease in OFC metabolism. Abelson et al.  showed decreased PET activity in the OFC after 3 to 6 weeks of ALIC DBS in two OCD responders, but not in nonresponders. The involvement of other brain areas, such as the OFC, anterior cingulated cortex, striatum, pallidus, and thalamus, was seen in a PET study in six OCD patients 2 weeks after implantation of the electrodes in the VC/VS . Increased activity in the frontal cortex and striatum after ALIC DBS activation was seen during postoperative MRI . In the same study, clinical response after 3 months of continuous stimulation was related to a relative decrease in OFC hyperactivity. Thus, as of yet, sparse neuroimaging research suggests that hyperactivity in the OFC correlates with the severity of OCD, and that OFC activity normalizes following DBS.
The 16 identified efficacy studies used different brain targets, study designs, numbers of participants, duration of follow-up, on-off phase duration, electrode models, and stimulation parameters. The reported percentages of responding patients and mean Y-BOCS decrease also differed considerably (Tables 1 and 2). Although one should be careful interpreting these studies because of the different designs and brain targets, an improvement of 35% in Y-BOCS scores was observed in 34 of 63 patients, with a Y-BOCS decrease that ranged from 6.8 to 31 points. Can we determine the currently optimal DBS brain target for OCD from these results? Monopolar NAc DBS with electrode model 3389 (Medtronic, Minneapolis, MN), as reported by our group, resulted in a mean Y-BOCS decrease of 72% in the 56% of responding patients, the largest follow-up Y-BOCS decrease published to date [19••]. Interestingly, clinical improvement was only observed when the upper two (ie, dorsal) electrode contacts were used, suggesting active stimulation at the border of the NAc core and ventral ALIC. Greenberg et al. [12••] reported an increase in responder rate (from 33.3% to 75%) and a larger mean Y-BOCS decrease in responders (from 29% to 54.3%) when the implantation site of their ALIC–VC/VS electrodes was moved posteriorly to the junction of anterior capsule, AC, and BST. However, the large size of their 3887 electrode (Medtronic) contacts makes it difficult to pinpoint the site of active stimulation to a specific anatomic structure. Jimenez-Ponce et al. [24•] reported a mean Y-BOCS decrease of 49% in 100% of the responding patients following bipolar ITP DBS using electrode model 3387 (Medtronic), with active stimulation just posterior of BST. Thus far, the mechanism of action of DBS for OCD remains unclear. Functional imaging studies of DBS in ALIC  and STN  showed normalization of OFC hyperactivity, suggesting a final common cortical-striatal pathway. Serious procedure-related events are rare, and side effects can be reversed by cessation or adjustment of the stimulation parameters. The present data suggest no decline or improvement in cognitive function caused by DBS. DBS may be a promising and safe therapy in patients with treatment-refractory OCD. Further research is needed to optimize this therapy with respect to patient selection and management, target location, and investigation of its mechanism of action.
Dr. Schuurman has served as a consultant for, received honoraria for service on an expert safety board from, and had travel expenses reimbursed by Medtronic. Drs. De Koning, Figee, van den Munckhof, and Denys reported no potential conflicts of interest relevant to this article.