To the Editor: When obsessive-compulsive disorder (OCD) symptoms are severe and refractory to both cognitive-behavioral therapy (CBT) and medication, deep brain stimulation (DBS) may be of value (1). Although OCD research has examined the effects of DBS on cognition (2), few studies have used translational computerized paradigms capable of fractionating dissociable aspects of cognition and their neural substrates. Neurocognitive assessments have the potential to help elucidate the underlying mechanisms of action of DBS in OCD and the optimal brain target.
“Mr. T” is a 30-year-old man with a 5-year history of OCD with primary contamination obsessions and washing compulsions. He had stopped socializing, had dropped out of school, and was unemployed because of his OCD. Past adequate trials of all serotonin reuptake inhibitors, both as monotherapy and with multiple augmentation strategies, and 20 weeks of CBT using exposure response prevention provided only limited benefits.
After ethical review board approval, Mr. T underwent bilateral implantation of electrodes targeting the nucleus accumbens. At the time of surgery, his Yale-Brown Obsessive Compulsive Scale score was 32 while taking the following medications: clomipramine, 250 mg/day; ziprasidone, 120 mg/day; and clonazepam, 1 mg t.i.d. His medication dosages were unchanged before and after the cognitive testing.
We performed cognitive assessments at baseline (prestimulation) and again 8 months after DBS began. Tasks from the Cambridge Neuropsychological Test Automated Battery included the stop-signal test (assessing ability to suppress prepotent motor responses), the intradimensional/extradimensional set shift task (examining rule learning and behavioral flexibility), and the Cambridge Gamble Task (assessing decision making). The results of these assessments are summarized in Table 1.
Cambridge Neuropsychological Test Automated Battery (CANTAB) Test Performance After Deep Brain Stimulation (DBS) Compared With Normative Dataa
| Add to My POL
|Individual Tests||Before DBS||After DBS||Before DBS||After DBS|
|Stop-signal test reaction time (msec)||211.58||204.70||–1.0||–0.8|
|Stop-signal test median go reaction time (msec)||449.00||379.00||0.1||0.5|
|Intradimensional/extradimensional set shift task, pre-extradimensional errors||6.00||7.00||0.2||–0.1|
|Intradimensional/extradimensional set shift task, extradimensional errors||1.00||2.00||0.8||0.6|
|Cambridge Gamble Task, overall proportion bet||0.60||0.57||0.6||0.4|
|Cambridge Gamble Task, quality of decision making||1.00||1.00||0.5||0.5|
Before the DBS, Mr. T generally exhibited cognitive performance akin to healthy comparison subjects except for evidence of stop-signal reaction time impairment (z=1.0). His stop-signal reaction time performance changed little after DBS (posttreatment z=0.8); however, the DBS resulted in significant improvement in Mr. T's OCD symptoms, and his Yale-Brown Obsessive Compulsive Scale score decreased to 10. Mr. T returned to college and now has a social life and works part-time.
This case indicates that DBS, targeting the nucleus accumbens, was associated with significant therapeutic benefits in treatment-refractory OCD in the absence of effects on response inhibition, set shifting, or decision making. The lack of effect of accumbens DBS on the stop-signal deficit accords with translational research indicating that accumbens damage (unlike cortical damage) has no effect on response inhibition on an equivalent animal task (4). While impaired response inhibition appears to be a trait marker for OCD, and was evident in Mr. T at baseline, DBS to the accumbens does not appear to ameliorate this problem. This case illustrates the value of DBS in patients with refractory OCD and the importance of including cognitive tests in such studies to identify meaningful predictors for successful DBS in OCD on an individual level.