To the Editor: Current options for treatment-resistant obsessive-compulsive disorder (OCD) include switching to an alternative selective serotonin reuptake inhibitor or augmentation with dopamine antagonists or other agents (1). Evidence from genetic, behavioral, and neuroimaging studies have indicated glutamatergic alteration in OCD (2). In pediatric OCD patients, the glutamate caudate concentration was abnormally increased, but it decreased after paroxetine treatment (3). Thus, attenuating glutamatergic hyperactivity might be beneficial in OCD. We report a therapeutic effect of add-on memantine, an N-methyl-d-aspartic acid glutamatergic receptor antagonist, in treatment-resistant OCD.
Ms. A, a 34-year-old woman, was seen with incapacitating ego-dystonic obsessions, including fear of harm to her daughter and of losing her mind. She developed compulsive checking behavior to decrease the associated anxiety. Obsessive-compulsive symptoms, initially detected at age 16, remitted spontaneously 2 years later. Subsequent postpartum exacerbation of DSM-IV OCD symptoms associated with major depression occurred at age 30. She also met DSM-IV criteria for schizotypal personality disorder.
Subsequent adequate trials with paroxetine and sertraline were ineffective. Add-on risperidone caused marked akathisia and was discontinued. At her presentation, oral clomipramine was initiated and titrated to 300 mg/day; however, 10 weeks later, there was no significant clinical improvement (Yale-Brown Obsessive Compulsive Scale  score=35). Addition of a selective dopamine D2 antagonist, sulpiride (up to 400 mg/day for 4 weeks), was also ineffective (Yale-Brown Obsessive Compulsive Scale score=34). At this point, adding memantine to Ms. A’s regimen of clomipramine (300 mg/day) and sulpiride (400 mg/day) was suggested, and she signed informed consent after explanation of this off-label therapy. Memantine was started at 5 mg/day and titrated to 20 mg/day within 2 weeks. Ms. A reported initial relief on day 7 of combined treatment, and a significant decrease in symptom severity was noted 3 weeks later (Yale-Brown Obsessive Compulsive Scale score=22). There was a substantial reduction in the time occupied by OCD and distress, followed by increased control over obsessions. No clinically significant side effects were noted. Improvement was maintained after 3 months.
Add-on memantine was well tolerated and resulted in clinically significant reduction of OCD symptom severity. Prior treatment resistance and the proximity between symptomatic improvement and the initiation of memantine point to its possible attenuating effect on the symptoms of OCD. The association of OCD-schizotypal comorbidity with the beneficial effect of memantine is noteworthy in view of a pertinence of glutamatergic dysfunction in both OCD and schizophrenia spectrum disorders (5). Our case suggests that memantine may be an option for treatment-resistant OCD, but controlled studies are needed to substantiate this observation.