While most patients with obsessive-compulsive disorder (OCD) can derive benefit from psychopharmacologic treatment, cognitive behavioral therapy (CBT) that includes exposure and response prevention is recommended without pharmacotherapy when feasible (American Psychiatric Association (APA). Practice guideline for the treatment of patients with obsessive-compulsive disorder. Arlington, VA: 2007. www.psychiatryonline.org).

For patients who prefer medication only, do not have access to CBT, or have severe OCD (or other severe mental conditions) that would benefit from decreased symptoms prior to starting CBT, psychopharmacologic monotherapy is a reasonable treatment option (APA practice guideline op.cit).

The mainstays of treatment for OCD are the selective serotonin reuptake inhibitors (SSRIs). Currently, the following SSRIs are approved for the treatment of OCD in the US. They are:

• Fluoxetine (Prozac)


• Fluvoxamine (Luvox)


• Paroxetine (Paxil, Pexeva)


• Sertraline (Zoloft)

While other SSRIs are not approved for the treatment of OCD, they are most likely also effective. Clomipramine (Anafranil), a tricyclic antidepressant (TCA), has significant serotonin reuptake inhibition properties and is also approved in the US for the treatment of OCD.

Data indicate that SSRIs are superior to placebo in both pediatric and adult populations with a moderate effect size (0.46 CI 0.37–0.55 and 0.48 CI 0.36–0.61, respectively) (Soomro GM et al, Cochrane Database Syst Rev 2008;1:CD001765). Most meta-analyses report clomipramine (Anafranil) is superior to SSRIs in adults and pediatric patients (see for example Watson HJ and Rees CS, J Child Psychol Psychiatry 2008;49(5):489–498). However, given the side effect profile of the TCAs, clomipramine is typically reserved for patients who have failed SSRI treatment (APA practice guideline ibid).

Most patients experience a significant improvement in symptoms, but psychopharmacologic interventions rarely lead to complete remission of symptoms. Data indicate that achieving at least a 25% reduction in symptoms is almost twice as likely with SSRI treatment than with placebo. Regardless of which medication is prescribed, most patients will not experience substantial improvement in symptoms until a minimum of four to six weeks after starting and some may not experience symptom improvement until as many as 10 to 12 weeks after starting (APA practice guideline ibid).

Adverse effects of SSRIs include mild gastrointestinal effects; anxiety, agitation and insomnia (particularly when initiating treatment); bleeding (especially in those on NSAIDs or anticoagulants); sexual dysfunction; weight gain (predominately paroxetine); and QTc prolongation (most recently associated with doses of citalopram greater than 40 mg daily). Clomipramine use is limited by anticholinergic and sexual side effects as well as dose-related seizures and cardiac toxicity.

Other Pharmacological Treatments

For patients who do not respond to the first psychopharmacologic treatment, it is important to ensure comorbid psychiatric illnesses are treated, the patient has had an adequate trial of the maximum recommended or tolerated dose of the medication, and has been adherent with treatment. For those who do not experience significant relief after these interventions, you should consider the addition of CBT, switching to a different SSRI, or use of clomipramine.

Patients who have failed multiple adequate trials of SSRIs, clomipramine, and CBT may benefit from augmentation strategies (APA practice guideline ibid; Lewin AB et al, Neuropsychiatr Dis Treat 2006;2(1):21–31). These augmentation strategies include pro-serotonergic agents (eg, buspirone, lithium), antipsychotics, and benzodiazepines. The best evidence exists for antipsychotics, particularly second generation antipsychotics such as aripiprazole (Abilify) or risperidone (Risperdal) (Ipser JC et al. Cochrane Database of Syst Rev 2006;4:CD005473; APA practice guideline ibid).

Ondansetron (Zofran), memantine (Namenda), riluzole (Rilutek), clomipramine, and mirtazapine (Remeron) have preliminary evidence to suggest benefit as augmentation of SSRIs for refractory OCD (APA practice guideline ibid; Abdel-Ahad P et al, Curr Clin Pharmacol 2013 Feb 4;online ahead of print).

While open-label studies support the use of lithium augmentation of SSRIs, controlled trials have not demonstrated benefit with the combination. Data with buspirone augmentation have been mixed. Therefore, use of buspirone and lithium augmentation should be reserved for patients with comorbidities responsive to these agents or those who have failed other treatments.

Antipsychotics have been shown to be beneficial as augmentation for SSRIs in randomized controlled trials. Placebo controlled trials of antipsychotics have found significantly higher response rates compared with placebo (46% to 71%) and relapse rates of more than 80% when antipsychotics are discontinued in responders.

Benzodiazepines have not been consistently shown to be of benefit for OCD as monotherapy or augmentation. In addition, their use is limited by sedation, motor side effects, and potential for disinhibition. However, given their quick onset of action, they may play a role in decreasing anxiety associated with initiation of SSRIs (APA practice guideline, Abdel-Ahad ibid).

Non-Pharmacologic Treatments

Other non-pharmacologic treatments that show promise for OCD include neurosurgery, deep brain stimulation, electroconvulsive therapy (ECT), and transcranial magnetic stimulation.

Given the risk associated with neurosurgery, this treatment is typically reserved for those with incapacitating OCD that does not respond to medication and CBT. Anterior cingulotomy, subcaudate tractotomy, and anterior capsulotomy have been shown (in uncontrolled studies) to lead to improvements in 35% to 70% of patients with refractory OCD and depression.

Deep brain stimulation is a promising treatment but current data are limited. In a randomized controlled trial, right bilateral prefrontal transcranial magnetic stimulation resulted in up to a 35% decrease in symptoms compared with a 6.2% decrease with sham treatment (Gomes PV et al, J Neuropsychiatry Clin Neurosci 2012;24(4):437–443). ECT has traditionally been reserved for those with comorbid depression and OCD, and is typically not used for OCD alone (Arumugham SS and Reddy JY, Expert Rev Neurother 2013;13(2):187–202).

TCRBH’s Take: Pharmacological “treatment response” in research on OCD is typically defined as achieving at least a 25% to 35% reduction of symptoms. Even though that bar is set quite low, that relatively small reduction in symptoms can make the difference between severe impairment and moderate impairment—and it can enable a patient to engage in CBT for OCD who might otherwise have been unable or unwilling to do so.