Obsessive-Compulsive Disorder: Augmentation Strategies

Augmentation Strategies

If a patient has had only a partial response to an antiobsessional agent of adequate dose and duration, the next question is whether to change the SRI or add an augmenting agent. Current clinical practice suggests that if there is no response at all to an SRI, it may be best to change to another SRI. However, if there has been some response to treatment, an augmentation trial of at least 2 to 8 weeks may be warranted. No augmentation agent has been firmly established as efficacious. Many questions about augmen-tation remain unanswered, including the optimal duration of aug-mentation, comparative efficacy of different agents, predictors of response and mechanism of action. Nonetheless, these agents do help some patients significantly, and thus their systematic use should be considered (see Table 51.3).

In patients with severe symptoms or comorbid psycho-sis or tic disorder, pimozide 1 to 3 mg/day, haldol 2 to 10 mg/ day, and other neuroleptic agents (risperdone 2–8 mg/day and olanzapine 2.5–10 mg/day) have been used with some success.

However, the use of a neuroleptic agent should be considered carefully in light of the risk of extrpyramidal symptoms and side effects such as weight gain, lethargy and tardive dyski-nesia. Thus, when a neuroleptic drug is used, target symptoms should be established before beginning treatment, and the med-ication discontinued within several months if target symptoms do not improve.

The use of lithium (300–600 mg/day) and buspirone (up to 60 mg/day) as augmentation agents has also been explored. Both agents looked promising in open trials but failed to be effective in more systematic trials. Augmentation with fenfluramine (up to 60 mg/day), clonazepam (up to 5 mg/day), clonidine (0.1–0.6 mg/day) and trazodone (100–200 mg/day), as well as the combination of clomipramine with any of the SSRIs, has had anecdotal success but has not been evaluated in methodologically rigorous studies. Some potential augmenting agents and their dosage ranges are presented in Table 51.3.

Behavioral Therapy

Behavioral therapy is effective for OCD both as a primary treat-ment and as an augmentation agent. This form of therapy is based on the principle of exposure and response prevention. The patient is asked to endure, in a graduated manner, the anxiety that a specific obsessional fear provokes while refraining from compulsions that allay that anxiety. The principles behind the ef-ficacy of treatment are explained to the patient in the following way. Although compulsions, either covert or overt, usually im-mediately relieve anxiety, this is only a short-term solution; the anxiety will ultimately return, requiring the performance of an-other compulsion. However, if the patient resists the anxiety and urge to ritualize, the anxiety will eventually decrease on its own (i.e., habituation will occur), and the need to perform the ritual will eventually disappear. Thus, behavioral therapy helps the pa-tient habituate to the anxiety and extinguish the compulsions.

Compulsions, especially overt behaviors like washing rituals, are more successfully treated by behavioral therapy than are obsessions alone or covert rituals like mental checking. This is because covert rituals are physically harder to resist than are rituals like handwashing and checking a door. In fact washing rituals are the most amenable to behavioral treatment, followed by checking rituals and then mental rituals.

For rituals that do not constitute overt behaviors, tech-niques other than exposure and response prevention have been used in conjunction with exposure and response prevention. These approaches include imaginal flooding and thought stop-ping. In imaginal flooding, the anxiety provoked by the obses-sions is evoked by continually repeating the thought, often with the help of a continuous-loop tape or the reading of a “script” composed by the patient and therapist, until the thought no longer provokes anxiety. In thought stopping, an compulsive mental rit-ual (e.g., continually repeating a short prayer in one’s head) is stopped by simply shouting, making a loud noise, or snapping a rubber band on the wrist in an attempt to interrupt the thought.

In the early stages of treatment, a behavioral assessment is performed. During this assessment, the content, frequency, dura-tion, amount of interference and distress, and attempts to resist or ignore the obsessions and compulsions are catalogued. An at-tempt is made to clarify the types of symptoms, any triggers that bring on the obsessions and compulsions, and the amount and type of avoidance used to deal with the symptoms. For instance, in the clinical vignette described later, the fact that Ms Z stopped preparing meals to deal with her obsessional concerns about con-tamination was carefully documented. The patient, usually with the help of a therapist, then develops a hierarchy of situations ac-cording to the amount of anxiety they provoke. During treatment, patients gradually engage in the anxiety-provoking situations in-cluded in their hierarchy without performing anxiety-reducing rituals.

Behavioral therapy can be used with patients of any age and has been used in young children, often with the help of a parent as a cotherapist. However, systematic trials of behavioral treatment in children have not been performed. More recently, behavioral therapy in a group setting has been explored and found as effective as, and perhaps even more effective than, indi-vidual behavioral treatment. The group seems to act as a catalyst for change by promoting group cohesion, support and encour-agement. Groups can include patients with different symptoms, though each has a personalized hierarchy, so that one patient can encourage another. Van Noppen and associates (1991) included family members in groups because families are often affected by the patient’s rituals and often function as unwilling participants in rituals. As members of the group, family members not only gain knowledge and understanding about OCD but can be co-therapists at home for homework assignments.

Despite its efficacy, behavioral therapy has limitations. To begin with, about 15 to 25% of patients refuse to engage in behav-ioral treatment initially or drop out early in treatment because it is so anxiety-provoking. Behavioral treatment fails in another 25% of patients for a variety of other reasons, including concomitant depression; the use of central nervous system depressants, which may inhibit the ability to habituate to anxiety; lack of insight; poor compliance with homework, resulting in inadequate expo-sure; and poor compliance on the part of the therapist in enforc-ing the behavioral paradigm. Thus, overall, 50 to 70% of patients are helped by this form of therapy.

One of the issues that has emerged in treating OCD with CBT is the lack of trained therapists and the cost of repeated indi-vidual exposure sessions. Thus, in addition to developing group treatments which allow therapists to treat a number of patients simultaneously, researchers have begun to develop computer-guided behavior therapy. Several recent reports have shown that while this modality is not as effective as individual behavior therapy, it does allow for significant improvement in symptoms over a control condition like relaxation therapy.

Behavior therapy can be used as the sole treatment of OCD, particularly with patients whose contamination fears or somatic obsessions make them resistant to taking medications. Behavioral treatment is also a powerful adjunct to pharmaco-therapy. Some research appears to indicate that combined treat-ment may be more effective than pharmacotherapy or behavio-ral therapy alone, although these findings are still preliminary. Some studies have even suggested that adding pharmacotherapy to behavior therapy may be particularly helpful in reducing ob-sessions while compulsions respond to behavior therapy. From aclinical perspective, it may be useful to have patients begin treat-ment with medication to reduce the intensity of their symptoms or comorbid depressive symptoms if present; patients may then be more amenable to experiencing the anxiety that will be evoked by the behavioral challenges they perform.

Work by Baxter and colleagues (1992) has illustrated some interesting correlations between treatment response and changes in neuroanatomy and neurophysiology. Positron emission tomog-raphy scans with 18F-fluorodeoxyglucose were performed on all patients before and after treatment. Compared with nonrespond-ers and control subjects, responders in both the medication and the behavioral treatment groups showed a decrease in activity in the right head of the caudate nucleus. This finding seems to sup-port the notion that both forms of treatment bring about similar changes in neurophysiology which lead to improvement in symp-toms. These results also provide important theoretical links with the serotonin hypothesis described earlier, as basal ganglion structures like the caudate nucleus have been postulated to mediate serotonin function.

Psychotherapy

The use of psychotherapeutic techniques of either a psychoana-lytic or a supportive nature has not been proved successful in treating the specific obsessions and compulsions that are a hall-mark of OCD. However, the more characterological aspects that are part of obsessive–compulsive personality disorder may be helped by a more psychoanalytically oriented approach. As noted earlier, the defense mechanisms of reaction formation, isolation and undoing, as well as a pervasive sense of doubt and need to be in control, are hallmarks of the obsessive–compulsive char-acter. Salzman (1983) and MacKinnon and Michels (1971) have written elegantly on how to approach the maladaptive aspects of this character style in therapy. In essence, the patient must be encouraged to take risks and learn to feel comfortable with, or at least less anxious about, making mistakes and to accept anxiety as a natural and normal part of human experience. Techniques for meeting such goals in treatment may include the psychiatrist’s being relatively active in therapy to ensure that the patient focuses on the present rather than getting lost in perfectly recounting the past, as well as the psychiatrist’s being willing to take risks and present herself or himself as less than perfect.

Neurosurgery

Occasionally, even after receiving adequate pharmacotherapy (including augmentation), adequate behavioral therapy, and a combination of behavioral therapy and pharmacotherapy, pa-tients may still experience intractable OCD symptoms. Such patients may be candidates for neurosurgery. Although criteria for who should receive neurosurgery vary, it has been suggested that failure to respond to at least 5 years of systematic treatment is a reasonable criterion. Frequently used criteria are the follow-ing: a minimum of two adequate medication trials with augmen-tation plus adequate behavioral therapy in the absence of severe personality disorder.

The procedures that have been most successful inter-rupt tracts involved in the serotonin system. The surgical pro-cedures used – anterior capsulotomy, cingulotomy and limbic leukotomy – all aim to interrupt the connection between the cortex and the basal ganglia and related structures. Current stereotactic surgical techniques involve the creation of precise lesions, which are often only 10 to 20 mm, to specific tracts. These procedures have often been done with radiofrequency heated electrodes andmore recently, with gamma knife techniques. Postsurgical risks have been minimized, and in some cases cognitive function and personality traits improve along with symptoms of OCD.

Data compiled from a number of small studies have yielded success rates of 25 to 84% with treatment. However, most samples are small, and the procedures have often differed in both lesion location and size, making it difficult to compare them. However, in a recent prospective long-term follow-up study, all 44 patients received the same procedure (cingulotomy), although some had single procedures and others multiple procedures (Dougherty et al., 2002). This study had several important findings. Clinical improvement occurred in 32 to 45% of patients, depending on the criteria used to rate full or partial response, and the average effect size was 1.27, comparable to that seen in pharmacologic trials (1.09–1.53). However, these changes were not immediately apparent postoperatively, and most patients were encouraged to engage in pharmacotherapy and/or behavior therapy postopera-tively. The longitudinal follow-up component of the study, which was a mean duration of 32 months, allowed the researchers to assess the longer-term impact of the procedure in ways other studies could not. Of particular note, patients continued to show improvement for up to 29 months after surgery without receiving further procedures. As a result, the authors noted that the typical 6-month wait before deciding whether to repeat and extend the lesion may be too brief. In conclusion, neurosurgical treatments offer hope to some of the most severely ill and treatment-resist-ant patients and should therefore be considered. However, which surgical lesions are most effective in which patients still needs much more study.