Home | | Psychiatry | Treatment of Cooccurring Psychiatric Disorders in Tourette’s Disorder

Chapter: Essentials of Psychiatry: Childhood Disorders: Tic Disorders

Treatment of Cooccurring Psychiatric Disorders in Tourette’s Disorder

Treatment of Attention-deficit/Hyperactivity Disorder

Treatment of Cooccurring Psychiatric Disorders in Tourette’s Disorder


Treatment of Attention-deficit/Hyperactivity Disorder


Nonpharmacological Approaches


The nonpharmacological approaches to ADHD in Tourette’s disorder are similar to approaches in children without Tourette’s disorder. The presence both at home and at school of a structured environment, consistent behavioral management, and a gener-ally positive, rewarding atmosphere can produce significant im-provement in ADHD symptoms. Increasingly, there are specific programs for children with ADHD that go beyond basic positive programming and include more intensive and specific behavioral approaches.


Pharmacological Treatments


The two major difficulties in the treatment of ADHD in Tourette’s disorder are the risk of side effects from the stimulants and desipramine, arguably the most potent treatment agents for ADHD and the lack of adequate alternatives.




In the early 1970s, a number of reports of induction or exacerbation of tics by stimulant medications raised concerns about the role of stimulants as a cause of Tourette’s disorder. At that time, the con-cern was that stimulants could be causing tics de novo or that in-creases in tic severity would endure even if stimulant medications were discontinued. Concurrent with these reports, other authors noted that tic induction or exacerbation was relatively infrequent and that the beneficial effects in some patients with Tourette’s dis-order outweighed any negative impact on tic severity.


As a result, stimulants have been used infrequently for ADHD in Tourette’s disorder. However, results of short-term and long-term double-blind, placebo-controlled trials with stimulants in Tourette’s disorder are positive and support a role for stimu-lants in some patients with Tourette’s disorder plus ADHD. In-creasingly, psychiatrists are cautiously, and with fully informed consent, using stimulant medication in selected children and adolescents with Tourette’s disorder and ADHD. In the patient in whom tics are increased by stimulants, combined treatment with stimulants and tic-suppressing agents can be used.


More recently, in a large (N 5 136) multicenter double-blind, placebo-controlled trial, children with ADHD and a chronic tic disorder were randomly assigned to clonidine alone, methylphenidate alone, clonidine plus methylphenidate or placebo for the treatment of their ADHD. The results suggest that the active treatments were superior to placebo, with the combination treatment being the most effective. With respect to worsening of tic severity the percentage of children with worsening of tics was similar in each of the medication conditions including placebo. For methylphenidate, 20% reported tic increases compared with 26% on clonidine alone and 22% on placebo. Interestingly tic severity lessened in all active treatment groups even in the methylphenidate group. Other side effects were predictable with sedation commonly associated with clonidine treatment (The Tourette Syndrome Study Group, 2002). Given the controversy regarding the coadministration of clonidine and methylphenidate it is important to note the absence of cardiac toxicity in children on combined medication.


Given the above, informed consent prior to initiating stim-ulant medication should include information regarding the risks for new onset of tics or tic exacerbations. In Table 30.2 is a listing of topics that might be useful to address during informed consent. This listing is not intended to be a comprehensive or exhaustive.




Desipramine is a TCA with prominent noradrenergic activity that has been noted to improve attention and concentration in children and adolescents with ADHD and Tourette’s disorder plus ADHD. Symptom improvement is often significant with lower doses than needed for depression. Side effects are generally limited. The cardiac side effects of increased heart rate and elevation in blood pressure are usually not clinically significant; however, reports of sudden death in children and adolescents taking desipramine have resulted in marked reductions nationally in the use of desipramine in children and adolescents.




Given the concern about the cardiac effects of desipramine, TCAs such as nortriptyline have been used for the treatment of ADHD. The only available report, a chart review, assessed the effect of nortriptyline in children and adolescents with Tourette’s disorder plus ADHD. The majority of subjects experienced mod-erate to marked improvement in both ADHD and tics (Wilens et al., 1993). Although the concern regarding sudden death is less with nortriptyline than with desipramine, it is prudent to obtain baseline and follow-up electrocardiograms.


Clonidine and Guanfacine


Data supporting the efficacy of these agents has been limited to open trial or small controlled trials. In the largest trial to date subjects (N 5 34; mean age of 10.4 years) were randomly assigned to 8 weeks of guanfacine or placebo. Guanfacine was superior to placebo (37 versus 8%) in reduction of the total score on the teacher-rated ADHD Rating Scale and significant differences were also observed in omission and commission errors on a continuous performance test (Scahill et al., 2001).


Selective Noradrenergic Reuptake Inhibitors (SNRIs)


One SNRI has recently been approved for the treatment of ADHD in children and one is under FDA review for major depression. Atomoxetine has been demonstrated to be effective and approved for use in children and adults with ADHD. Given its mechanism of action it is possible that atomoxetine could improve ADHD symptoms without exacerbating tic symptoms. Studies are cur-rently underway. Similarly reboxetine, which has an indication in Europe for depression, may be useful for ADHD in people with tics disorders.


Treatment of Obsessive–Compulsive Disorder


Nonpharmacological Approaches


The positive role of cognitive–behavioral treatments of OCD is well established in adults. Most reports of nonpharmacological treatment of OCD in children and adolescents are case studies. Only one report has a sufficient sample size and a protocol-driven, cognitive–behavioral treatment regimen to begin to establish the possible role of such treatment in children and adolescents with OCD. There are no specific reports of cognitive–behavioral treatment or other nonpharmacological treatments of OCD in adults or children with Tourette’s disorder. Given the success of cognitive–behavioral treatment in OCD, it is likely that patients with Tourette’s disorder and OCD will also be able to benefit from cognitive–behavioral treatment.


Pharmacological Treatments


The number of agents available for the treatment of OCD in pa-tients with and without Tourette’s disorder is increasing. Cur-rently available agents include the TCA clomipramine and the specific serotonin reuptake inhibitors fluoxetine, sertraline, par-oxetine, fluvoxamine and citalopram. Even though only a few of these agents have a specific indication for OCD (clomipramine and fluvoxamine), others may be effective in OCD given their serotoninergic activity. The choice of agent depends on the side-effect profile, the potential drug interactions and the psychia-trist’s familiarity with the drug.


With the exception of clomipramine, all of the available agents have mild and somewhat similar side-effect profiles. Drug interaction issues must be taken into account, especially in chil-dren with complex presentations or multiple medical conditions given the increased possibility of multiple drug regimens. The re-ports of elevated TCA levels in patients receiving TCA and fluox-etine are good examples of unforeseen drug interactions with the specific serotonin reuptake inhibitors.


Increasingly, augmentation strategies are pursued in pa-tients with OCD and with Tourette’s disorder plus OCD, when clinical symptoms remain after initial treatment. A number of strategies have been used, including augmentation with lithium, neuroleptics, buspirone, clonazepam, liothyronine sodium (T3) and fenfluramine. Although positive outcomes of these strate-gies have been reported in open trials, only neuroleptic aug-mentation has shown to be of any benefit in controlled trials. Interestingly, controlled trials of haloperidol combined with specific serotonin reuptake inhibitors in patients with Tourette’s disorder and OCD demonstrated improvement in both tic and OCD symptoms.


Treatment-refractory Cases


Strategies for approaching two types of treatment-refractory symptoms are discussed here: 1) patients who are truly treatment-refractory with severe and impairing symptoms of Tourette’s disorder and OCD, despite conventional and heroic treatments, and 2) patients, often children, who are clinically complex and enigmatic, and whose impairment is disproportionally greater than their tic, obsessive–compulsive, or ADHD symptoms would suggest.


Treatment-refractory Tics


Perhaps the most important “treatment” in patients with severe incapacitating tics is a full clinical reevaluation to assess the ad-equacy of previous evaluations and treatment efforts. It is not un-common for treatment-refractory patients to have had inadequate evaluations and treatment trials.


Two alternative treatment strategies are available for truly treatment-refractory tics. When a single tic or a few tics are re-fractory and impairing, the injection of botulinum toxin into the specific muscle group can be helpful. This strategy is most useful for painful, dystonic tics. Treatment has a long duration of action, but the effect does decrease in 2 to 4 months, and repeated injec-tions may be necessary. Specific side effects are few, other than weakness in the affected muscle. Some patients reported the loss of the premonitory sensation with their botulinum toxin treat-ment. For the psychiatrist, it is essential to work with a neurolo-gist experienced in using botulinum toxin.


There have been reports in the literature and the media concerning the use of neurosurgical approaches for the treat-ment of refractory tics. To date, the optimal size and location of the surgical treatment lesions are not known. There are no well-controlled trials, although some data are available from patients with OCD and tics who were treated for OCD. In these patients, the impact on tic severity was mixed. Because these approaches are particularly controversial, it is important, before considering neurosurgical approaches, to complete a detailed and exhaustive reevaluation to determine whether all other treatment options are exhausted. It is also important that patients who pursue neuro-surgical approaches consider centers of clinical excellence where controlled treatment trials are ongoing.


Treatment-refractory Obsessive– Compulsive Disorder


A similarly thorough and exhaustive reevaluation is critical for patients with Tourette’s disorder plus OCD who present as treat-ment-refractory. Diagnostic reevaluation focuses on whether other psychiatric disorders are present and disabling and whether the current hierarchy of clinical disability considers all conditions.


Pharmacological reevaluation is especially critical be-cause there are an increasing number of new medications and potential medication combinations. Rather than repeated change from one antiobsessional agent to another, consideration can be given to augmentation strategies, because they take less time than changing agents and may offer synergistic benefits. Low-dose neuroleptic augmentation is the best first choice; controlled trials support the use of low-dose neuroleptics for augmentation of serotonin reuptake inhibitors in OCD. Lithium and T3 are proven, effective augmenters of antidepressants for depression, yet neither is proven effective in OCD. Because of the frequent overlap of OCD and major depression, lithium or T3 augmenta-tion may be the next best choice.

Treatment-refractory or malignant OCD has been the psy-chiatric disorder most frequently treated with neurosurgical in-terventions in the modern era. Whereas it is a major treatment in-tervention, the surgical approaches are somewhat better defined, and the outcome in severe cases is often positive. Also, medical centers are available that specialize in the presurgical work-up and the neurosurgical procedure.


Clinically Complex Patients


Clinically complex patients may be severely impaired without having severe tic or OCD symptoms. The clinically complex patient is often a diagnostic dilemma with additional diagnoses complicating the clinical picture. In addition, patients can be-come clinically complex when otherwise straightforward treat-ments are a challenge to implement.




In clinically complex patients, the diagnostic challenge is not an accurate assessment of tics, ADHD, obsessive–compulsive symptoms, or LDs, although this is important. In clinically com-plex patients, the diagnostic goal is to identify what other condi-tions or factors may be present that make the current treatment approaches difficult.


From a strictly diagnostic point of view, it is the additional psychiatric conditions beyond Tourette’s disorder, OCD, ADHD and LDs that often escape clinical observation and result in diag-nostic dilemmas and treatment failures.


Treatment Implementation


Clinical problems occur when the treating psychiatrist does not have access to critical information or is not in control of the treat-ment process. Traditionally, psychiatrists develop a relationship with the patient and the other major figures in the patient’s life. Given the current clinical climate, a comprehensive level of in-volvement can be overwhelming and enormously time-consum-ing for the psychiatrist. Because it is increasingly difficult for the psychiatrist to be as involved as necessary, problems with poorly coordinated team efforts and the psychiatrist’s lack of awareness of important clinical issues can have a negative impact on the treatment of a patient.


Psychiatrists who work with children and adolescents may wish to consider changes in their treatment approaches to these patients. Experience in tertiary care centers suggests that expanded time with the parents is a critically important and ef-ficient approach to care. Psychiatrists who form a treatment part-nership with families, respecting and addressing their concerns, educating them about Tourette’s disorder, training them to evalu-ate and manage complex behaviors, and empowering them to be an effective advocate for their child, are providing good care. In working directly with families, the collection of important infor-mation regarding the family’s and patient’s functioning is direct and regular, and often small interventions can produce changes in family functioning that have a positive ripple effect throughout the life of the child.


Pharmacological Treatment Dilemmas


It is often difficult to obtain accurate information regarding side effects and treatment response in child patients. Parents, children and psychiatrists, in spite of a good collaborative effort, may have different understandings of the target symptoms, side-effect pro-file, and what constitutes a positive clinical response. This ambi-guity makes any but the most robust clinical responses difficult to observe. Again, experience at tertiary referral centers suggests that the lack of a clinical response to medication in complex pa-tients may often be related to inadequate monitoring of medicine effects and inadequate treatment trials.


Clinically complex patients may not have a robust re-sponse to a single medication but may require multiple medica-tion trials to identify which medications offer the most benefit, and in which combination. Sequential treatment trials are dif-ficult for all involved, especially children and families, who are often looking for a single powerful intervention. With the added complexity of treatment, there is the added risk of confusion and the need for an excellent psychiatrist–patient–family relation-ship. In those cases in which the relationship is not optimal, it is possible that a patient may not have the maximal clinical benefit of pharmacological interventions.


With increasing numbers of available psychotropic medi-cations, psychiatrists become increasingly less experienced with the range of clinical effects and side effects in individual medica-tions. In clinically complex patients, the prescription of unfamil-iar medications may be necessary but may add to the risk that a trial will be discontinued prematurely because of doubt about a side effect. In addition, unusual side effects, such as the apathy or disinhibition syndromes seen with some patients receiving the specific serotonin reuptake inhibitors, may go unnoticed and add to clinical morbidity.


Whereas pharmacological interventions offer great prom-ise, clinical experience suggests that excellent diagnostic skills, good relationships with the patient and family, time, and a keen eye for effects and side effects are necessary for benefits to be realized. Less intensive efforts may make patients appear more complex than necessary.

Study Material, Lecturing Notes, Assignment, Reference, Wiki description explanation, brief detail
Essentials of Psychiatry: Childhood Disorders: Tic Disorders : Treatment of Cooccurring Psychiatric Disorders in Tourette’s Disorder |

Privacy Policy, Terms and Conditions, DMCA Policy and Compliant

Copyright © 2018-2024 BrainKart.com; All Rights Reserved. Developed by Therithal info, Chennai.