•No clinical trials have assessed the utility of quetiapine for Tourette’s syndrome, although there have been a couple of positive case reports.

•Olanzapine has not been tested under double-blind conditions either, although it did show “modest to moderate” effectiveness in an open-label trial of 10 adults with Tourette’s syndrome, Dr. Scahill noted. Four of the patients had a reduction of 20 points or more in their Yale Global Tic Severity Scale scores, but two others quit the eight-week trial early due to excessive sedation. Weight gain was a problem for some participants, although the degree varied greatly from patient to patient. Olanzapine showed similar benefits in a second open trial, this one involving 14 patients (pediatric and adult) with Tourette’s syndrome who were treated with 10 to 20 mg/d.

•Ziprasidone has been tested in a single placebo-controlled trial, in which it showed “moderate effectiveness,” according to Dr. Scahill, who was one of the study’s authors. The study population consisted of 28 children (ages 7 to 17) with Tourette’s syndrome or, in one instance, chronic multiple tics disorder; most subjects also had attention-deficit/hyperactivity disorder (ADHD), obsessive-compulsive disorder (OCD), or obsessive-compulsive symptoms. Ziprasidone was started at 5 mg/d (a dose that is not available commercially) and subsequently titrated to a maximum of 40 mg/d (mean, 28 mg/d).

Tic severity declined by an average of 35% in the 16 children who received ziprasidone, compared to a 7% decrease among the 12 placebo recipients. Despite the small numbers of subjects in each group, the difference in response was statistically significant, with an effect size of about 0.7. The most common adverse effect was mild transient somnolence; other side effects (eg, akathisia, dyskinesia) were usually clinically insignificant. The ziprasidone group did not gain weight or experience EKG changes.

Following completion of the study, the researchers treated another 10 patients under an Investigational New Drug protocol, as the Food and Drug Administration (FDA) had temporarily delayed its approval of ziprasidone due to concerns about the drug’s cardiac effects. The 10 patients, all treatment-resistant, had severe Tourette’s syndrome (scores of at least 30 on the Yale tic scale). Unfortunately, treatment yielded an average improvement of only 26%, which is “really less than you’d hope for,” Dr. Scahill noted. Sedation and insomnia were the primary side effects; interestingly, some children had both problems. Again, there were no signs of weight gain or EKG changes (the latter were tracked very carefully at the FDA’s insistence, Dr. Scahill said).

•The best-studied atypical agent for the treatment of Tourette’s syndrome is risperidone, which has been tested in several open trials, a placebo-controlled study, and even a head-to-head comparison trial with pimozide. The results, as with olanzapine, suggest that risperidone is moderately effective.

The three open trials included a Cornell University study involving 38 patients who hadn’t tolerated or responded to conventional neuroleptics or clonidine. Eight of the patients (21%) were unable to tolerate risperidone either and discontinued the study, but 22 participants (58%) did show improvement after one month. Final doses ranged from 0.5 to 9 mg/d (mean, 2.7 mg/d). Another open-label study, conducted at Yale University, involved only seven children and adolescents; treatment with 1 to 1.25 mg/d of risperidone for 11 weeks resulted in a mean 42% reduction in tic scores. Weight gain ranged from 8 to 14 lbs.

In the comparison trial, 50 patients, ages 11 to 50, were treated for 12 weeks with flexible daily doses of either risperidone (mean, 3.8 mg) or pimozide (mean, 2.9 mg). Reduction in tic severity was essentially identical for the two agents (44% to 47%); the degree of improvement was similar when the pediatric patients were analyzed separately. Extrapyramidal symptoms were twice as common in the pimozide group (33%) as in the risperidone group (15%), though the latter medication produced a slightly greater degree of weight gain (4.5 vs 3 kg).

THE LATEST DATA

Finally, Dr. Scahill and his colleagues have completed an eight-week placebo-controlled trial (not yet published) of 34 patients with Tourette’s syndrome or chronic tic disorder of moderate or greater severity (ie, a Yale tic score of at least 20). Participants ranged in age from 6 to 62, though all but eight were children. About half of the subjects had OCD, ADHD, or obsessive-compulsive symptoms.

After a seven- to 14-day placebo wash-in period, each subject received either placebo or risperidone. The active medication was started at a dose of 0.5 mg/d and increased in a flexible manner during the first four weeks of the trial; doses remained constant thereafter. The maximum dose for children was 2.5 mg/d.

At the end of treatment, the 16 patients in the risperidone group showed a 32% reduction in tic symptoms, compared with a 7% decline in the control group. The difference was statistically and clinically significant, Dr. Scahill noted, and a similar degree of improvement was seen when the analysis was restricted to pediatric subjects, who showed a 36% reduction. He added that the decrease in tic severity was often not apparent until at least six weeks had passed, so clinicians should not assume that treatment is ineffective if improvement is not apparent after a month. Side effects of risperidone included weight gain (mean, 6 lbs), increased appetite, sedation, and two cases of social phobia (one of which resolved with dose reduction). Two adult male patients reported sexual dysfunction; there were no instances of extrapyramidal symptoms.

Overall, the evidence suggests that “risperidone is effective for the short-term treatment of tics,” Dr. Scahill concluded. The magnitude of improvement appears to be similar to that of ziprasidone, though not as good as the benefits seen in historical data with haloperidol and pimozide. Although risperidone is well tolerated, patients should be monitored for weight gain, social phobia, and (for adults) sexual dysfunction, he advised. Finally, the lack of extrapyramidal symptoms seen with risperidone and ziprasidone suggests that the 5-HT2 properties of these drugs may protect against such symptoms.

Suggested Reading

ATYPICALS FOR CONDUCT DISORDER

Another indication for which atypicals are widely prescribed is conduct disorder. Michael Aman, PhD, Professor of Psychology and Psychiatry at the Nisonger Center, Ohio State University, Columbus, noted that only two medications, neither an atypical, have been approved for conduct disorder: chlorpromazine and thoridazine. Moreover, he said, those drugs were approved on the basis of early studies that would not meet today’s standards.

Dr. Aman focused on the treatment of conduct disorder in patients who have a below-normal IQ, as such individuals are particularly vulnerable to behavioral disorders. Both traditional and newer antipsychotics have proven useful in this setting. Perhaps the earliest placebo-controlled study in this area, published in 1968, found that haloperidol greatly reduced destructive behavior and aggression against adults in 12 children with conduct disorder. (Not all of the children in the study were cognitively impaired, however.) A second placebo-controlled study, published three years later, also found that haloperidol reduced aggressive behavior, as well as disinhibition.

In 1984 Magda Campbell, MD, and colleagues published a paper that Dr. Aman described as “one of the most important studies [that has been conducted] in kids with conduct problems.” The investigators assigned 61 treatment-resistant children, ages 5 to 12, to receive either haloperidol (1 to 6 mg/d), lithium carbonate (0.5 to 2 g/d), or placebo. The well-designed study showed that both medications led to reductions in aggression, hostility, hyperactivity, and unresponsive behavior, as well as to better scores on the Clinical Global Improvement (CGI)–Severity of Illness scale. Haloperidol, but not lithium, had more side effects than placebo; the two active treatments were not statistically different, however. The one aspect of the findings that is “a little bit troubling,” Dr. Aman noted, is that the improvements were apparent only on physician-rated measures; teachers and nurses did not detect statistically significant changes.

THE DATA DROUGHT ENDS

More recently, in 2000, a small study led by Robert Findling, MD, provided the first controlled data for an atypical agent in conduct disorder; the authors found that risperidone reduced clinician-rated aggression, conduct problems, and delinquent behavior (but not scores on the aggressive behavior subscale of the CGI). Maximum daily doses were 1.5 mg for children who weighed less than 50 kg and 3.0 mg for heavier children.

Two studies published in the past year have also found support for the use of risperidone in behavioral problems. Buitelaar et al reported that scores on the CGI–Severity of Illness scale declined significantly in a sample of 38 adolescents with subaverage IQs who met DSM-IV criteria for disruptive behavior disorders. Daily doses at the end of the six-week trial ranged from 1.5 to 4.0 mg/d (mean, 2.9 mg/d); symptoms worsened during a two-week post-trial washout. The second study involved 13 children with persistent behavioral disturbances and IQs between 66 and 85; risperidone recipients showed a 65% reduction from baseline in Aberrant Behavioral Checklist scores, compared with a 7% reduction in the placebo group. Side effects (including extrapyramidal symptoms) were mild in both studies, though the authors of the first study reported a mean weight gain of 3.5% as well as some cases of “very mild” extrapyramidal symptoms in the risperidone group.

Finally, Dr. Aman presented preliminary findings from two large multicenter risperidone trials. The first trial, conducted at 11 US sites, involved 118 children (ages 5 to 10) with conduct disorder or oppositional defiant disorder; all subjects had IQs of 84 or less and had scores of 24 or greater on the Nisonger Child Behavior Rating Scale, a behavioral measure developed to assess children with intellectual handicaps. After a one-week placebo run-in, study participants were randomized to six weeks of treatment with either risperidone (.02 to .06 mg/kg/d) or placebo. Subjects who had been on psychostimulants prior to enrollment were allowed to continue taking the medication, a provision that not only aided recruitment (60% of the study population had comorbid ADHD) but also made the sample cohort more representative of the general behavioral disorder population.

Overall, Dr. Aman said, the risperidone group showed a “fairly meaty” response: Scores on the conduct problems subscale of the Nisonger Child Behavior Rating Scale declined by an average of 15 points, compared to a 6-point decline in the placebo group. Other subscales documented improvements in adaptive social behavior, insecure/anxious behavior, hyperactivity, and self-injurious/stereotypic behavior. In addition, risperidone users were much less likely than were placebo recipients to withdraw from the study due to poor response (7% vs 24%).

One issue that is sometimes raised when a patient’s behavioral problems improve is whether the medication is specifically targeting conduct problems or whether it is having a “sledgehammer effect” that tones down all behavior. The present study, Dr. Aman noted, offers an interesting clue that risperidone’s effects do have some specificity: Recipients not only showed improvements in conduct but also reductions in lethargy and social withdrawal. “I feel that this is important,” Dr. Aman said, “because if medication is acting as a blunt instrument and de-energizing these kids, you’d expect lethargy to increase, not decrease.”

Adverse effects were reported in 70% of placebo recipients and 98% of risperidone users, though most of these were minor (eg, drowsiness, headache, vomiting). Hyperprolactinemia was much more prevalent with risperidone (13 patients) than with placebo (two patients).

MORE OF THE SAME

The second multicenter risperidone trial, conducted in Canada, used essentially the same design as the US study, had roughly the same number of participants (110), and yielded similar results. For example, patients receiving active medication showed a mean decrease of nearly 50% on the Nisonger Child Behavior scale—benefits, in fact, were apparent after only one week—whereas patients in the placebo group were far more likely than risperidone recipients to withdraw from the trial due to insufficient response. Active treatment also led to improvements in insecurity/anxiety and hyperactivity, and to better scores on the CGI and the Aberrant Behavior Checklist. The primary adverse events were transient somnolence, weight gain, vomiting, and headaches.

Thus, the antipsychotic literature for patients with subnormal intelligence and conduct disorder is limited to five controlled trials of risperidone (totaling 287 patients), three of haloperidol (101 patients)—and not much else. Sedation was apparent in all of the studies; extrapyramidal symptoms were reported in all of the haloperidol trials but in only one of the risperidone studies, so “there seems to be an advantage, as one would expect, with an atypical antipsychotic,” Dr. Aman said. Hopefully, he said, future studies will determine whether atypical antipsychotics truly do have targeted effects in this setting, whether one can determine in advance which patients will gain weight on risperidone or other atypical antipsychotics, and whether antipsychotics are more effective than other medications (eg, anticonvulsants) for conduct disorder.

—Peter Doskoch

Campbell M, Small AM, Green WH, et al. Behavioral efficacy of haloperidol and lithium carbonate: a comparison in hospitalized aggressive children with conduct disorder. Arch Gen Psychiatry. 1984;41:650-656.

Return to table of contents