Antipsychotics. Typical antipsychotics were among the first agents assessed in an organized manner. Haloperidol (Haldol) decreased emotional outbursts and anger in placebo-controlled studies including young children with autism (Anderson et al., 1984). However, its role is limited due to concerns about dystonic reactions and dyskinesias (Campbell et al., 1997).
Atypical antipsychotic drugs have a reduced propensity to cause motor side effects and have largely replaced haloperidol in the clinical treatment of autism. Clozapine (Clozaril), quetiapine (Seroquel), olanzapine (Zyprexa) and ziprasidone (Geodon) have been evaluated in small or uncontrolled studies and case reports (Erickson et al., in press). Each of these drugs, with the exception of ziprasidone, is associated with frequent weight gain. Clozapine has additional liabilities, given its propensity to decrease the seizure threshold and requirement for frequent venipuncture.
Among the atypical antipsychotics, risperidone (Risperdal) has received the most attention in treating maladaptive behaviors in patients with autism. This drug has been the subject of controlled trials in both adults (McDougle et al., 1998) and children (McCracken et al., 2002; Shea et al., 2004). McCracken et al. (2002) compared risperidone (mean dose=1.8 mg/day) to placebo in an eight-week, randomized, double-blind trial in 101 children with autism. Risperidone was markedly efficacious in reducing aggression and irritability. Side effects included weight gain, increased appetite, sedation, tremor and hypersalivation. Shea and colleagues (2004) also reported that risperidone (mean dose=0.04 mg/kg/day) was efficacious in reducing irritability, conduct problems, anxiety and hyperactivity in an eight-week, double-blind, placebo-controlled study of risperidone in a sample of 79 children. They also found significant weight gain, as well as increased pulse and systolic blood pressure in the risperidone-treated group.
Our group recently reported that aripiprazole (Abilify) (mean dose=12 mg/day) was effective for reducing aggression, agitation and self-injury in five youth with PDD (Stigler et al., 2004). During open-label aripiprazole treatment (mean duration=12 weeks), two patients experienced mild somnolence, but no changes in vital signs occurred. Weight loss was also noted. This weight loss may have been associated with patients discontinuing other medications that had caused excess weight gain. We are currently conducting a large placebo-controlled trial of aripiprazole for children and adolescents with autism and aggression.
Other psychotropic agents. Clonidine (Catapres), an α2-adrenergic agonist, was efficacious in two short-term, placebo-controlled trials, but its long-term efficacy has been questioned (McDougle et al., 2003). Guanfacine, a longer-acting α2-adrenergic agonist, was only effective in 19 of 80 (24%) patients with PDD at mean treatment duration of one year (Posey et al., 2004b). Out of 69 patients with significant aggression, it was only effective in 10 (14%) patients.
Case reports and a retrospective review have reported on the use of lithium (Eskalith, Lithobid) or divalproex (Depakote) for aggressive behaviors in youth with autism (McDougle et al., 2003). While we expect more reports on the use of mood stabilizers in the future, our clinical experience, combined with available evidence, suggests that mood stabilizers will not be as effective as atypical antipsychotics in treating aggression.
The tricyclic antidepressant clomipramine (Anafranil) has shown some promise in treating aggression in one placebo-controlled trial (Gordon et al., 1993). Secondary to concerns over tolerability and the availability of safer selective serotonin reuptake inhibitors, the use of clomipramine has been limited.
Fluvoxamine (Luvox) (mean dose=277 mg/day) was evaluated in a 12-week, randomized, placebo-controlled trial in 30 adults with autism (McDougle et al., 1996). Fluvoxamine was associated with significant global symptom improvement, as well as a reduction in maladaptive behavior and aggression. The same group conducted a similar study in 34 children with PDD but noted increased aggression without overall improvement (McDougle et al., unpublished data). Hollander and colleagues (2005) reported on an eight-week, double-blind, placebo-controlled crossover trial of liquid fluoxetine (Prozac) in 45 children and adolescents with PDD. There was a reduction in repetitive behaviors but less reduction of global symptoms. Overall, SSRIs as first-line agents in targeting aggressive behaviors in children with autism is not currently supported by the available literature.
