Evidence-Based Reviews

Using atypicals for patients without psychosis: The strength of evidence varies with the diagnosis

Author and Disclosure Information

Antipsychotics are being investigated for pediatric conditions including anxiety disorder, autism, and Tourette’s syndrome. Studies show benefit in some—but not all—of these uses.



Antipsychotics—particularly the atypicals—have therapeutic properties that make them potential candidates for treating a variety of disorders in children and adolescents. As has occurred in adults, the use of atypical antipsychotics is expanding beyond schizophrenia to pediatric affective and nonpsychotic conditions.

In part 1 of this article, we examined the evidence for using atypical antipsychotics in childhood/adolescent-onset schizophrenia, bipolar disorder, and psychotic depression. Our search of the literature suggested two concerns to keep in mind when prescribing antipsychotics to children and adolescents:

  • Side effects—weight gain, metabolic disturbances, hyperprolactinemia, and cardiac conduction abnormalities—are health concerns for all patients but particularly for children and adolescents, who may require years of exposure to atypical antipsychotics.
  • Administering medications to children and adolescents requires special precautions because younger patients respond differently than do adults to psychotropic medications.

In part 2, we look at more limited evidence for using atypicals in children with anxiety disorders, autism and developmental disorders, Tourette’s and other tic disorders, disruptive behavior disorders, anorexia nervosa (Box 1)1-3, and stuttering (Box 2)4.

Box 1


Anorexia nervosa was treated in the 1960s with chlorpromazine at dosages as high as 1,600 mg/d and often in combination with insulin. Weight increased and hospital stays decreased over the short term, but side effects—including seizures in 5 of 30 patients in one study—greatly complicated treatment. With longer follow-up, weight gain did not improve significantly, and (interestingly) purging behavior emerged.1 Treatment outcomes were unclear in double-blind, placebo-controlled, crossover studies of pimozide and sulpiride.2

Atypical antipsychotics are being investigated for adjunctive treatment of anorexia nervosa. Case reports have described olanzapine’s efficacy in weight gain and psychological improvement in patients with severe symptoms.3

Weight restoration is the most essential step in treating children and adolescents with anorexia nervosa. In this regard, the weight gain associated with using atypical antipsychotics may offer adjunctive benefit. Atypicals also may decrease relapse rates by treating comorbid personality traits (e.g., rigidity and obsessionality) and by restoring cognition in delusional patients. Such claims are speculative but worthy of investigation.

Anxiety disorders: Limited use for antipsychotics

Anxiety disorders are among the most prevalent psychopathologies in the pediatric population, and current treatment recommendations strongly focus on psychotherapeutic interventions. Pharmacologic interventions, however—including imipramine, selective serotonin reuptake inhibitors, and even benzodiazepines—can offer an important adjunct to behavioral and other nonpharmacologic therapies, particularly at the onset of illness and before behavioral techniques are learned.

The American Academy of Child and Adolescent Psychiatry’s 1997 practice parameters for anxiety disorders do not recommend using neuroleptics in the absence of comorbidity—such as Tourette’s syndrome or psychosis—because of concerns about impaired cognition and tardive dyskinesia.5 The clinician should also be aware that “neuroleptic separation anxiety syndrome” has been described in children who developed school phobia in response to haloperidol or pimozide while being treated for Tourette’s disorder.

Similar reports describe separation anxiety in two adolescent boys and one prepubertal boy treated with adjunctive low-dose risperidone for obsessive-compulsive disorder (OCD), attention-deficit/hyperactivity disorder (ADHD), and behavioral disruption. Two of the boys were subsequently treated with olanzapine without a recurrence of anxiety.

It seems unlikely that atypical antipsychotics will play a significant role in managing pediatric anxiety. Controlled studies for this indication are limited. Alternate pharmacologic options are considered safer and are themselves recommended only as adjuncts to other interventions.

Clinicians will no doubt be tempted to try atypicals in lieu of benzodiazepines for severe OCD (and perhaps even for severe anxiety) in this young population. If an antipsychotic trial is initiated, we recommend clear documentation of poor response to other interventions, notations of comorbidity, judicious dosing, and close monitoring. Psychotic symptoms associated with posttraumatic stress disorder may be a reasonable indication for antipsychotic use, but more research is needed.

Autism: Improving behavioral symptoms

Pharmacotherapy for children with autism and pervasive developmental disorders (PDD) generally targets aggression, irritability, stereotypic behavior, hyperactivity, self-abusive behavior, and self-stimulatory behavior. Almost all classes of psychotropics—including antipsychotics, selective serotonin reuptake inhibitors, tricyclic antidepressants, lithium, mood stabilizers, and anxiolytics—have been tested in clinical trials, with varying degrees of success.

Haloperidol has been shown to improve behavioral symptoms, including educational learning.6 Dyskinesias—including tardive dyskinesia—remain a concern, however, with long-term use of haloperidol in children.

Recently, attention has turned to atypical antipsychotics, with their lower risk of extrapyramidal symptoms (EPS). Double-blind, placebo-controlled studies have demonstrated the efficacy of these agents in treating autistic and developmental disorders; risperidone and olanzapine have been studied most extensively.

Risperidone. A 16-week open-label trial of 24 children ages 3 to 6 with autistic disorders demonstrated modest improvement with risperidone, 0.5 mg/d. At least 25% improvement was seen in:


Next Article: