Treatment of Comorbid Disorders in Autism: Which Regimens Are Effective and for Whom?

Jessica A. Hellings, MD, Department of Psychiatry, University of Kansas Medical Center, Kansas City, KS

Abstract

In the past 10 years, many new medications with the potential for greater safety and fewer side effects have become available to treat mental illness. Yet, clinicians treating persons with autism continue to prescribe older neuroleptic agents targeting generic behavior problems -- despite the absence of psychosis and in the face of the serious accompanying risks of tardive dyskinesia and neuroleptic malignant syndrome. This article presents a model using basic medical and DSM IV-based comorbidity groups of seizure-related behavioral symptoms; hyperactive-inattentive cluster symptoms; tics, Tourette syndrome, and movement disorder symptoms; compulsive-sameness oriented-explosive symptoms; mood disorder symptoms; and other miscellaneous behavioral conditions to guide clinicians in selecting an effective treatment approach.

Introduction

Autistic disorder is one of a group of pervasive developmental disorders (PDDs) within a DSM-IV category that also includes Rett syndrome and childhood disintegrative disorder; Asperger syndrome is also sometimes included. Ideally, all individuals with PDDs should receive regular clinical psychiatric and neurologic assessments at a specialty center throughout their life span for identification of treatable conditions that may compound their impairment. Should seizures be suspected, a further clinical workup including an electroencephalogram can be pursued. Assessments should be obtained annually or more frequently if behavior problems are deemed to be significantly interfering with the child's progress.

These evaluations are generally available at centers specializing in the assessment and treatment of these disorders, such as university hospitals or university-affiliated programs. Treatable conditions associated with autism and other PDDs include seizure disorder, attention deficit hyperactivity disorder (ADHD), Tourette syndrome, obsessive-compulsive disorder, and bipolar mood disorder. In addition, autistic persons with severely disruptive behavioral disturbances, such as aggressive, destructive, or self-injurious behavior, are more likely to seek and receive treatment than individuals with comorbid disorders that cause withdrawn behavior, such as major depression. A great deal more research is needed before discussion focuses primarily on comorbid conditions.

Increasingly, diagnoses and treatment of the aforementioned types of comorbidity are based on DSM-IV criteria.[1,2] This specific clinical approach appears to produce more predictable and successful treatment responses in comparison with simply prescribing drugs for the non-specific diagnosis of "behavioral control". Rutter[3] discusses the methodologic difficulties in determining the accuracy of comorbidity in psychiatry, and whether this difficulty relates to errors in the prevailing psychiatric diagnostic concepts versus obscure knowledge of neurobiologic disturbances. More research is necessary before the comorbid groups can be validated as meaningfully different and underlying mechanisms can be identified. In the meantime, the syndromal diagnoses provide a valuable guide to which treatments are most likely to work in any given PDD patient who is thought to have a comorbid psychiatric or neuropsychiatric disorder.

In the community, the use of new pharmacotherapeutic agents in clinical practice often precedes the desired large-scale efficacy and safety studies that would evaluate the drugs' effects on populations such as children, the elderly, and the developmentally disabled. Thus, many new psychoactive drugs - including selective serotonin reuptake inhibitors (SSRIs), mood stabilizers such as valproic acid and gabapentin, and the serotonin-dopamine antagonists (SDAs), already in wide use in those with normal IQs - remain underutilized among individuals with autism and other PDDs. In many cases, clinicians today continue prescribing older drugs that were more frequently used to treat the nondevelopmentally disabled population 10 to 20 years ago.

Aman and colleagues[4] surveyed 838 individuals with autism and found that, apart from those on antiseizure medications, the largest group (12.2%) was still treated with the older "classic" antipsychotic drugs. These drugs, including chlorpromazine, thioridazine, and haloperidol, are widely prescribed and even in the absence of movement disorders such as tardive dyskinesia and neuroleptic malignant syndrome they can produce cognitive dulling, motor slowing, and other long-term effects. Apart from prescribed medications, many of the currently used nonscientific interventions for autism, such as megavitamin treatment, are expensive and potentially harmful. The newer medications, when prescribed with caution for comorbid syndromal conditions, afford more effective treatment with potentially greater safety and fewer side effects. Nevertheless, scientifically designed double-blind studies are urgently needed to establish the safety and efficacy profiles of these medications in the autism population. The Aberrant Behavior Checklist-Community[60] is a useful adjunctive standardized assessment instrument which a parent or teacher can fill out at clinic visits, rating behavior problems observed at home or at school.

Generally, before medications are introduced, nonpharmacologic interventions should be considered. These are best managed by a behavioral psychologist who specializes in developmental disabilities. However, seizure conditions (which may be subclinical), psychosis, mania, or severe ADHD require at least some pharmacologic treatment to allow the affected patient to function in behavioral programs. An improvement in attention span or stabilization of mood, for example, may render the patient more capable of working toward goals and appreciating rewards.

It is important to emphasize the additional vigilance needed on the part of clinicians and caregivers for the presence of underlying medical conditions that may lead to acute or insidious behavioral deterioration. Persons with autism may be unable to communicate feelings of unwellness verbally to caregivers. Multiple factors involving illness may then manifest as aggression or self-injury, especially if the person is pressured to continue with daily routine or work activities. Common underlying causes of a new onset of behavior problems include infections or untoward effects of medications. Influenza, ear infections, dental problems, migraine and pain of any kind are a few of the common causes of a re-emergence of problems. Every effort must be made in these cases to treat the underlying condition medically and provide a supportive environment for the person without adding unnecessary psychoactive agents. The reader is alerted to the fact that, while researchers at the University of Kansas, Kansas City, have found this approach to be of great practical use, it remains tentative until supported by further published research.

haut de la page

Clinical Approach to Pharmacotherapy

The neurobiologic/psychiatric conditions occurring with autism that may respond to pharmacologic treatment and thereby relieve confounding symptoms that impair the autistic individual's ability to function can be subdivided into 6 large categories:

Seizure-related behavioral symptoms
Hyperactive-inattentive impulsive-distractible symptom cluster
Tics, Tourette syndrome, and movement disorders
Compulsive-sameness oriented-explosive symptom cluster
Mood disorder symptom cluster
Other or nonspecific behavioral symptoms

haut de la page

Seizure-Related Behavioral Symptoms

Behavioral problems resulting from cerebral seizure activity or the medications used to treat seizures are discussed first and foremost, to stress the importance of this issue. At least 25% of persons with autism are likely to experience seizures.[5] Onset may occur during infancy, childhood, or adolescence, with a lower new onset rate later in life. A routine electroencephalogram (EEG) and brain scan should be obtained in all newly diagnosed cases of mental retardation or autism. Neurologic referral is appropriate for individuals with staring spells, syncopal episodes, or behavioral deterioration after stopping or starting antiseizure medications.

Behavioral side effects of the barbiturate-related agents, phenobarbital and phenytoin, may include irritability and depression as well as aggressive behaviors such as biting, pinching, and kicking.[6] Ethosuximide may produce psychosis. Carbamazepine not uncommonly causes behavioral disinhibition or worsening of mania in some persons with mental retardation [7] and can result in a low serum sodium.[8] Gabapentin has been associated with a worsening of hyperactivity in some cases.[9]

Valproic acid, however, appears to be associated with fewer behavioral side effects. This agent has been shown to have mood-stabilizing effects, as well as beneficial effects on agitation and aggression in animals and in humans.[10] Changing from another seizure medication to valproate by the add-on method to achieve therapeutic blood levels of 50 to 125 mcg/mL, followed by a gradual tapering of the other medication, has often proven extremely useful.

haut de la page

Hyperactive-Inattentive Impulsive-Distractible Symptom Cluster

Stimulants methylphenidate (MPH), dextroamphetamine sulfate (DEX), pemoline, and new combination long-acting preparations such as Adderall are useful but still understudied in children and adolescents with autism. Clear ADHD symptoms of inattention, distractibility, impulsivity, and hyperactivity are stronger predictors of response than are age, IQ, or other comorbid diagnoses.[11] Children with ADHD and less severe degrees of MR may show greater response to stimulants than those with more severe or profound MR [61] Use of these stimulants in those with autism and/or severe or profound mental retardation, has met with mixed results, though Demb reported that dramatic improvement may occur in individual cases.[12] At the University of Kansas at Kansas City, we have found that low to moderate doses are most useful, that is, MPH 1 mg/kg/day, DEX 0.5 mg/kg/day, or pemoline approximately 2 mg/kg/day. Higher doses may produce withdrawn behavior as well as worsening of stereotypic movements and self-picking.[13-15]

Other side effects reported with high-dose treatment include anxiety, worsening of tics, nail-biting, weight loss, and if extremely high doses are used, psychosis. Pemoline carries an additional (though rare) risk for hepatitis, which may prove fatal.[16-18] Therefore, pemoline is not used as a first-line medication, and when it is used, liver enzymes should be monitored. In these cases, it is important to educate family members, alerting them that they should contact the physician if jaundice or dark urine develops.

haut de la page

Tricyclic antidepressants (TCAs)

These drugs may be effective in persons who show a partial response to stimulants but remain overactive, especially if there is associated anxiety, insomnia, or low weight. Our experience suggests that the TCA amitriptyline is often more effective than imipramine or desipramine in the treatment of overactive disruptive children and adolescents with autism, although this observation requires scientific study. More side effects can be expected with combination treatment than with monotherapy, though we have not found this to be a problem. A study of 16 hospitalized normal IQ children treated with MPH and desipramine for ADHD and comorbid anxiety, dysthymia, and major depression showed no side effects that were unusual or serious.[19] Test scores on higher order learning showed greater improvement on this combination therapy as compared with either drug alone, in this same group of children with normal IQs.[20]

TCAs may be useful also as monotherapy or in combination with stimulants for autistic children and adolescents with ADHD, although further studies are needed to confirm our experience. Our preliminary results are also more favorable with amitriptyline alone rather than with imipramine or other TCAs in children with autism and ADHD, but the reason for this is unclear. It remains vital for parents to keep all TCAs literally locked up, because of their potential lethality in overdose. Both of the TCAs, clomipramine and desipramine were superior to placebo in treating hyperactivity in seven autistic children in a 10-week randomized crossover study; however, clomipramine was superior to desipramine on measures of obsessive-compulsive behaviors, anger, and autism. [21,22]

Before prescribing a TCA, clinicians should perform a thorough personal history, family history regarding sudden deaths, physical examination, and electrocardiogram (ECG). If any doubt remains, a cardiology consultation should be obtained. Monitoring of the pulse, blood pressure, weight, ECG, and blood levels of TCA is necessary at follow-up visits (intervals of at least 3 months) to ensure safe usage. Biederman[23] has discussed ECG and cardiac parameters, such as a QTC(complex) of less than 440 millivolts, for the purpose of monitoring.

At the University of Kansas, we initiate a TCA at a dose of approximately 1mg/kg/day.

Thereafter, a gradual increase is made in the dosage, by clinical response to achieve a total dose of 2 to 5 mg/kg/day. The risk of seizures, cardiac side effects, and sudden death increases sharply when blood levels rise above the upper limit of the therapeutic blood range. The blood levels are checked 12 hours after the last dose administered, for example at 9 am after a 9 pm dose the night before. The therapeutic range is recorded as 125-250 mg/mL although this range was established for the normal IQ population.

Several patients have demonstrated a good clinical response with blood levels in the subtherapeutic or low therapeutic range; the TCA dosage is adjusted in accordance with clinical status. Also, the dosage may be divided if side effects, such as dry mouth or urinary retention, become a problem.

haut de la page

Clonidine and Guanfacine

Some children can tolerate neither stimulants nor TCAs (eg, behavioral disinhibition with TCAs, worsening of stereotypies with stimulants). In these cases, clonidine or guanfacine may be more effective.

Children with a hypomanic mood, excitable behavior, or comorbid tics may respond best to low doses of the alpha-adrenergic agonist, clonidine. A double-blind crossover trial of clonidine versus placebo in eight autistic children showed mild improvement in hyperactivity and irritability on the active drug. Unwanted effects of sedation, decreased blood pressure, and eventual tolerance to the beneficial drug effects were also noted.[24] It remains unclear whether clonidine or guanfacine produce any significant cognitive benefits.

haut de la page

Antipsychotics

Trials of antipsychotic medication are not regarded as appropriate treatment for uncomplicated ADHD. These medications should be reserved for the most severe, resistant cases because of the chronicity of the condition as well as the potential risks of acute dystonia, tardive and withdrawal dyskinesia, and neuroleptic malignant syndrome. Although Campbell and coworkers[25] have demonstrated a significant therapeutic response in autistic children to the classic neuroleptic haloperidol, this drug is generally avoided because of the high incidence of associated dyskinesias.[26]

Although haloperidol remains the best validated antipsychotic at this time, our group and others have large double-blind placebo controlled studies in progress in persons with MR and/or autism. Classic antipsychotics of low potency, such as thioridazine and chlorpromazine, though still widely used, have considerable antimuscarinic and sedation side effects. These side effects may further impair the person's cognitive and motor functioning. Drug holidays appear especially problematic with neuroleptics, due to withdrawal dystonia and akathisia. A withdrawal syndrome comprised of irritability, agitation, insomnia, and aggression lasting 1 to 9 months may also occur if neuroleptics are tapered after chronic use. [27]
Newer antipsychotics, such as the SDAs, risperidone, and olanzapine, may have potentially lower risks in terms of serious side effects; however, long-term studies are needed to evaluate their proper use. Risperidone was associated with significant clinical improvement in an open trial involving 20 children and adolescents, most of whom had autism or PDD with ADHD or other disorders.[28] However, several of these patients were also on other medications including lithium, carbamazepine, or valproate, making it difficult to separate out the effects of risperidone.

haut de la page

Naltrexone

The value of naltrexone treatment as a routine trial for hyperactive, self-injurious autistic children remains debatable,[29] and generally our institution does not employ this regimen.

haut de la page

Tics, Tourette Syndrome, and Movement Disorders

As mentioned, researchers at the University of Kansas, Kansas City, have found low doses of clonidine or guanfacine to be useful in these cases. Preliminary results from an open-label study of five prepubertal autistic children showed a significant improvement in movement disorders using the serotonergic tricyclic clomipramine.[30] It is important to monitor morning trough levels of the tricyclic, using the aforementioned guidelines. Prescription of low doses of an SDA, such as risperidone, may be necessary for resistant cases.

haut de la page

Compulsive-Sameness Oriented-Explosive Symptom Cluster

Individuals with these symptoms have compulsive behaviors such as arranging, organizing, or hoarding a restricted range of objects, to a degree that interferes with their daily functioning. When prevented from engaging in these activities, or if redirected too rapidly to another activity, these persons may manifest explosive behaviors such as yelling, aggression, or property destruction. Dysthymic or depressive symptoms such as crying spells, loss of interest, loss of weight, or appetite and behavioral withdrawal may be present. In other cases, there may be associated manic-like symptoms of overactivity, pacing, insomnia, sexual preoccupation, and euphoria or irritability. In the latter cases, treatment with one or more mood stabilizers, such as valproic acid, lithium, and/or gabapentin, is often necessary before an impact can be made on the compulsive behavior. Conventional treatment with SSRIs is used for this indication.

In two open-label studies, explosive outbursts, aggression, and self-injurious behavior responded well to sertraline or fluoxetine.[31,32] In the fluoxetine series, hypomanic-like side effects, including restlessness, insomnia, overactivity, decreased appetite, and agitation, occurred in approximately 25% of subjects. In another double-blind placebo-controlled study, the SSRI, fluvoxamine, reduced compulsive behaviors and aggression in 15 (50% ) of 30 adults with autism.[33]

In our experience, starting low and going slow has proven effective. For example, we might begin with 12.5 mg daily of sertraline in 5-year-olds, with a gradual increase to 25mg daily or 50mg daily in older elementary school-age children. However, behavioral disinhibition is not uncommon with higher doses, and in many cases, is resolved by reducing the SSRI dose.[34]

Fluoxetine is available in liquid form, permitting doses of less than 10 mg daily to be administered. Clinicians should exercise caution regarding drug interactions related to the propensity for SSRIs, especially paroxetine and fluoxetine, to inhibit cytochrome P450 enzymes involved in the metabolism of many drugs.[35] Potential interactions include those with phenytoin, beta blockers, theophylline, ketoconazole, and others. Information about sexual side effects of SSRIs may be difficult to elicit in nonverbal individuals.

Results of trials with the serotonergic agent, fenfluramine, have generally been disappointing. An 8-week placebo-controlled parallel design study of fenfluramine in 28 children with autism (mean age, 4.57 years) not only failed to show significant results, but also suggested a negative effect on learning.[36]

haut de la page

Mood Disorder Symptom Cluster

Unipolar depressive illness often responds to low doses of the SSRIs, as described. For resistant cases, venlafaxine in low doses has proven useful, though controlled studies are needed in persons with this disorder. Bipolar illness in autistic persons is more often of a chronic, atypical, mixed, or rapid-cycling nature. Manic symptoms include euphoria, mood lability, irritability, excitability or laughing spells, or crying spells. Overactivity associated with mania may manifest as constant pacing, rocking, or walking great distances away from home. Insomnia, hypersexuality, pressure of speech or incessant loud vocalizations, and aggression are other symptoms.[37] Mood disorder symptoms may occur together with any of the other aforementioned categories of symptoms. Polypharmacy -- using mood stabilizers together with other psychoactive medications -- may then be justified.[11]

To treat aggression, an open-label trial showed valproate to be effective and safe as an add-on treatment in 10 adolescents with autism or PDD and accompanying overactivity, euphoria or irritability, decreased sleep, and hypersexuality.[38] Doses resulting in blood levels of 75-100 mcg/mL proved more effective than lower doses. Half of the patients were on other psychoactive medications, including MPH, clonidine, lithium, thioridazine, and sertraline.

We have again found that to "start low and go slow" minimizes side effects, although, controlled studies are needed. We start with 250 mg at night for 2 days, then 250 mg twice a day for 2 days. We then increase gradually in this fashion to approximately 20 mg/kg/day, or to achieve a morning trough blood level of 75 to 100 mcg/mL.

Results of another open-label series of 10 adolescents suggest the efficacy and safety of valproic acid for aggression.[39] In this series, patients with mild and moderate mental retardation were included; however, the number of subjects with autism and/or PDD was unclear. Another series, this time of three autistic children treated in an open-label trial with valproate at doses of 250 to 375 mg/day resulted in language and social improvements in all subjects[40]; all had documented EEG abnormalities. Valproate has the advantage of availability in liquid form (Depakene), although it may be associated with slightly more gastrointestinal side effects and should be given with food. It is worth noting that in 3 other open-label trials involving 25 patients with mental retardation, valproate produced marked improvement in assaultiveness, aggression, and self-injurious behavior.[8,41,42]

In chronic, mixed, or rapid-cycling mania, more than one mood stabilizer may be necessary. Addition of a low dose of lithium or gabapentin in divided doses with meals may enhance the partial improvement achieved with valproate. Gabapentin is administered at 10 to 20 mg/kg/day in divided doses with meals, and thus far, appears relatively safe, provided that renal function is not impaired.[42] Levels of gabapentin are not routinely monitored, though a level may be obtained to check patient compliance. In severe cases or during exacerbations, addition of a low-dose neuroleptic to two mood stabilizers may be necessary. Low doses of risperidone, olanzapine or loxapine may be best in terms of less movement side effects or sedation.

Lower-potency neuroleptics, such as thioridazine or chlorpromazine, often produce sedation and sluggishness, blurred vision, and postural hypotension. High-potency classic antipsychotics, such as haloperidol, often produce severe extrapyramidal side effects after a period of some delay. These may interfere with persons' ability to eat and drink without spilling, or may even cause the chair they are sitting on to shake. Addition of and rapid dosage increase of anticholinergic drugs, such as benztropine, may further complicate the situation by causing mental confusion. Tremor is more likely using the combination of lithium, valproate, and a neuroleptic. McElroy and Weller[43] describe the drawbacks of antipsychotic use for more than a 3-week period in bipolar disorders. These disadvantages include an increased risk for tardive dyskinesia, the masking of mood response, and the worsening of bipolar depressive symptoms.

haut de la page

Other or Nonspecific Behavioral Symptoms

Children with neurobiologic/psychiatric and developmental disabilities, including autism, are more likely to experience sleep-wake cycle disturbances. An open-label trial of oral melatonin 2.5 to 10 mg (fast-release preparation) given at bedtime produced positive results in 82% of 100 patients,[44] and appeared safe. Onset of effect appeared to be with the first dose in some, however, in others, several months of treatment together with initiation of rigorous sleep routines and a modified bedroom environment were needed. Some multidisabled children were able to discontinue melatonin treatment after several months, maintaining a normal sleep pattern. Double-blind placebo-controlled studies are needed.

Controlled clinical trials of the opiate antagonist naltrexone have been done in the autistic population. Although results show mild improvement in hyperactivity, no improvement in core symptoms or aggression were noted.[27,34,45-49]

Studies of the ACTH 4-9 analogue org 27766,[50] and of the 9-amino acid peptide oxytocin[51] are still preliminary.

Ratey and colleagues[52,53] reported on an open study in which propranolol reduced aggression in six of eight hospitalized autistic adults using doses of 100 to 420 mg/day. Seven of the eight subjects, however, were also treated with concomitant antipsychotics or mood stabilizers.

Several additional preliminary reports of risperidone efficacy, have been published.[54-58] In these studies, risperidone was used to target nonspecific "dysfunctional behavior," aggression, self-injury, and property destruction. Dosages varied between 1 and 12 mg/day. Major side effects were somnolence, weight gain, galactorrhea, extrapyramidal side effects, and akathisia at higher doses. The novel antipsychotic agent, clozapine, was effective in reducing treatment-resistant self-injurious behavior in one patient.[59] Newer antipsychotics with different receptor-blocking profiles, such as olanzapine, and quetiapine, may offer the advantages of risperidone without the risk of acute and chronic movement disorders.

haut de la page

Summary

The development of many new classes of medications, as well as the range of pharmacologically different drugs within each class, is broadening the options for understanding and treating individuals with autism and comorbid psychiatric illness. Similar DSM-IV diagnostic categories are a valuable guide to diagnosis in this population although some criteria cannot be met, for example, in nonverbal individuals ongoing vocalization may take the place of pressured speech. While maintaining a vigilance for seizures, physical illnesses, and medication side effects, clinicians are now able to use low doses of psychoactive agents with potentially greater safety and fewer side effects to achieve good outcomes. Nevertheless, scientifically controlled, long-term studies are urgently needed.

haut de la page

References

- American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, 4th ed. Washington, DC, American Psychiatric Association, 1994.

- Sovner R. Behavioral and affective disturbances in persons with mental retardation: a neuropsychiatric perspective: preface. Semin Clin Neuropsychiatry. 1996;1:90-93.

- Rutter M. Child psychiatric disorder: measures, causal mechanisms, and interventions. Arch Gen Psychiatry.1997;54:785-789.

- Aman MG, Van Bourgondien ME, Wolford PL, Sarphare G. Psychotropic and anticonvulsant drugs in subjects with autism: prevalence and patterns of use. J Am Acad Child Adolesc Psychiatry. 1995;34:1672-1681.

- Bailey A, Phillips W, Rutter M. Autism: towards an integration of clinical, genetic, neuropsychological, and neurobiological perspectives. J Child Psychology Psychiatry. 1996;37:89-126.

- Trimble MR. Anticonvulsants in children and adolescents. J Child Adolescent Psychopharmacology.1990;1:107-126.

- Freeman TW, Clothier JL, Pazzaglia P, Lesem MD, Swann AC. A double-blind comparison of valproate and lithium in the treatment of acute mania. Am J Psychiatry.1992 Jan;149(1):108-111.

- Kastner T, Friedman DL, Plummer AT. Valproic acid for the treatment of children with mental retardation and mood symptomology. Pediatrics. 1990;80:467-472.

- Lee DO, Steingard RT, Cesena M, Helmes SL, Riviello JJ, Mikati MA. Behavioral side effects of gabapentin in children. Epilepsia. 1996;37:87-90.

- Wilcox J. Divalproex sodium in the treatment of aggressive behavior. Ann Clin Psychiatry. 1994;6:17-20.

- Spencer T, Biederman J, Wilens T. Pharmacotherapy of attention-deficit/hyperactivity disorder: a life span perspective. In: Psychopharmacology Across the Life Span Section IV of Review of Psychiatry, Vol 16,

- Dickstein, JL, Riba, MB, Oldham, JM (eds). Washington, DC, American Psychiatric Press Inc; 1997:87-127.

- Demb H. Use of Ritalin in the treatment of children with mental retardation. In: Greenhill L, Osman B, eds. Ritalin: Theory and Patient Management. New York, Mary AnnHeliert; 1991:155-170.

- Birmaher B, Quintana H, Greenhill L. Methylphenidate treatment of hyperactive autistic children. J Am Acad Child Adolescent Psychiatry.1988;27:248-251.

- Campbell M, Green WH, Deutsch SI. Child and Adolescent Psychopharmacology, Vol. 2. Beverly Hills, Calif., Sage; 1985.

- Strayhorn J, Rapp N, Donina W, et al. Randomized trial of methylphenidate for an autistic child. J Am Acad Child Adolescent Psychiatry. 1988;27:244-247.

- Berkovitch M, Pope E, Phillips J, Koren G. Pemoline-associated fulminant liver failure: testing the evidence for causation. Clin Pharmacol Therapy. 1995;57:696-698.

- Nehra A, Mullick F, Ishak K, Zimmerman H. Pemoline-associated hepatic injury. Gastroenterology. 1990;99:1517-1519.

- Pratt A, Dubois RS. Hepatotoxicity due to pemoline (Cylert): a report of two cases. J Pediatr Gastroenterol Nutr.1990;10:239-241.

- Pataki C, Carlson G, Kelly K, et al. Side effects of methylphenidate and desipramine alone and in combination in children.Am J Psychiatry. 1993;32:1065-1072.

- Rapport M, Carlson G, Kelley K, et al. Methylphenidate and desipramine in hospitalized children, I: Separate and combined effects on cognitive function. J Am Acad Child Adolesc Psychiatry. 1993;32:333-342.

- Gordon CT, Rapoport JL, Hamburger SD, State RC, Mannheim GB. Differential response of seven subjects with autistic disorder to clomipramine and desipramine. Am J Psychiatry. 1992;49:363-366.

- Gordon CT, State RC, Nelson JE, et al. A double-blind comparison of clomipramine, desipramine, and placebo in the treatment of autistic disorder. Arch Gen Psychiatry. 1993;50:441-447.

- Biederman J. Sudden death in children treated with a tricyclic antidepressant: a commentary. Biological Therapies in Psychiatry Newsletter.1991;14:1-4.

- Jaselskis CA, Cook EH, Fletcher KE, Leventhal BL. Clonidine treatment of hyperactive and impulsive children with autistic disorder. J Clin Psychopharmacol. 1992;12:322-327.

- Anderson LT, Campbell M, Adams P, Small A, Perry R, Shell J. The effects of haloperidol on discrimination learning and behavioral symptoms in autistic children. J Autism Dev Dis. 1989;19:227-239.

- Campbell M, Adams P, Perry R, Spencer EK, Overall JE. Tardive and withdrawal dyskinesia in autistic children: a prospective study. Psychopharmacol Bull. 1988;24:251-255.

- Szymanski LS, Crocker AC. Mental retardation. In: Kaplan HI, Sadock BI, eds. Textbook of Psychiatry V, 5th ed. Baltimore, Md., Williams & Wilkins; 1989:1728-1771.

- Hardan A, Johnson K, Johnson C, Hoecznyj B. Case study: Risperidone treatment of children and adolescents with developmental disorders. J Am Acad Child Adolesc Psychiatry. 1996;35:1551-1556.

- Campbell M, Harris JC. Resolved: autistic children should have a trial of naltrexone: affirmative, negative, affirmative rebuttal, negative rebuttal. J Am Acad Child Adolesc Psychiatry. 1996;35:246-251.

- Brasic JR, Barnett JY, Kaplan D, et al. Clomipramine ameliorates adventitious movements and compulsions in prepubertal boys with autistic disorder and severe mental retardation. Neurology. 1994;44:1309-1312.

- Hellings JA, Kelley LA, Gabrielli WF, Kilgore ER, Shah P. Sertraline response in adults with mental retardation and autistic disorder. J Clin Psychiatry. 1996;57:333-336.

- Cook EH, Rowlett R, Jaselskis C, Leventhal BL. Fluoxetine treatment of children and adults with autistic disorder and mental retardation. J Am Acad Child Adolesc Psychiatry. 1992;31:739-745.

- McDougle C, Naylor S, Cohen D, Volkmar F, Heninger G, Price L. A double-blind, placebo-controlled study of fluvoxamine in adults with autistic disorder. Arch Gen Psychiatry. 1996; 53: 1001-1008.

- Riddle MA, King RA, Hardan MT, et al. Behavioral side effects of fluoxetine in children and adolescents. J Child Adolescent Psychopharmacol. 1990/1991;1:193-198.

- Preskorn SH. Clinical Pharmacology of Selective Serotonin Reuptake Inhibitors. 1st ed. Wichita, Ks, Professional Communications Inc.; 1996.

- Campbell M, Overall JE, Small AM, et al. Naltrexone in autistic children: an acute open dose range tolerance trial. J Am Acad Adolesc Psychiatry. 1989;8:200-206.

- Lowry MA. Unmasking mood disorders: recognizing and measuring symptomatic behaviors. The Habilitative Mental Healthcare Newsletter. 1997;16:1-6.

- Hellings JA, Chibber S, Kilgore ER, Nickel J. Valproate response as a function of blood levels in adolescents with autistic disorder and PDD NOS Proceedings of 150th Annual American Psychiatric Association Meeting, San Diego CA; 1997:260.

- Donovan SJ, Sussen ES, Nunes EV, Steward JW, Quitkin FM, Klein DF. Divalproex treatment of disruptive adolescents: a report of 10 cases. J Clin Psychiatry. 1997; 58:12-15.

- Plioplys AV. Autism: electroencephalogram abnormalities and clinical improvement with valproic acid. Arch Pediatr Adolescent Med. 1994;148:220-228.

- Kastner T, Finesmith R, Walsh K. Long-term administration of valproic acid in the treatment of affective symptoms with mental retardation. J Clin Psychopharmacol. 1993;13:448-451.

- Mattes J. Valproic acid for nonaffective aggression in the mentally retarded. J Nerv Ment Dis. 1992;180:601-602.

- McElroy SL, Weller E. Psychopharmacological treatment of bipolar disorder across the lifespan. In: McElroy SL, - Dickstein JL, Riba MB, Oldham JM, eds. Psychopharmacology Across the Life Span. Review of Psychiatry, vol. 16. Washington, DC, American Psychiatric Press Inc; 1997:31-86.

- Jan JE, O'Donnell ME. Use of melatonin in the treatment of pediatric sleep disorders. J Pineal Res. 1996;21:193-199.

- Aman M. Effects of psychotropic drugs for reducing self-injurious behavior in the developmental disabilities. Am J Clin Psychol. 1993;5:171-188.

- Campbell M, Anderson LT, Small AM, Adams P, Gonzalez NM, Ernst M. Naltrexone in autistic children: behavioral symptoms and attentional learning. J Am Acad Child Adolescent Psychiatry. 1993;32:1283-1291.

- Campbell M, Schopler E, Cueve J, Hallin A. Treatment of autistic disorder. J Am Acad Child Adolescent Psychiatry.1995;35:134-143.

- Herman BH, Asleson GS, Borghese IF, et al. Acute naltrexone in autism: selective decreases in hyperactivity. Scientific proceedings, 38th annual meeting of the American Academy of Child and Adolescent Psychiatry. J Am Acad Child Adolesc Psychiatry1991;34:223-231.

- Kolmen BK, Feldman HM, Handen BL, Janosky JE. Naltrexone in young autistic children: a double-blind, placebo controlled crossover study. J Am Acad Child Adolesc Psychiatry. 1995; 34:223-231.

- Buitelaar JK, van Engeland H, de Kogel K, van Hooff J, van Ree J. The adrenocorticotropic hormone (4-9) analog ORG 2766 benefits autistic children: report on a second controlled clinical trial. J Am Acad Child Adolesc Psychiatry.1992; 31:1149-1156.

- McDougle CJ, Price LH, Volkmar FR. Recent advances in the pharmacotherapy of autism and related conditions. Child Adolescent Psychiatric Clin North Am. 1994;3:71-89.

- Ratey JJ, Bemporad J, Sorgig P, et al. Brief report: open trial effects of beta-blockers on speech and social behaviors in 8 autistic adults. J Autism Dev Dis. 1987;17:439-446.

- Ratey JJ, Mikkelson E, Sorgi P, et al. Autism: the treatment of aggressive behaviors. J Clin Psychopharmacol. 1987;7:35-41.

- Findling RL, Maxwell K, Wiznitzer M. An open clinical trial of risperidone monotherapy in young children with autistic disorder. Psychopharmacol Bull. 1997;33:155-159.

- Lott RS, Kerrick JM, Cohen SA. Clinical and economic aspects of risperidone treatment in adults with mental retardation and behavioral disturbance. Psychopharmacol Bull. 1996;32:721-729.

- Purdon E, Lit W, Labelle A, Jones BDW. Risperidone in the treatment of pervasive developmental disorder. Can J Psychiatry.1994;39:400-405.

- Simon EW, Blubaugh KM, Pippidio M. Substituting traditional antipsychotics with risperidone for individuals with mental retardation. Ment Retardation. 1996;34:359-366.

- Van den Borre R, Vermote R, Buttiens M, et al. Risperidone as an add-on therapy in behavioral disturbances in mental retardation: a double-blind placebo-controlled cross-over study.Acta Psychiatr Scand.1993;87:167-171.

- Hammock RG, Schroeder SR, Levine WR. The effects of clozapine on self-injurious behavior. J Autism Dev Dis.1995;25:611-626.

- Aman MA, Burrow WH, Wolford PL. The Aberrant Behavior Checklist-Community: Factor validity and effect of subject variables for adults in group homes. Am J Ment Retard. 1995; 100(3):283-292.

- Aman MA, Marks RE, Turnbott SH et al. Clinical effects of methylphenidate and thioridazine in intellectually subaverage children. J Am Acad Child Adolesc Psychiatry. 1991 Mar 30(2):246-256.