U.S. PSYCH: Atypical Antipsychotic Medication Cuts Behavioral Symptoms in Autism

SAN FRANCISCO, April 25 -- While the behavioral symptoms of autism have been treated with a wide spectrum of medications, atypical antipsychotics may be the most effective drug class.

So said Lawrence D. Scahill, Ph.D., of Yale University, in a presentation here at the U.S. Psychiatric and Mental Health Congress regional extension.

Serious behavioral problems accompany autism in 20% to 30% of cases and include aggression, tantrums and self-injury.

To treat these and as well as the repetitive behaviors intrinsic to autism spectrum disorders, clinicians have tried a laundry list of medications ranging from stimulants to anticonvulsants to the substance abuse drug naltrexone (Vivitrol).

"Apparently we've been quite willing to try almost anything," Dr. Scahill said.

A survey of more than 800 autism spectrum disorder patients in North Carolina showed medication use in general rose from 30.5% in 1993 to 45.2% in 2001. The study, published in the Journal of Child and Adolescent Psychopharmacology in 2005, also reported increases in the percentage of patients on an antidepressant (6.1% to 21.4%), a stimulant (6.6% to 13.8%), or an antipsychotic (12.2% to 16.5%).

That increase in antipsychotic use, though small, "really represents a sea change," said Dr. Scahill, because the majority of patients were switched from typical antipsychotics to the newer atypical generation of antipsychotics.

However, the evidence for use of these medications in childhood autism is not very good with the exception of atypical antipsychotics, Dr. Scahill said.

His group, the Research Unit on Pediatric Psychopharmacology Autism Network, conducted a double-blind randomized trial of risperidone (Risperdal) among children and adolescents with autism to test its effect on aggression, tantrums, and self-injury. The study was published in the New England Journal of Medicine in 2002.

After eight weeks, irritability scores were improved 43% more with risperidone than placebo (from 26.2 to 11.3 for risperidone versus from 25.5 to 21.9 for placebo). Both patient groups started at an irritability score two standard deviations above the population mean, Dr. Scahill said.

"That's like a blood pressure of 240/160 or a fever of 105," and risperidone improved it more than one standard deviation, "a very big movement," he added.

Clinicians blinded to treatment also rated their global impression as "much improved" or "very much improved" for 75% of the risperidone group compared with just 12% of the placebo group.

However, weight gain was a problem. A quarter of the parents said appetite increase was a problem -- some reporting that they had to lock the refrigerator. In the extension study, patients gained a mean of 12 pounds over six months without any predictors of which patients would be affected the most.

"Monitoring and counseling about diet and weight is especially important at the start of treatment," he said. "Don't wait until it's a problem."

In October 2006, the FDA approved risperidone for the treatment of aggression, tantrums and self-injury among patients with autism based on the findings.

"Atypical antipsychotics are being used for a range of problems in children with pervasive developmental disorders," Dr. Scahill said. "The best studied is risperidone."

"It is very effective, but not without some problems," he concluded.

Dr. Scahill's group also conducted a study of the stimulant methylphenidate (Ritalin) for pervasive developmental disorder with hyperactivity. They found that the effect sizes ranging from 0.25 for low dose to 0.35 for high dose.

Though better than placebo, "these are not very big effects," Dr. Scahill said. Tolerability was a problem as well at the higher doses, including adverse events related to decreased appetite, irritability and trouble sleeping.

The data on the use of serotonin reuptake inhibitors encompassed just 386 subjects spread across 28 trials and five drugs with some of the trial participants being adults.

"The lack of data is frightful," Dr. Scahill said. The medications are being used to reduce repetitive behaviors and anxiety, which is not unreasonable, but again they are at increased risk of adverse events including activation, he said.

For the serotonin reuptake inhibitor best studied in children, fluoxetine (Prozac), one study showed a paltry 10% of children improved on the drug compared to 4% who improved with placebo.

"It just doesn't look that good," he said. "It certainly does not support that 20% of children are on it."

"If you feel strongly that this is a reasonable child to try a serotonin reuptake inhibitor because of repetitive behaviors," Dr. Scahill recommended, "the repetitive behaviors should be high or you won't see a difference and use a very conservative dosing strategy."