SSRIs Not Recommended for Autism in Children or Adults Based on Current Evidence

From Medscape Medical News
Pauline Anderson

August 12, 2010 — Based on the research to date, selective serotonin reuptake inhibitors (SSRIs) cannot be recommended for treating autism in children or adults, the results of a new Cochrane review of the literature show.

The analysis found no evidence that SSRIs are effective in children with autism and may even be harmful, and although there is limited evidence that SSRIs are effective in adults, the sample size of the trials is small and there is a risk of bias.

Decisions about treating conditions that might accompany autism spectrum disorder (ASD), for example, obsessive-compulsive disorder (OCD) or depression, should be made on an individual basis, said the study authors.

"It's not surprising that clinicians and parents have hoped that SSRIs will help core features of autism as well as associated problems, but there is no strong evidence that they do," said lead study author Katrina Williams, PhD, School of Women's and Children's Health, University of New South Wales & Sydney Children's Hospital, Australia, in email correspondence with Medscape Medical News.

"So when trying to balance benefit and harm from existing evidence, clinicians and families are still in a situation where decisions will need to take into account the severity of the problem, the type of problem, and the potential risks."

The study was published online August 8 in The Cochrane Library, issue 8.

Most Studies Small

For this report, researchers searched the Cochrane Central Register of Controlled Trials (CENTRAL), MEDLINE, EMBASE, CINAHL, and PsycINFO for randomized controlled trials of an oral SSRI compared with placebo in participants with an ASD. Characterized by impairment in social interaction and communications skills, stereotypic behaviors, and limited activities and interests, ASDs include autism, pervasive developmental disorder–not otherwise specified, and Asperger syndrome.

The analysis included 7 studies that evaluated 4 SSRIs: fluoxetine, fluvoxamine, fenfluramine, and citalopram. The studies included a total of 271 subjects and were performed in the United States (5 studies), France (1), and Japan (1). Five of the studies included only children, and 2 included only adults.

The studies, which continued for a maximum of 12 weeks, reported 17 different outcome measures. Most of the studies were small; although 1 multicenter trial included a sample size of over 100, the next largest study recruited 39 participants. For these and other reasons, the reviewers found that meta-analyses were not possible.

None of the studies assessed sleep disturbance, self-mutilation, attention and concentration problems, gastrointestinal function, or quality of life.

In the 5 studies of children, including 1 large good quality study of citalopram and 4 smaller studies of fluoxetine, fluvoxamine, and fenfluramine, there was no evidence of benefit. The study of citalopram reported significantly more adverse events in children taking this drug compared with placebo, including 1 serious adverse event, a prolonged seizure.

Different Adverse Effects in Children

Dr. Williams pointed out that children with autism may experience different side effects of SSRIs than adults with autism or other children without autism. "It's important that parents embark on SSRI treatment aware that, as yet, there are few proven benefits and some known harms. That's different from making a decision about commencing a treatment with likely benefit and no known harms."

With monitoring, dose adjustment, and time, all but one of the adverse effects in these studies were resolved, said the study authors.

Although none of the trials assessing "core features" of autism in children — communication, social interaction, and behavior problems — showed improvement, it is possible that SSRIs may still be effective in treating autism in children.

Lack of evidence is not the same as evidence that SSRIs don't work.
"This is a complex area because there are many SSRIs available — and not all have been studied in large well-conducted trials," said Dr. Williams. "Lack of evidence is not the same as evidence that SSRIs don't work."

According to the study authors, replication of the citalopram study will provide further information about the effectiveness and safety of SSRIs for childhood autism. "For completeness, an adequately powered [randomized controlled trial] should be conducted on at least 1 other SSRI," the study authors note, and recommend this SSRI be fluoxetine because of its favorable safety profile.

Studies in Adults

The studies that included adults reported significant improvements in clinical global impression (fluvoxamine and fluoxetine), OCD behaviors (fluvoxamine), anxiety (fluoxetine), and aggression (fluvoxamine). However, the 2 relevant studies were small (1 had only 6 subjects, the other 30 subjects). and the quality of the trials was uncertain, said Dr. Williams. Both adult studies reported that treatment was well tolerated.

As well as covering a wide age range, the studies in the analysis included subjects whose conditions were diagnosed using different classification systems and assessment procedures. The studies also varied in terms of the subjects' IQ, severity of their problems, and whether they had the problems that the treatment is suggested to ameliorate.

Despite such differences, there is consistency of findings for the studies conducted in both children and adults, said the study authors.

None of the studies evaluated sertraline, paroxetine, or escitalopram — drugs used in clinical practice to treat problems associated with ASD, for example, OCD and depression. "When conducting a review of a drug class, it's important for prescribers to know what has and has not been examined," commented Dr. Williams. "It's also important for those planning future research."

Prevalence of autism varies between 1 and 40 per 10,000, and for ASD it is between 3 and 82 per 10,000. Males are affected about 4 times more frequently than females.

Antidepressants, most of which are SSRIS, are the most commonly prescribed psychotropic medication for ASD, but the number of SSRI prescriptions for children have decreased because concerns have been raised about increased risk of suicide-related behaviors.

Off-Label Use

SSRIs are not approved for treatment of autism, so use of these drugs in children with this condition is either off-label or used to treat depression or OCD. The US Food and Drug Administration (FDA) has approved sertraline in children 6 years and older, fluoxetine in children 7 years and older, and fluvoxamine in children 8 years and older for the treatment of OCD. The FDA has also approved fluoxetine in children 8 years and older and escitalopram in adolescents 12 to 17 years for the treatment of depression.

In autism, there's a "fundamental difference" between treating a co-occurring problem and treating the underlying disorder, explained Dr. Williams. "If SSRIs were useful in treating core features of autism, their use would become widespread, and it's likely they would be used in younger age groups. That would require much more rigorous review and monitoring — especially with regard to side effects."

The thinking is different when it comes to treating co-occurring problems, said Dr. Williams. "Trials are building on existing evidence about effectiveness for that problem in individuals who don't have autism. The treatment would only need to be long enough to address the problem, if it were short term, or to allow receptiveness to other proven nonmedication therapies."

Not a Single Condition

Commenting on the findings, Isabelle Rapin, MD, professor of neurology and pediatric neurology at Albert Einstein College of Medicine, New York City, emphasized how extremely difficult it is to perform studies of autism.

"We know that autism is not a single condition; we know that it has multiple causes, marked differences in symptomatology, and marked differences in severity," she said.

Some children with autism might respond well to an SSRI "if they were properly classified," but because these patients are a heterogenous group, it is important to make decisions subject by subject, said Dr. Rapin.

She said there are "hundreds and hundreds" of children with autism who are prescribed an SSRI to treat symptoms such as self-injury, irritability, or repetitive movements.


Cochrane Database Syst Rev. 2010;8.