Autism and Pediatric Bipolar Disorder

Aggression in Autism

Posted Jul 11, 2014

The diagnosis of bipolar disorder in childhood is often incorrectly based on the symptom of severe irritability alone. It has been estimated that about 20 percent of autistic children are very irritable. The irritability associated with autism leaves autistic children vulnerable to the incorrect diagnosis of bipolar disorder. In my experience, this is especially true in those autistic children who are in residential care. Often these are the autistic children without language and with serious intellectual disabilities. Such children are often placed in residential care because of unmanageable aggression despite the best and often heroic efforts of concerned family members and other caretakers and despite the use of standard medications.  

Once in residential care, such children may acquire the incorrect diagnosis of bipolar disorder because of their irritability and other intractable management difficulties. The diagnosis of bipolar disorder makes more acceptable the prescription of a range of pharmaceutical agents including lithium and anticonvulsant medications with potentially lethal side effects and marginal therapeutic benefits.

The anticonvulsant most frequently prescribed for bipolar disorder in children is valproic acid (brand name Depakote). Both lithium and valproic acid have been proven effective in adult bipolar disorder but their usefulness in so called child bipolar disorder is at best unclear. Both medications are approved for the treatment of adult bipolar disorder by the FDA, but neither is approved for use in prepubertal children diagnosed with bipolar disorder.  Both lithium and anticonvulsants such as valproic acid have an array of toxic side effects that make them unsuitable for children with a disputable diagnosis of bipolar disorder. Based on my clinical experience, both lithium and anticonvulsants have been regularly used in residential centers with aggressive, intellectually disabled, autistic children misdiagnosed as having bipolar disorder.

The situation is more complex in the consideration of antipsychotic medications (such as risperidone, quetiapine, and aripiprazole, etc) for the treatment of irritability in autistic children.  Antipsychotics seem to be effective in treating aggression in children regardless of their underlying diagnosis. Antipsychotics are a nonspecific therapy for aggression. The underlying diagnosis of the patient does not seem to matter. If the patient has aggression as a symptom, antipsychotics might be helpful. For example, antipsychotics have been demonstrated to be effective in reducing the aggression of children diagnosed (correctly or incorrectly) as having bipolar disorder. 

Similarly, antipsychotics have been used to treat irritability in children with autism.  FDA approval for the psychopharmacological treatment of irritability in autistic children is a precedent setting landmark in FDA history.  In the past, the FDA refused to approve medications for the treatment of psychiatric symptoms alone. Medications could only be approved for DSM diagnoses.  Risperidone was the first medication to be approved by the FDA for the treatment of a symptom, and it was approved specifically for the treatment of the symptom of irritability associated with autism in children (1).  Subsequently, aripiprazole was approved for the same purpose (2).

One clinically useful organizing principle to understand autism and other psychiatric diagnoses that occur with it (comorbidities) is to consider autism as the basic underlying diagnosis with an appreciation that autistic children are vulnerable to all of the other DSM psychiatric diagnoses that can afflict children.

Unfortunately, distinguishing behaviors that are intrinsic to autism itself from those disorders that are separate from autism but occur with it can pose substantial difficulties. For example in DSM IV attention deficit hyperactivity disorder (ADHD) in autistic children was considered to be a part of the autistic disorder and  for this reason DSM IV specifically forbade making a diagnosis of ADHD in an autistic child. The result was that innumerable autistic children were denied stimulant medication treatment for their ADHD disorder. In DSM-V, ADHD is recognized as a legitimate comorbid diagnosis of autism and many autistic children are now receiving this diagnosis and having it treated.

Recent developments in biological psychiatry add a new layer of complexity to the discussion about the comorbidity between autism and ADHD. Despite their distinct clinical features, most recently there has developed a greater appreciation of important underlying biological genetic similarities between ADHD and autism (3).

The situation is quite different with regard to bipolar disorder in children with autism. Autistic children who meet DSM criteria for bipolar disorder are rare to non-existent. It is more useful to recognize the aggression of autistic disorder as  a symptom and treat it than to concoct  an exotic diagnosis such as bipolar disorder and attempt to treat the aggression with more toxic and less appropriate medications for a diagnosis the children do not have. 

Copyright, Stuart L. Kaplan, M.D. 2014

Stuart L. Kaplan, M.D., is the author of Your Child Does Not Have Bipolar Disorder: How Bad Science and Good Public Relations Created the Diagnosis. available at

  1. Shea S. Turgay A et al. Risperidone in the treatment of disruptive behavioral symptoms in children with autistic and other pervasive developmental disorders. Pediatrics   2004 Nov:e634-41.
  2. Owen R  Sikich L. et al. Aripiprazole in the treatment of irritability in children and adolescents with autistic disorder. Pediatrics 2009,124:1533.
  3. Martin J Cooper M. et al. Biological overlap of attention-deficit/hyperactivity disorder and autism  spectrum disorder: evidence from copy number variants. J Am  Acad Child Adolesc. Psychiatry. 2014;53:761-770.