Saturday, May 19, 2007

"A blood test to predict bipolar disorder? I don't think it will happen."

New Scientist

Bipolar children - is the US overdiagnosing?

  • 16 May 2007
  • news service
  • Andy Coghlan

Rebecca Riley seemed a normal, playful young child, if at times a little boisterous. Then, aged 2, she was diagnosed with attention deficit hyperactivity disorder, and at 3 as having a bipolar personality. By the age of 4, Rebecca was dead, killed by an overdose of the drug clonidine, which was being used to treat her condition. She was also taking the anti-convulsant Depakote (valproate) and the anti-psychotic Seroquel (quetiapine fumarate) to stabilise her mood.

Rebecca died next to her parents' bed in Hull, Massachusetts, on 13 December last year. This Wednesday a pre-trial hearing began that will be followed by a full trial to determine whether her parents, Carolyn and Michael Riley, intentionally gave Rebecca too much of the drug, or whether she died of an accidental overdose. Whatever the verdict, Rebecca's tragic story forms part of a wider narrative: the growing numbers of young children in the US being diagnosed with bipolar disorder and given drugs to treat it.

While the trend is clear, the reasons behind it are not. Psychiatrists also disagree over whether children are being helped or harmed by being diagnosed with the condition.

Bipolar disorder used to be called manic depression. People with it swing across an extreme spectrum of moods, usually from a state in which they feel euphoric and active, to one of depression and despondency, and back again.

In 1994, the official psychiatric manual, DSM-IV, widened the definition of bipolar disorder. The classic condition, characterised by the symptoms described above, was named bipolar disorder I, and was joined by three newly defined conditions: bipolar disorder II, which is a milder version of disorder I; cyclothymic disorder, in which mood swings manifest themselves more subtly; and bipolar disorder not otherwise specified, a catch-all category for people with bipolar-like symptoms who do not fit the other three diagnostic criteria.

Some specialists see this widening of the diagnostic criteria as the cause of the explosion in the numbers of children diagnosed with bipolar disorder. "When DSM expands its symptomology there's obviously a greater basis for diagnosis and prescription," says Lisa Cosgrove, a clinical psychologist at the University of Massachusetts, Boston, who studies the links between psychiatrists and the drug industry.

Child psychiatrist Gabrielle Carlson agrees. She and colleague Joseph Blader, both at Stony Brook University, New York, analysed data on people with bipolar disorders discharged from US hospitals. Between 1996 and 2004 the rate "went up minimally in adults, but rose astronomically in children", Carlson says.

In 1996, 13 out of every 100,000 children in the US were diagnosed as having bipolar disorder. In 2004, the figure was 73 in 100,000, a more than fivefold rise, they report in a paper to be published in Biological Psychiatry. Among children diagnosed with a psychiatric condition in 1996, 1 in 10 were deemed to have bipolar disorder. By 2004, 4 out of 10 children with a psychiatric condition were told they were bipolar (see Chart).

Some say this high prevalence of bipolar disorder in children in the US, which far exceeds that in most other countries, is genuine. "I think it's down to increased recognition that the condition exists in children, which wasn't accepted until a decade ago," says Susan Resko, executive director of the Child and Adolescent Bipolar Foundation, an advocacy group for families with a bipolar child based in Wilmette, Illinois.

According to a review of bipolar disorder published this month in the Harvard Mental Health Letter by Michael Miller, a psychiatrist at Harvard Medical School in Boston, the jury is still out. "We simply don't have all the information we need to connect the dots between behaviour, causes and symptoms. We don't know yet if kids are being under-medicated, over-medicated or mis-medicated," he told New Scientist. Any genetic basis for the condition is also proving elusive (see "On the trail of the bipolar gene").

"We simply don't yet have all the information to know if kids are being under-medicated, over-medicated or mis-medicated"

That uncertainty over both the cause and true prevalence of bipolar disorder raises concerns that children are being prescribed inappropriate drugs with damaging side effects. Lithium, for example, can cause excessive weight gain, thirst and acne in young children. It also makes children want to urinate more. Valproate can trigger excessive hair growth and, in adolescent girls, polycystic ovary syndrome, a leading cause of infertility.

Another worry is that drugs prescribed for other psychiatric conditions could be triggering bipolar disorder. Around 2 million children in the US are diagnosed as having attention deficit hyperactivity disorder (ADHD), which in children presents similar symptoms to bipolar disorder (see "Easy to confuse"), and the antidepressants used to treat ADHD can make bipolar symptoms worse.

This possibility is backed by a review of worldwide rates of bipolar disorder in children published in 2005 by C├ęsar Soutullo of the University of Navarra in Pamplona, Spain, and colleagues (Bipolar Disorders, vol 7, p 497). One study they cited found that children given antidepressants go on to develop a bipolar disorder at an earlier age (at 10, on average, rather than 14), and with greater severity, than those not prescribed such drugs. Resko, however, rejects the idea that prescribed drugs are causing more children to become bipolar.

There are also worries that over-diagnosis of bipolar disorder is leading to other psychiatric conditions being missed. In February, Ellen Leibenluft and her colleagues at the US National Institute of Mental Health in Bethesda, Maryland, suggested that brain imaging tests could help pick out children with a condition called "severe mood dysregulation", which is characterised by bipolar-like symptoms such as extreme irritability and hyperactivity. "We will end up with more clear-cut clinical tests that help avoid misdiagnosis," she says.

Meanwhile, debate continues over whether children such as Rebecca should be diagnosed as having bipolar disorder. Carlson says many parents and doctors would prefer to have a medical diagnosis rather than accept that a badly behaved child is psychologically normal, and she questions the validity of some diagnoses. "When children are only 2 to 4 years old, how can they be 'grandiose'?" she asks. Children of that age have mood swings every hour, she points out.

Resko maintains that a correct diagnosis of bipolar disorder can be helpful for children more than 5 or 6 years old, and says intervention may be required even for younger children if they have a history of the condition in the family. "These kids are suffering and committing suicide," she says. "It's a difficult call."

Easy to confuse


Irritability (not getting your own way)

Psychosis (grandiosity/inflated self-esteem)

Elation (expansive mood)

Rapid speech

Sleep (lack of)





Emotional lability (short temper)

On the trail of the bipolar gene

"A blood test to predict bipolar disorder? I don't think it will happen." So says Francis McMahon, who heads an effort coordinated by the US National Institute of Mental Health in Bethesda, Maryland, to scour the human genome for genes related to bipolar disorder.

This month, McMahon's team published a comparison of the genomes of 413 adults with bipolar disorder and 563 healthy controls. Eighty genes could be associated with bipolar disorder, eight of which influence how the brain responds to neurotransmitters such as dopamine. The gene most strongly correlated with the disorder codes for a variant of the enzyme diacylglycerol kinase known as DGKH, which could make this enzyme a target for future drug therapies (Molecular Psychiatry, DOI: 10.1038/

In March, a separate study by Colleen McClung and colleagues at the University of Texas Southwestern Medical Center in Dallas found that mice carrying mutations in the Clock gene, which is thought to influence the animal's circadian rhythm, show manic-like behaviours such as hyperactivity, reckless behaviour and a predilection for addictive substances such as cocaine. They behaved normally when treated with lithium (Proceedings of the National Academy of Sciences, DOI: 10.1073/pnas.0609625104).

Despite these insights, it is unlikely we will find a particular gene, or set of genes, that clearly trigger bipolar disorder, says McMahon. Even DGKH is carried by 8 out of 10 people in the general population. "Not only are there many genes involved, but the forms involved are very common," McMahon says. He believes many will also be associated with other psychiatric conditions such as ADHD.



No comments: