The striking rise in methylphenidate prescriptions for young children worries Demitri Papolos, MD—and not for the obvious reason. As many as a third of children diagnosed with attention-deficit/hyperactivity disorder (ADHD) may actually have early symptoms of childhood-onset bipolar disorder, Dr. Papolos believes, and this missed diagnosis could have insidious effects. In their book The Bipolar Child,Dr. Papolos and his wife, author Janice Papolos, warn that giving stimulants (or antidepressants) to children with incipient bipolar disorder without a mood stabilizer "may wreak havoc on a possible quiescent bipolar gene"—in many cases provoking psychotic episodes and violence, and perhaps inducing an ultra–rapid-cycling form of the disorder in which mood shifts dramatically as often as several times a day.

Tales of delayed or missed bipolar diagnoses are all too familiar to Dr. Papolos, who is associate professor of psychiatry at Albert Einstein College of Medicine in New York and codirector of its Behavioral Genetics Program. For more than a year, the Papoloses monitored messages to an Internet listserv for parents of bipolar children. This immersion in shared e-mail—sometimes written by parents whose children were in the midst of an explosive manic episode—allowed Dr. Papolos to observe the impact of the disorder in far more detail than is typically possible in clinical practice. "They really did provide us with a unique picture of the condition in childhood," he told Neuropsychiatry Reviews when we spoke with him in late April.

NPR: What led you to study childhood-onset bipolar disorder?

DP: I got into it through work I did with children with velo-cardio-facial syndrome (VCFS), which is a multiple anomaly condition that is very well characterized on a molecular basis. Robert Shprintzen [PhD], who was the first to identify the syndrome, had asked me to evaluate a number of these children who were having various psychiatric symptoms. So we did a study, which came out in 1996, that found that a very significant number had early-onset bipolar disorder. They had rapid-cycling and ultra–ultra–rapid-cycling forms of the condition. And they had many comorbidities, particularly attention-deficit disorder. The field was just beginning to define childhood-onset bipolar disorder in the early 1990s, and there was only a small literature available at the time. But what others were beginning to describe, we were seeing in children with VCFS.

NPR: Until the 1970s, the diagnosis of childhood-onset bipolar disorder was almost unheard of.

DP: Well, I wouldn't say it was unheard of. It's been described since antiquity in the medical literature. But in the 1930s, the myth developed in this country that it didn't exist. The child psychiatrist and researcher Leo Kanner, who wrote the primary textbook that was used all the way through the late 1960s, didn't even mention it as a condition that could happen in childhood. That omission was a major influence on the field.

NPR: You've cited data suggesting that there may be hundreds of thousands of cases of childhood bipolar disorder, with perhaps the majority of them misdiagnosed. Is the incidence rising?

DP: That seems to be the impression of most clinical researchers I come in contact with. My colleague Rosalie Greenberg [MD], who's a child psychiatrist in New Jersey, says that maybe 80% of her practice is now early-onset bipolar disorder. That wasn't the case 10 years ago.

The stories I hear are basically what is written in the book—missed diagnoses, treatment with agents that make the children worse and may kindle the condition. We're wondering if early introduction of antidepressants and stimulants turns on or off genes that would be better off not tampered with. That's a big concern, and there needs to be more scrutiny of that possibility.

NPR: An important point that you and others have made is that childhood-onset bipolar disorder does not resemble the adult-onset version, and hence the DSM-IVcriteria are not very useful.

DP: They really need to be significantly modified. There are so many features of the condition that are common to most of these kids but are not present in adults. I'd say that more than 70% of the kids we've seen have this ultra–ultra–rapid-cycling form of the condition. And there is a clear circadian pattern of arousal, energy levels, and mood. The typical pattern is that they can't get up in the morning; they're almost anesthetized, and their parents have to dress them in their sleep to get them off to school. For the first couple of periods of school they're pretty groggy. Then they get more energy; around 3 or 4 o'clock the energy begins to accelerate, and it peaks around 8 or 9 o'clock at night, depending on their age. So they have a lot of trouble getting to sleep, and there is a tendency for sleep-wake reversals, particularly in the summer, when they don't have the structure of school.

Many of them also have nightmares. They wake up in the middle of the night, screaming and thrashing as if they were being attacked. When they are able to remember the dreams, they often report that a monster or predator or dinosaur was attacking them, and they are usually very gory dreams. The other theme of their dreams is abandonment—they are fearful that their parents are not going to be able to protect them when the parents are gone or dead. This doesn't happen with every child, but it's almost a template.

They also often have night terrors and enuresis and parasomnias such as restless leg syndrome and sleep walking. So there's something awry with the arousal system, and I think it's easily traced to some kind of exacerbation of the fight-or-flight mechanism that carries over into the day. Because their behaviors generally go along two dimensions—either they are aggressive, oppositional, defiant, or they are fearful, anxious, withdrawn. So this dimension of fight or flight really comes out as part of their temperament. It's pretty unmistakable when you put it all together. The DSM-IVdoes not paint the picture very accurately at all.

NPR: We'll see if things are better in DSM-V.

DP: I think that will require a lot of political wrangling. But it's too late for a lot of families that are suffering with this now.

NPR: How well understood is the genetic basis of childhood-onset bipolar disorder?

DP: In the VCFS study, we found that there is a very specific deletion on chromosome 22, and there are about 30 genes that are often clipped off in the kids with VCFS who have the ultra–ultra–rapid-cycling form of bipolar disorder and other comorbidities. This makes it a gold mine for studying childhood psychiatric conditions.

One of the genes that is clipped off is COMT,which is the gene that produces the enzyme [catechol-O-methyltransferase] that degrades dopamine and norepinephrine. There are two forms of the COMTgene: a low-activity form and a high-activity form. And we found that all of the rapid cyclers in the VCFS population were missing the COMTgene on one chromosome and had the low-activity form on the other chromosome. Then we recruited a group of rapid-cycling adults, and there were seven or eight who had ultra–ultra–rapid cycling, and they all had histories of childhood-onset bipolar. And they were homozygous for the low-activity allele. So we may have found a specific marker for the ultra–ultra–rapid-cycling variant, and this finding has been confirmed by another group.

NPR: You've noted, as have others, that more than 90% of kids with bipolar disorder also meet criteria for ADHD.

DP: Without a doubt. And about the same number meet criteria for oppositional defiant disorder.

NPR: What's the key to making a correct diagnosis?

DP: One thing is identifying the rapid cycling and the particular circadian pattern of mood and energy. And there may be severe temper tantrums that last for more than half an hour, and sometimes up to seven hours. We've been assessing patients using the Overt Aggression Scale, and we now have about 60 reports from kids with the condition. There are 16 possible positive responses on the scale, and I have yet to see a score under 11 for any of these kids.

NPR: That would be rare in ADHD.

DP: Yes. And the other important thing is family history. There seems to be bilineal transmission of the condition; both paternal and maternal families are affected with bipolar disorder and alcoholism.

NPR: Given all the comorbidity associated with childhood-onset bipolar disorder, it's understandable that there has been so much diagnostic confusion.

DP: Absolutely. That's a big problem. A significant number of these kids have depression, which is the first thing that is recognized. But no one will ask about the cycling of moods, so they get put on an antidepressant. And the danger, once again, is that stimulants or antidepressants might have a kindling effect. The important point, I would say, is to rule out bipolar disorder before you treat for ADHD or depression.

NPR: What treatment trials are most urgently needed for childhood bipolar disorder?

DP: Everything. There's nothing but a couple of small lithium studies. I don't think any mood stabilizers have been tested in a controlled trial. There's a crossover design study that Robert Kowatch [MD] is doing; he did an effect size analysis and found that there are some minor differences between lithium, divalproex, and carbamazepine—but nothing to write home about.

Suggested Reading
1. Papolos DF, Papolos J. The Bipolar Child.New York, NY: Broadway Books; 1999.
2. Faedda GL, Baldessarini RJ, Suppes T, et al. Pediatric-onset bipolar disorder: a neglected clinical and public health problem. Harv Rev Psychiatry.1995;3:171-195.
3. Yudofsky SC, Silver JM, Jackson W, et al. The Overt Aggression Scale for the objective rating of verbal and physical aggression. Am J Psychiatry. 1986;143:35-39.

Return to table of contents