When youths get referred for help with symptoms that don’t quite meet diagnostic criteria for bipolar illness, there’s a 50-50 chance they’ll progress to a diagnosis of bipolar disorder I or II within 7 years. The odds are a coin toss.

The risk logically might be even lower in general clinical settings than in this defined group. That’s why Dr. David A. Axelson and his associates advocate using conservative criteria for diagnosing “bipolar not otherwise specified” (BP-NOS) in general clinics.

Sherry Boschert/IMNG Medical Media

Dr. David A. Axelson

Dr. Axelson won the American Academy of Child and Adolescent Psychiatry’s Klingenstein Third Generation Foundation Award for his longitudinal research on 140 children and adolescents who met an operationalized diagnosis of BP-NOS. At a median follow-up of 5 years, 45% had converted to bipolar disorder I or II (BP I/II) within a mean of 58 weeks after intake (J. Am. Acad. Child. Adolesc. Psychiatry 2011;50:1001-16.e3).

New data from the ongoing study show that 50% progressed to BP I/II at a median follow-up of 7 years, he said at the academy’s annual meeting. Symptoms for most of the youths in the COBY (Course and Outcome of Bipolar Youth) study far exceeded the minimum BP-NOS criteria at baseline.

Very few factors predicted whether patients would convert to BP I/II or not, and even those were not strong predictors, said Dr. Axelson, medical director of Child and Adolescent Bipolar Services Outpatient Program at the University of Pittsburgh’s Western Psychiatric Institute and Clinic.

Many children and adolescents present to clinics with manic symptomatology that does not meet diagnostic criteria for BP I/II in the DSM-IV. Clinicians walk a tightrope between intervening as early as possible for best treatment results and mislabeling (and then mistreating) some youths who don’t have bipolar illness.

The criteria for BP-NOS in the DSM-IV are vague and nonspecific, Dr. Axelson said. Based on his and other studies of “subthreshold” bipolar symptoms in children and adolescents, Dr. Axelson proposed that criteria for diagnosing BP-NOS in general clinical settings include:

• Use of full DSM symptom criteria for a hypomanic or manic episode.

Almost all of the children and adolescents in his COBY study met medical criteria for symptoms, he noted.

• Having hypomanic symptoms for most of the day.

“Similar to what we think about for major depression,” Dr. Axelson said. “This most-of-the-day specifier will be in DSM 5 for manic or hypomanic episodes.”

• At least one episode of 2-day duration.

• At least four recurrent episodes.

Almost all the children and adolescents with BP-NOS in the COBY study already had recurrent episodes.

These criteria are “probably the best balance between sensitivity and specificity, understanding the fact that this is going to miss some kids in the early phase of illness,” he said.

The COBY study enrolled 153 youths seen at three academic medical centers, 140 of whom had at least one follow-up visit. The main reasons the diagnosis was BP-NOS instead of BP I/II were because the duration of manic or hypomanic episodes was too short (only 1-3 days in 86% of patients); the youth had hypomania with no major depressive episode (11%); or the youth did not have the required number of symptoms for BP I/II (3%).

The investigators tracked at least 17 factors that they hypothesized might help predict which youths would progress to BP I/II. “Much to my surprise, very little of this actually predicted future onset,” Dr. Axelson said. The main predictor was a family history of mania or hypomania and, “the effect size isn’t huge.”

At intake, the 63 patients who later converted to BP I/II were significantly more likely to have a family history of mania or hypomania (64%) or depression (90%), compared with the 77 patients who did not convert to BP I/II (40% and 78%), respectively.

A total of 58% of youths with a family history of mania or hypomania converted to BP I/II by a median 5-year follow-up, compared with 36% of youths without this family history. The newest data suggest that by 8 years, two-thirds of youths with a family history of mania or hypomania convert to BP I/II, compared with just under half of youths without this family history, he reported.

“One thing that’s interesting is the progression rate keeps going up in both groups if you follow them longer,” Dr. Axelson said. “These kids continue to go forward in converting to bipolar illness.”

A multivariate analysis found that a family history of mania or hypomania tripled the risk for progression to BP I/II. So did white race, which “we can’t really explain,” he said. Any lifetime history of psychiatric hospitalization multiplied the risk for progression 2.5 times. Higher scores on the Young Mania Rating Scale in the past month increased the risk of progression by 3%, which was statistically significant.

Any lifetime history of psychotic symptoms, however, was significantly and negatively associated with progression to BP I/II, “something we still don’t fully understand,” Dr. Axelson said. Patients with a history of psychotic symptoms were 71% less likely to convert to BP I/II.

Having a family history of mania or hypomania is “a useful predictor, because more kids with family history did convert, however it’s not so strong that you can say it’s definitive,” he said. “Lots of kids who had a family history didn’t progress, and a full third of the kids who didn’t have a family history progressed.”

Dr. Axelson reported having no financial disclosures.