Novel Scan May Distinguish Unipolar and Bipolar Depression

Deborah Brauser

August 27, 2013

A novel brain scanning method that measures cerebral blood flow may help distinguish between bipolar disorder and depression early, new research suggests.

A small study of 54 adult women used a new imaging method called arterial spin labeling (ASL) to measure blood flow in subdivisions of the anterior cingulate cortex (ACC), brain regions that are associated with depression.

Using this measure resulted in an 81% accuracy rate in identifying which of the participants had unipolar depression and which had bipolar depression.

"This is the first time this type of technique has been able to be that accurate in distinguishing between these 2 conditions," principal investigator Mary L. Phillips, MRCPsych, MD (Cantab), professor of psychiatry and clinical translational science at the University of Pittsburgh School of Medicine in Pennsylvania and director of the school's Mood and Brain Laboratory, told Medscape Medical News.

The investigators note that bipolar disorder is often misdiagnosed as clinical depression — and that only 1 in 5 patients are correctly diagnosed with bipolar at first assessment.

"Earlier and more accurate diagnoses can make an enormous difference for patients and their families, and may even save lives," lead author Jorge Almeida, MD, PhD, assistant professor of psychiatry at the University of Pittsburgh, said in a release.

"These results also suggest that we may one day be able to predict future bipolar behavior in younger adults who haven't shown any symptoms, allowing for earlier and more accurate treatment," he added.

The study was published online August 22 in the British Journal of Psychiatry.

No Biological Markers

The investigators note that although unipolar depression and bipolar disorder are often difficult to tell apart, no biological markers have yet been discovered to help in differentiating between the 2 mental health conditions.

However, they voice some optimism that the recent development of pattern recognition analysis (PRA), and its use in combination with various neuroimaging techniques, could prove helpful at the diagnostic stage.

Dr. Mary Phillips

"We wanted to use neuroimaging as a tool to see if we could find biomarkers that would give us more information about bipolar and gives us more clues and measures which we can then use to better diagnose the disorder, rather than just relying on what people are telling us," said Dr. Phillips.

The investigators enrolled 54 adult women (mean age, 32.5 years), including 18 with recurrent unipolar depression and 18 with bipolar disorder type 1, all of whom were experiencing a depressive episode, and 18 age-matched women who had neither disorder ("control group").

PRA was applied for all participants, on a case-by-case basis, using subdivisions of ACC blood flow at rest, as measured with ASL.

"Before, research has looked at groups of patients. But we were interested in finding measures that could help the individual person," said Dr. Phillips.

Results showed significant brain pattern differences between the unipolar and bipolar depression groups in the subgenual ACC (BA25) subdivision (P = .001).

The measurement methods used yielded 83.3% sensitivity, 77.8% specificity, and 80.6% overall accuracy rates in distinguishing between these 2 groups.

In addition, "the positive predictive value was 79% and the negative predictive value was 82% for the same ACC subdivision," report the researchers.

Promising Diagnostic Tool

PRA was "only marginally significant" in predicting who had unipolar depression and who was in the control group (P = .02), and it had just a 50% sensitivity rate, based on patterns in the rostral/perigenual ACC (BA24) subdivision.

The method did not distinguish significantly between the bipolar depression group and the control group in any subdivision of the ACC.

"Our goal was to use ASL, a promising non-invasive neuroimaging technique, to identify the extent to which measures of ACC resting cerebral blood flow could have clinical utility to distinguish recurrent unipolar from bipolar depression," write the investigators.

Overall, the study "highlights the usefulness of neuroimaging to help identify biological markers associated with different mental health conditions," added Dr. Almeida.

However, he noted that further research is now needed with larger populations in multiple centers.

"Our ultimate goal is to use this as a diagnostic tool. This is a first study, and there needs to be replication, but we feel that differences in the subregion of the [ACC] is a really promising marker," Dr. Phillips said. "It's early days, but we're hopeful that in 2 to 5 years, these neuroimaging biomarkers will get into clinical practice."

She reported that the investigators continued to follow the study participants for 1 to 2 years and found some interesting further results. Originally, the 81% overall accuracy rate was due in part to the fact that the PRA predicted that 4 of the participants had bipolar when in fact they had a clinical diagnosis of unipolar depression.

However, follow-up showed that 2 of the supposedly misdiagnosed women went on to develop bipolar disorder.

"So this is really exciting. Not only can neuroimaging give us brain biomarkers of these disorders, but it can actually do that at an earlier stage than clinical diagnosis. That is so important, and we're following up on that now," she said.

Caution Warranted

"This is a very preliminary study, and there are several reasons to think it won't hold true," Michael Thase, MD, professor of psychiatry at the Perelman School of Medicine at the University of Pennsylvania in Philadelphia and from the Philadelphia Veterans Affairs Medical Center, told Medscape Medical News.

Dr. Michael Thase

"If it were to hold true, it would be an exciting development because you approach treatment from the onset differently in people who have the risk of having manic episodes than in people with little to no risk," he added.

For an individual in their first lifetime episode of depression, especially if they are younger than 25 years, "the odds are that it'll be unipolar depression," said Dr. Thase, who was not involved with this research.

"But there's a significant minority who have manic episodes. So the quicker you recognize this, the better."

However, although the study's results appear to show differences in the ACC subdivisions between the groups with unipolar and bipolar depression, he was concerned that neither group was highly significantly different from the control group.

"This just makes you worry about the pathophysiological significance of this."

Dr. Thase added that the type of treatment that the groups were receiving could also have contributed to the findings. More of those in the bipolar group than in the unipolar depression group were receiving mood stabilizers (55% vs 5.6%), and more of those in the unipolar group were using antidepressants than those in the bipolar group (83.3% vs 38.9%).

"We always thought that there were some subtle differences between bipolar depression and unipolar depression. For example, bipolar-depressed patients tend to have more psychomotor retardation. So it's not far-fetched that there would be differences in a brain region that could help clue you in," said Dr. Thase.

"I think these findings are exciting, certainly warrant further study, and could be potentially valuable. But there's reason to be cautious," he concluded.

The study was funded by grants from the National Institute of Mental Health. The study authors have reported several sources of funding, which are listed in the original article. Dr. Thase has disclosed no relevant financial relationships.

Br J Psychiatry. Published online August 22, 2013. Abstract, Supplemental Data


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.