Clinical Differentiation of Bipolar II Disorder From Borderline Personality Disorder

Adam Bayes; Gordon Parker; Kathryn Fletcher


Curr Opin Psychiatry. 2014;27(1):14-20. 

In This Article


We review a wide range of potential parameters differentiating BP II disorder from BPD, while accepting that a small percentage of individuals may have both conditions. The literature is limited by most studies considering bipolar in general (i.e. both BP I and BP II subtypes) and any extrapolation of findings from studies of bipolar groups to BP II alone risks false conclusions.

No clear distinctive neuroimaging differences have been identified. Although some neurocognitive differences have been suggested, their discrimination is too limited for application. Personality testing might be expected to show commonalities (reflecting shared emotional dysregulation and impulsiveness) and not particularly discriminating unless assessed as to whether present as a trait or only during mood states. Other parameters (e.g. family history and a 'breeding true' phenomenon, childhood trauma, self-harm, impulsivity) may be of some discriminating use but again require consideration of context rather than simply considering prevalence.

The seemingly most useful discriminating domains would appear to be phenomenological differences in terms of mood and age of onset. The majority of BP II hypomanic episodes are euphoric anxiety-free states and in sharp contrast with the hostility, irritability and anxiety-weighted periods experienced by those with BPD. The majority of depressive episodes in those with a BP II condition are melancholic in nature and contrast with the nonmelancholic reactive depressive episodes experienced by those with a BPD. BPD (being based in personality style) would appear to evolve from childhood and adolescence, whereas BP II is most likely to have a sharp onset period (i.e. the individual reporting episodes of hypomania and of depression emerging when no such distinct episodes were previously present).

Sharpening differentiation would benefit from more specific studies focussing on BP II and not bipolar in general, with multivariate analyses considering multiple potentially discriminating domains and refining the most distinctive. These are likely to weight clinical nuances that might then be expected to lead to follow-up studies pursuing underlying differentiating contributions from more precisely defined subsets of those with a BPD or BP II disorder, evaluating the likely relevance of differing drug and nondrug treatments. Misdiagnosis of BP II as a BPD can risk extensive periods on nondrug treatments when the individual might benefit from a mood stabilizer, whereas misdiagnosis of BPD as a bipolar disorder can conversely risk inappropriate pharmacological treatment.