Commentary on ‘Comorbidity of bipolar disorder and borderline personality disorder: Phenomenology, course and treatment considerations’ by Temes et al.

The co-occurrence of bipolar disorder (BD) symptoms and borderline personality disorder (BPD) symptoms has been consistently reported since the emergence of the BPD diagnosis in DSM-III. This co-occurrence of symptoms is greater than would be expected by chance if BD and BPD were truly distinct psychopathologies.1 Therefore, much effort has gone into studying the nature of the relationship. As Temes et al. note, some researchers theorised that BPD was part of the BD spectrum, although most evidence suggests that while there are common symptoms between the two disorders, they present these traits differently. The clinical relevance of the co-occurrence of BD and BPD symptoms is that those with symptoms of both disorders appear to have a worse course of illness. These individuals have higher rates of suicidal behaviours, other comorbid psychiatric conditions and poorer psychosocial functioning. My contention is that research is hampered by the imposition of categories onto what are dimensional forms of psychopathology. Both BPD and BD are empirically better conceptualised as a continuum of symptoms with varying intensity and severity, and their relationship would be easier to discern if this was how they were studied. BPD has always been difficult to justify as a PD category. All attempts to find a borderline personality factor have failed. While BPD symptoms group together, when examined in isolation, they disappear into a general severity measure of personality pathology when modelled alongside other PDs.2 One reason for BPD's success appears to be that the symptoms are related to the overall severity of personality pathology so that BPD is strongly related to an individual's total number of PD symptoms.3 A diagnosis of BPD is best seen as signifying that the patient has at least moderate personality pathology. Clinically, comorbidity of BD and BPD simply states that a patient with BD also has moderate to severe personality pathology. That is; in addition to BD, the patient has a marked disturbance in personality functioning with impairments in functioning of aspects of the self and/or in interpersonal functioning. Emphasising severity as a dimension rather than BPD as a category allows more sophisticated formulation. The evidence is that severity, not PD type, is the major predictor of outcome. Clinicians might choose to describe specific personality features such as affective instability, impulsivity and suicidality (as in BPD) but could also describe non-BPD symptoms such as detached or anankastic features. However, it is the extent of the core symptoms of impairment in self and interpersonal functioning that will predict the outcome. Similarly, some argue that BD symptoms are more accurately conceptualised as a dimension. Dividing BD into BD I and BD II with loosely applied 4- and 7-day time criteria creates a categorical distinction which does not conform to the distribution of manic symptoms.4 Conceptualising a unified BD allows a more coherent discussion of a spectrum of bipolar symptoms. Conceptualising both BPD and BD as dimensional rather than categorical constructs encourages a more coherent discussion of their relationship with a number of advantages. First, as noted, dimensional models reflect more accurately the reality of BD and PD. The intensity and severity of mood states and personality pathology exist on a continuum. Second, the consideration of the two dimensions in all patients would encourage and allow a more overarching set of treatment principles. For example, those with severe BD symptoms would require medication treatment regardless of the severity of personality pathology. However, the presence of severe personality disorder might lead to a more structured treatment setting with careful consideration around establishing the treatment alliance and awareness of the patient's potential for harm to themselves and others. In patients with less severe mood symptoms (what might now be called Bipolar II) but moderate-to-severe personality pathology the emphasis might shift to structured psychotherapies with more sparing and careful use of medication. Third, having two dimensions offers a better model to discuss and research the co-occurrence of BD and BPD symptoms. For example, the question of why those with less severe mania symptoms (classified as BD II) have higher rates of personality pathology (currently diagnosed as BPD) as Temes et al. note. Is mania protective in some way? Or is this related to measurement error where the more severe manic symptoms overwhelm the personality symptoms when the categories are assessed. Also, what is the effect of milder personality pathology (which would not reach BPD threshold) on treatment and prognosis in those suffering from BD? In conclusion, using dimensions of personality disorder severity and BD symptoms may help to better integrate the relationship between the disorders. It helps clinicians to more easily differentiate milder forms of BD from personality disorder. Importantly, it may allow a more coherent model focusing on both functional impairments from self and interpersonal problems and/or mood symptoms in formulating a treatment plan. There is no data included in the paper.