Bipolar II disorder: The need for clearer definition and improved management

Bipolar II disorder (BD-II) is associated with significant morbidity, yet has almost been employed as a ‘dustbin’ diagnosis for those failing to meet criteria for bipolar I disorder (BD-I). The Diagnostic and Statistical Manual of Mental Disorders (DSM) criteria lack sufficient clarity for reliable use and ignore empirical data, while research and treatment guidelines combine BD-I and BD-II as if ‘one size fits all’. There is an urgent need for the scientific community to examine BD-II as a distinct diagnostic category worthy of investigation in its own right. Only then can optimised treatments be developed to remediate this debilitating, misunderstood condition.

BD-II: playing second fiddle to BD-I?

BD-II was formally recognised in 1976 following an inpatient record review during the 1960s (Dunner, 2017). Suicidality was a marked feature of the group, with increased suicide rates following discharge. Over the next decade, a ‘bipolar other’ group was identified, who had never been hospitalised for manic-like episodes, yet received outpatient treatment for mood disorder. It was not until 1994 that the Diagnostic and Statistical Manual of Mental Disorders (4th ed.; DSM-IV) included BD-II as a distinct diagnostic group, based on a review of biological, pharmacological and family data, clinical features and longitudinal illness course. More recently, the largest family study to date demonstrated aetiological heterogeneity between BD-I and BD-II, supporting BD-II as a separate entity (Song et al., 2017). Despite this, an ongoing disregard of BD-II exists in the published literature.

From the few studies examining BD-II, a clear picture emerges: BD-II can no longer be considered a ‘milder’ manifestation of BD. The World Mental Health survey (Merikangas et al., 2011) provides compelling evidence for the comparable morbidity of BD-II to BD-I in community settings, highlighting severity of depression, suicide attempts and impairment. The most striking finding is the disproportionate longitudinal symptom burden of BD-II depression quantified by two large prospective studies. BD-I patients reported experiencing depression for 32% of weeks, whereas BD-II patients reported depression for 50% of weeks, more frequent depressive episodes and shorter inter-episode intervals (Judd et al., 2003).

The diagnosis

BD-II has been under-researched in part due to fuzzy and contentious diagnostic boundaries. The DSM-IV 4-day minimum duration criterion for hypomania was not empirically based. Diagnostic and Statistical Manual of Mental Disorders (5th ed.; DSM-5) retains this time frame despite studies suggesting 2–3 days as a more valid minimum duration (Angst et al., 2003). Muddying the waters further, DSM-5’s addition of ‘increased energy or activity’ as a necessary criterion for hypomania decreases the likelihood of a BD-II diagnosis, while recognising hypomania when symptoms emerge during antidepressant treatment has the opposite effect. DSM-5 now specifies criteria for short-duration hypomania (contributing to ‘other bipolar’ categories). These categories have questionable clinical utility given the diagnostic challenge of differentiating BD-II from borderline personality disorder (BPD). To advance research efforts, improved definition of BD-II is required. Descriptive and discriminant validity of existing diagnostic definitions could be tested in cross-sectional and prospective longitudinal studies exploring points of rarity between current BD diagnoses and related conditions (BPD, unipolar depression).

BD-II in the clinical context

A general lack of knowledge about BD-II exists among clinicians, patients and the community. Most only recognise the ‘classic’ BD-I: mania is unambiguously abnormal and often accompanied by psychotic symptoms. In contrast, BD-II patients generally present when depressed, failing to spontaneously report hypomania. This, combined with poor diagnostic precision among clinicians, contributes to missed BD-II diagnoses. Clinical studies suggest approximately half (30–60%) of BD-II patients are initially diagnosed with unipolar depression, with a correct diagnosis often a decade away. Delays result in inappropriate treatments, commonly a series of failed antidepressant treatment trials or non-specific psychotherapy. The collateral damage to patients in this predicament is substantial.

Once a BD-II diagnosis is established, clinicians commonly apply treatments for BD-I. Adequate evidence of the best interventions for BD-II is lacking as randomised controlled trials (RCTs) primarily comprise BD-I cases. Treatment guidelines reflect this, suggesting extrapolation from BD-I despite lack of evidence that such treatments are optimal for BD-II. The recent statement by Royal Australian and New Zealand College of Psychiatrists (RANZCP) recognises the need for differentiation in pharmacotherapy across the bipolar spectrum, cautioning against extrapolation from BD-I to BD-II. The higher prevalence of depressive symptoms, atypical features, rapid cycling, complex mixed mood states and absence of mania suggest different treatment targets in BD-II.

Where to from here?

The marked imbalance in research towards BD-I over BD-II needs to be addressed. The dearth of research into BD-II partly reflects long-standing scepticism about the validity of the phenotype. While the bipolar spectrum remains poorly understood, emerging data (Song et al., 2017) suggest there is important leverage in investigating BD-II as an entity. Specifically, investigation of pharmacological and psychological treatments targeting BD-II and its predominant depressive polarity has potential to redress clinical needs, while simultaneously collecting data on validity of the BD-II diagnosis. Improved understanding, timely detection and evidence-based intervention will help prevent the lifelong burden of chronic mood instability and reduced quality of life associated with BD-II.