It is perhaps prudent to preface any discussion about this diagnostic category by standing back and looking at diagnoses more broadly. In no other branch of medicine does pathophysiology linearly track phenomenology, yet we are limited to phenomenological-based diagnostic boundaries. Biomarkers have shown scant respect for the most commonly used diagnostic categories, and treatments too have minimal diagnostic specificity. The only clear caveat perhaps is lithium, whose profile of efficacy is in the domain of classical bipolar disorder and the cyclical recurrent mood disorders. For the rest, we need to acknowledge that we are parsing the clouds on an overcast day. The counterpoint is that we need to make major clinical and treatment decisions on the basis of a very incomplete evidence base, drawing on the basis of experience, interpretation and bias simultaneously from the bipolar and unipolar literature, without a clear pathophysiological foundation.