Introduction

The distinction between unipolar and bipolar forms, rooted in the work of Pierre Falret (1851) and Jules Baillarger (1854), was later established by Karl Kleist (1928, 1953) and his school (Neele, 1949; Leonhard, 1957), and subsequently validated by Angst (1966), Perris (1966), and Winokur et al. (1969), who showed that clinical, familial, and course features supported the nosological differentiation between unipolar and bipolar disorders (Angst and Marneros, 2001; Marneros, 2001). However, there are many areas of overlap between those extremes, pointing up the question of possible clinical subtypes in the interface of depressive and manic extremes of affective illness (Akiskal, 2002a).

Bipolar disorder occurs in multiple forms and degrees of severity. The recognition of the existence of so-called milder forms of manic-depressive illness has been a major endeavor in the last decade. The distinctions hinge on the classification of elated states and this poses some difficulty because it depends on the arbitrary gradation of severity and duration. Bipolar disorder with mania and strict unipolar depression without manic or hypomanic episodes would represent the extremes of a spectrum (Akiskal, 1983); recurrent depressions with hypomania would occupy a middle territory (Akiskal, 2002b). Goodwin and Jamison (1990) point out that the exploration of spectrum models of manic-depressive illness would enhance research on genetic markers and modes of genetic transmission, provide an approach for identifying individuals at risk for the development of bipolar illness, and permit the evaluation of treatments for milder forms, including the question of whether early intervention could lessen the chance of progression to bipolar illness.