Clinical data are largely focused on depressive comorbidity in OCD. However in practice, treating resistant or severe OCD sufferers revealed many cases who seem to have an authentic OCD with a hidden comorbid bipolar disorder. Most reports had evaluated the OCD comorbidity in unipolar and bipolar mood disorders (Kruger et al., 1995; Chen et Dilsaver, 1995). The only investigation in clinical population focused on the reverse issue was conducted in Pisa. Perugi et al. (1997) have showed in a consecutive series of 315 OCD outpatients, that 15.7% presented a bipolar comorbidity, mostly with BP-II disorder. Further analyses suggested that when comorbidity occurs with bipolar and unipolar depression, it has a differential impact on the clinical picture and course of OCD. The rate of bipolar comorbidity in OCD was analyzed in a recent epidemiological survey undertaken by the French Association of patients suffering from OCD (FA-OCD or AFTOC in French). In a sample of 453 OCD patients, 76% had suffered from a major depression, 11% from bipolar disorder (DSM IV mania or hypomania), 30% from hypomania (cases that obtained a score > or = 10 on the self-rated Angst Hypomania Checklist). According to the score > or = 10 on Self-rated Questionnaire for Cyclothymic Temperament, 50% were classified as cyclothymic. The self-assessment of soft-bipolar dimensions, such as hypomania and cyclothymia was previously validated in a multi-site study in major depression (Hantouche et al., 1998). Further analyses showed that comorbidity with soft bipolarity was characterized by significant interactions with high levels of impulsivity, anger attacks and suicidal behavior. In order to confirm these data, another cohort (n = 175 patients treated by psychiatrists for OCD) was formed and named "PSY-OCD". Comparative analyses between the two populations allowed showing very few demographic and clinical differences. The frequency rate of "bipolar OCD" was equivalent in both populations: BP-II disorder (DSM IV criteria) was present in 11% of FA-OCD and 16% of PSY-OCD. Furthermore using the Hypomania Checklist showed that BP-II disorder rate (score > or = 10) was higher: 32% of in both populations. Cyclothymic rate was also globally higher, but significant difference was obtained: 56% of FA-OCD versus 45% of PSY-OCD (p = 0.02). Moreover, mood switching rate under anti-OCD drugs was equivalent in both OCD populations (respectively 38% and 33%, p = ns). In case of BP comorbidity, patients had presented a greater number of concurrent major depressive episodes and suicidal attempts. When concurrent depression was considered, the rate diagnosis of soft bipolarity was 2.5 fold, and the number of suicidal attempts augmented by 7 fold (by comparison versus non-depressed OCD). Despite very early descriptions (since the beginning of the last century) of particular relationships between so-called "psychasthenia, folie de doute, folie raisonnante" and "circular and intermittent madness or cyclothymia", a few attention has been devoted to this complex pattern of comorbidity. The comparative data deriving from the collaborative survey with patients who are members of AFTOC and with a cohort of psychiatric outpatients, confirm the reality of bipolar-OCD comorbidity, which is largely under-recognized in clinical practice. More in depth analyses are now undertaken in order to investigate the characteristics of "bipolar OCD" by comparison to "non bipolar OCD".