Bulimia nervosa in atypical depression: the mediating role of cyclothymic temperament

Recent data indicate significant clinical, biological, and treatment response overlap between eating and bipolar disorders, especially when soft symptoms of either spectrum disorders are considered. The aim of the present analyses is to evaluate the lifetime prevalence of bulimia nervosa (BN) in patients with atypical depression (AD) and to delineate any demographic, clinical, personality or temperamental factors that may characterize this subgroup.

We examined in a semi-structured format 107 consecutive patients who met DSM-IV criteria for major depressive episode with atypical features and we separated them into two groups according to the co-occurring criteria for BN. They were further evaluated on the basis of the Atypical Depression Diagnostic Scale (ADDS), the Hopkins Symptoms Check-list (HSCL 90), and the Hamilton Rating Scale for Depression (HRSD), coupled with its modified form for reverse vegetative features, as well as Axis I and II comorbidity and temperamental dispositions.

Seventeen (17.8%) percent of AD met the DSM-IV criteria for Bulimia Nervosa (BN+). These patients, compared with those who did not meet criteria for BN (BN-), were indistinguishable on all demographic and most psychopathologic and clinical features (including bipolar I and II), but were significantly higher in lifetime comorbidity for Narcissistic, Histrionic, Borderline and Dependent personality disorders as well as that for Cyclothymic temperament. BN+ also scored higher on the ADDS items of reactivity of mood and interpersonal sensitivity.

Correlational clinical study in which doctors could not be entirely blind to the variables under investigation.

Cyclothymic temperament and related mood reactivity and interpersonal sensitivity may account for much of the relationship between AD and BN. Narcissistic, histrionic and borderline traits, too, seem to be related to the presence of a cyclothymic disposition. The data overall, in particular the cyclothymic reactivity in the absence of differences in BP-I and II, all support the hypothesis that places BN in the "ultra-soft" bipolar realm.