The role of affective temperaments in bipolar disorder: The solid role of the cyclothymic, the contentious role of the hyperthymic, and the neglected role of the irritable temperaments.

BACKGROUND: The aim of the present study is to evaluate the role of individual affective temperaments as clinical predictors of bipolarity in the clinical setting. METHODS: The affective temperaments of 1723 consecutive adult outpatients presenting for various symptoms to a university-based mental health clinical setting were assessed. Patients were administered the Hypomania Checklist-32 and the Temperament Evaluation of Memphis, Pisa, Paris, and San Diego - Auto-questionnaire (TEMPS-A) and were diagnosed by psychiatrists according to the DSM-5 criteria. TEMPS-A scores were studied as both continuous and normalized categorical z-scores from a previously established nationwide study on the general population of Lebanon. Simple and multiple binary logistic regressions were done on patients who have any of the DSM-5 defined bipolar types, as a combined group or separately, versus patients without any bipolar diagnosis. RESULTS: At the multivariable level and taking into account all temperaments, the irritable temperament is a consistent predictor of bipolar I and bipolar II disorders. Cyclothymic temperament also played a strong role in bipolarity but more decisively so in bipolar II and substance-induced bipolarity. The hyperthymic temperament had no role in bipolar I or bipolar II disorder.

Negative emotion dysregulation is linked to the intensity of suicidal ideation in a mixed inpatient sample.

BACKGROUND: Suicide has been associated with dysfunctional strategies for emotion regulation but, so far, research findings have been inconclusive. METHODS: To investigate how difficulties in emotion regulation impact suicidal ideation (SI) and behavior, 111 psychiatric inpatients were enrolled. Affective instability (AI), emotional impulsivity (EI), and negative and positive emotionality (NE and PE) were measured by the RIPoSt-40 questionnaire; the first three subscales have been summed to form a total negative emotion dysregulation (NED) score. RESULTS: In the sample, 55 subjects reported at least one-lifetime suicide attempt; 50 patients were diagnosed with mood-disorder (MD), 30 with the schizophrenia-spectrum disorder (SSD), and 15 with personality-disorder (PD). Diagnostic groups differed for NED scores (p=.008) but not for PE (p>0.05), with patients suffering from PD having higher scores (p=0.03). Compared to non-attempters, lifetime-suicide attempters were 6.5 times more likely to have a personality disorder (95% CI=1.34/31.83). Partial correlation analyses, controlling for the presence of suicide attempts, showed that lifetime SI-intensity score was significantly and positively associated with NED (r=.39, p<.001), AI (r=.40, p<.001), and NE (r=.42, p<.001). NED scores (p=.001) and the presence of lifetime suicide attempts (p<.001) were independently associated with lifetime SI-intensity scores. LIMITATIONS: The lack of a non-clinical control group and the cross sectional nature of the study limits the generalizability of the results. CONCLUSION: Our findings support the hypothesis that negative emotion dysregulation is independently associated with SI and behavior. Negative emotion dysregulation should be targeted in suicide prevention.

Reactivity, Intensity, Polarity and Stability questionnaire (RIPoSt-40) assessing emotional dysregulation: Development, reliability and validity.

BACKGROUND: Emotional dysregulation (ED) is a heterogenous construct with great relevance in psychiatric research and clinical practice. In the present study, we validated a 40-items version of the Reactivity, Intensity, Polarity and Stability questionnaire (RIPoSt-40), a self-report measure of ED. METHODS: A non-clinical sample (N = 396) and two clinical samples of patients with cyclothymia (N = 120) and ADHD (N = 54) were recruited. Items were selected and subscales were derived based on inter-item correlations and PCA with promax rotation in the non-clinical sample. Test-retest reliability was assessed in a subsample (N = 60). Internal consistency and concurrent validity with TEMPS-M factors were evaluated in each sample. The three groups results were compared to ascertain discriminant validity. RESULTS: Four subscales were identified as measures of affective instability, emotional impulsivity, negative and positive emotionality. The first three subscales also sum up to a negative ED score comprising thirty items. Measures of reliability (test-retest r = 0.71-0.84) and internal consistency (Cronbach's α = 0.72-0.95) were generally high. Concurrent validity was supported by correlations with TEMPS-M factors. Discriminant validity was significant (p < 0.001) with cyclothymic and ADHD patients showing higher scores for each subscale, except for positive emotionality. LIMITATIONS: The non-clinical sample was recruited through a web-survey and mainly included young and highly educated subjects. Mood and anxiety comorbidity of the clinical samples were not taken into consideration. CONCLUSION: RIPoSt-40 questionnaire has proved to be a valid, reliable and useful tool to assess ED both in clinical and non-clinical contexts.

Diagnosis and Treatment of Cyclothymia: The "Primacy" of Temperament.

BACKGROUND: Contrary to DSM-5 definition based on recurrence of low grade hypomanic and depressive symptoms, cyclothymia is better defined in a neurodevelopmental perspective as an exaggeration of cyclothymic temperament. Emotional dysregulation with extreme mood instability and reactivity is the core features of the complex symptomatology. METHOD: In the present article, we critically reviewed the literature on the diagnosis and treatment of cyclothymia, focusing on the temperamental and neurodevelopmental perspectives. RESULTS: Current epidemiological and clinical research showed the high prevalence and the validity of cyclothymia as a distinct form of bipolarity, frequently associated with multiple comorbidities with anxiety, impulse control, substance use, and so called "personality" disorders. Many patients receive correct diagnosis and treatments after many years of illness, when the superposition of complications reduces the possibility of complete remission. A therapeutic model combining the focus on symptomatic presentations with a temperamental perspective seems to represent an effective approach for cyclothymic patients with complex clinical presentations. CONCLUSION: Cyclothymic mood instability is an understudied issue despite the evidence of its clinical relevance. Unresolved issues concern its diagnostic delimitation and the possible relationships with emotional dysregulation observed in other neurodevelopmental disorders. We need to confirm the specificity of the disorder and to improve its recognition in early phase of the life, especially in youth. Early recognition means avoiding unnecessary complications and establishing specific treatments and clinical management since the beginning.

Temperament and suicide: A national study.

BACKGROUND: Several studies have shown temperament variants in suicidality. Yet, to our knowledge, the association between temperaments and suicide attempts has not been studied on a nationally representative level nor systematically in subjects with no mental disorders. Also, although hyperthymic temperament is recognized as protective of most mental disorders, its role in the protection from self-harm remains inconclusive. METHODS: The study is based on nationally representative data of all Lebanese adults. Mental disorders were assessed using the Composite International Diagnostic Interview, whereas the five affective temperaments were assessed using the TEMPS-A. RESULTS: Anxious temperament is a solid and strong risk factor for suicide attempts in subjects with (OR: 10.1) and without (OR: 9.0) mental disorders. Depressive (OR: 4.3) and irritable (OR: 5.1) temperaments are risk factors for suicide attempt among subjects with mental disorders. Hyperthymic temperament plays a dual role in females with mental disorders: while the hyperthymic trait "having self-confidence" is strongly protective of suicide attempts, "liking to be the boss", "getting into heated arguments", and "the right and privilege to do as I please" are hyperthymic risk traits for suicide attempts reflecting the "dark side" of the hyperthymic temperament. Interestingly, these three hyperthymic risk traits--in the absence of "having self-confidence"--are a universal risk for suicide attempt in females with mental disorder. LIMITATIONS: Social desirability could have led to the under-reporting of suicide attempts and mental disorders. CONCLUSIONS: The anxious temperament plays a strong role in predicting suicide attempts in the community, in the presence and absence of diagnosable mental disorders. The irritable and the depressive temperaments are additional risks in subjects with mental disorders. The dual role of the hyperthymic temperament is quite interesting: while it is protective of suicidal behavior, it also has a dark side in subjects with mental disorders.

Cyclothymia reloaded: A reappraisal of the most misconceived affective disorder.

Data emerging from both academic centers and from public and private outpatient facilities indicate that from 20% to 50% of all subjects that seek help for mood, anxiety, impulsive and addictive disorders turn out, after careful screening, to be affected by cyclothymia. The proportion of patients who can be classified as cyclothymic rises significantly if the diagnostic rules proposed by the DSM-5 are reconsidered and a broader approach is adopted. Unlike the DSM-5 definition based on the recurrence of low-grade hypomanic and depressive symptoms, cyclothymia is best identified as an exaggeration of cyclothymic temperament (basic mood and emotional instability) with early onset and extreme mood reactivity linked with interpersonal and separation sensitivity, frequent mixed features during depressive states, the dark side of hypomanic symptoms, multiple comorbidities, and a high risk of impulsive and suicidal behavior. Epidemiological and clinical research have shown the high prevalence of cyclothymia and the validity of the concept that it should be seen as a distinct form of bipolarity, not simply as a softer form. Misdiagnosis and consequent mistreatment are associated with a high risk of transforming cyclothymia into severe complex borderline-like bipolarity, especially with chronic and repetitive exposure to antidepressants and sedatives. The early detection and treatment of cyclothymia can guarantee a significant change in the long-term prognosis, when appropriate mood-stabilizing pharmacotherapy and specific psychological approaches and psychoeducation are adopted. The authors present and discuss clinical research in the field and their own expertise in the understanding and medical management of cyclothymia and its complex comorbidities.

Gender differences in a cohort of major depressive patients: further evidence for the male depression syndrome hypothesis.

BACKGROUND: Previous studies have shown that major depressive patients may differ in several features according to gender, but the existence of a specific male depressive syndrome remains controversial. METHODS: As part of the EPIDEP National Multisite French Study of 493 consecutive DSM-IV major depressive patients evaluated in at least two semi-structured interviews 1 month apart, 125 (27.7%) were of male gender, whereas 317 (72.3%) were female, after exclusion of bipolar I patients. RESULTS: Compared to women, men were more often married, had more associated mixed features, with more bipolar disorder NOS, more hyperthymic temperaments, and less depressive temperaments. Women had an earlier age at onset of depression, more depressive episodes and suicide attempts. A higher family loading was shown in men for bipolar disorder, alcohol use disorder, impulse control disorders and suicide, whereas their family loading for major depressive disorder was lower. Men displayed more comorbidities with alcohol use, impulse control, and cardiovascular disorders, with lower comorbidities with eating, anxiety and endocrine/metabolic disorders. The following independent variables were associated with male gender: hyperthymic temperament (+), alcohol use disorder (+), impulse control disorders (+), and depressive temperament (-). LIMITATIONS: The retrospective design and the lack of specific tools to assess the male depressive syndrome. CONCLUSION: Study findings may lend support to the male depression syndrome concept and draw attention to the role of hyperthymic temperament, soft bipolarity as well as comorbidities as determinants of this syndrome. The latter could help recognize an entity which is probably underdiagnosed, but conveys a high risk of suicide and cardiovascular morbidity.

DSM-IV diagnosis in depressed primary care patients with previous psychiatric ICD-10 bipolar disorder.

BACKGROUND: In the past 20 years, much evidence has accumulated against the overly restrictive diagnostic concepts of hypomania in DSM-IV and DSM-IV-TR. We tested DSM-IV-TR and a broader modified version (DSM-IV-TRm) for their ability to detect bipolarity in patients who had been treated for bipolar disorders (BD) in psychiatric settings, and who now consulted general practitioners (GPs) for new major depressive episodes (MDE). METHODS: Bipolact II was an observational, single-visit survey involving 390 adult patients attending primary care for MDE (DSM-IV-TR criteria) in 201 GP offices in France. The participating GPs (53.3 ± 6.5 years old, 80.1% male) were trained by the Bipolact Educational Program, and were familiar with the medical care of depressive patients. RESULTS: Of the 390 patients with MDE, 129 (33.1%) were previously known as bipolar patients (ICD-10 criteria). Most of the latter bipolar patients (89.7%) had previously been treated with antidepressants. Only 9.3% of them met DMS-IV-TR criteria for BD. Conversely, 79.1% of the 129 bipolar patients met DMS-IV-TRm criteria for BD and showed strong associations with impulse control disorders and manic/hypomanic switches during antidepressant treatment. LIMITATIONS: Limited training of participating GPs, recall bias of patients, and the study not being representative for untreated bipolar patients. CONCLUSIONS: Very few ICD-10 bipolar patients consulting French GPs for MDE met DSM-IV-TR criteria for bipolar diagnosis, which suggests that DSM-IV-TR criteria are insufficient and too restrictive for the diagnosis of BD. DSM-IV-TRm was more sensitive, but 20% of bipolar patients were undetected.

Risks associated with gender differences in bipolar I disorder.

BACKGROUND: Previous studies have demonstrated that bipolar patients may differ in several features according to gender, but a number of the differences found remain controversial. METHODS: The demographic, illness course, clinical, comorbidity and temperament characteristics of a total of 1090 consecutive DSM-IV bipolar I manic inpatients were compared according to gender. RESULTS: Bipolar illness in women was characterised by the predominance of depression, as indicated by a depressive polarity at onset, higher rates of mixed mania, more suicidal behaviour, and a greater number of temperaments with depressive propensities. In contrast, the manic component was found to predominate in men. Men also had an earlier onset of their illness. Women displayed more comorbidities with eating, anxiety, and endocrine/metabolic disorders, whereas men were more comorbid with alcoholism and other forms of substance abuse, neurological, and cancer disorders. The following independent variables were associated with male gender: being single (+), depressive temperament (-), excessive alcohol use (+), cyclothymic temperament (-), excessive other substance use (+), mood congruent psychotic features (+), and manic polarity at onset (+). LIMITATIONS: The retrospective design and the sample being potentially not representative of the bipolar disorder population are limitations. CONCLUSIONS: Findings from this study tend to confirm most of the differences previously observed among bipolar men and women. Furthermore, these results draw attention to the risks that may be specifically linked to gender differences in bipolar I patients.

Characteristics and profiles of bipolar I patients according to age-at-onset: findings from an admixture analysis.

BACKGROUND: Many studies have used admixture analysis to separate age-at-onset (AAO) subgroups in bipolar patients, but few have looked at the phenomenological characteristics of these subgroups, in order to find out phenotypic markers. METHODS: Admixture analysis was applied to identify the model best fitting the observed AAO distribution of a sample of 1082 consecutive DSM-IV bipolar I manic inpatients who were assessed for demographic, clinical, course of illness, comorbidity, and temperamental characteristics. RESULTS: The model best fitting the observed distribution of AAO was a mixture of three Gaussian distributions. We could identify three AAO subgroups: early, intermediate, and late age-at-onset (EAO, IAO, and LAO, respectively). Patients in the EAO subgroup were more often single young males exhibiting severe mania with psychotic features, a subcontinuous course of illness with substance use and panic comorbidity, more suicide attempts, and temperamental components sharing hypomanic features. Patients with LAO showed a less severe picture with more depressive temperamental components, alcohol use and comorbid general medical conditions. A less typical phenotype was present in IAO patients. LIMITATIONS: The following are the limitations of this study: retrospective design, and bias toward preferential enrollment of patients with manic predominant polarity. CONCLUSIONS: This study confirms that bipolar I disorder can be subdivided into three subgroups based on AAO distribution and shows that patients from these subgroups differ in phenotypes.

Self-assessment and characteristics of mixed depression in the French national EPIDEP study.

BACKGROUND: Studies on mixed depression have been conducted so far on the basis of DSM-IV manic symptoms, i.e., a list of 7 symptoms which may provide limited information on the subsyndromal features associated with a full depressive episode. METHODS: As part of the EPIDEP National Multisite French Study of 493 consecutive DSM-IV major depressive patients evaluated in at least two semi-structured interviews 1 month apart, 102 (23.8%) were classified as mixed depressives (≥3 hypomanic symptoms), and 146 (34%) as pure depressives (0 hypomanic symptom), after exclusion of bipolar I patients; hypomanic symptoms were assessed with the Multiple Visual Analog Scales of Bipolarity (MVAS-BP, 26 items) of Ahearn-Carroll in a self assessment format. A narrower definition of mixed depression, resting on those MVAS-BP items referring to DSM-IV hypomanic symptoms was also tested, as a sensitivity analysis. RESULTS: Compared to pure depressives, mixed depressive patients had more psychotic symptoms, atypical features and suicide attempts during their index episode; their illness course was characterized by early age at onset, frequent episodes, rapid cycling, and comorbidities. Mixed depressive patients were more frequently bipolar with a family history of bipolar disorder, alcohol abuse, and suicide. A dose-response relationship was found between intradepression hypomania and several clinical features, including temperament measures. The following independent variables were associated with mixed depression: hyperthymic temperament, cyclothymic temperament, irritable temperament, and alcohol abuse. Using the narrower definition of mixed depression missed risk factors such as suicidality and comorbidities. LIMITATIONS: The following are the limitations of this study: retrospective design, recall bias, lack of sample homogeneity, no cross-validation of findings by hetero-evaluation of hypomanic symptoms. CONCLUSIONS: EPIDEP data showed the feasibility and face validity of self-assessment of intradepressive hypomania. They replicated previous findings on the severity and high suicidal risk of mixed depression profile. They confirmed, for mixed depression, that mixed states occur when mood episodes are superimposed upon temperaments of opposite polarity. They finally suggested that a definition of mixed depression only based on DSM-IV-TR hypomanic symptoms may not allow to identify the most unstable subforms of the entity.

Correlates of first-episode polarity in a French cohort of 1089 bipolar I disorder patients: role of temperaments and triggering events.

OBJECTIVES: As only a few studies so far systematically reported on bipolar patients subtyped according to first-episode polarity, we took the opportunity of having at disposal a large sample of bipolar I patients to specify the characteristics of patients included in these subtypes, with a special focus on temperament and triggering events. METHODS: A total of 1089 consecutive DSM-IV bipolar I manic inpatients were subtyped in manic onset (MO), depressive onset (DO) and mixed onset (MXO), and assessed for demographic, illness course, clinical, psychometric, comorbidity and temperament characteristics. RESULTS: The main characteristics of MO patients were a hyperthymic temperamental predisposition, a first episode triggered by substance abuse and an illness course with pure, severe and psychotic mania. In comparison, DO patients had more depressive temperaments, a first episode triggered by stress and alcohol, an illness course with more episodes, cyclicity, suicide attempts, anxious comorbidity and residual symptoms. Although sharing characteristics with either MO or DO, MXO patients had more mixed episodes and cyclothymic temperament. LIMITATIONS: The following are the limitations of this study: retrospective design, bias toward preferential enrolment of MO patients, and lack of information on the number and polarity of lifetime episodes. CONCLUSIONS: Findings from this study tend to confirm most of the differences previously evidenced among patients subtyped according to first-episode polarity. Differences found in temperamental predisposition and illness onset triggering events are worth noting and may help target early preventive interventions as well as orientate the search for specific genetic risk factors.

The role of anxious and hyperthymic temperaments in mental disorders: a national epidemiologic study.

Temperament has been demonstrated clinically to be linked to mental disorders. We aimed to determine the possible role of temperament in mental disorders in a national epidemiologic study. A nationally representative sample of adults (n=1320) was administered the Lebanese-Arabic version of the Temperament Evaluation of the Memphis, Pisa, Paris and San Diego Autoquestionnaire (TEMPS-A), and the Arabic CIDI 3.0, as part of the LEBANON study. The association among temperaments and DSM-IV mood, anxiety, and impulse control disorders was assessed. The anxious temperament was shown to be a robust predictor of most disorders, especially within the anxiety and depressive clusters. The hyperthymic temperament had a uniquely protective effect on most mental disorders, with the exception of separation anxiety, bipolar, substance abuse and impulse control disorders. These effects were moderated by age and education. Temperaments, previously largely neglected in epidemiologic studies, could play a major role in the origin of mental disorders.

Explained factors of suicide attempts in major depression.

OBJECTIVE: The aim of this study is to identify risk factors for suicide attempts including bipolarity. METHOD: The paper presents the most recent data on suicide attempts and depression with or without hypomanic features from three French "Bipolact", studies including 2249 patients with recurrent or resistant depression. Hypomania and BP-II disorder were defined by a score of 10 or more on the Hypomania Checklist-20. Attempters and non-attempters were compared, and multivariate logistic regression analyses were performed on all the significant variables obtained in univariate tests. RESULTS: Rates of suicide attempts and of a family history of suicide were higher in BP-II disorder. Suicide attempts were best explained by a family history of suicide and mood disorders, recurrence of depression, the "irritable-risk-taking" dimension of hypomania, substance abuse, and need of psychiatric treatment. LIMITATIONS: The study does not deal with DSM-IV BP-II disorder. CONCLUSION: Clinicians need to be familiarized with these risk factors.

Psychopathological correlates of lifetime anxiety comorbidity in bipolar I patients: findings from a French national cohort.

BACKGROUND: Despite numerous studies on the comorbidity of bipolar and anxiety disorders, there is no satisfactory psychopathological model for their overlap. METHOD: 1,090 hospitalized patients meeting DSM-IV criteria for a manic episode of bipolar I disorder were subtyped according to the presence or not of lifetime anxiety comorbidity and assessed for demographic, illness course, clinical, associated condition, temperament, and treatment characteristics. RESULTS: Lifetime anxiety comorbidity, defined as presence of at least one anxiety disorder in lifetime, was found in 27.2% (n = 297) of the sample. Compared to patients without such a comorbidity (n = 793), those who had it experienced a higher number of mood episodes and suicide attempts in the previous year, more stressors, organic disorders and less free intervals; furthermore, they showed more temperaments with depressive features and complex treatment. At study entry, they also experienced manic episodes with higher levels of depression, psychosis and hostility. The following independent variables were associated with lifetime anxiety comorbidity: higher scores on the Montgomery-Asberg Depression Rating Scale, depressive temperament, irritable temperament, higher scores on the Scale for the Assessment of Positive Symptoms, episodes without free intervals and at least one stressor before the index episode. CONCLUSIONS: Factors associated with lifetime anxiety comorbidity in bipolar I patients may be integrated into a comprehensive diathesis-stress model emphasizing the role of irritable temperament as a source of mood instability and stress, and interacting with other temperamental characteristics to trigger the outbreak of both anxiety and bipolar symptoms.

Risk factors associated with lifetime suicide attempts in bipolar I patients: findings from a French National Cohort.

Risk factors that may be associated with suicide attempts in bipolar disorder are still a matter of debate. We compared demographic, illness course, clinical, and temperamental features of suicide attempters vs those of nonattempters in a large sample of bipolar I patients admitted for an index manic episode. One thousand ninety patients (attempters = 382, nonattempters = 708) were included in the study. Multivariate analysis evidenced 8 risk factors associated with lifetime suicide attempts as follows: multiple hospitalizations, depressive or mixed polarity of first episode, presence of stressful life events before illness onset, younger age at onset, no free intervals between episodes, female sex, higher number of previous episodes, and cyclothymic temperament. These characteristics may help identify subjects at risk for suicide attempt throughout the course of bipolar disorder. We finally propose to integrate such characteristics into a stress-diathesis model of suicidal behavior, adapted to bipolar patients.

Factors associated with rapid cycling in bipolar I manic patients: findings from a French national study.

INTRODUCTION: Despite numerous explanatory hypotheses, few studies have involved a large national clinical sample examining risk factors in the occurrence of rapid cycling during the course of bipolar illness. METHODS: From 1,090 manic bipolar I disorder inpatients included in a multicenter national study in France, 958 could be classified as rapid or nonrapid cyclers and assessed for demographic, illness course, clinical, psychometric, temperament, comorbidity, and treatment characteristics. RESULTS: Rapid cycling bipolar disorder occurred in 9% (n=86) of the study group. Compared to nonrapid cyclers (n=872), patients with rapid cycling experienced the onset of their illness at a younger age, a higher number of prior episodes, more depression during the first episode, and more suicide attempts. At study entry, they also experienced manic episodes with more depressive and anxious symptoms, but less psychotic features. The following independent variables were associated with rapid cycling: longer duration of illness, antidepressant treatment, episodes with no free intervals, cyclothymic temperament, lower scores on the Scale for Assessment of Positive Symptoms and presence of thyroid disorder. Retrospective study limited to bipolar I disorder inpatients; several factors previously associated with rapid cycling were not assessed. CONCLUSION: Our findings may confirm previous descriptions, according to which rapid cycling develops later in the course of illness following a sensitization process triggered by antidepressant use or thyroid dysfunction, in patients with a depression-mania-free interval course, and cyclothymic temperament.

Toward the delineation of mania subtypes in the French National EPIMAN-II Mille Cohort. Comparisons with prior cluster analytic investigations.

BACKGROUND: Knowledge about psychopathologic presentations of mania in current clinical practice has to be refined in order to improve diagnosis and treatment. METHODS: One thousand ninety manic patients included in the French National Study EPIMAN-II Mille were submitted to a cluster analysis on the basis of multiple variables related to the history of bipolar illness and symptoms of the current episode. RESULTS: Four clusters were identified: "classic mania" (29.3% of patients) with less severe mania; "psychotic mania" (22.7%) with psychotic symptoms, more severe mania, younger age and social impairment; "depressive mania" (30.4%) characterized by female gender, suicide attempts, high number of previous episodes and residual symptoms; and "dual mania" (17.6%) characterized by male gender, substance use, earlier onset and poor compliance. Patients groups also differed in manic symptoms, marital status, stressors preceding illness onset, prior diagnoses, first episode polarity and temperamental characteristics. LIMITATIONS: Cross-sectional assessment of patients. CONCLUSIONS: In comparing our findings with those of four prior cluster analytic studies, we integrate clinical characteristics of mania subtypes found in this very large representative French sample in contemporary practice, we suggest how such convergence of data may help improve earlier recognition, differential response to different treatments, and prevention of these subtypes. We finally suggest that such subtyping might provide clues to phenotype delineation suitable for pharmacogenetic investigations.