Cet article synthétise les études originales réalisées entre 1986 et 2008 sur le tempérament selon le modèle de Cloninger dans le trouble bipolaire de l’adulte. Ces mots clés ont été recherchés dans MedLine et PsycInfo : bipolar disorder, mania, temperament, TCI, Cloninger, TPQ, harm avoidance, novelty seeking, persistence, reward dependence. Il existe un évitement du danger et une recherche de nouveauté plus élevés et une détermination plus faible dans le trouble bipolaire comparé à la population générale. Lorsque le trouble bipolaire est comparé à la dépression, les résultats diffèrent entre les études dues à l’hétérogénéité de la phase affective des sujets. Un évitement de danger plus faible, mais une détermination et une coopération plus élevées sont observés dans le trouble bipolaire, comparativement au trouble de personnalité limite. L’évitement de danger est plus élevé chez les bipolaires à début précoce (versus tardif). Les bipolaires ayant un trouble comorbide lié aux substances ont une recherche de nouveauté supérieure et une persistance inférieure à ceux sans abus/dépendance. Les sujets avec une comorbidité anxieuse ont un évitement de danger plus élevé et une détermination plus faible que ceux sans trouble anxieux. Des différences tempéramentales existent dans le trouble bipolaire comparé à d’autres populations cliniques ou générales. La généralisation des observations doit être faite avec prudence, due à l’absence d’unité méthodologique pour l’âge, le sexe, l’appartenance culturelle, l’état affectif, le nombre de sujets recrutés et la version utilisée des questionnaires. Le risque suicidaire, le type clinique et la comorbidité sont inconstamment documentés.

Although psychiatric research uses clear diagnostic criteria to describe bipolar disorders, therapists in clinical practice are often confronted with patients presenting a number of symptoms with different degrees of intensity and belonging to more than one diagnostic category. With respect to this actual clinical complexity, there is an increasing interest in a dimensional approach of psychopathological traits to gain better understanding of mental disorders. In the 1980s, Robert Cloninger elaborated on a psychobiological model to explain personality in clinical groups as well as in general population. His model was then operationalised with a questionnaire evaluating temperament (harm avoidance, novelty seeking, reward dependence and persistence) and character (self-directeness, cooperativeness, self-transcendence): the Temperament and Character Inventory (TCI).

To review all studies conducted in adult bipolar samples on temperament and character according to Cloninger’s psychobiological model.

A search was conducted on MedLine and PsycInfo for all articles written in English or French, between 1986 and September 2008, on temperament and character in bipolar disorder. The words bipolar disorder or mania had to be associated with the following keywords temperament, TCI, Cloninger, TPQ, harm avoidance, novelty seeking, reward dependence.

Across studies, compared to the general population, bipolar subjects have significantly higher harm avoidance, higher novelty seeking and lower self-directness. Some studies have investigated differences between bipolar disorders and other psychopathologies like depression, and borderline personality disorder. Among studies on depression and bipolar disorder, there is no consensus on the findings. Compared to borderline personality disorder patients, bipolar disorder subjects have lower harm avoidance and higher self-directness and cooperativeness. This finding is consistent with Cloninger’s hypothesis that all personality disorders have lower self-directness than any axis I disorder. With respect to other temperament and character traits, studies yielded results either contradictory or non-significant. No difference was found when the bipolar group was subdivided according to the clinical presentation (type I vs II) and the suicidal risk, apart from harm avoidance. Bipolar subjects with substance related disorders displayed higher novelty seeking and lower persistence, which might be explained by a low dopaminergic activity that had to be compensated with drug intake. Low persistence causes greater difficulties to overcome substance dependence. However, it is not possible to determine whether these temperament characteristics are specifically linked to bipolar disorder, substance related disorders, or both. Similar limitations apply to bipolar patients with comorbid anxiety disorders, who presented higher harm avoidance and lower self-directness.

Across studies, there are limitations which impede the generalization of the findings to other clinical populations. Age, gender, cultural characteristics, mood status during evaluation, group size, versions of the TCI questionnaire, suicidal risk, clinical type (BP I vs II) and comorbidity differ from one study to another. These methodological variables should be controlled in future studies. Nevertheless, adult bipolar patients appear to present a different temperamental profile than other clinical groups and general population. Therefore, Cloninger’s psychobiological model of temperament remains an interesting avenue for future researches in bipolar disorder.