ObjectivesClinical heterogeneity during euthymic states is a crucial issue in bipolar disorders. Indeed, actual data are not sufficient to understand why some patients are unharmed by subthreshold symptoms and have functional impairments whereas others have a functional remission but have subthreshold symptoms. Based on the Ball model, cognitive and schematic vulnerability interact with genetic vulnerability and trigger affective symptoms with the intervention of stressful life events. Furthermore, according to this model, adjustment and adaptation to illness assessed by functional outcome and illness experience are associated with this cognitive and schematic vulnerability. So, theoretical arguments support that childhood adversity and temperamental deregulation characterize patients with bipolar disorders. Thus, the aim of this study is to systematically review studies of Early Maladaptive Schemas in bipolar disorder, to determine whether Early Maladaptive Schemas have specificity in bipolar disorder in comparison with other populations, and to identify which Early Maladaptive Schemas could be activated. The challenge of this review is to identify if the taking of early maladaptive schemas into account could allow us to better identify, understand and manage bipolar disorders. MethodsThis systematic review was led according to the Preferred Reporting Items for Systematic review and Meta-Analysis statement on the electronic databases Cochrane, PsycArticles, Psychology and Behavioral Sciences Collection, PsycInfo, PubMed, ScienceDirect and Scopus with « early maladaptive schemas » AND « bipolar disorder » as keywords. Only studies meeting eligibility criteria concerning publication status, language, population and outcomes were included after several screenings on basis of title, abstract and full-text. Then, we carried out data extraction in accordance with criteria defined in principle (about characteristics of participants, objectives, materiel and methods, principle results and bias). ResultsAmong 39 records identified, a total of 10 studies met eligibility criteria for inclusion in this review. Synthesizing findings across the studies revealed three important topics. First, early maladaptive schemas appear as potential cognitive characteristics that clinicians have to investigate in clinical practice. Indeed, patients with bipolar disorders present greater activation of the early maladaptive schemas in comparison with people who have no disorder. This point supports the first part of Ball's theoretical model that considers schemas as a vulnerability to bipolarity. Secondly, early maladaptive schemas are relevant to distinguish bipolar disorders from unipolar depression and borderline personality disorder. A greater and a lower activation are respectively identified among bipolar disorders. Thirdly, supporting the second part of Ball's model, early maladaptive schemas play a key role in recovery regarding their impact on the course of bipolarity, in particular on suicidality and functional impairment. Finally, these dysfunctional schemas allow us to understand the clinical heterogeneity of bipolar disorder, and among others, about the type of bipolarity. These results have several implications, but there are some limits in this systematic review. First, no French study has been done. Then, reduced sample sizes in these studies increased the risk to conclude wrongly to an activation difference between groups. Furthermore, probably due to the variety of methods and populations, we could not identify an homogeneous pattern of early activated maladaptive schemas. Overall, scientific approaches used in these studies are based on statistical models using mean and standard deviation. These types of statistical analyses are the main limit because they cannot represent the heterogeneity of early maladaptive schemas profiles. ConclusionsSchema theory proves to be a relevant approach in bipolar disorders, and early maladaptive schemas appear to be important to take into account in clinical practice. Nevertheless, in order to propose schemas therapy appropriately, it is necessary to specify if early maladaptive schemas are activated and to specify therapeutic indications because of clinical heterogeneity. Moreover, data do not yet allow us to understand the disparity of profiles during the inter-episode period. Indeed, a French research perspective is being considered that will prefer a person-oriented approach.