[Methodological approach to inter "guideline" variability in the management of bipolar disorders].

INTRODUCTION: In recent decades, an increasing number of pharmacologic agents have become available in bipolar disorder treatment. These therapeutic advances provide a new challenge for clinicians in the choice of medication for patients with bipolar disorder. In this context, tools have been developed for making medical decisions in the management of bipolar disorder: guidelines.

METHODS: Guidelines for bipolar disorder were compared on the basis of their construction methodology (evidence-based treatment guidelines or consensus-based treatment guidelines), results and recommendations for clinical practice.

RESULTS: There are differences between guidelines for treating bipolar disorder. For the American Psychiatric Association (APA), the severity of the manic episode is a primary endpoint of the decision-making tree for the choice of therapy. On the other hand, the National Institute for Health and Clinical Excellence (NICE) ruled that the choice of the initial treatment, in the case of manic episode, should be based first on the current patient's treatment (history of anti-manic therapy) while the World Federation of Societies of Biological Psychiatry (WFSBP) emphasizes the clinical classification of the type of mania. The sequencing of medication in the guidelines may vary according to the construction methodology, the date of elaboration, the geocultural context and experts' position. Recent guidelines consider the last randomized controlled trials (RCT) as those of aripiprazole in the treatment of mania, recommending it in first line as anti-manic agent. The recent updated WFSBP guidelines changed in its construction methodology taking into account the negative studies or those showing non-superiority compared to placebo. Thus, a recent study of non-superiority of lithium monotherapy compared to placebo in the treatment of bipolar depression downgraded lithium from level of evidence B to D. During recent years, a large number of RCT have demonstrated superior efficacy (particularly in mania treatment) of lithium or valproate combined with second-generation antipsychotic compared with lithium or valproate monotherapy. Consequently, according to geocultural context or experts' position, some guidelines recommended medication combinations in first line (Canadian Network for Mood and Anxiety Treatment) and other guidelines considered monotherapy in first line (except for particular cases) to promote tolerance and good therapeutic alliance (WFSBP). Malhi et al. recommended a sequencing of medication based on the benefit risk ratio for the management of each phase of bipolar disorder. These differences between guidelines may cause difficulties for clinicians in choosing clinical practice guidelines.

CONCLUSION: While there are a large number of guidelines for bipolar disorder, the recommendations may vary depending on multiple factors. It seems interesting to conduct a comparative study of guidelines for bipolar disorder on the basis of a validated scale (AGREE) or completed by other items such as date of elaboration and number of proposed recommendations. However, the methodological understanding of guidelines remains the central element for practitioners in their choice of guidelines. Thus, the initial objective of guidelines "to develop statements to assist clinician and patient decisions about the most appropriate health care for specific clinical situations" could be implemented in clinical practice.