What not to use in bipolar disorders: A systematic review of non-recommended treatments in clinical practice guidelines

BACKGROUND: Clinical practice guidelines (CPG) are an important tool for implementation of evidence-based clinical care. Despite clinical trials showing lack of efficacy of some agents in bipolar disorder (BD), they are still frequently prescribed in clinical practice. The objective of this study was to systematically review the CPG recommendations on pharmacological interventions with evidence against their use due to lack of efficacy data and/or due to serious safety concerns. METHODS: A systematic literature search identified 29 guidelines published by national and international organizations during the 1994-2020 period. Information was extracted regarding how the recommendations framed non-use of treatments in particular clinical situations as well as the actual recommendation in the guideline. RESULTS: Twenty-three guidelines (79%) mentioned at least one non-recommended treatment. The terms used to qualify recommendations varied amongst guidelines and included: "not recommended" "no recommendation" and "negative evidence". Lamotrigine, topiramate and gabapentin were commonly cited as non-recommended treatments for mania and most CPG did not recommend monotherapy with antidepressants, aripiprazole, risperidone, and ziprasidone for treatment of acute bipolar depression. Most guidelines made recommendations about lack of efficacy data or potential harm in treatments for BD but there is a significant variation in the way this information is conveyed to the reader. LIMITATIONS: Non-recommended treatments were based on their use for BD episodes or maintenance but specific medications may benefit patients when treating comorbid conditions. CONCLUSIONS: The absence of a uniform language and recommendations in current guidelines may be an additional complicating factor in the implementation of evidence-based treatments in BD.