Is there consensus across international evidence-based guidelines for the psychotropic drug management of bipolar disorder during the perinatal period?

Background: Clinicians treating a patient with bipolar disorder who is pregnant or breastfeeding may seek advice from bipolar management guidelines that provide recommendations for perinatal treatment. We examine the consistency of such recommendations across several evidence-based guidelines.

Methods: A literature search in the National Guideline Clearinghouse, the Cochrane Database of Systematic Reviews, PsycInfo and PubMed was undertaken using the search terms "bipolar disorder" and "guidelines," which generated 11 sets of evidence-based guidelines published by professional organizations during the 2005-2015 period. Information relevant to management during the perinatal period was reviewed by two independent reviewers, with key themes qualitatively analysed.

Results: There was a moderate level of agreement across guidelines regarding the potential teratogenic effects of lithium, sodium valproate and carbamazepine, with most highlighting caution in using these medications during the perinatal period. There was less agreement regarding the safety risks associated with lamotrigine, antipsychotics, and antidepressants, and little agreement regarding the risks and recommendations of medications during breastfeeding. Limitations: Some differences in recommendations are likely due to varying publication dates, with recent guidelines having more up-to-date evidence available to use when formulating recommendations. Further, due to ethical issues surrounding pregnancy and infant research, the evidence used to formulate perinatal recommendations is largely based on retrospective reports and/or case studies. It is therefore unrealistic to expect such recommendations to be entirely consistent and based on rigorous evidence.

Conclusions: While there was some consistency across guidelines on key recommendations, there were also substantial inconsistencies, with the latter risking compromising clinical management.