A systematic meta-review of predictors of adverse effect development in response to antidepressant medications

Main Article Content

Emily Snook
Kelly Perlman https://orcid.org/0000-0002-2716-0712
Eleanor H. Brown
Timothé Langlois-Therien
David Benrimoh
Myriam Tanguay-Sela
Colleen Rollins
Eunice You
Marcelo T. Berlim



The pharmacological antidepressant (AD) treatments currently available to treat major depressive disorder (MDD) are associated with numerous adverse events (AEs), including gastrointestinal disturbances, sexual dysfunction, and sleep difficulties. Intolerance of the AEs caused by ADs is cited as a main reason for treatment discontinuation, hindering the recovery process. Therefore, the ability to predict the profile of AEs that an individual patient is likely to experience would hold great clinical value. Here, we review the extant research identifying biological, clinical, and sociodemographic features that are predictive of the incidence and severity of AD-induced AEs. We searched Embase and MEDLINE electronic databases for relevant reviews of all types that discuss predictors of AEs of antidepressant treatment published through March 2018, using a combination of relevant search terms. This protocol, filtered through our inclusion and exclusion criteria, resulted in the inclusion of 29 reviews. Several genetic factors involved in convergent biological processes, including the serotonergic, glutamatergic, and noradrenergic systems, were identified as predictive factors of AE incidence and severity. Non-genetic factors such as age, sex, and pre-existing medical comorbidities were also shown to have predictive value in this context. We could not conclusively determine the directionality of each finding or assign formal levels of evidence; rather, we focused on the emerging patterns of predictive factors. In summary, we systematically curated factors predictive of AD-induced AEs and discussed biological mechanisms that may underlie their predictive value. Additionally, we highlighted the paucity of predictive biomarkers beyond genetics, and emphasized the import of differential prediction.