Systematic review of randomized controlled trials of candidate treatments for cognitive impairment in depression and methodological challenges in the field

摘要

Cognitive impairment is a core feature of Major Depressive Disorder (MDD) but treatments targeting cognition are lacking. Numerous pre-clinical and clinical studies have investigated potential cognition treatments, but overall the evidence is conflicting. We conducted a systematic search following the PRISMA guidelines on PubMed and Psychlnfo to evaluate the extant evidence and methodological challenges in randomized controlled trials (RCTs) of biological, psychological and behavioural candidate treatments targeting cognition in MDD. Inclusion criteria were RCTs with a placebo control assessing potential pro-cognitive effects of candidate treatments in MDD. Two independent authors reviewed the studies and assessed their risk of bias with the Cochrane Collaboration's Risk of Bias tool. Twenty-eight eligible studies (24 biological and four psychological or behavioural studies) were identified. Cognition was the primary treatment target in ten (36%) trials and an additional treatment outcome together with mood symptoms in 18 (64%) trials. The risk of bias was high or unclear in 93% of trials due to potential selective outcome reporting or 'pseudospecificity' (unspecific cognitive improvement due to reduced depression severity), and/or insufficient details on how the allocation sequence was generated or how blinding was maintained. Several promising treatments were identified, including vortioxetine, erythropoietin, transcranial direct current stimulation and cognitive remediation. However, several common methodological challenges may impede advances in the field. In particular, future trials should select one cognitive composite score as primary outcome, screen for cognitive impairment before inclusion of participants and address 'pseudospecificity' issues. Together, these strategies may improve the success of future cognition trials in MDD. (C) 2016 Elsevier B.V. and ECNP. All rights reserved.