Effect of pharmacological and neurostimulation interventions for cognitive domains in patients with bipolar disorder: A systematic review and network meta-analysis of randomized controlled trials

Abstract

Introduction: The priority of interventions to alleviate cognitive deficits in patients with bipolar disorder (BD) is inconclusive. We systematically evaluate the efficacy of pharmacological or neurostimulation interventions for cognitive function in BD through a network meta-analysis. Methods: The PubMed, PsycINFO, Embase, and Cochrane Library databases were searched from database inception to September 30, 2021. Following PRISMA guidelines, all eligible studies were randomized controlled trials of adult bipolar patients that provided detailed cognitive outcomes. Studies were excluded if participants limited to comorbid substance use disorder or the intervention was a psychotherapy. Network meta-analysis comparing different interventions was conducted for 8 cognitive domains. Partially ordered set with Hasse diagram was used to resolve conflicting rankings between outcomes. The study was preregistered on PROSPERO database (CRD42020152044). Results: Total 21 RCTs including 42 tests for assessing intervention effects on cognition were retrieved. Adjunctive erythropoietin (SMD = 0.61, 95% CI = 0.00–1.23), Withania somnifera (SMD = 0.58, 95% CI = 0.03–1.13), and galantamine (SMD = 1.22, 95% CI = 0.10–2.35) was more beneficial for attention, working memory, and verbal learning in euthymic BD patients than treatment as usual, respectively. Hasse diagram suggested ranking of choice when multiple domains were combined. Conclusion: Considerable variability in measurements of cognitive domains in BD was observed, and no intervention resulted in superior benefits across all domains. We suggested interventions priority can be tailored according to individual patients’ cognitive deficits. As current findings from relatively small and heterogeneous dataset, future trials with consensus should be applied for building further evidence.