The Comparative Efficacy of Noninvasive Brain Stimulation for Suicidal Ideation: A Network Meta-Analysis

OBJECTIVE: Noninvasive brain stimulation (NIBS) holds promise for reducing suicidal ideation (SI), but its cross-diagnostic efficacy remains unexamined. This systematic review and network meta-analysis examined the comparative efficacy of NIBS interventions on SI across psychiatric conditions and the lifespan. METHODS: The authors searched Embase, MEDLINE, PsycINFO, CINAHL, and the Cochrane CENTRAL Database from inception to March 2023 for randomized controlled trials (RCTs) with at least one NIBS arm and an outcome from the suicide item of a standardized measure or a suicide-specific measure. Studies that focused solely on invasive stimulation or excluded participants with SI were ineligible. Network meta-analyses examined the comparative efficacy of NIBS for treating SI, using standardized mean differences (Hedges' g). Analyses were first restricted to RCTs that recruited participants with SI at baseline (network 1), then were conducted on all identified RCTs (network 2), and, finally, were conducted to examine NIBS plus pharmacotherapy initiation (network 3). RESULTS: Seventy-five studies met inclusion criteria for the systematic review, and 58 were included in meta-analyses. Studies in network 1 (three RCTs) used accelerated transcranial magnetic stimulation (TMS) over the left dorsolateral prefrontal cortex (DLPFC); no difference from sham TMS was found. In network 2 (25 RCTs), bitemporal electroconvulsive therapy (ECT) was favorable compared to sham stimulation for reducing SI, and all other NIBS interventions showed no benefit over sham NIBS. In network 3 (five RCTs), left DLPFC repetitive TMS plus escitalopram significantly reduced SI compared to sham TMS plus a selective serotonin reuptake inhibitor (SSRI). CONCLUSIONS: Although the findings were limited by small numbers of studies in networks 1 and 3, large heterogeneity in study design and SI outcomes in network 2, and a focus mostly on treating patients with depression in networks 2 and 3, they support bitemporal ECT and high-frequency repetitive TMS combined with SSRIs as effective interventions for improving SI, and nonconvulsive NIBS without pharmacotherapy shows no benefit over sham NIBS.