The efficacy of lithium in the treatment of suicidal ideation, behavior and suicide: An updated systematic review of method analysis of randomized controlled trials

J Affect Disord. 2025 May 27:119487. doi: 10.1016/j.jad.2025.119487. Online ahead of print.

ABSTRACT

OBJECTIVES: Lithium is a first-line treatment for bipolar disorder, especially BD-I. In addition to efficacy in treating acute mania and recurrence prevention in bipolar disorder, lithium’s anti-suicide effects have been documented in studies showing reduced rates of completed suicide and serious attempts. To update current knowledge on the topic, we conducted a systematic review and meta-analysis of lithium’s impact in randomized controlled trials on suicidality (i.e., suicidal ideation, attempts, and suicide).

METHODS: Herein we followed Cochrane protocols and PRISMA guidelines to search OVID databases (Embase, MedLine, PsychINFO) from January 2013 to July 2024 for randomized trials evaluating lithium’s effect on suicidality, specifically reporting suicidal ideation, attempts, or related mortality outcomes. We identified earlier studies from manual reference searching. Two reviewers independently extracted data and assessed quality.

RESULTS: From 1793 articles, 15 studies (eight placebo-controlled, n = 1698; seven open-label, n = 1338) met eligibility criteria. Lithium was associated with nonsignificant reductions in suicide attempts (OR = 0.73; 95 % CI = [0.41, 1.31]; 25 in the lithium cohort vs. 63 on placebo) and completed suicide (OR = 0.61; 95 % CI = [0.25, 1.48]; lithium 4 vs. 13 placebo). Heterogeneity in methods and measures used to quantify suicidal ideation prevented quantitative analysis.

CONCLUSIONS: Methodological limitations affect assay sensitivity for suicidality measures, notably small sample sizes, diagnostic heterogeneity, inadequate treatment fidelity, subtherapeutic lithium levels, and intersubject differences in prior suicidality. Failure to reach statistical significance likely results from type II error, yet lithium’s beneficial effects on suicide attempts and completed suicide align with observational studies.

PMID:40441661 | DOI:10.1016/j.jad.2025.119487