The Effect of Creatine Monohydrate on Mental Disorders: A Systematic Review of Randomized Controlled Trials: Effet du monohydrate de créatine sur les troubles mentaux: examen systématique des essais contrôlés à répartition aléatoire

Abstract

Objective: The objective of this systematic review is to synthesize and evaluate the evidence involving creatine monohydrate supplementation (CrM) across mental disorders. Methods: MEDLINE, Embase, Cochrane, and PsycINFO were searched up to 09/30/2025 for randomized controlled trials (RCTs) investigating the effect of CrM on psychiatric symptoms and safety in participants with a mental disorder. Risk of bias was assessed. Results: Six articles from five RCTs were included (CrM: n = 126, placebo: n = 112; mean age=36 ± 14 years; male sex = 26%). Four RCTs reported on major depressive disorder (MDD), one bipolar depression. No other mental disorders were investigated. Two RCTs were low risk of bias and three had some concerns. CrM dosing ranged from 2 to 10 g/day for 4–8 weeks as adjunct treatment. In the treatment of MDD, CrM was tested as combination with escitalopram (k = 1, outperforming selective serotonin reuptake inhibitor (SSRI) + placebo; Cohen's d = 1.13 at 8 weeks), pharmacotherapy augmentation in adults (k = 1) and female adolescents (k = 1, no difference vs placebo), psychotherapy augmentation (k = 1, cognitive behavioural therapy (CBT) + CrM outperforming CBT + placebo) in MDD, and as pharmacotherapy augmentation in bipolar depression (k = 1, no difference vs placebo augmentation). Two trials in MDD found a correlation between CrM brain N-acetylaspartate and phosphocreatine, which was associated with larger improvement. CrM was generally well-tolerated. Two CrM out of 17 participants experienced hypomania/mania. Conclusion: CrM shows promise as a combination treatment with SSRIs or for augmenting psychotherapy in MDD in adults. Double-blind, large-scale RCTs investigating the efficacy of CrM, with and without first-line therapies, are needed across mental disorders.