Eckert · The British journal of nutrition 2025 · systematic review and meta-analysis · n=11 trials (1,093 participants)

Creatine supplementation for treating symptoms of depression: a systematic review and meta-analysis.

Cited 0 times in the scientific literature.

Level 1 - systematic review of randomized trials

Systematic review and meta-analysis of randomized trials

PubMed 41189312 · doi:10.1017/S0007114525105588 · record verified 2026-08-29

What was done

Systematic review and random-effects meta-analysis (with Hartung-Knapp-Sidik-Jonkman adjustment) of randomized trials searched across four databases up to February 2025. The authors compared creatine supplementation to placebo in individuals with or without depression. Risk of bias was evaluated via RoB 2, and certainty was assessed with GRADE.

What was found

Across 11 trials (1,093 participants), creatine produced a standardized mean difference in depressive symptoms of -0.34 (95% CI -0.68 to -0.00; GRADE: very low quality of evidence), equating to a 2.2-point reduction on the 17-item Hamilton Depression Rating Scale (below the minimal important difference of 3.0 points). Heterogeneity was substantial (I² = 71.3%), and trim-and-fill adjustments indicated substantial bias favoring creatine. Remission was significantly higher with creatine across 3 trials (OR 3.60; 95% CI 1.76 to 7.56), but treatment response was not significantly different across 2 trials (OR 0.72; 95% CI 0.28 to 1.88).

Why it matters

This review provides the first systematic synthesis of creatine for depressive symptoms, indicating that while small benefits may exist, the average effect is not clinically meaningful and may be an artifact of study bias.

Limits

The certainty of evidence was graded as very low due to substantial heterogeneity, wide confidence intervals including non-clinically important effects, and bias favoring creatine in subgroup/trim-and-fill analyses. Secondary outcomes were limited by very few reporting trials (three for remission, two for response).

Cited by