Tier 2 — strong

Effects of Arginine Supplementation on Athletic Performance Based on Energy Metabolism: A Systematic Review and Meta-Analysis

Viribay A, Burgos J, Fernández-Landa J, Seco-Calvo J, Mielgo-Ayuso J

Nutrients 2020 12(5):1300

Bibliography

DOI: 10.3390/nu12051300
PubMed: PMID 32370176
PubMed Central: PMC7282262
Funding: External funding: none declared
Competing interests: The authors declare no conflict of interest

Study snapshot

Design	Systematic review and meta-analysis (PRISMA + PICOS)
Model	Human — mostly trained or elite adult athletes (282 of 394 total participants were trained or elite)
Sample	394 participants across 18 studies systematically reviewed; 15 studies included in the quantitative meta-analysis
Intervention	Arginine supplementation (multiple forms — L-arginine, arginine aspartate, and arginine alpha-ketoglutarate/AAKG pooled) vs placebo. Acute protocols 30 minutes to 2 days pre-exercise; chronic protocols 4 to 8 weeks
Duration	Trials ranged from single acute doses to 8-week chronic supplementation
Endpoints	Primary: Aerobic performance (≤VO₂max tests — Harvard Step Test, incremental to exhaustion, 16.1 km time trial, cycloergometer endurance); Primary: Anaerobic performance (>VO₂max tests — Wingate Test, RSAT, 1 km time trial, 1RM strength, isokinetic flexion and extension); Secondary: Effective acute dose and timing; Secondary: Effective chronic dose and duration

What the study showed, in plain terms

This is a 2020 meta-analysis published in the journal Nutrients that pooled the results of 15 randomised trials testing whether arginine supplementation improved athletic performance. The researchers looked separately at aerobic performance (endurance tests) and anaerobic performance (short high-intensity efforts).

The pooled results were statistically positive for both. The authors concluded arginine improved aerobic performance with a large effect size and anaerobic performance with a small but significant effect. Their practical recommendations were 0.15 g of arginine per kg body weight taken 60 to 90 minutes before exercise for acute effect, or 1.5 to 2 g per day for 4 to 7 weeks for aerobic performance, or 10 to 12 g per day for 8 weeks for anaerobic performance.

The paper matters to the Ca-AKG conversation because the 15 studies pooled included arginine in multiple forms — plain L-arginine, arginine aspartate, and arginine alpha-ketoglutarate (AAKG). It is the most-cited meta-analysis in the ergogenic-arginine literature, and both the AAKG marketing case and the counter-case cite it.

The results should be read with care. Heterogeneity between studies was very high for the aerobic outcome (I²=89%), and the authors' own funnel plot analysis showed statistically significant publication bias for aerobic performance data. In 2025, a formal published comment by Eric Trexler in the same journal documented methodological errors — including outlier handling and confusion between standard errors and standard deviations — and Viribay's group published a reply acknowledging several of these criticisms. Subsequent narrative reviews and the shift in sports supplement formulations from arginine-based to citrulline-based have both moved in the direction of arginine's ergogenic effect being smaller and less reliable than this meta-analysis alone suggested.

Key findings

Fifteen randomised placebo-controlled trials of arginine supplementation in humans were pooled quantitatively out of 18 identified in the systematic review.

Aerobic performance (≤VO₂max): Hedges' g = 0.84 (95% CI 0.12 to 1.56), p = 0.02. The authors classify this as a large effect. Heterogeneity was I² = 89% — a large amount of inconsistency between studies. Egger's test for publication bias was statistically significant (z = 2.873, p < 0.05).
Anaerobic performance (>VO₂max): Hedges' g = 0.24 (95% CI 0.05 to 0.43), p = 0.01. The authors classify this as a small effect. Heterogeneity was I² = 0% — no meaningful inconsistency.
Effective acute dose: 0.15 g per kg body weight (roughly 10 to 11 g for a 70 kg adult) taken 60 to 90 minutes before exercise.
Effective chronic dose: 1.5 to 2 g per day for 4 to 7 weeks for aerobic performance; 10 to 12 g per day for 8 weeks for anaerobic performance.
Formulations pooled: plain L-arginine, arginine aspartate, and arginine alpha-ketoglutarate (AAKG). The meta-analysis did not separate results by formulation.
Participants: 394 total (386 male, 8 female). 282 were classified as trained or elite athletes.

What this study can and cannot tell us

Very high heterogeneity for the primary aerobic result. I² of 89% means the 15 pooled studies disagreed strongly with each other. A pooled effect size drawn from such heterogeneous data is a summary statistic of debatable interpretive value — the underlying trials did not converge on a common effect.

Statistically significant publication bias for aerobic outcomes. The authors' own funnel plot analysis showed asymmetry (Egger's test z = 2.873, p < 0.05), meaning small negative trials are likely missing from the literature. This tends to inflate the pooled effect estimate upward.

Post-publication methodological criticism. A 2025 formal comment by Eric Trexler in Nutrients (doi:10.3390/nu17142375) documented multiple methodological errors, including confusion between standard errors and standard deviations, inadequate handling of outliers, and other issues common in sports nutrition meta-analyses. Viribay et al. published a formal reply (doi:10.3390/nu17142376) acknowledging several of these criticisms and reporting that revised analyses were underway.

Pooled formulations. The 15 studies used L-arginine, arginine aspartate, and AAKG interchangeably. Effects specific to AAKG cannot be extracted from the aggregate result — the paper's conclusion applies to "arginine" as a category, not to any specific salt.

Severe sex imbalance. 386 male participants versus 8 female. Results cannot reasonably be generalised to women.

Population selection. Predominantly trained or elite athletes (282 of 394). Applicability to recreational or untrained populations is unclear.

Placebo-response inflation. Several included trials were single-dose acute studies where placebo response tends to be large, further complicating interpretation of small pooled effects.