⚠️ Auto-extracted by Claude on 2026-07-11 from the publisher abstract, figures, and supplementary material; the subscription-only full article was not available for end-to-end verification (DOI 10.1038/s41380-026-03727-9). Quantitative values below were cross-checked against those publisher materials, but the page remains unverified until the full article is read. gap/no-fulltext-access
Cognitive and affective effects of L-Theanine
TL;DR
This systematic review and meta-analysis synthesized 31 placebo-controlled oral L-theanine randomized trials (n=1,168) in healthy and clinical populations. The clearest result was a short-term improvement in choice reaction time, while acute-stress evidence was small and bias-sensitive, fatigue was null, anxiety findings were mostly null, and the depressive-symptom signal appeared only after excluding an outlying study. These data support limited evidence for acute attention, not prevention or treatment of age-related cognitive decline.
Design
| Field | Value |
|---|---|
| Included studies | 31 randomized, placebo-controlled trials: 17 crossover and 14 parallel-group |
| Participants | 1,168 total; healthy adults and clinical psychiatric populations |
| Intervention | Oral L-theanine, approximately 50-900 mg/day |
| Duration | Single-dose studies through 12 weeks |
| Primary outcome | Acute effect of a single dose on stress in healthy adults |
| Secondary outcomes | Repeated-dose stress, anxiety, depressive symptoms, fatigue, reaction time, attention, and dropout-based safety |
| Risk-of-bias method | Cochrane RoB 2 |
The review classified 13 trials as low risk of bias, 11 as having some concerns, and 7 as high risk. Endpoint-specific pools were much smaller than the overall review population.
Key results
| Outcome | Result | Interpretation |
|---|---|---|
| Acute choice reaction time | SMD 0.51 (95% CI 0.25-0.77); I2=3%; 7 comparisons, 151 participants | Short-term attention/psychomotor signal. The pool included 100.6 mg, 200 mg, 250 mg, and 6 mg/kg arms, so it should not be read as a uniform 200 mg intervention. |
| Acute stress in healthy adults | SMD 0.312 (95% CI 0.001-0.623); I2=25%; 7 comparisons, 120 participants | Small, borderline result; 4 of 7 contributing trials were high risk of bias. |
| Acute depressive symptoms | Full analysis: SMD 0.35 (95% CI -0.40 to 1.10), I2=77%. Leave-one-out analysis: SMD 0.69 (0.13-1.25), I2=30%; 3 comparisons, 52 participants | The positive result is sensitivity-only and depends on excluding Haskell et al. 2008. |
| Fatigue | SMD 0.30 (95% CI -0.26 to 0.86); I2=55% | No significant acute effect. |
| Anxiety | Generally inconsistent and non-significant | The highlighted psychotic-population estimate came from one 8-week adjunctive trial at 400 mg/day: SMD 0.54 (95% CI -0.01 to 1.10), whose CI crosses zero. |
| Dropout | RR 1.04 (95% CI 0.84-1.29); 507 participants per condition | No detected difference between L-theanine and placebo. |
| Serious adverse events | None reported | Reassuring for the studied durations, but not evidence for rare-event or long-term safety. |
Aging relevance
Only one included trial specifically enrolled middle-aged and older adults (n=52), and the review did not report an age-stratified pooled effect. Choice reaction time measured within an hour of dosing is not equivalent to durable cognitive improvement, prevention of mild cognitive impairment, or modification of neurodegeneration. This paper therefore does not establish an aging-hallmark effect or a cognitive-aging intervention.
| Dimension | Status | Notes |
|---|---|---|
| Human evidence? | yes | All included studies were human randomized trials. |
| Aging population directly tested? | limited | One small middle-aged/older trial; no age-stratified pooled estimate. |
| Durable aging phenotype tested? | no | Outcomes were neurobehavioral or psychiatric, mostly acute or short-term. |
| Replicated in aging-specific trials? | no | No adequately powered trial of age-related cognitive decline was identified. |
Limitations
- Most endpoint-specific pools contained few small trials despite the aggregate n=1,168.
- Doses, durations, populations, outcome scales, adjunctive treatments, and crossover/parallel designs varied.
- The acute-stress estimate was disproportionately supported by high-risk-of-bias studies.
- The depressive-symptom signal was not significant in the complete analysis and appeared only in a leave-one-out sensitivity analysis.
- Safety follow-up was too short and sparse to assess uncommon adverse events or long-term use.
- Guillaume Fond disclosed founding a company that sells dietary supplements; the review reported no specific grant funding.
- The full article was subscription-only at ingestion, so preregistration status and narrative-method details could not be verified end to end. gap/no-fulltext-access