Matcha: six small trials and mostly null results

The longest matcha trial gave 99 older adults with early cognitive decline 2 g a day in capsules for 12 months, and their main memory and daily-living scores did not change significantly against placebo. One secondary score, reading emotions in faces, improved by 1.39 points, but its 95% interval ran from minus 0.002 to 2.78, so it just crosses zero. The matcha came from the company that funded it. The other trials are shorter and smaller. Adults aged 27 to 64 with trouble sleeping took 2.7 g a day for four weeks, and sleep measured by EEG did not differ from placebo. In 30 young women a single 3 g dose did not raise brown fat activity overall. In 61 older adults taking 3 g a day for 12 weeks, a cognitive score rose only in women, found in a split by sex, and the abstract gives no size for it. In 36 untrained young men, with 8 to 10 per group, a matcha drink during 8 to 12 weeks of weight training lowered fatigue after exercise at week one. Muscle mass grew more in one of the two trials, and strength only trended up.

How much is actually in it

• Dry matcha powder analyzed from three packages contained 56.1 g fiber, 17.3 g protein and 7.3 g fat per 3.5 oz (100 g), and about 2.2 mg caffeine per gram. — organic Japanese matcha (Uji, Kyoto), three production runs (n = 3 packages) Evidence D source
  Fiber 56.1 g/100 g (insoluble 52.8), protein 17.3 g/100 g, fat 7.285 g/100 g, caffeine 2213 ug/g

What a real portion looks like

• A 2 g daily serving of the trial matcha contained 170.8 mg catechins, 48.1 mg theanine and 66.2 mg caffeine. — one commercial matcha product (ITO EN), analyzed for an RCT Evidence D source
  Per 2 g: 170.8 mg catechins (105.3 mg EGCG), 48.1 mg theanine, 66.2 mg caffeine
• One cup made with 1.5 g of matcha powder contained 45 mg caffeine, 94.5 mg EGCG and 27 mg theanine, plus vitamin K in an amount whose reported unit looks like a misprint. — one commercial matcha beverage (Nestle Japan), analyzed for an RCT Evidence D source
  Per 1.5 g cup: 45 mg caffeine, 94.5 mg EGCG, 27 mg theanine, 488 mg fiber, vitamin K printed as 46.4 mg, unit implausible

What it does to your health

• In a 12-month RCT of 99 older adults with cognitive decline, 2 g of matcha a day did not change the primary cognitive scores (MoCA, ADL) and improved only a facial-emotion perception score. — 99 older adults with subjective cognitive decline or mild cognitive impairment, 12 months (n = 99) Evidence B source
  Social acuity score difference -1.39 (95% CI -2.78 to 0.002), P = 0.028; primary outcomes not significant
• In two small RCTs on 36 untrained young men, a matcha drink twice daily during 8 to 12 weeks of resistance training lowered subjective fatigue after exercise at week 1, and in the 12-week trial gave a larger muscle mass gain than placebo. — 36 healthy untrained young men, 8-week (n=17) and 12-week (n=19) trials (n = 36) Evidence B source
  Leg strength gain trend only; larger skeletal muscle mass change in trial 2; lower salivary cortisol
• In a 12-week RCT of 61 community-dwelling older adults taking 3 g of matcha a day, MoCA scores improved only in women in a sex-specific analysis. — 61 community-dwelling elderly adults, 12 weeks, 3 g matcha powder per day (n = 61) Evidence B source
  Significant MoCA gain in women only (subgroup); no overall effect reported
• In a 4-week RCT of healthy adults aged 27 to 64 taking 2.7 g of matcha a day (71.5 mg caffeine), EEG-measured sleep time, latency and efficiency did not differ from placebo. — healthy Japanese men and women aged 27-64 with difficulty sleeping, 4 weeks Evidence B source
  No significant difference in EEG sleep parameters; subjective sleep satisfaction trend p < 0.1
• In a crossover RCT of 30 young women, a single 3 g dose of matcha did not raise cold-induced brown fat activity overall, with a difference only in the lowest-activity third. — 30 healthy young women, single dose, crossover (n = 30) Evidence B source
  No overall difference vs placebo; larger difference in low-BAT-activity tertile (post hoc stratification)

Safety, limits and interactions

• Over 12 months of 2 g matcha a day in 99 older adults, investigators found no adverse event causally linked to matcha. — 99 older adults aged 60-85 with cognitive decline, 12 months (n = 99) Evidence B source
  0 adverse events judged causally related; 8 dropouts for consent withdrawal or illness
• EFSA judged green tea catechins from traditional infusions generally safe, but found that 800 mg EGCG a day or more from supplements raised liver enzymes in trials. — adults in intervention trials of green tea extracts Evidence E source
  Threshold >= 800 mg EGCG/day for serum transaminase rise; mean EU intake from infusions 90-300 mg/day

What people believe that the data does not support

• The EU refused a health claim that green tea catechins cut carbohydrate absorption by 25% and slim by 12%. — — Evidence E source
  Claim status: non-authorized

What is still unknown

• EFSA noted that none of the green tea intervention studies it reviewed included pregnant women, breastfed infants or children. — pregnant women, breastfed infants, children Evidence E source
  No intervention data

The bottom line

What is in a cup depends heavily on the product. A 2 g daily serving of one trial matcha held 170.8 mg of catechins, 105.3 mg of them EGCG, plus 66.2 mg of caffeine and 48.1 mg of theanine. A 1.5 g cup of another commercial product held 45 mg of caffeine and 94.5 mg of EGCG. An organic brand from Kyoto, analyzed across three packages, came in at about 2.2 mg of caffeine per gram, far below the 33 mg per gram of the trial product. Dry matcha is 56.1 g fiber per 3.5 oz (100 g) in that analysis, though a real cup delivers under half a gram. Neither USDA nor the UK table lists matcha. The EU refused a claim that green tea catechins cut carbohydrate absorption by 25% and slim you by 12%, meaning the evidence was judged insufficient when assessed. On safety, the 12-month trial found no adverse event causally linked to 2 g a day, though safety was not what it was designed to measure and the sample was small. EFSA judged catechins from traditionally brewed green tea generally safe and noted rare, most probably idiosyncratic cases of liver injury after drinking it. In trials of supplements, 800 mg of EGCG a day or more raised liver enzymes. That opinion covers green tea in general, and a matcha serving carries more EGCG than an ordinary infusion. At 94.5 to 105.3 mg of EGCG per serving in the two analyzed products, 800 mg is about eight servings, or roughly 13 to 15 g of powder a day, and EGCG varies by product. None of the intervention studies EFSA reviewed included pregnant women, breastfed infants or children. No trial data does not mean it is safe for them. A 1.5 g cup also carries vitamin K, though the paper's unit looks like a misprint, so we cannot say how much. We have no cards on caffeine limits or on interactions with medicines, so if you are pregnant or take any regular prescription, ask your doctor or a pharmacist before making matcha a daily habit.

Sources

1. mtch-daily-r3-09 · composition analysis
   The total fibre content of matcha was 56.1 g/100 g, with 52.8 g/100 g (94.1% of total fibre) of insoluble dietary fibre and 3.3 g/100 g of soluble fibre (5.9% of total fibre). The total protein content was 17.3 g/100 g. The total fat content in dry matcha was 7.285 g/100 g, comprising varying proportions of individual fatty acids, the highest ones being those of linoleic acid and α-linolenic acid. The caffeine content of matcha tea was 2213.492 µg/g.
   Foods, 2024 · checked 2026-09-29
2. mtch-daily-r3-03 · composition analysis
   The matcha used, “Hojin no shiro” (provided by ITO EN, LTD., Tokyo, Japan) contained 170.8 mg of catechin, 48.1 mg of theanine, and 66.2 mg of caffeine per daily serving.
   PLoS One, 2024 · checked 2026-09-29
3. mtch-daily-r3-05 · composition analysis
   The matcha group consumed a beverage containing 1.5 g matcha green tea powder [contents/cup; 0.06 g water, 0.4 g protein, 0.2 g fat, 4.4 g carbohydrate, 166.5 mg total polyphenols, 94.5 mg epigallocatechin gallate (EGCG), 1.1 mg lutein, 46.4 mg vitamin K, 27 mg theanine, 488 mg dietary fiber, and 45 mg caffeine]
   Nutr J, 2023 · checked 2026-09-29
4. mtch-daily-r3-01 · RCT
   Matcha consumption led to significant improvements in social acuity score (difference; -1.39, 95% confidence interval; -2.78, 0.002) (P = 0.028) as evaluated by the perception of facial emotions in cognitive function. The primary outcomes, that is, Montreal Cognitive Assessment and Alzheimer's Disease Cooperative Study Activity of Daily Living scores, showed no significant changes with matcha intervention.
   PLoS One, 2024 · checked 2026-09-29
5. mtch-daily-r3-04 · RCT
   In the matcha group, subjective fatigue after exercise at 1 week of training was lower than that in the placebo group. Gut microbe analysis showed that the abundance of five genera changed after matcha intake. The change in Ruminococcus, Butyricimonas, and Oscillospira compositions positively correlated with the change in maximum strength. In trial 2, the change in skeletal muscle mass in response to training was larger in the matcha group.
   Nutr J, 2023 · checked 2026-09-29
6. mtch-daily-r3-07 · RCT
   Sixty-one participants were recruited and randomly assigned to receive test drink containing 3g powder from fresh Matcha or placebo powder per day. Changes in cognitive function were assessed utilizing a psychometric test battery. Daily food intake was assessed by a Brief-type Self-administered Diet History Questionnaire (BDHQ). In the gender-specific analysis, a significant cognitive enhancement was observed in the Montreal Cognitive Assessment (MoCA) score in the active group of women.
   Nutrients, 2020 · checked 2026-09-29
7. mtch-daily-r3-08 · RCT
   After 4 weeks of consuming 2.7 g of matcha daily (containing 50.3 mg theanine, 301.4 mg catechins, and 71.5 mg caffeine), no significant differences were observed between the control and matcha groups on total sleep time, sleep latency, wake after sleep onset, or sleep efficiency measured by electroencephalography (EEG).
   Nutrients, 2024 · checked 2026-09-29
8. mtch-daily-r3-13 · RCT
   Healthy young women (n = 30) were enrolled in a randomized single-blind crossover trial. Participants consumed 3 g of matcha powder or a placebo and then received the alternate condition in a subsequent trial. BAT activity was analyzed using thermography with cold exposure. A stratified analysis was performed based on tertiles of maximal changes in BAT activity. BAT activity was increased for all participants following cold exposure; however, no significant difference in the increased levels was found between the matcha and placebo trials.
   Physiol Rep, 2026 · checked 2026-09-29
9. mtch-daily-r3-02 · RCT
   There were no cases of a causal relationship between harmful events and matcha consumption.
   PLoS One, 2024 · checked 2026-09-29
10. mtch-daily-r3-10 · agency opinion
    The Panel concluded that catechins from green tea infusion, prepared in a traditional way, and reconstituted drinks with an equivalent composition to traditional green tea infusions, are in general considered to be safe according to the presumption of safety approach provided the intake corresponds to reported intakes in European Member States. However, rare cases of liver injury have been reported after consumption of green tea infusions, most probably due to an idiosyncratic reaction. Based on the available data on the potential adverse effects of green tea catechins on the liver, the Panel concluded that there is evidence from interventional clinical trials that intake of doses equal or above 800 mg EGCG/day taken as a food supplement has been shown to induce a statistically significant increase of serum transaminases in treated subjects compared to control.
    EFSA Journal, 2018 · checked 2026-09-29
11. mtch-daily-r3-12 · EU claim register
    POL-HC-7101 Flavonoids from green tea, apple and onion Flavonoids, especially catechins from green tea, reduce the absorption of carbohydrates by 25%. Carbohydrates account for 49% of total energy. Thus the weight slimming effect is a total of 12%. Reduces visceral fat Non-authorised
    EU Register on nutrition and health claims, claim POL-HC-7101, snapshot 2026-09-21 · checked 2026-09-29
12. mtch-daily-r3-11 · agency opinion
    None of the intervention studies addressed pregnant women, breast‐fed infants or children.
    EFSA Journal, 2018 · checked 2026-09-29