Lemongrass

What the trials found

Goes 2015: 40 healthy men, randomly placed in four groups. They breathed lemongrass oil (3 or 6 drops), tea tree oil (control smell) or water, then did a stressful colour-word test on video. State anxiety and tension were lower right after the lemongrass aroma, and the lemongrass groups recovered from the test in 5 minutes, unlike controls. Heart rate and muscle tension changes during the test were not prevented, the same as seen with diazepam in earlier work.

Results

Positive on self-rated anxiety and tension; no protection of body measures (heart rate, muscle activity).

Effect sizes as reported

Statistically significant versus tea tree and water on state anxiety and subjective tension; numbers not given in the abstract.

Why this rating

One trial of 40 young men, single short exposure, anxiety induced in the lab, no replication found. Self-rated scores are easily affected by smell and expectation.

Form and dose studied

Lemongrass (Cymbopogon citratus) essential oil, 3 or 6 drops inhaled once.

Who was studied

Healthy adult men (about 10 per group).

Limits

Healthy volunteers, not people with an anxiety disorder; single exposure; the abstract gives no numbers; no long-term data.

Why researchers looked

Lemongrass tea is used in Brazilian folk medicine for nervous complaints, and the oil is popular in aromatherapy, so researchers tested the aroma directly.

Traditional use

Brazilian folk medicine uses a lemongrass leaf tea (called an abafado) for nervous and stomach complaints, according to a 1986 paper.

Proposed mechanism Animal research

Not worked out in people. In mice, lemongrass oil effects were blocked by flumazenil, which points to the brain's GABA calming system, but that was with injected oil.

A proposed mechanism does not show that the effect happens in people.

Safety is covered in the safety section. Lemongrass essential oil is a strong oil: it can cause skin allergy, and it should not be swallowed on the strength of these results.

¹⁶

What the trials found

Maybodi 2025: 38 patients having scaling and root planing (a deep gum clean). One group inhaled lemongrass oil during the work; the control group had the work with no aroma. The lemongrass group had lower systolic and diastolic blood pressure, heart rate and anxiety scores. Oxygen saturation, pain and satisfaction did not differ. Registered as IRCT20230615058492N2.

Results

Better on anxiety, blood pressure and heart rate after one session; no difference in pain or satisfaction.

Effect sizes as reported

Significant reductions reported for anxiety, systolic and diastolic blood pressure and heart rate; numbers not given in the abstract.

Why this rating

Very small (38), one session, no placebo aroma so people could tell their group, and anxiety is self-rated. Authors say lemongrass 'appears beneficial'.

Form and dose studied

Lemongrass essential oil aroma during one treatment session; the oil and amount are not given in the abstract.

Who was studied

Adults with gum disease treated at a university dental clinic in Yazd, Iran.

Limits

No blinding to aroma, single centre, short effect; the blood pressure change is a response to a nervous situation, not treatment of high blood pressure.

Why researchers looked

Dental visits cause worry that raises blood pressure and heart rate. Aromatherapy is a low-cost add-on researchers have tried.

Traditional use

No traditional use of lemongrass in the dental chair; aromatherapy use is modern.

Proposed mechanism Pharmacological theory

Not known. Possible effects of the smell on the brain's stress response are theory only.

A proposed mechanism does not show that the effect happens in people.

Safety is covered in the safety section. Lemongrass essential oil is a strong oil: it can cause skin allergy, and it should not be swallowed on the strength of these results.

²¹

What the trials found

Chawla 2025: 350 children aged 6 to 9, 70 in each of five groups: cards, lemongrass aroma, music, acupressure, honey taste. Every group had lower anxiety after treatment (P < 0.001). The abstract says music was best (mean change 1.10), visual next (1.04), and that smell and taste were less effective than those two. No lemongrass-specific number is given.

Results

Anxiety lower in the lemongrass aroma group, but less than with music or visual distraction; no figure in the abstract.

Effect sizes as reported

All groups P < 0.001 before versus after; lemongrass group number not in the abstract.

Why this rating

The abstract does not show a no-treatment group or a lemongrass figure, so it cannot show that the lemongrass smell, rather than the dental setting or the passing of time, lowered anxiety.

Form and dose studied

Lemongrass oil aromatherapy during one dental visit; amount not in the abstract.

Who was studied

Children aged 6 to 9 in India.

Limits

No lemongrass result in the abstract, no untreated comparison described, children's anxiety scale.

Why researchers looked

Fear of the dentist is common in children, and non-drug distraction methods are tried first.

Proposed mechanism Pharmacological theory

Not known; distraction and smell effects are theory.

A proposed mechanism does not show that the effect happens in people.

Essential oils should not be swallowed by children. This trial used the aroma only. Poison-centre guidance on children was not read here.

⁷

What the trials found

Leite 1986: in 50 volunteers, lemongrass tea and a placebo were swallowed under double-blind conditions. Sleep induction, sleep quality, dream recall and waking were no different from placebo. In 18 people with high trait anxiety given an anxiety-inducing test after tea or placebo, anxiety levels were similar. The paper's authors concluded the tea 'lacks hypnotic or anxiolytic properties'. The companion rat and mouse study found no oral calming, sleep or anti-seizure effects either (Carlini 1986).

Results

Null for sleep and for anxiety in a small double-blind study; also null by mouth in animals.

Effect sizes as reported

No difference from placebo; numbers not given in the abstract.

Why this rating

Only one small human study (50 sleep volunteers, 18 anxious volunteers) and the abstract has no numbers, but it was double blind and found no effect.

Form and dose studied

Lemongrass leaf tea (abafado), dose not given in the abstract.

Who was studied

Healthy volunteers (sleep) and people with high trait anxiety (anxiety test), Brazil.

Limits

1986 study, small, single dose or short use, abstract only, no numbers. Does not test the oil or long-term use.

Why researchers looked

Lemongrass tea (abafado) is among the best-known Brazilian folk teas for calming the nerves, so researchers tested it.

Traditional use

Folk medicine in Brazil uses the leaf tea for nervous and stomach disorders and for fever (Carlini 1986).

Proposed mechanism Animal research

Not applicable: tests found no effect.

A proposed mechanism does not show that the effect happens in people.

The same study found normal blood tests after the tea for up to 2 weeks (see the safety section).

^{20 4}

What the trials found

Blanco 2009: oral lemongrass oil at 0.5 or 1.0 g/kg in mice lengthened pentobarbital sleep, increased open-arm time on the maze and delayed seizures. Silva 2010: oil injected at 50 to 200 mg/kg delayed seizures and lengthened barbiturate sleep; effects were blocked by flumazenil. Carlini 1986: lemongrass tea by mouth had no such effects in rats and mice. No human study tests this.

Results

Positive in two mouse studies with oil; null for oral tea in rodents.

Effect sizes as reported

Significant effects in mice; numbers not in the abstracts.

Why this rating

Animal only, huge oil doses compared with aroma or food use, results conflict with the oral tea findings, and the human tea trial was null.

Form and dose studied

Essential oil, 0.5 to 1.0 g/kg by mouth (Blanco) or 50 to 200 mg/kg by injection (Silva) in mice.

Who was studied

Mice; not applicable to people.

Limits

Cannot be translated to people. Oil at these doses would be dangerous to swallow; do not copy these doses.

Why researchers looked

To explain the folk use of the tea for the nerves.

Traditional use

Folk medicine uses lemongrass tea for nervous disorders (Brazil).

Proposed mechanism Animal research

Mouse work links the oil effect to the GABA system (the brain's main calming system), because flumazenil, a drug that blocks it, blocked the effect (Silva 2010, injected oil). Low doses of oil added to low doses of diazepam made the effect stronger.

A proposed mechanism does not show that the effect happens in people.

The animal results suggest that lemongrass oil could add to sedative medicines (see the interactions in the safety section).

^{1 31 4}

What the trials found

Chaisripipat 2015: 30 volunteers with moderate dandruff in a double-blind, placebo-controlled split-head test applied tonics twice a day. Dandruff scores fell 33%, 75% and 51% at day 7 and 52%, 81% and 74% at day 14 with 5%, 10% and 15% oil. No details on side effects are in the abstract.

Results

Large, rapid reductions in dandruff score in a small trial.

Effect sizes as reported

Reductions of 52%, 81% and 74% at day 14 (5%, 10%, 15% oil), P < 0.005; the placebo result is not given in the abstract.

Why this rating

A double-blind placebo-controlled design is a strength, but there are only 30 people, 2 weeks, no repeat, and the oil was from Cymbopogon flexuosus, not C. citratus.

Form and dose studied

Hair tonic with 5%, 10% or 15% lemongrass (C. flexuosus) essential oil, twice a day for 14 days.

Who was studied

Thai adults aged 20 to 60 with dandruff.

Limits

Different species, tiny trial, short, no follow-up, cosmetic-science team. The abstract does not report skin irritation.

Why researchers looked

Dandruff is linked to Malassezia yeasts, and lemongrass oil kills yeasts in the dish.

Traditional use

Not a documented traditional use in the sources read.

Proposed mechanism Laboratory research

Citral and other oil compounds kill or slow yeasts in laboratory tests; Malassezia was targeted in the trial's design.

A proposed mechanism does not show that the effect happens in people.

Lemongrass oil is a known skin allergen (see the safety section); do not put undiluted oil on skin.

⁶

What the trials found

Carmo 2013: Phase I in 20 healthy volunteers found no significant adverse events. In phase II, 47 people received the oil shampoo and cream and 29 received ketoconazole versions for 40 days. 30 (63.83%) on oil and 18 (62.07%) on ketoconazole finished. Mycological cure was 60% with the oil and over 80% with ketoconazole.

Results

Some antifungal effect, less than the standard drug; many dropped out.

Effect sizes as reported

60% cure (oil) versus over 80% (ketoconazole) at 40 days, both P < 0.05.

Why this rating

A randomised pilot against a standard drug with a measured cure rate, but small, about a third dropped out, no placebo, and the drug was better.

Form and dose studied

Lemongrass (C. citratus) essential oil at 1.25 microlitres/mL in shampoo (3 times a week) and cream (twice a day) for 40 days.

Who was studied

76 people with pityriasis versicolor, Brazil (plus 20 healthy volunteers for safety).

Limits

Pilot, high dropout, no placebo arm, culture-based cure.

Why researchers looked

Lemongrass oil is active against Malassezia yeasts in the dish.

Traditional use

Not a documented traditional use in the sources read.

Proposed mechanism Laboratory research

Oil compounds damage yeast cells in laboratory tests.

A proposed mechanism does not show that the effect happens in people.

Skin allergy to lemongrass oil is possible; see the safety section.

⁵

What the trials found

Satthanakul 2015: 20 healthy volunteers in a double-blind trial measured bad-breath gases with a Halimeter on day 0 and day 8. The rinse significantly reduced them on day 8 and appeared safe. Spiciness and taste were drawbacks.

Results

Positive, small, short.

Effect sizes as reported

Significant fall in volatile sulphur compounds on day 8; numbers not in the abstract.

Why this rating

Twenty healthy people, a pilot, 8 days, no numbers in the abstract.

Form and dose studied

Lemongrass oil mouth rinse; concentration not in the abstract.

Who was studied

Healthy adult volunteers, Thailand.

Limits

Pilot; healthy people; no long-term data.

Why researchers looked

Bad breath comes from mouth bacteria making sulphur gases, and lemongrass oil kills several of them in the dish.

Proposed mechanism Laboratory research

In the dish the oil was active against Aggregatibacter actinomycetemcomitans and Porphyromonas gingivalis and less so against Streptococcus mutans.

A proposed mechanism does not show that the effect happens in people.

Rinses should not be swallowed. Mouth and gum irritation was not reported in the abstract.

²⁹

What the trials found

Wasey 2023: 60 orthodontic patients with mild-to-moderate gingivitis, 20 per group: 0.25% lemongrass oil mouthwash, 0.2% chlorhexidine, or oral cleaning only, for 21 days. Plaque and gum scores were lower with lemongrass than chlorhexidine by day 14 and day 21 (P < 0.001). In another small trial of severe gum disease (Subha 2017, 15 per group, 3 months), pocket depth and attachment loss improved in the lemongrass group and in the chlorhexidine group compared with cleaning alone.

Results

Positive in two small trials from India; one reports lemongrass better than chlorhexidine.

Effect sizes as reported

P < 0.001 versus chlorhexidine for plaque and gum index (scores not in the abstract).

Why this rating

Groups of 15 to 20, short, one setting each, comparisons with a gold-standard rinse that came out unusually in favour of the herbal one, no independent repeats found.

Form and dose studied

Lemongrass oil 0.25% mouthwash, twice a day (21 days; Wasey) or 3 months (Subha).

Who was studied

Adults with gum inflammation or periodontitis, India.

Limits

Small, no numbers in the abstracts, authors' links to products not stated.

Why researchers looked

Chlorhexidine mouthwash stains teeth and tastes unpleasant, so a herbal alternative is of interest.

Proposed mechanism Laboratory research

Oil compounds kill plaque bacteria in the dish.

A proposed mechanism does not show that the effect happens in people.

Do not swallow mouthwash. Contact allergy to lemongrass oil can affect the mouth and lips, though not reported in these trials' abstracts.

^{37 34}

What the trials found

Wright 2009: 90 HIV-positive patients with oral thrush were randomly given gentian violet, lemon juice or lemongrass infusion for 11 days; 83 finished. In the intention-to-treat analysis none of the p-values were significant. In completers, lemon juice beat gentian violet (P < 0.02), and a difference between lemongrass and gentian violet was also reported (P < 0.05), with direction not stated in the abstract.

Results

Null in the main analysis; completers' result is unclear for lemongrass.

Effect sizes as reported

ITT: no significant differences; completers: lemongrass versus gentian violet P < 0.05, direction not given.

Why this rating

A null main analysis and an unclear secondary result in a small hospice trial.

Form and dose studied

Lemongrass infusion (grown and dried at the hospice) as a mouth treatment for 11 days; amount not in the abstract.

Who was studied

HIV-positive adults with oral thrush, South Africa.

Limits

Small, ITT null, abstract unclear on the direction for lemongrass.

Why researchers looked

Lemongrass infusion is a low-cost treatment the hospice already used.

Traditional use

The hospice grew and dried its own lemongrass and used the infusion routinely (the trial authors describe this).

Proposed mechanism Laboratory research

Antifungal lemongrass compounds act on Candida in the dish.

A proposed mechanism does not show that the effect happens in people.

Anyone with HIV and thrush needs medical care; this is not a substitute.

³⁸

What the trials found

Silva 2022 (narrative review) says lemongrass tea lowers blood pressure in healthy and high-blood-pressure people, but its abstract names no trials and we could not find primary human tea trials in PubMed. Maybodi 2025 saw lower systolic and diastolic pressure with lemongrass aroma during gum treatment (38 patients; see the mind file). Carbajal 1989: a leaf decoction lowered rat blood pressure when injected into a vein, with weak diuretic effect by mouth.

Results

Positive in animals and in a one-session aroma trial; no verified human tea or long-term trial.

Effect sizes as reported

Numbers are not given in the abstracts read.

Why this rating

No verified human treatment trial for high blood pressure; the review's human claim could not be traced.

Form and dose studied

Leaf decoction (rats); aroma during one dental session (people); tea claimed in the review.

Who was studied

Rats; dental patients; review population unclear.

Limits

Aroma-session result is a stress response. Review claim not verified. Animal results used intravenous doses.

Why researchers looked

Cuban folk medicine uses lemongrass as an antihypertensive.

Traditional use

Used in Cuba as an antihypertensive and anti-inflammatory folk medicine (Carbajal 1989).

Proposed mechanism Animal research

Lab work (summarised in a review) shows lemongrass and citral relax blood vessels through nitric oxide and calcium channels, and that citral slows the heart in animals.

A proposed mechanism does not show that the effect happens in people.

People taking blood-pressure medicine or with low blood pressure should be careful with medicinal amounts, because an extra drop in blood pressure is plausible (theory).

^{32 3 21}

What the trials found

Elson 1989: 22 people with total cholesterol about 315 mg/dl took 140 mg of lemongrass oil a day for 90 days. Across all 22 the change approached significance (P < 0.06). Eight responders fell by 25, 33 and 38 mg/dl at 30, 60 and 90 days; 14 non-responders changed by +2, +2 and -1 mg/dl. Six responders measured 90 days after stopping had returned toward baseline (+8 mg/dl). In Subha 2017 (45 people with severe gum disease) total cholesterol and LDL fell in the lemongrass mouthwash group compared with its own baseline, but these were not significantly different between groups. In Leite 1986, cholesterol, triglycerides and lipids did not change in healthy volunteers after tea. Rats: Junior 2024 reports less abnormal blood fat with lemongrass oil in diabetic rats.

Results

Unclear: no overall significant effect in the one dedicated study, and within-group changes in a mouthwash trial.

Effect sizes as reported

Responders -38 +/- 10 mg/dl at 90 days (P < 0.025); non-responders -1 +/- 6; whole group P < 0.06.

Why this rating

No control group, a result that missed significance overall, and splitting into responders afterwards. Healthy-volunteer tea data show no change.

Form and dose studied

Lemongrass oil capsule, 140 mg daily for 90 days (Elson); mouthwash 0.25% (Subha); tea (Leite).

Who was studied

People with high cholesterol; people with periodontitis; healthy volunteers.

Limits

Different products and populations, small, old.

Why researchers looked

A theory that compounds in lemongrass oil (geraniol and citral) could lower cholesterol by slowing its production.

Traditional use

Not a documented traditional use in the sources read.

Proposed mechanism Laboratory research

Oil compounds that touch the cholesterol-building pathway (mevalonate) were thought to lower cholesterol; citral also switches on PPAR-alpha and gamma in cells, which handle fats.

A proposed mechanism does not show that the effect happens in people.

Swallowed lemongrass oil capsules are not an approved use; the study did not report side effects in its abstract.

^{11 34 20 17 18}

What the trials found

Leite 1986: healthy volunteers had unchanged blood glucose after a single dose or 2 weeks of lemongrass tea. Junior 2024: C. flexuosus oil (32 or 64 mg/kg) and citral (32 mg/kg) for 14 days lessened the rise in glucose in rats with streptozotocin-induced type 1 diabetes. No trial in people with diabetes was found.

Results

No human evidence of benefit; positive in rats.

Effect sizes as reported

Numbers not given in the abstracts.

Why this rating

Only rat data; healthy volunteers show no change.

Form and dose studied

Oil by mouth in rats; tea in healthy people.

Who was studied

Rats; healthy adults.

Limits

No human diabetes trial; the rat study used a different species of lemongrass.

Why researchers looked

Folk and lab interest in lemongrass for diabetes; citral activates PPAR switches that affect sugar and fat handling in cells.

Traditional use

Not a documented traditional use in the sources read.

Proposed mechanism Laboratory research

Citral activated PPAR alpha and gamma in cell work (Katsukawa 2010).

A proposed mechanism does not show that the effect happens in people.

People using diabetes medicines should not add medicinal lemongrass expecting a sugar-lowering effect, and should tell their doctor if they use it in larger amounts (theory).

^{20 17 18}

What the trials found

Carlini 1986: lemongrass teas by mouth did not change gut transit in mice or defecation in rats. Venzon 2018: lemongrass oil 10 mg/kg reduced ethanol-induced ulcer area by 51.67% and sped healing of acetic-acid ulcers by 34.52% in mice; geraniol worked, citral did not heal established ulcers. Leite 1986 saw slight rises in amylase in some volunteers after tea, with no symptoms. We found no human trial for indigestion, nausea, cramps or bowel complaints.

Results

No human evidence; mixed in animals.

Effect sizes as reported

Ulcer area down 51.67% with 10 mg/kg oil in mice.

Why this rating

Animal and dish work only, and the older tea animal study was null.

Form and dose studied

Oil by mouth in mice; tea in rodents.

Who was studied

Mice, rats.

Limits

No human trial.

Why researchers looked

Brazilian folk medicine uses the tea for stomach and gut complaints.

Traditional use

Brazilian folk use for nervous and gastrointestinal disorders (Carlini 1986).

Proposed mechanism Animal research

In mice, the oil reduced ethanol ulcer area and, with geraniol, sped healing; in the dish it partly blocked the stomach acid pump (Venzon 2018).

A proposed mechanism does not show that the effect happens in people.

Swallowed oil can irritate the stomach (theory; see the safety section).

^{36 4 20}

What the trials found

Ekpenyong 2015 (kidney): 105 healthy adults aged 18 to 35 drank tea made from 2, 4 or 8 g of lemongrass leaf powder a day for 30 days. At day 10, urine volume and diuretic indices rose in the 2 g and 4 g groups. By day 30, creatinine clearance and eGFR fell significantly in all groups, serum and urine creatinine rose, and serum urea rose in the 4 g and 8 g groups. The authors describe dose- and time-dependent adverse effects. The reason for the fall was not established. Carbajal 1989: weak diuretic effect in rats given decoction by mouth.

Results

A diuretic effect, together with a fall in measured kidney filtration after 30 days.

Effect sizes as reported

Creatinine clearance and eGFR down significantly at day 30 (P < .05); values not in the abstract.

Why this rating

Single research group, healthy volunteers, no placebo group described; not independently repeated.

Form and dose studied

Lemongrass leaf tea made from 2, 4 or 8 g leaf powder daily for 30 days.

Who was studied

Healthy adults aged 18 to 35, Nigeria.

Limits

No control group in the abstract; mechanism unknown.

Why researchers looked

Folk use as a diuretic (water pill) and a rat study found a weak diuretic effect.

Traditional use

Folk use as a diuretic is mentioned in reviews; the Cuban study tested it in rats (Carbajal 1989).

Proposed mechanism Pharmacological theory

Not established.

A proposed mechanism does not show that the effect happens in people.

People with kidney disease, on diuretics or on drugs that affect the kidneys should avoid regular medicinal-strength lemongrass tea until a doctor says otherwise.

^{8 3}

What the trials found

Ekpenyong 2015 (blood): in 105 healthy adults, tea from 2, 4 or 8 g of leaf powder daily for 30 days raised packed cell volume, haemoglobin and red cell count (P < .05), especially at 30 days. MCH, MCV and MCHC did not change. Most white cell counts fell, but neutrophils and lymphocytes rose in some groups. The same volunteers had a drop in kidney filtration (see the kidney benefit).

Results

Changes in blood counts in healthy people without anaemia, no clinical benefit shown.

Effect sizes as reported

Rises in PCV, Hb, RBC significant (P < .05); values not in the abstract.

Why this rating

Healthy volunteers, no placebo group described, one research group. A rise in red cell counts can also reflect less body water, which fits the diuretic effect.

Form and dose studied

Lemongrass leaf tea from 2, 4 or 8 g of powder daily for 30 days.

Who was studied

Healthy adults aged 18 to 35, Nigeria.

Limits

No clinical outcome; no independent repeat; could be a dehydration effect.

Why researchers looked

An exploratory look at blood counts in the same volunteers.

Proposed mechanism Pharmacological theory

The authors suggest nutrients and antioxidants in the leaf; not tested.

A proposed mechanism does not show that the effect happens in people.

A fall in white cell counts was also reported; clinical meaning is unknown.

⁹

What the trials found

Sari 2025: women at a public health centre received a 20-minute aroma session daily for 4 weeks with ginger oil, lemongrass oil, or both. All three arms had significantly lower Menopause Assessment Scale scores. The abstract says 355 women but 89 per arm (3 x 89 = 267), so the real size is unclear.

Results

Symptom scores fell in all arms, including lemongrass alone.

Effect sizes as reported

Statistically significant reductions in all three groups; numbers not given.

Why this rating

No comparison with no aroma is described, the numbers are inconsistent, and a combination arm is mixed with ginger.

Form and dose studied

Lemongrass oil aroma, 20 minutes a day for 4 weeks (alone, or combined with ginger oil).

Who was studied

Menopausal women, Indonesia.

Limits

No clear control; participant numbers inconsistent.

Why researchers looked

Aromatherapy is used for menopausal discomfort.

Proposed mechanism Pharmacological theory

Not known.

A proposed mechanism does not show that the effect happens in people.

Safety is covered in the safety section. Lemongrass essential oil is a strong oil: it can cause skin allergy, and it should not be swallowed on the strength of these results.

²⁸

What the trials found

Carbajal 1989: a decoction showed weak anti-swelling effect by mouth in rats. No human inflammation trial was found.

Results

Lab and animal only.

Effect sizes as reported

Dose-dependent COX-2 suppression in cells.

Why this rating

Cell and rat work; not a human result.

Form and dose studied

Citral and oil in cells; decoction in rats.

Who was studied

Cells; rats.

Limits

No human trial.

Why researchers looked

To explain folk use for pain and swelling.

Traditional use

Cuban folk use as an anti-inflammatory (Carbajal 1989).

Proposed mechanism Laboratory research

Citral activated PPAR alpha and gamma and lowered COX-2 expression in human macrophage-like cells (Katsukawa 2010). Both Cymbopogon oils blocked release of the inflammation marker MPO from human white cells in the dish (Silva 2010).

A proposed mechanism does not show that the effect happens in people.

None specific.

^{18 31 3}