Benefits
Blood sugar and insulin measures over 90 days of twice-daily use
In a 90-day trial in 24 adults with impaired glucose tolerance, HbA1c and insulin output after a glucose drink fell within the tarragon group, but no direct comparison with placebo was reported. Systolic blood pressure also fell and HDL cholesterol rose in that group, while the placebo group showed no significant changes. The dose was 1,000 mg before breakfast and before dinner.
Glucose and insulin response after a single dose before a glucose drink
In a double-blind crossover test, 12 healthy young men took 2 g of a Russian tarragon water extract or placebo 15 minutes before a 75 g glucose drink. Glucose and insulin curves did not differ significantly between conditions (glucose area 4.5% lower, insulin area 17.4% lower, P = 0.54 and 0.53). A company whose employee co-authored the study helped fund it.
Creatine retention in muscle when taken before creatine
In a double-blind crossover study of 10 trained men, 500 mg of a Russian tarragon water extract taken 30 minutes before each 5 g creatine dose (twice daily for 5 days) did not change whole-body creatine retention, muscle creatine or sprint power compared with placebo. The extract supplier funded the study.
Insulin signaling in muscle cells and blood sugar in mice
Lab and animal work only. An ethanol extract of Russian tarragon raised glucose uptake and lowered the enzyme PTP1B in muscle cells grown from donors with type 2 diabetes, and at 500 mg/kg a day for 7 days it lowered blood glucose by 20 to 24% in two diabetic mouse models but not in non-diabetic animals. These effects have not been confirmed in people.
Long-standing culinary herb with US food flavoring status
Federal rules list tarragon (Artemisia dracunculus) among spices and among essential oils and natural extractives generally recognized as safe for their intended use in food. That status covers seasoning and flavoring amounts, not concentrated supplement doses taken daily for weeks.
Mechanism of action
Insulin receptor signaling in muscle cells (lab)
In human muscle cells grown in the lab, the ethanol extract PMI 5011 increased glucose uptake, partly restored glycogen storage under a fatty acid load and lowered the protein PTP1B, which normally dampens insulin signaling. These are cell findings only.
Liver glucose production enzyme in rodents
In diabetic rats the extract Tarralin lowered expression of PEPCK, an enzyme that drives glucose production in the liver, by 28%, and in a test tube it increased binding of the gut hormone GLP-1 to its receptor. Neither effect has been measured in people.
Platelet activity in lab tests
A methanol extract of tarragon leaves and its chloroform fraction reduced the adhesion of thrombin-activated human platelets to coated plates by 50 and 60% at 200 micrograms per mL (leaf powder equivalent), and lowered platelet aggregation and secretion in the dish. Whether eating tarragon affects clotting in people is unknown.
Clinical trials
Randomized, double-blind, placebo-controlled trial at the University of Guadalajara, Mexico, in adults with impaired glucose tolerance; A. dracunculus 1,000 mg before breakfast and before dinner, or placebo, for 90 days; outcomes oral glucose tolerance test glucose and insulin curves, insulin secretion and sensitivity indexes, HbA1c and lipids. Wilcoxon (within group) and Mann-Whitney (between group) tests were named. (Méndez-Del Villar et al. 2016, J Med Food)
24 adults with impaired glucose tolerance (12 tarragon, 12 placebo).
Within the tarragon group, HbA1c (5.8 to 5.6%), insulin area under the curve, total insulin secretion and systolic blood pressure (120 to 113 mmHg) fell and HDL cholesterol rose (P < 0.05); the placebo group showed no significant changes. The abstract gives no between-group P values and reports no significant change in insulin sensitivity or first-phase insulin secretion. Funding and adverse events are not stated in the abstract.
Randomized, double-blind, placebo-controlled crossover pilot study at the University of Memphis; 2 g of an aqueous Russian tarragon extract (essential oil removed, 30% maltodextrin carrier) or cellulose placebo 15 minutes before a 75 g dextrose drink, 1 to 2 weeks apart. Funded by the University of Memphis and PhytoLab, whose employee is a co-author; Finzelberg supplied the extract. Not PubMed-indexed. (Bloomer et al. 2011, Nutr Diet Suppl)
12 healthy, non-diabetic young men (mean age 24.8); all completed both sessions.
Serum glucose showed no condition effect (P = 0.19) and no difference in area under the curve (569 vs 596, P = 0.54); insulin area was 17.4% lower with tarragon but not significantly (P = 0.53). Heart rate and blood pressure were lower on the tarragon day, which the authors attribute mostly to lower values before dosing. The paper reports no side effects from the extract or the glucose drink.
Double-blind, randomized, placebo-controlled crossover study at Texas A&M University; 500 mg of an aqueous Russian tarragon extract or placebo taken 30 minutes before 5 g creatine monohydrate, twice daily for 5 days, with a 6-week washout; outcomes urinary creatine retention, muscle biopsy free creatine and Wingate sprint power. Funded by the Martin Bauer Group, Finzelberg, the extract supplier; three authors were named inventors on the group's pending patents. (Oliver et al. 2014, J Int Soc Sports Nutr)
10 recreationally trained men aged about 20.
Creatine raised whole-body creatine retention and muscle free creatine with or without tarragon, with no difference between conditions (P = 0.59 and P = 0.34). Sprint mean power and total work rose over time in both conditions without a between-condition difference. No side effects were reported.
Animal study of Tarralin, an ethanol extract of Russian tarragon, given by stomach tube at 500 mg/kg body weight a day for 7 days, with troglitazone and metformin as comparison drugs, plus a test-tube GLP-1 receptor binding assay. (Ribnicky et al. 2006, Phytomedicine)
Genetically diabetic KK-Ay mice, mice with chemically induced diabetes, diabetic rats and non-diabetic animals; no human participants.
Blood glucose fell 24% in KK-Ay mice and 20% in chemically diabetic mice versus controls, and blood insulin fell 33% in KK-Ay mice. PEPCK mRNA fell 28% in diabetic rats. Glucose, insulin and PEPCK did not change significantly in non-diabetic animals.
Cell study of the ethanol extract PMI 5011 in primary skeletal muscle cultures taken from donors with type 2 diabetes; outcomes glucose uptake, glycogen accumulation, insulin signaling proteins and PTP1B. (Wang et al. 2008, Metabolism)
Human muscle cells grown in culture; no participants received the extract.
Glucose uptake rose with increasing extract concentration, and glycogen storage, lowered by free fatty acids, was partly restored. The abundance of IRS-1, IRS-2, PI-3 kinase, Akt, the insulin receptor and Glut-4 did not change significantly, but PTP1B protein fell.
Genotoxicity study comparing pure methylchavicol (estragole) with deep-frozen and dried tarragon, using unscheduled DNA synthesis tests in cultured rat liver cells and in the livers of rats dosed by mouth. (Nesslany et al. 2010, Mutat Res)
Rat liver cells in culture and rats; no human participants.
Pure estragole was genotoxic in both tests. Deep-frozen tarragon was clearly less genotoxic than estragole at equal doses, and dried tarragon was negative, including in rats given up to 6.25 g/kg (18.75 mg/kg of estragole). In cells, tarragon reduced the genotoxicity of added estragole, but dried tarragon did not do so in rats given high estragole doses.