Fenugreek is a legume whose seeds flavor curries and whose extracts fill testosterone-booster capsules. The seeds carry steroidal saponins, diosgenin in particular, which is why plant-steroid language attaches to it.

Two literatures sit side by side. The older one is about blood sugar: fenugreek seed has been tested in people with diabetes for decades and the pooled trials find lower fasting glucose and HbA1c. The newer, smaller one is about men, testosterone and strength, mostly in trials of branded extracts, and the effect it finds is real but small.

The saponins and their aglycone diosgenin are proposed to act on steroid synthesis, and on aromatase and 5-alpha-reductase, the enzymes that turn testosterone into estradiol and DHT. Inhibiting those would leave more testosterone in circulation. The glucose effect runs through a different set of constituents, soluble fiber and the amino acid 4-hydroxyisoleucine, which slow carbohydrate absorption and act on insulin release. None of the human trials measured the mechanism; they measured the hormone or the glucose.

Testosterone. A 2023 meta-analysis of seven studies with 449 participants (378 men) found a small effect on total testosterone (SMD 0.32, 95% CI 0.09 to 0.55) in male athletes, with free testosterone, lean mass, fat mass and leg press all pointing the same way but not reaching significance. A 2020 meta-analysis of four trials also found a significant effect on total testosterone, with no numbers in its abstract.

The branded trial. 120 healthy men aged 43 to 70 took 600 mg a day of a standardized extract or placebo for 12 weeks. Aging-male symptom scores fell, morning erections and sexual activity increased, and total and free testosterone rose against placebo. No numbers in the abstract.

The independent dose-ranging trial. 95 men aged 40 to 80 took 0, 600, 1,200 or 1,800 mg a day of another extract for 12 weeks. Plasma testosterone rose 13% from baseline on any dose but not significantly against placebo (9%); saliva testosterone rose 37.2% against placebo. Read against the branded trial, this is what a modest effect looks like when the placebo arm is reported honestly.

Blood sugar. Fourteen trials in 894 people found HbA1c 0.88 percentage points lower with fenugreek; fasting and post-meal glucose moved in the same direction without reaching significance, with quality and heterogeneity flagged. A separate 12-trial meta-analysis found fasting glucose 12.9 mg/dL lower, HbA1c 0.58 points lower, and total and LDL cholesterol lower.

The trials report it as well tolerated, and the dose-ranging trial saw no side effects at up to 1,800 mg a day for 12 weeks. Fenugreek is a food, and the extract doses in trials are within what a curry-heavy diet delivers in seed.

Three practical cautions. It lowers blood sugar, so it adds to diabetes drugs. It is a legume related to peanuts and chickpeas, and allergy has been reported. And it gives sweat and urine a maple-syrup smell, which is harmless but alarming if you were not warned. Pregnancy is a traditional contraindication because of effects on the uterus in animals; no trial here addresses it.

A food and a dietary supplement in the United States, the United Kingdom and the European Union, with no claims reviewed. The testosterone and glucose claims the trials support are not claims a supplement label may make. No regulator has licensed any fenugreek extract as a medicine for either purpose.