When looking for articles on testosterone, I notice a lot of hysteria on the subject of soy's affect on testosterone levels. People fail to recognize that the isoflavones in soy that are supposedly responsible for lowering testosterone aren't found in soybean oil, which is a refined product, and refusing to eat any food whose ingredient list includes it. There's even a claim that the presence of soy in processed foods is a plot to feminize men and keep them placid and docile. A recent article alerted me to an older placebo controlled trial of the effect of soy and soy isoflavones on testosterone levels. Not only did they find no affect, they found that consuming soy isoflavones had a positive effect on markers of heart disease and diabetes.
The full text of the study is available.
Isoflavones found in soy products have a chemical structure similar to estrogen, leading to concerns of an adverse estrogenic effect in men, particularly in those with type 2 diabetes mellitus (T2DM) who have low testosterone levels due to hypogonadism. Two hundred men with T2DM and a total testosterone level ≤12 nmol/L were included [in the study]. Fifteen grams of soy protein with 66 mg of isoflavones (SPI) or 15 g soy protein alone without isoflavones (SP) daily as snack bars for 3 months were administered.
There was no change in either total testosterone or in absolute free testosterone levels with either SPI or SP. There was an increase in thyrotropin (TSH) and reduction in free thyroxine (fT4; P < 0.01) after SPI supplementation. Glycemic control improved with a significant reduction in hemoglobin A1c (−4.19 [7.29] mmol/mol, P < 0.01) and homeostasis model of assessment - insulin resistance after SPI. Cardiovascular risk improved with a reduction in triglycerides, C-reactive protein, and diastolic blood pressure (DBP; P < 0.05) with SPI vs SP supplementation. There was a 6% improvement in 10-year coronary heart disease risk after 3 months of SPI supplementation. Endothelial function improved with both SPI and SP supplementation (P < 0.01), with an increased reactive hyperemia index that was greater for the SPI group (P < 0.05).
The full text of the study is available.