Blood Testosterone Threshold for Androgen Deficiency Symptoms

There are few systematic studies of the relationship between blood testosterone concentrations and the symptoms of overt androgen deficiency. Because most testosterone preparations are relatively short-term, the rapid changes in blood testosterone concentrations they cause make it difficult to define any testosterone threshold. By contrast, subdermal testosterone implants provide stable blood testosterone concentrations over days to weeks, while gradually declining to baseline over 5-7 months. Hence, this provides an opportunity to define a blood testosterone threshold for androgen deficiency symptoms by observing androgen-deficient men as their familiar androgen deficiency symptoms return as testosterone pellets slowly dissolve. Among 52 androgen-deficient men who underwent 260 implantations over 5 yr, at the time of return of androgen deficiency symptoms the blood total and free testosterone concentrations were highly reproducible within individuals (F = 0.8, P = 0.49 and F = 1.4, 0.24, respectively) but varied markedly between men (F = 167 and F = 138, both P < 0.001), indicating that each person had a consistent testosterone threshold for androgen deficiency symptoms that differed markedly between individuals. The most reported symptoms of androgen deficiency were lack of energy, lack of motivation, and reduced libido. The symptomatic threshold was significantly lower in men with secondary hypogonadism compared with men with primary or mixed hypogonadism (total, 9.7 +/- 0.5 nmol/liter vs. 11.7 +/- 0.4 nmol/liter and 10.2 +/- 0.3 nmol/liter, P = 0.006; free, 146 +/- 10 pmol/liter vs. 165 +/- 6 pmol/liter and 211 +/- 18 pmol/liter, P = 0.002) but was not affected by the underlying cause of hypogonadism or by specific symptoms of any severity. Despite a wide range in individual thresholds for androgen deficiency symptoms, the mean blood testosterone threshold corresponded to the lower end of the eugonadal reference range for young men. The implications of these observations for the development of more specific quality-of-life measures, as well as for other potential androgen deficiency states such as chronic diseases and aging, remain to be determined.