A recent study published in the prestigious New England Journal of Medicine (NEJM) (1) casts a shadow on the common use of testosterone therapy in older men for boosting their energy and vitality. An FDA-approved testosterone gel was shown to significantly increase risk for adverse cardiovascular, respiratory, and dermatological events in men 65 or older who had impaired mobility and increased health risks.
These unexpected negative results fly in the face of decades of good research on testosterone therapy in older men showing it to be beneficial to the cardiovascular system, and improve sexual function, mood, energy level, and muscle and bone mass and strength (2,3,4,5).
What went wrong? Why were the results so different from hundreds of other publications painting testosterone as the elixir of youth and vitality? Was it the dose of testosterone these men were using, or the way it was delivered as a gel through the skin?
In the NEJM study the testosterone treated men had mean venous serum testosterone levels of 574+/- 403 ng/dL, compared with 292+/- 160 ng/dL in the placebo group. The Endocrine Society Clinical Practice Guidelines suggests that testosterone should be dosed to achieve serum levels about mid-range for a healthy young male (5). The 50-150 mg topical testosterone gel put the testosterone-treated men well within the expected mid-physiological range seen in healthy young men (about 500-1000 ng/dL). So based on conventional serum testing one would conclude that overdosing would not likely be responsible for the adverse reactions these men had to testosterone. Or was it?
Let’s look at some assumptions made about using conventional venous serum to monitor how much testosterone is entering the body following topical testosterone therapy. It is “assumed” that when testosterone is delivered as a topical gel only 10% is absorbed and utilized by the body. This is based on measuring the total amount of testosterone that can be detected in venous serum over a specific time frame, usually about 12-24 hours, after the testosterone has been applied to the skin. With this conventional wisdom of 10% absorption, topical testosterone gel has been FDA-approved at a pharmacological dose about 10 times higher (50-100 mg) than the amount a healthy young male’s testes will manufacture in a day, which is about 5-10 mg of testosterone.
The NEJM study (1), and many others preceding it (4,5), confirmed that the daily topical dosage of 50-150 mg testosterone results in a level of serum testosterone (mean 574 ng/dL) that would be considered healthy in young men. So why was this dose of topical testosterone not healthy for the older men?
It is “assumed” that when testosterone is delivered into the body using a topical gel that serum testosterone measurements, derived from a venous blood draw, actually reflect how much testosterone is entering tissues throughout the body.
What evidence do we have that the pharmacological dose (50-150 mg) of topical testosterone used in the NEJM study was actually “delivering” a physiological dose of testosterone into the tissues of these older men? The correct answer is none. This would require measuring the levels of testosterone in different tissue biopsies, which for obvious reasons is not practical and such studies would not likely find many volunteers.
However, there are other means to look a little closer at “tissue” exposure to testosterone, which is to measure the level in the fluid of tissues bathed by capillary beds (blood flowing to tissues and delivering nutrients and hormones) as opposed to blood flowing away from tissues (i.e. venous blood). Capillary blood and saliva provide the next-best alternative as they are easily accessible body fluids that are more representative of the direct interaction of capillary beds with tissues (6,7).
Over the past 10 years ZRT Laboratory has measured the levels of steroid hormones (estradiol, progesterone, testosterone, DHEAS, cortisol) in saliva and capillary blood (finger stick) from men and women who have used a plethora of different hormone doses and delivery systems (topical creams and alcoholic gels, oral, im-injections, sc-pellets, sublingual drops, and troches). Information on hormone type, dosage, delivery, and timing on well over a million people has been recorded in our database.
In mining this extensive database, we find that when steroid hormones are produced endogenously by the ovaries and testes and released into the bloodstream the levels in venous blood serum are quantitatively equivalent to whole capillary blood drawn from the finger (7). Under the same conditions, salivary hormones are about 2-3% of venous serum or capillary blood hormone levels, as reported extensively in the literature (6).
In contrast to what we see with endogenously produced hormones, when testosterone, or any other steroid hormone, is applied topically as a cream, gel, or spray, levels in capillary blood (finger) are remarkably higher, by 10-20 fold, than venous blood levels. Moreover, salivary hormones are often >100% of venous serum levels, which ordinarily are only 2-3%. This remarkable increase in capillary blood and salivary testosterone is NOT seen when it, or other hormones, are delivered orally, or as im-injections or sc-pellet implants. With these delivery systems capillary blood and venous serum levels are about the same.
This discrepancy in venous serum vs capillary blood and salivary hormone levels following topical hormone delivery has created enormous confusion among physicians attempting to interpret hormone test results. Those using venous serum are often frustrated that pharmacological dosing, as seen in the NEJM article, only leads to lower to mid-physiological serum testosterone levels, but side effects such as excessive buildup of red blood cells associated with pharmacological dosing. On the other side, those using saliva and capillary blood testing are shocked to see that pharmacological dosing leads to pharmacological levels much higher than expected.
Specifically, with reference to the use of topical testosterone gels in men at the 50-150 mg dosage, we find mean capillary (finger tip) blood levels of testosterone to be about 5000 ng/dL, approximately 10 x higher than the venipuncture serum blood levels reported in the NEJM study (574 +/- 403 ng/dL).
Of further relevance to the NEJM study, we find that men using the higher pharmacological topical testosterone gel have capillary blood levels of estradiol that usually exceed 100 pg/ml, which is very high for males. While physiological levels of estrogens are beneficial to the male cardiovascular system, higher levels can be harmful (8). Unfortunately, estrogens were not monitored in the NEJM study so it is not possible to know if levels, even in venous serum, were higher than physiological range. It is very possible that some of the cardiovascular events observed in the older high risk men participating in the NEJM study were precipitated by excess estrogens derived from excess testosterone.
We believe these differences in conventional venipuncture serum vs capillary blood and saliva testosterone provide a rational explanation for the unexpected adverse clinical events seen in the NEJM study.
Based on the exceptionally high testosterone and estradiol levels we have documented in capillary blood and saliva following use of topical testosterone gels in the 50-150 mg range, we suggest that dosage for topical delivery of testosterone, or any steroid hormone, should not be determined by venipuncture serum levels as this will result in over dosage and potential for toxicity, as seen in the men participating in the NEJM study.
Perhaps a lesson is learned from The Swiss alchemist, Paracelsus, who noted some 500 years ago: "All things are poison and nothing is without poison. Only the dose permits something not to be poisonous."
Hormones are no different. A physiological level of testosterone is unquestionably life-enhancing in men of all ages and health conditions who suffer from testosterone deficiency. However, at a high enough dose testosterone, like all hormones, has the potential to transition from an elixir to a dangerous drug.
(1) Basaria S et.al. N Eng J Med June 30, 2010. Adverse Events Associated with Testosterone Administration
(2) Bhasin S, et. al. J Clin Endocrinol Metab 90: 678-688, 2005. Older men are as responsive as young men to the anabolic effects of graded doses of testosterone on skeletal muscle.
(3) Wu FCW et. al. N Engl J Med 363: 123-35, 2010. Identification of late-onset hypogonadism in middle-aged and elderly men.
(4) Fernandez-Balsells et. al. J Clin Endocrinol Metab 95: 2560-2010, 2010. Adverse effects of testosterone therapy in adult men: a systematic review and meta-analysis.
(5) Bhasin S et.al. J Clin Endocrinol Metab 95: 2536-2559, 2010. Testosterone therapy in men with androgen deficiency syndromes: an endocrine society clinical practice guideline.
(6) Vining RF, McGinley RA. J Steroid Biochem 27: 81-94, 1987. The measurement of hormones in saliva: possibilities and pitfalls.
(7) Edelman et.al. Fert steril 88(5): 1404-7, 2007. A comparison of blood spot vs plasma analysis of gonadotrophin and ovarian steroid hormone levels in reproductive age women.
(8) Sudhir K, Komesaroff PA. J Clin Endocrinol Metab 84, 3411-3415, 1999. Cardiovascular actions of estrogens in men.