Journal Logo

Basic Sciences: Original Investigations

Effects of dehydroepiandrosterone vs androstenedione supplementation in men


Author Information
Medicine & Science in Sports & Exercise: December 1999 - Volume 31 - Issue 12 - p 1788
  • Free


Dehydroepiandrosterone (DHEA), its sulfate ester (DHEA-S), and androstenedione are androgenic hormones produced primarily by the adrenal glands, which serve as precursors in the endogenous production of both testosterone and estrogens (7,20). In the adrenal cortex, DHEA is converted to androstenedione that, in turn, can either be dehyrogenated in the liver to testosterone or aromatized to estrone (13). It is, in fact, their role in testosterone synthesis that has stimulated much of the recent interest in their potential as anabolic/ergogenic aids.

Like testosterone, the production of these hormones peaks in the mid-20s and then declines steadily with age after the third decade of life (19). It has been speculated, therefore, that supplementation with these “precursor” hormones may help keep androstenedione and/or testosterone levels elevated thereby optimizing the anabolic state and attenuating age-associated metabolic/hormonal changes and related sequelae (e.g., decrease in immune function, lean body mass, and strength levels; and increase in body fat, insulin resistance, and coronary artery disease). Several recent studies have, in fact, demonstrated age-related benefits with DHEA supplementation including increases in lean body mass and strength (17,18,25) as well as a reduced risk of heart disease, improved lipid levels, and enhanced immune system function and glucose tolerance/insulin sensitivity (1,3,4,7,9,11,14,18,24,25). These findings have generated considerable interest in the potential health and longevity benefits of androgenic hormone supplementation.

However, studies with humans have produced equivocal results (22,23). These discrepancies may be related to a number of factors including: the basal physiologic milieu (initial hormonal/metabolic profile); body composition (lean body mass vs fat mass); nutritional status (quantity, composition, and timing); training index (frequency × intensity × volume); dosing criteria (timing, quantity, and route of administration); and age.

Supplementation with androgenic hormones also poses a potential risk for promoting growth of hormone-sensitive tumors, most notably prostatic hypertrophy, as well as promoting hepatotoxicity, and increasing the risk of cardiovascular disease (8,10,21). However, there is little evidence of acute toxicity associated with DHEA or androstenedione supplementation in humans. Further, there is little information regarding the effects of normal physiological doses in middle-aged men participating in a strength-training program. The purpose of this study, therefore, was to compare the effects of short-term (12 wk) DHEA versus androstenedione supplementation on body composition, strength levels, and hormonal profiles in middle-age men experienced in weight training. A key question to be addressed is whether or not such supplementation will induce any adverse side effects in these subjects.



Forty healthy men between the ages of 40 and 60 (mean ± SD age: 48.1 ± 3.9 yr) experienced in weight training (>1 yr) participated in this study. All subjects completed a medical history, lifestyle and training inventory, and written informed consent before participation. All procedures were in compliance with human subject guidelines established by the U.S. Department of Health, Education, and Welfare and the American Physiological Society. Subjects were required to maintain their normal training index (frequency × intensity × volume), physical activity patterns, and dietary regimen (including no use of any other supplements) throughout the duration of the study.

Experimental Design

Subjects participated in one of three randomly assigned experimental trials. Each supplement phase consisted of blind assignment of either a placebo (P), DHEA (D), or androstenedione (A) capsules. Supplements were measured in 50-mg quantities and placed in generic capsules coded for identification. Capsules were ingested two times daily for 3 months (i.e., 100 mg·d−1). All measurements (including a body composition profile, blood chemistry and hormone profile, and performance measures) were made at three times points: presupplementation (1 d), 6 wk, and postsupplementation (12 wk).

Body Composition

Body composition was determined using hydrostatic-weighing techniques and corrected for residual volume using the helium dilution method. Body fat and fat free mass were calculated from body density values. The waist-to-hip ratio, an index of intra-abdominal adiposity, was calculated by dividing the circumference of the abdomen (at the level of the umbilicus) by the circumference of the buttocks where the hips are largest.

Blood Samples

Fasting blood samples (12 h) were taken from an antecubital vein by a trained phlebotomist between 0700 and 0800 on three consecutive days. Samples were immediately refrigerated and transported to the lab and analyzed for the following: a blood chemistry profile (including a lipid profile, liver function, and glucose); hormones (including testosterone, cortisol, DHEA-S, androstenedione, insulin-like growth factor-1 (IGF-1), and insulin; sex-hormone-binding-globulin (SHBG); and prostate-specific antigen (PSA). All hormone measurements for each subject were performed in the same assay. Total cholesterol, glucose, and triglycerides were determined using enzymatic methods/spectrophotometry; high-density lipoprotein (HDL)-C was determined by phosphotungstic precipitation/spectrophotometry; and serum concentrations of DHEA-S, testosterone, insulin, SHBG, IGF-1, and androstenedione were measured by specific radioimmunoassays (RIA). The intra- and inter-assay coefficients of variation were 4 and 7% for DHEA-S, 5 and 8% for testosterone, 4 and 9% for androstenedione, 5 and 10% for IGF-1, 3 and 6% for SHBG, 2 and 3% for PSA, and 5 and 7% for insulin.


Each subject’s maximal functional capacity was tested for both muscular strength and aerobic capacity.


Subjects were tested for one repetition maximum (1RM) upper-body (barbell bench press) and lower-body (leg press) strength. After two to three progressive warm-up sets, sets of 1RM were completed with increasing loads until the subject could not complete one repetition. Each set was followed by a 2- to 3-min rest period. The last successful weight lifted was determined to be the 1RM.

Aerobic capacity.

A symptom-limited maximal exercise test was performed to assess cardiopulmonary and metabolic functional capacity, to screen for coronary artery disease (ischemic and arrhythmic responses), and to evaluate any potential contraindications to exercise training. All tests were performed using an electromagnetically braked cycle ergometer using a 25-W RAMP protocol. Each test consisted of 2 min of unloaded pedaling at 60 rpm, followed by a 2-min warm-up at 25 W and a subsequent constant rate of increase of power output (25 W·min−1). An electrocardiogram was used to monitor cardiac rate and rhythm continuously throughout the test using standard 12-lead procedures. Metabolic and ventilatory responses were assessed using a digital computer based exercise system (SensorMedics, Yorba Linda, CA).

Training/Dietary Control

Subjects met individually with a nutritionist who instructed them to maintain their normal dietary pattern throughout the course of the study and to refrain from using any other supplementation. Compliance was monitored by analyzing 3-d food records pre-, mid-, and post-supplementation. Exercise logs/training programs were also reviewed at this time by an exercise physiologist to insure compliance with their training program. All subjects increased their training loads progressively over the course of the study to maintain an 8–12 RM intensity. A visual analog scale of libido and an open-ended questionnaire for self-assessment of well-being were administered pre- and post-supplementation. A 24-h history questionnaire consisted of questions related to factors such as sleep patterns, physical health, exercise levels, and general state of well-being.

Statistical Analysis

All results are reported as means ± SE. Data were analyzed using a two-way (3 × 3, time × group) analysis of variance. Interactions were analyzed on pre/post scores using a t-test. Pearson product moment correlations were performed to determine significant correlations between variables. When significant effects were revealed, a Scheffe post hoc test was used to locate the pairwise differences among means. For all statistical tests, significance was accepted at the P < 0.05 level.


Performance Measures

Figure 1 summarizes the results of the maximal exercise test and 1RM testing for each trial. There were no significant differences (NS) demonstrated between groups on any baseline (pre-) measures. There was a small increase in power output (W) and V̇O2max (L·min −1) for all groups that were not significantly different (NS) between groups. There was also an increase from baseline in both strength measures, bench press (4.3 ± 1.2 and 3.4 ± 1.0 kg), and leg press (13.0 ± 3.1 and 10.8 ± 2.4 kg) in both D and A, respectively, that were also not significantly different from changes in P or from each other (P < 0.05).

Figure 1:
Percentage change of performance measures pre- to post-supplementation. No significant differences were demonstrated (P < 0.05).

Body Composition

There were no significant differences observed between groups on baseline measures of body composition, including body mass (BM), percent fat (%), lean body mass (LBM), and fat mass (FM). As seen in Table 1, there was a small increase in LBM relative to baseline in D and A that was not significantly different from that demonstrated in P or from each other (P < 0.05).

Table 1:
Body composition measures pre- and post-supplementation; values are mean ± SD.

Hormone Profile

Figure 2 summarizes the results of the salient (androgenic) hormonal changes. There were no significant differences observed between groups on any baseline hormone levels. Initial basal levels of all hormones were within the normal range for men. However, there was a significant correlation (r = 0.56) between baseline DHEA values and LBM that did not increase with supplementation. Further, there was a significant increase in serum DHEA-S levels and a NS increase in IGF-1 demonstrated in D only (P < 0.05). No other statistically significant hormonal changes were observed in any group.

Figure 2:
Percentage change of androgenic hormone profile pre- to post-supplementation. * Statistically different from baseline (P < 0.05). ** Statistically different from placebo (P < 0.05).

Risk Factor Profile

To monitor the occurrence of potential adverse side effects of supplementation, the following variables were determined: PSA, glucose, insulin, lipids, and liver function (ALT transaminase). Table 2 summarizes the results of this risk factor profile. There were no significant differences observed in any baseline “risk factor” measures. Small, NS decreases in glucose, cholesterol, and triglycerides were demonstrated in D. There were no significant differences reported for any of these risk factors relative to baseline or P.

Table 2:
Risk factor profile pre- and post-supplementation; values are mean ± SE.

Well-Being and Libido

A number of subjects reported an improved sense of well-being after supplementation (D = 4, A = 3, P = 3). Subjective statements included enhanced feelings of relaxation, sleep quality, and energy and stress management. Two subjects each in D and A, and one in P, reported improved feelings of libido. Further, two subjects in D reported a reduction in joint pain. Only one subject each in D and A reported an adverse effect on sleep quality. No other adverse effects were reported by any subjects.


The purpose of this study was to compare the effects of DHEA versus androstenedione supplementation in healthy, middle-aged men currently involved in a weight training program. The results demonstrate that supplementation did not significantly increase lean body mass or strength levels relative to changes observed in P. These results are consistent with previous studies reporting no increase in LBM or decrease in percent fat (23) but contradict others reporting an increase in both LBM and strength with DHEA supplementation (17,25). As mentioned, these discrepancies may be related to a number of factors including: the basal physiologic milieu (e.g., initial hormonal/metabolic profile), body composition (lean body mass vs fat mass), nutritional status (quantity, composition, timing), training index (frequency × intensity × volume), dosing criteria (timing, quantity, route of administration), and age. Though of similar age and body mass, only two of our subjects (both in D) were below the normal reference range for serum DHEA and androstenedione concentrations. Such low levels may be critical to inducing a statistically significant response with replacement dose supplementation.

Though there is evidence that androstenedione can raise serum testosterone into normal range from low basal levels (2,13), in the present study, supplementation with androstenedione did not significantly increase serum androstenedione or testosterone levels. This lack of a conversion response in A may be related to the short half-life of androstenedione and its rapid conversion to estrogens in peripheral tissues (12). Further, it has been characterized as a weak androgen with only a minimal amount converted to testosterone and more to estrogen (16). Muscle, along with adipose tissue, is a major source of serum estrogens derived from androstenedione (15). In fact, increasing androstenedione levels in men may not provide the anabolic environment desired. Overproduction of androstenedione may cause feminizing signs such as gynecomastia, whereas elevated levels cause increased estrone levels. On the other hand, supplemental androstenedione in low doses is not prone to excess elevation of estrogen levels (13), particularly in hypogonadism. It also appears that a small additional amount of estrogen produced by supplemental androstenedione may trigger formation of more androgen receptors where they are wanted most—skeletal muscles (5). It is also likely that the body rapidly adapts to excess androstenedione and may down-regulate its production and subsequent conversion to testosterone or estrogens as needed. Finally, because androstenedione and DHEA have been shown to decrease with long-term strength training (2), supplementation may have been “sacrificed” to maintain testosterone levels and prevent overtraining. Replacement therapy may help reduce overtraining syndrome and prevent a deficiency induced by intense training and delayed strength plateaus normally observed. Though the subjects in this study were involved in a weight training program 3–4 d·wk−1, the intensity and/or volume were not sufficient to elicit an overtraining syndrome.

In contrast to A, there was a significant increase in DHEA-S levels and a mild, NS increase in IGF-1 in D. Although the precise functions of DHEA (and thus the implications of an elevated DHEA level with supplementation) remain unclear (8), it has been postulated that it may play a role as a discriminator of life expectancy and aging (4). It has also been reported that DHEA-S level is independently and inversely predictive of death from any cause and from cardiovascular disease (3,4). Thus, it is tempting to speculate that supplementation with DHEA may help confer some protection against chronic disease and/or premature death. Future prospective studies would be needed to confirm this potential benefit.

Though smaller in magnitude, the increase in IGF-1 observed in this study is consistent with others’ (17,25) who have suggested that an increased bioavailability to target tissues over time may manifest as improved capacity in physical and psychological performance in deficient subjects. However, this was not demonstrated in the present study nor was the increase in IGF-1 statistically significant relative to P. Although a number of subjects reported an improved sense of well-being after supplementation, the large majority of subjects did not report this benefit. Again, there may be a responder subset who may benefit with an enhanced sense of well-being and libido. Moreover, two subjects in P were convinced they were taking one of the hormone supplements, which, of course, demonstrates a significant placebo and/or seasonal effect. Results may have, in fact, been confounded by the timing of this study (spring-summer) which is often characterized by an improved sense of well-being and motivation.

Although several studies have reported that androgen levels may be related to insulin levels (18), there was not a significant relationship between fasting insulin and DHEA-S or androstenedione levels in this study. This confirms more recent studies also failing to demonstrate a relationship (6). The potential benefit of supplementation with these hormones may only be manifest in specific population subsets (characterized by low basal hormone levels and/or high-intensity training) using much longer supplementation periods. Researchers have, in fact, recently demonstrated the ability of DHEA supplementation to undergo biotransformation into potent androgens and estrogens in subjects with panhypopituitarism, characterized by the absence of adrenal and gonadal steroid secretion (26).

A key question to be addressed by this investigation was whether or not (replacement dose) supplementation with these androgenic hormones produces any adverse side effects. The results suggest that supplementation with 100 mg·d−1 of either DHEA or androstenedione do not elicit any significant adverse side effects over a 3-month period. This includes changes in PSA, glucose, insulin, lipids, and liver function. In fact, it appears that even doubling serum androstenedione levels is not harmful (18). Moreover, bioavailability of oral steroid use is only approximately 5% largely due to the effects of digestion, liver metabolism, and specific tissue enzyme activity. Like other fat-soluble nutrients, it is assumed that percent uptake also decreases with higher doses, thereby demonstrating another intrinsic safety mechanism.

In conclusion, despite several subtle physiologic “trends” in performance and hormone measures, any effect of supplementation in A or D was not statistically greater than that observed in P. Given the low bioavailability of these “weak” androgens coupled with internal feedback mechanisms (e.g., down-regulation), normal hormonal profiles, and potential placebo/seasonal affects, it is not surprising that significant differences were not observed in this relatively short-term study. Potential benefits would more likely be found with long-term supplementation and in special subset populations with very low initial basal levels and/or involved in high intensity training. Future studies should further delineate these populations and dosing criteria.

This research was supported by the United States Sports Academy and Weider Research International.


1. Abbassi, A., E. H. Duthrie, L. Sheldabl, et al. Association of dehydroepiandrosterone sulfate, body composition, and physical fitness in independent community-dwelling older men and women. J. Am. Geriatr. Soc. 46:263–273, 1998.
2. Alen, M., A. Pakarinen, K. Hakkinen, and P. V. Komi. Responses of serum androgenic-anabolic and catabolic hormones to prolonged strength training. Int. J. Sports Med. 9:220–233, 1988.
3. Barrett-Connor, E., and D. Goodman-Gruen. The epidemiology of DHEAS and cardiovascular disease. Ann. N. Y. Acad. Sci. 774:259–270, 1995.
4. Barrett-Connor, E., K. T. Khaw, and S. S. Yen. A prospective study of dehydroepiandrosterone sulfate, mortality, and cardiovascular disease. N. Engl. J. Med. 315:1519–1524, 1986.
5. Chaikovskii, V. S., I. V. Evtinova, and O. B. Basharina. Steroid levels and androgen receptors in skeletal muscles during adaptation to physical effort. Vopr. Med. Khim. 31:81–86, 1985.
6. Denti, L., G. Pasolini, L. Sanfelici, F. Ablondi, et al. Effects of aging on dehydroepiandrosterone sulfate in relation to fasting insulin levels and body composition. Metabolism 46:826–831, 1997.
7. Ebeling, P., and V. Koivisto. Physiological importance of dehydroepiandrosterone. Lancet 343:1479–1481, 1994.
8. Gann, P. H., C. H. Hennekens, C. Longcope, W. Verhoek-Oftedahl, W. Grodstein, and M. J. Stampfer. A prospective study of plasma hormone levels, nonhormonal factors, and development of benign prostatic hyperplasia. Prostate 26:40–49, 1995.
9. Haffner, S. M., R. A. Valdez, L. Mykkanen, M. P. Stern, and M. S. Katz. Decreased testosterone and dehydroandrosterone sulfate concentrations are associated with increased insulin and glucose concentrations in nondiabetic men. Metabolism 43:599–603, 1994.
10. Jones, J. A., A. Nguyen, and M. Straub. Use of DHEA in a patient with advanced prostate cancer: a case report and review. Urology 50:784–788,.
11. Khaw, K. T. Dehydroepiandrosterone, dehydroepiandrosterone-sulfate, and cardiovascular disease. J. Endocrinol. 150:(Suppl.)S149–S153, 1996.
12. Longcope, C., R. B. Billiar, Y. Takaoka, P. S. Reddy, D. Richardson, and B. Little. Tissue sites of aromatization in the female rhesus monkey. Endocrinology 113:1679–1682, 1983.
13. Mahesh, V. B., and R. B. Greenblatt. The in vivo conversion of dehydroepiandrosterone and androstenedione to testosterone in the human. Acta Endocrinol. 41:400–406, 1962.
14. Marin, P. Androgen treatment of abdominally obese men. Obes. Res. 1:245–251, 1993.
15. Mastrogiacomo, I., G. Bonanni, E. Menegazzo, et al. Clinical and hormonal aspects of male hypogonadism in myotonic dystrophy. Ital. J. Neurol. Sci. 17:59–65, 1996.
16. Matsumine, H., K. Hirato, T. Yanaihara, T. Tamada, and M. Yoshida. Aromatization by skeletal muscle. J. Clin. Endocrinol. Metab. 63:717–720, 1986.
17. Morales, A. J., J. J. Nolan, and J. C. Nelson. Effects of replacement dose dehydroepiandrosterone in men and women of advancing age. J. Clin. Endocrinol. Metab. 78:1360–1367, 1994.
18. Nestler, J. E., C. O. Barlascini, J. N. Clore, and W. G. Blackard. Dehydroepiandrosterone reduces serum low density lipoprotein levels and body fat but does not alter insulin sensitivity in normal men. J. Clin. Endocrinol. Metab. 66:57–61, 1988.
19. Orenteich, N., J. L. Brind, R. L. Rizer, and J. H. Vogelman. Age changes and sex differences in serum DHEA-S concentrations throughout adulthood. J. Clin. Endocrinol. Metab. 59:551–555, 1984.
20. Shackleton, C., E. Roitman, A. Phillips, and T. Chang. Androstenediol, and 5-androstenediol profiling for detecting exogenously administered dihydrotestosterone, epitestosterone, and dehydroepiandrosterone: potential use in gas chromatography isotope ratio mass spectrometry. Steroids 62:665–673, 1997.
21. Tenover, J. L. Testosterone and the aging male. J. Androl. 18 (2):103–106, 1997.
22. Usiskin, K. S. Lack of effect of dehydroepiandrosterone in obese men. Int. J. Obes. 14:457–463, 1990.
23. Welle, S., R. Jozefowicz, and M. Statt. Failure of dehydroepiandrosterone to influence energy and protein metabolism in humans. J. Clin. Endocrinol. 71:1259–1264, 1990.
24. Wisniewski, T. L., C. W. Hilton, E. V. Morse, and F. Svec. The relationship of serum DHEA-S and cortisol levels to measures of immune function in human immunodeficiency virus-related illness. Am. J. Med. Sci. 305:79–83, 1993.
25. Yen, S. S., A. J. Morales, and O. Khorram. Replacement of DHEA in aging men and women: potential remedial effects. Ann. N. Y. Acad. Sci. 774:128–142, 1995.
26. Young, J., B. Couzinet, K. Nahoul, S. Brailly, et al. Panhypopituitarism as a model to study the metabolism of dehydroepiandrosterone (DHEA) in humans. J. of Clin. Endocrinol. Metab. 82:2578–2585, 1997.


© 1999 Lippincott Williams & Wilkins, Inc.