Intended for healthcare professionals


Hormone replacement for men

BMJ 1996; 312 doi: (Published 06 April 1996) Cite this as: BMJ 1996;312:859
  1. Marc E Weksler
  1. Wright professor of medicine Division of Geriatrics and Gerontology, Cornell University Medical College, New York, NY 10021, USA

    Not enough evidence to recommend routine treatment with dehydroepiandosterone

    Some journalists, some patients, and some doctors believe that dehydroepiandosterone (DHEA) is the fountain of youth in a bottle. Most members of the biomedical community remain sceptical. The recent publication of a report of a conference at the New York Academy of Science has provided a collection of studies relating the hormone to the disabilities and diseases of aging.1 Many of the clinicians and investigators attending the conference had more than a scientific interest in the subject—as many as a quarter of them were thought to be taking DHEA.2

    DHEA is produced by the adrenal gland and circulates largely as its water soluble sulphate. Molecule for molecule, it reaches a serum concentration 10 times that of any other steroid hormone in young men, but between the ages of 35 and 70 the concentration falls to 20% of its peak value. Within populations, serum values show great variation,3 and in the only published longitudinal study 15 of 97 men actually showed an increase in their serum DHEA over the 15 years of observation.4 Little is known of DHEA's physiological functions, though it may serve as a precursor of both testosterone and oestradiol. However, epidemiological studies have indicated that a low serum concentration of DHEA is associated with an increased risk of cardiovascular disease in men over 50 years of age5 and of breast cancer in premenopausal women.6

    The mass media have simplified the issues, which they have presented as making four points. Firstly, the secretion of DHEA declines with age; secondly, this decline is associated with an increasing frequency of disability and disease, with low serum concentrations of DHEA apparently increasing the risks; thirdly, experiments on animals indicate that treatment with DHEA reduces the severity of a variety of diseases; and, fourthly, replacement therapy carries little risk. As DHEA can be obtained in the United States by mail order without a prescription, self experimentation is not only possible but is common, reflecting a growing interest in and acceptance of “alternative” medicine in the United States.7 Some distributers claim that DHEA is now a best seller. Health food shops are cashing in on the public's interest in DHEA by offering products labelled DHEA but which do not contain DHEA. Instead they contain plant sterols said to be the “botanical building blocks of DHEA.”

    By contrast with all this public attention given to DHEA, little has appeared in general medical journals apart from an editorial note in the Lancet last year.8 DHEA was not mentioned in a review of androgens in men in the New England Journal of Medicine in March 1996.9 Clearly we need to know whether DHEA does enhance “well-being” in middle aged and elderly people and whether it delays the onset of disability and disease, and we also need to know whether it is safe. The United States National Institutes of Health is supporting a few studies of DHEA. One is investigating the role of DHEA in Alzheimer's disease and another is examining the effect of DHEA supplementation on disease and lifespan in mice. The literature reviewed in the New York conference report included many reports of experiments on animals—but since none of the animals studied synthesise DHEA, this work is of doubtful relevance to the question of replacement treatment in humans.

    The best available clinical evidence comes from three placebo controlled, double blind, crossover studies of DHEA replacement treatment in older adults.10 11 12 13 In the first, 17 women and 13 men between the ages of 40 and 70 years (average 54 years) who did not smoke, were not obese, and enjoyed good health took 50 mg of DHEA by mouth at bedtime for three months or placebo for three months followed by three months on the other preparation.10 The order of DHEA and placebo was random. Compliance was checked by pill count and confirmed by measuring serum concentrations of DHEA and its sulphated form. Within two weeks of their starting to take the treatment the serum concentration of DHEA in the volunteers rose between threefold and fivefold, reaching the range found in serum of young adults. Serum concentrations of androstenediol, testosterone, and dihydrotestosterone rose in the women taking DHEA, but only androstenediol rose in the men. The treatment did not affect the amount of body fat or glucose metabolism. In both men and women the serum concentration of insulin-like growth factor rose by around 10%.

    When the volunteers were asked how they felt, the most striking changes they associated with DHEA were in the self reported sense of “well-being” (improved quality of sleep, greater energy, and increased ability to handle stress). Positive responses were reported by 82% of the women and 67% of the men—and fewer than 10% of the volunteers reported these effects while taking placebo. No change in libido was reported by volunteers while taking DHEA. Increased facial hair was reported by one woman while taking placebo and by another woman while on DHEA. No change in liver or thyroid function was detected (although hepatitis has been reported elsewhere.9)

    Another study looked at changes in the lymphocytes in 11 postmenopausal women (average age 56 years) given either 50 mg of DHEA or placebo each morning for three weeks.11 12 The treatment phases of the study were separated by a two week washout period. Previous oestrogen replacement was stopped at least three weeks before entry into the study. The results showed that the number and activity of natural killer cells (CD8+/CD56+) increased by the end of the period of treatment with DHEA. Treatment was also associated with an increase in insulin binding and degradation by T lymphocytes, and this was thought to reflect increased insulin sensitivity. No adverse effects were reported.

    In the third study eight men and eight women aged 50-65 were given 100 mg of DHEA or placebo for six months.13 The investigators confirmed their previous observation10 that the intake of DHEA was associated with an increase in insulin like growth factor. Treatment at the higher dose led to an increase in lean body mass that reached statistical significance when both men and women were analysed together. Muscle strength at the knee was increased in men but not in women. Because the 100 mg dose of DHEA stimulated a greater increase in serum androgens in women than had been observed with a dose of 50 mg, the authors concluded that this amount “may induce undesirable androgenic effects with time.” The effect of DHEA on well being was not discussed.

    Writing in the Lancet, Herbert concluded: “Enough is known or suspected to warrant investigation of DHEA as an effective, worthwhile, and relatively risk-free replacement therapy in advancing age. This can only be done by a controlled prospective trial of DHEA treatment.”8 This recommendation may be scientifically sound but it will be difficult to follow. The pharmaceutical industry may hesitate to support studies with DHEA, a molecule that cannot be patented. Similarly, it may not be easy to attract research funds from government agencies or private philanthropies at a time when support is shifting from clinical trials to fundamental research. Further research studies confirming or rejecting the reported results of DHEA replacement treatment seem unlikely to appear in the near future.

    What advice should a doctor give when asked about the treatment? In my opinion there is not at present sufficient evidence for routinely recommending replacement treatment with DHEA to middle aged and elderly people. Furthermore, patients with a family or personal history of a tumour responsive to hormones—for example, breast cancer or prostate cancer—should be dissuaded from taking DHEA. Treatment should be restricted to patients with low serum levels of DHEA and normal levels of prostate antigen. This means that patients who say that they want to take DHEA should have their serum DHEA and prostate antigen levels determined before treatment. All patients receiving DHEA should be asked to attend their doctor for regular clinical review and monitoring of their serum levels of DHEA and prostate specific antigen.


    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    6. 6.
    7. 7.
    8. 8.
    9. 9.
    10. 10.
    11. 11.
    12. 12.
    13. 13.

    Log in

    Log in through your institution


    * For online subscription