DHEA or Dehydroepiandrosterone

By Dr. Abraham Kryger, MD, DMD

I. Introduction 

DHEA has an interesting history. DHEA is produced by the zona reticularis of the adrenal cortex, in response to ACTH stimulatory signaling and is the hormone that is present in the greatest abundance in circulation. DHEA levels peak at around age twenty, and then decline by approximately ten percent per year, with great individual to individual variability. By age 70, DHEA levels reach a relatively stable trough level of 10-20 percent of its young adult levels.

DHEA was first isolated in 1931, by Dr. Adolf Buternandt, whose data were published in 1934. ( Butenandt A. et Dannenbaum H., 1934. Z physiol, Vol. 229, 192.) The publication of the structure of DHEA preceded that of testosterone by one year. 

In the 1950's, DHEA was isolated from human serum, and in the latter part of that decade, the first reports of DHEA demonstrating a decline in level with aging were reported. This was noted for both males and females. 

In the 1960's, DHEA was found to be an intermediate molecule in the testosterone synthesis pathway, and this data was the basis for the initial embrace of DHEA supplementation as a potential body building/performance enhancing aid. The utility of DHEA in this regard was generally unspectacular. High dose DHEA supplementation favored conversion to estradiol in addition to any small benefit in terms of testosterogenic effect, which limited its value as an androgenic supplement, and it's use as a high dose supplement was short lived. By the 1980's, DHEA was universally stocked in nutrition and health stores at lower doses and marketed as a more general aid to good health and function with aging. 

DHEA followed a checkered path in the subsequent decade. In 1986 the FDA reclassified DHEA as a controlled drug based on the lack of well delineated dosing data and a paucity of literature regarding the risks of long-term supplementation; but in 1994, the U.S. Dietary Supplement Health and Education Act re-reclassified DHEA (this time based on the lack of data demonstrating long term supplement risk) as a food supplement, once again allowing it to be sold over the counter, where it has remained since. 

When looking at DHEA as a weapon in the medical armamentarium, it is not typically mentioned in the same tones and with the same volume as testosterone, growth hormone, thyroid hormones, or estradiol and progesterone, but like a good character actor, it still plays a role in a balanced approach to the consideration of the utility of hormone supplementation. DHEA would be the Harvey Keitel of hormones, to Testosterone's John Travolta, or Joan Cusack to estradiol's Meryl Streep – important in the big picture, integral to the fabric of the film, but not the star. 

No matter the subject of a hormone discussion, the standard caveat applies. We should require more than mere association between the presence or absence of a hormone and an associated finding; we should also require data that demonstrates the effect of altering that state. Whenever possible, Cenegenics strives to follow literature that is not only based on the association of low or high hormone levels and disease risk, but also examines outcomes associated with directed intervention. Only meteorologists and economists get the luxury of only telling people what has happened, we want data that give us an idea of what our interventions will do in the future. 

So, today we will endeavor to put the use of DHEA in a rational clinical context. As with testosterone in last month's conference, today we will focus on the literature of DHEA and its replacement data in a system by system review. 

Quick note regarding DHEA levels: 

DHEA, like cortisol, follows a diurnal secretion/level pattern. This makes sense, given DHEA's secretion is, like that of cortisol, stimulated by pituitary derived ACTH. Unlike cortisol, DHEA secretion is prolonged rather than pulsatile, and DHEA has a longer serum half-life (10-20 hours), making for a much less exaggerated daily peak and trough levels. This also makes for readily valid measurement that is not as time dependent as it is for cortisol. 


II. General

A good starting point in the DHEA literature is a study published in the Proceedings of the National Academy of Science in 1996, by Claudine Berr. This study followed 622 adults, with an average age of 74 upon study entry. The study population was followed for DHEA levels and total mortality risk, functional status, psychological state, and mental status over the proceeding 4 years. In women, DHEA levels were directly related to scores of well-being, cognitive function, and functional status. In men, DHEA levels were inversely related to total mortality risk. Neither group demonstrated a specific Alzheimer's Disease (AD) risk based on DHEA levels, and the mortality data for women at the end of 4 years was not significant. 


Barrett-Connor, in the New England Journal of Medicine, 1996, published a prospective study of baseline dehydroepiandrosterone sulfate levels and associated risk for mortality and cardiovascular disease in 242 men, aged 50-79, who were followed over a 12 year period. In men with no history of heart disease at base line, the age-adjusted relative risk associated with a DHEAS level below 140 micrograms per deciliter was 3.3 (P less than 0.05) for death from cardiovascular disease, and 3.2 (P less than 0.05) for death from ischemic heart disease. In multivariate analyses, an increase in DHEAS level of 100 micrograms per deciliter was associated with a 36 percent reduction in mortality from any causes (P less than 0.05) and a 48 percent reduction in mortality from cardiovascular disease (P less than 0.05), after adjustment for age, systolic blood pressure, serum cholesterol level, obesity, fasting plasma glucose level, cigarette smoking status, and personal history of heart disease. Of greatest interest here is that baseline determination of DHEA levels was associated with long term health risk, even in subjects with no diagnostic history of previous pathology. 


III. DHEA and Heart Disease

The relationship between low DHEA levels and risk for heart disease have been shown in several studies. 

The Massachusetts Male Aging Study (MMAS) included DHEA as part of their thorough prospective evaluation of heart disease mortality. The MMAS followed a random sample of 1,709 men aged 40-70 years at baseline over a nine-year period. Men in the lowest baseline quartile for DHEA levels had a relative CAD risk of 1.6 compared to those in the other three quartiles. This risk was independent of confounding risk factors including age, obesity, diabetes, hypertension, smoking, serum lipids, alcohol intake, and physical activity. 


Muller, in the Journal of Clinical Endocrinology and Metabolism (JCEM), 3002, performed a meta-analysis of the androgen/CAD literature, pointing out that 21 of 33 cited studies associated low DHEA levels with increased CAD risk. 11 of the studies were neutral, and one study associated higher values of DHEA levels with higher CAD risk. Muller's conclusion was that higher levels of DHEA are associated with a mild reduction in CAD risk. 


A few interesting side notes regarding DHEA and the heart: 

Osorio, in Hormone Research, 2002, reported that DHEA levels fell after acute myocardial infarction. The study broadly concluded that drops in DHEA level were an epiphenomenon of CAD rather than playing any risk related role, and only followed subjects nine days post-MI, but it does raise the intriguing point that cardiac tissue may interact with as yet to be elucidated hormone synthesis/inhibition pathways. 


Nakamura, in Circulation, 2003, demonstrated localized production of DHEA by cardiac tissue, and showed a decline in DHEA production in association with a rise in aldosterone production in the failing heart. His hypothesis is that DHEA exerts a cardioprotective effect and is associated with more favorable non-androgenic hormone levels in healthy hearts. Whether or not this is accurate, its association between hormones and their effects on each other is intriguing. 


In a study published in the JCEM this month, Muller reports the results of a 400 subject study demonstrating that a 1 SD above the mean DHEA level was associated with a 0.76 relative risk for the presence of metabolic syndrome, an important CAD risk marker. 


IV. DHEA and its Effect on Mood and Well-Being: 

For decades, DHEA has been touted as a useful intervention for treating depression or improving mood. In the study previously mentioned by Berr, women whose DHEA levels were better maintained scored higher on measures of mood. This correlation has been shown in several articles: 

Cawood, in 1996, evaluated mood in relation to ovarian and adrenal steroids and found that only DHA was associated with their measurement modality of well being. 


Barrett-Connor, in the Journal of the American Geriatric Society , 1999, published data from one of my favorite study groups, the Rancho Bernardo longitudinal survey database. Plasma levels of estradiol, testosterone, estrone, androstenedione, cortisol, and DHEA were measured. A study of 699 women, aged 50-90 demonstrated a significant inverse correlation between DHEA levels and mood scores on the Beck Depression Inventory. These findings were age independent. DHEA had better predictive value than any other hormone evaluated in the study. 


In addition to correlation between DHEA levels and mood, interventional studies have demonstrated efficacy of DHEA supplementation and measures of mood. 

A small prospective pilot study published by Bloch in 1999, looking at “mid-life dysthymia,” found that sixty percent of their subjects had responded to DHEA monotherapy at the end of the 6-week treatment period compared with 20% in the placebo group. The symptoms that improved most significantly were anhedonia, loss of energy, lack of motivation, emotional "numbness," sadness, inability to cope, and worry. 


Wolkowitz, in 1999, demonstrated similar findings:


Schmidt, in the Archives of General Psychiatry, February, 2005, demonstrated the utility of DHEA as mono-therapy for both major and minor mid-life onset depression in both men and women. DHEA produced a greater than 50% reduction in depression scores for 50% of subjects, vs. a 23% score reduction rate for placebo. Treatment subjects also outperformed the placebo group in terms of self reports of sexual function. 


Just as Co-Q10 levels are affected by statin therapy, and demonstrate the importance of maintaining levels of endogenous compounds that can be iatrogenically affected, the same may hold true for DHEA during pharmacologic therapy for depression. In one study (Deuschle, in Neuropsychobiology, 2004) showed that amitryptiline therapy was associated with a 39% reduction in circulating DHEA levels. 


V. DHEA and Autoimmune/Inflammatory Disorders: 

Some disease states outside the typical hormonal realm have been demonstrated, as well. There are many literature citations dealing with disease state and DHEA levels in addition to demonstrating some clinical benefit associated with DHEA replacement. 

Patients with SLE/Lupus have been shown to have diminished levels of DHEA. (Small, but representative study) 


This low DHEA state also correlates with diseases specific pathophysiology. For example, SLE patients demonstrate an increase in terminal differentiation of peripheral blood mononuclear cells, which correlates with disease activity, and exposure to DHEA was associated in vitro with a diminishment in the pathological differentiation patterns. 


SLE patients demonstrate lower levels of DHEA than in normal controls, these diminished levels are associated with specific disease enhancing pathology, and low levels are also associated with loss of “pathology preventing” signaling between immune system cells, as well. Interferon gamma (IFN g) is associated with negative feedback inhibition of immune response in normal controls, and DHEA facilitates this signaling. In SLE, normal DHEA levels and signaling effects are lost, with the subsequent change favoring an exaggerated immune response rather than a normal one. 


The above findings would lead one to predict a therapeutic role for DHEA in the treatment of SLE, and this has been shown to be the case. Studies have repeatedly demonstrated the utility of DHEA supplementation in terms of reduction of symptom scores or reduction in accompanying medication requirements. This effect has been greatest in patients with mild or moderate disease, but DHEA has had clinical value as part of multi-drug therapy in severe cases and may also be associated with mitigating the osteoporosis associated with long-term corticosteroid regimens: 




SDHEA has small utility in severe SLE but associated with possible reduction in steroid requirements and steroid related bone loss: 


DHEA suppresses IL-10 secretion by T-cells in patients with SLE. 


The effect of DHEA on IL-10 secretion demonstrates the substantial “crossover” that is seen in the pathophysiology of inflammatory disorders. It has also been demonstrated that DHEA supplementation has disease score modulating effects on other inflammation based disorders such as Crohn's Disease and Ulcerative Colitis. 


In addition to SLE, DHEA therapy may be of utility in Rheumatoid Arthritis, Polymyalgia Rheumatica, and some aspects of Sjogren's Syndrome, but there are only preliminary reports of diminished DHEA secretion or function and scattered early speculative reports regarding DHEA therapy. These are tidbits only, and are not definitive, in any regard. 


DHEA may aid in regulation of some atopic inflammation:


In addition to affecting disease physiology, DHEA may also be associated with minimizing the negative effects of other therapies. Even low dose inhaled corticosteroid therapy in the management of asthma can be associated with an increase in osteoporosis risk, and these therapies are associated with diminished DHEA levels, as well: 



This effect has been shown for other steroid requiring disorders, as well: 


Which leads us to… 

VI. DHEA and Bone: 

Even in otherwise normal patients, low DHEA levels correlate with lower bone mineral density (BMD) and osteoporosis risk, making it yet another disease risk marker for loss of BMD. 




Females in the lowest quartile for DHEA levels have been demonstrated to be at increased fracture risk compared to normal risk: 


Conversely, well maintained DHEA levels are associated with retained BMD: 


Per our usual discussions, an association between level and risk is not sufficient validation for therapy, and DHEA supplementation has been shown to improve BMD in men with osteoporosis. Over a 12 month study period, BMD increased in the lumbar spine and femoral neck by nearly 3%. 


The same types of findings have been seen in female replacement studies, with increases in BMD and decreases in markers of bone resorption. 


The segue from DHEA and bone to other organ systems can be found in the next reference. Interleukin 6 (IL-6), is a pro-inflammatory cytokine whose presence is associated with inhibition of osteoblast function – leading to declines in BMD. IL-6 levels are also associated with other potential disease risk, such as heart disease and dementia. Lower DHEA levels are associated with an elevation in IL-6 levels, and replacement is associated with declines in IL-6. So, in this study by Straub in JCEM, 1998, we have results in terms of a disease risk marker that is multi-system in terms of risk and in terms of risk reduction. 


Another aside, done sotto voce. There are null result studies for DHEA, as well. When reviewing the literature, it may be best to take search results in an electoral fashion and add up the number of studies or total number of subjects studied, and then determine one's opinion. The following references show no effect on BMD of DHEA, reminding us that perhaps universal dosing is not always an optimal method of treating a patient group, nor is treating without titration to a given level vs. using even a dose/weight method. These results also act as a reminder of what we promise, and not all patients will receive all the benefits of a given intervention. As one additional note, these results also point out the lack of any negative outcome, which is useful in determined the safety part of the safety and efficacy of a given intervention. 



So, transition to… 

VII. DHEA and the Brain: 

Some studies have demonstrated a correlation between Alzheimer's Disease (AD) and DHEA levels. Again, these results exist within the context of multiple risk factors, but there is a statistical inverse correlation between DHEA levels and AD. 

Weill-Engerrer found a significant negative correlation between the levels of cortical beta-amyloid peptides DHEA in the hypothalamus. 


Another study found higher cortisol levels, lower DHEA levels, and a higher cortisol/DHEA ratio in subjects with AD than is seen in control subjects. 


Yet another study demonstrated diminished circulating DHEA levels AD subjects. 


In patients with pre-existing AD, those with higher DHEA levels performed better on cognitive function testing. 


Here comes the important part, and why we insist on more than correlation between a finding and assuming that altering that finding will have an effect. AD patients treated with DHEA showed no improvement on cognitive measures. 


Neither did studies on normal older people. 


At present, I would not try to claim any certain value for DHEA in the course of preventing or treating AD. Given the large variability of lesion number and precise location of AD plaques or pathology, different study results may only be expressing findings for the myriad sites of AD damage and anatomical rather than metabolic consequences with regard to AD and its effect on the brain/HPA/adrenal axis. 

Rat studies have demonstrated a neuroprotective effect of DHEA on experimentally induced oxidative damage to neurons, so this may one day show a role for DHEA in cytoporotection. This may also overlap with the clinical benefit of DHEA in autoimmune disorders. 


VIII. DHEA and Erectile Dysfunction: 

As a member of the androgen family, DHEA has been looked at regarding its possible usefulness in the treatment of erectile dysfunction (ED). The ED studies are not exhaustive in their attention to detail… 

DHEA has been shown to be lower in men with ED. Keep in mind that many times, this is in conjunction with low levels of all androgens or may have been measured without consideration for other androgen levels. Needless to say, if monotherapy works, however, then effective is still effective. 


DHEA supplementation was shown to improve ED in subjects with hypertension, but no mention was made as to which, if any, medications were also involved, and whether any specific class of anti-hypertensive was more or less likely to contribute to the incidence of ED or the outcome of DHEA intervention. 


IX. DHEA, General Considerations: 

DHEA levels in patients who subsequently developed prostate cancer over a 12 year period, had DHEA levels 11-12% below those of controls. 


Today's final words on DHEA belong to women. DHEA supplementation was shown in one 12 month study on postmenopausal women to increase HDL by 11%, decrease LDL by 11%, and improve insulin sensitivity, at a dose of 25 mg/day. Again, global reduction in markers of disease risk. 


Finally, a “who knows what this will mean for the future” study on the effects of DHEA on postmenopausal women showed a change in T-cell profile from CD4+ to CD56+ (NK) cell profiles in association with inhibition of IL-6 and T cell mitogenic responses with a simultaneous dramatic increase in NK cell cytotoxicity. These effect were consistent with an “antioncogenic” profile.