| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Endocrine Care |
Clinical Gerontology Unit, University of Cambridge School of Clinical Medicine, Addenbrooke’s Hospital, Cambridge CB2 2QQ, United Kingdom
Address all correspondence to: Dr. D. Trivedi, Clinical Gerontology Unit, Level 2, Box 251, University of Cambridge School of Clinical Medicine, Addenbrooke’s Hospital, Cambridge CB2 2QQ, United Kingdom. E-mail: dpt21{at}medschl.cam.ac.uk
Abstract
Dehydroepiandrosterone sulfate levels have been inversely related with cardiovascular mortality in men, but findings have been inconsistent, and there are few data in women.
We examined the relationship between baseline circulating dehydroepiandrosterone sulfate levels and subsequent all-cause and cardiovascular mortality in 963 men and 1171 women, 65–76 yr old, surveyed in 1991–1995, and followed up until August 2000 (when 296 deaths had occurred).
All-cause and cardiovascular disease mortality rates were highest in the lowest dehydroepiandrosterone sulfate quartile in men; and thereafter, rates did not differ significantly in the upper three quartiles. This pattern remained after excluding those with previous history of cardiovascular disease and, in multivariate analyses, was independent of age, cigarette smoking habit, systolic blood pressure, body mass index, blood cholesterol, and steroid use. There was no significant association of dehydroepiandrosterone sulfate and mortality in women.
The multivariate adjusted relative risks for all-cause mortality by sex-specific increasing quartile of dehydroepiandrosterone sulfate were 1.00, 0.66 (95% confidence interval, 0.44–1.01), 0.70 (0.46–1.07), 0.73 (0.48–1.10), respectively, for men and 1.00, 0.71 (95% confidence interval, 0.41–1.24), 0.97 (0.58–1.62), and 1.14 (0.69–1.88), respectively, for women.
In older men and women, there is no consistent relationship between dehydroepiandrosterone sulfate and all-cause or cardiovascular mortality. The highest mortality rates were observed in the lowest quartile in men, but the highest rates were in the highest quartile in women.
THE SECRETION AND circulating levels of the adrenal steroid dehydroepiandrosterone (DHEA) and its sulfate (DHEAS) decrease with aging (1, 2). A number of reports have suggested that circulating DHEAS levels are inversely related to cardiovascular disease and mortality, but findings have not been entirely consistent (3, 4, 5, 6). Most studies have been in men (7), and only one study has reported results in men who are 30–82 yr old and women who are 50–88 yr old, from the same cohort (8); this showed a reduction in the risk of cardiovascular mortality with increasing DHEAS in men but not in women. A U-shaped relationship in mortality rates, with increasing DHEAS level in 60- to 79-yr-old women from the same cohort, has been reported (9). We examined the relationship between DHEAS and mortality in a cohort of older men and women (65–76 yr) living in the community in Cambridge.
Materials and Methods
The Cambridge General Practice Health Study invited men and women, 65- to 76-yr-old residents in the community, identified using general practice age-sex registers, to participate in a health survey, and about 50% of those mailed agreed to participate. All those who consented completed a health and lifestyle questionnaire, which included questions on past medical history and cigarette smoking habit. They then underwent a health examination at Addenbrooke’s Hospital, Cambridge.
At the health check, height and weight were measured in light clothing without shoes. Body mass index (BMI) was calculated as weight in kilograms divided by height in meters squared. Blood pressure was measured twice using an Accutor, after the participant had been seated at rest for 5 min, and the mean of the two readings was used for analysis. Serum cholesterol was measured at the routine biochemistry laboratory at Hinchingbrooke Hospital, Huntingdon, Cambridge, which conforms to the National External Quality Assurance scheme.
DHEAS assays were performed on nonfasting serum collected at baseline and subsequently frozen and stored at -70 C. The serum was thawed in 1995, and DHEAS was measured on all available samples in the Department of Clinical Biochemistry, University of Cambridge. A previous study (1) has shown that plasma sex hormone levels show no deterioration over a period of up to 15 yr, provided the specimens are frozen and stored in sealed containers. The hormonal assays were performed by RIA using a commercial kit (Coat-a-Count, Glyn Rhonwy, Llanberis, Caernarfon, Gwynedd). Studies in the laboratory, using this technique, showed intraassay coefficients of variation ranging from 11% for results less than 1 µmol to 5.5% for results above 12 µmol. The interassay coefficient of variation on quality control batches ranged from 5.9–10.7%.
Current smokers were defined as those who reported "yes" to the question "Do you smoke cigarettes now?" Steroid users were defined as those who reported "yes" to the question "Have you ever taken steroid tablets or injections for 3 months or more?" Women receiving hormone replacement therapy (HRT) were defined as those who reported "yes" to the question "Are you currently receiving any HRT?"
All participants have been flagged at the Office of National Statistics for mortality by cause and have been followed up to the present. We present data for mortality up to the end of August 2000. Death certificates were coded by a trained nosologist using the International Classification of Diseases, version 9 (10). Cardiovascular mortality was defined as International Classification of Diseases 400–438 for underlying cause.
Data were analyzed using SPSS, Inc. software, version 10.0 for Windows. Sex-specific quartiles of DHEAS and their association with covariates were analyzed using ANOVA. Because the DHEAS distribution was slightly skewed, we repeated analyses on log-transformed values, and findings were similar (not shown). Mortality rates were calculated for each quartile of DHEAS, with adjustments for age (in 5-yr age groups) using Mantel-Haenszel direct age adjustment (11).
The relationship of mortality to DHEAS was not linear in either men or women, so linear modeling in multivariate analysis was not appropriate. Using the DHEAS quartile as a categorical variable, we conducted a multivariate analysis to estimate the relative risks (RRs) of mortality for each DHEAS quartile in comparison to the lowest quartile using the Cox proportional hazards model (12), adjusting for age, blood pressure, BMI, cholesterol, smoking, past history of cardiovascular disease or cancer, and steroid or HRT use. This analysis was repeated separately, using men and women who had no previous history of cardiovascular disease or cancer, were nonsmokers, and were not taking steroids or HRT. All significance tests were two-sided.
Results
We present data for 963 men and 1171 women, 65–76 yr old, who had
DHEAS measurements available from 1991–1995. The baseline
characteristics for men and women who were alive and for those who had
died are shown in Table 1
. In total, 177
men (18.4%) and 119 women (10.2%) had died during a mean follow-up
period of 7.4 yr. Age-adjusted mean DHEAS levels were slightly lower in
men who had died (compared with men who were alive) and were higher in
women who had died (compared with women who were alive), but the
differences were not significant. Women had lower mean DHEAS levels
than men.
|
The baseline characteristics for men, according to sex-specific
quartile of DHEAS, are shown in Table 2.
Compared with the highest quartile of DHEAS, men and women in the
lowest quartile were older and more likely to have taken steroids.
Blood pressure was positively related to DHEAS quartile, significantly
so in men but not in women; women were more likely to be current
smokers in the highest DHEAS quartile and to report past history of
cardiovascular disease. Prevalence of steroid use increased with
decreasing DHEAS quartile.
|
. Mortality rates were higher in
men than in women overall. Though the mortality rates seemed to show a
U-shaped relationship with DHEAS quartile, there were no significant
consistent trends. For men, the highest rates were in the lowest DHEAS
quartile; whereas for women, the highest rates were in the highest
quartile. Exclusion of men and women with past history of
cardiovascular disease or cancer patterns did not change the U-shaped
relationship.
|
In the Cox regression model (adjusting for age, BMI, cigarette smoking
status, serum cholesterol, systolic blood pressure, past history of
cardiovascular disease or cancer, steroid or HRT use), analysis showed
a 34% significant decrease in RR for quartile 2, compared with the
lowest quartile in all men, for all-cause mortality [RR, 0.66; 95%
confidence interval (CI), 0.44–1.01; P = 0.05] (Table 4A), but quartiles 3 and 4 showed
approximately 30% nonsignificant decrease. The lower risk in quartile
2 was also apparent for risk for cardiovascular mortality (66%) (RR,
0.34; 95% CI, 0.18–0.64; P = 0.001).
In a post hoc analysis, we
compared quartile 1 with quartiles 2, 3, and 4 combined and observed
30% significant reduction for all-cause mortality (RR, 0.70; 95% CI,
0.50–0.96; P = 0.03). This reduction was greater and
significant for cardiovascular mortality (RR, 0.57; 95% CI,
0.38–0.86; P = 0.007). We repeated the analysis in
healthy men who did not report a past history of cardiovascular disease
or cancer, were nonsmokers, and were not taking steroids. The findings
were similar. In women, there were no significant trends concerning
mortality (Table 4B).
|
|
and 2).
|
|
There was no consistent relationship between DHEAS level and subsequent all-cause mortality or cardiovascular mortality in this cohort of elderly men and women; highest mortality rates were observed in the lowest DHEAS quartile in men, but highest rates were in the highest DHEAS quartile in women. Multivariate analyses (after adjusting for age, systolic blood pressure, BMI, serum cholesterol, smoking status, and steroid use or exclusion of persons with existing cardiovascular disease, cancer exclusion of current smokers, or steroid users) did not substantially change the results. These findings broadly reflect the lack of consistency reported from other studies.
A recent review of DHEAS and cardiovascular disease (7) indicated that, of 4 prospective studies that have reported on DHEAS and mortality, 2 reported significant reduction in RR for cardiovascular mortality, with increasing levels of DHEAS in men. (3, 4), both of which included a younger age range in men, compared with our older cohort. From the same cohort in the Rancho Bernardo Study, in 289 women, 60–79 yr old, a U-shaped relationship between DHEAS levels and cardiovascular mortality was reported in a 12-yr follow-up study (9). In a further 19-yr follow-up study of the same cohort of 942 women, 50 yr and over, DHEAS was unrelated to the risk of fatal cardiovascular disease (13). A nonsignificant RR of 1.11 was reported for a 50-µg/dl (1.36-µM) decrease in DHEAS. However, the study did not show mortality rates per quartile of DHEAS, and it is not clear whether a U-shaped relationship, originally reported in the older women, persisted in all women. If the relationship was U-shaped, analyses using a regression model with linear assumption or mortality rates, comparing groups above and below the median level, may not be appropriate and may obscure a U shaped relationship. Our findings for the elderly women in our cohort are similar to those reported earlier in 289 elderly women.
A more recent update on 19-yr follow-up for both men and women (8) reported a much lower reduction (RR, 0.85) in cardiovascular mortality in men, compared with those followed up for 12 yr (3). The authors attributed this difference to the comparison of survivors with cardiovascular deaths in the recent study, rather than with noncardiovascular deaths in the earlier study.
Our findings do not provide strong support for an inverse risk reported in men. In another report of DHEAS and cognitive function of the same cohort (14), 1547 men and women (50 yr and older) in the highest quartile of DHEAS had all-cause mortality rates 15% and 25%, respectively, lower than those in the lowest quartile. Surprisingly, age-adjusted rates did not show a trend from the lowest to the highest quartile for men but a trend was reported for women. Tilvis et al., 1999 (15), showed that DHEAS did not predict risk of all-cause or cardiovascular mortality in men and women, 75 yr and older, during a 5-yr follow-up period. Mortality rates in 75-yr-old women showed a U-shaped relationship with quartiles. A more recent prospective study, the Bruneck study (16) of men and women, 40–79 yr old, reported no association between DHEAS and incident atherosclerosis. Their findings did not support a role for endogenous DHEAS in the development of human atherosclerosis. Birkenhage-Gillesse et al. (17) found that men with higher DHEAS levels tended to survive less long and women with higher DHEAS levels to survive longer, in contrast to the inverse relationship between DHEAS and mortality (4).
In this cohort of men and women, two factors that were associated with DHEAS levels were cigarette smoking status (which was positively associated with DHEAS) and steroid use (which was negatively associated with DHEAS). Both cigarette smoking habit and steroid use were also associated positively with mortality in men and women. Steroid use has not previously been reported or adjusted for in studies, and it is possible that observed associations previously reported between DHEAS and mortality could be confounded by smoking and steroid use; the variable relationships observed could reflect varying prevalence of these two factors.
It is also possible that, in men of 65 yr and older, any possible protective effect of DHEAS would not be apparent, because levels decrease with age, and the effect of low DHEAS may be stronger in the younger population. Although age-specific survival curves from the Rancho Bernardo study (8) suggest that protective effect is not limited to younger age groups, the study only reported findings for DHEAS levels above and below the median level of DHEAS.
It is also difficult to make much of the observed sex difference, with the highest mortality rates observed in men in the lowest quartile but women in the highest quartile. Physiological effects of DHEAS may differ in men and women and may depend on the predominance of other sex hormones, such as estrogens in elderly men and women. Ebeling and Koivisto (18) have proposed that DHEA has either an estrogen-like or androgen-like effect, depending on the hormonal milieu. In postmenopausal women, its metabolism to testosterone may increase the risk of cardiovascular disease. In men, with a more androgenic environment, DHEAS may act like an estrogen and may protect against cardiovascular disease; this might explain the lower risk seen in higher quartiles. In this elderly cohort, men have a lower androgenic environment than the younger men previously reported; and, though this may explain the difference in RRs across each quartile, the U-shaped relationship is seen in both men and women. Elderly men, 60–80 yr old, in the highest quartile of DHEAS, have higher testosterone levels and a favorable lipid profile, compared with men in the lowest quartile (19); but there is no difference in elderly women. Low testosterone levels are associated with cardiovascular risk factors in men (20), which might also explain the higher risk in the lowest quartile. However, a review of studies examining the relationship between circulating levels of testosterone and DHEAS and coronary heart disease in men suggests a neutral or a favorable effect of testosterone and DHEAS on coronary heart disease (21), although the age range of the cohorts in the studies was wide. DHEAS may have an effect on different physiological mechanisms in elderly men and women, because they show differences in response to DHEAS administration (22, 23).
Data from clinical trials are rare; and, though they have reported on short-term surrogate endpoints [such as hormones or other biological or physiological parameters (23)], there is no trial using hard clinical endpoints such as mortality. Additionally, many of these trials were not adequately randomized or blinded.
Many uncertainties considering the role of DHEAS and its association with chronic diseases have yet to be addressed. Identifying the nature of this relationship may help clarify important lifestyle, physiological, and other determinants of cardiovascular and total mortality. In the interim, the current enthusiasm for DHEAS supplementation for health may still be premature in men and certainly is unfounded in women.
Acknowledgments
We thank the general practitioners and participants in the Cambridge General Practice Health Study.
Footnotes
This work was supported by a research grant from the Wellcome Trust.
Abbreviations: BMI, Body mass index; CI, confidence interval; DHEAS, dehydroepiandrosterone sulfate; HRT, hormone replacement therapy; RR, relative risk.
Received December 15, 2000.
Accepted May 5, 2001.
References
This article has been cited by other articles:
 |
|
 |
|
  S. P. L. Rice, N. Agarwal, H. Bolusani, R. Newcombe, M. F. Scanlon, M. Ludgate, and D. A. Rees Effects of Dehydroepiandrosterone Replacement on Vascular Function in Primary and Secondary Adrenal Insufficiency: A Randomized Crossover Trial J. Clin. Endocrinol. Metab., June 1, 2009; 94(6): 1966 - 1972. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Cocco, D. Fadda, and A.G. Schwartz Subjects expressing the glucose-6-phosphate dehydrogenase deficient phenotype experience a lower cardiovascular mortality QJM, February 1, 2008; 101(2): 161 - 163. [Full Text] [PDF]
|
|
|
|
 |
|
 M. Maggio, F. Lauretani, G. P. Ceda, S. Bandinelli, S. M. Ling, E. J. Metter, A. Artoni, L. Carassale, A. Cazzato, G. Ceresini, et al. Relationship Between Low Levels of Anabolic Hormones and 6-Year Mortality in Older Men: The Aging in the Chianti Area (InCHIANTI) Study Arch Intern Med, November 12, 2007; 167(20): 2249 - 2254. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. R. N. Pour, D. E Grobbee, M. Muller, M. Emmelot-Vonk, and Y. T van der Schouw Serum sex hormone and plasma homocysteine levels in middle-aged and elderly men Eur. J. Endocrinol., December 1, 2006; 155(6): 887 - 893. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Goldman, C. M. Turra, D. A. Glei, C. L. Seplaki, Y.-H. Lin, and M. Weinstein Predicting mortality from clinical and nonclinical biomarkers. J. Gerontol. A Biol. Sci. Med. Sci., October 1, 2006; 61(10): 1070 - 1074. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. R. Cappola, Q.-L. Xue, J. D. Walston, S. X. Leng, L. Ferrucci, J. Guralnik, and L. P. Fried DHEAS Levels and Mortality in Disabled Older Women: The Women's Health and Aging Study I. J. Gerontol. A Biol. Sci. Med. Sci., September 1, 2006; 61(9): 957 - 962. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Arnlov, M. J. Pencina, S. Amin, B.-H. Nam, E. J. Benjamin, J. M. Murabito, T. J. Wang, P. E. Knapp, R. B. D'Agostino Sr., S. Bhasin, et al. Endogenous sex hormones and cardiovascular disease incidence in men. Ann Intern Med, August 1, 2006; 145(3): 176 - 184. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. K. C. Ng, S. Nakhla, A. Baoutina, W. Jessup, D. J. Handelsman, and D. S. Celermajer Dehydroepiandrosterone, an adrenalandrogen, increases human foam cell formation: a potentially pro-atherogenic effect J. Am. Coll. Cardiol., December 3, 2003; 42(11): 1967 - 1974. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Muller, Y. T. van der Schouw, J. H. H. Thijssen, and D. E. Grobbee Endogenous Sex Hormones and Cardiovascular Disease in Men J. Clin. Endocrinol. Metab., November 1, 2003; 88(11): 5076 - 5086. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. C. W. Wu and A. von Eckardstein Androgens and Coronary Artery Disease Endocr. Rev., April 1, 2003; 24(2): 183 - 217. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. L. Lasley, N. Santoro, J. F. Randolf, E. B. Gold, S. Crawford, G. Weiss, D. S. McConnell, and M. F. Sowers The Relationship of Circulating Dehydroepiandrosterone, Testosterone, and Estradiol to Stages of the Menopausal Transition and Ethnicity J. Clin. Endocrinol. Metab., August 1, 2002; 87(8): 3760 - 3767. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
