| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
BRIEF REPORT |
Department of Obstetrics and Gynecology (A.K., R.A.), Cedars-Sinai Medical Center, Los Angeles, California 90048; Department of Obstetrics and Gynecology (A.K., R.A.), David Geffen School of Medicine at the University of California, Los Angeles, Los Angeles, California 90095; Department of Gynecology and Obstetrics (J.B.S.), Emory University, Atlanta, Georgia 30308; and School of Public Health (M.K.), Emory University, Atlanta, Georgia 30322
Address all correspondence and requests for reprints to: Ricardo Azziz, M.D., M.P.H., M.B.A., Department of Obstetrics and Gynecology, Cedars-Sinai Medical Center, 8635 West Third Street, Suite 160W, Los Angeles, California 90048. E-mail: azzizr{at}cshs.org.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Objective: Our objective was to determine the impact of racial differences and body mass index (BMI) on the change in androgen levels during a woman’s reproductive and early menopausal years.
Design and Setting: We conducted a frequency-matched cross-sectional study at a tertiary academic medical center.
Patients or Other Participants: Subjects included 260 healthy, nonhirsute and eumenorrheic, self-identified Black and White women, ages 15–60 yr.
Interventions: A medical and reproductive history, physical exam, and blood sampling were determined in the fasting state during the early follicular phase.
Main Outcome Measures: Serum levels of androgens or androgen metabolites (dehydroepiandrosterone sulfate, androstenedione, and total and free testosterone) and SHBG were measured and the BMI, the waist-to-hip ratio (WHR), and the basal insulin resistance estimated by the homeostasis model assessment for insulin resistance determined.
Results: After controlling for differences in BMI, insulin resistance, and WHR, Black women had lower androgen levels than age-matched White women. All androgens, or androgen metabolites, declined similarly across the reproductive lifespan and menopausal transition in both Black and White women. Race was a significant predictor of dehydroepiandrosterone sulfate, androstenedione, and total and free testosterone but not SHBG.
Conclusions: Eumenorrheic, nonhirsute Black women have a lower range of normal androgen levels than White women of the same age, BMI, WHR, and homeostasis model assessment index for insulin resistance. Race and age-adjusted data should be considered when evaluating androgen levels in women between the ages of 15 and 60 yr.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Androgens decline with age. The clearest decline observed is with the adrenal products DHEA and DHEAS, which peak in women during their late teens and then decline progressively with age (2, 3). The circulating levels of other androgens also appear to decline with age, although not as clearly or pronounced. A4 may not start to decline until after their mid-40s and may plateau after a 25% decrease by their 60s (4). TT appears to decline up to 50% in a woman’s midlife years with only a minimal reduction after menopause (3, 5). Finally, it is unclear whether SHBG and FT change significantly before menopause.
Several large studies have confirmed that mean androgen levels vary by race, with most suggesting that Black individuals have lower levels of androgens than Whites. Young Black women between the ages of 18 and 36 yr were found to have significantly lower levels of TT, FT, A4, and estrogen, even after controlling for age and body mass index (BMI) (6). This relationship persisted in the larger population of older reproductive-aged women participating in the SWAN study (3). Furthermore, some investigators have demonstrated that circulating levels of SHBG vary by race and age (2, 7). However, it is not clear whether these differences are prevalent during only part of the reproductive lifespan or persist throughout and whether the age-related decline in androgens normally observed differs between Black and White women.
We have hypothesized that after accounting for differences in BMI and insulin resistance, race is associated with differences in circulating androgen levels and that this difference persists throughout the reproductive lifespan, possibly altering the rate of the age-related decline. To test this hypothesis, we studied 260 healthy eumenorrheic, nonhirsute Black and White women, who were recruited by age.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
A total of 260 (121 Black and 139 White) healthy, eumenorrheic women were recruited; none had clinical evidence of androgen excess (e.g. hirsutism) or were on medications known to affect androgen levels. The subjects were recruited to meet specific age groups (15–19, 20–29, 30–39, 40–49, and 50–60 yr) to ensure even distribution across the reproductive age and menopausal transition. The Institutional Review Board at the University of Alabama at Birmingham approved the study. Subjects responded to posted advertisements and provided written informed consent.
All women underwent a brief history and physical exam, primarily to exclude the presence of medical disorders. Medications, menstrual dysfunction, acne, hirsutism, and anthropometric measures, including height, weight, and waist and hip circumferences, were recorded. Subjects underwent blood sampling one time in the morning, between 0700 and 1000 h, in the fasting state, and between d 3 and 8 of the menstrual cycle. Serum was separated and stored at –80 C until assayed. These subjects were included in a previous report assessing the prevalence of DHEAS excess in women with polycystic ovary syndrome (8).
Assays
Serum samples were analyzed for A4, TT, FT, DHEAS, and SHBG levels. TT was measured by an in-house RIA method (for low and high T levels, the interassay coefficients of variation were 1.2 and 5.3%, respectively, and the intraassay coefficients of variation were 2.4 and 3.1%, respectively) after serum extraction with ether and using dextran-coated charcoal for separation of bound and free, as previously described (9). SHBG activity was measured by diffusion equilibrium dialysis, using Sephadex G-25 and [3H]T as the ligand, and the FT was calculated as previously described (10). The levels of DHEAS and A4 were measured by direct RIA using a commercially available kit (Diagnostic Products Corp., Los Angeles, CA). Samples were assayed in batches to minimize the effect of interassay variation. The intraassay coefficient of variation for all steroids did not exceed 9%.
Statistical analysis
The homeostasis model assessment of insulin resistance (HOMA-IR) was calculated from fasting glucose and fasting insulin levels with the following equation: [glucose (mmol/liter) x insulin (µIU/ml)]/22.5.
A two-sample Student’s t test was first used to determine age-matched but otherwise uncontrolled differences in BMI, waist-to-hip ratio (WHR), HOMA-IR index, and androgen and SHBG levels in our Black and White subgroups.
Androgens and SHBG were reasonably normally distributed. Hormone levels in Blacks and Whites were then compared with multivariate linear regression in SAS version 9.0. Covariates chosen for the initial model were age, race, BMI, HOMA-IR index, and WHR, with the androgen or SHBG level as the dependent variable. Pearson’s correlation was used to assess colinearity, and Lowess regression of each androgen confirmed that the linear assumption was appropriate. Interaction variables were compared among race, BMI, HOMA-IR index, and WHR, but none of them were significant in the model. However, because our primary area of interest was the relationship between race and age, the interaction term [(race)(age)] was included. We used a significance level of 0.05.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
). DHEAS was also lower in Black women, but the difference was not statistically significant (P = 0.069). There was no difference in SHBG levels between the two races. Given the significant differences in weight and insulin resistance, linear regression was used for subsequent analysis. Among all the covariates, only TT, FT, and A4 were strongly correlated with each other (Pearson correlation coefficient > 0.60).
|
The results of our linear regression (Fig. 1) are the predicted androgen or SHBG level as a function of age for the average woman in this cohort. Specifically, the lines shown represent the theoretical androgen level for a woman who is either Black or White with a BMI of 27.75 kg/m2, a HOMA-IR index of 2.39, and a WHR of 0.806. Race and age were significant parameters in the model for DHEAS, A4, FT, and TT. WHR was a significant parameter in A4, FT, and TT. The effect of race on the age-related decline of each androgen was not significant (evaluated with the product term race x age). In other words, the rate of decline did not significantly differ between Black and White women. We did not detect a significant difference in SHBG between Black and White women, and SHBG did not decline with age in any group.
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Black women had lower levels of androgens than White women. As expected with age, the androgen levels of both Black and White women declined significantly. A nonsignificant trend suggested that the difference between the mean androgen levels in Black and White women became smaller with age; thus, the greatest impact of race on androgen levels was in the younger reproductive-aged women. SHBG levels were not different between Black and White women and did not change significantly with age.
Our results are consistent with other reports that demonstrate that androgens decline during reproductive aging (3, 12), whereas SHBG is relatively stable during this period of time (12). Several other studies have also remarked on finding significantly lower androgen levels in Black women compared with White women (3, 6); however, these cohorts did not span the entire reproductive lifespan and menopausal transition. Our data differ from another report of a large cohort of women (ages 35–47 yr) that found a more rapid decline in DHEAS levels in Black compared with White women and no racial variation for TT (13).
Body mass is known to be inversely correlated with SHBG levels (14); alternatively, SHBG does not seem to be altered by race once BMI is controlled for. Our data are consistent with the older SWAN cohort (ages 42–52 yr) (2) and a study of younger subjects (ages 18–36 yr) (6) that observed that SHBG did not vary among the races. However, the HERITAGE family study (ages 17–65 yr) did observe differences in SHBG levels between Black and White women (12).
Body fat distribution and insulin resistance likely play a role in the observed androgen differences between Black and White women. In a study of 106 premenopausal Black women, those subjects with an upper body fat distribution had 50% higher levels of TT and FT than those with a lower body fat distribution (15). Hyperinsulinemia may result in augmented androgen production by the ovarian theca (16), and the extent of insulin’s effect on ovarian androgen production may vary by race. Furthermore, smoking has been associated with elevated androgen levels in women, but this information is not available in this cohort.
As suggested by the data in the present study, it is likely that androgen levels are determined, to a significant extent, by genetic or inherited factors. For example, androgen levels were highly heritable among both Black and White parent-offspring and sibling pairs (17). Part of this variation may be due to an inherited variation in the CAG repeat of the androgen receptor (18). Smaller repeats were associated with higher androgen receptor transcriptional activity and higher androgen levels. The variation with race observed may also be partially explained by inherited differences in the peripheral production and conversion of circulating androgens, which may be overlooked by serum measurements of androgens alone (19). Finally, genome-wide linkage studies have implicated various loci that influence androgen levels differently for Black and White women (20).
The present study demonstrates the role of race in determining androgen levels, particularly among younger subjects. Higher levels of androgens have been associated with an increased risk for heart disease (21); however, this would not explain the higher rates of cardiovascular disease and mortality seen in Black women (22). These data also provide an estimate of the normal age-related decline in androgen levels in Black and White women that should be considered when establishing the presence of hyperandrogenemia in women.
|
Footnotes |
|---|
Disclosure Summary: The authors have nothing to declare.
First Published Online September 25, 2007
Abbreviations: A4, Androstenedione; BMI, body mass index; DHEA, dehydroepiandrosterone; DHEAS, DHEA sulfate; FT, free testosterone; HOMA-IR, homeostasis model assessment of insulin resistance; TT, total testosterone; WHR, waist-to-hip ratio.
Received October 30, 2006.
Accepted September 14, 2007.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  F. Laghi, N. Adiguzel, and M. J. Tobin Endocrinological derangements in COPD Eur. Respir. J., October 1, 2009; 34(4): 975 - 996. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. J. Broekmans, M. R. Soules, and B. C. Fauser Ovarian Aging: Mechanisms and Clinical Consequences Endocr. Rev., August 1, 2009; 30(5): 465 - 493. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Azziz Diagnosing the diagnosis: why we must standardize the defining features of polycystic ovary syndrome Ann Clin Biochem, January 1, 2008; 45(1): 3 - 5. [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 |
